When a cell divides to generate daughters of different cell fates, the division is said to be asymmetric. In C. elegans, 807 of the 949 nongonadal cell divisions that occur during development are asymmetric. An important feature of most asymmetric cell divisions in C. elegans is that they occur with a defined orientation with respect to the body axes. The relative orientation of the different daughter cells to each other and to the body axes gives an asymmetric cell division a polarity. Mutations in
lin-44 cause the polarities of certain asymmetric cell divisions in the tail to be frequently reversed, while maintaining the asymmetries and the orientations of division planes. Genetic analysis indicated that the polarity reversal defect is the result of a complete loss of
lin-44 function (Herman and Horvitz, 1994, Development, 120: 1035-1047). Molecular analysis has revealed that
lin-44 encodes a member of the Wnt family of secretory glycoproteins, which have been shown to function as short range signaling molecules. Examination of animals containing reporter constructs designed to mimic patterns of
lin-44 expression, as well as in situ hybridization experiments, have demonstrated that
lin-44 is expressed in the developing tail hypodermis, which is posterior to the cells affected by
lin-44 mutation. Mosaic analysis has shown that
lin-44 function is not required in the cells affected by
lin-44 mutation, but in cells in which we observe expression. Based upon these results we propose that
lin-44 protein is secreted by the tail hypodermal cells and affects the polarity of asymmetric cell divisions that occur more anteriorly in the tail (Herman et al. Cell, 83: 101-110). As a result of the reversal of T cell polarity in
lin-44 mutants, the phasmid socket cells are misplaced, and dye filling by the phasmid neurons is blocked. To identify other genes that act with
lin-44 to specify cell polarity, we are screening for additional mutations that reduce or abolish the dye filling of phasmid neurons without affecting amphid neuron dye-filling. We have isolated over 40 such mutations. Many of these have been scored for the presence of the phasmid socket cells in their normal positions. Two mutations lacked phasmid socket cells in the normal positions: one is a new
lin-44 allele and the other,
mn553, identifies a new gene, which maps to LGII between
dpy-10 and
unc-4 and appears to cause defects in gonad morphology as well as in T cell polarity. The map position and phenotype of
mn553 suggest that it might be an allele of
mig-5, which has been shown by Guo and Hedgecock (WBG
v14 n1, p. 52) to be involved in cell fate determination and cell migration and to encode a homolog of the Drosophila dishevelled (dsh) gene. We thus suggest that
mig-5 may also be involved in the control of cell polarity. We also expect to identify mutations in
lin-17, which cause a loss of T cell polarity and hence block phasmid dye-filling.
lin-17 has been shown by Sawa and Horvitz (1995 Meeting Abstracts, p. 455) to encode a homolog of the Drosophila frizzled (fz) gene. In Drosophila, fz , wingless (a member of the Wnt gene family) and dsh mutants have defects in tissue polarity. Many structures on the cuticular surface of Drosophila display a tissue polarity characterized by arrays of polarized sensory bristles or hairs. This polarity appears to be controlled by a cell signaling pathway involving fz (
lin-17), dsh (
mig-5) and wingless (
lin-44). We are exploring the possibility that these genes may interact in a similar manner to control cell polarity in C. elegans.