Q12 - A Three-Dimensional iPSC-derived Midbrain Neuron Culture for Modeling Opioid Addiction

Introduction: : Understanding the complex interplay of cell-cell and cell-matrix interactions is essential for modeling development, disease, and therapeutic responses in the central nervous system (CNS). Traditional two-dimensional (2D) culture systems often fail to recapitulate the structural and functional characteristics of native neural tissue, limiting their utility in neuroscience and drug discovery. To address this, we developed a three-dimensional (3D) hydrogel platform based on hyaluronic acid (HA), a key component of the brain extracellular matrix known for its roles in neural signaling and tissue architecture. Using this biomimetic scaffold, we differentiated human induced pluripotent stem cells (iPSCs) into midbrain-fated neural populations, generating both dopaminergic (TH-positive) and GABAergic (GAD1-positive) neurons. The 3D cultures exhibited enhanced neuronal maturation and network formation compared to 2D conditions, as evidenced by single-nuclei RNAsequencing, immunostaining, and electrophysiological recordings. Importantly, this platform supported dynamic neurochemical responses to pharmacological stimuli, including robust dopamine release following potassium depolarization and chronic morphine treatment. The observed dopamine dynamics—characterized by sensitization, tolerance, and relapse upon morphine re-exposure—mirror in vivo patterns of opioid neuroadaptation. These findings highlight the potential of this 3D iPSC-derived midbrain model as a physiologically relevant tool for studying addiction mechanisms and for advancing neural tissue engineering strategies.

Materials and

Methods: : Human iPSC lines were differentiated to midbrain-fated neural cultures based on previously published methods (1). Cells were encapsulated in 3D hydrogels as cell aggregates (~120 µm diameter) and cultured in maturation media for an additional 28 days before assessment. Hydrogels were crosslinked from thiolated HA (~750 kDa, 5-6% thiolation (2)), 4-arm PEG-SH, 8-arm PEG Norbornene, and RGD-SH using lithium phenyl(2,4,6-trimethylbenzoyl) phosphinate at UV exposure (365 nm, 15 s). Immunostaining was performed on whole gel and crysections (14um) with TH, GAD67 and Tuj-1 antibodies. Single-nuclei RNAseq was performed using 10x genomics. Nuclei were dissociated using the gentleMACS machine and the umbilical cord kit. snRNAseq analysis was performed using the Seurat Package in R. Electrophysiological data in 2D was recorded using the Maestro Edge (Svendsen Lab) and in 3D was recorded using Innovative Neurophysiology Micro-Array Electrode and Ripple Neuro (Santacruz Lab).

Results, Conclusions, and Discussions:: 3D iPSC-derived midbrain neuron cultures exhibited robust differentiation into key neuronal subtypes: immunostaining revealed TH-positive dopaminergic neurons alongside GAD1 (GABAergic) neurons, mimicking native midbrain composition. Importantly, neurons in the 3D hydrogel extended robust neurite networks and formed synaptic connections; functional recordings confirmed that 3D cultures, unlike 2D, had much higher action potential amplitudes and frequencies, indicating a more mature neural circuit. Single-nuclei RNAseq analysis showed that 3D cultures upregulate genes associated with neuronal maturation and synaptic function relative to 2D: gene ontology analysis found enrichment of neural-development processes in 3D. Consistently, electrophysiological profiling revealed ~2-fold higher spike amplitudes in 3D (~60 μV) versus 2D (~30 μV), reflecting greater functional maturity and network excitability. When we applied a chronic morphine regimen, the 3D cultures recapitulated classic in vivo addiction dynamics: an initial sensitization (surge in dopamine release), followed by tolerance (progressive decline of dopamine signal) and a rebound (relapse-like overshoot) upon re-exposure (data not shown). These patterns align with known opioid neuroadaptations, where progressive tolerance such that more opioid is needed to release the same amount of DA and long-lasting dopamine deficits after withdrawal. Overall, the data demonstrate that 3D iPSC-derived midbrain models achieve far more mature, in vivo–like phenotypes than 2D cultures, with mixed neuron subtypes, synaptic networks, and physiologic drug responses, underscoring their value as bioengineering platforms for addiction biology and neural interface applications.

Acknowledgements and/or References (Optional):: 1. Laperle A. H. Nat Med. 2020; 26: 289–299. 2. Ehsanipour. A. Cel. Mol. Bioeng. 2019; 12: 399–413. 3. Patriarchi. T. Science. 2018. 360(6396): eaat4422.