Worm Breeder's Gazette 13(5): 62 (February 1, 1995)
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Institute of Molecular Biology, University of Oregon, Eugene, Oregon 97403 Mutations in the maternal genes par-1, par-2, and par-3 result in developmental defects during embryogenesis that in many cases appear very similar (Kemphues et al., Cell 52, 311-20, 1988). The first cleavage in all three types of mutant embryos results in daughters typically of equal size, in contrast to wild-type embryos in which a smaller posterior daughter (P1) and larger anterior daughter (AB) are born at the 2-cell stage. In par-1, par-2, and par-3 embryos, defects are also observed in the segregation of cytoplasmic P-granules to posterior blastomeres during early embryonic cleavages (hence the name partitioning defective). Subsequent cleavages in all three types of mutant embryos are much more synchronous than in wild-type embryos, and the terminally differentiated phenotypes of all three are superficially similar, characterized in part by the production of excess pharynx and excess body wall muscle. More recent studies (Bowerman et al., Cell, 74,1993) have shown that in addition to being defective in the localization of P-granules, par-1 mutant embryos also fail to properly localize SKN-1 protein to P1 descendants. SKN-1 is a transcriptional regulator (Blackwell et al., Science 266, 621-28, 1994) which is required to specify the identity of EMS, a daughter of P1 (Bowerman et al., Cell 68, 1061-75, 1992). In wild- type embryos, SKN-1 accumulates to high levels only in P1 and its 4-cell and 8-cell stage descendants; in par-1 embryos SKN-1 is evenly distributed in all early blastomeres. AB descendants in par-1 embryos produce extra pharyngeal cells and require skn-1 function to do so (Bowerman et al., 1993), consistent with models in which mislocalization of SKN-1 (and possibly other factors) results in ectopic production of cell types normally made only by EMS. Furthermore, glp-1;par-1 double mutant embryos appear to produce as many ectopic, excess pharyngeal cells as par-1 embryos, suggesting that the production of pharyngeal cells in par-1 embryos is dependent only on skn-1 function. In wild-type embryos, the GLP-1 receptor is required for the induction of anterior pharyngeal cell production by the AB daughter, ABa (Priess et al., 1987). Finally, skn-1 par-1 double mutant embryos usually fail to produce any body wall muscle cells. This result is surprising because both par-1 and skn-1 single mutant embryos produce large numbers of body wall muscle cells, suggesting that skn-1-independent body wall muscle pathway(s) are not functional in par-1 mutant embryos. We have undertaken similar studies of the par-2 and par-3 mutant phentoypes. We have (i) used laser ablation to isolate the development of individual blastomeres and thereby determine the cell fate patterns they produce, (ii) constructed double mutants with skn-1 and glp-1 to address the genetic requirements for pharyngeal and body wall muscle cell production in these mutants, and (iii) used antibody staining to determine the distribution of SKN-1 protein in par-2 and par-3 embryos. The alleles analyzed for par-2 are it5 and lw32; for par-3, it62 and it71 (kindly provided by K. Kemphues). We have found that while the terminal phenotypes of par-1, par-2, and par-3 embryos appear similar, the defects in the different mutants are substantially different. First, skn-1- independent body muscle pathways are active in par-2;skn-1 and par-3;skn-1 double mutant embryos, in contrast to the loss of these pathway(s) in skn-1;par-1 embryos. We also have found that the skn-1-independent body wall muscle pathway(s) are restricted to P1 descendants in both par-2 and par-3 embryos, as in wild-type embryos. However, while SKN-1 protein is present at similar levels in all 4-cell stage blastomeres in most par-3 mutant embryos, SKN-1 localization appears normal in par-2 embryos (present at high levels only in P1 and its 4-cell and 8-cell stage descendants). As in par-1 embryos, mislocalization of SKN-1 to AB in par-3 mutants correlates with the ecopic production of skn-1-dependent pharyngeal cells after laser ablation of P1. Also as in par-1 embryos, glp-1 function appears unnecessary for the production of excess pharyngeal cells in par-3 embryos. In contrast, AB does not produce pharyngeal cells in par-2 mutant embryos if P1 is killed. However, if one waits until the 8-cell stage to kill P1 descendants, AB descendants do produce pharyngeal cells, suggesting that some pharyngeal cells in par-2 embryos are made in response to inductive signaling from P1 descendants. Consistent with this hypothesis, AB descendants fail to produce any pharyngeal cells in par-2 glp-1 double mutant embryos even if one waits until the 8-cell stage to kill P1 descendants. Finally, while par-1 embryos never, and par-3 mutants only rarely, produce intestinal cells, par-2 mutant embryos often produce extra intestinal cells. The production of a pharyngeal and intestinal cell excess by par-2 mutant embryos resembles the phenotype of pie-1 mutant embryos. In pie-1 embryos, both P2 and EMS (instead of just EMS) produce pharyngeal and intestinal cells (Mello et al., Cell 70, 163-76, 1993) and both are capable of signaling AB descendants to produce pharyngeal cells (Mello et al., 1993; Mango et al., Development 120, 2305-15, 1994). Further supporting this idea, both daughters of P1 in par-2 embryos typically divide synchronously but slightly later than do the two daughters of AB (similar in timing to the division of EMS in wild- type embryos). We are currently conducting further studies to determine if P2 and EMS adopt identical fates in most par-2 mutant embryos, as is usually the case in pie-1 mutant embryos. In summary, these studies permit the following conclusions. First, localization of SKN-1 protein to posterior blastomeres is not coupled to proper positioning of the first embryonic cleavage, or to proper localization of skn-1-independent body wall muscle pathway(s) at the first cleavage. Second, par-1 function is required for skn-1-independent body wall muscle pathway(s), but par-2 and par-3 are not. Third, glp-1 function is required for some of the extra pharyngeal cells made in par-2 embryos, but glp-1 appears not to be required for the production of extra pharyngeal cells in par-1 and par-3 mutant embryos. Finally, par-2 mutant embryos may esemble pie-1 mutant embryos in terms of cell fate patterning, although par-2 embryos exhibit early cleavage defects and P-granule localization abnormalities not seen in pie-1 mutant embryos. These studies indicate that while par-1, par-2, and par-3 all appear to be required for partioning events during early cleavages, different localization events and different developmental pathways are affected by mutations in each of these genes. While the loss of skn-1-independent body wall muscle in par-1 embryos, and the loss of glp-1-dependent pharyngeal cell production in par-1 and par-3 embryos, may be due to indirect effects of early cleavage abnormalities, it is also conceivable that some par gene products may have more direct roles in regulating developmental pathways in the early embryo.