Neural stem cells and their contribution to neurogenesis in the adult mammalian brain

Abstract

For a long time it was believed that neurogenesis in the mammalian central nervous system was restricted to the embryonic and early postnatal period. Almost four decades ago Altman and colleagues challenged this notion, but it took more than thirty years before new studies with refined methods convincingly demonstrated neurogenesis in the adult mammalian brain, including the adult monkey and human brain. Another major discovery was the demonstration by Reynolds and Weiss (1992) that neural stem cells persist in the mature brain that could be grown in culture and had the hallmark properties of stem cells: multipotency and self-renewal capacity. An obvious question following the discovery of stem cells in the adult brain was the localization and identity of these cells. We tested the hypothesis if ependymal cells, which delineate the ventricular system, might be a neural stem cell population in the adult brain.

Ependymal cells were labeled in vivo, using either the fluoroescent label DiI or an adenovirus expressing the reporter gene lacZ. Immediately following an intraventricular injection, labeling was restricted to ependymal cells. At later timepoints labeled cells could be observed to enter the migratory stream to the olfactory bulb. There they differentiated into neurons as demonstrated by double labeling with neuronal markers. Based on fluorescent labeling and on their specific morphology single ependymal cells could be isolated, cultured and induced to differentiate into the major cell types of the CNS. Exploiting the fact that ependymal cells express the notch- I receptor on their surface, neural stem cells could be enriched by magnetic sorting. Long term labeling of cells in the subventricular zone and the spinal cord with BrdU revealed that neural stem cells are slowly dividing. After a traumatic injury to the spinal cord however, proliferation of ependymal cells increased. Newborn cells migrated from the central canal to the site of injury where they contributed to the formation of the glial scar.

A very slow rate of proliferation can be indicative for a stem cell in certain tissues and is widely considered as a primary step leading towards their identification. We devised a method that combined postembedding, detection and ultrastructural characterization of immunogold labeled cells, thus allowing for the relatively rapid screening of rarely dividing cells in the adult central nervous system. This technique was applied to identify the ultrastructure of slowly proliferating putative stern cells in the adult mouse spinal cord.

A specific supplement added to the culture medium was shown to have a selective effect on the propagation of distinct neural stem cell populations, This was in contrast to hypoxic culture conditions which were shown not to have this specific effect on neural stem cell propagation. There were no apparent differences between the distinct neural stem cell populations in their morphology, capacity for self-renewal, or ability to differentiate into glia and neurons. Intraventricular infusion of EphB2 receptor leads to a cellular rearrangement in the subventricular zone with astrocytes contacting the ventricular lumen. After infusing EphB2 receptors or vehicle solution, an upregulation of GFAP expression in ependymal cells was seen in both cases. We conclude that this is an injury related reaction rather than a specific effect of EphB2.

The adult mammalian hippocampus and olfactory bulb are structures with an extensive, continuous generation of interneurons derived from stem cells. We asked whether there may also be a turnover of neurons in other regions of the adult brain, and focused on the substantia nigra pars compacta in the midbrain, where the dopamine- producing neurons that are lost in Parkinson's disease reside. We found that despite ongoing neuronal cell death in the substantia nigra, total cell number remained constant over a large part of the life span of the adult mouse. This indicated that there must be the generation of new neurons to compensate for cell loss. After long term labeling with either BrdU or DiI, we found TH-positive neurons with BrdU positive nuclei or DiI labeled membranes were found. We obtained similar results after labeling with tritiated thymidine followed by autoradiography. As the most likely origin for newborn dopaminergic neurons, we identified ependymal cells lining the third ventricular recess and the cerebral aqueduct by labeling with BrdU or DiI. Newborn neurons were found to send projections to their target area and integrate into the local synaptic circuitry as assessed by retrograde tracing and pseudorabies virus labeling studies. Similar to other injury studies, we could show that the rate of neurogenesis is increased after selectively ablating dopaminergic neurons with the toxin MPTP.