66 Recently, a number of molecules originally thought to have a primary role in cell determination have been shown to affect the cell cycle at specific check points, while other molecules discovered for their roles in the cell cycle progression are known to affect the determination and differentiation of neurons. These discoveries have led to a more detailed investigation of the complex molecular machinery that co- ordinates proliferation and differentiation. Addresses * ‡ Department of Anatomy, University of Cambridge, Downing Street, CB2 3DY, UK † Department of Oncology, University of Cambridge, Wellcome Trust Centre for the Study of Disease, Cambridge Institute for Medical Research, Wellcome/MRC Building, Addenbrookes Hospital, Hills Road, Cambridge, CB2 2XY, UK; e-mail: ap113@hermes.cam.ac.uk *e-mail: so218@cam.ac.uk ‡ e-mail: harris@mole.bio.cam.ac.uk Current Opinion in Neurobiology 2001, 11:66–73 0959-4388/01/$ —see front matter © 2001 Elsevier Science Ltd. All rights reserved. Abbreviations bHLH basic helix-loop-helix Cdk cyclin-dependent kinase Id inhibitor of differentiation NGF nerve growth factor Rb retinoblastoma protein Shh Sonic hedgehog TH thyroid hormone Introduction In the nervous systems of almost all species, neurons of par- ticular types arise from cells that divide at particular times. In vertebrates, this is clearly manifest in a stratified or his- togenetic neuronal architecture in which different layers are composed of cells with different birthdates, the cerebral cortex being the archetypal example with early-born cells in the deeper layers and later-born cells in more superficial layers. Although no stratified organisation is evident in invertebrates, different ganglion mother cells and their dis- tinct neuronal progeny arise from successive asymmetric divisions of central nervous system (CNS) neuroblasts [1,2]. Different neural fates thus appear to be determined at spe- cific developmental stages or after certain numbers of cell divisions; however, there are no simple rules that dictate when these decisions are made. Some fates, such as whether to be a neural or a glial precursor, may be deter- mined when a cell is actively proliferating. Other decisions, such as which type of neuron to form in the retina or spinal cord, may be made around the time of the final cell cycle. The mature phenotype of a neuron is often determined postmitotically, or at adult stages. Because there are no obvious rules, the histogenetic order that we see in some neural structures may then simply be the result of a cascad- ing developmental process that has no mechanistic relationship to proliferation or the cell cycle. To test whether birth order affected neural cell fate, Harris and Hartenstein [3] arrested the cell cycle of developing Xenopus embryos using inhibitors of DNA synthesis. Surprisingly, all types of neurons differentiated in the reti- na of these animals. This result suggests that cell cycle arrest by itself does not preclude diverse neural cell fate determination. It further shows that precursor cells can receive determination signals without actively cycling. The result, however, does not rule out the possibility that com- ponents of the cell cycle machinery influence the fate determination machinery. The converse, in fact, is more likely. The striking relationship between cellular differen- tiation and cell cycle exit in many systems, including the nervous system [4], strongly suggests that the determina- tion machinery does influence the cell cycle. Work in other model systems such as yeast, Dictyostelium, nematodes, the Drosophila wing and vertebrate muscle [5–11] has shed light on some of the interactions that co-ordinate the cell cycle with determination and differentiation. The interaction between neuronal fate and the cell cycle was most clearly uncovered by the elegant work of McConnell and Kaznowski [12] in the layered neocortex. By labelling neural precursors from a young animal with [ 3 H] thymidine and then transplanting these cells at dif- ferent phases of the cell cycle to an older animal, they found that these proliferating cells tended to receive deter- mination signals for layer identity sometime during their final cell cycle. The radiolabelled early precursor cells that were transplanted during S phase of their final cell cycle were found in layer 2/3 of the older host instead of layer 6, which would have been their natural fate. When the same experiment was performed with cells that had been allowed to enter M phase of the final cell cycle, the cells retained their deep layer fate in the older host cortex. The conclusion is that the laminar fate of cortical neurons is determined during the S phase or G2 phase of the final cell cycle, and clearly indicates that the cell cycle status has some influence over the determination process. Possible relationships between the cell cycle components and the determination/differentiation process are shown in Figure 1 and raise fundamental questions. Do cell cycle gene products regulate the function of determination gene prod- ucts and vice versa? If so, how do they do it? Our review of the present data suggests that there are indeed highly involved bidirectional mechanisms for coordinating the cell cycle with the process of neural determination and differentiation. The role of cell cycle molecules in regulation of determination/ differentiation Some of the key factors that regulate progression through the cell cycle are shown in Figure 2a. Cyclin-dependent Cell cycle and cell fate in the nervous system Shin-ichi Ohnuma*, Anna Philpott † and William A Harris ‡