In This Issue of Diabetes

A protein called AIFM2, or apoptosis-inducing mitochondrion-associated factor 2, appears to be a critical metabolic regulator of glucose utilization in glycolytic muscles during exercise, according to Nguyen et al. (p. 2084). The findings collectively start to explain some of the molecular steps that determine fuel efficiency during exercise, according to the authors, but there might also be leads for potential therapies in obesity and diabetes, as skeletal muscle is a major metabolic tissue involved in glycemic control. The findings come from mouse- and cell-based mechanistic studies focusing mainly on the effects of treadmill exercise in mice with overexpression or depletion of AIFM2 in muscles. Initially, the authors found that AIFM2 appeared to be selectively inducible in the glycolytic extensor digitorum longus (a leg muscle) following exercise. Using myotubes in vitro and overexpression of AIFM2, they then found that the protein appeared to alter the NAD+-to-NADH ratio, which led to promotion of glucose oxidation. In mice, they found that overexpression of AIFM2 greatly increased exercise capacity and glucose utilization. Conversely, knockdown of AIFM2 decreased exercise capacity and glucose utilization, in this case demonstrating the effect with two different methods. Finally, they also found that impaired glucose utilization and low capacity for exercise could be rescued with the introduction of NADH dehydrogenase 1. According to the authors, the findings help with understanding the effects of exercise, but perhaps more importantly they note the importance of skeletal muscle in glycemic control. They also suggest that a potential line of research is to investigate the effects of promoting AIFM2 at rest and thus not when exercising, setting it up as a potential starting point for therapies in diabetes and obesity. Commenting further, author Sona Kang said: “It is exciting that we identified AIFM2 as a novel molecular factor to support high intensity exercise by supporting NADH supply for efficient glucose utilization. Our findings will enhance our understanding of the mechanistic basis of fuel utilization to achieve a better exercise capacity.”

A transcription factor called forkhead box i2, or Foxi2, may have an important role in controlling energy metabolism, according to Fan et al. (p. 2106). More specifically, it seems that hypothalamic Foxi2 exerts its effects on energy intake and expenditure through the induction of agouti-related protein (AgRP) expression in neurons. According to the authors, this sets up the targeting of Foxi2 in the hypothalamus as a potential treatment route in obesity and other metabolic diseases such as diabetes. The findings come from a series of experiments with mice and cells that looked at the effects of overexpression or knockdown of Foxi2 under different nutritional and temperature conditions. Initial experiments showed that Foxi2 expression was increased most clearly in the hypothalamus of mice with high-fat-diet–induced obesity. Fasting further increased mRNA and protein levels of Foxi2, while refeeding somewhat reversed the effect. Of note, the authors also observed that expression levels of AgRP were like those of Foxi2. Virus-induced Foxi2 overexpression in mice led to its overexpression in AgRP neurons and to increased AgRP expression. This increased food intake and reduced energy expenditure compared with controls effectively led to obesity and insulin resistance. Conversely, mice with Foxi2 knockdown became leaner with age and resistant to high-fat-diet–induced obesity and other metabolic disturbances. Based on further (mechanistic) experiments, the authors note that Foxi2 appears to stimulate AgRP by directly binding to it, activating transcription. Using whole-cell patch clamp experiments, they also found Foxi2 overexpression activates AgRP neuron activity. Based on the findings, they suggest that “targeting Foxi2 in the hypothalamus might be a therapeutic strategy for treating obesity and metabolic diseases.” Author Yongsheng Chang added: “Given that Foxi2 is also highly expressed in paraventricular nucleus, cerebellum and brain stem, it will be interesting to reveal the physiological functions of Foxi2 in these brain areas.”

Four markers of glycemic status that are commonly measured in clinical settings appear to be associated with increased risk for a range of coronary atherosclerosis-related diseases and symptoms, according to Yuan et al (p. 2222). Importantly, the findings come from a Mendelian randomization analysis of the relationships, meaning that causal inference is stronger and residual confounding and reverse causality are (largely) eliminated. Previous studies have looked at glycemia and cardiovascular outcome risks, but most have been observational in nature, meaning causality is impossible to assign. Using data from large genetic consortia and the FinnGen and UK Biobank studies, the authors looked at the relationships between fasting glucose, fasting insulin, glucose levels 2 h after a glucose challenge, and HbA1c with 12 atherosclerotic and 4 thrombotic outcomes. In total the study used data from just under 200,000 individuals. They found that except for HbA1c, increases in all the glycemic traits were associated with increased peripheral arterial disease risks. Fasting glucose and glucose levels 2 h post-challenge were positively associated with large artery stroke risks. Fasting insulin levels were positively associated with risks for ischemic stroke and chronic kidney disease. Glucose levels 2 h post-challenge also showed positive association with small vessel stroke. Numerous other relationships existed, although some were more suggestive than others, according to the authors. Notably, however, not all higher levels of the glycemic traits were associated with increased risks for thrombotic outcomes. The authors note that most individual glycemic trait–outcome relationships remained following adjustment for other glycemic traits. “These results imply the universally detrimental role of high glycemic status in both coronary and peripheral artery atherosclerosis and the importance of glycemic management for preventing and delaying progression of related diseases,” they write.

Insulin sensitivity appears to be enhanced by glucagon agonism during conditions of both hypoglycemia and euglycemia, according to Kim et al (p. 2123). Moreover, they also identify a complex termed mTORC2 (mammalian target of rapamycin complex 2) as key to this glucagon-dependent enhancement in sensitivity. Previously it was assumed that glucagon opposes the actions of insulin, driving up blood glucose levels under conditions such as hypoglycemia. However, according to the authors, its actions may be much more subtle and more reliant on the physiological context in which glucagon is stimulated. The findings come from mouse- and cell-based studies that center on various glucose or insulin clamp–based approaches to explore more deeply hepatic mTORC2 signaling and its relationship with insulin. According to the authors, their previous studies identified an unexpected improvement in glucose metabolism following acute glucagon agonism, motivating them to investigate the mechanisms involved in their current study. Focusing on the protein called RICTOR, one of the seven proteins that make up the mTORC2 complex, they found that mice with a liver-specific genetic knockout of Rictor had glucose intolerance and impaired AKT signaling. They were also resistant to hyperglycemia following stimulation with glucagon. A series of clamp studies and in vitro experiments with hepatocytes, essentially working down the glucagon signaling pathway, then demonstrated the essential nature of RICTOR in the mTORC2 complex and thus, by implication, its role in the action of insulin. According to the authors, the findings should provide more insights into the paradoxical benefits of potential treatments based on glucagon-like protein 1/glucagon co-agonism that are in development. They add that the insights might also aid in the further technological development of a bihormonal (i.e., glucagon and insulin) artificial pancreas. “In sum, these data suggest that [glucagon receptor] agonism acts via the mTORC2 kinase complex to enhance hepatic insulin-stimulated AKTSer473 phosphorylation and activity and thereby potentiates whole-body and hepatic insulin sensitivity,” they conclude.