Mutations in
smu-2 were isolated as extragenic suppressors of the synthetic lethal phenotype of
mec-8 unc-52(viable) double mutants. The
smu-2 mutations also suppress other phenes of
mec-8 mutants, such as mechanosensory and chemosensory defects, apparently by a bypass mechanism;
mec-8 encodes a putative RNA binding protein that affects the accumulation of certain alternatively-spliced transcripts of
unc-52 and other genes (Lundquist et al. Development 1996 122: 1601-1610). Finally,
smu-2 mutations suppress the uncoordination conferred by nonsense mutations in exon 17 but not exon 18 of
unc-52 . We hypothesized that
smu-2 encodes a factor that regulates the splicing of various target genes, at least some of which are also targets of splicing control by MEC-8. Indeed, our RT-PCR experiments indicate that mutation in
smu-2 leads to enhanced skipping of exon 17 but not exon 18 of
unc-52 pre-mRNA. We identified
smu-2 by positional cloning and transformation rescue. We rescued the
smu-2 mutant phenotype (recessive suppression of
unc-52 and
mec-8 ) with the YAC Y37F3 and with a long-range PCR product that contained the single gene Y49F6B.4. We also found that a construct containing 2 kb of the promoter region of Y49F6B.4 inserted 5' to a full length cDNA (YK563h8) rescued
smu-2(
mn416) and that a frameshift mutation introduced in the same construct abolished rescue. Finally, we identified sequence alterations in Y49F6B.4 associated with all three known
smu-2 mutations. Database searches showed that
smu-2 is 37% identical to a mammalian nuclear protein called RED. The similarities between these proteins occur throughout their full extents. SMU-2 is the only protein with significant similarity to RED in the C. elegans genomic sequence database. Neubauer et al. (Nat. Genet. 1998 20: 46-50) identified RED in purified human splicesomes using mass spectrometry and peptide sequencing. The stringent conditions under which the spliceosomes were isolated--proteins transiently associated with spliceosomes were not isolated--suggest that RED is tightly associated with the spliceosome. RED was named after the most distinctive feature of the protein, a domain consisting of alternating basic (arginine) and acidic residues (aspartic and glutamic acid). RED domains are similar to RD domains that consist mainly of arginine and glutamic acid di-peptide repeats. RD domains have been found in a number of proteins that associate with the spliceosome. The human RED domain has diverged from SMU-2's RED domain--SMU-2 contains far more RD di-peptide repeats, and in addition it contains five serine residues. Surprisingly, Drosophila and Arabidopsis homologues do not contain a RED domain at all, although there is high conservation in regions outside this domain. None of our three
smu-2 mutations is a molecular null.
mn416 is a 3' splice site mutation,
mn610 is a single base pair deletion in the last quarter of the protein, and
mn611 is a rearrangement with a breakpoint in the last quarter of the protein; thus, both
mn610 and
mn611 are predicted to make the N-terminal three-quarters of the protein, including the RED domain. We gained further evidence that these mutations are not null when we did transformation rescue experiments using long-range PCR products generated from mutant DNA templates: mutant DNA was able to rescue the
smu-2 phenotype, presumably as a consequence of overexpression. We did RNAi experiments to see if we could detect a more severe
smu-2 phenotype. When we injected
smu-2 dsRNA (including the RED domain) into
unc-52(ts) mutants, we mimicked the
smu-2 mutant phenotype: over 98% suppression of
unc-52 and no embryonic lethality. We have monitored expression of a rescuing
smu-2::gfp reporter construct: the SMU-2::GFP is a nuclear protein that is ubiquitously expressed at all stages of development.