Hyperglycaemia-induced reactive oxygen species production in cardiac ventricular myocytes differs among mammals

Diabetic cardiomyopathy (DbCM) preclinical studies have used animal models from different species and implicated altered redox regulation and reactive oxygen species (ROS) in this disease. However hyperglycaemia-induced ROS signalling may differ in ventricular myocytes from different species, and here we compare mice, rats, rabbits and humans with respect to hyperglycaemia-induced ROS production.

Using the chemical sensor H₂DCFDA and genetically encoded redox sensors, we measured time-dependent ROS signals in response to high glucose (HiGlu). Live cell confocal imaging was used to determine redox and glucose levels, contractility and Ca²⁺ transients in isolated cardiomyocytes. Western blotting and activity assays were used to evaluate superoxide dismutase (SOD) expression and activity.

Consistent with our prior mouse studies, HiGlu exposure in rat and human ventricular myocytes increased cytosolic (not mitochondrial) ROS production via a mechanism involving Ca-calmodulin-dependent protein kinase (CaMKII) O-GlcNAcylation (at Ser280) and NADPH oxidase 2 (NOX2) activation. However in rabbit ventricular myocytes HiGlu alone did not induce significant ROS production, unless complemented by low levels of angiotensin II and inhibition of the enzyme that reverses O-GlcNAcylation using 100 nM Thiamet G, where both cytosolic and mitochondrial ROS accumulation were involved by more complex signalling networks.

Acute hyperglycaemia readily induces robust ROS production in mouse, rat and human ventricular myocytes, but rabbit myocytes are intrinsically better protected from this HiGlu-induced ROS responses. In most regards rabbit are much more human-like than rodents for many aspects of myocyte function, but this surprisingly is not the case for HiGlu-induced ROS production.

In mouse ventricular myocytes we previously showed that acute hyperglycaemia (HiGlu) induced increase in reactive oxygen species (ROS) mediated by O-GlcNAcylation of Ca-calmodulin-dependent protein kinase at Ser280 and consequent NADPH oxidase 2 (NOX2) activation. Here we demonstrate that this same pathway is functional in human and rat ventricular myocytes but not in rabbit ventricular myocytes. In rat and human myocytes HiGlu induced an increase in cytosolic, but not mitochondrial, ROS production. In rabbit myocytes HiGlu only produced increased ROS (cytosolic and mitochondrial) only when it was accompanied by both low concentration angiotensin II and an O-GlcNAcylase inhibitor (that individually did not promote ROS production). This direct HiGlu-induced ROS production in cardiac myocytes may contribute to pathological signalling in diabetes.