By Max Bingham, PhD

Although the kidney is known to play a crucial role in glucose homeostasis and metabolism, the underlying mechanisms of regulation of gluconeogenesis in the proximal tubule are surprisingly little understood. On that basis, Sasaki et al. (p. 2339) describe a series of mouse experiments that systematically try to uncover the mechanisms involved, revealing that gluconeogenesis is likely regulated by both insulin signaling and glucose reabsorption in the proximal tubules. They report that this likely affects gluconeogenic gene expression variously, depending on whether there was a fasted or fed state. Using different models, including tissue-specific knockout mice, they then unravel the pathways involved. They go on to suggest that various factors including PGC1α, FoxO1, sirtuin 1, and sodium–glucose cotransporter (SGLT) may be involved. Specifically, in mice with proximal tubule–specific knockout of insulin receptor substrates, they found impaired insulin signaling and upregulated gluconeogenesis gene expression, with corresponding increases in renal gluconeogenesis and insulin resistance. The hint that two pathways might be involved came from experiments involving mice given streptozotocin. Further experiments in cells then helped unravel what might be going on. In addition to insights into the mechanisms involved, the authors point out that some of the findings might have clinical significance, including consequences for the use of the SGLT2 inhibitors. Author Naoto Kubota told Diabetes: “We assumed that insulin is the dominant regulator of glucose metabolism regardless of particular tissues. However, our data show that this is unlikely to apply to the proximal tubules of the kidney, at least in some conditions. We also find it curious that the proximal tubules do not seem to care for blood glucose levels: they sense primary urine glucose levels instead, via SGLTs on the luminal side. As is revealed in this paper, glucose metabolism in the kidney does have unique aspects, which should be addressed in detail in future research.”

Sasaki et al. Dual regulation of gluconeogenesis by insulin and glucose in the proximal tubules of the kidney. Diabetes 2017;66:2339–2350

Gluconeogenic gene expression is dually regulated by insulin and the reabsorbed glucose signaling in the proximal tubules of the kidney.

Gluconeogenic gene expression is dually regulated by insulin and the reabsorbed glucose signaling in the proximal tubules of the kidney.

Close modal

Genome-wide association studies have identified numerous genetic loci that are likely associated with type 2 diabetes. Linking these to actual underlying molecular mechanisms (and therefore opening them up as potential drug targets) has proven difficult. Progress, however, seems possible. Roman et al. (p. 2521) report a molecular investigation into the role of ADCY5 gene variants that encode adenylate cyclase 5, an enzyme involved in the catalysis of the production of cAMP, which is a crucial messenger molecule involved in pancreatic β-cell insulin production. Previous genome-wide association studies have consistently flagged an association between variants in ADCY5 and type 2 diabetes risk, but according to the authors, underlying mechanisms linking the two have never been uncovered. Using a series of experiments in cells and various sequencing approaches, they report that a number of type 2 diabetes risk alleles are likely associated with decreased ADCY5 expression in human islets and that the variant rs11708067 appears to overlap a predicted enhancer region in the cells. More specifically, the rs11708067-A allele appeared to exhibit lower transcriptional activity, specific binding of certain nuclear proteins, and enhancer activity in human islets—all of which would presumably influence the expression of the enzyme adenylate cyclase 5, the subsequent catalysis of cAMP, and hence insulin production. To confirm the links, the authors report using CRISPR-Cas9 gene editing to delete the enhancer element in rat insulinoma cells, which they say resulted in diminished Adcy5 gene expression and ultimately resulted in reduced insulin secretion when exposed to glucose. As a result, they propose that rs11708067 is a regulatory variant at the ADCY5 gene locus and that the mechanisms they propose can be linked to altered glucose levels and hence type 2 diabetes risk via a number of potential steps. Author Karen L. Mohlke said: “Establishing direct links between noncoding regulatory variants and their target genes can help translate association study results into causal mechanisms and disease pathways. Studies such as this one help explain the molecular basis for disease susceptibility.”

Roman et al. A type 2 diabetes–associated functional regulatory variant in a pancreatic islet enhancer at the ADCY5 locus. Diabetes 2017;66:2521–2530

Empagliflozin may be able to reduce fasting plasma glucose and improve β-cell function in individuals with impaired fasting glucose, i.e., a state widely regarded as having prediabetes and thus an increased risk of type 2 diabetes. According to Abdul-Ghani et al. (p. 2495), the result suggests that the glucosuria promoted by the drug may help reduce fasting plasma glucose and potentially glucotoxicity to such an extent that β-cell function and hence insulin production can kick in again. If the result stands up to further scrutiny, it would help explain some of the effects of empagliflozin but moreover suggests it might have clinical utility at a much earlier point in type 2 diabetes development, at the prediabetes stage. The study involved eight individuals with impaired fasting glucose and eight with normal fasting glucose and used the gold standard stepped hyperglycemic clamp technique to track the effects of empagliflozin over 14 days on fasting plasma glucose levels and β-cell function. The clamp approach was administered at baseline, 48 h, and 14 days. In short, the drug induced glucosuria at about the same rate in both groups, which was maintained over 14 days. In individuals with impaired fasting glucose (i.e., prediabetes), glucose levels dropped significantly over 14 days. However, in the group with normal fasting glucose, levels remained unchanged, despite the glucosuria. β-Cell function improved by about 20% in individuals with impaired fasting glucose but remained unchanged in those with normal glucose levels. Author Ralph A. DeFronzo commented: “For every patient with diabetes, there are at least two individuals with prediabetes. Early treatment, at the prediabetic stage, has the potential to prevent the development of overt diabetes and thus the onset of microvascular complications. Since progressive β-cell failure is the major determinant of progression to overt diabetes, the SGLT2 inhibitor class of antidiabetic medications has great promise in preventing the development of diabetes and its associated microvascular complications.”

Abdul-Ghani et al. Inhibition of renal sodium–glucose cotransport with empagliflozin lowers fasting plasma glucose and improves β-cell function in subjects with impaired fasting glucose. Diabetes 2017;66:2495–2502

An investigation by Voss et al. (p. 2483) into the actions of insulin during acute hypoglycemia suggests that insulin-stimulated glucose disposal and clearance in skeletal muscle is likely impaired, whereas levels of a variety of counterregulatory hormones, including epinephrine, are increased. Likewise, insulin-stimulated glycogen synthase activity may also be reduced during hypoglycemia. Taken together, the authors explain, the data illustrate the possible biological processes under way following a hypoglycemia episode and that the dual outcome of reduced glucose disposal and increased endogenous glucose production is likely to be essential for recovery. Although the study focused on healthy individuals, the authors say that the results are still likely to be applicable to individuals with diabetes even when there is an impaired glucagon response. The study focused on nine healthy individuals who were examined on three separate study days in a crossover design. The examinations included periods with hyperinsulinemic hypoglycemia (with bolus insulin), hyperinsulinemic euglycemia (with bolus insulin and glucose), and a saline control period. Muscle biopsies were also taken before and after insulin/saline injection. Subsequently, a range of different clinical values, signaling molecules, and hormones were assessed. Although glucose levels fell during hypoglycemia, a range of other changes occurred including a peak in epinephrine, complete reduction of glycogen synthase activity, and increases in catecholamine signaling and protein kinase activity. Going further, the authors describe in depth many potential effects of insulin signaling and counterregulatory hormones and their signaling pathways and effects of hypoglycemia on glycogen synthesis. Overall, they suggest their data should fill many of the knowledge gaps relating to insulin and its effects during hypoglycemia. Author Niels Jessen said: “Fear of hypoglycemia is a commonly reported problem among patients with insulin-treated diabetes. In order to prevent this condition, we need to understand the complicated interplay between hormones and substrates under these conditions by performing experiments in humans. Our results give insight into compensatory mechanisms in skeletal muscle and may help facilitate better treatment.”

Voss et al. Acute hypoglycemia in healthy humans impairs insulin-stimulated glucose uptake and glycogen synthase in skeletal muscle: a randomized clinical study. Diabetes 2017;66:2483–2494

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at