We have previously described the interactions between alleles of mei- 1(I) and the mutations
ct46(I) and
ct61(I) [Mains et al., WBG 10(1) :102]. The latter two mutations produce gain-of-function 'poison' gene products that result in dominant, temperature-sensitive maternal- effect lethality. The mutations are linked, but represent different loci separated by 2 cM. Mutations of
mei-1 show nonconditional recessive maternal-effect lethality resulting from a failure of meiosis; the canonical allele (
b284) was isolated in a screen for maternal-effect lethals [Sprunger and Kemphues, WBG 10(1):100]. This mutation is an efficient trans-dominant suppressor of both
ct46 and
ct61. Previously, we had isolated twelve additional alleles of
mei-1 as suppressors of
ct46. Here we show that
ct46 is an allele of
mei-1. One of these suppressor mutations,
ct99, was isolated in cis to
ct46. The eggs from
ct46ct99/+ + mothers show wild-type levels of hatching compared to l% for the eggs of
ct46/+ mothers.
ct46ct99 shows recessive nonconditional maternal effect lethality and fails to complement
mei-1(
b284).
ct46ct99/ct46+ shows no trans-suppression of
ct46. An additional suppressor mutation, isolated as
ct46ct101, shows identical behavior. Therefore,
ct101 and
ct99 are cis-dominant suppressors of
ct46 apparently resulting from intragenic loss-of- function reversion of the
ct46 gain-of-function poison product. Thus,
ct46 is an allele of
mei-1. Consistent with this interpretation, mutations in
mei-1 map <0.01 cM from
ct46. These revertants occurred at a high frequency: 2/5500 after EMS mutagenesis, again implying that they are loss-of-function mutations. While the above interpretation of the cis-dominant suppression of
ct46 by
ct99 and
ct101 is straightforward, the majority of
mei-1 alleles, for example
ct82, shows the surprising property of also being trans-dominant. As reported previously [Mains et al., op. cit.], 92% of the eggs from
ct46ct82/ct46+ mothers hatch. Twelve alleles of mei- 1 (defined by the failure to complement for the recessive maternal effect lethality) show this property, and the trans-suppression ranges from 28-92%. Like
ct99 and
ct101, these alleles appear to represent general loss-of-function rather than specific gain-of-function mutations that compensate for the
ct46 lesion. They were isolated at a very high frequency after EMS mutagenesis (11/5500); moreover, the canonical allele
b284, which was isolated in the absence of
ct46 (see above), is also an efficient trans-suppressor (77% of the eggs from
ct46+/+
b284 hatch). Unlike
ct99 and
ct101, these mutations are also trans-dominant suppressors of
ct61. How can we reconcile these two classes of apparent loss-of-function alleles of
mei-1, one which suppresses
ct46 only in cis and the other which can suppress
ct46 (and
ct61) in cis or trans? It seems likely that the cis-only suppressors (
ct99 and
ct101) are true null mutations (e.g., causing absence of all gene-product functions) while the trans suppression (
ct82,
b284, etc.) is mediated by the presence of a partially defective gene product. Deficiencies for the region (nDf23 and nDf24) show no trans-suppression, consistent with the interpretation that the cis only suppressors represent the null phenotype (although this interpretation is complicated by the fact that the deficiencies tested remove the
ct61 locus in addition to mei- 1). One possibility consistent with the evidence would be that a partially defective
mei-1 product can form a complex with the poison
ct46 (or
ct61) gene product and thereby inactivate it, whereas a completely defective (or missing)
mei-1 product resulting from a null mutation cannot. It should be noted that Kusch and Edgar [Genetics 113: 621 (1986)] saw a similar pattern of interactions for several genes affecting the cuticle (i.e. loss-of-function mutations, but not true nulls, could suppress dominant gain-of-function mutations). We may have also identified yet another class of
mei-1 alleles. The mutation
ct103(I) is a trans-dominant suppressor of both
ct46 and
ct61, but unlike the previously described
mei-1 alleles shows no recessive maternal-effect lethality. It complements other
mei-1 alleles. However, it maps between
lin-10 and
lin-28(I), the closest markers that flank
mei-1.Therefore, we have an allelic series of
mei-1: (1)
ct46, a dominant gain-of-function poison, (2)
ct99 and
ct101, putative null alleles that are cis-dominant suppressors of
ct46 and recessive maternal-effect lethals, (3)
ct82,
b284 (and 9 others), trans-dominant suppressors of
ct46 and
ct61 and recessive maternal effect lethals, and (4) possibly
ct103, a trans-dominant suppressor of
ct46 and
ct61 with no recessive phenotype. We have identified a mutation at another locus,
ct102(I), which is also a trans-dominant suppressor of both
ct46 and
ct61. It shows recessive maternal-effect lethality with defects that closely resemble those of
mei-1 mutants (Sprunger and Kemphues, op. cit.). However,
ct102 complements recessive alleles of
mei-1 and maps to a different interval, between
bli-4 and
unc-37. This locus will likely be designated '
mei-2'. Finally, a second mutation that maps to this same interval,
ct98, is also a trans-dominant suppressor of
ct46 and
ct61, but complements
ct102 and shows no recessive maternal-effect lethality. Therefore, this could be a viable allele of '
mei-2'.[See Figure 1]