Portal de Periódicos da CAPES

Putting a Notch in Our Understanding of Nuclear Migration

2008; Cell Press; Volume: 134; Issue: 6 Linguagem: Inglês

10.1016/j.cell.2008.09.007

1097-4172

Joshua J. Buchman, Li‐Huei Tsai,

Microtubule and mitosis dynamics

The nuclei of progenitor cells in developing neural epithelia change their position during the cell cycle through a process called interkinetic nuclear migration. Del Bene et al., 2008Del Bene F. Wehman A.M. Link B.A. Baier H. Cell. 2008; (this issue)Google Scholar report that defects in the machinery controlling this process lead to altered exposure to Notch signals and systemic effects on neurogenesis in the retina. The nuclei of progenitor cells in developing neural epithelia change their position during the cell cycle through a process called interkinetic nuclear migration. Del Bene et al., 2008Del Bene F. Wehman A.M. Link B.A. Baier H. Cell. 2008; (this issue)Google Scholar report that defects in the machinery controlling this process lead to altered exposure to Notch signals and systemic effects on neurogenesis in the retina. In the proliferative zones of developing neural structures, neural progenitors make cytoplasmic connections with both the apical and basal sides of the neural epithelium, whereas the position of the nucleus along the apical-basal axis varies depending on the stage of the cell cycle. This phenomenon relies on a process known as interkinetic nuclear migration (INM) (Figure 1). INM has been documented in multiple systems, including the developing spinal cord, cerebral cortex, and retina. In the retina, the cell body and nucleus are on the apical side of the neuroepithelium closest to the pigmented epithelium upon entry into G1. Progression through G1 coincides with movement of the nucleus to the basal side of the neuroepithelium, closer to the site where the lens forms. Upon completion of this basal migration, cells enter S phase, and at the onset of G2, the cell body and nucleus begin to reverse migration toward the apical side. Completion of this nuclear migration is marked by mitosis and repetition of the entire process by daughter progenitor cells (Baye and Link, 2008Baye L.M. Link B.A. Brain Res. 2008; 1192: 29-36Crossref Scopus (92) Google Scholar) (Figure 1A). Although this phenomenon was documented over 70 years ago (Sauer, 1935Sauer F.C. J. Comp. Neurol. 1935; 62: 377-405Crossref Scopus (477) Google Scholar), the cellular mechanisms that regulate INM remain elusive. Moreover, it remains unclear why neural progenitors undergo this process, although it has been suggested to promote diversity in cell fates (Baye and Link, 2008Baye L.M. Link B.A. Brain Res. 2008; 1192: 29-36Crossref Scopus (92) Google Scholar). In this issue, Baier and colleagues report that a mutation in zebrafish (called mikre oko, moks309) that alters the proliferative capacity of the retinal ciliary marginal zone also results in defects in INM and embryonic neurogenesis (Del Bene et al., 2008Del Bene F. Wehman A.M. Link B.A. Baier H. Cell. 2008; (this issue)Google Scholar, Wehman et al., 2005Wehman A.M. Staub W. Meyers J.R. Raymond P.A. Baier H. Dev. Biol. 2005; 281: 53-65Crossref Scopus (61) Google Scholar). moks309 zebrafish harbor a nonsense mutation in the Dynactin-1 (Dnct1) locus that eliminates the C terminus of the protein, and mutants express lower levels of dynactin overall. The dynein complex, with which dynactin associates, plays a critical role in crosslinking microtubules to the nuclear envelope and mediating nuclear movement (Tsai and Gleeson, 2005Tsai L.H. Gleeson J.G. Neuron. 2005; 46: 383-388Abstract Full Text Full Text PDF PubMed Scopus (258) Google Scholar), a key aspect of interkinetic nuclear migration. Previous studies have documented mispositioning of photoreceptor cell nuclei in other mok mutants, and this mispositioning phenotype is associated with decreased photoreceptor cell survival (Doerre and Malicki, 2001Doerre G. Malicki J. J. Neurosci. 2001; 21: 6745-6757Google Scholar, Tsujikawa et al., 2007Tsujikawa M. Omori Y. Biyanwila J. Malicki J. Proc. Natl. Acad. Sci. USA. 2007; 104: 14819-14824Crossref PubMed Scopus (70) Google Scholar). Del Bene et al., 2008Del Bene F. Wehman A.M. Link B.A. Baier H. Cell. 2008; (this issue)Google Scholar find that moks309 mutant zebrafish produce more of the earliest born neurons of the retina, the retinal ganglion cells. In moks309 mutants, the time window for the production of retinal ganglion cells is extended, as determined by expression of atoh7, a transcription factor that promotes the retinal ganglion cell fate. Additionally, the overproduction of retinal ganglion cells appears to come at the expense of Müller glia and bipolar cells, which are normally produced later in retinogenesis. The authors conclude that this alteration in the generation of different cell types occurs in a cell-autonomous manner because neural progenitors from moks309 mutants continue to overproduce retinal ganglion cells even when transplanted to wild-type retinas (Figure 1B). Interestingly, although there is a clear defect in neurogenesis in the moks309 mutant animals, initial production of some retinal cell types is unaffected. Given that all neuronal cell types of the retina are generated from a single progenitor pool, it is intriguing that only certain cell types are affected. Thymidine analog labeling experiments performed in moks309 animals during early retinal neurogenesis demonstrate that many cells exit the cell cycle prematurely during the initial wave of neurogenesis. This observation explains why moks309 animals that have smaller than average eyes contain a larger population of retinal ganglion cells than do wild-type animals; premature cell-cycle exit biases the progenitor population to differentiate into the earliest possible fate, the retinal ganglion cell, and may simultaneously deplete the progenitor pool available for later neurogenesis. Del Bene et al. are able to link the dynactin mutation to the defect in neurogenesis by observing that INM is altered in mutant retinal progenitors. Specifically, mutant cell nuclei migrate to the basal side of the neuroepithelium further and faster, while moving more slowly in the apical direction than wild-type cells. Additionally, basal mitoses occur more frequently in moks309 retinas. Previous work from one of the contributing groups has demonstrated that the extent of basal nuclear migration is a predictor of neurogenic versus proliferative divisions—greater basal migration distances are correlated with neurogenic divisions (Baye and Link, 2007Baye L.M. Link B.A. J. Neurosci. 2007; 27: 10143-10152Crossref PubMed Scopus (129) Google Scholar). Thus, altered INM dynamics resulting from the moks309 mutation may cause premature cell-cycle exit and retinal ganglion cell overproduction in this system. Because INM depends on nuclear coupling to the cytoskeleton, defects in nuclear envelope proteins that mediate this association through the dynactin complex are also likely to affect neurogenesis. Consistent with this hypothesis, expression of one domain of the Syne2a nuclear envelope protein and knockdown of Syne2a have been shown to alter nuclear positioning and photoreceptor cell survival (Tsujikawa et al., 2007Tsujikawa M. Omori Y. Biyanwila J. Malicki J. Proc. Natl. Acad. Sci. USA. 2007; 104: 14819-14824Crossref PubMed Scopus (70) Google Scholar). Expression of the same domain during early retinogenesis results in retinal ganglion cell overproduction and decreased production of later-born cell types. Although not shown directly, this result is also likely due to defects in INM. Although mutations that effect INM can be linked to altered neurogenesis, previous work has not explained why rounds of INM that extend further basally should result in neurogenic mitoses. One possibility is that the extent of nuclear migration could alter cellular exposure to extracellular signals. One candidate molecule that plays a role in regulating progenitor proliferation versus cell-cycle exit is Notch, activation of which can inhibit cell-cycle exit and neuronal differentiation. Conveniently, Notch is expressed in an apical-to-basal gradient in a number of tissues, including the retina (Murciano et al., 2002Murciano A. Zamora J. Lopez-Sanchez J. Frade J.M. Mol. Cell. Neurosci. 2002; 21: 285-300Crossref PubMed Scopus (102) Google Scholar). The authors demonstrate that greater basal nuclear migrations diminished exposure to Notch signaling and Notch pathway activation. Furthermore, either crossing moks309 zebrafish to a line in which cell-cycle exit is delayed or overexpressing activated Notch in mutant cells can rescue the production of cell types normally depleted in the moks309 mutant. This demonstrates convincingly that altered Notch signaling is a cause of the neurogenesis defect observed in these animals. The work of Del Bene and colleagues can be related to another recent study regarding the importance of INM on cortical neurogenesis. Xie et al., 2007Xie Z. Moy L.Y. Sanada K. Zhou Y. Buchman J.J. Tsai L.H. Neuron. 2007; 56: 79-93Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar recently demonstrated that depletion of either Cep120 or TACC proteins, which are associated with the centrosome and the microtubule lattice that links the centrosome and nucleus, results in altered INM and neurogenesis. Similar to the effect of decreased Dynactin or Syne2a expression, depletion of Cep120 or TACC protein levels resulted in increased cell-cycle exit, increased neuron production, more basal mitoses, and slower basal-to-apical nuclear movement. Xie et al. looked at proteins coupling the centrosome to the microtubule lattice, whereas the present study focuses more on the opposite end of this cytoskeletal connection, the interface of the cytoskeleton and nuclear envelope. In both cases, however, the net effect of the loss of these connections is similar. One caveat is that Xie and colleagues did not look at the effect of Cep120 or TACC knockdown on production of specific classes of neurons. Therefore, it is not known whether knockdown of Cep120 or TACC results in global alterations of cortical cell types produced or only specific classes of neurons. Finally, given that the cerebral cortex displays a similar gradient of Notch expression (Murciano et al., 2002Murciano A. Zamora J. Lopez-Sanchez J. Frade J.M. Mol. Cell. Neurosci. 2002; 21: 285-300Crossref PubMed Scopus (102) Google Scholar), it will be interesting to see whether lessons from the present study can be applied to INM defects in this system as well. The authors would like to thank Drs. H. Sive and L. Pan for helpful comments.