When a stem cell divides, one sister cell differentiates and the other retains its stem-cell identity. Differences in the age of an organelle — the centriole — inherited at cell division may determine these differing fates.
One of the enduring mysteries of biology is how two genetically identical sister cells become different from each other after cell division. Stem cells are particularly interesting in this respect because they can divide so that one of the two resulting cells remains an undifferentiated stem cell while the other becomes a differentiated cell type. It has long been thought that such asymmetric cell division may reflect an underlying asymmetry in the segregation of a cellular component at division; the asymmetrically inherited component would have properties that allow it to control the fate of its recipient cell. In this issue (page 947), Wang et al.1 present evidence that the centrosome, a multifunctional organelle that is common to all animal cells, might be such a determinant.
The centrosome is an ancient organelle2 and is one of the cell structures that distinguishes eukaryotic cells (animal and plant cells) from prokaryotes (bacteria and archaea). It helps to form the microtubule cytoskeleton, a network of protein filaments that serve as tracks for moving cellular cargo. It also organizes the primary cilium, a whip-like structure that extends from the surface of cells. In most cells the primary cilium is non-motile, in contrast to the beating cilium of sperm cells, but it is responsive to chemical and mechanical signals outside the cell. For more than a century, the main function of the centrosome was thought to be organization of the mitotic spindle — the filamentous network that carries out the segregation of chromosomes at cell division. But it is now clear that the spindle can form without the centrosome, and that formation of a cilium is actually the centrosome's essential function3. This revelation is particularly exciting because it has coincided with the recognition that the primary cilium is a key signalling centre in vertebrate organisms, thereby placing it, and the centrosome, in the thick of important regulatory processes4.
Each centrosome consists of a pair of cylindrical centrioles and associated microtubule-organizing material. The two centrioles in a pair usually lie in close association at right angles to each other. One centriole, the mother, has structural appendages that confer the ability to anchor microtubules and to organize a cilium; the other centriole, the daughter, lacks these appendages.
The centrosome duplicates once during the cell cycle, and it derives an intrinsic asymmetry from its mechanism of duplication. Centriole duplication is initiated by disengagement of the centriole pair at the end of mitosis, followed in S phase (the phase of DNA synthesis) by assembly of two new daughter centrioles, each adjacent to one of the existing centrioles. This pattern of duplication and segregation results in an age difference in the two centrosomes that are segregated to sister cells at division. One sister cell receives a centrosome containing a newly minted mother centriole (one that was a daughter centriole before duplication and cell division), and the other sister cell receives a centrosome containing the older mother centriole.
Might there be a correlation between the asymmetric fates of dividing stem cells and differences in the age of the centrioles inherited at cell division? Wang et al.1 addressed this question by studying the asymmetric divisions of radial glial cells, a type of neural stem cell that is important for the development of the mammalian cerebral cortex. These cells are highly polarized, stretching between the epithelial surface of the cerebral ventricles and the adjoining layer of cells. The nucleus in these elongated radial glial cells moves up and down during the cell-division cycle, with mitosis occurring at the apical end, adjacent to the ventricle. After division, one cell remains a radial glial cell while the other differentiates into a neuron or a neuronal precursor that migrates away from the ventricular zone.
When Wang et al. labelled the centrioles in developing mouse brain with fluorescently tagged proteins, they found that the older mother centriole was preferentially inherited by the cell that retained the stem-cell fate. To test whether this pattern is important for stem-cell function, they used RNA interference to remove the protein ninein, a component of the centriolar appendages required for mother-centriole functions. Strikingly, when ninein was removed, centriole asymmetry was lost, and the pool of stem cells became depleted, suggesting that inheritance of the older mother centriole is crucial for maintaining stem-cell fate in radial glial cells.
How does the older mother centriole specify stem-cell fate after cell division? On the basis of the properties of centrioles, I consider three possibilities. First, the mother centriole initiates the formation of a primary cilium at the beginning of the cell cycle in most cells. A recent report5 indicates that the cell that inherits the older mother centriole usually projects a cilium before its sister, and that the sister cell thus differs in its response to signals mediated by the cilium. Such a temporal difference in receptiveness to external differentiation signals might result in a cell-fate difference in recently divided cells.
A second possibility is that the older and newer mother centrioles differ in their complement of anchored microtubules during the cell cycle before division. As anchored microtubules can serve as tracks on which to move components towards the centrosome, the older mother centriole might accumulate proteins6 or RNA7 that influence cell fate after division.
Last, it has been proposed that stem cells are maintained by asymmetric segregation of a set of 'stem' chromosomes, all of a similar replicative age8. Such asymmetric chromosome segregation is at odds with the known mechanisms of mitosis, but has been observed in some types of mammalian stem-cell division9. Perhaps the older mother centriole maintains a connection to the chromosomes (the nuclear envelope notwithstanding) from one mitosis, through interphase to the next mitosis, allowing all similarly aged sister chromatids (the copies of a replicated chromosome) to segregate together.
Possibly the most exciting result from Wang and colleagues' work1 is that their findings are remarkably similar to those of studies10 of male germline stem cells in the fruitfly Drosophila melanogaster. In that system, the older centrosome also stays in the stem cell, and this asymmetric segregation is part of a stereotyped division choreographed by signals from the stem-cell niche. We can hope that a unifying mechanistic principle of differentiation will be revealed by future experiments investigating this remarkable organelle and its behaviour during division.
Wang, X. et al. Nature 461, 947–955 (2009).
Marshall, W. F. Curr. Opin. Cell Biol. 21, 14–19 (2009).
Basto, R. et al. Cell 125, 1375–1386 (2006).
Berbari, N. F., O'Connor, A. K., Haycraft, C. J. & Yoder, B. K. Curr. Biol. 19, R526–R535 (2009).
Anderson, C. T. & Stearns, T. Curr. Biol. 19, 1498–1502 (2009).
Fuentealba, L. C., Eivers, E., Geissert, D., Taelman, V. & De Robertis, E. M. Proc. Natl Acad. Sci. USA 105, 7732–7737 (2008).
Lambert, J. D. & Nagy, L. M. Nature 420, 682–686 (2002).
Cairns, J. Genetics 174, 1069–1072 (2006).
Rando, T. A. Cell 129, 1239–1243 (2007).
Cheng, J. et al. Nature 456, 599–604 (2008).
About this article
Journal of Neurochemistry (2011)
Current Opinion in Genetics & Development (2010)
Stem Cells and Development (2010)