Figure 9

(A) Co-immunoprecipitation with MANSMA1 anti-SMN mAb shows that SMN and gemin5 exist as complexes in cytoplasmic extracts, but not in nuclear extracts (arrows show lanes to be compared). SMN and gemin4 controls are pulled down from both extracts equally. (B) Addition of KCl to the cytoplasmic extract to the same concentration as nuclear extraction buffer did not disrupt the SMN-gemin5 interaction. (C) All gemins in SMN complexes were accessible to appropriate antibodies. All anti-gemin antibodies pulled down SMN from total RIPA extracts of HeLa cells, except GEM5G which only recognizes denatured gemin5 on western blots (this also acts as a negative control). Antibodies were (left to right; see Table 1): No mAb (input control), GEM6B, GEM7B, GEM5G, GEM5P, GEM4C, GEM4D, MANSIP1B, MANSIP1A, rabbit anti-gemin3 and MANSMA1. In each case, 0.05 ml of Dynabead magnetic beads (Dynal, Oslo, Cat. No. 100.41), with anti-mouse Ig (or anti-rabbit Ig) attached covalently, were incubated with 0.1 ml of undiluted mAb culture supernatant (or 1/100 dilution antiserum) for 1 h at 4°C, washed 3× with PBS containing 0.1% BSA, and then incubated for 16 h at 4°C with 0.08 ml of HeLa extract (also sampled as "input"). After removing the "unbound" extract, the beads were washed 5× with PBS and boiled in 0.02 ml of SDS sample buffer. Gels (10% or 12.5% polyacrylamide) were loaded with 0.01 ml of SDS extract for SDS-PAGE and western blotting with anti-SMN mAb MANSMA12, as described in Methods. All lanes loaded with SDS extracts of beads contain a 50 KDa band of mouse Ig heavy chain which reacts with the HRP anti-mouse Ig used to develop the blot (band is absent from input and when rabbit antisera are used on beads).