Worm Breeder's Gazette 12(5): 25 (February 1, 1993)

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.

More on smg-Dependent Dominance and the Role of Disruptive Polypeptide Fragments

Brian Cali, Phil Anderson

Laboratory of Genetics, University of Wisconsin-Madison

In smg(-) backgrounds, otherwise unstable nonsense mutant mRNAs accumulate to near WT levels (WBG 11(5):90). In addition specific alleles of a number of genes that are recessive (or only weakly dominant) in smg(+) backgrounds are dominant (or more strongly dominant) in smg(-) backgrounds. Pulalk and Anderson proposed that this "synthetic dominance" results from increased acacumulation of aberrant mRNAs in smg(-) backgrounds which, when translated, produce disruptive polypeptides (WBG 11(5):91).

We previausly reported that smg-dependent and smg-enhanced dominant mutations are common (WBG 12(2):98). Of 32 dominant mutations that we isolated in a smg(-) baclcground, 22 were either smg-dependent or smg-enhanced. We now know these 22 mutations represent alleles of 7 different genes. Nineteen are alleles of known genes ( dpy-5 [1 allele], unc-1 [1 allele], unc-20 [6 alleles], and unc-70 [11 alleles] ). We do not yet know whether the three remaining synthetic dominant mutations are alleles of known genes, but based on map position and phenotypic differences they are distinct from each other and from the other 19 mutations. By the same criteria, the 10 smg-independent dominant mutations represent alleles of 5-7 different genes. Thus, 7 of 12-14 genes represented in our collection have alleles whose dominant phenotypes are dependent upon, or enhanced i4 a smg(-) background. As suggested by the large number of alleles, unc-70 seems to be particularly susceptable to synthetic dominant mutations. Of five EMS-induced unc-70 alleles isolated independently in a screen for recessive lethals in D. Baillie's lab (Genetcs 129:735), three show synthetic dominance (strains kindly provided by D. Baillie).

Pulak, DeStasio and Anderson described four unc-54 nonsense mutations clustered near the head/rod junction of the myosin molecule that show smg-dependent dominance (WBG 11(5):91). Two models seem most reasonable to explain the smg-dependent dominance of these alleles: (1) Increased mutant mRNA levels allow increased producdon of truncated myosin polypeptides that disrupt muscle assembly and/or function. These fragments are generated either by termination of ribosomes at the premature STOP codon (to generate N-terminal myosin head fragments) or by in-frame reinitiation of ribosomes dowmstream of the premature STOP (to generate C-terminal myosin rod fragments). Efforts to detect these fragment polypeptides have been unsuccessful, suggesting that, if present, they are unstable (WBG 11(5):91). Evidence that mutant myosin polypeptides can be unstable yet disruptive has been reported by Besjovec and Anderson (Cell 60:133). (2) Increased mutant mRNA levels allow greater opportunity for read-through of the premature STOP codon. This produces full-length polypeptides with a single amino acid change at the site of the STOP codon. These mutant polypeptides are disruptive to muscle assembly and/or function. Jonathan Hodgkin has provided evidence that the suppression of tra-3 nonsense mutations in smg(-) backgrounds is due to increased levels of mutant mRNA allowing increased read-through of the premature STOP codon. In this case, read-through leads to functional, not disruptive, full-length product (WBG 12(1):49).

To differentiate between these models, we tested two unc-54 out-of-frame deletion mutations ( r95 and r325 ,118 bp and 470 bp deletions, respectively) for dominance in smg(-) backgrounds. The 5' breakpoints of both deletions are within the region delimited by the four synthetic dominant nonsense mutant alleles described above. These deletions can not encode full-length myosin molecules with a single amino acid substitution. They are deleted for coding sequenÏ and, if a STOP codon downstream of the deletion frameshift is mistranslated, the ribosome will be out-of-frame. However, they can encode either N-terminal or C-terminal fragments of approximately the same size as those encoded by the nonsense mutant alleles. We found that while both r95 and r325 are recessive in smg(+) backgrounds, they are both dominant in smg(-) backgrounds.

This result argues that the synthetic dominance of these unc-54 alleles is due to increased production of toxic polypeptide fragments from mutant mRNAs that accumulate in smg(-) backgrounds. Other examples of synthetic dominance are likely due to similar mechanisms. If so, the diversity of genes that have smg-dependent or smg-enhanced dominant alleles suggests that many processes are susceptible to disruption by truncated polypeptides. The aberrant mRNAs that encode the disruptive polypeptides described above were generated by mutation. It has been suggested that the smg gene products also degrade aberrant RNAs generated from errors in transcription or RNA processing (WBG I11 (5):90).Such errors are expected to occur more frequently than DNA lesions in vivo. The smg gene products may function to protect cells from toxic polypeptide fragments by degrading the mRNAs that encode them.