The muscle phenotype of Unc-98 mutant animals, when viewed by polarized light, was first described by Zengel and Epstein (1980) as having poorly organized myofibrils and birefringent "needle-like" structures at the ends of muscle cells. Consistent with these findings, examination of these animals by EM revealed shortened M-lines and broken dense bodies (Z-line analogs). The needle-like structures, originally thought to be accumulations of thin filaments by EM, stained with neither phalloidin (actin staining) nor antibodies to myosin. In addition, immunofluorescence with antibodies to alpha-actinin and vinculin revealed a less regular staining of dense bodies as compared to wild type. The original mutant allele of
unc-98, (
su130), was recovered as motility defective. Though not obvious by casual observation, this mutant does have slower motility when compared to w.t. (p<.0001) in a sensitive, liquid motility assay. We have since recovered two additional alleles of
unc-98 (
sf19, sfDf1), both with motility defects. We began genetic mapping using deficiencies, duplications, and 3-factor mapping. After placing
unc-98 to a narrow interval between
dpy-7 and
unc-18, we tested for transgenic rescue of
unc-98 using cosmids in the region. We obtained rescue of the mutant muscle phenotype with cosmid F08C6 and then with an 8kb PCR fragment predicted to contain just a single gene, F08C6.7. The cDNA sequence of FO8C6.7 revealed a gene of 7 exons, which translates into a protein of 310 amino acids. Double stranded RNA from the full-length cDNA was injected to produce RNAi resulting in a phenotype identical to that of the existing mutants. Sequencing of two alleles revealed that both are 3' intron splice acceptor site mutations (replacing AG with AA) that may result in aberrant splicing products but may also produce simply lower levels of the full-length protein. By western analysis we see antibody binding to an UNC-98 protein present in w.t. but clearly absent in the mutants. We can conclude that these mutants are loss of function, producing significantly lower levels of the full-length UNC-98 protein, if any at all. By genetic criteria, our worst allele
sf19 is not null; the motility phenotype worsens when placed over a deficiency. The UNC-98 protein contains four C2H2 zinc finger domains and predicted nuclear localization and export signals (primarily at the N-terminus). Transgenic lines containing an UNC-98::GFP fusion protein (full length UNC-98 plus promoter) were created to determine UNC-98 localization. UNC-98 localizes to M-line and dense bodies, but also localizes to the nucleus, excluding the nucleolus. Immunofluorescence with antibodies to UNC-98 resulted in prominent M-line staining but undetectable levels of staining in either the dense body or nucleus. The absence of antibody staining, likely due to antigen masking or lower protein levels in these areas of the cell, was recovered by utilizing a transgenically rescued line with higher expression levels of the UNC-98 protein. To date, there has only been one other muscle specific protein found in both the myofibrils and the nucleus - UNC-97, a LIM domain protein (Hobert et al, 1999). Remarkably, UNC-98 showed strong interaction with UNC-97 when used as bait to screen a yeast 2-hybrid "bookshelf" of known dense body and M-line components. Upon further examination, we found this interaction to require the first two LIM domains of UNC-97 and all four zinc fingers from UNC-98. Curious as to how important it was to have all zinc fingers for proper localization of UNC-98, we developed four GFP constructs each missing one, two, three, or four zinc fingers. Individual constructs were injected and resulting transgenic lines examined for localization. Interestingly, in all cases, nuclear localization was maintained, dense body localization was less prominent, and previously prominent M-line localization was now diffusely localized throughout the A-band. Our working hypothesis is that UNC-98 functions both as a focal adhesion protein and a transcription factor, or acts together with transcription factors, to influence gene expression. We plan to 1) conduct microarray experiments to determine which genes might be regulated by UNC-98, 2) examine the role of the NLS and NES's, 3) further examine the UNC-98-UNC-97 interactions, and 4) obtain additional alleles of
unc-98.