Liver failure, or shunting of intestinal blood around the liver, results in hyperammonemia and cerebral dysfunction. Recently it was shown that ammonia caused some of the metabolic signs of hepatic encephalopathy only after it was metabolized by glutamine synthetase in the brain. In the present study, small doses of methionine sulfoximine, an inhibitor of cerebral glutamine synthetase, were given to rats either at the time of portacaval shunting or 3-4 weeks later. The effects on several characteristic cerebral metabolic abnormalities produced by portacaval shunting were measured 1-3 days after injection of the inhibitor. All untreated portacaval-shunted rats had elevated plasma and brain ammonia concentrations, increased brain glutamine and tryptophan content, decreased brain glucose consumption, and increased permeability of the blood-brain barrier to tryptophan. All treated rats had high ammonia concentrations, but the brain glutamine content was normal, indicating inhibition of glutamine synthesis. One day after shunting and methionine sulfoximine administration, glucose consumption, tryptophan transport, and tryptophan brain content remained near control values. In the 3-4-week-shunted rats, which were studied 1-3 days after methionine sulfoximine administration, the effect was less pronounced. Brain glucose consumption and tryptophan content were partially normalized, but tryptophan transport was unaffected. The results agree with our earlier conclusion that glutamine synthesis is an essential step in the development of cerebral metabolic abnormalities in hyperammonemic states.

1. It has been established that chronic hyperammonaemia, whether caused by portacaval shunting or other means, leads to a variety of metabolic changes, including a depression in the cerebral metabolic rate of glucose (CMRGIC), increased permeability of the blood-brain barrier to neutral amino acids, and an increase in the brain content of aromatic amino acids. The preceding paper [Jessy, DeJoseph & Hawkins (1991) Biochem. J. 277, 693-696] showed that the depression in CMRGlC caused by hyperammonaemia correlated more closely with glutamine, a metabolite of ammonia, than with ammonia itself. This suggested that ammonia (NH3 and NH41) was without effect. The present experiments address the question whether ammonia, in the absence of net glutamine synthesis, induces any of the metabolic symptoms of cerebral dysfunction associated with hyperammonaemia. 2. Small doses of methionine sulphoximine, an inhibitor of glutamine synthetase, were used to raise the plasma ammonia levels of normal rats without increasing the brain glutamine content. These hyperammonaemic rats, with plasma and brain ammonia levels equivalent to those known to depress brain function, behaved normally over 48 h. There was no depression of cerebral energy metabolism (i.e. the rate of glucose consumption). Contents of key intermediary metabolites and high-energy phosphates were normal. Neutral amino acid transport (tryptophan and leucine) and the brain contents of aromatic amino acids were unchanged. 3. The data suggest that ammonia is without effect at concentrations less than 1 /tmol/ml if it is not converted into glutamine. The deleterious effect of chronic hyperammonaemia seems to begin with the synthesis of glutamine.

1. Portacaval shunting in rats results in several metabolic alterations similar to those seen in patients with hepatic encephalopathy. The characteristic changes include: (a) diminution of cerebral function; (b) raised plasma ammonia and brain glutamine levels; (c) increased neutral amino acid transport across the blood-brain barrier; (d) altered brain and plasma amino acid levels; and (e) changes in brain neurotransmitter content. The aetiology of these abnormalities remains unknown. 2. To study the degree to which ammonia could be responsible, rats were made hyperammonaemic by administering 40 units of urease/kg body weight every 12 h and killing the rats 48 h after the first injection. 3. The changes observed in the urease-treated rats were: (a) whole-brain glucose use was significantly depressed, whereas the levels of high-energy phosphates remained unchanged; (b) the permeability of the blood-brain to barrier to two large neutral amino acids, tryptophan and leucine, was increased; (c) blood-brain barrier integrity was maintained, as indicated by the unchanged permeability-to-surface-area product for acetate; (d) plasma and brain amino acid concentrations were altered; and (e) dopamine, 5-hydroxytryptamine (serotonin) and noradrenaline levels in brain were unchanged, but 5-hydroxyindoleacetic acid (5-HIAA), a metabolite of 5-hydroxytryptamine, was elevated. 4. The depressed brain glucose use, increased tryptophan permeability-to-surface-area product, elevated brain tryptophan content and rise in the level of cerebral 5-HIAA were closely correlated with the observed rise in brain glutamine content. 5. These results suggest that many of the metabolic alterations seen in rats with portacaval shunts could be due to elevated ammonia levels. Furthermore, the synthesis or accumulation of glutamine may be closely linked to cerebral dysfunction in hyperammonaemia.

No abstract

Recent studies showed that hyperammonaemia caused many of the metabolic changes in portacaval-shunted rats, a model of hepatic encephalopathy. These changes included a depression in the cerebral metabolic rate of glucose (CMRGlc), an indication of decreased brain function. 2. The purpose of the present experiments was to determine whether the depression of CMRGlc caused by ammonia is confined to certain brain structures, or whether the depression is an overall decrease in all structures, such as occurs in portacaval-shunted rats. To accomplish this objective, rats were made hyperammonaemic by giving them intraperitoneal injections of 40 units of urease/kg body wt. every 12 h; control rats received 0.154 m-NaCl. CMRGlc was measured 48 h after the first injection, by using quantitative autoradiography with [6-14C]glucose as a tracer. 3. The experimental rats had high plasma ammonia concentrations (control 70 nmol/ml, experimental 610 nmol/ml) and brain glutamine levels (control 5.4 mumol/ml). Hyperammonaemia decreased CMRGlc throughout the brain by an average of 19%. CMRGlc showed an inverse correlation with plasma ammonia, but a stronger correlation with the brain glutamine content. 4. Hyperammonaemia led to a decrease in CMRGlc throughout the brain that was indistinguishable from the pattern seen in portacaval-shunted rats. This is taken as further evidence that the cerebral depression found in portacaval-shunted rats is a consequence of hyperammonaemia. The observation that depression of CMRGlc correlated more closely with brain glutamine content than with plasma ammonia suggests that metabolism of ammonia is an important step in the pathological sequence.