Inhibition of Contraction-Stimulated AMP-Activated Protein Kinase Inhibits Contraction-Stimulated Increases in PAS-TBC1D1 and Glucose Transport Without Altering PAS-AS160 in Rat Skeletal Muscle

OBJECTIVE

Phosphorylation of two members of the TBC1 domain family of proteins, Akt substrate of 160 kDa (AS160, also known as TBC1D4) and TBC1D1, has been implicated in the regulation of glucose transport in skeletal muscle. Insulin-stimulated phosphorylation (measured using the phospho-Akt substrate [PAS] antibody) of AS160 and TBC1D1 appears to occur in an Akt-dependent manner, but the kinases responsible for contraction-stimulated PAS-AS160 and PAS-TBC1D1 remain unclear. AMP-activated protein kinase (AMPK) and Akt, both activated by contraction, can each phosphorylate AS160 and TBC1D1 in cell-free assays.

RESEARCH DESIGN AND METHODS

To evaluate the roles of AMPK and Akt on insulin- or contraction-stimulated PAS-AS160, PAS-TBC1D1, and glucose transport, rat epitrochlearis was incubated with and without compound C (inhibitor of AMPK) or Wortmannin (inhibitor of phosphatidylinositol [PI] 3-kinase, which is upstream of Akt) before and during insulin stimulation or contraction.

RESULTS

Insulin-stimulated glucose transport and phosphorylation of both AS160 and TBC1D1 were completely inhibited by Wortmannin. Wortmannin eliminated contraction stimulation of phospho-Ser^21/9glycogen synthase kinase 3α/β (pGSK3; Akt substrate) and PAS-AS160 but did not significantly alter pAMPK, phospho-Ser⁷⁹acetyl CoA carboxylase (pACC; AMPK substrate), PAS-TBC1D1, or glucose transport in contraction-stimulated muscle. Compound C completely inhibited contraction-stimulated pACC and PAS-TBC1D1 and partially blocked glucose transport, but it did not significantly alter pAkt, pGSK3, or PAS-AS160.

CONCLUSIONS

These data suggest that 1) insulin stimulates glucose transport and phosphorylation of AS160 and TBC1D1 in a PI 3-kinase/Akt–dependent manner, 2) contraction stimulates PAS-AS160 (but not PAS-TBC1D1 or glucose transport) in a PI 3-kinase/Akt–dependent manner, and 3) contraction stimulates PAS-TBC1D1 and glucose transport (but not PAS-AS160) in an AMPK-dependent manner.Insulin and contractile activity, the two most important physiological stimuli that increase glucose transport in skeletal muscle, can each induce the translocation of GLUT4 glucose transporters from the cell''s interior to its surface membranes (1,2). However, they regulate glucose transport via distinct signaling pathways (3). Insulin-stimulated glucose transport requires phosphatidylinositol (PI) 3-kinase activation, which leads to Akt activation without stimulating AMP-activated protein kinase (AMPK) (3–6). A great deal of evidence suggests that contraction stimulates glucose transport by a mechanism independent of PI 3-kinase/Akt (7–10) and attributable to the effects of multiple inputs, with AMPK- and calcium-mediated processes being major factors (11,12).In 3T3-L1 adipocytes, insulin stimulates phosphorylation of Akt substrate of 160 kDa (AS160; also called TBC1D4) in an Akt-dependent manner on sites identifiable by the phospho-Akt substrate (PAS) antibody (13,14). AS160 includes a Rab GTPase-activating protein domain (RabGAP) that inhibits Rab proteins involved in regulating vesicular traffic (15). The insulin-mediated increase in PAS phosphorylation of AS160 (PAS-AS160) appears to inhibit RabGAP activity, thereby allowing GLUT4 to be recruited to surface membranes and elevate glucose transport (14–17). In skeletal muscle, insulin or contraction results in elevated PAS-AS160 (18,19), and AS160 phosphorylation appears to regulate glucose transport (20).Recently, TBC1D1, a RabGAP protein paralog to AS160, was also shown to become PAS-phosphorylated (PAS-TBC1D1) in response to insulin in an Akt-dependent manner (21). However, whereas AS160 knockdown in 3T3-L1 adipocytes resulted in elevated basal cell-surface GLUT4 (17,22), TBC1D1 knockdown had no effect on basal cell-surface GLUT4 in 3T3-L1 cells (23). TBC1D1 protein is only ∼5% as abundant as AS160 protein in 3T3-L1 adipocytes, which may explain why TBC1D1 does not appear to play a major role in regulating glucose transport in these cells (23). TBC1D1 protein abundance is much greater in skeletal muscle versus adipose tissue (24), and silencing TBC1D1 in L6 myotubes resulted in increased basal cell-surface GLUT4 (25), supporting the idea that TBC1D1 inhibits GLUT4 translocation in the basal state. However, in contrast to the results for L6 cells with AS160 knockdown (which did not alter the insulin-stimulated net increase in cell-surface GLUT4), silencing TBC1D1 in L6 cells resulted in greater insulin-induced GLUT4 translocation versus control cells (25). In other words, TBC1D1 knockdown allowed insulin to induce a greater amount of GLUT4 translocation than in cells that express TBC1D1. These findings suggest that at least a portion of the inhibitory effects of TBC1D1 on GLUT4 may not be restrained by insulin. However, they do not eliminate the possibility that TBC1D1 can regulate an insulin-independent increase in glucose transport (e.g., with contraction). PAS-TBC1D1 is elevated in response to contraction in rodent skeletal muscle (19,24). Therefore, it seems possible that PAS-TBC1D1 may play a role in mediating contraction-stimulated glucose transport.Experiments using purified Akt or AMPK demonstrated that each kinase can phosphorylate both AS160 and TBC1D1 in cell-free assays (26,27). Considerable evidence indicates that the insulin-stimulated increase in PAS-AS160 is Akt dependent in skeletal muscle (18,28), and increased AS160 phosphorylation appears to be important for the full effect of insulin on glucose transport (20). However, the specific kinases responsible for contraction-stimulated PAS-AS160 need to be clarified because: 1) Wortmannin can completely inhibit the contraction-stimulated increase in PAS-AS160 in rat skeletal muscle, suggesting that Akt is responsible for the increased PAS-phosphorylation of AS160 during contraction (18), but 2) muscles from mice with genetically disrupted AMPK versus wild-type littermates had reduced contraction-stimulated increase in immunoreactivity toward PAS antibody at ∼160 kDa (PAS-160) (28,29).The primary aim of this study was to elucidate the contributions of Akt and AMPK on increases in PAS-AS160 and PAS-TBC1D1 in skeletal muscle stimulated by insulin or contraction. The PI 3-kinase inhibitor Wortmannin was used to prevent Akt activation (without altering AMPK activation), and compound C, a potent AMPK inhibitor (30), was used to prevent AMPK activation (without altering Akt activation). A secondary aim was to determine whether inhibition of insulin- or contraction-stimulated increases in PAS-AS160 or PAS-TBC1D1 was accompanied by attenuated insulin- or contraction-stimulated glucose transport. We hypothesized that in isolated rat epitrochlearis muscle: 1) Akt-dependent mechanisms are essential for the insulin-stimulated increases in glucose transport and phosphorylation of AS160 and TBC1D1; 2) Akt-dependent (but not AMPK-dependent) mechanisms are essential for contraction-stimulated increases in PAS-AS160, but not glucose transport; and 3) AMPK-dependent (but not Akt-dependent) mechanisms are essential for contraction-stimulated increases in PAS-TBC1D1 (but not PAS-AS160) and glucose transport.