During development, myelinating oligodendrocytes develop from migratory, proliferative oligodendrocyte precursors, also known as NG2 cells. These cells can be recognized because they express the PDGF receptor (PDGFRa) and the NG2 proteoglycan (hence their name). Large numbers of NG2 cells persist past the developmental period into adulthood, when they comprise ~5% of all cells in the CNS.

Recently we and other labs have shown, by Cre-lox fate mapping in transgenic mice, that NG2 cells continue to divide and generate significant numbers of myelinating oligodendrocytes thoughout adult life. It is not known whether these adult-born oligodendrocytes replace oligodendrocytes that die through normal wear-and-tear, or whether the new cells add to the existing population. However, it is clear that they can replace oligodendrocytes that are lost during demyelinating diseases or spinal cord injury. Whether they can form other cell types like astrocytes or neurons is controversial.

It has been shown that NG2 cells form synapse-like structures with unmyelinated axons and receive synaptic input from passing action potentials. One possibility is that they are "listening in" to electrical activity, which at some threshold might stimulate proliferation and/or differentiation of the NG2 cells, resulting in de novo myelination of the active axons. This intriguing idea is currently under investigation by several labs worldwide. The implications of such a mechanism are far-reaching: could adult-born oligodendrocytes, like adult-born neurons, contribute in some way to neural plasticity? Magnetic resonance imaging has shown that white matter volume increases during the first 4 decades of human life and decreases thereafter, in line with general cognitive and motor decline. Complex motor tasks like piano playing or juggling have also been linked to changes in the volume or structure of specific white matter tracts. Is it possible that adult myelin genesis is required for motor skills learning and procedural memory?