Elevation of glutathione levels by ammonium ions in primary cultures of rat astrocytes.

It is well established that ammonia is detoxified in the brain to form glutamine and that astrocytes play a major role in this process. The synthesis of glutamine requires glutamate and ATP. Since glutamate and ATP are also required for the synthesis of glutathione (GSH), we examined the effect of pathophysiological concentrations of ammonia on levels of GSH in primary cultures of astrocytes. GSH content in the medium increased in a dose- and time-dependent manner in the presence of ammonia. After an initial decrease, cellular GSH content increased in a similar manner. The levels of glutathione disulfide (GSSG) were also increased. A linear relationship was observed between ammonia concentration and the increase in GSH levels. An increase in the efflux of GSH from cells into medium was also observed under these conditions. Buthionine sulfoximine and acivicin, but not methionine sulfoximine, blocked the ammonia induced increase in GSH levels. No, or minor, changes in the activities of enzymes (gamma-glutamyl transpeptidase, GSH reductase and GSH-peroxidase) that might influence GSH levels were identified and thus could not account for the ammonia induced increase in GSH levels in astrocytes. These findings indicate that pathophysiological concentrations of ammonium ions result in increased astroglial levels of GSH which may affect the metabolism and function of astrocytes.

Association between cell swelling and glycogen content in cultured astrocytes.

Treatment of cultured rat astrocytes with hypotonic media or with 1 mM glutamate for 90 min caused cell swelling and a significant increase in glycogen content. Conversely, treatment with hypertonic media caused cell shrinkage with a corresponding decrease in astrocyte glycogen, which was proportional to the increasing osmolality of the hypertonic media. The glutamate receptor antagonist, MK-801, lowered both the glutamate-induced swelling and glycogen increase. These findings demonstrate a correlation between changes in cell volume and astrocyte glycogen content. This may explain the increased astrocytic glycogen observed in many neuropathological conditions where astrocyte swelling occurs. Because glycogen represents the largest energy reserve in the central nervous system, a swelling-induced disturbance in glycogen metabolism may lead to abnormal glial-neuronal interactions resulting in impaired brain bioenergetics.

Acute liver failure and hyperammonemia increase peripheral-type benzodiazepine receptor binding and pregnenolone synthesis in mouse brain.

We investigated the role of brain peripheral-type benzodiazepine receptors (PBRs) and pregnenolone (a product of PBRs activation) in hepatic encephalopathy (HE)/hyperammonemia. Administration of the hepatotoxin, thioacetamide, or ammonium acetate to mice for 3 days significantly increased the number of brain PBRs (138-146% of control) and the affinity of the ligands for these receptors (2-fold). The total content of pregnenolone and its rate of synthesis in brain of the experimental animals were significantly increased. Our results suggest a novel integrated mechanism by which ammonia-induced activation of PBRs leads to elevated levels of pregnenolone-derived neurosteroids which are known to enhance GABA-ergic neurotransmission. This mechanism may play a pivotal role in pathogenesis of HE.

Glycogen synthesis in the astrocyte: from glycogenin to proglycogen to glycogen.

The astrocyte of the newborn rat brain has proven to be a versatile system in which to study glycogen biogenesis. We have taken advantage of the rapid stimulation of glycogen synthesis that occurs when glucose is fed to astrocytes, and the marked limitation on this synthesis that occurs in astrocytes previously exposed to ammonium ions. These observations have been related to our earlier reports of the initiation of glycogen synthesis on a protein primer, glycogenin, and the discovery of a low-molecular-weight form of glycogen, proglycogen. The following conclusions have been drawn: 1) In the ammonia-treated astrocytes starved of glucose, free glycogenin is present. 2) When these astrocytes are fed with glucose, proglycogen is synthesized from the glycogenin primer by a glycogen-synthase-like UDPglucose transglucosylase activity (proglycogen synthase) distinct from the well-recognized glycogen synthase, and synthesis stops at this point. 3) Proglycogen is the precursor of macromolecular glycogen, which is synthesized from proglycogen by glycogen synthase when glucose is fed to untreated astrocytes, accounting for the much greater accumulation of total glycogen. 4) The stimulus to proglycogen and macroglycogen synthesis that occurs on feeding glucose to untreated or ammonia-treated astrocytes is the result of the activation of proglycogen synthase, not of glycogen synthase. 5) Therefore, in the synthesis of macromolecular glycogen from glycogenin via proglycogen, the step between glycogenin and proglycogen is rate-limiting. 6) The discovery of additional potential control points in glycogen synthesis, now emerging, may assist the identification of so-far-unexplained aberrations of glycogen metabolism.

The action of ammonia on astrocyte glycogen and glycogenolysis.

Most of the brain glycogen, a major energy reserve that can be mobilized in response to increased neuronal activity, resides in the astrocyte, the site of the neuropathological abnormality found in hepatic encephalopathy (HE). Ammonia, a neurotoxin implicated in the pathogenesis of HE, has been reported to cause a depletion of glycogen in primary astrocyte cultures. To further investigate the action of ammonia on glycogen levels, cultured astrocytes were exposed to ammonium chloride (1-5 mM) for various times up to 7 d. Treatment with ammonia for 24 h did not alter deoxyglucose uptake, but significantly lowered peak glycogen values (found at 1.5 h following feeding with medium containing 5.5 mM glucose) in a concentration-dependent manner. This inhibitory effect was not observed after longer exposure times to ammonia. Three day treatment of cells did, however, significantly reduce norepinephrine-stimulated glycogenolysis, an effect not seen after 1 d of ammonia treatment. Part of the neurotoxic action of long term ammonia exposure in humans and experimental animals may be to inhibit the breakdown of glycogen. The effect of ammonia on astrocyte glycogen synthesis and/or breakdown may disrupt glial neuronal signaling and thus play a role in the pathogenesis of HE.

Lowering of plasma free tryptophan in normal and portacaval shunt dogs.

The purpose of this study was to develop a method which would produce immediate and sustained lowering of plasma free tryptophan. Glucose, insulin, and mixtures of the two in different concentrations were infused into dogs over a period of 6 hr. Blood was sampled during the course of the infusion for assay of plasma free and total tryptophan, free fatty acids, glucose, and insulin. The infusate found to produce the most significant and sustained reduction in plasma free tryptophan in normal dogs was a mixture of 25% glucose and insulin. The lowering ranged from 26 to 47% during the course of the infusion. When the same infusion mixture was administered to dogs with established portacaval shunts, and elevated plasma free tryptophan concentrations (3.3 +/- 0.3 micrograms/ml compared to normal dogs which had 1.9 +/- 0.1 micrograms/ml), there was a 35% reduction to normal levels (2.0-2.2 micrograms/ml). A glucose-insulin mixture has, therefore, been found which, upon infusion into dogs, significantly lowers plasma free tryptophan levels. The effect of such an infusion on the plasma levels of free tryptophan, and the clinical course of patients with hepatic encephalopathy remains to be determined.

Insulin degradation in hepatic cirrhosis.

Hepatic insulinase activity was studied in cirrhosis in an attempt to explain the hyperinsulinemia known to occur in this disease. Liver tissue was obtained during laparotomy in seven patients with cirrhosis of the liver and five patients without liver disease. Insulin degradation was significantly decreased in the cirrhotic liver at each time interval measured (p less than 0.05). Insulinase inhibitor activity was measured in the plasma of 12 patients with cirrhosis of the liver and six controls. There was no significant difference between the two groups. Decreased insulinase activity in cirrhosis may account, in part, for the hyperinsulinemia seen in this disease.

Chronic catheterization of the portal vein in dogs.

The portal vein of dogs was catheterized using a silicone catheter with a short, straight, retroperitoneal route. This technique did not require daily irrigation to maintain patency, but permitted weekly collection of portal blood to be obtained readily.

A possible hepatic factor in the control of plasma free fatty acid levels.

Insulin acts as a regulator of lipid metabolism by inhibiting the rate of lipolysis. Hyperinsulinemia, produced by the peripheral infusion of glucose, therefore, causes a lowering of plasma free fatty acids (FFA) by decreasing the mobilization of fats from peripheral deposits. Based on previous studies, it was postulated that a lowering of plasma FFA could be produced, in the absence of hyperinsulinemia, by the infusion of glucose directly into the liver. This study was undertaken to investigate the existence of endogenous factors other than insulin, which inhibit lipolysis. It was determined that glucose at a dosage of 0.125 gm/kg/hr could be infused into the portal vein of dogs without increasing plasma insulin levels. This dosage significantly lowered plasma FFA levels when injected into the portal vein but had no effect on plasma FFA when infused peripherally. There were significant decreases in FFA turnover following portal glucose administration, with no changes in fractional turnover. This indicated that the lowered plasma FFA concentrations found following portal infusions of glucose, were due to reduced lipid mobilization from adipose tissue, rather than increased tissue uptake of FFA. These results suggested the existence of an hepatic factor, stimulated by portal glucose infusion, which lowered FFA mobilization from adipose tissue.

A modified procedure for the determination of 5-hydroxyindoleacetic acid in the urine and cerebrospinal fluid of patients with hepatic cirrhosis.

A method has been developed for the separation of 5-hydroxyindoleacetic acid from substances in urine and cerebrospinal fluid which interfere with its fluorometric determination. This involves the use of anion exchange column chromatography with elution of the 5-hydroxyindoleacetic acid by 3 mol/l formic acid in 60% methanol. The indole acid is then assayed fluorometrically after reaction with o-phtalaldehyde. Urinary 5-hydroxyindoleacetic acid values are reported for patients with stable cirrhosis and hepatic encephalopathy and compared with those found in control subjects without liver disease.

A longitudinal study of tryptophan involvement in hepatic coma.

Previous cross-sectional studies have demonstrated that hepatic coma is associated with abnormally high levels of plasma free tryptophan. Establishment of a more definitive relationship between this biochemical abnormality and hepatic coma requires an evaluation of biochemical changes in individual patients as they undergo alterations in the function of their central nervous systems. The present report is an evaluation of tryptophan and substances believed to influence its entry into the brain in six patients whose clinical status progressed from hepatic coma to complete recovery. Consistent significant decreases in plasma free tryptophan and free fatty acids were demonsrated to occur as the patients recovered. No consistent changes, however, were found in the plasma levels of the amino acids which have been reported to compete with tryptophan for transport across the blood-brain barrier, except for leucine which was significantly decreased in all six patients upon recovery from coma. Assay of lumbar cerebrospinal fluid for both tryptophan and the serotonin metabolite, 5-hydroxyindoleacetic acid (5-HIAA), showed no consistent changes as the patients recovered from hepatic coma.

Tryptophan and hepatic coma.

To clarify the involvement of tryptophan in the pathogenesis of hepatic coma, plasma and cerebrospinal fluid (CSF) tryptophan was studied in three patient groups (hepatic coma, stable cirrhosis, and control). An assessment of free fatty acids, some of the amino acids reported to compete with tryptophan for brain uptake, and albumin was also made. The data demonstrated that, whereas the elevated CSF tryptophan levels in cirrhotic patients compared to controls may have been attributable to decreased plasma branched chain amino acids, the elevated CSF tryptophan levels in hepatic coma compared to stable cirrhotic patients were probably attributable to increased plasma free tryptophan concentrations. Associated with the elevated plasma free tryptophan in coma patients was an increase in plasma free fatty acids and a marked decrease in serum albumin levels. Of all the amino acids investigated in the CSF, only tryptophan was significantly increased in patients in hepatic coma compared to cirrhotic patients not in coma.