The precocious heterochronic mutation
lin-28(
n719) possesses an unexpected phenotype: in hermaphrodites and males, an ectopic ray neuroblast lineage is frequently found in place of a postdeirid ( figure l; Ambros and Horvitz, Science 226, 409). How can a ray, a male- specific structure, form in the temporally transformed
lin-28 hermaphrodite? Kenyon and Hedgecock (WBG 8 #2, 19) have proposed that the fate of neuroblasts is influenced by the developmental stage in which they arise: the neuroblast in the postdeirid position is normally formed in L2, and therefore follows a postdeirid fate, but the equivalent
lin-28 neuroblast finds itself in a later temporal environment, becoming a ray instead. Consistent with this hypothesis, I have found that the ectopic ray does not result from a general sexual transformation in
lin-28 hermaphrodites, because it can still form in
lin-28(
n719);
tra-1dom(
e1575)/+ XX and XO females (4/12 sides examined by DAPI staining had the ectopic ray). The
lin-28 ectopic ray can be used to test whether
mab-5 or
mab-3, two genes involved in male development, are required for the expression of a ray neuroblast lineage. The null mutation
mab-5(
e1239) results in ray precursor cells failing to form in the V5 and V6 lineages (Kenyon, Cell 46, 477); the null mutation
mab-3(
e1240) causes ray precursor sublineages to produce seam cells instead of rays (at least for V5 and V6 descendants; WBG 9 #2, 62). Compared with
lin-28 alone (33/82 sides examined = 40%), the frequency of ectopic ray formation is not significantly altered in
mab-5;
lin-28 (22/68 = 32%) or in
mab-3;
lin-28 (19/67 = 28%) hermaphrodites or males, as assayed by DAPI staining. It is therefore likely that the
mab-5 and
mab-3 gene products are not necessary for a ray neuroblast lineage (unless the
lin-28 ectopic ray is abnormal, or behaves as a T-derived ray). In contrast, the gene
lin-32 does appear to be required for both wild- type and
lin-28 ray neuroblast lineages (0/15
lin-28;
lin-32 sides had ectopic rays; Kenyon and Hedgecock, ibid; C. Kenyon, personal communication).
mab-5, s each affect a different step in the sequence of divisions specific to the male lateral hypodermis, all three causing seam cells to be made instead of rays. It is interesting to speculate that their gene products may be involved in a series of decisions which switches these lineages from a hermaphroditic seam cell 'ground state' to the specialized male ray sublineages (figure 2). In addition,
tra-1 might interact negatively with
mab-5, and perhaps
mab-3, but not
lin-32, to control these sexually dimorphic decisions (based on epistasis experiments). In this sequential switch model,
mab-5 and
mab-3 products are not directly involved in specifying a ray neuroblast lineage, consistent with the observations above. This reasoning supports the neuroblast fate hypothesis of Kenyon and Hedgecock in explaining how the
lin-28 ectopic ray could arise in the presence of
tra-1 activity, and argues against a role for
lin-28 in the generation of sexual dimorphism, as has been suggested (Hodgkin, JEEM 83 Supp., 103). Furthermore, these speculations provide a framework for understanding how the selector gene
tra-1 might control 'downstream' functions to implement sexually dimorphic decisions at the level of individual cell fates. {Figure 1}