Neuroscience (SfN) 2026

Human 3D midbrain organoids hold promise as physiologically relevant in vitro models to capture selected features of developing human midbrain tissue, including dopaminergic identity, tissue-like organization, and emerging circuit activity. These properties make them well suited to investigate Parkinson’s disease (PD)- and synucleinopathy-relevant alterations in neuronal excitability, network synchronization, and axonal output, which may precede overt neurodegeneration.

However, resolving these phenotypes in 3D tissues requires non-invasive electrophysiological methods that combine longitudinal access, high spatial sampling, and reproducible acquisition across wells and timepoints to detect low amplitude signals, quantify network bursting, and track axonal propagation at subcellular resolution. Here, in collaboration with STEMCELL Technologies, we used high-density microelectrode arrays (HD-MEAs) to longitudinally profile the functional maturation of commercially available human midbrain organoids. Organoids were plated at 77 days of age onto PEDOT-coated MaxTwo HD-MEA 24-well plates, each containing 26,400 electrodes per well, and spontaneous activity was tracked for approximately one month. Organoids showed progressive strengthening of firing activity, including increased firing rate, spike amplitude, and active area, indicating neuronal maturation and improved tissue-electrode coupling. At the network scale, we tracked the emergence of increasingly organized synchronized bursting, with increased burst cadence and enhanced neuronal recruitment over time. To assess pharmacological responsiveness, organoids were challenged with 4-AP and TTX, which respectively enhanced and suppressed firing activity in a highly reproducible manner. Finally, AxonTracking enabled label-free extraction of soma-to-axon propagation patterns at the organoid-electrode interface, providing a complementary readout of intra-organoid axonal functionality. Together, these results establish a scalable, non-invasive multiscale electrophysiology workflow for human midbrain organoids that captures spontaneous network maturation, pharmacological modulation of excitability, and axonal propagation phenotypes. By supporting consistent acquisition, processing, and interpretation of functional readouts across wells and timepoints, this approach provides a foundation for Parkinson’s disease-relevant phenotyping in 3D human neural models using HD-MEAs.

‍

Heterozygous loss-of-function (LOF) mutations in the granulin precursor gene (GRN) are a major genetic cause of frontotemporal dementia (FTD), leading to GRN haploinsufficiency. Reduced neuronal progranulin impairs lysosomal function and proteostasis, promoting synaptic dysfunction and TDP-43 pathology, contributing to altered functional connectivity and selective neuronal vulnerability. Human iPSC-derived models of GRN haploinsufficiency are valuable tools for disease phenotyping and therapeutic discovery; however, translating these models into disease-relevant functional endpoints remains challenging.

Using the MaxTwo High-Density Microelectrode Array (HD-MEA) multiwell platform (MaxWell Biosystems, 26,400 electrodes per well), which enables non-invasive, label-free, real-time electrophysiological recordings, we longitudinally profiled functional phenotypes in a human iPSC-derived FTD model across network, cellular, and axonal scales.  

iCell Induced Excitatory Neurons were co-cultured with normal iCell Astrocytes 2.0 and compared between an Apparently Healthy Normal (AHN) control line and an isogenic line carrying a heterozygous GRN R493X LOF mutation. Spontaneous activity was monitored from DIV 6-50. MaxWell’s ActivityScan and Network assays detected reproducible disease-associated changes across replicates and timepoints, with significantly elevated firing rates and reduced inter-spike intervals emerging by DIV 15. These early abnormalities indicate that HD-MEAs capture disease-related dysfunction before overt structural degeneration. GRN mutant cultures also showed significantly altered network synchrony, including more frequent and robust network bursts, shorter inter-burst intervals, and prolonged bursting, supporting a hyperexcitable network phenotype.  

MaxWell’s AxonTracking Assay reconstructed action potential propagation along individual axonal arbors. GRN mutant neurons displayed reduced axonal branching and shorter neurites, extending the disease-associated phenotype from early network and single-cell alterations to axonal architecture.  

Together, these data identify an early, reproducible, multiscale, and multiparametric electrophysiological signature of GRN haploinsufficiency in a human iPSC-derived FTD neuronal model, longitudinally captured on a single HD-MEA platform with scalable assay workflows. This approach links network-level dysfunction with single-cell and axonal phenotypes, providing robust functional endpoints for therapeutic window definition, progression monitoring, and future therapy-response studies in human neurodegeneration models.