Vertebrate adducin is an actin-binding protein which can promote assembly of a spectrin-actin complex and can also act as a barbed-end actin capping protein. Biochemical and biophysical characterization in our lab has shown that active adducin is probably a tetramer consisting of two alpha subunits tightly complexed with either beta or gamma subunits. Adducin binds calmodulin and is an in vivo substrate for protein kinases C and A, all of which modulate adducin's spectrin-actin binding activity in different ways in in vitro assays. Nevertheless, vertebrate adducin has never been definitively linked to any known mutant phenotype. However mutations in a Drosophila adducin homologue, hu-li tai shao (hts), result in a female sterile phenotype in which egg chambers have fewer nurse cells, with abnormal actin distribution in the ring canals. Here we report preliminary results of our attempt to develop C. elegans as a genetic model system for adducin. We made a database search for C. elegans adducin and originally identified two partial cDNAs from Yuji Kohara's EST sequences. The first clone,
yk26f2, which we call
add-1 (C. elegans Adducin 1), maps to cosmid F39C12 on the left arm of LGX, close to
lon-2. The 659 residue predicted protein shows significant homology to vertebrate adducin and hts in the head, neck and tail regions which suggests it may behave like the "canonical" adducins. Interestingly though,
yk120b7(
add-2) which lies on Cosmid F57F5 on the right arm of LGV, shows a novel structure. The C-terminal 261 residues show homology to the adducin "head" which normally lies at the N-terminus of other adducins. Instead Cad-2 has a novel 350 residue N-terminus that shows no homology in a BLAST search to known proteins. Cad-2 may use its adducin "head" to interact with Cad-1 in a heteromeric manner similar to that demonstrated for vertebrate adducins. But it also raises the exciting prospect that the adducin "head" domain may actually be a component of multiple proteins with different functions, in a manner analogous to that demonstrated for the kinesin and myosin head domains. We aim to resolve these possibilities using C. elegans as a model system.