Recessive, loss-of-function mutations in
smu-2 were previously isolated and characterized as extragenic suppressors of
mec-8(ts);
unc-52(ts) synthetic lethality. In addition, these mutants suppress several phenotypes associated with
mec-8 alleles and the Unc phenotype of
unc-52(viable) alleles that have nonsense mutations in exon 17. MEC-8 is an RNA binding protein that regulates the production of certain alternatively spliced
unc-52 transcripts. Our preliminary evidence suggests that
smu-2 also regulates splicing of
unc-52. RT-PCR experiments show that in
smu-2 mutants, the levels of some alternatively spliced
unc-52 transcripts that skip exon 17 increase relative to internal controls. Increased levels of these transcripts are thus able to compensate for the loss of
unc-52 transcripts resulting from mutation in exon 17. The genetic interactions among
mec-8,
unc-52 and
smu-2 suggest that
smu-2 may be directly involved in the splicing process. We have cloned the
smu-2 gene and find evidence consistent with this hypothesis.
smu-2 is the homologue of a ubiquitously expressed nuclear protein known as RED (or IK factor). Although the function of RED is unknown, it has been identified in purified human spliceosomes by mass spectrometry (Neubauer et al., Nature Genetics 20:46). SMU-2 contains a domain rich in arginine (R), serine (S), and aspartic acid (D) residues. These types of domains (RS and RD/RED domains) have been identified in several splicing factors and are proposed to mediate protein-protein interactions. Using a rescuing in-frame fusion construct, we find that SMU-2::GFP is present in the nucleus and is often concentrated in sub-nuclear structures. Similar nuclear "speckles" are observed for many mammalian splicing factors including SR proteins and snRNPs. Additionally,
smu-2::gfp is expressed at many stages of development and in many cell types. Interestingly, when we sequenced the
smu-2 alleles obtained in the suppressor screen, we found that none is a candidate null. In fact, we are able to rescue
smu-2 mutant phenotypes by transformation with
smu-2 mutant DNA, presumably by overexpression. Transformation with a
smu-2 null made in vitro does not rescue the
smu-2 mutant phenotypes. Currently we are addressing the issue of the null phenotype using RNA interference.