The
lin-32 gene functions in a switch between certain hypodermal and neuronal cell fates; mutation of the
lin-32 gene causes the Q, postdeirid, and ray neuroblast cells to adopt hypodermal fates. As a result,
lin-32 mutant males lack rays.
mab-5 ,
mab-3 ,
lin-22 ,
pal-1 ,and
egl-5 affect V rays; while
mab-19 causes loss of T rays.
lin-32 is unique among these genes in causing both V and T ray loss. Furthermore,
lin-32 acts downstream of the control gene
lin-22 (1). Therefore,
lin-32 might function in the initiation of the ray sublineage program, while the other genes control
lin-32 expression. We observed that in
tra-1 (
e1099);Dp+ males (where Df is eT2 ,covering
lin-32 )ray 1 was lost 50% of the time, indicating that the
lin-32 locus is haplo-insufficient for generating ray 1. The strong allele of
lin-32 ,
u282 ,behaves similarly, causing 40% ray 1 loss in
tra-1 (
e1099);
lin-32 (
u282)/+heterozygous XX males. Therefore,
u282 may be close to a null allele (although we know that it may not be a null since
u282 /Dfhas never been seen and therefore is probably lethal). All three alleles of
lin-32 show a graded extent of ray loss. Among V5 and V6 -derivedrays (
r1 -
r6 ),anterior rays are lost more frequently than posterior rays. Thus, there is a gradient of ray loss from anterior to posterior. In the T-derived rays (
r7 -
r9 ),this gradient is reversed. Statistical analysis revealed that the presence or absence of any ray is independent of the presence or absence of its neighbors. Furthermore, laser ablation has failed to demonstrate interactions between neighboring ray precursor cells. Thus, we propose that the probability that a ray precursor cell makes a ray depends solely on the
lin-32 activity in that cell. A simple model is that
lin-32 gene function in wild type increases from anterior to posterior among V rays, and the level of the gradient is lowered by the existing, presumably hypomorphic
lin-32 alleles. Since it has been shown that there is a gradient of
mab-5 activity (2), we performed the following experiments to test if the gradient of
mab-5 may establish the gradient of
lin-32 .We constructed the
mab-5 (
bx54);
lin-32 (
e1926)double mutant, and found that the double mutant had a much more severe ray loss phenotype than either single mutant. This synergistic effect suggests that
mab-5 and
lin-32 may function in the same pathway of ray development. Moreover, an extra copy of wild type
mab-5 carried on the free duplication sDp3 increased the number of rays in
lin-32 (
e1926);sDp3 males. We interpret this to mean that a higher level of
mab-5 (+)product increases the level of
lin-32 (+)activity in the hypomorphic background. Molecular characterization of
lin-32 will demonstrate if there is a gradient of
lin-32 protein or mRNA.