Inactivating mutations in a majority of the 17,000 protein-coding genes of C. elegans may result in very subtle phenotypes, at least under laboratory culture conditions. Some of these mutationally-silent genes may encode essential functions redundantly. We have screened for mutations that when homozygous by themselves result in no obvious phenotype but that are lethal when the
mec-8 gene is also homozygous mutant. In our screen, which is modeled after work in yeast (Bender and Pringle 1991, Mol. Cell. Biol. 11: 1295), we used a homozygous
mec-8(
u74) strain containing
mec-8(+) on an extrachromosomal array and looked for synthetic lethal mutants in which
mec-8(+) on the array had become essential for viability. We have identified five independent mutations that are synthetic lethal with
mec-8 and assigned the five mutations to the genes
sym-1 through
sym-4. Three of these genes map to the X:
sym-1 is 1.0 map unit left of
unc-3,
sym-3 maps between
dpy-8 and
unc-6, and
sym-4 (defined by two alleles) is about 2-4 map units left of
unc-3. The fourth gene,
sym-2, is on LGII between
rol-6 and
unc-4.
mec-8 encodes a protein with two RNA recognition motifs and affects the accumulation of certain alternatively spliced transcripts of
unc-52 (Lundquist et al. 1996, Development 122: 1601). The pleiotropic phenotype of
mec-8 mutants suggests that MEC-8 affects the processing of RNA from several target genes. We envisage three mechanisms by which a mutation could be synthetic lethal with a
mec-8 mutation. First, as is the case with certain alleles of
unc-52, the synthetic lethal mutation could occur in a target gene with alternatively processed transcripts where at least one class of transcripts is required for viability;
mec-8 mutation would affect certain transcripts, and the synthetic lethal mutation would affect others. Second, the synthetic lethal mutation could be in a gene that provides an RNA processing function overlapping that of MEC-8 such that the products of
mec-8(+) and sym(+) would each be capable of processing the transcripts of an essential target gene. Third, the synthetic lethal mutation could be in a gene that provides a function overlapping that of a gene whose transcripts are processed by MEC-8; in this case, a mutation in
mec-8 would lead to the loss of the target gene function, which would make the otherwise-redundant gene essential. We have identified
sym-1 as the Genefinder-predicted gene C44H4.3. Injection of a 4.4-kb BamHI/PstI restriction fragment containing C44H4.3 rescued the temperature-sensitive lethality of
mec-8(
u218ts);
sym-1 animals. A frameshift mutation introduced early in the predicted open reading frame of C44H4.3 abolished rescue. We identified our one
sym-1 mutant allele as a nonsense mutation that would shorten the SYM-1 polypeptide to about two-fifths wild-type length. SYM-1 is predicted to have an amino-terminal signal peptide, 15 contiguous leucine-rich repeats (LRRs, each about 24 amino acid residues long) and a threonine/glutamic acid-rich carboxyl terminus. LRRs have been implicated in protein-protein interactions in a broad range of proteins. A rescuing
sym-1::gfp construct is expressed in hypodermis and leads to secretion of GFP to the outer surface of the embryo and the extracellular space between the embryo and the eggshell. All
mec-8;
sym-1 embryos arrest at the twofold stage, with detached muscle quadrants.
sym-1 is synthetic lethal with all ten of the
mec-8 alleles we tested; homozygous
sym-1 by itself has no detectable effect on embryogenesis and depresses the average brood size only mildly. Neither northern blots nor cDNAs give any evidence for alternative splicing of
sym-1 transcripts. We favor the idea that
sym-1 provides a function that is redundant with that of a gene whose transcripts are processed by MEC-8.