Figure 7. Phosphorylation of GIV at S245 increases its ability to bind α-tubulin.
(A) Bacterially expressed and purified His-GIV-NT WT were phosphorylated (+) or not (-) with recombinant AMPK heterotrimers (α/β/γ) in vitro prior to their use in binding assays with GST-α-Tubulin-CT or GST-β-Tubulin-CT immobilized on glutathione-agarose beads. Bound proteins were analyzed for GIV-NT (anti-His) by immunoblotting (IB). His-GIV-NT(1–440) binds GST-tagged carboxyl terminus of α-Tubulin exclusively after phosphorylation by AMPK. No binding is seen with β-Tubulin regardless of phosphorylation. (B) WT and non-phosphorylatable SA mutant His-GIV-NT proteins were phosphorylated (+) or not (-) by recombinant AMPK as in A prior to use in binding assays with GST-tagged carboxyl terminus of α-Tubulin. Bound proteins were analyzed for GIV (His) by immunoblotting (IB). (C–D) Tubulin cosedimentation assays (see Materials and methods) were carried out with equal aliquots of pre-cleared lysates of Cos7 cells that were grown in the presence (G) or absence (NG) of glucose for 18 hr. Samples were centrifuged at high speed to separate the microtubule polymer (P; pellet) from the soluble tubulin (S; supernatant), resolved by SDS-PAGE, and analyzed for the presence of GIV and tubulin by immunoblotting (IB). The ratio of GIV in pellet vs supernatant was quantified by band densitometry and displayed as bar graphs in D. Error bars represent mean ± S.E.M; n = 3; *p<0.05. (E–F) Tubulin cosedimentation assays were carried out with equal aliquots of pre-cleared lysates of Cos7 cells expressing either vector control, or FLAG tagged GIV-WT or SD mutant as in C, and analyzed for the presence of GIV, GAPDH (negative control) and tubulin by immunoblotting (IB). The ratio of GIV in pellet vs supernatant was quantified by band densitometry and displayed as bar graphs in F. Error bars represent mean ± S.E.M; n = 3; **p<0.01. (G) Schematic summarizing the role of GIV in the regulation of cell-cell junction stability during an energetic stress and illustrating how the current findings relate to prior work. Exposure of epithelial cells to conditions that induce energetic stress result in depletion of cellular ATP stores and accumulation of AMP (step 1); the latter activates AMPK kinase (step 2). Once activated, AMPK phosphorylates GIV at S245 (step 3) triggering its localization to the cell-cell junction (TJs) via increased ability to bind TJ-associated microtubules (Lee et al., 2007) (step 4). Once localized to the cell-cell junctions, GIV has been shown (Houssin et al., 2015) to bind AJ-localized protein complexes, e.g., α- and β-Catenins and E-cadherin and links the catenin-cadherin complexes to the actin cytoskeleton (steps 5 and 8). GIV has also been shown to bind TJ proteins, e.g., aPKC/Par3/Par6 complex (Ohara et al., 2012) (step 6), and link these proteins to G proteins and the actin cytoskeleton (Sasaki et al., 2015) (steps 7 and 8).
Figure 7—figure supplement 1. The carboxyl terminus of α-tubulin, but not β-tubulin binds full length GIV from lysates of cells exposed to energetic stress by glucose starvation.
