Assistant Professor Tufts University Medford, Massachusetts, United States
Introduction: Region-specific heterogeneity in the spinal cord is essential for shaping complex behaviors. Though neuronal diversity is commonly appreciated, astrocytes also display distinct transcriptional differences across the rostrocaudal (head to tail) and dorsoventral (back to front) axes that contribute to local neural circuits. However, there remains a significant gap in our ability to rapidly generate and replicate this regional diversity in vitro, especially for astrocytes corresponding to the spinal cord versus the brain. Our goal is to develop a strategy using human pluripotent stem cells (hPSCs) to generate spinal astrocyte diversity that can be used in diverse tissue engineered microenvironments.
Materials and
Methods: We have developed methods to rapidly produce region-specific astrocytes from hPSC-derived, region-restricted spinal progenitors. First, cervical or thoracic spinal progenitors were differentiated using methods we previously established in Iyer et al. (2022) using FGF8, CHIR, RA, and SMAD inhibitors. Next, ventral or dorsal progenitors were patterned using Shh agonists or BMP4, respectively. Then, these progenitors with multi-lineage potential (neurons, oligodendrocytes, astrocytes) were expanded and selected for astrocytic fates using EGF and LIF. Finally, the glial progenitors were cultured in CNTF for up to 14 days to generate mature astrocytes. Calcium imaging and neuronal co-culture were implemented to assess electrophysiological function and differential growth capacities.
Results, Conclusions, and Discussions: Preliminary methods resulted in robust expression of astrocyte markers, SOX9 and GFAP, and characteristic star-shaped morphology in all four groups within 55 days (Figure 1). This is significantly shorter than comparable 100+ day protocols for dorsal and ventral spinal astrocytes, which also require 3D suspension culture. Notably, we qualitatively observed differential morphology between dorsal and ventral astrocytes that suggest phenotypic heterogeneity. Similar to established literature, we expect that mature astrocytes (100+ days) will show characteristic calcium intake kinetics and differentially support neurite extension of dorsal versus ventral neurons. As we continue to optimize our protocols to improve astrocyte yield, future work will explore whether matching the soft in vitro microenvironment will further facilitate astrocyte differentiation efficiency and shorten the time to express mature astrocyte markers. Ultimately, we hypothesize that combining region-specific differentiation with region-matched tissue mechanics will improve the accuracy of in vitro models to study of region-specific differences in neural injury, disorder, and repair.
