Diabetic Cardiomyopathy in OVE26 Mice Shows Mitochondrial ROS Production and Divergence Between In Vivo and In Vitro Contractility

Abstract

Many diabetic patients suffer from a cardiomyopathy that cannot be explained solely by poor coronary perfusion. This cardiomyopathy may be due to either organ-based damage like fibrosis, or to direct damage to cardiomyocytes. Mitochondrial-derived reactive oxygen species (ROS) have been proposed to contribute to this cardiomyopathy. To address these questions, we used the OVE26 mouse model of severe type 1 diabetes to measure contractility in isolated cardiomyocytes by edge detection and in vivo with echocardiography. We also assessed the source of ROS generation using both a general and a mitochondrial specific indicator. When contractility was assayed in freshly isolated myocytes, contraction was much stronger in control myocytes. However, contractility of normal myocytes became weaker during 24 hours of in vitro culture. In contrast, contractility of diabetic OVE26 myocytes remains stable during culture. Echocardiography revealed normal or hyperdynamic function in OVE26 hearts under basal conditions but with a sharply reduced response to isoproterenol, a beta-adrenergic agonist. For ROS generation, we found that ROS production in diabetic myocytes was elevated after exposure to either high glucose or angiotensin II (AngII). Superoxide detection with the mitochondrial sensor MitoSOX Red confirmed that mitochondria are a major source of ROS generation in diabetic myocytes. These results show that contractile deficits in OVE26 diabetic hearts are due primarily to cardiomyocyte impairment and that ROS from mitochondria are a cause of that impairment.

Figure 1. Contractile properties of isolated myocytes

Myocytes were assayed immediately after isolation (fresh) and after culture (1 h and 24 h) from five month old FVB mice (white columns) or OVE26 mice (black columns). L: lengthening (in μm). t: time (in seconds). A: maximal velocities of cell shortening (+dL/dt). B: re-lengthening (-dL/dt). C: cardiomyocyte shortening indicated as the peak contraction expressed as % of basal myocyte length (bl_peak). Values are mean ± SE, from at least 4 mice per time point and 15 cells measured per mouse. Asterisk indicates that OVE26 cardiomyocytes have lower function than FVB myocytes at the same time point. Number sign indicates that FVB myocytes have stronger function at that time point than those at all longer time points. OVE26 myocyte contractility was not significantly affected by time after isolation.

Figure 2

Quantitation of ROS production from FVB (white columns) and OVE26 (black columns) myocytes measured with CM-H₂DCFDA after exposure to normal glucose (NG, 5.5 mmol/l), high glucose (HG, 25 mmol/l), angiotensin II (AngII, 50 nmol/l) and H₂O₂ (40 μmol/l) for 1 h. DCF: dichlorodihydrofluorescein. Asterisk indicates that OVE26 fluorescence was greater than FVB under the same culture conditions and greater than OVE26 under normal glucose (p < 0.05, by 2-way ANOVA). Fluorescence after H₂O₂ was greater than all other conditions (p < 0.01) but OVE26 was no different from FVB. Cardiomyocytes were isolated from 180-220 day old mice of both sexes. Data are mean ± SE. n = 4 mice/group, assayed in triplicate.

Figure 3

Mitochondrial ROS levels detected by MitoSOX Red after exposure of FVB (white columns) and OVE26 (black columns) cardiomyocytes to high glucose or AngII. Asterisk indicates p < 0.05 vs. all other values except OVE26 with Ang II or high glucose by 2-way ANOVA. NG: normal glucose (5.5 mmol/l). HG: high glucose (25 mmol/l). AngII: 5.5 mmol/l glucose + angiotensin II (50 nmol/l). Values are the mean ± SE of at least 120 myocytes from 3 mice of each type.

Figure 4

Representative images of cardiomyocytes after incubation with CM-H₂DCFDA (A and B) and MitoSOX Red (C and D, 20×; E and F, 100×). Myocytes on the left (A, C and E) represent ROS baseline, myocytes on the right (B, D and F) represent ROS induced by high glucose or angiotensin II in OVE26 mice.