Edited by Helaine E. Resnick, PhD, MPH
Drug Repurposing: Potential Utility of Paroxetine in Diabetes Complications
Data presented in this issue of Diabetes (p. ) raise the potential of repurposing paroxetine for future exploratory clinical trials in patients with diabetes vascular complications. Work by Gerö et al. examines phenotypic screening as a means to identify existing compounds that may impact key mechanisms involved in promoting diabetes complications. Reactive oxygen species (ROS) formation is thought to be involved in the pathogenesis of diabetic endothelial dysfunction, which in turn suggests ROS involvement in both large- and small-vessel complications. Gerö et al. conducted the first systematic survey to identify drugs and drug-like molecules with the ability to protect endothelial cells from hyperglycemic mitochondrial ROS overproduction. After phenotypic screening in endothelial cells exposed to elevated extracellular glucose, fewer than 20 compounds from a focused library of more than 6,000 compounds met criteria for significant inhibition of ROS production without adverse effects on cell viability. Some of these were glucocorticoids and nonsteroidal anti-inflammatory compounds. The investigators chose to focus on characterizing paroxetine, a commonly used antidepressant agent. Importantly, the wide clinical use and acceptance of paroxetine suggests that, if effective in reducing ROS, this compound might be efficiently repurposed for experimental therapy of diabetes complications. In a series of in vitro experiments, paroxetine reduced hyperglycemia-induced mitochondrial ROS formation and mitochondrial protein oxidation, as well as mitochondrial and nuclear DNA damage. Notably, paroxetine did not interfere with mitochondrial electron transport or cellular bioenergetics. Decomposition studies demonstrated that the primary site of paroxetine’s antioxidant effect resides within its sesamol moiety. In a further series of experiments, aortic rings isolated from streptozotocin-injected diabetic rats showed endothelial dysfunction, as evidenced by an impaired relaxation in response to acetylcholine. Paroxetine prevented the impairment of the endothelium-dependent relaxations. These intriguing research methods provide insight into the potential benefits of leveraging existing compounds for future trials in diabetes vascular complications. — Laura Gehl, PhD
A Potential Role for Endoplasmic Reticulum Stress in Diabetic Neuropathy
New data suggest that endoplasmic reticulum (ER) stress, particularly CAAT/enhancer-binding protein homologous protein (CHOP), contributes to diabetic peripheral neuropathy (DPN). Although DPN affects at least half of all patients with both type 1 and type 2 diabetes, drug candidates studied in clinical trials thus far have shown limited efficacy, or they promote adverse side effects. In this issue of Diabetes (p. ), Lupachyk et al. identify ER stress as an important contributor to DPN. The ER responds to stress with the unfolded protein response (UPR), which leads to deficits in motor nerve conduction velocity (MNCV) and sensory nerve conduction velocity (SNCV), intraepidermal nerve fiber degeneration, and oxidative-nitrative stress in the peripheral nerve. The newly published data suggest that CHOP, a specific component of UPR, contributes to the development of DPN. In experiments with streptozotocin (STZ)-injected diabetic rats, rats with 12-week duration of STZ-induced diabetes responded to ER stress with upregulation of UPR in the spinal cord and sciatic nerve. Two structurally unrelated chemical chaperones, trimethyloxide (TMAO) and 4-phenylbutyric acid (PBA), both attenuated ER stress in the spinal cord and sciatic nerve. This report shows that although a 12-week duration of diabetes caused MNCV (26%) and SNCV (13%) deficits in the STZ-injected diabetic rats, the TMAO-treated and PBA-treated rats maintained normal SNCV and showed less severe MNCV deficits. TMAO also decreased intraepidermal nerve fiber loss and oxidative-nitrative stress in the sciatic nerve and spinal cord. Neither TMAO nor PBA reduced diabetic hyperglycemia. In a second series of experiments, nondiabetic CHOP−/− mice displayed normal MNCV and SNCV. Sciatic nerve CHOP levels were increased by 22% in wild-type STZ-injected diabetic mice, which also exhibited MNCV and SNCV deficits. Diabetic CHOP−/− mice exhibited reduced MNCV and SNCV deficits, intraepidermal nerve fiber loss, and oxidative-nitrative stress in the sciatic nerve compared with the wild-type STZ-injected diabetic mice. These data suggest that mitigation of ER stress may be a promising new therapeutic target in DPN. — Laura Gehl, PhD
Inhibition of Class I Histone Deacetylases Promotes Insulin Sensitivity and Weight Loss in Mice
In this issue of Diabetes (p. ), Galmozzi et al. evaluated the therapeutic potential of class I selective histone deacetylase (HDAC) inhibitors and found that they improved various physiological parameters indb/db mice. In a series of experiments, the researchers evaluated a class I selective HDAC inhibitor (MS275), a class II selective HDAC inhibitor (MC1568), and suberoyl anilide hydroxamic acid (SAHA), a nonselective HDAC inhibitor. Myotubes treated with either MS275 or SAHA showed an increase in the expression of genes involved in glucose and lipid metabolism, and a corresponding increase in oxidative metabolism and basal respiration. The three HDAC inhibitors were also tested in db/db mice. Mice treated with either MS275 or SAHA—but not MC1568—showed lower fasting glucose and insulin blood levels, improved glucose clearance during glucose tolerance testing, decreased triglycerides and nonesterified fatty acids, as well as reduced lipid accumulation in the liver. Mice treated with MS275 also had significant weight loss. Other results showed that when db/db mice were treated for 15 days with MS275, they demonstrated increased oxygen consumption and heat production. These effects were shown to be mediated by increased Pgc-1α expression resulting from HDAC3 inhibition. The class I HDAC inhibitors had a number of effects on white adipose tissue (WAT), including reducing WAT by 18% in db/db mice. They also induced a partial brown adipose tissue (BAT) phenotype in WAT, as evidenced by increased expression of several genes associated with BAT. These data indicate that inhibitors of class I HDACs show promise as a new treatment paradigm. While the findings of decreased insulin resistance and weight loss indicate that these agents may show promise for the treatment of diabetes and obesity, the reduction of circulating lipids also points to a possible benefit in the treatment of cardiovascular disease risk factors. — Deborah Elbaum, MD
A Causative Role for CRP in Glucose Dysregulation
In this issue of Diabetes (p. ), Tanigaki et al. demonstrate that in mice, elevated C-reactive protein (CRP) plays a causative role in insulin resistance. Although numerous human studies have shown that modestelevations of CRP are associated with insulin resistance and diabetes, it has been unclear whether the increase in CRP or another factor that is associated with both CRP and diabetes—such as obesity—is the root cause of glucose dysregulation. By employing a series of experiments using wild-type and CRP-overexpressing transgenic (TG-CRP) mice, the researchers show that CRP’s role in glucose abnormalities appears to be mediated by Fcγ receptors (FcγRs). At both 5 and 13 weeks of age, TG-CRP mice had elevated fasting glucose and insulin levels, as well as abnormal glucose and insulin tolerance relative to their wild-type counterparts. These manifestations of abnormal glucose homeostasis preceded any increase in fat mass. It was not until 13 weeks of age that TG-CRP mice demonstrated significantly higher fat mass. The researchers studied CRP’s effects on Fcγ receptor IIB (FcγRIIB) using FcγRIIB−/−;TG-CRP and FcγRIIB+/+;TG-CRP mice. While FcγRIIB−/−;TG-CRP mice had normal glucose uptake, FcγRIIB+/+;TG-CRP mice had significantly elevated fasting glucose and insulin levels, pointing to the role of FcγRs in cellular glucose uptake. FcγRs were abundant in skeletal muscle microvascular endothelium, and compared with wild-type mice, TG-CRP mice had 39% lower glucose uptake in skeletal muscle. FcγRs were not abundant in myocytes, adipocytes, or hepatocytes, and CRP had no significant effect on pancreatic insulin secretion, hepatic glucose production, or adipose glucose uptake. These new data not only suggest that assessing and monitoring CRP levels in patients at risk for diabetes may be clinically relevant for risk stratification, but they also point to the potential value of new therapies targeting the FcγRIIB-mediated effects of CRP. — Deborah Elbaum, MD
FcγRIIB expression in skeletal muscle assessed by immunofluorescence in tissue from FcγRIIB+/+ vs. FcγRIIB−/− mice
FcγRIIB expression in skeletal muscle assessed by immunofluorescence in tissue from FcγRIIB+/+ vs. FcγRIIB−/− mice
