PubMed2001
Studies of ischemic brain injury in neonatal rodents have focused upon the pathophysiology of neuronal damage. Much less consideration has been given to white matter injury, even though it is a major contributor to chronic neurological dysfunction in children. In the human neonate, particularly in those born prematurely, periventricular white matter is highly susceptible to hypoxic--ischemic (H--I) injury. To understand the basis for this selective vulnerability, we examined myelin gene expression and cell death in the subventricular layer and the surrounding white matter of neonatal mice following H--I insult. Using an in situ hybridization technique that gives high resolution and is very sensitive, we examined myelin basic protein and proteolipid protein gene expression three and twenty-four hours after a H-I insult. To elicit unilateral forebrain hypoxic and ischemic injury, 9--10-day-old mice underwent right carotid artery ligation followed by timed (40--70 min) exposure to 10% oxygen. Twenty-four hours following H--I, myelin basic protein and proteolipid protein transcripts were markedly reduced in striatum, external capsule, fornix, and corpus callosum in the injured side. Three hours after lesioning (ligation+70 min hypoxic exposure) myelin basic protein gene transcripts were visibly reduced in the ipsilateral white matter tracts. Interestingly, some cells in the subventricular layer expressed proteolipid protein transcripts, and 3 h after a H--I insult they were degenerating in the injured but not contralateral side. TUNEL staining showed an increase in the number of positive cells in the injured subventricular layer and corpus callosum but the adjacent striatum did not show a corresponding change in the number of TUNEL labeled cells. Ultrastructural studies of the subventricular zone and corpus callosum 3 h after H--I revealed that many subventricular cells, glial cells in the corpus callosum, and callosal axons in the injured side had already degenerated. However, the subventricular cells, glia and axons in the contralateral corpus callosum were spared. Many cells in the injured corpus callosum exhibited a apoptotic morphology; yet more mature oligodendrocytes in this region appeared normal. Our results show that a H--I insult causes a surprisingly swift and dramatic degenerative response in the subventricular layer and adjacent white matter. Within 3 h after H--I, the programmed cell death cascade was initiated; internucleosomal DNA degradation took place in subventricular and glial cells; oligodendrocyte progenitors died and axonal degeneration in the ipsilateral corpus callosum was extensive. The swiftness of the subventricular and glial cell degeneration suggests the H--I insult directly targets glia, as well as neurons, and raises the provocative question of whether glia exert damaging effects upon neurons and axons. Since the severity of the H--I insult can be modulated by varying the duration of hypoxia, the model is ideal to study whether oligodendrocyte progenitors are more susceptible to death than mature oligodendrocytes, whether mature oligodendrocytes de-differentiate and then are induced to remyelinate surviving axons, and/or whether oligodendrocyte progenitors in the subventricular layer can be stimulated to proliferate, migrate, and remyelinate the surviving axons.