Low Concentrations of Hydrogen Peroxide Improve Post-ischaemic Metabolic and Functional Recovery in Isolated Perfused Rat Hearts

J Mol Cell Cardiol 29, 2779–2787 (1997)

Low Concentrations of Hydrogen Peroxide Improve Post-ischaemic Metabolic and Functional Recovery in Isolated Perfused Rat Hearts Anna-Catharina Hegstad1, Olaf H. Antonsen2 and Kirsti Ytrehus2 1

Department of Morphology and 2Department of Medical Physiology, Institute of Medical Biology, University of Tromsø, Norway (Received 15 April 1996, accepted in revised form 7 July 1997) A.-C. H, O. H. A  K. Y. Low Concentrations of Hydrogen Peroxide Improve Postischaemic Metabolic and Functional Recovery in Isolated Perfused Rat Hearts. Journal of Molecular and Cellular Cardiology (1997) 29, 2779–2787. The aim of the present study was to test the hypothesis that low concentrations of hydrogen peroxide (H2O2) have a beneficial effect on post-ischaemic myocardial recovery. Functional and metabolic measurements were performed in isolated buffer-perfused rat hearts exposed to 30 min perfusion with 0 (control group A), 25, 50, 100 or 200 l H2O2 or 30 min global ischaemia followed by 30 min reperfusion with 0 (control group B), 25, 50 or 100 l H2O2. Catalase (200 U/ml) was added as scavenger during reperfusion with 25 l H2O2. Non-ischaemic perfusion: All concentrations of H2O2 induced an immediate vasodilatation, which was maintained in the 50 l group, but it was followed by vasoconstriction in the 100 and 200 l group. Left ventricular developed pressure (LVDP) was significantly increased at the end of perfusion in the 50 l group compared to the control group. Exposure to 100 and 200 l H2O2 significantly decreased LVDP and increased end-diastolic pressure. ATP was reduced in the 100 l group. Post-ischaemic perfusion: Exposure to 25 l H2O2 caused improved coronary flow during the first 20 min of reperfusion compared to the control group (accumulated coronary flow; 235.5 ± 10.8 v 172.7 ± 8.6 ml). LVDP was significantly higher in the 25 l group compared to the control (59.8 ± 10.2 v 22.1 ± 7.3 mmHg), and end-diastolic pressure was significantly lower (32.1 ± 19.6 v 78.8 ± 2.2 mmHg) at the end of reperfusion. Improved recovery was not observed in the group exposed to 25 l H2O2 plus catalase. Treatment with 25 l H2O2 caused significantly improved recovery of tissue ATP and creatine phosphate. In conclusion, the present study showed that exposure to 25 l H2O2 improved post-ischaemic recovery in hearts subjected to global ischaemia.  1997 Academic Press Limited K W: ATP; Coronary flow; Creatine phosphate; Free radicals; Hydrogen peroxide; Ischaemia; Isolated rat hearts; Left ventricular developed pressure; Nucleotides; Oxygen radicals; Reperfusion.

Introduction Oxygen radicals are reactive compounds capable of inducing myocardial damage (Ytrehus et al., 1986, 1987; Miki et al., 1988; Loesser et al., 1991; Ambrosio et al., 1992). The toxic effect of oxygen radicals is dose-dependent. As with other oxygen radical scavengers, superoxide dismutase (SOD) has been shown to exert a beneficial effect on impaired

ventricular functon, myocardial stunning and arrhythmias induced by ischaemia-reperfusion (Burton, 1985; Ambrosio et al., 1987; Bernier et al., 1989; Bolli et al., 1989; Nejima et al., 1989). Both the source of, and the actual level of, oxygen radicals during reperfusion of the myocardium remains a controversy. Some experimental studies have shown that SOD reduces infarct size (Werns et al., 1985; Ambrosio et al., 1986), while others

Please address all correspondence to: Anna-Catharina Hegstad, Department of Morphology, Institute of Medical Biology, University of Tromsø, 9037 Tromsø, Norway.

0022–2828/97/102779+09 $25.00/0

mc970513

 1997 Academic Press Limited

2780

A.-C. Hegstad et al.

have failed to demonstrate a sustained protection (Uraizee et al., 1987; Na¨slund et al., 1992). SOD is assumed to be cardioprotective because the enzyme removes the toxic superoxide anion radical by catalysing the dismutation of superoxide anions to H2O2. Few studies have examined in detail the role of H2O2 during reperfusion. The aim of the present study was to test the hypothesis that low concentrations of H2O2 have a beneficial effect on post-ischaemic myocardial recovery. The interest in H2O2 is based on results showing that H2O2 cause vasodilatation (Kontos, 1985; Burke and Wolin, 1987; Rubanyi and Vanhoutte, 1986). Furthermore, it has been reported that H2O2 caused reduction in infarct size in isolated rabbit hearts (Ytrehus et al., 1995). Concentrations of H2O2 above 100–200 l have been shown conclusively to have a deleterious effect on the structure and function of rat myocardium (Miki et al., 1988; Janero et al., 1991; Onodera et al., 1992; Evans et al., 1995). In the present study, isolated Langendorffperfused rat hearts were exposed to 25–200 l H2O2. Contractile function, coronary flow and metabolic alterations were studied both during nonischaemic conditions and during reperfusion after 30 min global ischaemia.

ventricular end-diastolic pressure and peak-systolic pressure and heart rate were recorded (Gould Recorder, Cleveland, OH, USA). Left venticular developed pressure (LVDP) was calculated as the difference between left ventricular peak-systolic and end-diastolic pressure. Coronary flow was measured by continuously timed collection of effluent at regular intervals, and expressed in ml/min. Accumulated coronary flow was calculated during defined parts of reperfusion.

Experimental protocol All hearts (n=105) were stabilised for 20–25 min prior to intervention. In the first protocol (A), the hearts were exposed to 30 min standard perfusion with 0 l H2O2 (control group A, n=8), 25 l H2O2 (n=10), 50 l H2O2 (n=8), 100 l H2O2 (n=9), 200 l H2O2 (n=5) or catalase (200 U/ ml) (n=9). In the second protocol (B), the hearts were exposed to 30 min global ischaemia followed by 30 min reperfusion. During reperfusion the hearts were exposed to 0 l H2O2 (control group B, n=10), 25 l H2O2 (n=8), 50 l H2O2 (n= 10), 100 l H2O2 (n=10), catalase (200 U/ml) (n=9) or 25 l H2O2 plus catalase (200 U/ml) (n=9).

Materials and Methods Heart perfusion

Metabolic measurements

Male Wistar rats weighing 280–340 g were anaesthetized with diethyl ether, and anticoagulated by injecting heparin (200 IU) into the femoral vein. The hearts were rapidly excised and perfused in a non-recirculating Langendorff perfusion system maintained at 37°C. The perfusion pressure was 100 cm H2O, and the perfuson medium was Krebs– Henseleit bicarbonate buffer containing 2.4 m calcium and 11.1 m glucose. The perfusate was continuously gassed with 95% O2 and 5% CO2. The perfusion system consisted of two reservoirs, allowing H2O2 as additive in different concentrations to one reservoir. H2O2 in a 30% solution containing no antioxidants was purchased from the Norwegian Medical Depot. The stability of the concentration of this solution was checked by spectrophotometry (240 nm) and found to vary between 30–31%. The diluted solution used for perfusion was made fresh daily. For isovolumetric functional measurements a water-filled latex balloon was placed in the left ventricle through a left atriotomy. The balloon was connected to a pressure transducer, and left

The hearts from the two control groups (A and B) and the groups exposed to H2O2 were freeze-clamped at the temperature of liquid nitrogen at the end of the experiment. Tissue content of adenosine triphosphate (ATP), adenosine diphosphate (ADP), adenosine monophosphate (AMP) and creatine phosphate (CP) was measured by high-performance liquid chromatography (HPLC) (Sellevold et al., 1986).

Statistics The results are presented as mean ± ... One-way analysis of variance was performed, and Student’s ttest with Bonferroni’s correction was applied to identify significant differences between the groups (P<0.05). Statistical comparisons between the different groups were performed before and at the end of intervention, and at selected timepoints during reperfusion.

2781

H2O2 and Myocardial Reperfusion

Table 1 Left ventricular developed pressure (LVDP) and coronary flow (CF) during the stabilisation period in each group, prior to exposure either to 30 min of non-ischaemic perfusion (NP) or 30 min of ischaemia followed by 30 min of post-ischaemic perfusion (PP). Group 0 l H2O2 25 l H2O2 50 l H2O2 100 l H2O2 200 l H2O2 Catalase 25 l H2O2 + catalase

LVDPNP (mmHg)

CFNP (ml/min)

LVDPPP (mmHg)

CFpp (ml/min)

105.0 ± 7.1 109.3 ± 4.2 118.2 ± 3.4 100.4 ± 3.2 124.3 ± 10.9 103.0 ± 3.1 —

13.6 ± 0.6 11.8 ± 0.7 13.1 ± 1.0 13.8 ± 0.8 13.3 ± 0.8 12.7 ± 0.4 —

113.4 ± 8.1 112.8 ± 3.5 107.9 ± 6.4 120.8 ± 3.1 — 105.3 ± 3.7 108.3 ± 4.8

13.5 ± 0.6 10.8 ± 0.6 12.8 ± 0.6 12.9 ± 0.7 — 12.0 ± 0.4 11.1 ± 0.4

Values represent mean ± ... There were no significant differences between the groups.

Results Effects of H2O2 during 30 min of non-ischaemic perfusion There were no significant differences in functional performance between the groups during the stabilisation period (Table 1). Exposure to H2O2 induced an immediate increase in coronary flow for all examined concentrations [Fig. 1(a)]. Treatment with 25 l H2O2 caused a transient increase in coronary flow. The vasodilatation was followed by a significant decrease in coronary flow in the 100 and 200 l group compared to the control group. Exposure to 25 and 50 l H2O2 had no effect on end-diastolic pressure [Fig. 1(b)]. Perfusion with 100 and 200 l H2O2 caused a gradual and prominent increase in end-diastolic pressure, starting after 12 and 5 min, respectively. Treatment with 25 l H2O2 induced an initial increase in LVDP which was gradually reversed during the perfusion period [Fig. 1(c)]. LVDP was significantly increased at the end of perfusion in the group exposed to 50 l H2O2, while perfusion with 100 and 200 l H2O2 caused a significant decrease in LVDP, compared to the control group. Exposure to 200 l H2O2 induced an initial increase in heart rate, which gradually was reduced to 12.8 ± 7.9% of stabilisation value at the end of the experiments [Fig. 1(d)]. Addition of catalase to the perfusion buffer had no significant effect on coronary flow, end-diastolic pressure, LVDP or heart rate compared to the control group when evaluated at the end of perfusion (12.7 ± 0.4 v control 13.8 ± 0.7 ml/min, 2.6 ± 0.4 v control 3.3 ± 0.5 mmHg, 95.6±3.7 v 91.1±6.7 mmHg, 305± 9 v control 297 ± 12 beats/min). Creatine phosphate (CP) was significantly higher after 30 min of perfusion in the group exposed to 25 l H2O2 compared to the control group (Table

2). In the 100 l group ATP was significantly reduced at the end of perfusion compared to the control group.

Effects of H2O2 during 30 min of post-ischaemic perfusion There were no significant differences in functional performance between the groups during the stabilisation period (Table 1). In all groups, the onset of ischaemia induced an initial increase in LVDP (first minute) which was followed by a decrease in LVDP and, eventually, left ventricular standstill. After 5–10 min of global ischaemia, the development of a left ventricular contracture was observed. The contracture was partially reversed at the end of the ischaemic period. Treatment with 25 l H2O2 during post-ischaemic perfusion caused an immediate improvement of coronary flow compared to the control group [Fig. 2(a)]. Coronary flow was significantly higher during the first 20 min of reperfusion in the group exposed to 25 l H2O2 compared to the control group (accumulated coronary flow; 235.5 ± 10.8 v 172.7 ± 8.6 ml). Addition of catalase to the perfusion buffer during post-ischaemic perfusion with 25 l H2O2 obviated the improvement in accumulative coronary flow during the first 20 min (accumulated coronary flow; 136.9 ± 6.5 ml). Coronary flow was significantly reduced in the group exposed to 100 l H2O2 [Fig. 2(a)]. There was no significant difference in coronary flow between the control group and the group exposed to catalase (8.7 ± 0.8 v 7.7 ± 0.7 ml/min), or between the control group and the group exposed to 25 l H2O2 and catalase (6.9 ± 0.5 ml/min) at the end of reperfusion.

A.-C. Hegstad et al.

(a)

(b) End-diastolic pressure (mmHg)

2782

Coronary flow (%)

175 150

*

125 100 75

*

50

*

25 0

0

5

10 15 Time (min)

20

25

80 * 60 * 40 20 0

30

5

10 15 Time (min)

20

25

30

H2O2 (d)

(c) 140 120

Heart rate (beats/min)

Left ventricular developed pressure (%)

H2O2

0

*

100 80 60 40 *

20 0

0

5

10 15 Time (min)

20

25

300 250 200 150 100 50

* 30

0

* 0

5

10 15 Time (min)

20

25

30

H2O2

H2O2

Figure 1 (a) Effect of H2O2 on coronary flow during non-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group A, open circles), 25 m (closed circles), 50 m (closed triangles), 100 m (closed diamonds) or 200 m (closed squares) H2O2. Coronary flow is presented as percentage of the respective flow in the stabilisation period for each heart. ∗, P<0.05 v controls. (b) Effect of H2O2 on end-diastolic pressure (mmHg) during non-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group A, open circles), 25 m (closed circles), 50 m (closed triangles), 100 m (closed diamonds) or 200 m (closed squares) H2O2. ∗, P<0.05 v controls. (c) Effect of H2O2 on left ventricular developed pressure (LVDP) during non-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group A, open circles), 25 m (closed circles), 50 m (closed triangles), 100 m (closed diamonds) or 200 m (closed squares) H2O2. LVDP is presented as percentage of the respective LVDP in the stabilisation period for each heart. ∗, P<0.05 v controls. (d) Effect of H2O2 on heart rate (beats/min) during non-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group A, open circles), 25 m (closed circles), 50 m (closed triangles), 100 m (closed diamonds) or 200 m (closed squares) H2O2 ∗, P<0.05 v controls.

Table 2 Tissue content of high energy phosphates (lmol/g dry weight) after 30 min non-ischaemic perfusion with 0 (control group A), 25, 50 or 100 l H2O2 as additive in isolated buffer-perfused rat hearts. Group 0 25 50 100

l l l l

H2O2 H2O2 H2O2 H2O2

CP

ATP

ADP

AMP

24.95 ± 0.99 54.56 ± 3.81∗ 44.13 ± 6.15 18.01 ± 2.46

20.39 ± 0.37 22.96 ± 0.79 21.78 ± 2.02 7.11 ± 0.79∗

7.14 ± 0.08 5.41 ± 0.21∗ 7.36 ± 1.28 4.53 ± 0.30∗

1.19 ± 0.11 1.16 ± 0.19 1.65 ± 0.39 4.07 ± 0.43∗

The values represent mean ± ... where ∗ denotes values significant different (P<0.05) from the control (0 l H2O2). CP, creatine phosphate; ATP, adenosine triphosphate; ADP, adenosine diphosphate; AMP, adenosine monophosphate.

2783

H2O2 and Myocardial Reperfusion

50 25 0

*

0

5

10 15 Time (min)

20

25

30

100 80 60 40 20 0

(d)

100

350

40 20 0

*

Heart rate (beats/min)

(c)

60

300 250 200 150 100 50 0

0

*

*

0

5

10 15 Time (min)

20

25

30

5

10 15 Time (min)

20

25

30

H2O2

30 min global ischaemia

Left ventricular developed pressure (%)

H2O2

80

30 min global ischaemia

75

120

5

10 15 Time (min)

20

25

30

H2O2

30 min global ischaemia

100

End-diastolic pressure (mmHg)

Coronary flow (%)

125

(b) 30 min global ischaemia

(a)

0 H2O2

Figure 2 (a) Effect of H2O2 on coronary flow during post-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group B, open circles), 25 m (closed circles), 50 m (closed triangles) or 100 m (closed diamonds) H2O2. Coronary flow is presented as percentage of the respective flow in the stabilisation period for each heart. ∗, P<0.05 v controls. (b) Effect of H2O2 on end-diastolic pressure (mmHg) during post-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group B, open circles), 25 m (closed circles), 50 m (closed triangles) or 100 m (closed diamonds) H2O2. ∗, P<0.05 v controls. (c) Effects of H2O2 on left ventricular developed pressure (LVDP) during post-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group B, open circles), 25 m (closed circles), 50 m (closed triangles) or 100 m (closed diamonds) H2O2. LVDP is presented as percentage of the respective LVDP in the stabilisation period for each heart. ∗, P<0.05 v controls. (d) Effect on H2O2 on heart rate (beats/min) during post-ischaemic perfusion in isolated buffer-perfused rat hearts. The hearts were exposed to 0 m (control group B, open circles), 25 m (closed circles), 50 m (closed triangles) or 100 m (closed diamonds) H2O2. ∗, P<0.05 v controls.

In all groups, post-ischaemic perfusion was followed by a further increase of the ischaemic-induced contracture. End-diastolic pressure increased by treatment with 100 l H2O2 compared with the control group (end values; 101.7 ± 3.7 v control 78.8 ± 2.2 mmHg) [Fig. 2(b)]. Exposure to 25 l H2O2, however, significantly decreased end-diastolic pressure (end value; 32.1 ± 19.6 mmHg). Reperfusion with 25 l H2O2 had furthermore a positive effect on the recovery of contractile function. LVDP was higher in the 25 l group compared to the control group throughout the reperfusion period (end values; 59.8 ± 10.2 v control 22.1 ± 7.3

mmHg) [Fig. 2(c)]. The improved mechanical recovery observed during reperfusion with 25 l H2O2 was obviated when catalase was added to the perfusion buffer containing 25 l H2O2 (end values; end-diastolic pressure 71.2 ± 4.5 mmHg, LVDP 25.6 ± 5.9 mmHg). Reperfusion with catalase alone had no significant effect on mechanical performance compared to the control group at the end of reperfusion (end-diastolic pressure 65.9 ± 4.5 mmHg, LVDP 18.0 ± 6.1 mmHg). Reperfusion with 100 l H2O2 reduced LVDP (end value; 9.7 ± 3.2 mmHg). There were no significant differences in heart rate between the groups at the end of the

2784

A.-C. Hegstad et al.

Table 3 Tissue content of high energy phosphates (lmol/g dry weight) after 30 min global ischaemia followed by 30 min post-ischaemic perfusion with 0 (control group B), 25, 50 or 100 l H2O2 as additive in isolated buffer-perfused rat hearts. Group 0 25 50 100

l l l l

H2O2 H2O2 H2O2 H2O2

CP

ATP

ADP

AMP

16.45 ± 2.23 39.68 ± 3.65∗ 13.98 ± 1.65 9.25 ± 1.81

4.33 ± 0.46 10.14 ± 0.83∗ 3.58 ± 0.43 2.10 ± 0.47∗

3.65 ± 0.09 3.75 ± 0.11 3.72 ± 0.10 3.44 ± 0.47

3.95 ± 0.33 2.05 ± 0.24∗ 4.62 ± 0.18 4.48 ± 0.41

The values represent mean ± ... where ∗ denotes values significant different (P<0.05) from the control (0 l H2O2). CP, creatine phosphate; ATP, adenosine triphosphate; ADP, adenosine diphosphate; AMP, adenosine monophosphate.

reperfusion period [Fig. 2(d)] (control 177 ± 44, catalase 184 ± 36, 25 l H2O2 and catalase 207 ± 33 beats/min). Treatment with 25 l H2O2 during reperfusion caused increased tissue content of adenosine triphosphate (10.14 ± 0.83 v control 4.33 ± 0.46 lmol/g dry weight) and creatine phosphate (39.68 ± 3.65 v control 16.45 ± 2.23 lmol/g dry weight) at the end of reperfusion (Table 3).

Discussion The present study shows that treatment with 25 l H2O2 during reperfusion after 30 min ischaemia has a beneficial effect on functional and metabolic recovery in isolated rat hearts. Coronary flow and left ventricular developed pressure and relaxation were improved during reperfusion in the group exposed to 25 l H2O2 compared to the untreated control group. The improved post-ischaemic functional recovery was obviated when catalase was added to the H2O2-solution, thus ascribing the beneficial effect observed to be a direct effect of H2O2. Addition of 25 l H2O2 further caused increased levels of high energy phosphates (ATP and CP) at the end of reperfusion compared to the control group. These results are in contrast with earlier studies which support the view that oxygen radicals and H2O2 induce reperfusion injury (Miki et al., 1988; Janero et al., 1991; Onodera et al., 1992; Evans et al., 1995). This discrepancy can partly be explained by the dose-dependent effect of H2O2 observed during non-ischaemic perfusion in the present study. Higher concentrations of H2O2 (100 and 200 l) exerted a deleterious effect on mechanical performance, coronary flow and levels of ATP, while lower concentrations (25 and 50 l), however, caused increased contractility, vasodilatation and preserved levels of high energy phosphates. The mechanism behind the protective effect of

low dose H2O2 has not been explored in the present study. Some indications could be obtained, however, from the results of perfusion with different concentrations of H2O2 during non-ischaemic perfusion. An initial response to H2O2 intervention was increased coronary flow due to decrease in coronary resistance. Although it is not clear if vascular dysfunction impairs post-ischaemic recovery, the improved recovery during reperfusion with low concentrations of H2O2 in the present study could be a function of improved circulation. Vasodilatory effect of H2O2 has been reported in several organs (Kontos, 1985; Rubanyi and Vanhoutte, 1986; Burke and Wolin, 1987). H2O2 can modulate the tone of vascular smoooth muscle by acting directly on the smooth muscle cells, or indirectly by changing the production or the biological activity of endogenous vasoactive mediators like endotheliumderived relaxing factor (EDRF) (NO) or arachidonic acid metabolites. Fraile et al. (1994) reported that exogenous H2O2 elicited endothelium independent concentration-dependent relaxation in precontracted isolated cat cerebral artery segments. Mu¨gge et al. (1991) showed that endothelial SOD activity, but not catalase or glutathione, is necessary for the biological activity of EDRF (NO) in the rabbit aorta, thus proposing that H2O2 might have a reinforcing effect on EDRF (NO). H2O2 have furthermore been shown to have an effect on the vasoactive arachidonic derivate prostacyclin which promotes vasodilatation. High concentrations of H2O2 inhibit prostacyclin production, while low concentrations of H2O2 stimulate it (Panganamala et al., 1986; Buckley et al., 1991). These results are compatible with the present findings revealing a vasodilatation during non-ischaemic perfusion with low micromolar concentrations of H2O2 while higher concentrations induced a vasoconstriction. Burke et al. (1987) reported the H2O2 produces concentration-dependent relaxation of intrapulmonary atrial rings by a mechanism independent of both the endothelium and

H2O2 and Myocardial Reperfusion

prostaglandin mediators. They found that H2O2induced relaxation was associated with direct activation of guanylate cyclase and formation of the intracellular mediator of relaxation; c-GMP. The improved post-ischaemic myocardial function in the present study could thus be due to a decrease in the “no-relow” phenomenon and/or rapid washout of metabolites with negative inotropic effect as a result of H2O2-induced vasodilatation. The increase in coronary flow during perfusion with low concentrations of H2O2 was accompanied by an increase of LVDP. The positive inotropic effect observed in the present study could be the result of H2O2-induced rise in intracellular calcium. Kaminishi et al. (1989) demonstrated that addition of H2O2 (2.5 m) to adult cardiomyocytes labelled with exogenous [45 Ca2+] raised the content of rapidly exchangeable intracellular Ca2+ two-fold. H2O2-induced enhancement of intracellular calcium is further reflected in a study by Beresewicz and Horackova (1991), reporting that treatment with 30 l H2O2 caused prolongation of the action potential duration and accelerated cell contraction of ventricular myocytes. Ward and Moffat (1995) have recently confirmed that exposure to low micromolar concentrations of H2O2 (25 or 75 l) cause increased levels of intracellular calcium and cell shortening in isolated ventricular myocytes. Enhanced levels of intracellular calcium could be caused by augmentation of sarcolemmal Ca2+-influx via activated Na+–Ca2+ exchange or increased Ca2+-leak channel activity (Bhatnagar et al., 1990; Wang et al., 1995). H2O2 may furthermore cause accumulation of intracellular calcium levels by depressing the sarcolemmal Ca2+-pumps via modifying sulfhydryl groups or by impairing mitochondrial or sarcoplasmatic reticulum function (Kaneko et al., 1989; Ytrehus et al., 1989). It is thus possible that the improved mechanical recovery observed in the present study during reperfusion with low concentrations of H2O2 could be the effect of moderately increased levels of intracellular calcium, whereas higher concentrations of H2O2 caused reduced contractility and development of ventricular contracture by promoting toxic calcium overload (Ytrehus et al., 1989; Barrington, 1990; Burton et al., 1990; Josephson et al., 1991; Wang et al., 1995). Combining the results of Burke et al. (1987), showing that low concentrations of H2O2 produces relaxation of intrapulmonary atrial rings via c-GMP, and the findings of Shah et al. (1995), indicating that pretreatment with a c-GMP analogue prevented impaired post-hypoxic relaxation in isolated cardiac myocytes without altering cytosolic Ca2+, it is tempting to propose that: the improved

2785

ventricular relaxation observed in the present study during post-ischaemic perfusion with 25 l H2O2 could be the result of the interaction between cytosolic Ca2+ and the myofilaments, mediated by cGMP. Treatment with 25 l H2O2 had a beneficial effect on post-ischaemic recovery of both ATP and creatine phosphate compared to the control group. Increased levels of high energy phosphates could contribute to explain the improved mechanical function during reperfusion with 25 l H2O2. Reduction of left ventricular contracture will further improve tissue perfusion as observed in the present study in the group exposed to 25 l H2O2. The improved metabolic recovery in the present study finally confirms that the improved contractile function did not cause metabolic deterioration. Griesmacher et al. (1993) have reported that addition of H2O2 (100 l) to human endothelial cells caused increased purine salvage resulting from changes in purine enzyme activities. These results are in accordance with the present metabolic results in isolated rat hearts. It is further known that creatine kinase is very sensitive to H2O2 (Suzuki et al., 1992). Reduction in creatine kinase activity may have an effect on the high energy phosphates content in the present study, but does not explain the pattern observed after non-ischaemic perfusion with 25 l H2O2. The result showing that H2O2 exposure causes a reduction of the infarct size in isolated rabbit hearts (Ytrehus et al., 1995), raises the question whether low levels of H2O2 have a direct effect on postischaemic cell death. It is well established that a burst of oxygen radicals is released at reperfusion (Bolli et al., 1989; Babbs et al., 1992). However, there is still controversy with respect to post-ischaemic treatment aimed at reducing the level of oxygen radicals (Ziegelstein et al., 1992). Takemura et al. (1992) recently questioned the importance of oxygen radical-related reperfusion damage, since they were not able to correlate a quantified amount of hydroxyl radicals released with reduced functional recovery, enzyme release or morphological damage. Addition of SOD as antioxidant treatment in ischaemia-reperfusion injury shows a bell-shaped dose-response curve, whereby SOD at higher concentrations loses its effectiveness and may even enhance the extent of reperfusion injury (Bernier et al., 1989; Omar et al., 1990, Mao et al., 1993). An explanation could be that the studies describe a dose-related effect of SOD promoted generation of H2O2. In conclusion, the present study shows that addition of 25 l H2O2 during reperfusion has a

2786

A.-C. Hegstad et al.

beneficial effect on both coronary flow, contractile function and level of high energy phosphates in isolated rat hearts subjected to 30 min global ischaemia. It is possible that to maintain a low stable level of H2O2 in the cardiac tissue during reperfusion is more important than to reduce the levels of shortlived oxygen radicals, like superoxide anions and hydroxyl radicals.

Acknowledgements We thank Professor Leif Jørgensen for his valuable comments on the manuscript, and Marit N. Nilsen and Elisabeth Børde for their excellent technical assistance. This study was supported by grants from the Norwegian Council on Cardiovascular Diseases, the Norwegian Research Council for Science and the Humanities and Laerdals Fond.

References A G, B LC, H GM, W HF, W ML, 1986. Reduction in experimental infarct size by recombinant human superoxide dismutase: insights into the pathophysiology of reperfusion injury. Circulation 74: 1424–1433 A G, W ML, J WE, F JT, 1987. Evidence for a reversible oxygen radical-mediated component of reperfusion injury: reduction by recombinant human superoxide dismutase administered at the time of reflow. Circulation 75: 282– 291. A G, S G, T I, E PP, D C, B A, S A, C M. 1992. Effects of ischemia and reperfusion on cardiac tolerance to oxidative stress. Am J Physiol 262: H23–H30. B CF, C MD, B SF, 1992. Histochemical demonstration of endothelial superoxide and hydrogen peroxide generation in ischaemic and reoxygenated rat tissues. Cardiovasc Res 26: 593–602. B PL, 1990. Effects of free radicals on the electrophysiological function of cardiac membranes. Free Radic Biol Med 9: 355–365. B A, H M, 1991. Alterations in electrical and contractile behaviour of isolated cardiomyocytes by hydrogen peroxide: possible ionic mechanisms. J Mol Cell Cardiol 23: 899–918. B M, M AS, H DJ, 1989. Reperfusion arrhythmias: dose-related protection by anti-free radical interventions. Am J Physiol 256: H1344–H1352. B A, S SK, S G, 1990. Oxidative stress alters specific membrane currents in isolated cardiac myocytes. Circ Res 67: 535–549. B R, J MO, P BS, A OI, H B, L EK, MC PB, 1989. Marked reduction of free radical generation and contractile dysfuncton by antioxidant therapy begun at the time of reperfusion. Evidence that myocardial stunning is a manifestation of reperfusion injury. Circ Res 65: 607–622.

B BJ, K RS, W AR, 1991. Regulation of endothelial cell prostaglandin synthesis by glutathione. J Biol Chem 266: 16659–16666. B TM, W MS, 1987. Hydrogen peroxide elicits pulmonary arterial relaxation and guanylate cyclase activation. Am J Physiol 252: H721–H732. B KP, 1985. Superoxide dismutase enhances recovery following myocardial ischemia. Am J Physiol 248: H637–H643. B KP, M AC, M KD, B LM, H HK, 1990. Free radicals alter ionic calcium levels and membrane phospholipids in cultured rat ventricular myocytes. J Mol Cell Cardiol 22: 1035–1047. E T, J H, E N, S JI, 1995. Effect of acidosis on hydrogen peroxide injury to the isolated perfused rat heart. Am J Physiol 269: H308–H312. F ML, C MV, S L, M MJ, M EJ, L D P AL, 1994. Different influence of superoxide anions and hydrogen peroxide on endothelial function of isolated cat cerebral and pulmonary arteries. Gen Pharmacol 25: 1197–1205. G A, W G, S I, W A, M MM, 1993. The H2O2 induced effects on purine metabolism in human endothelial cells. Free Radic Biol Med 15: 603–609. J DR, H D, S HM, 1991. Hydrogen peroxide-induced oxidative stress to mammalian heartmuscle cell (cardiomyocyte): lethal peroxidative membrane injury. J Cell Physiol 149: 347–364. J RA, S HS, L EG, S MD, Z JL, 1991. Study of the mechanisms of hydrogen peroxide and hydroxyl free radical-induced cellular injury and calcium overload in cardiac myocytes. J Biol Chem 266: 2354–2361. K T, M T, Y T, K KJ, 1989. Increase vs. decrease of calcium uptake by isolated heart cells induced by H2O2 vs. HOCl. Am J Physiol 256: C598–C607. K M, B RE, D NS, 1989. Depression of heart sarcolemmal Ca2+-pump activity by oxygen free radicals. Am J Physiol 256: H368–H374. K HA, 1985. Oxygen radicals in cerebral vascular injury. Circ Res 57: 508–516. L KE, K RC, K SY, J RL, H ML, 1991. Oxidative damage to the myocardium: a fundamental mechanism of myocardial injury. Cardioscience 2: 199–216. M GD, T PD, L GD, P MJ, 1993. Superoxide dismutase (SOD)-catalase conjugates. Role of hydrogen peroxide and the Fenton reaction in SOD toxicity. J Biol Chem 268: 416–420. M S, A M, S S, S N, 1988. Myocardial dysfunction and ultrastructural alterations mediated by oxygen metabolites. J Mol Cell Cardiol 20: 1009–1024. M¨  A, E JH, P TE, H DG, 1991. Release of intact endothelium-derived relaxing factor depends on endothelial dismutase activity. Am J Physiol 260: C219–C225. N¨  U, H¨  S, J G, P K, R S, M SL, 1992. Effects of reperfusion and peroxide dismutase on myocardial infarct size in a closed chest pig model. Cardiovasc Res 26: 170–178. N J, K DR, F JT, U N, M WT, C DR, C MV, V SF, 1989. Superoxide dismutase reduces reperfusion arrhythmias but fails to

H2O2 and Myocardial Reperfusion salvage regional function or myocardium at risk in conscious dogs. Circulation 79: 143–153. O BA, G NM, J MC, S SP, R WJ, D JM, MC JM, 1990. Cardioprotection by Cu,Zn-superoxide dismutase is lost at high doses in the reoxygenated heart. Free Radic Biol Med 9: 465–471. O T, T G, O T, A M, 1992. Effect of exogenous hydrogen peroxide on myocardial function and structure in isolated rat heart. Can J Cardiol 8: 989–997. P RV, K CW, M AJ, 1986. Peroxide mediated effects of homocysteine on arterial prostacyclin synthesis. Prostaglandins Leukot Med 22: 349–356. R T, T N, Z¨  A, S J, S L, 1993. Suppression of reperfusion induced arrhythmias in the isolated rat heart: pretreatment with 7-oxo prostacyclin in vivo. Cardiovasc Res 27: 1051–1055. R GM, V PM, 1986. Oxygen-derived free radicals, endothelium, and responsiveness of vascular smooth muscle. Am J Physiol 250: H815–H821. S OF, J P, A K, 1986. High performance liquid chromatography: a rapid isocratic method for determination of creatine compounds and adenine nucleotides in myocardial tissue. J Mol Cell Cardiol 18: 517–527. S AM, S HS, G EJ, S HA, L EG, 1995. cGMP prevents delayed relaxation at reoxygenation after brief hypoxia in isolated cardiac myocytes. Am J Physiol 268: H2396–H2404. S YJ, E JD, F GD, 1992. Inactivation of rabbit muscle creatine kinase by hydrogen peroxide. Free Radic Res Commun 16: 131–136. T G, O T, A M, 1992. Quantification of hydroxyl radical and its lack of relevance to myocardial injury during early reperfusion after graded ischemia in rat hearts. Circ Res 71: 96–105.

2787

U A, R KA, M CE, J RB, 1987. Failure of superoxide dismutase to limit size of myocardial infarction after 40 minutes of ischemia and 4 days of reperfusion in dogs. Circulation 75: 1237–1248. W SY, C JR, L GA, 1995. Increase in calcium leak channel activity by metabolic inhibition or hydrogen peroxide in rat ventricular myocytes and its inhibition by polycation. J Mol Cell Cardiol 27: 211–222. W CA, M MP, 1995. Role of protein kinase C in mediating effects of hydrogen peroxide in guineapig ventricular myocytes. J Mol Cell Cardiol 27: 1089– 1097. W SW, S MJ, D EM, C C, A GD, P B, L BR, 1985. The independent effects of oxygen radical scavengers on canine infarct size. Reduction by superoxide dismutase but not catalase. Circ Res 56: 895–898. Y K, M R, Mø OD, 1986. Influence of oxygen radicals generated by xanthine oxidase in the isolated perfused rat heart. Cardiovasc Res 20: 597–603. Y K, M R, O R, Mø OD, 1987. Ultrastructural changes induced in the isolated rat heart by enzymatically generated oxygen radicals. J Mol Cell Cardiol 19: 379–389. Y K, R S, Lø E, S TS, Mø OD, 1989. Mitochondrial calcium in hearts subjected to lipid peroxidation with contracture development. Basic Res Cardiol 84: 646–652. Y K, W RS, R SC, D JM, 1995. Hydrogen peroxide as a protective agent during reperfusion. A study in isolated perfused rabbit heart subjected to regional ischemia. Cardiovasc Res 30: 1033–1037. Z RC, Z JL, M ED, Y A, L EG, S MD, S HS, 1992. Dimethylthiourea, an oxygen radical scavenger, protects isolated cardiac myocytes from hypoxic injury by inhibition of Na+-Ca2+ exchange and not by its antioxidant effects. Circ Res 70: 804–811.