5′-AMP-activated protein kinase (AMPK) functions as a metabolic switch in mammalian cells and can be artificially activated by 5-aminoimidazole-4-carboxamide (AICA)-riboside. AMPK activation during muscle contraction is dependent on muscle glycogen concentrations, but whether glycogen also modifies the activation of AMPK and its possible downstream effectors (glycogen synthase and glucose transport) by AICA-riboside in resting muscle is not known. Thus, we have altered muscle glycogen levels in rats by a combination of swimming exercise and diet and investigated the effects of AICA-riboside in the perfused rat hindlimb muscle. Two groups of rats, one with super-compensated muscle glycogen content (∼200–300% of normal; high glycogen [HG]) and one with moderately lowered muscle glycogen content (∼80% of normal; low glycogen [LG]), were generated. In both groups, the degree of activation of the α2 isoform of AMPK by AICA-riboside depended on muscle type (white gastrocnemius ≫ red gastrocnemius > soleus). Basal and AICA-riboside-induced α2-AMPK activity were markedly lowered in the HG group (∼50%) compared with the LG group. Muscle 2-deoxyglucose uptake was also increased and glycogen synthase activity decreased by AICA-riboside. Especially in white gastrocnemius, these effects, as well as the absolute activity levels of AMPK-α2, were markedly reduced in the HG group compared with the LG group. The inactivation of glycogen synthase by AICA-riboside was accompanied by decreased gel mobility and was eliminated by protein phosphatase treatment. We conclude that acute AICA-riboside treatment leads to phosphorylation and deactivation of glycogen synthase in skeletal muscle. Although the data do not exclude a role of other kinases/phosphatases, they suggest that glycogen synthase may be a target for AMPK in vivo. Both basal and AICA-riboside-induced AMPK-α2 and glycogen synthase activities, as well as glucose transport, are depressed when the glycogen stores are plentiful. Because the glycogen level did not affect adenine nucleotide concentrations, our data suggest that glycogen may directly affect the activation state of AMPK in skeletal muscle.
Address correspondence and reprint requests to Jørgen F.P. Wojtaszewski, Copenhagen Muscle Research Centre, Institute of Exercise and Sports Sciences, 13 Universitetsparken, University of Copenhagen, DK-2100 Copenhagen, Denmark. E-mail:.
Received for publication 2 April 2001 and accepted in revised form 5 November 2001.
AICA, 5-aminoimidazole-4-carboxamide; AMPK, 5′-AMP-activated protein kinase; CCD, charge-coupled device; CrP, creatin phosphate; ECF, extracellular fluid; G6P, glucose-6-phosphate; GS, glycogen synthase; HG, high glycogen; LG, low glycogen; PP1, protein phosphatase-1; PVDF, polyvinylidine difluoride; RG, red gastrocnemius; TBST, Tris-buffered saline with Tween; WG, white gastrocnemius; ZMP, 5-aminoimidazole-4-carboxamide ribonucleotide.