A defining feature of stem cells is their ability to continuously maintain a balanced number of stem cells (self-renewal) while being able to generate specialized progeny (differentiation). An additional feature of stem cells is their ability to migrate. Therefore, stem cells are unique as they are able to balance four possible fates within a single cell: quiescence, migration, proliferation, and differentiation. These fate decisions are made in the context of the supporting stroma cells they adhere to, also referred to as the niche (Fig. ). Is there a role for polarity in stem cell fate decisions?
Stem cells that divide asymmetrically have to orient their mitotic spindle to allow for cell fate determinants to segregate asymmetrically into daughter cells. In theory, stem cells could also divide exclusively symmetrically, provided that extrinsic signals postdivision induce their postmitotic cell fate (inductive postdivision model). Growing experimental evidences demonstrate that stem cells have the ability to divide asymmetrically in vitro and in vivo, determining stem cell fate [
31,
32,
33–
35]. Even more interestingly, Wu et al. demonstrated that hematopoietic stem cells (HSCs) can undergo both symmetric and asymmetric division, and that the balance between them is not hard-wired but responsive to extrinsic and intrinsic cues [
36–
38].
One obvious hypothesis currently supported by experimental evidence is that polarity establishment during mitosis regulates the mode/outcome (symmetric vs. asymmetric) of stem cell divisions. For example, in
D. melanogaster, male germ line stem cells (GLSCs) are attached to somatic hub stem cells, which constitute the stem cell niche. On division, GLSCs polarize and produce one daughter cell or gonialblast that initiates differentiation and one daughter stem cell, which remains attached to the hub [
16,
39–
43]. Asymmetric GLSC division is controlled by the orientation of the mitotic spindle, and by the programmed anchoring of the mother centrosome to the self-renewing stem cell, while the gonialblast receives the newly synthesized centrosome [
31,
16,
34]. A stem cell in close contact with its niche will thus orient its mitotic spindle perpendicularly to the niche surface, ensuring that only one daughter cell maintains contacts with the niche, and thus retains the ability to self-renew. Signaling molecules such as Dpp and Hh, the bone-morphogenic protein (BMP)2/4 homolog in
D. melanogaster, released from the niche/hub cells are involved in regulating the mode of division, implying an extrinsic regulation of this polarity by the niche.
In the developing mouse brain, progenitors located in the apical surface of the ventricular zone are self-renewing and display an apical-basal polarity with a basolateral domain in contact with the basement membrane [
1]. The nuclei of these neuronal precursors move basally away from the ventricular surface for DNA synthesis, and apically return to the surface for mitotic division; a process known as interkinetic migration or “to-and-fro” nuclear translocation [
44,
45]. In comparison with progenitors located in the subventricular zone that gradually deplete, ventricular zone progenitors contain a specialized apical membrane domain whose activity is regulated by Cdc42. In these cells,
Cdc42 deletion results in an immediate increase of basal mitosis, a gradual loss of apical membrane protein location, and increasing failure of apically directed interkinetic nuclear migration. Therefore, these Cdc42-deficient progenitors acquire the fate of the progenitors located in the subventricular zone that cannot self-renew for long time and gradually deplete [
1]. It was also recently demonstrated that a planar cell polarity pathway activated by Wnt7
a controls the number of muscle stem cells and the regenerative potential of muscle tissue [
46], again most likely by regulating the mode of stem cell divisions.
On division, murine HSCs distribute Numb asymmetrically to daughter cells, a mechanism already described for asymmetric division of
D. melanogaster neuroblasts [
47]. Mammalian Numb displays a complex pattern of functions such as controlling cell fate decision, endocytosis, cell adhesion, cell migration, and ubiquitination of specific substrates and can interact with several signaling pathways (i.e., Notch, Hedgehog, p53). Alterations of Numb-dependent events and/or of Numb distribution during asymmetric cell division suggest an important role for Numb in disease and cancer progression [
48] (Fig. ).
An additional example implying a role for polarity in stem cell function comes from analysis of human hematopoietic stem/progenitor cells that can differentially localize the tetraspanins CD53, CD63, the transferrin receptor or CD71, and CD62 or
l-selectin while dividing in vitro [
32,
35]. An asymmetric distribution of cellular components on division has also been shown by a fluorescent Notch-activity indicator system [
36]. Cytokine distribution has been shown to correlate with cell fate determination on division [
49]; however, it is unclear whether cytokines are actually instructive in this process. Collectively, these and other published data support that both modes of cell division (asymmetric/polar and symmetric/nonpolar) are used by HSCs, with both intrinsic as well as extrinsic signals determining polarity on division.
Recently, a stem cell stroma synapse-like structure has been postulated in analogy to the well-characterized immune cell synapse [
50–
52] that describes the contact plane between T-cells and antigen-presenting cells [
53–
55]. Recent data demonstrating polarity in nondividing HSCs interacting with niche cells support a role for polarity in the stem cell synapse, like the reported T-cell-polarity on interaction with antigen-presenting cells [
56,
57]. Adult stem cells residing in their niche are mostly in a quiescent cell cycle state. Although cell cycle quiescence has so far not been frequently associated with cellular polarity, recent results analyzing mice deficient for Cdc42 in HSCs suggest that polarity established by Cdc42 might be necessary for both adhesion of HSCs to the niche as well as their quiescence, as these mice show an increase in the number and the frequency of phenotypic short-term HSCs and a loss of long-term HSCs [
58,
59]. Therefore, albeit supported by only few experimental results so far, polarity might be necessary in maintaining HSC quiescence by functioning in the formation of the stem cell-niche synapse, and polarity alterations might importantly impair stem cell quiescence or function. Such a polarity-based synapse model though leaves the question open whether adhesion to the niche induces polarity in stem cells (extrinsic regulation of polarity) or whether stem cells present an intrinsic polarity axis, in which a polar interaction with the niche might only be secondary to this intrinsically established polarity [
16,
50–
51].
Although polarity in migration has been extensively studied in differentiated progeny of stem cells-like neutrophils [
5,
60,
61], the role of polarity in stem cell migration has not been investigated in great detail. Obviously, more research in this area is necessary, although there is evidence that stem cell migration and migration-associated polarity are also regulated by small RhoGTPases [
62].