The genetic properties of smg mutants suggest that most existing alleles are loss of function mutations (Genetics 123:301). However, two observations make these arguments less compelling for the existing alleles of
smg-6 than for alleles of other smg genes: (1)
smg-6 alleles were isolated at a relatively low frequency. Of the 54 EMS-induced smg mutationsoriginally isolated, only 2 were alleles of
smg-6 (Genetics 123:301). Furthermore, none of 33
mut-2 (
r459)-inducedsmg mutations were alleles of
smg-6 .(2)
smg-6 mutations have a less pronounced effect on nonsense mutant mRNA accumulation than mutations in
smg-1 through
smg-5 .
unc-54 nonsense mutant mRNAs accumulate to only 5% of WT levels in smg(+) backgrounds. Mutations in
smg-1 through
smg-5 restore
unc-54 nonsense mutant mRNAs levels to approximately 100% of WT. These same mutant mRNAs accumulate to only 50-60% of WT levels in
smg-6 backgrounds (WBG 11(5):90). These observations suggested that the 2 existing
smg-6 alleles are hypomorphic, and that a
smg-6 null mutant might have a more severe phenotype that precluded its isolation in previous screens. We theerefore undertook a non-complementation screen to isolate new alleles of
smg-6 .Mutations in any of the smg genes suppress
unc-54 (
r293)(Genetics 123:301). Using suppression of
unc-54 (
r293)as a screen for new
smg-6 alleles, we isolated 8 EMSintucet alleles at a frequency of 7 X 10 -4 per mutagenized gamete. This forward mutation frequency is similar to that for an average-sized gene in C. elegans. These alleles were isolated by their failure to complement
smg-6 (
r896),and animals heteroallelic for
r896 and any of the newly isolated alleles are phenotypically indistinguishable from
r896 homozygotes (i.e.-hermaphrodites have protruding vulvae and males have abnormal tails). Five alleles are homozygous viable aod indistinguishable from the previously isolatet
smg-6 alleles. Three alleles (
r1014 ,
r1059 ,
r1062 )are recessive lethals.
r1059 is an embryonic lethal.
r1062 and
r1014 are early and late larval lethals, respectively. Both
r1059 and
r1062 fail to complement
r1014 for its late larval arrest phenotype. The lethality is tightly linked to the suppression phenotype of
smg-6 in all three cases (p<0. 1 m.u.). These data indicate that
smg-6 may be an essential gene, and that animals lacking
smg-6 activity arrest early in development. The different arrest phenotypes of these lethal alleles suggests they represent an allelic series,
r1059 being the strongest allele. We have not completely excluded the possibility that the lethality of these alleles is unrelated to the
smg-6 phenotype and results from very tightly linked mutations or small deficiencies affecting nearby essential gene(s). However, as EMS is primarily a point mutagen and as the lethality associated with these alleles maps within 0.1 m.u. of
smg-6 ,this possibility seems unlikely. The cloning of
smg-6 and subsequent molecular analysis of these alleles will differentiate between these possibilities.