INTRODUCTIONOur interest in developmental timing began with a surprising experimental result obtained by Erika Abney, an immunologist who joined our group in the late 1970s. We had earlier defined a set of cell-type-specific markers that allowed us to distinguish the major types of glial cells found in suspensions and cultures of rat central nervous system (CNS) cells: astrocytes, which are heterogeneous and have many functions; oligodendrocytes, which myelinate neuronal axons; and ependymal cells, which line the fluid-filled ventricles of the brain (Raff et al. 1979). Analyzing cell suspensions prepared from developing rat brains from embryonic day eleven (E11) through to birth at approximately E21, she determined when the differentiated cells of each type first appeared. She found that the first astrocytes appeared at E15-E16, the first ependymal cells at E17-E18, and the first oligodendrocytes at around the time of birth (E21-E22). Each cell type always first appeared in very small numbers and rapidly increased over the following days. Remarkably, when she isolated cells from E10 brain and cultured them in 10% fetal calf serum (FCS), the times of first appearance of these three cell types were the same as if the cells had been left in the developing brain; moreover, when cultures were prepared from E13 brain, all three cell types first appeared 3 days earlier, just as in vivo (Abney et al. 1981). These results are as surprising today as they were when we published them more than 25 years ago, because the axial cues and morphogen gradients that play such an important part in controlling cell specification in early animal development are presumably missing in these cultures. Moreover, it seems unlikely that the sequential cell-cell interactions that are thought to help time events in development could occur normally in such cultures.Whatever the nature of the timing mechanisms involved, the finding that they apparently could operate normally in dissociated-cell culture over a 10-day period encouraged us to study them. The complexity of the brain cell cultures, however, was daunting, and so we turned to the developing optic nerve, which is much simpler, and we focused on the timing of oligodendrocyte development.
AN INTRACELLULAR TIMER IN OLIGODENDROCYTE PRECURSOR CELLSThe optic nerve contains no neurons, although it contains the axons of retinal ganglion neurons. The main cell types in the nerve are astrocytes and oligodendrocytes, with smaller numbers of macrophages (microglia), blood vessel cells, and glial precursor cells. Whereas the astrocytes in the optic nerve develop from the neuroepithelial cells of the optic stalk (the primordium of the nerve), the oligodendrocytes develop from precursor cells that migrate into the developing nerve from the brain, beginning before birth (Small et al. 1987). The oligodendrocyte precursor cells (OPCs) divide a limited number of times before they stop and terminally differentiate into postmitotic oligodendrocytes. As in the brain, the first oligodendrocytes appear in the...