Human neural stem cells and neural progenitor cells have been tested in a variety of CNS disorders, including spinal cord injury (SCI) (Blaya et al., 2015; Han et al., 2013; Itakura et al., 2017; Kadoya et al., 2016; Kobayashi et al., 2012; Lepore and Fischer, 2005; Lu et al., 2012; Lu et al., 2014; Pieczonka and Fehlings, 2023; Tsuji et al., 2010; Tsuji et al., 2019; Wictorin et al., 1990). Potential donor cell types include primary spinal cord neural progenitor cells isolated from the developing spinal cord (Bonner et al., 2011; Fischer et al., 2020; Kadoya et al., 2016; Lu et al., 2012; Rosenzweig et al., 2018), embryonic stem cell-derived neural stem cells (McDonald et al., 2004), induced pluripotent stem cells driven to a neural fate (Inoue et al., 2023; Lee-Kubli and Lu, 2015; Lu et al., 2014; Tsuji et al., 2010), and primary adult cells directly differentiated to neural stem cells (Hou and Lu, 2016; Son et al., 2023). Each donor cell type has distinct advantages and limitations. We have pursued the generation of neural stem cells from pluripotent (embryonic) stem cells because they represent an unlimited source of cells from well characterized sources that can undergo efficient and bulk preparation and testing for clinical application. However, methods for generating sustainable cultures of neural stem cells driven to a specific spinal cord identity had been challenging to establish; while spinal cord neural stem cells had been successfully derived by Ashton and colleagues (Gouti et al., 2015; Lippmann et al., 2015), the cells spontaneously differentiated in cell culture at very low passage number and therefore did not represent a practical cell source for expansion in number and application for human clinical trials.

In 2018 we succeeded in developing methods for generating spinal cord neural stem cells from embryonic stem cells that could be sustained in vitro over multiple passage numbers (Kumamaru et al., 2018). Thus, an expandable source of cells was available for potential clinical translation. It was critical to generate neural stem cells specifically driven to a spinal cord stem cell fate because these cell types support optimal regeneration of host axons into the graft placed into the lesion site; for example, grafts of neural stem cells of forebrain or hindbrain identity supported more limited host axonal regeneration than spinal cord stem cells or progenitor cells (Kadoya et al., 2016; Kumamaru et al., 2018). Host axons regenerating into spinal cord stem cell grafts form synapses onto spinal cord neural stem cells, and the stem cells in turn extend very large numbers (tens of thousands in rats; hundreds of thousands in primates) of axons into the distal host spinal cord, where they form synapses onto host neurons below the injury (Bonner et al., 2011; Brock et al., 2018; Chambers et al., 2012; Lu et al., 2017; Lu et al., 2012; Rosenzweig et al., 2018). As a result, significant functional improvement occurs in both rodent and non-human primate studies (Bonner et al., 2011; Brock et al., 2018; Chambers et al., 2012; Kadoya et al., 2016; Lu et al., 2017; Lu et al., 2012; Rosenzweig et al., 2018; Son et al., 2023).

Ongoing study of the spinal cord neural stem cells that we developed in 2018 revealed that they occasionally exhibited karyotypic abnormalities, and that some of these abnormalities were clonal in nature. Accordingly, in the present study we sought to develop modified methods for generating spinal cord neural stem cells based on the hypothesis that more modest concentrations of cell growth-inducing and differentiation agents would result in reduced rates of cell division, minimizing the risk that errors would occur during cell division that resulted in chromosomal aberrations. Accordingly, we tested a range of concentrations of fibroblast growth factors 2 and 8 that drive cell division, and a range of concentrations of two small molecules CHIR 99041 (a Gsk3ß inhibitor) and SB 431542 (an Alk4, 5, 7 inhibitor) that are used to induce neural differentiation. We now report an optimized method for generating cells that results in fewer chromosomal aberrations while continuing to exhibit favorable propagation properties in vitro and engraftment properties in vivo.