We determined the functional role of nitric oxide (NO) and endothelins (ET), two potent vasoactive mediator systems in the liver, for the pathogenesis of sinusoidal perfusion failure and lethal hepatocyte injury after low-flow ischemia/reperfusion in the isolated perfused rat liver. NO synthase blockade with N -nitro-L-arginine methyl ester (L-NAME) (10 Οͺ3 mol/L) before reperfusion prevented increased NO 2 Οͺ /NO 3 Ψ , the final products of NO oxidation, which could be observed in the vehicle group. Epifluorescence microscopy revealed that the decrease in functional sinusoid density during reperfusion was much more profound compared with vehicle. This was associated with a lower surface PO 2 , a substantially higher number of nonviable hepatocytes, as assessed by in situ propidium iodide staining, and enhanced enzyme release into the perfusate compared with vehicle. In contrast, reperfusion in the presence of the endothelin AΟ©B receptor antagonist bosentan (2 Ψ 10 Οͺ4 mol/L) restored functional sinusoid density and surface PO 2 to baseline values, resulted in a small reduction in the number of propidium iodide-positive hepatocytes, and caused similar increases in enzyme release as compared with vehicle. This indicates that hepatic generation of NO attenuates sinusoidal perfusion failure and improves liver tissue oxygenation, thus limiting hepatocyte injury during early reperfusion after hepatic low-flow ischemia. In contrast, endothelins counteract the microcirculatory effects of NO, i.e., mediate the no-reflow in hepatic sinusoids; however, the restoration of functional sinusoid density with bosentan resulted only in a small reduction in tissue damage, suggesting that additional components, which are independent of microcirculatory failure, contribute to hepatic reperfusion injury under these conditions. (HEPATOL-OGY 1998;27:755-764.)
Reperfusion following hepatic ischemia is frequently associated with an exacerbation of injury, which may ultimately result in liver dysfunction or failure. 1 Based on the central role that the liver plays in the organism' s metabolic and immunologic response to injury, 2 the characterization of the pathophysiological mechanisms that cause hepatic reperfusion injury is of particular importance.
An accumulating body of evidence suggests that failure of the hepatic microcirculation is a major component of reperfusion injury in the liver. For example, we observed a substantial decrease in the number of perfused centrilobular areas and perfused sinusoids per unit area on the surface of the rat liver after ischemia/reflow in vivo. 3 Similarly, Koo et al. showed that reperfusion after hepatic ischemia results in an initial transient return of blood flow followed by a progressive reduction or even complete cessation of sinusoidal flow. 4 Furthermore, we previously showed that physical prevention of microvascular shut down, using a flow-controlled reperfusion mode, largely prevented parenchymal cell necrosis in isolated perfused rat livers after ischemia/reperfusion, i.e., the degree of microcirculatory failure determined the extent of lethal hepatocyte injury. 5 However, the vasoactive mediators involved in the regulation of sinusoidal flow during reperfusion following hepatic ischemia remain to be identified.
Recent data provided some evidence that endotheliumderived vasoconstrictors and vasodilators may be functionally important to the control of total hepatic blood flow under these conditions. Reperfusion after hepatic ischemia results in increased liver tissue concentrations and hepatic venous plasma levels of endothelin (ET)-1, 6-7 a member of a family of potent and long-acting vasoconstrictive mediators. 8 In addition, Goto et al. showed that administration of ET-1 antiserum before reperfusion following complete lobar hepatic ischemia resulted in a higher index of hepatic blood volume when compared with reperfusion in the absence of a specific anti-ET serum. 9 Furthermore, we recently showed that ETs reduce total liver blood flow during reperfusion after hepatic low-flow ischemia in the intact rat in vivo. 10 Because ET-1 is a potent vasoconstrictor in the hepatic microcirculation that acts at both extrasinusoidal and sinusoidal sites, 11 and because our previous results suggest that ETs contribute significantly to sinusoidal perfusion failure under other pathologic conditions such as endotoxemia [12][13] or chronic ethanol consumption, 14 we hypothesized that the no-reflow phenomenon in hepatic sinusoids during reperfusion following hepatic low-flow ischemia is mediated by ETs.
On the other hand, ET-induced constriction of hepatic sinusoids can be inhibited by nitric oxide (NO)-mediated vasodilatory action 15 ; we have previously shown that a dynamic balance between the two mediator systems becomes evident and regulates sinusoidal flow after endotoxin challenge. 13 Interestingly, it has been recently reported that