Worm Breeder's Gazette 11(5): 90
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
We previously described the effect of smg mutations on the accumulation of two unc-54 mutant mRNAs (WBG 11(1):54). Both of these mutations (an out-of-frame deletion and an amber mutation) accumulate reduced levels of unc-54 mRNA in smg(+) backgrounds and approximately wild-type levels of unc-54 mRNA in smg mutant backgrounds. To better characterize the effect of smg mutations on mRNA accumulation, we analyzed a number of additional nonsense mutations affecting unc-54. Using RNase protection assays, we measured the steady-state level of mRNA in 14 sequenced unc-54 nonsense mutations in both smg(+) and smg(- ) backgrounds. All three classes of nonsense mutations were represented (UAA, UGA, and UAG). The unc-54 gene is 1966 codons in length; the nonsense mutations we analyzed are distributed along the length of the gene from codon 420 (unc-54(r316)) to codon 1906 (unc-54( e1300)). All fourteen nonsense mutations accumulate low levels of unc- 54 mRNA in smg(+) backgrounds. For twelve of the fourteen, the amount of steady-state mRNA is 5-10% of wild type. The two exceptions are the two mutations closest to the unc-54 3' end. e1328 (codon 1863) and e1300 (codon 1906) accumulate 20-30% of wild-type levels of unc-54 mRNA We constructed all unc-54(nonsense);smg double mutants, using one allele each of smg-1 through smg-6. We have analyzed most of the 84 double mutant combinations. In no case is the unc-54 mutation phenotypically suppressed by the smg mutation, but in all cases we observe a striking effect on the steady-state levels of unc-54 mRNA. Near wildtype amounts of unc-54 mRNA are present in all strains. The amounts of unc-54 mRNA average about 82% of wild type for smg-1 through smg-5, and about 46% of wild type for smg-6. Based on genetic observations described in an accompanying WBG article (this issue, Pulak et al.), we suspect that smg-6(r896), the allele used, is a weak allele of smg-6. We conclude that all unc-54 nonsense mutations accumulate low levels of mRNA. This effect is not uniform along the entire length of unc-54. Nonsense mutations near the unc-54 3' end are less affected and accumulate greater amounts of mRNA. In smg mutant backgrounds, the mRNA levels are nearly wild type. smg-6 is an exception only in the degree to which it affects the message level. These data suggest that the smg gene system may be involved in a cell surveillance process that minimizes the potentially disruptive effects of errors of transcription or mRNA processing. mRNAs that are recognized as being aberrant (due to premature translation termination) are degraded by the wild-type products of the smg genes. This system may protect cells from potentially toxic effects of truncated gene products (see accompanying WBG article).