Axonal pathology is a major cause of neurological disability in multiple sclerosis. Axonal transection begins at disease onset but remains clinically silent because of compensatory brain mechanisms. Noninvasive surrogate markers for axonal injury are therefore essential to monitor cumulative disease burden in vivo. The neuronal compound N-acetylaspartate, as measured by magnetic resonance spectroscopy, is currently the best and most specific noninvasive marker of axonal pathology in multiple sclerosis. The possibility has been raised, however, that N-acetylaspartate is expressed also by oligodendroglial lineage cells. In order to investigate N-acetylaspartate specificity for white matter axons, transected rat optic nerves were analyzed by high-performance liquid chromatography and immunohistochemistry. In transected adult nerves, N-acetylaspartate and N-acetyl aspartylglutamate decreased in concordance with axonal degeneration and were undetectable 24 days posttransection. Nonproliferating oligodendrocyte progenitor cells, oligodendrocytes, and myelin were abundant in these axon-free nerves. At 24 days posttransection, N-acetylaspartate was increased (42%; p = 0.02) in nontransected contralateral nerves. After transection at postnatal day 4, total N-acetylaspartate decreased by 80% (P14; p = 0.002) and 94% (P20; p = 0.003). In these developing axon-free nerves, 25 to 33% of oligodendrocyte progenitor cells were proliferating. These data validate magnetic resonance spectroscopy measurements of N-acetylaspartate as an axon-specific monitor of central nervous system white matter in vivo. In addition, the results indicate that neuronal adaptation can increase N-acetylaspartate levels, and that 5 to 20% of the N-acetylaspartate in developing white matter is synthesized by proliferating oligodendrocyte progenitor cells.