There has been a concern that sodium-glucose cotransporter 2 (SGLT2) inhibitors could reduce skeletal muscle mass and function. Here, we examine the effect of canagliflozin (CANA), an SGLT2 inhibitor, on slow and fast muscles from nondiabetic C57BL/6J mice. In this study, mice were fed with or without CANA under ad libitum feeding, and then evaluated for metabolic valuables as well as slow and fast muscle mass and function. We also examined the effect of CANA on gene expressions and metabolites in slow and fast muscles. During SGLT2 inhibition, fast muscle function is increased, as accompanied by increased food intake, whereas slow muscle function is unaffected, although slow and fast muscle mass is maintained. When the amount of food in CANA-treated mice is adjusted to that in vehicle-treated mice, fast muscle mass and function are reduced, but slow muscle was unaffected during SGLT2 inhibition. In metabolome analysis, glycolytic metabolites and ATP are increased in fast muscle, whereas glycolytic metabolites are reduced but ATP is maintained in slow muscle during SGLT2 inhibition. Amino acids and free fatty acids are increased in slow muscle, but unchanged in fast muscle during SGLT2 inhibition. The metabolic effects on slow and fast muscles are exaggerated when food intake is restricted. This study demonstrates the differential effects of an SGLT2 inhibitor on slow and fast muscles independent of impaired glucose metabolism, thereby providing new insights into how they should be used in patients with diabetes, who are at a high risk of sarcopenia.

No abstract

Background: Diabetes is associated with loss of skeletal muscle mass and function. Specifically, diabetes may affect the function of oxidative muscle fibers with lower fatigability and higher oxidative capacity in skeletal muscle. The effects of SGLT2 inhibition on skeletal muscle metabolism and function in patients with type 2 diabetes remain unclear. Methods: Male db/db (diabetic) and db/+ (non-diabetic) mice at 8 weeks of age were fed a normal chow diet or a diet containing canagliflozin (CANA) for 4 weeks. Grip strength and running distance were assessed during CANA treatment. At the end of CANA treatment, we measured muscle weight and analyzed metabolites in skeletal muscle. Results: The weights of the oxidative soleus and glycolytic extensor digitorum longus muscles were reduced by approximately 50% in the diabetic group compared to the non-diabetic group, but neither was altered by CANA treatment. Grip strength did not differ between the CANA-treated diabetic, diabetic, and non-diabetic groups. However, the running distance decreased in the diabetic group compared to the non-diabetic group (89 ± 60 m vs. 1980 ± 276 m). This was partially restored by CANA treatment (89 ± 60 m vs. 415 ± 140 m). Metabolomic analysis revealed that long-chain acyl-CoA increased and medium-chain and short-chain acyl-CoA decreased in diabetic soleus muscle. In contrast, CANA treatment increased medium-chain and short-chain acyl-CoA. Furthermore, CANA increased 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranosyl 5’-monophosphate (AICARP), an endogenous AMPK activator, by approximately 3-fold in diabetic soleus muscle with increased levels of AMPK phosphorylation (Thr172). Conclusion: These results suggest that SGLT2 inhibitors may activate the AICARP/AMPK pathway and improve impaired oxidative muscle function in patients with type 2 diabetes. Disclosure S. Nakamura: None. Y. Miyachi: None. H. Yokomizo: None. H. Otsuka: None. R. Sakamoto: None. M. Takahashi: None. Y. Izumi: None. T. Miyazawa: None. T. Bamba: None. Y. Ogawa: None.