The study of stem cells holds immense promise for furthering our understanding of processes such as embryonic development, adult aging, and tumor formation. This is due to their remarkable ability to self-renew, to produce more stem cells and to differentiate into one or more specialized cell types. Many recent studies have focused not on the stem cells themselves, but on the cells surrounding them and their extracellular environment. It is now thought that for most stem cell types this environment, or "niche," provides signals necessary for the stem cells to continue to self-renew, and that upon exit from this niche they begin to undergo differentiation (1). Thus, the mechanism by which stem cells decide either to remain in the niche or to leave it should be a major player in the balancing act between stem cell self-renewal and differentiation. On page 1547 of this issue, Yamashita et al. (2) explore this mechanism in the germline stem cell niche of the Drosophila testis. They find that the stem cells themselves control this process directly by orienting the plane of their division. Surprisingly, this orientation is established by an apparently new method of asymmetric cell division, which could potentially be used in other systems where external signals dictate cell fate.
The Drosophila testis contains an average of nine germline stem cells surrounding a small cluster of nondividing somatic cells known as the hub (see the figure). Two recent studies have shown that the hub is responsible for creating the germline stem cell niche by secreting a signal that is required by germline stem cells for their self-renewal. Stem cells next to the hub receive high levels of this signal and are thus instructed to self-renew, whereas cells further away receive less signal and begin to differentiate into spermatogonia (3, 4). Using tubulin tagged with green fluorescent protein, which marks the mitotic spindle in dividing cells, Yamashita et al. find that germline stem cells always orient their divisions perpendicular to the hub, so that only one daughter cell contacts the hub, whereas the other is displaced from the niche (see the figure). Remarkably, the positioning of the spindle in the stem cells appears to be set up early during interphase, because the single centrosome present shortly after cell division is already found consistently localized to a cortical region of the stem cell next to the hub. Upon duplication, one centrosome remains at the hub while the other migrates to the opposite side of the cell to set up the mitotic spindle. This is in contrast to all other well-studied examples of oriented cell division. Such examples include division of another type of Drosophila stem cell, the neuroblast, and the first divisions of the Caenorhabditis elegans embryo--here, the spindle is formed before it rotates 90º into its final position during mitosis (see the figure) (5). The Yamashita et al. findings in Drosophila germline stem cells suggest a new mechanism of asymmetric cell division.
Spindle orientation and asymmetric cell division. In the model systems shown, the fate of cell progeny is determined by an asymmetrically positioned determinant or signal (purple). During interphase, the initial single centrosome (dotted yellow) duplicates and separates (solid yellow). The two centrosomes then will orchestrate formation of the mitotic spindle. In the P1 blastomere of the C. elegans embryo and in Drosophila neuroblasts, final spindle position is established by a 90º rotation of the centrosome-spindle complex during mitosis. In contrast, in Drosophila germline stem cells of the testis, one centrosome remains anchored to the region of the cortex at the interface between germ cells and somatic hub cells, while the other centrosome migrates to the opposite side to establish mitotic spindle orientation. The orientation of the mitotic spindle ensures that as the stem cell divides, the daughter cell nearest the hub remains in the niche and is marked for self-renewal, whereas the daughter cell farther away from the hub is edged out of the niche and begins to differentiate. GMC, ganglion mother cell.
CREDIT: KATHARINE SUTLIFF/SCIENCE
Germline stem cells in Drosophila testes that carry a mutation in centrosomin, an integral centrosome component, provide clues as to how the spindle-positioning mechanism may operate. These mutant stem cells display defects in positioning of the centrosomes during interphase, and the resultant mitotic spindles are often misoriented. This is consistent with a direct role for the centrosomes in setting up the division plane, as suggested by the early localization of the centrosomes during interphase. Strikingly, the number of stem cells in the testes of the centrosomin mutant flies increases significantly. These stem cells become crowded around the hub, presumably because of the symmetric divisions of stem cells that have misoriented spindles. It thus appears that in Drosophila testes, the balance between stem cell self-renewal and differentiation is not dictated entirely by the amount of available space in the niche; rather, this balance is influenced directly by the orientation of stem cell division. The authors observe a similar misorientation of mitotic spindles and increase in stem cell number in flies with mutations in the Drosophila homologs of the mammalian adenomatous polyposis coli (APC) tumor suppressor protein, which has been implicated in spindle orientation and cell adhesion (6, 7). Because these fly APC proteins are enriched at the cell cortex and at the centrosome during cell division, the authors propose that they may play a structural role in linking the centrosome to the cell adhesion molecule E-cadherin, which they find enriched at the stem cell-hub interface. APC is also an integral component of the Wnt signaling pathway, which participates in the regulation of stem cell division (8, 9). It will be interesting to determine whether Wnt signaling is important for stem cell division in the testis, contributing to the phenotypes observed in the APC mutant flies.
The study by Yamashita et al. raises a number of intriguing questions for further investigation. Although the authors do not observe symmetric divisions in wild-type Drosophila male germline stem cells, can such divisions take place and replenish a niche depleted of stem cells, as found in the Drosophila ovary (10)? How is the specialized cortical region recognized by the centrosome established? The authors suggest that homotypic interactions between cadherins at the germ cell-hub interface may be involved. If this is indeed the case, something must be preventing these interactions at interfaces between the germ cells and the somatic cyst cells that flank them. Given that after centrosome separation only one centrosome migrates away to the opposite cortex, how are the differences between the two centrosomes established and recognized? A recent study suggests that the mother and daughter centriole differ substantially in the degree to which they associate with microtubules and move within the cell (11). Such a mechanism might be used to differentiate between the duplicated centrioles in the Drosophila germline stem cells. It will be interesting to determine whether the centrosome differences observed are further exploited by testis stem cells to deliver cell fate information to the daughter cells (12). The Yamashita et al. study points to previously unappreciated mechanisms within stem cells that orient their divisions. Together with external cues, these mechanisms regulate the balance between stem cell self-renewal and differentiation.
References
- A. Spradling et al., Nature 414, 98 (2001) [Medline].
- Y. M. Yamashita, D. L. Jones, M. T. Fuller, Science 301, 1547 (2003).
- A. A. Kiger et al., Science 294, 2542 (2001).
- N. Tulina, E. Matunis, Science 294, 2546 (2001).
- C. Q. Doe, B. Bowerman, Curr. Opin. Cell Biol. 13, 68 (2001) [Medline].
- B. M. McCartney et al., Nature Cell Biol. 3, 933 (2001) [Medline].
- B. Lu et al., Nature 409, 522 (2001) [Medline].
- X. Song, T. Xie, Development 130, 3259 (2003) [Medline].
- T. Reya et al., Nature 423, 409 (2003) [Medline].
- C. H. Zhu, T. Xie, Development 130, 2579 (2003) [Medline].
- M. Piel et al., J. Cell Biol. 149, 317 (2000) [Medline].
- J. D. Lambert, L. M. Nagy, Nature 420, 682 (2002) [Medline].
R. M. Wallenfang is in the Department of Cell and Developmental Biology, University of Pennsylvania, Philadelphia, PA 19104, USA. E. Matunis is in the Department of Cell Biology, The Johns Hopkins Medical Institutions, Baltimore, MD 21210, USA. E-mail: matunis@jhmi.edu 10.1126/science.1090070
Include this information when citing this paper.
