The increase in skeletal muscle glucose uptake during exercise results from a coordinated increase in rates of glucose delivery (higher capillary perfusion), surface membrane glucose transport, and intracellular substrate flux through glycolysis. The mechanism behind the movement of GLUT4 to surface membranes and the subsequent increase in transport by muscle contractions is largely unresolved, but it is likely to occur through intracellular signaling involving Ca²⁺-calmodulin-dependent protein kinase, 5′-AMP-activated protein kinase, and possibly protein kinase C.

The Ca²⁺/calmodulin (CaM) competitive inhibitor KN-93 has previously been used to evaluate 5′-AMP-activated protein kinase (AMPK)-independent Ca²⁺-signaling to contraction-stimulated glucose uptake in muscle during intense electrical stimulation ex vivo. With the use of low-intensity tetanic contraction of mouse soleus and extensor digitorum longus (EDL) muscles ex vivo, this study demonstrates that KN-93 can potently inhibit AMPK phosphorylation and activity after 2 min but not 10 min of contraction while strongly inhibiting contraction-stimulated 2-deoxyglucose uptake at both the 2- and 10-min time points. These data suggest inhibition of Ca²⁺/CaM-dependent signaling events upstream of AMPK, the most likely candidate being the novel AMPK kinase CaM-dependent protein kinase kinase (CaMKK). CaMKK protein expression was detected in mouse skeletal muscle. Similar to KN-93, the CaMKK inhibitor STO-609 strongly reduced AMPK phosphorylation and activity at 2 min and less potently at 10 min. Pretreatment with STO-609 inhibited contraction-stimulated glucose uptake at 2 min in soleus, but not EDL, and in both muscles after 10 min. Neither KN-93 nor STO-609 inhibited 5-aminoimidazole-4-carboxamide-1-β-4-ribofuranoside-stimulated glucose uptake, AMPK phosphorylation, or recombinant LKB1 activity, suggestive of an LKB1-independent effect. Finally, neither KN-93 nor STO-609 had effects on the reductions in glucose uptake seen in mice overexpressing a kinase-dead AMPK construct, indicating that the effects of KN-93 and STO-609 on glucose uptake require inhibition of AMPK activity. We propose that CaMKKs act in mouse skeletal muscle regulating AMPK phosphorylation and glucose uptake at the onset of mild tetanic contraction and that an intensity- and/or time-dependent switch occurs in the relative importance of AMPKKs during contraction.

Dietary protein intake is linked to an increased incidence of type 2 diabetes (T2D). Although dietary protein dilution (DPD) can slow the progression of some aging-related disorders, whether this strategy affects the development and risk for obesity-associated metabolic disease such as T2D is unclear. Here, we determined that DPD in mice and humans increases serum markers of metabolic health. In lean mice, DPD promoted metabolic inefficiency by increasing carbohydrate and fat oxidation. In nutritional and polygenic murine models of obesity, DPD prevented and curtailed the development of impaired glucose homeostasis independently of obesity and food intake. DPD-mediated metabolic inefficiency and improvement of glucose homeostasis were independent of uncoupling protein 1 (UCP1), but required expression of liver-derived fibroblast growth factor 21 (FGF21) in both lean and obese mice. FGF21 expression and secretion as well as the associated metabolic remodeling induced by DPD also required induction of liver-integrated stress response–driven nuclear protein 1 (NUPR1). Insufficiency of select nonessential amino acids (NEAAs) was necessary and adequate for NUPR1 and subsequent FGF21 induction and secretion in hepatocytes in vitro and in vivo. Taken together, these data indicate that DPD promotes improved glucose homeostasis through an NEAA insufficiency–induced liver NUPR1/FGF21 axis.