AIM: To assess whether portacaval anastomosis (PCA) in rats affects the protein expression and/or activity of glutaminase in kidneys, intestines and in three brain areas of cortex, basal ganglia and cerebellum and to explain the neurological alterations found in hepatic en-cephalopathy (HE). METHODS: Sixteen male Wistar rats weighing 250-350 g were grouped into sham-operation control (n=8) or portacaval shunt (n=8). Twenty-eight days after the procedure, the animals were sacrificed. The duodenum, kidney and brain were removed, homogenised and mitochondria were isolated. Ammonia was measured in brain and blood. Phosphate-activated glutaminase (PAG) activity was determined by measuring ammonia production following incubation for one hour at 37℃with O-phthalaldehyde (OPA) and specific activity expressed in units per gram of protein (μkat/g of protein). Protein expression was measured by immunoblotting. RESULTS: Duodenal and kidney PAG activities together with protein content were significantly higher in PCA group than in control or sham-operated rats (duodenum PAG activity was 976.95±268.87μkat/g of protein in PCA rats vs 429.19±126.92.μkat/g of protein in sham-operated rats; kidneys PAG activity was 1259.18±228.79μkat/g protein in PCA rats vs 669.67±400.8μkat/g of protein in controls, P<0.05; duodenal protein content: 173% in PCA vs sham-operated rats; in kidneys the content of protein was 152% in PCA vs sham-operated rats). PAG activity and protein expression in PCA rats were higher in cortex and basal ganglia than those in sham-operated rats (cortex: 6646.6±1870.4μkat/g of protein vs 3573.8±2037.4μkat/g of protein in control rats, P<0.01; basal ganglia, PAG activity was 3657.3±1469.6μkat/g of protein in PCA rats vs 2271.2±384μkat/g of protein in sham operated rats, P<0.05; In the cerebellum, the PAG activity was 2471.6±701.4μkat/g of protein vs 1452.9±567.8μkat/g of protein in the PCA and sham rats, respectively, P<0.05; content of protein: cerebral cortex: 162%±40% vs 100%±26%, P< 0.009; and basal ganglia: 140%±39% vs 100%±14%, P<0.05; but not in cerebellum: 100%±25% vs 100%±16%,P=ns). CONCLUSION: Increased PAG activity in kidney and duodenum could contribute significantly to the hyperam-monaemia in PCA rats, animal model of encephalopathy. PAG is increased in non-synaptic mitochondria from the cortex and basal ganglia and could be implicated in the pathogenesis of hepatic encephalopathy. Therefore, PAG could be a possible target for the treatment of HE or liver dysfunction. AIM: To assess whether portacaval anastomosis (PCA) in rats affects the protein expression and / or activity of glutaminase in kidneys, intestines and in three brain areas of cortex, basal ganglia and cerebellum and to explain the neurological alterations found in hepatic en-cephalopathy METHODS: Sixteen male Wistar rats weighing 250-350 g were grouped into sham-operation control (n = 8) or portacaval shunt (n = 8). Twenty-eight days after the procedure, the animals were sacrificed. The Ammonia was measured in brain and blood. Phosphate-activated glutaminase (PAG) activity was determined by measuring ammonia production following incubation for one hour at 37 ° C with O-phthalaldehyde (OPA ) and specific activity expressed in units per gram of protein (μkat / g of protein). Protein expression was measured by immunoblotting. RESULTS: Duodenal and kidney PAG activities together with protein content were signi ficantly higher in PCA group than in control or sham-operated rats (duodenum PAG activity was 976.95 ± 268.87μkat / g of protein in PCA rats vs 429.19 ± 126.92.μkat / g of protein in sham-operated rats; kidneys PAG activity was 1259.18 ± 228.79 μ kat / g protein in PCA rats vs 669.67 ± 400.8 μ kat / g of protein in controls, P <0.05; duodenal protein content: 173% in PCA vs sham-operated rats; in kidneys the content of protein was 152% in PCA vs sham-operated rats). PAG activity and protein expression in PCA rats were higher in cortex and basal ganglia than those in sham-operated rats (cortex: 6646.6 ± 1870.4 μ kat / g of protein vs 3573.8 ± 2037.4 μ kat / g of protein in control rats, P <0.01; basal ganglia, PAG activity was 3657.3 ± 1469.6 μ kat / g of protein in PCA rats vs 2271.2 ± 384 μ kat / g of protein in sham operated rats, P <0.05; In the cerebellum, the PAG activity was 2471.6 ± 701.4 μ kat / g of protein vs 1452.9 ± 567.8 μ kat / g of protein in the PCA and sham rats, respectively, P <0.05; content of protein: ce rebral cortex: 162% ± 40%vs 100% ± 26%, P <0.009; and basal ganglia: 140% ± 39% vs 100% ± 14%, P <0.05; but not in cerebellum: 100% ± 25% vs 100% ± 16%, P = ns). CONCLUSION: Increased PAG activity in kidney and duodenum could contribute significantly to the hyperam-monaemia in PCA rats, animal model of encephalopathy. PAG is increased in non-synaptic mitochondria from the cortex and basal ganglia and could be implicated in the pathogenesis of hepatic encephalopathy. Thus, PAG could be a possible target for the treatment of HE or liver dysfunction.