Specification and potency of human neural stem cells for clinical transplantation

Abstract

Neural stem cells hold promise for future treatment of spinal cord injury. Various aspects regarding cell fate specification, manufacturing and monitoring, with implications for clinical applications of these cells, are discussed herein. Neural stem cells can be obtained from a number of sources, including fetal tissue and pluripotent embryonic stem cells. Transplantation of cells derived from immature sources is associated with tumor risk, which needs to be thoroughly investigated.

We have shown that assessment of pluripotency should preferably be performed in the intended target compartment, as commonly used teratoma tests failed to detect pluripotent cells remaining after neural induction, cells that after transplantation gave rise to tumors in the central nervous system of model animals. We also found that among nonpluripotent neural stem/progenitor cells (NPCs) from human fetal tissue, occasional cells expressed mRNA for genes associated with the pluripotent phenotype. The phenotype of these NPCs showed no overt differences from the others, but live-cell imaging showed that all NPCs constantly changed their morphology. Surprisingly, mRNA for pluripotency genes was not restricted to a certain subpopulation of cells. Rather, transcripts of these genes transiently appeared in most cells, but during short periods. In a similar way, we found that expression of markers such as PSA-NCAM and A2B5, associated with more differentiated phenotypes, entailed a propensity for differentiation, but not fate restriction. Isolated cell populations with either high or low immunoreactivity for CD133, CD15, CD24, CD29, PSA-NCAM or A2B5 both reconstructed the parental distribution of immunoreactivity after about two weeks in culture.

Transcriptome analysis and in vitro studies confirmed that the reversible expression of markers was a reflection of reversible phenotypic identity. This finding requires that phenotype interconversion is added to the hierarchical model of neural fate determination in vitro. Furthermore, we have developed and evaluated a device for automatized mechanical dissociation of cell aggregates in culture, in compliance with regulatory guidelines for production of cells for transplantation, and shown its usefulness in long-term NPC cultures.