Poster V8 - Hybrid Differentiation Strategy to Efficiently Derive Region Specific Spinal Neurons

Introduction: Diverse spinal cord neurons along the body axis coordinate sensory processing, proprioception, and motor control, but human pluripotent stem cell (hPSC) differentiation methods often overlook the critical regional specificity of these cells. We have previously developed a direct differentiation method to efficiently generate any spinal neuron along the rostrocaudal (R/C) axis; however, rapid and consistent replication is limited by protocol duration, user expertise, and cell line-specific optimizations. Here we apply a dual-strategy combining direct differentiation and direct reprogramming with a TET-on NGN2 system, to efficiently generate region-specific spinal neurons. We expect that this hybrid differentiation strategy will provide an efficient and easily replicated way to obtain cells for both in vitro models and future transplantation applications.

Materials and

Methods: Kolf 2.1J hiPSCs were transfected with a TET-on NGN2 cassette using the PiggyBac transposon-transposase system and selected with puromycin to generate a polyclonal hPSC pool. Transduced and un-transfected controls were differentiated by adapting established methods for cortical progenitors (Lippmann et. al. 2014) or neuromesodermal progenitors (NMPs) (Iyer et. al. 2022) with or without addition of doxycycline (DOX). To generate spinal vs. peripheral fates, RA, SB, and/or LDN were added during DOX induction. Neuronal vs. progenitor differentiation and region-specific phenotypes were characterized using immunocytochemistry (ICC).

Results, Conclusions, and Discussions: Induction of NGN2 by DOX in minimal media conditions efficiently produced >95% TUJ1+ neurons within 5 days (Fig. 1B) compared to non-transfected controls which remained PAX6+ neural progenitors. We found that the robust proliferation of Kolf 2.1J hiPSCs necessitated optimization of region-specific spinal differentiation. By reducing initial seeding densities and increasing splitting ratios, we successfully generated SOX2+/T+ NMPs expressing CDX2 (Fig. 1C), indicative of region-specific gene expression. Cultures treated with DOX showed traces of both spinal (PAX6) and neural crest (SOX10) lineages but were predominantly TUJ1+ neurons (Fig. 1D). This indicates that R/C patterning drives rapid conversion of NMPs to neurons from posterior nervous system fates. Moreover, this hybrid strategy generated a mix of peripheral sensory (ISL1+/BRN3A+/ PRPH+) or spinal dorsal horn (BRN3A+/PRPH- or ISL1+/PRPH-) neurons (Fig. 1E). The addition of spinal vs. peripheral and dorsoventral patterning factors during neuronal induction will help further specify lineage and neuron type.

Direct reprogramming using NGN2 in hiPSC-derived NMPs provides an efficient way to make neurons from the posterior nervous system within 5 days of DOX induction. As we continue to further optimize neuronal purity and implement dorsoventral patterning, we also anticipate using NMPs patterned to different regions along the body axis to demonstrate the potential to efficiently convert neurons expressing region-specific R/C transcription factors. Altogether, this method maintains phenotypic diversity while enhancing differentiation efficiency and reducing media costs. We believe this method will be particularly beneficial for differentiating patient derived hiPSCs for personalized drug screens, disease modelling, and cell transplantation therapies, applications that currently lack a source for region specific spinal populations.

Acknowledgements (Optional):