MEASUREMENT OF TIME IN OLIGODENDROCYTE-TYPE-2 ASTROCYTE (O-2A) PROGENITORS IS A CELLULAR PROCESS DISTINCT FROM DIFFERENTIATION OR DIVISION

When stimulated by platelet-derived growth factor (PDGF), oligodendroc yte-type-2 astrocyte (O-2A) progenitors derived from perinatal rat opt ic nerves undergo a limited number of cell divisions before clonally r elated cells synchronously and symmetrically differentiate into nondiv iding oligodendrocytes. The duration of this mitotic period is thought to be controlled by a cell-intrinsic biological clock. Thus, in the p resence of PDGF, the measurement of time by the biological clock is in timately coupled to the control of division and differentiation. In co ntrast, O-2A progenitors grown in the presence of PDGF plus basic fibr oblast growth factor (bFGF) divide indefinitely in the absence of diff erentiation and so do not exhibit a limited period of division. We hav e tested whether growth in PDGF plus bFGF alters the duration of the l imited period of division O-2A progenitors exhibit in response to PDGF alone. Accordingly, O-2A progenitors were grown in the presence of PD GF plus bFGF for varying lengths of time, before being switched to con ditions that promote timed differentiation (PDGF but not bFGF). Increa sing duration of culture in PDGF plus bFGF led to a gradual shortening of the period for which O-2A progenitors were subsequently responsive to PDGF alone, until eventually all cells differentiated without divi ding after switching. In contrast, a short exposure to bFGF was not su fficient to cause a similar alteration in the pattern of differentiati on. These results indicate that O-2A progenitors prevented from underg oing timed differentiation nevertheless retain the ability to measure elapsed time, implying that the biological clock in this cell type can be uncoupled from differentiation. Furthermore, they demonstrate that the biological clock does not impose an absolute limit on the number of divisions that an O-2A progenitor can undergo. In contrast with exi sting hypotheses, our observations suggest that the molecular mechanis m that controls timed differentiation must consist of at least two com ponents, with the clock itself being in some manner distinct from mech anisms that limit cell division and/or directly regulate differentiati on. (C) 1994 Academic Press, Inc.