3',4'-Dihydroxyflavonol antioxidant attenuates diastolic dysfunction and cardiac remodeling in streptozotocin-induced diabetic m(Ren2)27 rats

Abstract

Background: Diabetic cardiomyopathy (DCM) is an increasingly recognized cause of chronic heart failure amongst diabetic patients. Both increased reactive oxygen species (ROS) generation and impaired ROS scavenging have been implicated in the pathogenesis of hyperglycemia-induced left ventricular dysfunction, cardiac fibrosis, apoptosis and hypertrophy. We hypothesized that 3',4'-dihydroxyflavonol (DiOHF), a small highly lipid soluble synthetic flavonol, may prevent DCM by scavenging ROS, thus preventing ROS-induced cardiac damage.

Methodology/principal findings: Six week old homozygous Ren-2 rats were randomized to receive either streptozotocin or citrate buffer, then further randomized to receive either DiOHF (1 mg/kg/day) by oral gavage or vehicle for six weeks. Cardiac function was assessed via echocardiography and left ventricular cardiac catheterization before the animals were sacrificed and hearts removed for histological and molecular analyses. Diabetic Ren-2 rats showed evidence of diastolic dysfunction with prolonged deceleration time, reduced E/A ratio, and increased slope of end-diastolic pressure volume relationship (EDPVR) in association with marked interstitial fibrosis and oxidative stress (all P<0.05 vs control Ren-2). Treatment with DiOHF prevented the development of diastolic dysfunction and was associated with reduced oxidative stress and interstitial fibrosis (all P<0.05 vs untreated diabetic Ren-2 rats). In contrast, few changes were seen in non-diabetic treated animals compared to untreated counterparts.

Conclusions: Inhibition of ROS production and action by DiOHF improved diastolic function and reduced myocyte hypertrophy as well as collagen deposition. These findings suggest the potential clinical utility of antioxidative compounds such as flavonols in the prevention of diabetes-associated cardiac dysfunction.

Figure 1. Representative pressure volume (PV) loops during preload reduction.

Note the steeper slope of EDPVR (green line) and rightward shift in the diabetic group (C) compared to control (A). An increase in the slope of EDPVR indicated decreased chamber compliance as seen in the diabetic animals which was reduced with DiOHF treatment (D). *P<0.05 vs control (non-diabetic) rats; †P<0.05 vs diabetic rats.

Figure 2. Representative apical view pulsed wave Doppler studies.

In control rats (A) “normal” left ventricular (LV) filling pattern were observed, whereas diabetic rats (B) had reduced LV function. These changes were distinguished by delayed ventricular relaxation, characterized by reduction in the E wave and an increase in the late atrial A wave and prolongation of the deceleration time. Treatment with DiOHF showed improvement in LV function of diabetic rats (C).

Figure 3. Immumohistochemistry staining for collagen type I.

(A) Representative images. Brown region represents positive immunostaining for collagen type I which was substantially increased in diabetic rats and reduced by DiOHF treatment. The amount of collagen type I was found to be similar in control and control-treated rats. Collagen type I interstitial accumulation as assessed by percentage proportional area showing positive immunostaining in control and diabetic rats treated with or without DiOHF (B). Data expressed as mean ± SE. *P<0.05 vs control (non-diabetic) rats; †P<0.05 vs diabetic rats. Original magnification ×200.

Figure 4. Immunohistochemistry staining for collagen type III.

(A) Representative images. Brown region represents positive immunostaining for collagen type III which was substantially increased in diabetic rats and reduced by DiOHF treatment. The amount of collagen type III was found to be similar in control and control-treated rats. Collagen type III interstitial accumulation as assessed by proportional area on sections showing positive immunostaining in control and diabetic rats treated with or without DiOHF (B). Data expressed as mean ± SE. Data expressed as mean ± SE. *P<0.05 vs control (non-diabetic) rats; †P<0.05 vs diabetic rats. Original magnification ×200.

Figure 5. Representative images for myocyte hypertrophy.

(A) Diabetic rats demonstrated myocyte hypertrophy as evidenced by increased cross sectional area when compared with control rats. Treatment with DiOHF reduced cross sectional area in diabetic rats but had no effect on control rats. (B) Quantitative data for myocyte cross sectional area. *P<0.05 vs control (non-diabetic) rats; †P<0.05 vs diabetic rats. Original magnification ×200.

Figure 6. Measurements of oxidative stress.

Representative images for the localization of 3-nitrotyrosine in LV tissues, as assessed by the proportional area of positive immunostaining in control and diabetic rats treated with or without DiOHF (A). The marked increase in positive immunostaining (brown region) for 3-nitrotyrosine in diabetic rats was reduced by DiOHF treatment. The amount of 3-nitrotyrosine was found to be comparable in control and control-treated Ren-2 rats (B). Original magnification ×200. (C) NADPH-activated superoxide production in LV tissues as measured by lucigenin-enhanced chemiluminescence. Data expressed as mean ± SE. *P<0.05 vs control (non-diabetic) rats; †P<0.05 vs diabetic rats.

Figure 7. Intracellular antioxidants mRNA gene expression.

(A) Measurement of Cu/Zn-superoxide dismutase (SOD1) and (B) gluthathione peroxidase (Gpx1) mRNA by real time RT-PCR, normalized to the housekeeping gene 18s. Diabetic rats demonstrated increased gene expression of SOD1 and Gpx1, which were not affected by DiOHF. Data expressed as mean ± SE. *P<0.001 vs control (non-diabetic) rats.

Figure 8. Txnip mRNA gene expression.

(A) Representative autoradiographs from in situ hybridization of Txnip. The magnitude of gene expression is indicated quantitatively as a pseudocolourized computer image (blue, nil; green, low; yellow, moderate; red, high). (B) Quantitation of Txnip gene expression by quantitative autoradiography. (C) Measurement of Txnip mRNA by real time RT-PCR, normalized to the housekeeping gene 18s. Data expressed as mean ± SE. *P<0.05 vs control (non-diabetic) rats; †P<0.05 vs diabetic rats.

Figure 9. Txnip protein expression.

(A) Representative western blot and (B) quantitation of Txnip normalized to loading control pan-actin. The significant increase in Txnip protein in diabetic rats was moderately reduced by DiOHF treatment. Data expressed as mean ± SE. *P<0.05 vs control (non-diabetic) rats.