ABSTRACT
330One of the purposes of these experiments was to test the hypothesis that maximal rates of O2 uptake are regulated in the perfused liver by O2 concentrations far above the Km of cytochrome oxidase for O2. Oxygen uptake by the perfused liver decreased at O2 concentrations considerably higher than levels that caused NADH reduction when the input O2 concentration was varied. The maximal rate of O2 uptake measured with miniature O2 electrodes was two- to threefold higher in periportal (137 ± 8 µmol/g/h; O2 concentration = 478 ± 37 µM) than pericentral regions of the liver lobule (59 ± 5 (Jimol/g/h; O2 concentration = 263 ± 21 µM). The infusion of atractyloside, antimycin A, or KCN inhibited O2 uptake in both zones by 50–85%, indicating that respiration in both regions was dependent on mitochondrial electron transport. The content of ATP and ADP and ATP/ADP ratios were similar in microdissected samples from periportal and pericentral areas. On the other hand, when livers were perfused in the retrograde direction, O2 uptake was two- to threefold greater in pericentral than in periportal regions. Maximal rates of O2 uptake correlated with local O2 concentrations irrespective of the direction of flow when an electrode was moved across the liver lobule with a micromanipulator. Lower rates of O2 uptake in pericentral areas were not altered appreciably by infusion of agents known to uncouple oxidative phosphorylation (DNP), increase ADP supply (fructose), or elevate the NADH redox state (ethanol). In studies with microdissected sublobular regions (plugs), O2 uptake was also significantly higher in periportal than pericentral regions, and O2 uptake was directly proportional to O2 tension. However, the effect of O2 tension on O2 uptake was minimal in isolated hepatocytes and was absent in isolated mitochondria. Taken together, these studies are consistent with the hypothesis that an O2 sensor exists in the liver, most likely in nonparenchymal cells. In support of this hypothesis, we observed that O2 uptake was stimulated by arachidonic acid and inhibited by the lipoxygenase inhibitor, NDHGA.
In parallel studies, we evaluated whether modification of the energy state could protect against hypoxic injury. Perfusion of livers from fasted rats with hypoxic buffer caused hepatocellular damage within 30 min. LDH was released at maximal rates of ~300 U/g/h under these conditions, and virtually all cells in both regions of the liver lobule were stained with trypan blue. Infusion of glucose, xylitol, sorbitol, or mannitol (20 mM) did not appreciably change the time course or extent of damage; however, fructose (20 mM) prevented damage completely. A dose-dependent increase in glycolytic lactate production from fructose correlated well with cellular protection reflected by decreases in LDH release, and ATP: ADP ratios were also increased from 0.4 to 1.8 in a dose-dependent manner by fructose. The results indicate that fructose protects the liver against hypoxic cell death by the glycolytic production of ATP in the absence of O2.
