Identification of neural oscillations and epileptiform changes in human brain organoids

Abstract

Brain organoids represent a powerful tool for studying human neurological diseases, particularly those that affect brain growth and structure. However, many diseases manifest with clear evidence of physiological and network abnormality in the absence of anatomical changes, raising the question of whether organoids possess sufficient neural network complexity to model these conditions. Here, we explore the network-level functions of brain organoids using calcium sensor imaging and extracellular recording approaches that together reveal the existence of complex network dynamics reminiscent of intact brain preparations. We demonstrate highly abnormal and epileptiform-like activity in organoids derived from induced pluripotent stem cells from individuals with Rett syndrome, accompanied by transcriptomic differences revealed by single-cell analyses. We also rescue key physiological activities with an unconventional neuroregulatory drug, pifithrin-α. Together, these findings provide an essential foundation for the utilization of brain organoids to study intact and disordered human brain network formation and illustrate their utility in therapeutic discovery.

Extended Data Fig. 1. Plots and table of batch and patient line variability for key experimental measures.

a, Plots of experimental results from different batches of iCtrl and Mut Cx+GE fusion organoids analyzed for the percentage of cells in the cortical compartment that expressed tdTomato (tdTom) after the GE portion was labeled with AAV1-CAG:tdTom virus (left panel) or GAD65 antibodies (right panel). Each dot represents an individual organoid section used for analysis and numbered elements on the x-axis represent individual experiments. No significant within or across genotype differences were noted for either percentage of Cx expressing tdTom or GAD65. b,c, Plots of individual experimental results from iCtrl and Mut Cx+GE fusion calcium indicator and LFP experiments. Each dot represents results from an independent experiment, numbered elements on the x-axis represent independent organoid batches. Blue dots represent hiPSC line I (Rett patient with a 705delG frameshift mutation), green dots represent hiPSC line II (Rett patient with 1461A>G missense mutation), and orange and red circles indicate independently isolated hiPSC lines from the same patient. For calcium indicator and LFP data, plots were generated for all experiments in which significant differences between Mut and iCtrl Cx+GE fusions were reported. In all cases in which the same measure resulted in statistically significant differences between Mut and iCtrl in both hiPSC patient lines, the two patient lines were combined for within genotype statistical analyses (e.g., proportion of multispiking neurons). d, Table with mean, standard deviation (Std Dev), within genotype P value, and between genotype P value for all measures shown in a-c. The results show relatively low Std Dev within genotypes as reflected in non-significant P values, yet highly significant differences between the iCtrl and Mut groups in nearly all functional measurements. All between batch statistical analyses were by ANOVA. All between genotype analyses by ANOVA with correction for multiple comparisons by Tukey’s test, unless otherwise specified in the main text.

Extended Data Fig. 2. Constrained non-negative matrix factorization (CNMF) based Ca²⁺ data extraction workflow and output

a, Raw image of an GCaMP6f infected Cx+GE organoid (left) and CNMF based identification of fluorescently active (spiking) GCaMP regions of interest (right). b-d, Identification and analysis of individual neuronal Ca²⁺ spiking data. b, Changes in GCaMP6f fluorescence (normalized ΔF/F) for each neuron in a displayed as individual spike trains (left) or the same data displayed as a colorized amplitude plot (right). Individual spiking data are then used to determine various measures of spiking behavior including overall synchronicity based on a threshold level determined following spike shuffling c and calculation of interspike intervals d. e, Simultaneous to b-d, Ca²⁺ spiking data are categorized into neuronal microcircuits (clusters) based on correlations between individual Ca²⁺ spikes. f, during initial analyses, alternative clustering approaches including cross-correlation was used and the neural microcircuits resulting from multiple approaches were compared to determine the optimal clustering paradigm.

Extended Data Fig. 3. Immunohistochemical analyses reveal similar cell composition in iCtrl and Mut fusion organoids.

a, Day ~100 iCtrl and Mut Cx+GE fusion organoids have comparable numbers of GAD65⁺ positive cells in both the GE and Cx end (quantification in Fig. 3c). b, Both unfused Mut and unfused iCtrl day ~100 GE organoids contain multiple interneuron subtypes including CALRETININ, CALBINDIN, and SOMATOSTATIN (SST) expressing cells. c, Mut and iCtrl Day ~100 GE and Cx organoids also contain GFAP⁺ astrocytes. All images are representative examples from 3 or more independently imaged sections. See Supplementary Table 4 for additional details.

Extended Data Fig. 4. Rett syndrome fusion organoids from a second patient hiPSC line exhibit neural network irregularities in calcium indicator measurements.

a, Immunohistochemical analyses of isogenic Cx and GE organoids from a second Rett syndrome patient hiPSC line (harboring a 1461A>G missense mutation, indicated by “II”) reveals either the presence (iCtrl-II) or absence (Mut-II) of MECP2 expression. Representative images from n=2 independent experiments and 6 imaged sections. b, Mut-II Cx+GE fusions contain hyperexcitable neurons as indicated by the red boxed regions in the bottom ΔF/F colorized amplitude plot and spike plot. These plots show trains of repeatedly firing Ca²⁺ transients with short interspike intervals that are not present in iCtrl-II Cx+GE (top plots). c, There is no discernible change in synchronization of calcium transients between Mut and iCtrl as reflected in the clustergrams. d, The hyperexcitable phenotype in Mut-II Cx+GE fusions is reflected in the pooled data both by significant increases in multispiking neurons and decreases in mean and median interpeak intervals. Pooled data quantifications, n = 10 iCtrl-II and n =6 Mut-II fusion organoids, where each n is an independently generated organoid. Two-sided Mann-Whitney tests were used, *P = 0.0071 for the proportion of multispiking neurons, **P = 0.0047 for the mean interspike interval, **P = 0.0017 for the median interspike interval, ns = not significant. Plot in d displays the full distribution of individual data points with dotted lines indicating the median and quartile values.

Extended Data Fig. 5. Enrichment of autism and epilepsy risk genes in up/downregulated genes in MECP2 mutant and isogenic control organoids.

a, Overlap of differentially expressed genes in MECP2 mutant organoids (all cell groups) with SFARI autism spectrum disorder (ASD) gene categories 1-3 and DisGeNET epilepsy Gene-Disease Association list (CUI: C0014544). Overlaps between data are indicated by red and green shading and displayed as Venn diagrams in b. c, Two-sided Fisher’s Exact Test was used to determine if Up/Downregulated genes show enrichment for genes in SFARI and epilepsy gene lists. Odds ratio from the test are displayed along with Bonferroni-corrected P values. Up/Epilepsy: ***P = 0.0016, Down/Epilepsy: ****P =1.81×10⁻⁵, Up/ASD: ****P = 5.72×10⁻⁹, Down/ASD: P = 1.00.

Extended Data Fig. 6. UMAP representation of select genes associated with synaptogenesis and kainate responsivity.

a, UMAP representation of select genes associated with axonal projections and synaptogenesis found to be upregulated in MECP2 mutant Cx+GE fusion organoids. Violin plots display the relative expression level of each gene across the indicated cell clusters. b, UMAP representation of kainate receptor gene expression within the Cx+GE fusion organoids.

Extended Data Fig. 7. Gene ontology analysis of neuronal subtype clusters.

Top 10 most enriched Gene Ontology biological process (GO BP) terms associated with upregulated or downregulated differentially expressed genes when comparing Mut and iCtrl within the main excitatory (CPN and CFuPN) and interneuron (IN) clusters. Upregulated genes in the excitatory clusters are highly enriched for terms associated with synaptogenesis and axonal morphogenesis while downregulated genes are associated with mRNA catabolism and translation. In contrast, synaptogenesis terms are absent among the upregulated genes in the IN cluster, with this set populated by terms associated with forebrain differentiation and axonal morphogenesis. Downregulated genes in the IN cluster are enriched for metabolism and cellular cytoskeleton associated terms.

Extended Data Fig. 8. Spatially restricted microcircuit clusters and fewer synchronous events in MECP2 Mut Cx+GE organoids.

a, Pooled data for neuronal clusters derived here using Ca²⁺ activity correlations, reveal spatially restricted (smaller) microcircuit clusters with fewer average neurons per cluster in Mut compared to iCtrl. b, Pooled data of synchronous events demonstrates significantly fewer events (but with each event having a significantly higher amplitude, see Fig. 3) in Mut compared to iCtrl. Synchronous events have similar overall duration in both conditions (n = 6 for iCtrl, n = 7 for Mut and represents independently generated organoids, *P = 0.0436 for Pairwise Distances, *P = 0.0203 for Cluster Circumference, *P = 0.0321 for Cluster Area, **P = 0.0089 for Neurons per Cluster, and *P = 0.0180 for Number of Synchronized Transients). Plots display the full distribution of individual data points with dotted lines to indicate the median and quartile values. Following a normality test, statistical significance was determined using a two-sided Mann-Whitney U-test.

Extended Data Fig. 9. Additional independent examples of local field potential recordings.

a,d, Representative raw 10-minute LFP traces (top) and time expanded segments (bottom) from either unmixed iCtrl or Mut Cx+GE fusion organoids, or Mut Cx+iCtrl GE or iCtrl Cx+Mut GE mixed fusion organoids. b,e, Morlet plots derived from the time expanded segments shown in a,d. c, f, Periodogram derived from the entire 10 min traces shown in a,d.

Extended Data Fig. 10. Rett syndrome fusion organoids from a second patient hiPSC line demonstrate epileptiform changes in extracellular recordings.

a, Raw trace of a representative 10-minute LFP recording (top) and time expanded window (bottom) from iCtrl-II, Mut-II, or Mut-II +PFT-α Cx+GE fusion organoids. b, Morlet plots showing high frequency activity associated with the time expanded segments shown in (a). (c) Periodograms derived from the entire recordings shown in a. d, Quantification of high and low gamma spectral power from LFP recordings demonstrates a significant decrease of low gamma power and a sizeable but non-significant loss of high gamma power in Mut-II Cx+GE fusions. PFT-α treatment of Mut-II Cx+GE fusions results in a statistically significant rescue of both low and high gamma oscillatory power. Low gamma; Ordinary ANOVA, overall P = 0.0024, Tukey’s Multiple comparisons, *P =0.0313 iCtrl II vs Mut II, *P=0.0211 Mut II vs Mut II +PFT. High gamma; Ordinary ANOVA, overall P = 0.0091, Tukey’s multiple comparisons, *P = 0.0243 Mut II vs Mut II + PFT, P = 0.09 between iCtrl-II and Mut. e, Spike frequency across multiple independent experiments Kruskal-Wallis test, overall P = 0.0003, Dunn’s multiple comparisons **P = 0.0028, *P = 0.0276. For d and e, n = 5 for iCtrl-II and Mut-II +PFT-α, n = 6 for Mut-II (total n = 16). f, Plots of high and low gamma spectral power versus spike frequency demonstrates an inverse relationship between gamma power and spiking. The solid black line is the best fit following linear regression, and the dashed magenta lines indicate 95% confidence intervals for the estimated line of best fit. The slope of the line of best fit is indicated above each graph. Plots in d and e display the full distribution of individual data points with dotted lines to indicate the median and quartile values.

Fig. 1 |. Generation and characterization of fusion brain organoids.

a, Schematic outlining the generation, patterning, and fusion of dorsal cortex (Cx) and ventral ganglionic eminence (GE) organoids. b, Immunohistochemical analysis of H9 hESC or non-mutant hiPSC-derived Cx and GE organoids prior to fusion at the indicated days (D) of differentiation in vitro. c, Prior to fusion, D56 Cx or GE organoids were infected with AAV1 CAG:tdTomato virus, allowing for tracking of cells emanating from each compartment. Two weeks after fusion, labeled Cx cells showed limited migration into adjacent Cx or GE structures (left and middle images) while labeled GE progenitors display robust migration and colonization of their Cx partner (right image). d, Immunohistochemical analysis showing intermingling of SATB2⁺ cortical neurons with DLX5⁺ GAD65⁺ inhibitory interneurons in the cortical compartment of D106 Cx+GE but not Cx+Cx fusion organoids. e, At day 84, Cx+Cx fusions (left panels) contain numerous excitatory synapses reflected by prominent colocalization of the pre- and post-synaptic markers VGLUT1 and PSD95, yet sparse numbers of inhibitory synapses detected by VGAT/GEPHYRIN costaining. Cx+GE fusions (right panels) on the other hand contain numerous VGLUT1⁺/PSD95⁺ excitatory and VGAT⁺/GEPHYRIN⁺ inhibitory synapses (right panels). All images in b-e are representative images from multiple experiments and represent one of at least 3 or more imaged sections. For specific details refer to Supplementary Table 4.

Fig. 2 |. Cx+GE fusion organoids demonstrate complex neural network activities including oscillatory rhythms.

a, Schematic illustrating the identification of active neurons by virtue of their Ca²⁺ transients (I), representation of their network organization (II), and methods used to collect extracellular recordings (III). b, Example of live 2-photon microscopy imaging of an H9 hESC derived fusion organoid demonstrating acquisition of regions of interest (red circles) and the resulting activity profile shown as normalized change in fluorescence (ΔF/F), where each line is an individual neuron (middle plot) and representation of the same data as a colorized amplitude plot (right). c, Addition of 100 μM bicuculline methiodide (BMI) has a minimal effect on Cx+Cx fusions (top) yet elicits spontaneous synchronization of neural activities in Cx+GE organoids (bottom). d, These synchronizations can be transformed into a normalized amplitude versus time plot for quantitative analyses (left) and further visualized as a clustergram following hierarchical clustering of calcium spiking data (right). e, Pooled data of the amplitude measurements. Plots display the full distribution of individual data points with dotted lines indicating the median and quartile values. n = 3 independent experiments for Cx+Cx and Cx+GE. ANOVA P = 0.0011, F = 8.301, DF (between columns) = 3 followed by Tukey’s multiple comparison; **P = 0.0028 for Cx+Cx vs Cx+GE BMI; **P = 0.0100 for Cx+Cx BMI vs Cx+GE BMI; **P = 0.0031 for Cx+GE vs Cx+GE BMI. f-h, Local field potentials measured from a representative Cx+GE fusion reveal robust oscillatory activities at multiple frequencies during a 5-minute period, reflected in both raw traces f and spectrogram g. Spectral density analysis in h demonstrates the presence of multiple distinct oscillatory peaks ranging from ~1-100 Hz. i-k, Cx+Cx fusion organoids by contrast lack measurable oscillatory activities. Representative traces in (f-h) are taken from 3 independent experiments and in (i-k) from 4 independent experiments.

Fig. 3 |. Rett syndrome fused and unfused organoids have similar cortical organization and cell type expression profiles.

a,b, Generation and immunohistochemical analyses of isogenic Cx and GE organoids from Rett syndrome patient hiPSC that either contain (iCtrl) or lack (Mut) MECP2 expression (see also). iCtrl and Mut Cx organoids exhibit comparable formation of neural progenitors (SOX2, TBR2), both deep and superficial layer neurons (CTIP2, BRN2), and inhibitory interneurons (GAD65, SST, and GABA) All images reflect representative images from at least 3 independently imaged sections, refer to supplementary table 4 for further details. c,d, D100 unfused iCtrl and Mut Cx organoids show minimal expression of GAD65, whereas ~20-25% of the cells in the Cx end of age matched Cx+GE organoids express GAD65, n = 3 independently generated organoids, 2631 cells, ns, not significant by a two-sided t-test. e,f, Immunohistochemical analysis of interneuron subtypes by the co-expression of GAD65 with SST, Calretinin, NPY or Calbindin in the Cx portion of day 100 iCtrl or Mut Cx+GE fusion organoids reveals the presence of multiple interneuron subtypes e. f, Cell counting reveals trends for all comparisons and a statistically significant difference between iCtrl and Mut samples with respect to the percentage of cells expressing calretinin, n = 3 fusion organoids per genotype, ≥ 980 cells for each sample counted, ns = not significant for all groups except calretinin. ANOVA P = 0.0003, F = 4.665, DF (between columns) = 7, followed by Sidak’s correction for multiple comparison between iCtrl and Mut for each marker; *P = 0.0391 for iCtrl vs Mut calretinin; ns for all other comparisons. Plots in d,f display the full distribution of individual data points with dotted lines indicating the median and quartile values.

Fig. 4 |. Single-cell transcriptomic analysis reveals the presence of diverse cellular populations in fusion organoids with a trend towards accelerated maturation and alterations in interneuron formation in MECP2 mutant samples.

a, Uniform Manifold Approximation and Projection (UMAP) of combined iCtrl and Mut Cx and GE organoids. The plot includes cells from 3 D56 Cx and 3 GE organoids collected before fusion, 3 D70 Cx+GE fusion organoids, and 3 D100 Cx+GE fusion organoids. The total number of cells sequenced were as follows: D56 iCtrl, 9306; D56 Mut, 9186; D70 iCtrl, 10931; D70 Mut, 6260; D100 iCtrl, 7561; and D100 Mut, 6698 cells. b, Plots display the mean percentage of cells in the fusion organoids representing each of the clusters in a. Separation of the data by iCtrl and Mut status shows a trend of reduced progenitors and more differentiated neurons in Mut organoids compared to iCtrl samples. c, UMAPs of key genes associated with each of the major clusters identified in a. d, Re-clustered UMAP of the interneuron subset from a with interneuron subtype markers identifying each re-clustered subset. e, Percentage of cells for each of the clusters in d segregated by iCtrl and Mut reveals increased numbers of interneurons expressing PVALB/SST and CALB1 in iCtrl organoids and cells expressing VIP and CALB2 in Mut samples. f, Heat map with the relative expression of canonical interneuron-related genes within the re-clustered groups.

Fig. 5 |. Gene ontology and synaptic staining analyses reveal defects in the balance of excitatory and inhibitory synapses in MECP2 mutant fusion organoids.

a, Top 10 most enriched Gene Ontology biological process (GO BP) terms associated with upregulated or downregulated differentially expressed genes when comparing Mut and iCtrl across all cells. b, Immunohistochemical analysis of excitatory (VGLUT1/PSD) and inhibitory (VGAT/GEPHYRIN) pre-/post-synaptic puncta reveals an increase in excitatory synapses in Mut Cx+GE fusion organoids. The yellow dotted boxes in the right most panels display representative TUBB3⁺ regions that were used for analyses. The adjacent two panels demonstrate the raw immunohistochemical image followed by Imaris software renderings of the colocalized pre- and post-synaptic markers. The final two panels are magnified versions of the boxed areas. c, Plots of the number of synapses per cell. Data were pooled from multiple organoids. VGLUT1/PSD95 (Excit), n = 3 organoids for both iCtrl Mut samples, 1180 cells; VGAT/GEPHYRIN (Inhibit), n = 4 organoids for iCtrl and Mut samples, 1654 cells. ANOVA P = 0.0002, F = 8.387, DF (between columns) = 3, followed by Tukey’s multiple comparison; **P = 0.0088 for Excit iCtrl vs Excit Mut; **P = 0.0014 for Excit Mut vs Inhib Mut; ns = not significant. Plots display the full distribution of individual data points with dotted lines indicating the median and quartile values.

Fig. 6 |. Rett syndrome fusion organoids display GE dependent hypersynchronous neural network activity.

a, Mut Cx+GE fusions exhibit spontaneous synchronized Ca²⁺ transients that are not seen in iCtrl Cx+GE, reflected in the raw ΔF/F colorized amplitude plot (top), synchronization amplitude plot (bottom), and clustergram b. c, Pooled data quantifications, n = 12 independently generated iCtrl and n = 7 independently generated Mut fusion organoids, ****P < 0.0001 for the average amplitude of synchronized transients; **P = 0.0012 for multi-spiking neurons, significance assessed by two-sided Mann-Whitney U. d, Mixed fusions with iCtrl Cx and Mut GE exhibit spontaneously synchronized calcium transients, whereas as mixed fusions with Mut Cx and iCtrl GE do not, as seen in the raw ΔF/F colorized amplitude plot (top), synchronization amplitude plot (middle), and clustergram e. f, Pooled data quantifications, n = 10 independently generated iCtrl Cx+Mut GE and n = 11 Mut Cx+iCtrl GE, *P =0.0308, DF = 19, t = 2.334, two tailed student’s t-test. Plots in c,f display the full distribution of individual data points with dotted lines to indicate the median and quartile values.

Fig. 7 |. Rett syndrome fusion organoids display GE-dependent epileptiform changes.

a, Raw trace of a representative 10-minute LFP recording (top) and time expanded window (bottom) from unmixed Mut or iCtrl Cx+GE fusion organoids and Mut/iCtrl mixed Cx+GE fusions. b,c, Spectrograms and periodograms derived from the entire recordings shown in a. d, Morlet plot showing high frequency activity associated with the time expanded segments shown in a. e, Frequency histogram of interspike intervals derived from the raw trace in a. f, Quantification of high and low gamma spectral power from LFP recordings demonstrates a significant decrease of gamma power in Mut Cx+GE fusions and mixed fusions with a Mut GE. High gamma; Ordinary ANOVA, overall P = 0.0020, F = 7.089, DF (between columns) = 3, Tukey’s multiple comparisons, **P = 0.0018, *P = 0.0353 iCtrl vs Mut and *P = 0.0345 Mut Cx + iCtrl GE vs iCtrl Cx + Mut GE. Low gamma; Ordinary ANOVA, overall P = 0.0174, F = 8.038, DF (between columns) = 3, Tukey’s multiple comparisons, ***P = 0.0009, *P = 0.0174 iCtrl vs Mut, *P = 0.0309 Mut Cx + iCtrl GE vs iCtrl Cx + Mut GE. g, Spike frequency across multiple independent experiments Kruskal-Wallis test, overall P = 0.0002, Dunn’s multiple comparisons ***P = 0.002, *P = 0.0159 iCtrl vs iCtrl GE+ Mut Cx and *P = 0.0416 Mut vs Mut Cx + iCtrl GE. n = 6 independently generated organoids for each condition (iCtrl and Mut) in f and g. Plots in f,g display the full distribution of individual data points with dotted lines to indicate the median and quartile values.

Fig. 8 |. Partial restoration of gamma oscillations and suppression of HFOs in Rett syndrome fusion organoids by administration of Pifithrin-α.

a, Raw trace (top), time expanded window (middle), and periodogram (bottom) from representative Mut Cx+GE fusion organoids treated for 48 h with vehicle (DMSO, Veh), 2 mM sodium valproate (VPA), or 10 μM Pifithrin-α (PFT). b, Morlet plot derived from the time expanded segment in a. c, Quantification of high gamma oscillations and spike frequency in Mut Cx+GE shows a highly significant rescue of both high gamma spectral power and a reduction in spike frequency following treatment with PFT and more modest, but significant, rescue in both measures following VPA treatment. High gamma quantification; Ordinary ANOVA, overall P = 0.0085, F = 8.476, DF (between columns) = 2, Tukey’s Multiple comparisons, **P = 0.0093, *P = 0.0299, n = 4 independently generated organoids for each of the 3 conditions (Veh, VPA, and PFT). Spike Frequency following drug addition; Kruskal-Wallis test, overall P = 0.0020, Dunn’s multiple comparisons **P = 0.0042, *P = 0.0420. Plot displays the full distribution of individual data points with dotted lines to indicate the median and quartile values. d, Plots of high and low gamma spectral power versus spike frequency demonstrates an inverse relationship between gamma power and spiking. The solid black line is the best fit following simple linear regression, and the dotted magenta lines indicate 95% confidence intervals for the estimated line of best fit. The slope of the line of best fit is indicated at the top of the graph. The calculated slope is significantly different from zero with P < 0.0001 for high gamma and P = 0.0007 for low gamma,