Reactive oxygen intermediates and ischemia-reperfusion injury release tissue plasminogen activator from isolated rat hearts

Ischaemia-reperfusion injury impairs tissue plasminogen activator release in man

AIMS

Ischaemia-reperfusion (IR) injury causes endothelium-dependent vasomotor dysfunction that can be prevented by ischaemic preconditioning. The effects of IR injury and preconditioning on endothelium-dependent tissue plasminogen activator (t-PA) release, an important mediator of endogenous fibrinolysis, remain unknown.

METHODS AND RESULTS

Ischaemia-reperfusion injury (limb occlusion at 200 mmHg for 20 min) was induced in 22 healthy subjects. In 12 subjects, IR injury was preceded by local or remote ischaemic preconditioning (three 5 min episodes of ipsilateral or contralateral limb occlusion, respectively) or sham in a randomized, cross-over trial. Forearm blood flow (FBF) and endothelial t-PA release were assessed using venous occlusion plethysmography and venous blood sampling during intra-arterial infusion of acetylcholine (5-20 µg/min) or substance P (2-8 pmol/min). Acetylcholine and substance P caused dose-dependent increases in FBF (P<0.05 for all). Substance P caused a dose-dependent increase in t-PA release (P<0.05 for all). Acetylcholine and substanceP-mediated vasodilatation and substanceP-mediated t-PA release were impaired following IR injury (P<0.05 for all). Neither local nor remote ischaemic preconditioning protected against the impairment of substance P-mediated vasodilatation or t-PA release.

CONCLUSION

Ischaemia-reperfusion injury induced substanceP-mediated, endothelium-dependent vasomotor and fibrinolytic dysfunction in man that could not be prevented by ischaemic preconditioning.

CLINICAL TRIAL REGISTRATION INFORMATION

Reference number: NCT00789243, URL: http://clinicaltrials.gov/ct2/show/NCT00789243?term=NCT00789243&rank=1.

Variables affecting the transduction efficiency of adenovirus vectors in bovine aortic endothelial cells

Features and kinetics of Adenovirus (Ad)-mediated gene transfer to endothelial cells (EC) are not ultimately determined. We tested variables pertinent to the efficiency of Ad-mediated gene transfer to bovine aortic endothelial cells (BAEC) including: (1) Ad-vectors with different promoters, (2) kinetics of transduction efficiency of LacZ gene to BAEC, (3) the concentration and volume of vector-containing medium, (4) the period of incubation time of Ad vectors with BAEC, (5) the duration of transgene expression. An Ad5-LacZ vector with a cytomegalovirus (CMV) promoter transduced the LacZ gene to the cells more efficiently than vectors with the Rous sarcoma virus (RSV) promoter. However, both vectors exhibited a dose-dependent relationship between the vector multiplicity of infection (moi) and the percentage of LacZ-expressing cells. The higher moi of both vectors achieved nearly 100% of transduction efficiency in cultured BAEC. Although the Ad-CMV-LacZ vector better transduced the LacZ gene to BAEC than Ad-RSV-LacZ, a long period of vector exposure to BAEC could overcome the slightly difference in transduction efficiency between the two vectors. These results indicate that both Ad vectors are efficient for gene transfer to endothelial cells, and higher moi of vectors or a longer period exposure of vectors to EC can facilitate efficient transduction of foreign gene into EC in culture.

Aortic cross-clamping influences regional net release and uptake rates of tissue-type plasminogen activator in pigs

BACKGROUND

The key regulator of intravascular fibrinolysis, tissue-type plasminogen activator (t-PA), is released from a dynamic endothelial storage pool. The aim of the study was to investigate regional t-PA net release and uptake rates in response to infra-renal aortic cross-clamping (AXC) and declamping (DC).

METHODS

Anesthetized pigs were studied during 5 min of AXC, followed by a 35-min declamping (DC) period. Arterio-venous concentration gradients of total and active t-PA, as well as respective plasma flows, were simultaneously obtained across the preportal, hepatic, coronary and pulmonary vascular beds. Plasma levels of total t-PA (ELISA with purified porcine t-PA as standard), and active t-PA (spectrophotometric functional assay) were determined.

RESULTS

Prior to AXC, we found a high net release rate of total t-PA across the preportal vascular bed (1700 ng.min-1 P < 0.001), and a high hepatic net uptake (4900 ng.min-1, P < 0.001), while coronary and pulmonary t-PA net fluxes were small and variable. AXC per se did not induce significant alterations in net fluxes of t-PA. Following DC, preportal and coronary net releases of total t-PA increased (to 2900 ng.min-1 and 60 ng.min-1, respectively). Despite an increase in hepatic net uptake of total t-PA (to 6100 ng.min-1) after DC, a significant increase in hepatic venous total t-PA occurred.

CONCLUSIONS

The release and uptake of t-PA is indicated to be dynamic and organ-specific. DC induces an acute profibrinolytic reaction in preportal organs. The high hepatic t-PA uptake capacity restricts preportal profibrinolytic events to affect the systemic circulation.

Release of creatine kinase, troponin-T, and tissue plasminogen activator in arterial and coronary venous blood during coronary artery bypass surgery

Tissue plasminogen activator (t-PA) as a possible marker of endothelial injury during elective coronary artery bypass surgery was studied. T-PA antigen and activity were measured in arterial and coronary venous plasma in 14 patients, and compared to the markers of myocyte injury creatine kinase (CK-MB) and troponin-T (TnT). Cardiopulmonary bypass (CPB) lasted 86 (55-107) min, and aortic cross-clamping (cold, crystalloid cardioplegia) lasted 41 (25-62) min (median (central 90% percentile)). Blood flow in the great cardiac vein was measured by retrograde thermodilution, and increased from 49 (27-90) ml/min before CPB to a maximum of 92 (55-125) ml/min 40 min after declamping (not significant). CK-MB, TnT, and t-PA antigen and activity all increased during CPB, and were significantly higher in coronary sinus than arterial plasma after declamping the aorta. Net cardiac release ([coronary sinus-arterial concentration] x coronary flow) of TnT increased after the aorta was declamped, and was higher in the seven patients with the longest cross-clamping time than in the seven with the shortest time (p < 0.01). Cardiac release of CK-MB and t-PA antigen also increased after declamping, but with no significant difference between long and short cross-clamp times. t-PA activity, however, increased more in the patients with the longest cross-clamp times (p < 0.008). In conclusion, CK-MB, TnT and t-PA were released from the postcardioplegic heart. Release of t-PA indicates that postcardioplegic coronary endothelial activation or injury occurred t-PA activity as well as TnT increased more in patients with long times of cross-clamping, indicating that t-PA activity may be a possible marker of postcardioplegic endothelial injury or activation.

Oxidative stress and release of tissue plasminogen activator in isolated rat hearts

To evaluate the potential of tissue plasminogen activator (t-PA) as a marker of endothelial activation or injury, the dose-response relationship between reactive oxygen intermediates and t-PA release was investigated in isolated rat hearts. After stabilization the hearts were perfused for 10 minutes with different concentrations of hydrogen peroxide (H2O2) (0 (control perfusion), 20, 40, 80, 120, 160, or 200 microM) (n = 8 hearts/group), followed by 30 minutes recovery. Higher concentrations than 80 microM induced cardiac dysfunction and a dose-dependent release of lactate dehydrogenase, indicating myocyte injury. H2O2-concentrations of 80 microM and more caused a significant, but temporary t-PA release. Peak t-PA release occurred more rapidly with higher concentrations, but otherwise there was no difference dependent on the H2O2-dose. The effects of H2O2 (120 or 200 microM) on t-PA release were also compared to the effects of bradykinin. Both were given for 10 minutes as above, and the procedure was repeated after 10 minutes recovery. Bradykinin (50 or 500 nM) released t-PA with the same magnitude, but with peak values occurring earlier than t-PA release induced by H2O2. Bradykinin, but not H2O2, induced t-PA release during the second exposure, suggesting different mechanisms of release.

IN CONCLUSION

Perfusion with H2O2 leads to a dose-dependent myocardial injury in isolated rat hearts. H2O2 also causes an acute t-PA release without dose-dependency, suggesting an all or nothing response of the endothelium. t-PA may be used as an indicator of, but cannot quantify endothelial activation or injury.

Release of markers of myocardial and endothelial injury following cold cardioplegic arrest in pigs

Cold cardioplegic arrest causes reperfusion injury to both endothelium and myocardium. We investigated release of troponin-T (TnT), tissue plasminogen activator activity (t-PA) and histamine (HA) from the heart before and after 2h of cold crystalloid cardioplegia in eight Swedish landrace pigs. Coronary sinus blood flow was measured in an external shunt between the coronary sinus and the right atrium. TnT, t-PA and HA were measured concomitantly in arterial and coronary sinus plasma, and the cardiac release was calculated. Cardiac release of TnT increased from 18 (15-25) micrograms/min (median (central 90% percentile)) before cold cardioplegia to maximum 281 (132-510) micrograms/min 30 min after aortic declamping (p < 0.02 vs initial value). t-PA rose from -4 (-52-34) to maximum 249 (75-691) IU/min 2 min after declamping (p < 0.01) and thereafter returned to baseline levels. The net cardiac release of HA was 72 (-80-1321) nmol/min before cardioplegia, rising to 234 (-188-524) after 2 min of reperfusion (p < 0.02) and returning to baseline after 30 minutes. We conclude that the porcine heart releases t-PA, Tn-T and HA during postcardioplegic reperfusion. The differing kinetics of their release may indicate different affection of the myocardium and the endothelium. Tn-T, t-PA and HA are potential markers of myocardial and endothelial injury in the porcine heart.

Perfusing isolated rat hearts with hydrogen peroxide: an experimental model of cardiac dysfunction caused by reactive oxygen species

A model of cardiac dysfunction induced by reactive oxygen species (ROS) was established by adding hydrogen peroxide (H2O2) to the perfusate of isolated, Langendorff-perfused rat hearts, and the mechanism of functional injury was investigated. The following groups were included: 1 (n = 7), control perfusion; 2 (n = 11), perfusion with H2O2 (180 mumol 1(-1) for 10 min followed by recovery for 50 min; 3 (n = 4), control perfusion with N-acetylcysteine (NAC, 100 mumol 1(-1); 4 (n = 7), perfusion with H2O2 and NAC; 5 (n = 4), control perfusion with thiourea (15 mmol 1(-1), 6 (n = 7), H2O2 and thiourea together; 7 (n = 4), control perfusion with catalase (150 U ml-1); 8 (n = 7), catalase and H2O2, 9 (n = 4), control perfusion with deferoxamine (5 mmol 1(-1); and 10 (n = 7), deferoxamine and H2O2. coronary flow (CF), left ventricular developed pressure (LVDP), left ventricular end-diastolic pressure (LVEDP), and heart rate (HR) were measured. All values are mean +/- SEM. When given alone, catalase, thiourea, NAC and deferoxamine did not influence left ventricular pressures, but NAC, catalase and thiourea increased CF. H2O2 increased CF (maximum 146 +/- 6% of baseline value after 5 min, p < 0.001 compared to group 1), decreased LVDP (minimum 14 +/- 5% of baseline value after 10 min, p < 0.0004), and increased LVEDP (from 0 mmHg to a maximum of 54 +/- 7 mmHg after 5 min recovery, p < 0.0003). All these changes gradually reversed during recovery. Catalase and thiourea both inhibited the H2O2-induced effects, but catalase inhibition was more complete. Neither NAC nor deferoxamine had any effect on H2O2-induced cardiac dysfunction. In conclusion, H2O2 perfusion is a convenient and reversible model of ROS-induced functional injury to isolated rat hearts. H2O2, rather than the hydroxyl radical, seems to be the main injurious ROS in this model.

Release of tissue plasminogen activator during reperfusion after different times of ischaemia in isolated, perfused rat hearts

Tissue plasminogen activator (t-PA) is a potential marker of endothelial cell activation or injury. The relationship between duration of ischaemia and release of t-PA during reperfusion was investigated in isolated rat hearts exposed to either 5, 10, 20, 30, 40, or 60 min of global, normothermic ischaemia followed by 30 min of reperfusion (n = 8 in each group). t-PA activity was measured (chromogenic peptide substrate assay) in the effluent before ischaemia, and after 2.5, 5, 7.5, 10, 20, and 30 min of reperfusion. Release of lactate dehydrogenase (LD), a marker of myocyte injury, was measured before ischaemia and after 5 min reperfusion. Left ventricular pressures were measured by a balloon in the left ventricle. Ischaemia for 20 min or less had only minor effects on cardiac function. Thirty min or more of ischaemia induced ventricular fibrillation during reperfusion in most hearts. After ischaemia t-PA outflow increased, but without any significant difference between groups. Peak release occurred after either 2.5 or 5 min of reperfusion. After 10 min reperfusion the release was not different from the basal value. In contrast, postischaemic release of LD correlated to the length of ischaemia. To conclude, t-PA release from the ischaemic-reperfused rat heart is independent of the length of ischaemia. Thus the potential of t-PA to quantify endothelial injury appears to be limited.

The role of nitric oxide in the cardiac effects of hydrogen peroxide

Oxidative stress mediated by hydrogen peroxide (H2O2) increases coronary flow (CF) in Langendorff-perfused rat hearts. We investigated the possible role of nitric oxide (NO) in H2O2-induced vasodilation. A dose-response study was conducted to find a concentration of H2O2 which increased CF without influencing left ventricular developed (LVDP) or end-diastolic (LVEDP) pressures. 80(n = 10), 100 (n = 7), 120 (n = 7), 140 (n = 7), 160 (n = 7), and 180 (n = 10) microM H2O2 was infused for 10 min, followed by recovery for 50 min. 80 microM H2O2 increased CF to a maximum of 143 +/- 4 (mean +/- S.E.M) percent of initial value after 15 min observation (p < 0.001 compared to buffer only), with no effect on LVDP or LVEDP. Another series of hearts were perfused with N-nitro-L-Arginine methylester (L-NAME, 1 mM), methylene blue (MB, 50 microM), or haemoglobin (Hb, 10 microM), without (n = 7 in each) or with (n = 10 in each) 80 microM H2O2 for 10 min. L-NAME, MB, and Hb alone increased CF, but attenuated the H2O2-induced increase of CF.LVDP was depressed when L-NAME, MB or Hb were given in conjunction with 80 microM H2O2. In summary, H2O2 concentration-dependently increased LVEDP and depressed LVDP. The H2O2-induced increase of CF was independent of concentration. Inhibition of NO synthesis, action, or soluble guanylate cyclase attenuated the H2O2-induced increase of CF, and depressed LVDP when given together with H2O2. H2O2 induces a NO-dependent vasodilation, and inhibition of NO is detrimental to left ventricular function after H2O2-mediated oxidative stress.

The effect of exogenous adenosine on functional injury caused by hydrogen peroxide in the isolated rat heart

Adenosine is an endogenous cardioprotective substance. The present study examines whether exogenous adenosine attenuates cardiac injury induced by oxidative stress. Rat hearts (Langendorff model) were perfused with H2O2 (180 microM) for 10 min, then recovered for 60 min (n = 10). In other groups adenosine 55 microM, 11 0 microM, or 220 microM (n = 10 in each) was given in addition to H2O2 throughout perfusion. Control perfusion with Krebs Henseleit only (n = 7), adenosine 110 microM throughout perfusion (n = 7), and adenosine 110 microM as an intervention (n = 7) was performed. The hearts were paced at 320 beats/min. Left ventricular systolic (LVSP) and end-diastolic (LVEDP) pressures were measured together with coronary flow (CF), and left ventricular developed pressure (LVDP = LVSP - LVEDP) was calculated. H2O2 decreased LVSP from 105 +/- 8 to 60 +/- 5 mmHg (mean +/- SEM) after 10 min infusion (p < 0.008). Adenosine did not attenuate the decrease of LVSP. LVEDP increased from 0 to 59 +/- 10 mmHg (p < 0.004) and 62 +/- 11 mmHg 5 and 15 min after end of infusion of H2O2, respectively. Neither 55 microM nor 220 microM adenosine inhibited the H2O2-induced increase of LVEDP. Adenosine 110 microM attenuated the increase after 15 (15 +/- 4 mmHg, p < 0.004) and 25 min observation (26 +/- 7 mmHg, p < 0.012). Adenosine did not attenuate the reduction of LVDP. CF initially increased during infusion of H2O2, thereafter decreased. Hearts given adenosine had higher basal CF, and CF did not increase after H2O2. Control perfusion with adenosine, given throughout perfusion or as an intervention, increased CF and tended to increase LVSP. In summary, adenosine did not inhibit H2O2-induced depression of contractility or reduction of CF. One concentration of adenosine (110 microM) attenuated H2O2-induced impairment of relaxation. Exogenous adenosine does not have an important influence on functional injury caused by exogenous oxidants.