Portacaval shunting causes differential mitochondrial superoxide production in brain regions.

The portacaval shunting (PCS) prevents portal hypertension and recurrent bleeding of esophageal varices. On the other hand, it can induce chronic hyperammonemia and is considered to be the best model of mild hepatic encephalopathy (HE). Pathogenic mechanisms of HE and dysfunction of the brain in hyperammonemia are not fully elucidated, but it was originally suggested that the pathogenetic defect causes destruction of antioxidant defense which leads to an increase in the production of reactive oxygen species (ROS) and the occurrence of oxidative stress. In order to gain insight into the pathogenic mechanisms of HE in the brain tissue, we investigated the effects of PCS in rats on free radicals production and activity levels of antioxidant and prooxidant enzymes in mitochondria isolated from different brain areas. We found that O production, activities of Mn-superoxide dismutase (Mn-SOD), glutathione peroxidase (GPx), glutathione reductase (GR), glutathione transferase (GT), nitric oxide synthase (NOS), and levels of carbonylated proteins differed between the four brain regions both in the amount and response to PCS. In PCS rats, Mn-SOD activity in the cerebellum was significantly decreased, and remained unchanged in the neocortex, hippocampus and striatum compared with that in sham-operated animals. Among the four brain regions in control rats, the levels of the carbonyl groups in mitochondrial proteins were maximal in the cerebellum. 4 weeks after PCS, the content of carbonylated proteins were higher only in mitochondria of this brain region. Under control conditions, O production by submitochondrial particles in the cerebellum was significantly higher than in other brain regions, but was significantly increased in each brain region from PCS animals. Indeed, the production of O by submitochondrial particles correlated with mitochondrial ammonia levels in the four brain regions of control and PCS-animals. These findings are the first to suggest that in vivo levels of ammonia in the brain directly affect the rate of mitochondrial O production.