Back

One-step induction of human GABAergic neurons promotes presynaptic development & synapse maturation

van Voorst, T. W.; van boven, M. A.; Marinus, K. I.; Colon-Mercado, J. M.; Schretzmeir, J.; Haag, C.; Toonen, R. F.; Koopmans, F.; Ward, M.; Smit, A. B.; van Kesteren, R. E.; Verhage, M.; Cornelisse, L. N.

2025-07-04 neuroscience
10.1101/2025.06.30.662293 bioRxiv
Show abstract

Human induced pluripotent stem cells (iPSCs) present a powerful approach to study human brain physiology and disease, yet robust, pure GABAergic induction has remained difficult. Here we present improved, single-step, transposon-based GABAergic induction with Ascl1/Dlx2, which yields, unlike lentiviral approaches, exclusively GABAergic neurons and was validated across three independent iPSC lines. Co-seeding with Ngn2-induced excitatory neurons created stable networks of predefined excitation/inhibition ratios, with corresponding synapse ratios. Proteomic and electrophysiological characterization at different developmental time points showed that the single-step induced GABAergic neurons gain a proteomic profile that maps to different cortical interneuron subtypes and display typical GABAergic synaptic properties, producing large, synchronous and picrotoxin-sensitive currents. During early development, synaptic strength increased threefold, which was accompanied by an increase in expression of proteins exclusively enriched for presynaptic SYNGO terms. Synaptic strength continued to increase during late development but with only minor proteomic changes. Taken together, transposon-based GABAergic induction yields exclusively mature GABAergic neurons suitable for studying genes involved in synaptic maturation and to build excitation/inhibition networks for disease modelling.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.