The KH domain protein MEX-3 is key to spatial and temporal control of PAL-1 expression in the C. elegans early embryo.
pal-1 encodes a Caudal-like homeodomain protein that specifies the fate of posterior blastomeres (3). While
pal-1 mRNA is present throughout the oocyte and early embryo, PAL-1 protein is only detected in posterior blastomeres starting at the 4-cell stage. This spatial and temporal patterning of PAL-1 is dependant on the
pal-1 3'UTR and the putative RNA binding protein MEX-3, whose localization correlates with PAL-1 repression (2). MEX-3 localization and activity is dependent on the early-acting par genes. In
par-1 and
par-4 mutants, MEX-3 in all cells is coincident with failure to express PAL-1 in any cell (1). In contrast, 4-cell
par-3 embryos express PAL-1 despite high levels of MEX-3 in all cells, indicating that MEX-3 is not sufficient to repress PAL-1 at the 4-cell stage. To identify co-factors that work with MEX-3 or regulatory factors that modulate MEX-3 activity, we conducted a two-hybrid screen using MEX-3 as bait. We identified two M EX-3 I nteracting P roteins (MIPs), MEX-6 and PIP-1, that regulate MEX-3 expression levels and control spatial but not temporal patterning of PAL-1.
mex-6 encodes a CCCH type zinc finger protein that shares high sequence identity and partial functional redundancy with
mex-5 (6). In fact,
mex-6 (RNAi) reduces the function of both
mex-5 and
mex-6 . Anteriorly localized MEX-5,6 prevent premature degradation of MEX-3 and repress anterior PAL-1 expression. ZC404.8 encodes an RNA recognition motif protein also identified as a POS-1 interacting protein,
pip-1 (5 ), and as a mutant,
spn-4 (4), that affects spindle rotation.
pip-1 is required to repress weak anterior PAL-1 expression starting at the 4-cell stage and for the timely degradation of MEX-3. Genetic analysis suggests that MEX-6 and PIP-1 define parallel pathways that control MEX-3 stability and activity. First, mip; par double mutants with
par-3 or
par-1 are similar to the mip(RNAi) single mutants, indicating that
par-3 and
par-1 function upstream of
mex-5,6 and
pip-1 . In contrast,
par-4 mutations affect the RNAi phenotype of both
mex-6 and
pip-1 , suggesting that
par-4 ,
mex-5,6 , and
pip-1 act in parallel pathways. In
pip-1(RNAi)
par-4 mutants, MEX-3 accumulates abnormally as in
pip-1 (RNAi) embryos, but like
par-4 mutants, PAL-1 is not expressed.
mex-6(RNAi);
par-4 embryos express PAL-1 like
mex-6 (RNAi) embryos, but contain stable MEX-3 like
par-4 embryos. In all cases in which
mex-5,6 are absent, PAL-1 is ectopically expressed, suggesting that
mex-5,6 are absolutely required for PAL-1 spatial repression. In all cases in which
pip-1 is absent, MEX-3 persists through the 16-cell stage, suggesting that
pip-1 is absolutely required for the timely degradation of MEX-3. In all cases in which
par-4 is absent, MEX-3 persists at least through the 8 cell stage, suggesting that
par-4 plays a role in MEX-3 inactivation and degradation. We suggest that
par-4 normally acts to both inactivate MEX-3 and mark it for
pip-1 -mediated degradation. In the anterior,
mex-5,6 activate MEX-3 and protect it from inactivation and degradation. Furthermore, we suggest that a small amount of inactive MEX-3 in the anterior of
pip-1 (RNAi) embryos, which is normally degraded, interferes with active MEX-3 , causing weak PAL-1 expression. From these results we propose that MEX-3 and the MIPs act through multiple independent pathways to translate the polarity information provided by the early acting par genes into the asymmetric distribution of PAL-1. 1. Bowerman, Ingram and Hunter (1997) Development 124, 3815-3826. 2. Draper, Mello, Bowerman, Hardin and Priess (1997) Cell 87, 205-216. 3. Hunter and Kenyon (1996) Cell 87, 217-226. 4. Gomes, Swan, Shelton and Bowerman (2000) West Coast Worm Meeting 5. Ogura and Kohara (1999) International Worm Meeting 6. Schubert, Lin, Plasterk and Priess (2000) Mol. Cell 5, 671-682.