The
bus-8 gene encodes a diverged homolog of the conserved bifunctional glycosyltransferase ALG2, which probably acts in N-linked glycosylation. As previously published (Partridge et al. 2008, PMID: 18395708),
bus-8 has essential functions in embryogenesis and in maintaining cuticle integrity, as well as affecting drug sensitivity, pathogen susceptibility, locomotion and mate recognition. Most
bus-8 transcripts contain two 5' exons (exons 1 and 2) that are out of frame with the rest of the coding sequence (exons 3-10). A missense mutation in exon 1,
lj22, fails to complement missense alleles in exons 5 and 10, and has similar phenotypic properties. This suggested that a long form of BUS-8 could be made from all ten exons by translational frameshifting. However, mass-spectrometry data obtained using purified C-terminally-tagged BUS-8 constructs revealed peptides only from exons 3-10 and none from exons 1 and 2, so translational frameshifting is unlikely. Moreover, transgenes expressing only exons 1 and 2 partially rescued the
lj22 mutation, but did not rescue other
bus-8 mutations. Therefore, it seems that exons 1 and 2 can act on their own, as a 72aa peptide, modifying the activity of the BUS-8 protein encoded by exons 3-10. Most
bus-8 mutants, including
lj22, are killed by Leucobacter Verde1, unlike wildtype worms; this sensitivity allows strong selection for suppressors. Verde1 selections carried out with different
bus-8 alleles have yielded both extragenic and intragenic suppressors, whose sequences and properties add further complexity. The unusual gene structure of
bus-8 seems to be conserved in other rhabditids at least, and
bus-8 homologs are found in more distant nematode genomes (Brugia, Onchocerca). Why nematodes apparently need both
bus-8 and a less diverged ALG2 homolog (
algn-2 in C. elegans) is mysterious, as is the role of the regulatory peptide.