Does oligodendrocyte survival depend on axons? B.A. Barres*, M.D. Jacobson*, IR. Schmidt, M. Sendtnerf and M.C. Raff* ‘Medical R?warch Council Developmental Neurobiology Programme, Department of Biology, Medawar Building, University College, London WCIE 6BT, UK and tMax-Planck-Institute for Psychiatry, Department of Neurochemistry, 8033 Planegg-Martinsried, Germany. Background: We have shown previously that oiigoden- drocytes and their precursors require signals from other cells in order to survive In culture. In addition, we have shown that about 50 % of the oiigodendrocytes pro- duced in the developing rat optic nerve norrnaiiy die, ap- parently in a competition for the limiting amounts of sur- vival factors. We have hypothesized that axons may con- trol the levels of such oiigodendrocyte survival factors and that the competition-dependent death of oiigoden- drocytes serves to match their numbers to the number of axons that they myeiinate. Here we test one predic- tion of this hypothesis - that the survival of developing oligodendrocytes depends on axons. Results: We show that oiigodendrocyte death occurs selectively in transected nerves in which the axons degenerate. This ceil death is prevented by the deiiv- cry of exogenous ciiiary neurotrophic factor (CNTF) or insulin-like growth factor I (IGF-I), both of which have been shown to promote oiigodendrocyte survival in vitro. We also show that purified neurons promote the survival of purified oligodendrocytes in vitro. Conclusion: These results strongly suggest that oiigo- dendrocyte survival depends upon the presence of axons; they also support the hypothesis that a competi- tion for axon-dependent survival signals normally helps adjust the number of oiigodendrocytes to the number of axons that require myeiination. The identities of these signals remain to be determined. Current Biology 1993, 3:489-497 Background We have recently shown that the developing oligo- dendrocytes in the rat optic nerve behave like many developing vertebrate neurons, in that they depend on signals from other cells to survive in culture and about 50% of them die during normal development [ 11. Moreover, normal oligodendrocyte death, like nor- mal neuronal death, can be prevented by the delivery of exogenous survival factors [ 11. These findings sug- gest that developing oligodendrocytes, like many de- veloping neurons, may normally compete for limiting amounts of survival factors and that this competition might be an important mechanism for controlling their number. The only known function of oligodendrocytes is to myelinate axons in the vertebrate central nervous sys- tem. It would, therefore, make sense for axons to play an important role in regulating oligodendrocyte survival during the development of the central ner- vous system (CNS). We have found previously that the number of oligodendrocytes and their precursors de- creases in transected neonatal optic nerves, thus indi- cating that axons normally influence oligodendrocyte development [ 21. These findings were consistent with the possibility that axons might promote oligodendro- cyte survival. But because the optic nerves were trans- ected before the appearance of oligodendrocytes in the nerve, the results did not exclude the equally likely pos- sibilities that axons are required for oligodendrocyte precursor cell proliferation [3] or for the migration of oligodendrocyte precursor cells into the optic nerve 141. -. Here we provide strong evidence that axons are required for oligodehdrocyte survival. When the optic nerve is cut after many oligodendrocytes have ap- peared, the observed number of dead oligodendro- cytes increases and the number of live oligodendro- cytes decreases dramatically. This cell death is pre- vented by the delivery of exogenous oligodendrocyte survival factors, and it does not occur in optic nerves from mutant mice, whose axons do not degenerate after transection. These results suggest that a compe- tition for an axon-dependent survival signal may help to match the number of oligodendrocytes with the number of axons that require myelination, just as a competition for target-cell derived neurotrophic factors is thought to help match the number of developing presyrraptic neurons to the number of postsynaptic cells that require innervation [ 51. Results Effect of transection on cell death in the developing rat optic nerve To test whether axons influence the survival of devel- oping glial cells, we transected the right optic nerve be- hind the eye in rats on postnatal day 12 (~12), inducing the axons in the nerve to degenerate. At various times thereafter, we perfused the rats with paraformalde- hyde under anaesthesia, cut longitudinal frozen sec- tions ‘of both optic nerves, and stained them with the fluorescent nuclear dye propidium iodide. Dead (py- knotic) cells were identified, using phase-contrast mi- croscopy, by their shrunken phase-dark appearance, or, with fluorescence microscopy, by their condensed and often fragmented nucleus, which stained intensely with propidium iodide [ 11. As shown in Figure 1, the number of pyknotic cells rapidly increased following Correspondence to: B.A. Barres. @ Current Biology 1993, Vol 3 No 8