We described earlier the isolation of
ct31 as a gamma-ray-induced, semi-dominant, feminizing mutation that suppresses the masculinization of XX animals caused by the
her-1(gf) allele
n695, and reported that
ct31 increases the penetrance of the X-linked hypomorph
lin-15(
n765), suggesting that
ct31 might cause a decrease in X-chromosome expression (Burgess et al., WBG May, 1986).
ct31 XX animals are Dpy, Egl hermaphrodites; XO animals are Dpy, nonfertile, partially feminized males. Here we present evidence that
ct31 lowers expression of the X- linked
act-4 gene and that it is probably a gain-of-function mutation involving duplication of a region of the X chromosome. Assay of
act-4 mRNA in
ct31 and N2 L1 hermaphrodites using an RNA dot blot assay ( Donahue et al., PNAS 84:7600,1987) showed that the level of this transcript in
ct31 animals is only 46% + 4% of the level in N2 animals. This finding supports the genetic evidence cited above that
ct31 causes a decrease in expression of X-linked genes. Three-factor mapping data places
ct31 in the
lin-2 -
unc-9 interval on X. Deficiencies of this region have not been obtained, but it is covered by the duplication mnDp10. XX hermaphrodites of genotype mnDp10/+;
ct31 lin15, which have two mutant and one wild-type copy of the
ct31 locus, are more severely Dpy and Egl and generally sicker and more slow-growing than
ct31 homozygotes. When the former animals are allowed to self-fertilize, one quarter of the progeny arrest as L1 larvae. We surmise that these arrested larvae are the class carrying two copies of mnDp10 (
ct31/ct31/+/+). Since the
ct31 phenotype appears to be enhanced by additional copies of the wildtype locus, these results are consistent with
ct31 being a gain-of-function mutation. However, given the large size of mnDp10, and the fact that it carries at least one other locus known to affect X expression (sdc- 1), more definitive evidence on this point must await isolation and characterization of deficiencies in the region. The high reversion rate of
ct31 described earlier (Burgess et al., 1986) suggested that it might involve an unstable chromosomal rearrangement. In an attempt to test this suggestion, we used a probe derived from the cloned
myo-2 gene, which is located on the physical map at a position that should be near the
ct31 locus, to carry out quantitative Southern blot analyses. The copy number of the
myo-2 gene in the
ct31 strain is clearly higher than in N2 or in the
her-1(
n695) parent strain of
ct31. The elevated copy number of
myo-2 appears identical to that measured in a strain homozygous for mnDp10. Seven independent nonDpy
ct31 revertants do not show the elevated
myo-2 copy number, correlating the visible
ct31 phenotype and the apparent duplication. Preliminary experiments using as probes cosmid clones from the
myo-2 contig ( obtained from A. Coulson and J. Sulston) indicate that at least 40 kb including and to the left of the
myo-2 gene is duplicated in the
ct31 mutant. Experiments are in progress to characterize further the extent and nature of the rearrangement.