We have been studying
unc-86 because mutations in this gene alter developmental decisions in several neural cell lineages (Chalfie et al. , Cell 24, 59-69; Finney and Horvitz, WBG vol. 7 no. 2). A minority of alleles (seven of 34) cause, in addition to lineage abnormalities, a low-level (<2%) Him phenotype and a defect in male mating. We would like to determine how the latter phenotypes are related to the former. In the course of constructing strains for other experiments, we made a double between the
unc-86 1416 and
him-5(
e1490). Animals of this strain appeared healthy but produced large numbers of dead eggs (76% inviable zygotes at 20 C;
e1490 by itself produces 20-30% dead eggs and
e1416 produces very few). An interaction between the
unc-86 e and
him-5 was suspected. We constructed all doubles between four
unc-86 alleles (three Him:
e1416,
n306, and
n843; one non-Him:
e1507) and two
him-5 alleles (
e1490 and
e1467). All
him-5 doubles with
unc-86 produced mostly inviable zygotes, while the
e1507 doubles did not. Several lines of evidence suggest that the reduced viability of
e1416:
e1490 is caused by increased autosomal nondisjunction. First, the lethality has a maternal effect -- there is no decreased viability of
e1416:
e1490 progeny of an
e1416 +/+
unc-32(
e189);
e1490 parent. This result implies that an
e1416:
e1490 parent produces defective sperm or ova. Second, the lethality can be partly but not completely rescued by mating with wild-type males (40-50 inviable zygotes at 20 C in broods in which the surviving animals were mostly cross progeny), implying that both sperm and ova are defective. Third,
e1416:
e1490 produces a few progeny likely to be abnormal in chromosome number. Some of these are almost certainly polyploids -- Long animals that segregate both Long and WT-length progeny. Others may be aneuploids -- animals that appear sick, grow slowly, and produce few or no eggs: some of these eggs hatch into healthy-appearing individuals. Direct confirmation of the production of aneuploid gametes is easiest if males mate, but
e1416;
e1490 males do not. The original characterization of him mutations (Hodgkin et al., Genetics 91, 67-94) showed that mutations in some genes (e.g.
him-5) cause a higher incidence of males than of inviable zygotes, indicating preferential nondisjunction of the X chromosome.
unc-86(
e1416), however, produced 1.1% males and 3.5% dead eggs, suggesting that low- level nondisjunction of all chromosomes may be occurring.
him-5 may enhance this phenotype of
e1416. Lethality in combination with other him mutations is not a general feature of
him-5 since
him-1(
e879) is quite viable in a double with
e1467 (ibid.). The above data establish that some
unc-86 alleles affect processes other than developmental decisions in neural lineages. Two possible models that may explain the data are 1) the Him alleles are small deletions that reduce or eliminate the function of an adjacent gene, and 2) the Him alleles are the most severe
unc-86 alleles. The second model implies that the same gene product is involved in chromosome segregation and the determination of cell fate.