The drug levamisole is a potent anthelminthic causing nematode muscles to hypercontract, which eventually leads to death. In C. elegans, mutants resistant to the cholinergic agonist levamisole define 9 genes postulated to encode both musde acetylcholine receptor ( AChR) subunits and, perhaps more interestingly, products required for the regulation and assembly of this receptor. Three of these genes (
lev-8,
lev-9 and
lev-10) may affect receptor function only indirectly, while the other six (lev-l,
unc-29,
unc-38,
unc-50,
unc-63 and
unc-74) have more direct effects. Two types of resistant mutants are seen: strong (receptor nulls) and weak (partial receptor function). One of these genes, lev-l, is distinct in two respects: the null phenotype is weak resistance, with strongly resistant mutants being much rarer; and these strongly resistant mutants are semidominant. Previous genetic analysis and binding studies using extracts from both classes of lev-l mutants have suggested that the lev-l product is not essential for the function of the receptor, but rather has a peripheral modifying role. Similar types of evidence has led to the notion that
unc-29 and
unc-38 may encode structural subunits, while
unc-50,
unc-63 and
unc-74 may be involved in receptor biogenesis. Using a combination of RFLP mapping and transposon tagging the genomic location of the lev-l gene has been established. The gene sequence has been determined from both genomic and cDNA clones, and surprisingly its putative product is a homologue of the non-ligand-binding (non-a) subunits of vertebrate and invertebrate AChRs. The gene has fewer introns than its vertebrate counterparts, but the positions of two introns have been conserved in nematode, insect and vertebrate genes. The
unc-29 and
unc-38 genes have also recently been cloned by transposon tagging (see Fleming et al. abstract) and both appear to encode structural subunits, in accord with the earlier predictions. A comparison of the three subunits with other known insect and vertebrate subunits shows that the
unc-29 and lev-l products are very similar to each other but not to anything else. The
unc-38 product is similarly very different from other a subunits. This is not consistent with subunit specialization predating phylogenetic divergence. That lev-l encodes a subunit leaves three questions: l) Why is weak resistance the null phenotype? 2) Why does only lev-l mutate to semidominance? 3) What is the basis of the semidominance? Reinterpretation of previous data concerning lev-l mutants leads to a model where the position in the receptor normally occupied by the lev-l product uniquely has fewer structural requirements for receptor assembly than do other subunits, which property may be intimately tied to the normal function of this subunit.