By Max Bingham, PhD

A genome-wide association study investigating diabetic kidney disease (DKD) in type 2 diabetes has turned up only a modest set of potential leads that might explain progression of the disease. Despite using a sample size of just over 40,000 patients with diabetes, van Zuydam et al. (p. 1414) found only one novel locus that is potentially linked to microalbuminuria in European subjects. The study reportedly combined genome-wide data from numerous European and Asian study cohorts and involved the definition of eight phenotypes covering DKD overall as well as specific aspects of the disease. They also investigated possible links between various risk factors, genetics, and DKD and whether previously identified genetic associations with DKD held up in the expanded sample size they used. While there were no specific associations with the broad DKD phenotype, the authors did identify one novel locus called GABRR1 associated with microalbuminuria in European subjects with type 2 diabetes. However, they could not replicate the association in European patients with type 1 diabetes or Asian subjects with type 2 diabetes. In terms of risk factors, they found some genetic support for obesity and insulin sensitivity as causal factors for DKD in type 2 diabetes but urge caution because of the potential of collider bias. For previously identified genetic associations, they report replicating very few, with many of them only nominally significant and, in some cases, directionally inconsistent. Author Natalie R. van Zuydam told Diabetes: “Diabetic kidney disease is a serious complication of diabetes. It has a considerable effect on patients’ quality of life. Better understanding of the disease mechanisms is key to developing new therapies and identifying patients at high risk of disease development. This study highlights the complexity of genetic studies of diabetic kidney disease and provides a basis for follow-up studies, in even larger sample sizes, that are essential to the understanding of the genetic basis of diabetic kidney disease and its underlying mechanisms.”

van Zuydam et al. A genome-wide association study of diabetic kidney disease in subjects with type 2 diabetes. Diabetes 2018;67:1414–1427

Phosphatidylinositol 3-kinase (PI3K) appears to have separate roles in the development of obesity and insulin resistance/hyperglycemia. The findings by Solheim et al. (p. 1297), which have come from research in mice with a specific mutation, might help in the development of specific therapies targeting the PI3K pathway as a way to regulate insulin levels and also potentially obesity. The study focuses on a mouse model with a specific heterozygous missense mutation called R649W that is in the p85α regulatory subunit gene PIK3R1 of PI3K. The mutation has been identified in patients with SHORT syndrome, a rare disorder that results in growth retardation, insulin resistance, and partial lipodystrophy (where the body fails to produce or maintain fat tissues). Mice with the mutation have a very similar phenotype, meaning that it should be possible to tease apart the role of PI3K in obesity and insulin resistance. According to the authors, when compared with controls, mice with the mutation had reduced weight gain and adipose tissue accumulation when fed a high-fat diet designed to induce obesity. The mice with the mutation also had reduced levels of expression of several genes involved in lipid metabolism but conversely had significant insulin resistance and hyperglycemia. To confirm whether the effects were due to obesity and not specifically the high-fat diet, the authors then repeated the experiments with mice with the mutation but this time on a background of spontaneous obesity. Again, the mutation protected the mice from obesity and also from fatty liver disease, but they still developed a severe diabetic state. Author C. Ronald Kahn commented: “It is remarkable how well mice carrying this human mutation mimic the human disease. We are now exploring how we can capture the benefits of this mutation that protect against obesity and fatty liver while avoiding the effects that create generalized insulin resistance.”

Solheim et al. Mice carrying a dominant-negative human PI3K mutation are protected from obesity and hepatic steatosis but not diabetes. Diabetes 2018;67:1297–1309

In a Perspective for Diabetes, Michels and Gottlieb (p. 1211) reflect on the failure of type 1 diabetes prevention trials with oral insulin and advocate for an alternative approach that focuses on mechanistic studies and a more personalized approach toward treatment. Starting from a position that if a disease is to be prevented, it must first be predicted, the authors highlight a series of studies that have led to type 1 diabetes becoming a predictable disease. This has led to screening efforts being made in the U.S. and beyond in an attempt to identify individuals at risk. However, this poses a dilemma, as being able to predict risk for a disease necessitates treatments to delay and prevent the disease—and for type 1 diabetes, that ambition is still a long way off. Michels and Gottlieb describe two large randomized trials of oral insulin that both proved negative for the primary outcomes of delaying or preventing onset of type 1 diabetes at the dose tested. However, in both cases there were hints that certain subgroups might have responded to the therapy. While further studies on oral insulin are under way, the authors say it might now be the right time to reconsider the approach. Turning to adaptive trial designs, they suggest that a “learn as we go” approach that relies on disease biomarkers might be relevant to type 1 diabetes prevention. However, biomarkers for short-term readouts are not well defined, meaning that trial adaptations would be difficult. As a result, they describe a whole series of potential “personalized” therapies that are on the horizon and that the key is to target specific molecular targets and readout biomarkers in hypothesis-driven mechanistic research studies. Author Aaron W. Michels said: “Type 1 diabetes is now predictable, and disease prevention should naturally follow. However, prevention has proven much more difficult than originally thought. By reflecting on these challenges, we are cautiously optimistic about the future efforts at delaying and eventually preventing type 1 diabetes onset.”

Michels and Gottlieb. Learning from past failures of oral insulin trials. Diabetes 2018;67:1211–1215

GPR119, a G-protein–coupled receptor expressed in pancreatic islets, can reportedly be stimulated with small-molecule agonists to enhance glucagon secretion specifically in response to hypoglycemia. According to Li et al. (p. 1401), the findings suggest that adjunctive pharmacological therapy might be possible to prevent hypoglycemia episodes associated with insulin therapy in both type 1 and type 2 diabetes. Using a series of cell- and rodent-based experiments, the authors describe how pretreatment with various GPR119 agonists affected insulin and glucagon secretion during conditions of hypoglycemia and hyperglycemia. They reportedly found that pretreatment with various GPR119 agonists did result in enhanced glucagon secretion in isolated islets and pancreata but only during low glucose perfusion rates. Likewise, in healthy rats and streptozotocin-treated rats with diabetes, pretreatment with the agonists increased glucagon secretion but only during hypoglycemia. To determine whether it was a class effect of the compounds, the authors then tested the effects of the agonists in mice with GPR119 receptors knocked out. In this case, they found that while wild-type mice did mount an enhanced glucagon response following treatment with GPR119 compounds, no effect was evident when the receptors were knocked out. The authors say that, overall, the results indicate that the agonists appear to have the ability to pharmacologically augment glucagon secretion but, notably, only in response to hypoglycemia. While evidently relevant to potentially avoiding hypoglycemia, particularly due to intensive treatments for diabetes, they caution that there is likely a long way to go before such preventative approaches are available for patients. Author Nina Xiaoyan Li commented: “There is no current therapeutic, apart from exogenous glucagon injection, to address the life-threatening complication of hypoglycemia. We are hopeful that our novel discovery in rodents concerning GPR119 biology and pharmacology, if translatable to the clinic, could be a game-changing adjunctive therapy restoring α-cell counterregulatory response in type 1 and type 2 diabetes.”

Li et al. GPR119 agonism increases glucagon secretion during insulin-induced hypoglycemia. Diabetes 2018;67:1401–1413

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