c-Kit, the cytokine receptor that mediates SCF-induced cell proliferation, has recently been found to trigger cell polarity establishment in mouse HSPC. SCF stimulation of cultured bone marrow cells induces polarization of c-Kit and the aPKC-Par6-Par3 cell polarity machinery to one pole, followed by asymmetric division into dissimilar daughter cells. The SCF-induced redistribution of proteins in HSPC resembles that of polarized epithelial cells, in which cell surface proteins selectively traffic to either an apical or basolateral domain. We hypothesize that c-Kit might behave like an “apical” protein in this context. To test this hypothesis, we retrovirally transduced c-Kit constructs into Madin-Darby Canine Kidney (MDCK) cells, an epithelial cell line that forms polarized cysts when cultured in collagen gel. Each polarized MDCK cyst forms an apical surface facing inside, and a basolateral surface facing outside. When expressed in these cysts, a wild-type c-Kit construct traveled only to the apical surface and never to the basolateral surface, in agreement with our hypothesis. We next examined the distribution of c-Kit-BL, a c-Kit mutant that contained in its cytoplasmic domain a basolateral-trafficking (BL) signal sequence derived from TGFβ receptor. The c-Kit-BL mutant lost its apical trafficking preference and distributed instead to the BL surface, suggesting that the apical distribution of c-Kit could be blocked by an aberrant trafficking signal. We then examined the distribution of c-Kit expressed in MDCK cells cultured in liquid suspension. Under this condition, MDCK cells aggregated to form inside-out spheres, with an apical outer surface and a BL inner surface. Along with the polarity reversal of the spheres, the distribution patterns of c-Kit and c-Kit-BL were also reversed. Embedding the inside-out spheres in collagen gel re-established the original polarity, and the c-Kit constructs regained their original distributions. We then tested the behavior of c-Kit-BL-A531G and c-Kit-BL-Q532A, revertant and pseudo-revertant of c-Kit-BL, in which a critical residue (A531G) or non-critical residue (Q532A) of the BL signal sequence was mutated. We found that c-Kit-BL-A531G re-established apical distribution, whereas c-Kit-BL-Q532A remained at the basolateral surface. These results show that c-Kit behaved like an apical protein in the milieu of polarized epithelial cells and followed similar protein trafficking rules. We then asked if the apical trafficking capacity of c-Kit correlates with its ability to mediate SCF-induced proliferation of blood cells. We transduced c-Kit or c-Kit-BL into 32D cells, a mouse myelomonocytic cell line that normally does not express c-Kit and requires stimulation with hematopoietic growth factors to proliferate. Untransduced 32D cells proliferated only in response to IL3 stimulation, whereas c-Kit-transduced 32D cells also proliferated upon SCF simulation. By contrast, c-Kit-BL-transduced 32D cells behaved like the parental 32D cells and did not proliferate upon SCF stimulation, suggesting that a c-Kit mutant deficient in apical trafficking could not mediate proliferation. We next examined the effect of c-Kit or c-Kit-BL expression on polarization and proliferation of primary HSPC. We transduced mouse bone marrow cells to express either c-Kit or c-Kit-BL and examined for SCF-induced polarization. 76–96% of c-Kit-transduced HPSC exhibited polarization of markers Sca1, aPKC, β-catenin and annexin 2 upon SCF stimulation, whereas only 16–26% of c-Kit-BL-transduced cells underwent such polarization. We then cultured transduced bone marrow cells in methylcellulose in the presence of SCF to form colonies. c-Kit-transduced cells gave rise to discrete colonies but c-Kit-BL-transduced cells failed to produce any colonies, indicating that the c-Kit-BL mutant dominantly interfered with the ability of the endogenous c-Kit to promote proliferation. In summary, our data show that c-Kit trafficked selectively to an “apical” domain in epithelial cells and HSPC. A c-Kit mutant defective in the apical trafficking not only failed to induce blood cell proliferation but also suppressed the proliferative function of the endogenous c-Kit. Further delineation of the mechanism of apical trafficking of c-Kit should help elucidate the relationship between the cell polarity machinery and the regulation of HSPC proliferation.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.