lin-29 is considered a retarded heterochronic mutant because its L4 ( and later) seam cells appear to assume an L3 fate. In
lin-29 mutants, development occurs normally until just prior to the L4 molt, at which point the seam cells, which normally differentiate, continue to divide. Concomitantly, supernumerary molts occur, and larval cuticles are synthesized rather than an adult cuticle. Thus, animals go through an 'L5' stage, and often an 'L6.' The stage- and tissue-specificity of the defects caused by apparent
lin-29 null alleles suggests that lin- 29 is specifically required for cessation of molting, cessation of seam cell division and the switch from larval to adult cuticle synthesis. In addition to this role in wild-type development, epistasis studies between mutations of
lin-29, and the heterochronic genes
lin-14 and
lin-28, show that
lin-29 function is required for the expression of adult seam cell fates in precocious and retarded development, and suggest that these genes act via
lin-29 to specify the normal timing of the switch from larval to adult hypodermal cell fate. Unlike the phenotype of other heterochronic mutations,
lin-29 defects are not suppressed by development from dauer larvae. Thus, the larval-adult switch appears to require
lin-29 during both non-dauer and post-dauer development. To further elucidate the nature of
lin-29's action in the larval- adult hypodermal switch, we have performed a screen for mutations that either revert the
lin-29 phenotype to wild-type, or 'reverse' it, causing an epistatic precocious adult cuticle formation. Since we were interested in isolating both revertants and 'reversants,' the screen was designed to detect the formation of adult cuticle at any stage. The phenotypic expression of
rol-1(
e91), an adult-specific roller, can be regulated by heterochronic genes. Precocious mutants begin to roll at the first stage in which they produce adult cuticle, while the
lin-29 utant never rolls because
lin-29 mutants fail to make adult cuticle. Among approximately 25,000 progeny of EMS mutagenized
lin-29(
n546)
rol-1(
e91) hermaphrodites, nine independent revertants were isolated containing suppressor mutations. They are all recessive, autosomal, unlinked to
lin-29, and together define four complementation groups:
mab-1,
mab-13, and
mab-14 (see Hodgkin, Papp, and Ambros, this issue). They are allele specific to
lin-29(
n546) in that they do not suppress the
lin-29 phenotype of the 4 other existing lin- 29 alleles, even when these alleles are in trans to
n546. Since
lin-29 is a recessive mutation, one might expect that a single copy of
lin-29(
n546) in a homozygous suppressor background would be sufficient to alleviate or reduce
lin-29 defects, even if the other allele was
lin-29 null. Thus, the fact that animals homozygous for suppressor, and heterozygous for
n546 and other
lin-29 alleles, are not suppressed presents a curious situation. Either all of the other alleles produce some neomorphic product which poisons
n546 suppression, or some critical level of
lin-29 activity is required for the wild- type phenotype. Clearly a single wild-type
lin-29 allele is adequate, but perhaps a single suppressed
n546 is not. To test this possibility, we put
n546 over mnDF87, a lethal deficiency which covers
lin-29, to see if this combination is suppressible. These heterozygotes have much lower viability than
n546 homozygotes, but we can say that no suppressed heterozygotes have been seen so far. This implies that one suppressed
n546 allele produces some low level of
lin-29 activity insufficient for the larval-adult switch, but that two suppressed copies are sufficient for complete switching. In each genetic background described above, animals have either the full
lin-29 phenotype, or are fully wild-type for the larval-adult switch. This is in contrast to certain
lin-14 mutations which produce animals with patches of both larval and adult cuticle (showing that seam cells have the ability to differentiate independently), and to the incomplete and variable suppression of
tra-2 by members of the same set of mab loci ( J. Hodgkin, personal communication). It is consistent with a model in which a very quantized
lin-29 activity is required for the switch in cell fate.