mRNAs that contain premature translational termination codons are unstable in most eukaryotic cells. For example, the steady-state level of
unc-54 nonsense-containing mRNA is only 5-2096 of that found in wild-type
unc-54 mRNA (Pulak, 1991 C. elegans abstracts, pg. 134). The six smg genes are necessary for the instability of nonsense mutation-containing mRNAs. Mutations in any of the smg genes result in increased accumulation of nonsense-containing mRNAs. We think that the smg genes constitute all or part of a system for the removal of potentially harmful mRNAs. Such mRNAs may arise via errors of mRNA synthesis and/or processing. We are cloning
smg-2 in order to understand the nature and action of this system. We are attempting to clone the
smg-2 gene by transposon tagging. Suppression of
unc-54 (
r293)is a strong screen for smg(-) mutations. Consequently, we have isolated many spontaneous (TR679-derived) and EMS-induced alleles of
smg-2 .One particular TR679 -derivedallele,
smg-2 (
r920),reverts spontaneously to wild-type at high frequency.
smg-2 (
r920)was backcrossed and recombined extensively with Bristol strains to yield multiple, independently derived,
smg-2 (
r920)lines. We hybridized Southern blots of these
r920 mutant lines, three independent revertants of
r920 ,and wild-type (N2) with probes for the known C. elegans transposable elements. The copy numbers of all transposons, including Tc1 ,are very low in the backcrossed
r920 lines. One particular Tc4 element is present in all
r920 -containinglines, but is absent from N2 and from three independent revertants of
r920 .We believe that this Tc4 element is responsible for the
r920 mutation. The Tc4 element believed to be responsible for the
r920 mutation is contained within a 4.0 kb Sac I fragment. We have clone size-selected (3.8 4.3 kb) DNA from a Sac I digest of
r920 DNA into a plasmid vector. Only two Tc4 elements should be represented in this mini-library; the putative
r920::Tc4 and an additional Tc4 that was contained in the size-selected range. We have identified 13 colonies that hybridize to Tc4 ,at least two of which appear to correspond to the putative
r920::Tc4 insert. We are currently testing these Tc4 -containingclones to see if they contain sequences flanking Tc4 that are altered by other
smg-2 mutations. We will then use this flanking sequence to isolate cosmid or lambda clones from the region, and attempt and rescue the
smg-2 mutant phenotype by microinjection.