Attenuation of Ca 2+ homeostasis, oxidative stress, and mitochondrial dysfunctions in diabetic rats heart: insulin therapy or aerobic exercise?

1

1

Attenuation of Ca2+ homeostasis, oxidative stress, and mitochondrial dysfunctions in diabetic rat heart: insulin therapy or aerobic 2

exercise? 3

4

Márcia F. da Silvaa, Antônio J. Natalib*; Edson da Silvaac, Gilton J. Gomesb, Bruno G. Teodorod, Daise N.Q. Cunhab, Lucas R. Drummondb, 5

Filipe R. Drummondb, Anselmo G. Mourab, Felipe G. Belfortb, Alessandro de Oliveirab, Izabel R. S. C. Maldonadoa*, Luciane C. Albericid*# 6

7

*The authors share senior authorship. 8 9 aDepartamento de Biologia Geral, Universidade Federal de Viçosa, Viçosa, MG, Brasil. 10 bDepartamento de Educação Física, Universidade Federal de Viçosa, Viçosa, MG, Brasil. 11 cDepartamento de Ciências Básicas, Universidade Federal do Vale do Jequitinhonha e Mucuri, Diamantina, MG, Brasil. 12 dDepartamento de Física e Química, Universidade de São Paulo, Ribeirão Preto, SP, Brasil. 13 14 #Corresponding author: Luciane Carla Alberici 15

Address: Universidade de São Paulo, Faculdade de Ciências Farmacêuticas de Ribeirão Preto 16

Av. Do Café s/n, Ribeirão Preto, São Paulo, Brazil. 17

Post code: 14040-903 18

Phone: +55 (16) 36024435 Fax: +55 (16) 36024880 19

e-mail: [email protected] 20

21

Articles in

PresS

. J Ap

pl P

hysio

l (May 21, 2015). d

oi:10.1152/jap

plp

hysio

l.00915.2014

C

op

yrigh

t © 2015 b

y the A

merican

Ph

ysiolo

gical S

ociety.

2

Abbreviations: 22

ATP - Adenosine triphosphate 23

BG - Blood glucose 24

Ca2+ - Calcium 25

CaMKII - Ca2+/calmodulin-dependent protein kinase 26

CE - Control exercised 27

CS - Control sedentary 28

DE - Diabetic exercised 29

DEI - Diabetic trained treated plus insulin 30

DS - Diabetic sedentary 31

DSI - Diabetic sedentary 32

EF - Ejection fraction 33

EGTA - Ethylene glycol tetra-acetate 34

FS - Fractional shortening 35

GSH - Reduced glutathione 36

GSSH - Oxidized glutathione 37

IRS - insulin receptor substrate 38

LV- Left ventricle 39

LVM- Left ventricular myocytes 40

MPTP - Mitochondrial permeability transition pore 41

NADPH - Nicotinamide adenine dinucleotide phosphate 42

NCX - Sodium-calcium exchanger 43

NHE - Sodium-hydrogen exchanger 44

3

Nox-4 - NADPH oxidase 4 45

RaM - Rapid uptake mode 46

RHR - Resting heart rate 47

ROS - Reactive oxygen species 48

RyR - Ryanodine receptor 49

SERCA - Sarcoplasmc reticulum calcium ATPase 50

SOD - Superoxide dismutase 51

STZ - Streptozootocin 52

UCP - Uncoupling protein 53

54

4

Abstract 55

We tested the effects of swimming training and insulin therapy, either alone or in combination, on the intracellular calcium ([Ca2+]i) homeostasis, 56

oxidative stress and mitochondrial functions in diabetic rat hearts. Male Wistar rats were separated into control, diabetic or diabetic plus insulin 57

groups. Type 1 diabetes mellitus was induced by streptozotocin (STZ). Insulin treated groups received 1 to 4 UI of insulin daily for eight weeks. 58

Each group was divided into sedentary or exercised rats. Trained groups were submitted to a swimming (90 min/day, 5 days/week, 8 weeks). 59

[Ca2+]i transient in left ventricular myocytes (LVM), oxidative stress in left ventricular (LV) tissue and mitochondrial functions in the heart were 60

assessed. Diabetes reduced the amplitude and prolonged the times to peak and to half decay of the [Ca2+]i transient in LVM, increased NADPH 61

oxidase-4 (Nox-4) expression, decreased superoxide dismutase (SOD) and increased carbonyl protein contents in LV tissue. In isolated 62

mitochondria, diabetes increased Ca2+ uptake, susceptibility to permeability transition pore (MPTP) opening, UCP-2 expression and oxygen 63

consumption, but reduced H2O2 release. Swimming training corrected the time course of the [Ca2+]i transient, UCP-2 expression and 64

mitochondrial Ca2+ uptake. Insulin replacement further normalized [Ca2+]i transient amplitude, Nox-4 expression and carbonyl content. 65

Alongside these benefits, the combination of both therapies restored the LV tissue SOD and mitochondrial O2 consumption, H2O2 release and 66

MPTP opening. In conclusion, the combination of swimming training with insulin replacement was more effective in attenuating intracellular 67

Ca2+ disruptions, oxidative stress and mitochondrial dysfunctions in STZ-induced diabetic rat hearts. 68

69

Keywords: Physical activity; Diabetes mellitus; Mitochondria; Nox-4 70

71

72

73

74

5

1. Introduction 75

Diabetic cardiomyopathy leads initially to diastolic dysfunction which frequently progress to heart failure and sudden death (12). Cardiac 76

systolic and diastolic dysfunctions are associated with impaired intracellular calcium (Ca2+) homeostasis (16, 23, 43, 48) due to decreased 77

expression and/or activity of Ca2+ regulatory proteins (48, 23, 43, 16, 10, 47). 78

Heart failure induced by diabetes is also associated with increased reactive oxygen species (ROS) (19) as well as with nicotinamide 79

adenine dinucleotide phosphate (NADPH) oxidase activation by glycated proteins, and mitochondrial dysfunction (47). Mitochondrial 80

dysfunctions in diabetic hearts are related with increased expression of mitochondrial uncoupling protein-3 (UCP-3) (8, 19, 44) impaired 81

respiratory capacity, altered expression of respiratory chain complexes (8) and Ca2+ uptake, higher susceptibility to mitochondrial permeability 82

transition pore (MPTP) opening and elevated apoptotic signaling molecules (8, 26). Recently, it has been demonstrated that in diabetic hearts the 83

expression of phosphorylated Ca2+/calmodulin-dependent protein kinase II (CaMKII) and NADPH oxidase were augmented and its activation by 84

impaired Ca2+ metabolism would increase ROS production (32). 85

Aerobic exercise and insulin replacement are strategies to management of diabetes (41, 46). Endurance exercise training was shown to 86

improve cardiomyocyte Ca2+ cycling and restore its intracellular calcium ([Ca2+]i) transient and hence contractile function in diabetic rats (41). 87

Endurance exercise is also known to protect the rat myocardium against oxidative stress (22). The stimulation of Ca2+ uptake by insulin 88

replacement is involved in the regulation of heart metabolism and transporter activities (34). However, the effects of combined endurance 89

exercise training and insulin treatment on cardiac oxidative stress and heart mitochondrial function of diabetic rats are poorly understood. Then, 90

this study sought to examine the effects of swimming training combined with insulin treatment on cardiac oxidative stress and mitochondrial 91

dysfunctions in streptozotocin (STZ)-induced diabetic rats. 92

93

6

2. Materials and methods 94

95

2.1. Experimental animals 96

Male Wistar rats (30 days; 80.2 ± 1.8g) had free access to chow and water and were housed at 22 ± 2 °C on a 12h light-dark cycle, and 97

were separated in control sedentary (CS), control exercised (CE), diabetic sedentary (DS), diabetic exercised (DE), diabetic sedentary insulin 98

(DSI) and diabetic exercised insulin (DEI). Experiments were approved by the Ethics Committee of the Federal University of Viçosa (Protocol 99

number 51/2011). 100

101

2.2. Induction of diabetes and insulin treatment 102

Diabetic groups received an intraperitoneal injection [60 mg/kg of body weight (BW)] of STZ (Sigma, St. Louis, USA) diluted in sodium 103

citrate buffer (0.1 M, pH 4.5), and control groups received the buffer. Seven days after, animals with fasting blood glucose (BG) above 300 104

mg/dL were considered diabetic. Animals from DEI and DSI received a daily dose of human insulin (1-4 U/day). 105

106

2.3. Exercise training protocol 107

Animals from CE, DE and DEI groups were subjected to a swimming training program (five days/week/eight weeks) (Adapted from 108

[13]). On the first week the animals exercised with no load for 10-50 min/day and exercise duration was increased by 10 min/day. In the second 109

week, the animals exercised with a load corresponding to 1% of BW and the exercise duration was increased by 10 min/day up to a total of 90 110

min of continuous swimming in one session. From the third week on, the load was increased weekly (1% of BW/wk) up to a load of 5% of BW 111

on the eighth week. 112

113

2.4. Resting heart rate assessment 114

7

Resting heart rate (RHR) was obtained from electrocardiogram. Animals were placed inside a chamber for anesthesia (isoflurane 2% and 115

oxygen 100%) at a constant flow of 1L/minute. The electrocardiogram (DII) was acquired using the data acquisition system PowerLab ® (AD 116

Instr., SP, Brazil) and data analyzed using the program Lab Chart Pro ® (AD Instr. Lab Chart 7, SP, Brazil). Heart rate was obtained by average 117

of five consecutive cardiac cycles. 118

119

2.4. Echocardiographic examinations 120

Animals were anesthetized with isoflurane via mask (isoflurane 3% and oxygen 100% constant flow (1 L/min). Cardiac contraction and 121

relaxation were assessed noninvasively by transthoracic echocardiography using parasternal long- and short-axis images. Two-dimensional, M-122

mode echocardiographic images and color-guided pulsed-wave Doppler images were obtained by standard echocardiographic techniques (11) 123

(MyLabTM30, Esaote, Genova, Itália). The systolic function was assessed using the ejection fraction (EF) and fractional shortening (FS) while the 124

diastolic function was assessed using the mitral flow data [i.e. early filling wave (peak E); late filling wave (peak A) and E/A ratio)]. 125

126

2.5. Isolation of cardiomyocytes and mitochondria from left ventricle (LV) 127

Two days after the last exercise training session the rats were euthanized by cervical dislocation and their hearts were removed. LV 128

myocytes were enzymatically isolated [Adapted from (9)], using 1 mg/mL of collagenase type II (Worthington, USA) and 0.1 mg/mL of protease 129

(Sigma). LV mitochondria were isolated by standard differential centrifugation (40). 130

131

132

2.6. Intracellular Ca2+ measurements 133

Intracellular calcium ([Ca2+] i) transients in cardiomyocytes were evaluated as described (35) using 5 μM fluo-4 AM (Molecular Probes, 134

Eugene, OR, USA). A Meta LSM 510 scanning system (Carl Zeiss GmbH, Germany) with an x63 oil immersion objective was used for confocal 135

8

fluorescence imaging (488/510 nm ex/em). Digital image processing was performed using routines custom-written in the Matlab® platform. The 136

amplitude of the [Ca2+]i transient, measured as fluorescence ratio (F/F0), with fluorescence intensity (F) normalized to the minimal intensity 137

measured between 1-Hz contractions at diastolic phase (F0), the time to peak and the time from the peak to half resting level of the [Ca2+]i 138

transient were determined. 139

140

2.7. Heart Redox State 141

Reduced (GSH) and oxidized (GSSG) glutathione and protein carbonyl levels were assessed in LV tissue (50 mg/mL in cold 0.1 M Tris-142

HCl buffer, pH 7.4) by the fluorimetric ortho-phthalaldehyde method (20). Protein carbonyl were detected spectrophotometrically by 143

derivatisation of the carbonyl group with 2,4-dinitrophenylhydrazine (39, 40). Superoxide dismutase (SOD) content in LV tissue was determined 144

spectrophotometrically by the tetrazolium salt formation using a Superoxide Dismutase Assay Kit (Cayman®). 145

146

2.8. Mitochondrial standard incubation procedure and assays 147

Mitochondria were energized with 5 mM glutamate and malate in an incubation medium containing of 125 mM sucrose, 65 mM KCl, 2 148

mM K2HPO4, 10 mM HEPES-KOH pH 7.4, at 30 °C. Mitochondrial respiration was monitored using a temperature-controlled computer-149

interfaced Clark-type oxygen electrode (Oxytherm, Hansatech Inst, UK). Hydrogen peroxide production was monitored spectrofluorimetrically 150

using 1 µM Amplex red (Molecular Probes, OR, USA) and 1 UI/mL horseradish peroxidase (49); these assays were performed in the presence of 151

0.1 mM ethylene glycol tetra-acetate (EGTA). Ca2+ influx and efflux were monitored spectrofluorimetrically using 150 nM Calcium Green 5N 152

(Molecular Probes, OR, USA) (36). Ca2+ flux and H2O2 production were monitored in a Model F-4500 Hitachi fluorescence spectrophotometer 153

at the 506/531 and 563/587 (ex/em) wavelength pair, respectively. 154

155

2.9. Analysis of mRNA expression 156

9

RNA was isolated using the TRIzol reagent (Invitrogen®). For real-time PCR analysis, RNA was reverse transcribed using the Reverse 157

Transcriptase IMPROM II (PROMEGA®) and used in quantitative PCR reactions containing EVA-green fluorescent dye (BioRad®). Relative 158

expression of mRNAs was determined after normalization by β-actin using the ΔΔCt method (4). Quantitative PCR was performed using 159

Eppendorf Realplex4 Mastercycler Instrument (Eppendorf®). Primers for NOX4, β-actin and uncoupling protein-2 (UCP-2) were designed as 160

described in Table 1. 161

162

2.10. Statistical analysis 163

Normal distribution of the data was determined by the Shapiro–Wilk test. Non-normally distributed variables were log-transformed 164

before statistical analyses. Comparisons between groups were performed by using the factorial analysis 2 (sedentary x exercised) by 3 (control x 165

diabetes x insulin), followed by the Tukey post hoc test (Statistical Analysis System software - SAS version 9.3). Statistical significance was 166

defined at P ≤ 0.05. Results are presented as means ± SEM. 167

168

3. Results 169

170

3.1. General characteristics of rats 171

The initial BG levels were no different among the groups (Table 2). Streptozotocin augmented the final BG levels to approximately 500 172

mg/dL during the experimental period (factor effect: p<0.05). Insulin treatment reduced the final PG (factor effect: p<0.05) in diabetic rats by 173

approximately 35% (control: 88.70 ± 24.35; diabetes: 510.70 ± 24.3; insulin: 327.51± 24.34 – in mg/dL). However, neither exercise training 174

effect nor interaction between factors were observed (p>0.05). 175

The initial BW was not different among groups, however STZ-injected animals gained less BW (factor effect: p<0.05) than controls until 176

the end of the experimental period (control: 318.90 ± 12.39 g; diabetes: 194.80 ± 12.38; insulin: 216.71± 12.37 – in grams). Neither insulin 177

10

treatment nor exercise training alone altered significantly the BW gain in diabetic rats (factor effect: p<0.05). Interaction between factors was 178

observed (factor effect: p<0.05). For example, animals from DEI group exhibited higher BW as compared to those of both DS and DE groups. 179

Diabetes reduced the heart weight (HW) (factor effect: p<0.05), while insulin treatment attenuated such reduction (factor effect: p<0.05) 180

in diabetic rats (control: 1.92 ± 0.08; diabetes: 1.01 ± 0.08; insulin: 1.53 ± 0.08 – in grams). However, neither exercise training effect nor 181

interaction between factors were observed (p>0.05). The HW/BW was reduced in diabetic rats (factor effect: p<0.05) when compared to insulin-182

treated animals (control: 6.26 ± 0.48; diabetes: 5.28 ± 0.48; insulin: 7.25 ± 0.48 – in mg/g). Nevertheless, neither exercise training effect nor 183

interaction between factors were observed (p>0.05). 184

Resting heart rate was reduced by diabetes (factor effect: p<0.05), whereas insulin treatment corrected such effect (control: 341.60 ± 8.99; 185

diabetes: 272.10 ± 8.99; insulin: 346.70 ± 9.52 – in bpm). However, neither exercise training effect nor interaction between factors were 186

observed (p>0.05). 187

Diabetic animals exhibited reduced ejection fraction (control: 81.00 ± 1.16%; diabetes: 62.25 ± 3.21%; insulin: 67.46 ± 1.84%) and 188

fractional shortening (control: 45.57 ± 1.53%; diabetes: 30.05 ± 1.39%; insulin: 33.04 ± 1.32%) by the end of the experiment (factor effect: 189

p<0.05) which indicates systolic dysfunction (Table 3). Ejection fraction was not affected by either exercise training or insulin treatment (factor 190

effect: p>0.05), although interaction between factors was found (p<0.05). Animals from DE group showed higher EF than those from DS and 191

DEI groups, and animals from DEI group higher than those from DE and DSI groups. Exercise training increased the FS (sedentary: 34.18 ± 192

1.78%; exercised: 39.20 ± 2.12%), whereas no effect of insulin treatment was observed. However, interaction between factors was observed 193

(p<0.05). Rats from DEI group exhibited higher FS as compared to those from DSI and DE groups. Diabetic rats also showed higher E/A ratio 194

(factor effect: p<0.05) than controls (control: 1.72 ± 0.07; diabetes: 2.81 ± 0.21; insulin: 2.99 ± 0.11) which characterizes diastolic dysfunction. 195

Nevertheless, exercise training and insulin treatment, either isolated or in combination, did not reverse the diastolic dysfunction induced by 196

diabetes. 197

198

11

3.2. Intracellular global calcium transients 199

STZ-induced diabetes reduced (factor effect: p<0.05) the amplitude of the [Ca2+]i transient, whereas insulin treatment restored it to the 200

levels of controls (control: 2.27 ± 0.02 F/F0; diabetes: 2.12 ± 0.02 F/F0; insulin: 2.25 ± 0.02 - in F/F0). However, no effect of exercise was 201

observed (factor effect: p>0.05). There was interaction (p<0.05) between factors. Figure 1A shows that the [Ca2+]i transient amplitude in diabetic 202

myocytes (DS and DE) was significantly lower than that in controls (CS). In contrast, DSI group showed higher amplitude as compared to that 203

of DE group. The combination of exercise training with insulin treatment significantly increased the [Ca2+]i transient amplitude in DEI group as 204

compared to those of DS and DE groups. 205

The time to peak of the [Ca2+] i transient (Fig. 1B) was not affected by either diabetes or insulin treatment (factor effect: p>0.05). 206

Nevertheless, exercise training reduced (factor effect: p<0.05) the time to peak of the [Ca2+]i transient (sedentary: 77.00 ± 1.00; exercised: 71.00 207

± 1.00 – in ms). Such exercise effect was observed in CE, DE and DEI groups. No interaction between factors was found (p>0.05). 208

Diabetes prolonged the time to half decay of the [Ca2+] i transient (factor effect: p<0.05), while insulin treatment shortened it towards the 209

control levels (control: 221.00 ± 2.00; diabetes: 251.00 ± 2.00; insulin: 233.00 ± 2.00 in ms). In addition, exercise training reduced (factor effect: 210

p<0.05) the time to half decay of the [Ca2+] i transient (sedentary: 238.00 ± 2.00; exercised: 233.00 ± 2.00 in ms). Interaction (p<0.05) between 211

factors was observed. Figure 1C shows that the time to half decay of the [Ca2+] i transient in diabetic myocytes was longer than that of controls. 212

In contrast, insulin treatment itself (DSI group) as well as exercise training (DE group) and its combination (DEI group) shortened the time to 213

half decay of the [Ca2+] i transient in diabetic animals, as compared to DS group. 214

215

3.3. Heart Redox State 216

217

We observed that hearts of diabetic sedentary rats exhibited an increased expression of NADPH oxidase-4 (Nox-4) (factor effect: p<0.05), 218

as compared to control rats, which was reduced by insulin treatment (control: 0.86 ± 0.09; diabetes: 1.56 ± 0.19; insulin: 0.74 ± 0.12 – in RNA 219

12

relative expression). Exercise training had no isolated effect, however interaction between factors was observed (p<0.05). Control levels of Nox-220

4 expression in hearts of diabetic rats were reached when exercise training and insulin treatment were combined (Fig. 2A). 221

In normal conditions, the excess of O2•- is inactivated and converted to H2O2 by superoxide dismutase (SOD). However, along with the 222

increased Nox-4 expression hearts of our diabetic rats displayed a diminished SOD content (Fig. 2B; factor effect: p<0.05), which was not 223

reversed by insulin treatment (control: 43.82 ± 3.86; diabetes: 28.86 ± 1.43; insulin: 32.83 ± 2.43 – in U.mg-1). Exercise training showed no 224

independent effect (factor effect: p>0.05), however interaction between factors was found (p<0.05). Hearts from DEI group showed increased 225

SOD content as compared to those from DS, DSI and DE. 226

An additional indicative of low SOD content in hearts of our diabetic sedentary rats was verified by the reduced/oxidized glutathione 227

(GSH/GSSG), since GSH is consumed by glutathione peroxidase to degrade H2O2 generating GSSG and H2O. We found a slight increase (factor 228

effect: p<0.05; control: 1.77 ± 0.07; diabetes: 2.30 ± 0.11; insulin: 1.89 ± 0.16) in GSH/GSSG in diabetic hearts (Fig. 2C), indicating low 229

productions of H2O2, despite the apparent increase of O2•- production by elevated Nox-4 expression. Neither exercise training nor insulin 230

treatment had independent effect on GSH/GSSG (factor effect: p>0.05), nevertheless there was interaction between factors (p<0.05). Hearts from 231

DEI group showed lower GSH/GSSG than those from DE and DS groups. 232

The oxidative damage, as a consequence of an unbalanced production/inactivation of O2•-, was verified by the content of carbonyl 233

proteins (Fig. 2D). Proteins of diabetic rat hearts were more oxidized, an effect that was reversed by insulin treatment (factor effect: p<0.05; 234

control: 0.40 ± 0.07; diabetes: 0.84 ± 0.10; insulin: 0.37 ± 0.07 – in nmol.mg-1). Exercise training had no independent effect (factor effect: 235

p>0.05), however interaction between factors was observed (p<0.05). Hearts from DEI group showed lower carbonyl content than those from DS 236

and DE groups. Furthermore, the modulation of carbonyl contents promoted by diabetes and treatments (insulin or exercise) follow the profile 237

found for Nox-4 expression, suggesting linked effects. 238

239

3.4. Mitochondrial function 240

13

241

Diabetes did not affect (factor effect: p>0.05) the respiration rates (Table 4) in the phosphorylating state (State III) but increased the rates 242

in the resting state (State IV). Diabetes decreased (factor effect: p<0.05) the respiratory control ratios (RCR; control: 10.05 ± 0.58; diabetes: 7.39 243

± 0.60; insulin: 7.94 ± 0.36), indicating a weakness of the coupling between respiration and phosphorylation. Neither swimming training nor 244

insulin treatment of diabetic rats restored these respiratory parameters (factor effect: p>0.05) and no interaction between factors was observed for 245

such parameters (p>0.05). However, mitocondria from DEI group exhibited lower O2 consumption in State IV than those from DS, DSI and DE 246

groups. 247

In addition, LV of diabetic rats also presented an augment (factor effect: p<0.05) in the expression of UCP-2 (control: 0.77 ± 0.09; 248

diabetes: 1.83 ± 0.22; insulin: 1.13 ± 0.19 - in UCP/β-actin) (Fig. 3). These proteins located at the inner membrane that can dissipate the proton 249

gradient built by respiratory chain, promoting a mild uncoupling of the oxidative phosphorylation. UCP-2 mRNA was partially restored by 250

insulin treatment as an independent factor. Likewise, exercise training reduced (factor effect: p<0.05) the expression of UCP-2 (sedentary: 1.60 ± 251

0.21; exercised: 0.94 ± 0.11 - in UCP/β-actin), as an independent factor, as well as in diabetic rats when combined with insulin treatment 252

(interaction: p<0.05). For example, LV from DEI group exhibited lower UCP-2 mRNA expression than those from DS, DSI and DE groups. 253

The release of H2O2 in isolated heart mitochondria was monitored (Fig. 4). The representative experiment (Fig. 4A) and average data 254

(Fig. 4B) show that heart mitochondria from sedentary diabetic rats release lower amounts (factor effect: p<0.05) of H2O2 as compared to those 255

from controls (control: 0.67 ± 0.02; diabetes: 0.41± 0.01; insulin: 0.52 ± 0.04 – in nmol.mg-1.min-1). Neither exercise training nor insulin 256

treatment alone affected the reduced mitochondrial H2O2 release (factor effect: p>0.05). However, there was interaction between factors 257

(p<0.05). Swimming training associated with insulin reversed the reduced H2O2 release in heart mitochondria from diabetic rats, since heart 258

mitochondria from DEI group presented higher H2O2 values as compared to those of DS, DSI and DE groups. 259

Diabetes slightly increased the capacity of Ca2+ uptake (Fig. 5A) in isolated heart mitochondria (control: 180.00 ± 12.00; diabetes: 260

209.33 ± 20.27; insulin: 147.27 ± 15.55 – in nmol.mg-1), which was reduced below control levels by insulin treatment (factor effect: p<0.05). 261

14

Exercise training had no independent effect (factor effect: p<0.05), nevertheless interaction between factors was observed (p<0.05). Heart 262

mitochondria from DS group showed higher capacity of Ca2+ uptake than those from DE and DSI and mitochondria from DEI group exhibited 263

higher capacity of Ca2+ uptake when compared to those from DSI group. It indicates that exercise training either alone or in combination with 264

insulin restored the Ca2+ uptake to control levels in diabetic animals. 265

The Ca2+ retention capacity was monitored to check the susceptibility to MPTP opening in isolated heart mitochondria (Fig. 5B). Traces 266

of the external Ca2+ concentration dynamics in response to sequential additions of Ca2+ show that heart mitochondria isolated from DS rats did 267

not sustain the preloaded Ca2+ (after 6 additions of 20 nmol Ca2+), in contrast to those isolated from CS rats, indicating an increased susceptibility 268

to Ca2+-induced inner membrane permeabilization (Fig. 5B). This condition was only reversed by exercise training combined with insulin 269

treatment (DEI group). Surprisingly, both swimming training and insulin treatment separately further reduced the capacity of Ca2+ retention in 270

heart mitochondria of diabetic rats. Cyclosporin A (CSA), a classical inhibitor of MPTP opening, totally prevented the release of accumulated 271

Ca2+ in heart mitochondria of DS, DE or DEI groups, and partially prevented it in heart mitochondria of DSI (Fig. 5C). Together, these results 272

indicate that diabetes induces high Ca2+ uptake and thus increases the susceptibility to MPTP opening in heart mitochondria. Higher Ca2+ uptake 273

and the susceptibility to MPTP opening, as well as the respiratory, UCP-2 expression and H2O2 release patterns were reversed by insulin 274

treatment combined with exercise training. 275

276

4. Discussion 277

278

We examined the effects of swimming training and insulin therapy, either alone or in combination, on the intracellular Ca2+ homeostasis, 279

oxidative stress and mitochondrial functions in diabetic rat hearts. Our data showed that endurance training associated with insulin treatment was 280

more effective in attenuating intracellular Ca2+ homeostasis disruptions, cardiac oxidative stress and mitochondrial dysfunctions caused by STZ-281

induced diabetes in rat hearts. 282

15

Here, diabetes impaired the in vivo cardiac systolic and diastolic functions as it reduced the ejection fraction and fractional shortening as 283

well as it increased peak E/peak A ratio in rats. These dysfunctions were also evident at the cellular level inasmuch as diabetes reduced the 284

amplitude and prolonged the time to half decay of the [Ca2+]i transient in left ventricular myocytes. These disruptions in experimental diabetes 285

have been shown previously and are related to dysfunctions in the cellular Ca2+ regulatory proteins (23). Regarding the systolic dysfunction, the 286

Ca2+ release from the SR is not uniform in cardiomyocytes of diabetic rodents which is due to reductions in the number of functional ryanodine 287

receptor 2 (RyR2) (6), reduced Ca2+ channel activities (23) and depressed sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2) expression or 288

activity and hence intra-SR Ca2+ stores (23). This would help to explain the observed reduced [Ca2+]i transient amplitude. As for the diastolic 289

dysfunction, the prolonged Ca2+ transient decay observed here is consistent with a prolonged cytosolic Ca2+ removal via decreased SERCA2 290

expression or activity in diabetic hearts (23). It is noteworthy that an increased activity of sodium-calcium exchanger (NCX) may contribute to 291

reduced SR Ca2+ load in diabetic myocytes, although the relative contribution of NCX to cytosolic Ca2+ removal in rats is small (~ 7%) and in 292

diabetic hearts increased or unchanged NCX activity or decreased NCX expression were reported (10). Nevertheless, as we did not measure 293

either the expression or the activity of these proteins, their role in these results found in this study should be taken with caution. 294

Despite the fact that cytosolic removal of Ca2+ into mitochondria in rats is small (~ 1%) and seems not to impact on the [Ca2+]i transient 295

(5), we observed an increased capacity of mitochondrial Ca2+ uptake in mitochondria isolated from diabetic hearts. Mitochondria are sensitive to 296

Ca2+ concentration, which is important for the modulation of mitochondrial metabolism. These organelles may act as a Ca2+ buffering system 297

removing and modulating the local Ca2+ concentration. Calcium is transported through the inner membrane matrix by two mechanisms: a 298

uniporter (MCU) and a rapid uptake mode (RaM) that are dependent of the electrochemical gradient for Ca2+. Calcium flux rates through the 299

MCU are fast, equivalent to fast gated pores but slower than the most channels. Calcium uptake through the RaM occurs very rapidly, faster than 300

MCU and at the beginning of Ca2+ pulse. The localization of mitochondria close to Ca2+ release sites of the endoplasmic or sarcoplasmic 301

reticulum or even near plasma membrane Ca2+ channels also facilitates mitochondrial Ca2+ handling [see (15) for review]. Furthermore, in 302

different cell types an increased peak of mitochondrial Ca2+ concentration in response to Ca2+ mobilizing stimuli has been described to be 303

16

increased by UCP-2 and UCP-3 overexpression (45). In fact, we found elevated UCP-2 expression and indications of increased UCP activity, 304

such as enhanced resting respiration rate and reduced mitochondrial H2O2 release in diabetic hearts. Although the role of UCPs in the heart be 305

controversial and incompletely understood, mild uncoupling promoted by UCP decreases ROS formation by accelerating electron transport rates 306

through respiratory chain, thus decreasing the lifetime of intermediates capable of donating electrons toward superoxide radical formation (42). 307

This process could represent a mechanism to protect the cell against oxidative damage. Anyway, our findings agree with previous works (18) 308

showing that mitochondria per se apart from the insulin-deficient diabetic profile remain free from ROS production, despite considerable 309

evidences suggesting that diabetes is associated with oxidative tissue damage in the heart of diabetic rodents (2). However, mitochondrial 310

uncoupling may augment oxygen consumption without proportionately increasing mitochondrial ATP production. The resulting energy deficit 311

may explain the lack of increase in cardiac systolic and diastolic functions, resulting in reduced cardiac efficiency, as found hearts of db/db mice 312

(7). 313

Our results showed that cardiac oxidative damage could be related to upregulation of Nox-4 expression, as showed previously by 314

Maalouf and coworkers (27) and also found in LV dysfunctions such as heart failure progression and aging (1). Nox-4 is localized in perinuclear 315

organelles, including mitochondria (1), and is supposed to be constitutively active, not requiring cytosolic factors for its activation. Therefore, its 316

expression levels determine the amount of O2•- production in the cells. Oxidative damage is also favored when O2

•- production is associated to an 317

impaired antioxidant system, as demonstrated here by diminished SOD in hearts of diabetic rats. Reductions in SOD content and/or activity have 318

been shown previously in STZ-induced diabetic cardiomyopathy (28). Thus, oxidative damage in LV of diabetic rats could be a consequence of 319

high production and low inactivation of O2•- promoted by high Nox-4 expression and low SOD content. We also suggest that increased Nox-4 320

activity could also sensitize MPTP opening in the heart mitochondria of STZ-induced diabetic rats, inasmuch as Nox-4 activity leads to cysteine 321

oxidation of mitochondrial proteins including components of the MPTP complex and mitochondrial damage as reported previously (1). This 322

sensibility associated to an excessive mitochondrial Ca2+ load could trigger MPTP opening, following dissipation of the inner mitochondrial 323

membrane potential and swelling. 324

17

It is possible that the impaired intracellular Ca2+ homeostasis in our diabetic rat heart have increased oxidative stress by activating 325

CaMKII. Nishio et al. (30) demonstrated an augmented [Ca2+]i in cardiomyocytes exposed to high glucose concentrations due to increased 326

sodium-hydrogen exchanger (NHE) expression and activity which activated NCX in reverse mode. High glucose also up-regulated the 327

phosphorylated CaMKII expression that was suppressed by inhibiting NCX in reverse mode. In addition, a CaMKII inhibitor attenuated the ROS 328

level in these myocytes. In STZ-induced diabetic rat hearts they observed up-regulation of ROS level and components of NADPH oxidase, 329

p47phox and p67phox, which were attenuated by a CaMKII inhibitor. In fact, our diabetic rat hearts exhibited augmented expression Nox-4, the 330

major catalytic component of NADPH oxidase. 331

More important, our results showed that eight weeks of combined swimming training and insulin therapy is more effective in restoring 332

intracellular Ca2+ homeostasis as well as cardiac oxidative stress and mitochondrial dysfunctions. Regarding the Ca2+ homeostasis, both exercise 333

training and insulin treatment either alone or in combination restored the time course as well as the amplitude of the [Ca2+]i transient in left 334

ventricular myocytes of diabetic rats. Although we did not measure Ca2+ handling proteins, exercise training has been shown to improve SR Ca2+ 335

resequestration via increases in SERCA2a and phospholamban expression and/or activity along with augments of Ca2+ efflux via NCX in 336

diabetic rat hearts (24, 31). It is noteworthy that SR Ca2+-induced Ca2+ release is insulin-dependent as insulin regulates the cardiac function by 337

stimulating ICaL. Insulin also interacts with SERCA2 via IRS indicating that IRS proteins bind to the SERCA2 in an insulin-regulated fashion (3). 338

Our echocardiographic data show that diabetic rats exercised and treated with insulin had their systolic function partially restored, despite no 339

recovery of diastolic function. 340

On the subject of cardiac oxidative stress, the combined treatments normalized Nox-4 expression and reduced the content of carbonyl 341

proteins. Among the factors known for NADPH oxidase modulation in diabetic heart [see (44) for review], exogenous insulin replacement can 342

act controlling blood glucose levels (27) and thereby the intracellular Ca2+ homeostasis (32) as described above, drastically down regulating Nox-343

4 expression; exercise training can act increasing the levels of protein kinase C (17), reestablishing Nox-4 expression to the control levels. 344

18

As for the mitochondrial STZ-induced dysfunctions, the combination of insulin with exercise training was able to reduce UCP-2 345

expression, Ca2+ uptake, O2 consumption and MPTP opening susceptibility and increased H2O2 release. Insulin itself is known to improve the 346

activities of complexes I, II and/or IV after four weeks, as shown by others (38). It is probably related to the levels of mRNA for the peroxisome 347

proliferator-activated receptor γ co-activator 1α that up-regulate nuclear genes required for mitochondrial biogenesis (31). Long-term treadmill 348

running (14 weeks) prevented the elevation of proteins involved in MPTP pore formation and apoptotic signaling in hearts of diabetic rats (26). 349

Nevertheless, in our eight-week treatment, endurance exercise training or insulin alone restored the mitochondrial Ca2+ uptake only. 350

Finally, we observed that diabetes reduced the RHR in rats, as shown elsewhere (21). This can be explained in part by the reduction in the 351

expression of beta adrenergic receptors (β1 e β2) in diabetic rats (29). As expected, bradycardia was normalized in sedentary and exercised 352

animals by insulin, as demonstrated previously (37). Insulin exerts positive inotropic and chronotropic effects on the myocardium (25). Along 353

with the impaired growth of the animals induced by STZ, by the end of the experiment diabetic rats exhibit lower HW. In rats with diabetes, in 354

addition to insulin, the secretion of hormones such as the growth hormone, glucagon, pancreatic polypeptide and, consequently, growth factor 355

similar to that of insulin, are altered and affect their growth (14, 30). We observed that insulin therapy increased the HW gain and the HW to BW 356

ratio in both sedentary and exercised diabetic animals. The increased HW/BW in diabetic rats treated with insulin reflects the recovering of 357

growth in the heart of these animals, inasmuch as their final HW did not change. 358

359

360

4. Conclusion 361

The combination of an eight-week swimming training program with daily insulin replacement was more effective than isolated treatments 362

in attenuating oxidative stress, Ca2+ homeostasis disruptions and mitochondrial dysfunctions in the hearts of rats with STZ-induced type 1 363

diabetes. 364

365

19

Acknowledgments 366

This study was funded by the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG – CDS APQ 01171/11) and 367

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP – 2010/17259-9). A. J. Natali is a CNPq fellow. M. F. da Silva was recipient 368

of a doctoral scholarship from FAPEMIG. The confocal experiments were performed in the facilities of the Núcleo de Microscopia e 369

Microanálise at the Federal University of Viçosa (UFV). We acknowledge Ieda M.R. Prado for the technical support. 370

371

Duality of interest 372

The authors declare that there is no duality of interest associated with this manuscript. 373

374

5. References 375

1. Ago T, Kuroda J, Pain J, Fu C, Li H, Sadoshima J. Upregulation of Nox4 by hypertrophic stimuli promotes apoptosis and 376

mitochondrial dysfunction in cardiac myocytes. Circ Res. 2010 Apr 16;106(7):1253-64. 377

2. Aragno M, Mastrocola R, GAlloatti G. Vercellinatto I, Bardini P, Geuna S, Catalano MG, Danni O, Boccuzzi G. Oxidative stress 378

triggers cardiac fibrosis in the heart of diabetic rats. Endocrinology. 2008; 149:380-388 379

3. Aulbach F, Simm A, Maier S, Langenfeld H, Walter U, Kersting U, Kirstein M. Insulin stimulates the L-type Ca2+ current in rat 380

cardiac myocytes. Cardiovasc Res. 1999 Apr;42(1):113-20. 381

4. Barbosa MR, Sampaio IH, Teodoro BG, Sousa TA, Zoppi CC, Queiroz AL, Passos MA, Alberici LC, Teixeira FR, Manfiolli AO, 382

Batista TM, Cappelli AP, Reis RI, Frasson D, Kettelhut IC, Parreiras-e-Silva LT, Costa-Neto CM, Carneiro EM, Curi R, Silveira 383

20

LR. Hydrogen peroxide production regulates the mitochondrial function in insulin resistant muscle cells: effect of catalase 384

overexpression. Biochim Biophys Acta. 2013 Oct;1832(10):1591-604. 385

5. Bers DM. Calcium cycling and signaling in cardiac myocytes. Annu Rev Physiol. 2008;70:23-49. 386

6. Bidasee KR, Dinçer UD, Besch HR Jr. Ryanodine receptor dysfunction in hearts of streptozotocin-induced diabetic rats. Mol 387

Pharmacol. 2001 Dec;60(6):1356-64. 388

7. Boudina S, Sena S, Theobald H, Sheng X, Wright JJ, Hu XX, Aziz S, Johnson JI, Bugger H, Zaha VG, Abel ED. Mitochondrial 389

energetics in the heart in obesity-related diabetes: direct evidence for increased uncoupled respiration and activation of uncoupling 390

proteins. Diabetes. 2007; 56: 2457-66. 391

8. Bugger H, Abel ED. Mitochondria in the diabetic heart. Cardiovasc Res. 2010 Nov 1;88(2):229-40. 392

9. Carneiro-Júnior MA, Prímola-Gomes TN, Quintão-Júnior JF, Drummond LR, Lavorato VN, Drummond FR, Felix LB, 393

Oliveira EM, Cruz JS, Natali AJ, Mill JG. Regional effects of low-intensity endurance training on structural and mechanical properties 394

of rat ventricular myocytes. J Appl Physiol (1985). 2013 Jul 1;115(1):107-15. 395

10. Choi KM, Zhong Y, Hoit BD, Grupp IL, Hahn H, Dilly KW, Guatimosim S, Lederer WJ, Matlib MA. Defective intracellular 396

Ca(2+) signaling contributes to cardiomyopathy in Type 1 diabetic rats. Am J Physiol Heart Circ Physiol. 2002 Oct;283(4):H1398-408. 397

11. Dent CL, Bowman AW, Scott MJ, Allen JS, Lisauskas JB, Janif M, Wickline SA, Kovács SJ. Echocardiographic characterization of 398

fundamental mechanisms of abnormal diastolic filling in diabetic rats with a parameterized diastolic filling formalism. J Am Soc 399

Echocardiogr. 2001 Dec;14(12):1166-72. 400

12. Fang ZY, Prins JB, Marwick TH. Diabetic cardiomyopathy: evidence, mechanisms, and therapeutic implications.. Endocr Rev. 2004 401

Aug;25(4):543-67. 402

13. Gomes RJ, de Mello MA, Caetano FH, Sibuya CY, Anaruma CA, Rogatto GP, Pauli JR, Luciano E. Effects of swimming training 403

on bone mass and the GH/IGF-1 axis in diabetic rats. Growth Horm IGF Res. 2006 Oct-Dec;16(5-6):326-31. 404

21

14. Gomes RJ1, Leme JA, de Moura LP, de Araújo MB, Rogatto GP, de Moura RF, Luciano E, de Mello MA. Growth factors and 405

glucose homeostasis in diabetic rats: effects of exercise training Cell Biochem Funct. 2009 Jun;27(4):199-204. 406

15. Gunter TE, Buntinas L, Sparagna G, Eliseev R, Gunter K. Mitochondrial calcium transport: mechanisms and functions. Cell Calcium. 407

2000 Nov-Dec;28(5-6):285-96. 408

16. Hattori Y, Matsuda N, Kimura J, Ishitani T, Tamada A, Gando S, Kemmotsu O, Kanno M. Diminished function and expression of 409

the cardiac Na+-Ca2+ exchanger in diabetic rats: implication in Ca2+ overload. J Physiol. 2000 Aug 15;527 Pt 1:85-94. 410

17. Hao Z, Pan SS, Shen YJ, Ge J. Exercise preconditioning-induced early and late phase of cardioprotection is associated with protein 411

kinase C epsilon translocation. Circ J. 2014;78:1636-45. 412

18. Herlein JA, Fink BD, O'Malley Y, Sivitz WI. Superoxide and respiratory coupling in mitochondria of insulin-deficient diabetic rats. 413

Endocrinology. 2009 Jan;150(1):46-55. 414

19. Hidaka S, Kakuma T, Yoshimatsu H, Sakino H, Fukuchi S, Sakata T. Streptozotocin treatment upregulates uncoupling protein 3 415

expression in the rat heart. Diabetes. 1999 Feb;48(2):430-5. 416

20. Hissin PJ, Hilf R. A fluorometric method for determination of oxidized and reduced glutathione in tissues. Anal Biochem. 1976 417

Jul;74(1):214-26. 418

21. Howarth FC, Jacobson M, Shafiullah M, Adeghate E. Effects of insulin treatment on heart rhythm, body temperature and physical 419

activity in streptozotocin-induced diabetic rat. Clin Exp Pharmacol Physiol. 2006 Apr;33(4):327-31. 420

22. Kim JD; Yu BP; MCCarter, RJM; Lee SY; Herlihy JT. Exercise and diet modulate cardiac lipid peroxidation and antioxidant 421

defenses. F Radic Biol & Medic. 1996; 20:283-288. [16] 422

23. Lacombe VA, Viatchenko-Karpinski S, Terentyev D, Sridhar A, Emani S, Bonagura JD, Feldman DS, Györke S, Carnes CA. 423

Mechanisms of impaired calcium handling underlying subclinical diastolic dysfunction in diabetes. Am J Physiol Regul Integr Comp 424

Physiol. 2007 Nov;293(5):R1787-97. 425

22

24. Le Douairon Lahaye S, Gratas-Delamarche A, Malardé L, Zguira S, Vincent S, Lemoine Morel S, Carré F, Rannou Bekono F. 426

Combined insulin treatment and intense exercise training improved basal cardiac function and Ca(2+)-cycling proteins expression in type 427

1 diabetic rats. Appl Physiol Nutr Metab. 2012 Feb;37(1):53-62 428

25. Lucchesi BR, Medina M, Kniffen, FJ. The positive inotropic action of insulin in the canine heart. Eur J Pharmacol. 1972 429

Apr;18(1):107-15. 430

26. Lumini-Oliveira J, Magalhaes J, Pereira CV, Moreira AC, Oliveira PJ, Ascensao A. Endurance training reverts heart mitochondrial 431

dysfunction, permeability transition and apoptotic signaling in long-term severe hyperglycemia. Mitochondrion. 2010; 11:54-63. 432

27. Maalouf RM, Eid AA, Gorin YC, Block K, Escobar GP, Bailey S, Abboud HE. Nox4-derived reactive oxygen species mediate 433

cardiomyocyte injury in early type 1 diabetes. Am J Physiol Cell Physiol. 2012 Feb 1;302(3):C597-604. 434

28. Maritim AC, Sanders RA, Watkins JB. Effects of alpha-lipoic acid on biomarkers of oxidative stress in streptozotocin-induced diabetic 435

rats. J Nutr Biochem. 2003 May;14(5):288-94. 436

29. Matsuda N, Hattori Y, Kimura J, Ishitani T, Tamada A, Gando S, Kemmotsu O, Kanno M. Diminished function and expression of 437

the cardiac Na+-Ca2+ exchanger in diabetic rats: implication in Ca2+ overload. J Physiol. 2000 Aug 15;527 Pt 1:85-94. 438

30. Menon RK, Stephan DA, Rao RH, Shen-Orr Z, Downs LS Jr, Roberts CT Jr, LeRoith D, Sperling MA. Tissue-specific regulation 439

of the growth hormone receptor gene in streptozocin-induced diabetes in the rat. J Endocrinol. 1994 Sep;142(3):453-62. 440

31. Mootha VK, Bunkenborg J, Olsen JV, Hjerrild M, Wisniewski JR, Stahl E, Bolouri MS, Ray HN, Sihag S, Kamal M, Patterson N, 441

Lander ES, Mann M. Integrated analysis of protein composition, tissue diversity, and gene regulation in mouse mitochondria. Cell. 2003 442

Nov 26;115(5):629-40. 443

32. Nishio S, Teshima Y, Takahashi N, Thuc LC, Saito S, Fukui A, Kume O, Fukunaga N, Hara M, Nakagawa M, Saikawa T. 444

Activation of CaMKII as a key regulator of reactive oxygen species production in diabetic rat heart. J Mol Cell Cardiol. 2012 445

May;52(5):1103-11. 446

23

33. Pedersen PL, Greenawalt JW, Reynafarje B, Hullihen J, Decker GL, Soper JW Bustamante E. Preparation and characterization of 447

mitochondria and submitochondrial particles of rat liver and liver derived tissues. Methods Cell Biol. 1978;20:411-81. 448

34. Pierce GN, Ganguly PK, Dzurba A, Dhalla NS. Modification of the function of cardiac subcellular organelles by insulin. Adv 449

Myocardiol. 1985; 6:113–125. [17] 450

35. Prímola-Gomes TN, Campos LA, Lauton-Santos S, Balthazar CH, Guatimosim S, Capettini LS, Lemos VS, Coimbra CC, Soares 451

DD, Carneiro-Júnior MA, Quintão-Júnior JF, Souza MO, Cruz JS, Natali AJ. Exercise capacity is related to calcium transients in 452

ventricular cardiomyocytes, J Appl Physiol (1985). 2009 Aug;107(2):593-8. 453

36. Rajdev S, Reynolds IJ. Calcium green-5N, a novel fluorescent probe for monitoring high intracellular free Ca2+ concentrations 454

associated with glutamate excitotoxicity in cultured rat brain neurons. Neurosci. Lett. 1993; 162:149–152. [27] 455

37. Ramanadham S, Decker P, Tenner TE. Effect of insulin replacement on streptozotocin-induced effects in the rat heart. Life Sci. 1983; 456

33:289-6. 457

38. Remor AP, de Matos FJ, Ghisoni K, da Silva TL, Eidt G, Búrigo M, de Bem AF, Silveira PC, de León A, Sanchez MC, Hohl A, 458

Glaser V, Gonçalves CA, Quincozes-Santos A, Borba Rosa R, Latini A. Differential effects of insulin on peripheral diabetes-related 459

changes in mitochondrial bioenergetics: involvement of advanced glycosylated end products. Biochim Biophys Acta. 2011 460

Nov;1812(11):1460-71. 461

39. Reznick AZ, Packer L. Oxidative damage to proteins: spectrophotometric method for the carbonyl assay. Methods Enzymol. 1994; 233: 462

357–63. 463

40. Saks VA, Khuchua ZA, Kuznetsov AV, Veksler VI, Sharov VG. Heart mitochondria in physiological salt solution: not ionic strength 464

but salt composition is important for association of creatine kinase with the inner membrane surface. Biochem Biophys Res Commun 465

1986; 139:1262–71. 466

24

41. Shao CH, Wehrens XH, Wyatt TA, Parbhu S, Rozanski GJ, Patel KP, Bidasee KR. Exercise training during diabetes attenuates 467

cardiac ryanodine receptor dysregulation. J Appl Physiol. 2009 Apr;106(4):1280-92. 468

42. Skulachev VP. Uncoupling: new approaches to an old problem of bioenergetics. Biochim. Biophys. Acta 1998; 1363: 100–24. 469

43. Teshima Y, Takahashi N, Saikawa T, Hara M, Yasunaga S, Hidaka S, Sakata T. Diminished expression of sarcoplasmic reticulum 470

Ca2+-ATPase and ryanodine sensitive Ca2+ Channel mRNA in streptozotocin-induced diabetic rat heart J Mol Cell Cardiol. 2000 471

Apr;32(4):655-64. 472

44. Teshima Y, Takahashi N, Nishio S, Saito S, Kondo H, Fukui A, Aoki K, Yufu K, Nakagawa M, Saikawa T. Production of Reactive 473

Oxygen Species in the Diabetic Heart – Roles of Mitochondria and NADPH Oxidase. Circ J. 2014;78(2):300-6. 474

45. Trenker M, Malli R, Fertschai I, Levak-Frank S, Graier WF. Uncoupling proteins 2 and 3 are fundamental for mitochondrial Ca2+ 475

uniport. Nat Cell Biol. 2007; 9:445-52. [32] 476

46. Winnick JJ, Sherman WM, Habash DL, Stout MB, Failla ML, Belury MA, Schuster DP. Short-term aerobic exercise training in 477

obese humans with type 2 diabetes mellitus improves whole-body insulin sensitivity through gains in peripheral, not hepatic insulin 478

sensitivity. J Clin Endocrinol Metab. 2008 Mar;93(3):771-8. 479

47. Yu T, Sheu SS, Robotham JL, Yoon Y. Mitochondrial fission mediates high glucoseinduced cell death through elevated production of 480

reactive oxygen species. Cardiovasc Res. 2008; 79:341–51. 481

48. Zhao XY, Hu SJ, Li J, Mou Y, Chen BP, Xia Q. Decreased cardiac sarcoplasmic reticulum Ca2+-ATPase activity contributes to cardiac 482

dysfunction in streptozotocininduced diabetic rats. J Physiol Biochem. 2006; 62: 1–8. 483

49. Zhou M, Diwu Z, Panchuk-Voloshina N, Haugland R P. A stable nonfluorescent derivative of resorufin for the fluorometric 484

determination of trace hydrogen peroxide: applications in detecting the activity of phagocyte NADPH oxidase and other oxidases. Anal. 485

Biochem. 2007; 15: 162–168. 486

487

25

488

Table 1. Forward and Reverse sequences of primers used in the real time RT-PCR assays. 489

Gene Forward Reverse

Nox-4 5′TTCTGGACCTTTGTGCCTATAC3′ 5′CCATGACATCTGAGGGATGATT3′

UCP 5′ATGTGGTAAAGGTCCGCTTC3′ 5′CATTTCGGGCAACATTGGG3′

β-actin 5′CACTTTCTACAATGAGCTGCG3′ 5′CTGGATGGCTACGTACATGG3′

UCP-2, uncoupling protein-2; Nox-4, NADPH oxidase-4. 490

491

492

26

493

Table 2. General characteristics of animals in the experimental groups. 494

CS DS DSI CE DE DEI

Initial BG (mg/dL) 66.61±3.81 68.61±3.01 64.21±3.21 70.20±1.60 67.21±1.71 62.60±2.61

Final BG (mg/dL) 90.21±2.81 529.21±22.70a 338.20±39.00ad 87.21±3.80 492.20±34.50ab 316.81±67.01abcd

Initial BW (g) 83.40±1.61 85.20±3.40 85.80±2.71 84.81±2.91 85.80±2.71 83.60±1.80

Final BW (g) 351.20±23.31 198.80±20.40a 200.81±16.11a 286.60±25.10d 190.80±16.60ab 232.60±11.80abcd

HW (g) 1.86±0.08 1.00±0.10a 1.43±0.15ad 1.95±0.12d 1.01±0.07ab 1.62±0.07cd

HW/BW (mg/g) 5.43±0.47 5.07±0.20 7.46±1.28d 7.07±0.83d 5.48±0.68 7.02±0.39cd

RHR (bpm) 345.90±9.01 267.00±11.3a 345.3±17.8d 337.30±7.61 277.20±15.8b 349.21±9.81cd

Data expressed as mean ± SEM of 10 animals in each group. BW, body weight. BG, blood glucose. HW, heart weight. RHR, resting heart rate. 495

CS, control sedentary. DS, diabetic sedentary. DSI, diabetic sedentary with insulin. CE, control exercised. DE, diabetic exercised. DEI, diabetic 496

exercised with insulin. a different from CS; b different from CE; c different from DE; d different from DS. 497

498

27

Table 3. Left ventricular systolic and diastolic functions measured by resting echocardiography at the end of the experimental period. 499

CS

(n=7)

DS

(n=7)

DSI

(n=7)

CE

(n=7)

DE

(n=7)

DEI

(n=7)

Ejection fraction (%) 77.33±2.86 64.80±3.14a 63.33±2.86a 83.75±2.48 58.66±2.86ab 71.41±2.86bce

Fractional shortening (%) 41.00 ± 1.54 30.80 ± 1.69a 30.16 ± 1.54a 47.43 ± 1.33a 29.41 ± 1.54ab 35.91 ± 1.54bce

Peak E (m/s) 0.497±0.041 0.621±0.045 0.615±0.038 0.504±0.038 0.539±0.045 0.563±0.041

Peak A (m/s) 0.316±0.031 0.223±0.031 0.208±0.032 0.285±0.031 0.216±0.033 0.208±0.030

E/A ratio 1.66 ± 0.16 3.10 ± 0.18ab 3.02 ± 0.15ab 1.77 ± 0.15 2.54 ± 0.18 2.95 ± 0.16

Data expressed as mean ± SEM of n (number of animals) in each group. Peak E, early filling wave. Peak A, late filling wave. CS, control 500

sedentary. DS, diabetic sedentary. DSI, diabetic sedentary with insulin. CE, control exercised. DE, diabetic exercised. DEI, diabetic exercised 501

with insulin. a different from CS; b different from CE; c different from DE; e different from DSI. 502

503

504 505 506

28

Table 4. Phosphorylating (State III) and resting (State IV) respiration rates, and respiratory control ratio (RCR, State III/State IV) in heart 507

mitochondria. 508

CS DS DSI CE DE DEI

State III 26.50±2.20 23.50±3.00 23.80±2.40 27.90±1.80 20.10±0.13 21.00±1.5

State IV 2.32±0.13 3.03±0.2a 3.13±0.22a 3.34±0.48 3.09±0.09a 2.59±0.17

RCR 10.66±0.65 7.71±0.81a 7.61±0.55a 8.82±1.01 6.51±0.16a 8.13±0.50

Data expressed as mean ± SEM of 6 animals in each group. Mitochondria (0.5 mg protein/ml) were added in the standard medium supplemented 509

with 0.1 mM EGTA, under the conditions described in Materials and methods. Respiration rates given in nmols oxygen/mg protein/min. State III 510

was induced by the addition of 200 nmol ADP. State IV was determined by the addition of 1 µg/ml oligomycin, a classical inhibitor of FOF1 ATP 511

synthase. CS, control sedentary. DS, diabetic sedentary. DSI, diabetic sedentary with insulin. CE, control exercised. DE, diabetic exercised. DEI, 512

diabetic exercised with insulin. a Different from CS. 513

514

515

516 517

518

29

6. Figure legends 519

520

Figure 1. Intracellular global Ca2+ transient in isolated left ventricular myocytes. (A) Amplitude of transient. (B) Time to peak. (C) Time from 521

peak transient to half resting value. F/F0, fluorescence ratio [fluorescence intensity (F) normalized to the minimal intensity measured between 1-522

Hz contractions at diastolic phase (F0)]. CS, control sedentary. DS, diabetic sedentary. DSI, diabetic sedentary with insulin. CE, control 523

exercised. DE, diabetic exercised. DEI, diabetic exercised with insulin. a different from CS; b different from CE; c different from DE; d different 524

from DS; e different from DSI. n = 87 to103 cells per group. N = 87 to103 cells per group. 525

526

Figure 2. Redox state parameters in rat hearts. (A) mRNA relative expression of Nox-4. (B) SOD content. (C) GSH/GSSG. (D) Protein carbonyl 527

content. CS, control sedentary. DS, diabetic sedentary. DSI, diabetic sedentary with insulin. CE, control exercised. DE, diabetic exercised. DEI, 528

diabetic exercised with insulin. a different from CS; b different from CE; c different from DE; d different from DS; e different from DSI. n = 6 529

animals per group. 530

531

Figure 3. mRNA relative expression of UCP-2 in left ventricles. CS, control sedentary. DS, diabetic sedentary. DSI, diabetic sedentary with 532

insulin. CE, control exercised. DE, diabetic exercised. DEI, diabetic exercised with insulin. a different from CS; b different from CE; c different 533

from DE; d different from DS; e different from DSI. n = 6 animals per group. 534

535

Figure 4. H2O2 release in heart isolated mitochondria. (A) Typical traces and (B) Mean ± SEM data of H2O2 release rates are presented. Traces 536

are representative of five experiments with different mitochondrial preparations of each group. CS, control sedentary. DS, diabetic sedentary. 537

DSI, diabetic sedentary with insulin. CE, control exercised. DE, diabetic exercised. DEI, diabetic exercised with insulin. a different from CS; b 538

different from CE; c different from DE; d different from DS; e different from DSI. n = 6 animals per group. 539

30

540

Figure 5. Calcium dynamics in heart isolated mitochondria. (A) Mean ± SEM data of Ca2+ uptake measured in the presence of 1 µM CSA (to 541

prevent MPTP opening). (B) Representative raw experiments of Ca2+ retention capacity induced by MPTP opening in the absence of 1 µM CSA. 542

(C) Representative raw experiments of Ca2+ retention capacity induced by MPTP opening in the presence of 1 µM CSA. Where indicated were 543

added 20 nmol Ca2+ or 1 μM CCCP. CCCP was used to collapse membrane potential and release the full accumulated Ca2+. Traces are 544

representative of 5 experiments with different mitochondrial preparations of each group. CS, control sedentary. DS, diabetic sedentary. DSI, 545

diabetic sedentary with insulin. CE, control exercised. DE, diabetic exercised. DEI, diabetic exercised with insulin. a different from CS; d 546

different from DS; e different from DSI. 547

548

549

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5