By Max Bingham, PhD

A study published in this issue of Diabetes (p. 3573) suggests a process called O-GlcNAcylation in cardiac muscle proteins may be responsible for the weaker heart function often seen in diabetic patients. The authors, Ramirez-Correa et al., suggest that displacement of two enzymes in cardiac muscle proteins associated with the O-GlcNAcylation process might now represent a target for treating cardiomyopathy in diabetes. The process is thought to involve cardiac muscle becoming desensitized to Ca2+. Both intracellular calcium concentrations and myofilament properties govern how the heart beats. However, these processes are not currently well understood. More widely, the reason why heart muscle becomes weaker with the onset of diabetes is also still not fully understood at the molecular level in particular. Using an advanced set of proteomics techniques, the authors investigated the effects of removing abnormal myofilament–associated O-GlcNAcylation in a diabetic animal model. They found that the removal resulted in restoration of Ca2+ sensitivity. At the cellular level the displacement of O-GlcNAcase (OGA) and O-GlcNAc (OGT) transferase appears to occur between Z-line and A-band locations in the cardiac muscle structure in both the animal and human cell diabetes models. This suggests a fundamental change occurred in cardiac muscle structure upon the occurrence of diabetes and, as the authors propose in their conclusions, might represent a “new therapeutic avenue for treating diabetic cardiomyopathy.” Commenting further on the study, lead author Genaro A. Ramirez-Correa told us: “The fact that we show that O-GlcNAcylation plays an important role in diabetic cardiomyopathy is a valuable opportunity to develop new therapies targeted to either fix the myofilament apparatus (for example with gene transfer or calcium sensitizing drugs) or to modulate the cardiac OGT interactome. For example, in skeletal cells, OGT cross talk with AMPK might present many more drugable interactor partners of cardiac OGT—in much the same way that AMPK is a known target of metformin.”

Ramirez-Correa et al. Removal of abnormal myofilament O-GlcNAcylation restores Ca2+ sensitivity in diabetic cardiac muscle. Diabetes 2015;64:3573–3587

A vaccine for type 1 diabetes might be possible if targeted towards central immunological tolerance. That is according to a series of animal experiments described by Culina et al. in this issue of Diabetes (p. 3532). Preproinsulin fused to a fragment of IgG can be delivered from mother to fetus via the placenta (and specifically via the neonatal Fc receptor pathway), ferried to the thymus by dendritic cells, and resulted in the prevention of autoimmune (type 1) diabetes development, at least in mice prone to developing diabetes. According to the authors, the protection relied on a resulting increase in regulatory T cells expressing transforming growth factor-β and also an increase in effector T cells displaying impaired cytotoxicity. Since this occurred at a crucial stage in the development of the immune system, it would have boosted central immune tolerance toward pancreatic β-cells and hence protected them from autoimmune destruction. The idea, at least in principle, is based on the theory that the immune system of the fetus is very much geared to defining what immunological “self” is at the early stages of development. By directly intervening in the process, the study shows that it should be possible to reshape the defective “self-image” of the diabetic immune system in order to prevent autoimmune destruction of β-cells. While this approach is at an early stage of development, it should be interesting for helping children with a high genetic risk of type 1 diabetes according to the authors. The authors also caution that a less invasive administration route would be needed before any clinical application can be considered. Commenting more widely on the study, author Roberto Mallone stated that: “Recent efforts toward type 1 diabetes prevention are being implemented earlier than previously attempted in genetically at-risk children without any sign of ongoing autoimmunity. The earliest checkpoint in autoimmune progression that takes place in the thymus has however remained therapeutically unexploited. This vaccination strategy is the first targeting thymic central tolerance. Although considerable work remains to be done to make the risk-benefit balance more suitable for preventative purposes, the same concept could be applied to other autoimmune diseases arising from an inappropriate activation against self-antigens.”

Culina et al. Materno-fetal transfer of preproinsulin through the neonatal Fc receptor prevents autoimmune diabetes. Diabetes 2015;64:3532–3542

Optimal glycemic control in the management of type 1 diabetes is often hampered by the risk of hypoglycemic events. Gregory et al. asked whether the normal route of administration of insulin (subcutaneous injection) into peripheral circulation might be improved upon by injecting insulin directly into portal circulation. As reported in this issue of Diabetes (p. 3439), they found that—in dogs at least—portal vein administration lessens hypoglycemia and that peripheral insulin delivery may well exacerbate hypoglycemia, particularly when there is a diminished glucagon response (as is the case in type 1 diabetes). Glucose fell faster and glucagon peaked earlier when insulin was administered peripherally. Epinephrine levels were also increased more with peripheral insulin administration. In a second protocol where glucagon levels were maintained at basal levels, glucose levels fell faster with peripheral insulin and epinephrine responses were much larger than observed in the first protocol. The authors suggest that the increased propensity for hypoglycemia associated with the peripheral delivery of insulin “…stems from having higher insulin concentrations at muscle, leading to a larger increase in Rd [glucose utilization] than when insulin is delivered Po [via portal vein].” The outcome of the study suggests that we need to rethink insulin delivery in type 1 diabetes. Commenting on the wider implications of the research, lead author Justin M. Gregory told us: “Emerging technologies such as encapsulated islet cell transplantation and the artificial pancreas carry great potential to reduce glycemic excursions in type 1 diabetes, but these therapies are generally conceived to deliver insulin into the peripheral circulation. To date, the metabolic consequences of this delivery route utilizing these novel approaches—as opposed to the more physiologic route of hepatic portal insulin delivery—have not been well characterized. Our present study suggests that these developing therapies could be enhanced with insulin delivery into the intraperitoneal space or if short-acting, hepatopreferential insulin was used. Therapies that bring into balance the physiologic portal-to-peripheral ratio of insulin would result in a shift to a lower portion of the dose-response curve relating insulin concentration to glucose utilization. This shift would not only be expected to reduce hypoglycemia, but would likely make artificial pancreas control algorithms more robust. Our future investigations will further characterize these proposed benefits.”

Gregory et al. Insulin delivery into the peripheral circulation: a key contributor to hypoglycemia in type 1 diabetes. Diabetes 2015;64:3439–3451

Vision loss due to diabetic macular edema (DME) in diabetic patients is often treated with therapies that block vascular endothelial growth factor (VEGF) in the vitreous of the eye (the gel between the lens of the eye and the retina). However, around 50% of patients do not fully respond to the therapy, suggesting that other mechanisms are at play. Kita et al. now suggest that the kallikrein-kinin system in the vitreous of the eye might be a mediator of diabetic macular edema and propose mechanisms that might contribute to retinal thickening and retinal vascular permeability that ultimately lead to vision loss. Reporting in this issue of Diabetes (p. 3588), the study first reveals that plasma prekallikrein and kallikrein in the vitreous of patients with DME are significantly elevated in comparison to patients with a macular hole. VEGF was also elevated but kallikrein levels did not correlate in the group of patients studied suggesting that different mechanisms may be important in some patients. Proteomics studies then identified 30 proteins likely associated with DME that had a higher correlation with plasma prekallikrein than with VEGF concentrations. Subsequent studies in diabetic rats showed that DME vitreous with relatively high levels of kallikrein and low levels of VEGF could induce retinal vascular permeability. The response could be blocked by bradykinin receptor antagonists but not by bevacizumab, a VEGF inhibitor. In further mice studies, plasma prekallikrein deficiency reduced retinal vascular permeability induced by diabetes. Taken together, the authors suggest that targeting this kallikrein-kinin system might represent an option to improve vision in patients that do not fully respond to anti-VEGF DME therapy. Commenting more widely on the study, author Edward P. Feener said: “Our findings demonstrate that plasma kallikrein is increased in the vitreous in a large proportion of DME patients and activation of the plasma kallikrein-kinin system is a potent VEGF-independent pathway of retinal edema. Plasma kallikrein inhibitors may provide opportunities to treat DME both as a stand-alone therapy and in combination with anti-VEGF agents. Clinical trials are currently underway to examine plasma kallikrein’s role as a therapeutic target for DME.”

Kita et al. Plasma kallikrein-kinin system as a VEGF-independent mediator of diabetic macular edema. Diabetes 2015;64:3588–3599