Fig. 3. SAD-1 and NAB-1 interact in vitro and in vivo. (A) Yeast two-hybrid assays to determine the NAB-1-interacting domain of SAD-1. Upper left panel; schematic representation of SAD-1 deletions used to map the region for NAB-1 interaction. Lower left panel; Y274 yeast strain was cotransformed with a NAB-1-AD prey plasmid and various LexA-SAD-1-deletion bait plasmids, and tested for ß-gal activity on X-gal plates. Blue color indicates interaction. Right panels; yeast transformed with different bait- and prey-plasmid combinations (as indicated) were grown in trp^- leu^- SD media, and color development with X-gal allowed to occur. LexA-SAD-1 amino acids 280-914 (SAD-1) or LexA-SAD-1 amino acids 280-911 (SAD-1DKV) were used as bait and NAB-1 as prey. The lacZ reporter was expressed in only yeast strains carrying both full-length SAD-1 and NAB-1. (B) GST-pull-down assays showed that GST-NAB-1 (or the PDZ domain) selectively interacts with the 110 kD isoform of SAD-1. Upper panels; GST, GST-NAB-1 PDZ domain and GST-full-length NAB-1 precipitated FLAG-tagged SAD-1 from the total-protein lysate from C. elegans strains carrying an integrated, fully functional SAD-1::FLAG array. Lysate input and GST input are shown at the bottom. Lower panel; GST, GST-SAI-2 and GST-NAB-1 PDZ precipitated endogenous SAD-1 from wild-type C. elegans lysate. SAI-2, another SAD-1-interacting protein identified from the yeast two-hybrid screen, pulled-down both isoforms of SAD-1, whereas NAB-1 precipitated only the 110 kD isoform. (C) Co-immunoprecipitation experiments showed that SAD-1 and NAB-1 interact in vivo. C. elegans lysates prepared from wild type, hpIs66 or hpIs66; sad-1 were immunoprecipitated with either anti-SAD-1 or anti-GFP antibody (for NAB-1::GFP), probed with anti-GFP antibody and then stripped and re-probed with anti-SAD-1 antibody, or vice versa. (D) Two alternatively spliced forms of sad-1. Schematic representation of the two splice variants encoded by the sad-1 gene is shown. We sequenced all the existing cDNA clones of sad-1 and discovered that two clones (yk134f11 and yk238h3) in which an additional exon was present in the C-terminal region of the clone, which leads to an earlier stop than the predicted SAD-1 coding region. The corresponding amino acid sequence truncated in this short isoform is shown.