Effect of portacaval anastomosis on electrically stimulated release of glutamate from rat hippocampal slices

Neuroprotective effects of guanosine administration on behavioral, brain activity, neurochemical and redox parameters in a rat model of chronic hepatic encephalopathy

It is well known that glutamatergic excitotoxicity and oxidative stress are implicated in the pathogenesis of hepatic encephalopathy (HE). The nucleoside guanosine exerts neuroprotective effects through the antagonism against glutamate neurotoxicity and antioxidant properties. In this study, we evaluated the neuroprotective effect of guanosine in an animal model of chronic HE. Rats underwent bile duct ligation (BDL) and 2 weeks later they were treated with i.p. injection of guanosine 7.5 mg/kg once a day for 1-week. We evaluated the effects of guanosine in HE studying several aspects: a) animal behavior using open field and Y-maze tasks; b) brain rhythm changes in electroencephalogram (EEG) recordings; c) purines and glutamate levels in the cerebral spinal fluid (CSF); and d) oxidative stress parameters in the brain. BDL rats presented increased levels of glutamate, purines and metabolites in the CSF, as well as increased oxidative damage. Guanosine was able not only to prevent these effects but also to attenuate the behavioral and EEG impairment induced by BDL. Our study shows the neuroprotective effects of systemic administration of guanosine in a rat model of HE and highlights the involvement of purinergic system in the physiopathology of this disease.

Effects of CA1 glutamatergic systems upon memory impairments in cholestatic rats

BACKGROUND

Bile duct ligation (BDL) is shown to induce cholestasis-related liver function impairments as well as consequent cognitive dysfunctions (i.e. impaired learning and memory formation). Glutamatergic neurotransmission plays an important role in hippocampal modulation of learning and memory function. The present study aimed to investigate the possible involvement of dorsal hippocampal (CA1) glutamatergic systems upon cholestasis-induced amnesia.

METHOD

Cholestasis was induced in male Wistar rats through double-ligation of the main bile duct (at two points) and transection of the interposed segment. Step-through passive avoidance test was employed to examine rats' learning and memory function. All drugs were injected into CA1 region of the hippocampus.

RESULTS

our results indicated a decrease in memory retrieval following cholestasis (11, 17 and 24 days post BDL). Only subthreshold doses of N-methyl-d-aspartate (NMDA; 0.125 and 0.25 μg/μl) but not its effective dose (0.5 μg/μl), restored the cholestasis-induced amnesia in step-through passive avoidance test, 11, 17 and 24 days post BDL. This effect was blocked by the subthreshold dose of D-[1]-2-amino-7-phosphonoheptanoic acid (D-AP7, NMDA receptor antagonist; 0.0625 μg/μl, intra-CA1) at 0.125 μg/μl and 0.25 μg/μl doses of NMDA. Moreover, our data revealed that only effective doses of D-AP7 (0.125 and 0.25 μg/μl, intra-CA1) potentiate memory impairments in 11 days after BDL. It was noted that none of applied drugs/doses exerted an effect on memory acquisition and locomotors activity, 10 and 12 days post laparotomy, respectively.

CONCLUSION

Our findings suggest the potential involvement of CA1 glutamatergic system(s) in cholestasis-induced memory deficits.

Persistence of cognitive impairment after resolution of overt hepatic encephalopathy

BACKGROUND & AIMS

In patients with cirrhosis, hepatic encephalopathy (HE) has acute but reversible as well as chronic components. We investigated the extent of residual cognitive impairment following clinical resolution of overt HE (OHE).

METHODS

Cognitive function of cirrhotic patients was evaluated using psychometric tests (digit symbol, block design, and number connection [NCT-A and B]) and the inhibitory control test (ICT). Improvement (reduction) in ICT lures and first minus second halves (DeltaL(1-2)) were used to determine learning of response inhibition. Two cross-sectional studies (A and B) compared data from stable cirrhotic patients with or without prior OHE. We then prospectively assessed cognitive performance, before and after the first episode of OHE.

RESULTS

In study A (226 cirrhotic patients), 54 had experienced OHE, 120 had minimal HE, and 52 with no minimal HE. Despite normal mental status on lactulose after OHE, cirrhotic patients were cognitively impaired, based on results from all tests. Learning of response inhibition (DeltaL(1-2) > or =1) was evident in patients with minimal HE and no minimal HE but was lost after OHE. In study B (50 additional patients who developed > or =1 documented OHE episode during follow-up), the number of OHE hospitalizations correlated with severity of residual impairment, indicated by ICT lures (r = 0.5, P = .0001), digit symbol test (r = -0.39, P = .002), and number connection test-B (r = 0.33, P = .04). In the prospective study (59 cirrhotic patients without OHE), 15 developed OHE; ICT lure response worsened significantly after OHE (12 before vs 18 after, P = .0003), and learning of response inhibition was lost. The 44 patients who did not experience OHE did not have deteriorations in cognitive function in serial testing.

CONCLUSIONS

In cirrhosis, episodes of OHE are associated with persistent and cumulative deficits in working memory, response inhibition, and learning.

Persistence of cognitive impairment after resolution of overt hepatic encephalopathy

BACKGROUND & AIMS

In patients with cirrhosis, hepatic encephalopathy (HE) has acute but reversible as well as chronic components. We investigated the extent of residual cognitive impairment following clinical resolution of overt HE (OHE).

METHODS

Cognitive function of cirrhotic patients was evaluated using psychometric tests (digit symbol, block design, and number connection [NCT-A and B]) and the inhibitory control test (ICT). Improvement (reduction) in ICT lures and first minus second halves (DeltaL(1-2)) were used to determine learning of response inhibition. Two cross-sectional studies (A and B) compared data from stable cirrhotic patients with or without prior OHE. We then prospectively assessed cognitive performance, before and after the first episode of OHE.

RESULTS

In study A (226 cirrhotic patients), 54 had experienced OHE, 120 had minimal HE, and 52 with no minimal HE. Despite normal mental status on lactulose after OHE, cirrhotic patients were cognitively impaired, based on results from all tests. Learning of response inhibition (DeltaL(1-2) > or =1) was evident in patients with minimal HE and no minimal HE but was lost after OHE. In study B (50 additional patients who developed > or =1 documented OHE episode during follow-up), the number of OHE hospitalizations correlated with severity of residual impairment, indicated by ICT lures (r = 0.5, P = .0001), digit symbol test (r = -0.39, P = .002), and number connection test-B (r = 0.33, P = .04). In the prospective study (59 cirrhotic patients without OHE), 15 developed OHE; ICT lure response worsened significantly after OHE (12 before vs 18 after, P = .0003), and learning of response inhibition was lost. The 44 patients who did not experience OHE did not have deteriorations in cognitive function in serial testing.

CONCLUSIONS

In cirrhosis, episodes of OHE are associated with persistent and cumulative deficits in working memory, response inhibition, and learning.

The expression of nNOS, iNOS and nitrotyrosine is increased in the rat cerebral cortex in experimental hepatic encephalopathy

The changes in the distribution and amount of nitric oxide (NO) synthases (nNOS and iNOS) and the appearance of nitrotyrosine (NT) in the rat cerebral cortex were investigated following portacaval anastomosis (PCA), an experimental hepatic encephalopathy (HE) model. One month after PCA, rats showed more neurones immunoreactive to nNOS than did control animals. At 6 months post PCA, the number of neurones expressing nNOS had again increased and the intensity of the immunoreactions was stronger. Immunohistochemical analysis also showed that iNOS was increasingly expressed in pyramidal-like cortical neurones and in perivascular astrocytes from 1 to 6 months post PCA. In addition, a significant increase in cerebral iNOS concentration, at both post-PCA periods, was determined by Western blotting. The iNOS induction appears to be correlated with the length of the post-PCA period. PCA also induced the expression of NT, a nitration product of peroxynitrite. NT immunoreactivity was found in pyramidal-like cortical neurones. At 6 months, NT immunoreactivity was also evident in perivascular astrocytes, which was concomitant with a significant increase in NT protein level. PCA therefore not only increases the expression of nNOS but also induces the expression of iNOS and NT in both neurones and astrocytes. Taken together, these findings indicate that the induction of iNOS in pyramidal neurones and cortical astrocytes 6 months after PCA contributes to the generation of NT, and demonstrate the clear participation of NO in the pathogenic process of HE in this model.

Neuronal and inducible nitric oxide synthase expression in the rat cerebellum following portacaval anastomosis

In order to determine the role of neuronal nitric oxide synthase (nNOS) and inducible nitric oxide synthase (iNOS) in the pathogenesis of experimental hepatic encephalopathy (HE), the expression of both was analyzed in the cerebellum of rats 1 month and 6 months after performing portacaval anastomosis (PCA). In control cerebella, nNOS immunoreactivity was mainly observed in the molecular layer (ML), whereas the Purkinje cells did not express nNOS. However, nNOS expression was detected in the Purkinje cells at 1 month after PCA, correlating with a decrease in nNOS expression in the ML--part of an overall reduction in cerebellar nNOS concentrations (as determined by Western blotting). At 6 months post-PCA, a significant increase in nNOS expression was observed in the ML, as well as increased nNOS immunoreactivity in the Purkinje cells. nNOS immunoreactivity was also observed in the Bergmann glial cells of PCA-treated rats. While no immunoreactivity for iNOS was seen in the cerebella of control rats, iNOS immunoreactivity was significantly induced in the cerebellum 1 month after PCA. In addition, the expression of iNOS was greater at 6 months than at 1 month post-PCA. Immunohistochemical analysis revealed this iNOS to be localized in the Purkinje cells and Bergmann glial cells. The induction of iNOS in astroglial cells has been associated with pathological conditions. Therefore, the iNOS expression observed in the Bergmann glial cells might play a role in the pathogenesis of HE, the harmful effects of PCA being caused by them via the production of excess nitric oxide. These results show that nNOS and iNOS are produced in the Purkinje cells and Bergmann glial cells following PCA, implicating nitric oxide in the pathology of HE.

Down-regulation of astroglial proteins in the rat cerebellum after portacaval anastomosis

The effect of short-term portacaval anastomosis (PCA) on the expression of specific astroglial markers [glial fibrillary acidic protein (GFAP) and glutamine synthetase (GS)] in the rat cerebellum was examined to determine the influences of PCA on astroglial cells. The results suggest that PCA directly interferes with astroglial cytoskeleton, as indicated by the irregular distribution and reduced expression of GFAP observed after 1 month. PCA also decreased GS immunoreactivity in the Bergmann glial processes of the molecular layer, as well as in astrocytes of the granule cell layer. It might also modulate glutamatergic nervous activity as GS expression was reduced in 1 month post-PCA brains. Moreover, the GFAP and GS levels in PCA-exposed rats were lower than in control rats. This might contribute to the appearance of encephalopathy by increasing extracellular glutamate and/or ammonia concentrations. These results show that short-term PCA interferes with astroglial protein expression, with both GFAP and GS levels falling in astroglial cells.

Cell-selective effects of ammonia on glutamate transporter and receptor function in the mammalian brain

Increased brain ammonia concentrations are a hallmark feature of several neurological disorders including congenital urea cycle disorders, Reye's syndrome and hepatic encephalopathy (HE) associated with liver failure. Over the last decade, increasing evidence suggests that hyperammonemia leads to alterations in the glutamatergic neurotransmitter system. Studies utilizing in vivo and in vitro models of hyperammonemia reveal significant changes in brain glutamate levels, glutamate uptake and glutamate receptor function. Extracellular brain glutamate levels are consistently increased in rat models of acute liver failure. Furthermore, glutamate transport studies in both cultured neurons and astrocytes demonstrate a significant suppression in the high affinity uptake of glutamate following exposure to ammonia. Reductions in NMDA and non-NMDA glutamate receptor sites in animal models of acute liver failure suggest a compensatory decrease in receptor levels in the wake of rising extracellular levels of glutamate. Ammonia exposure also has significant effects on metabotropic glutamate receptor activation with implications, although less clear, that may relate to the brain edema and seizures associated with clinical hyperammonemic pathologies. Therapeutic measures aimed at these targets could result in effective measures for the prevention of CNS consequences in hyperammonemic syndromes.

Effects of hyperammonemia and liver failure on glutamatergic neurotransmission

Glutamate is the main excitatory neurotransmitter in mammals. Glutamatergic neurotransmission involves several steps, beginning with release of glutamate from the presynaptic neuron. Glutamate in the extracellular space activates glutamate receptors present in the synaptic membranes, leading to activation of signal transduction pathways associated with these receptors. To avoid continuous activation of glutamate receptors, glutamate is removed from the synaptic cleft by specific glutamate transporters located mainly on astrocytes. All these steps are tightly modulated under physiological conditions, and alterations of any of the above steps may result in impairment of glutamatergic neurotransmission, leading to neurological alterations. There are studies in the literature reporting alterations in all these steps in hyperammonemia and/or hepatic failure. Glutamatergic neurotransmission modulates important cerebral processes. Some of these processes are altered in patients with liver disease and hepatic encephalopathy, who show altered sleep-wake patterns, neuromuscular coordination, and decreased intellectual capacity. The alterations in glutamatergic neurotransmission may be responsible for some of these neurological alterations found in hepatic encephalopathy. The effects of hyperammonemia and liver failure on different steps of glutamatergic neurotransmission including alterations of glutamate concentration in the extracellular fluid in brain, transport and transporters of glutamate, the content and function of different types of glutamate receptors and signal transduction pathways. Alterations induced by hyperammonemia and liver failure on the glutamate-nitric oxide-cGMP pathway in brain may result in changes in long-term potetiation and learning ability.

Glutamine synthetase in brain: effect of ammonia

Glutamine synthetase (GS) in brain is located mainly in astrocytes. One of the primary roles of astrocytes is to protect neurons against excitotoxicity by taking up excess ammonia and glutamate and converting it into glutamine via the enzyme GS. Changes in GS expression may reflect changes in astroglial function, which can affect neuronal functions. Hyperammonemia is an important factor responsible of hepatic encephalopathy (HE) and causes astroglial swelling. Hyperammonemia can be experimentally induced and an adaptive astroglial response to high levels of ammonia and glutamate seems to occur in long-term studies. In hyperammonemic states, astroglial cells can experience morphological changes that may alter different astrocyte functions, such as protein synthesis or neurotransmitters uptake. One of the observed changes is the increase in the GS expression in astrocytes located in glutamatergic areas. The induction of GS expression in these specific areas would balance the increased ammonia and glutamate uptake and protect against neuronal degeneration, whereas, decrease of GS expression in non-glutamatergic areas could disrupt the neuron-glial metabolic interactions as a consequence of hyperammonemia. Induction of GS has been described in astrocytes in response to the action of glutamate on active glutamate receptors. The over-stimulation of glutamate receptors may also favour nitric oxide (NO) formation by activation of NO synthase (NOS), and NO has been implicated in the pathogenesis of several CNS diseases. Hyperammonemia could induce the formation of inducible NOS in astroglial cells, with the consequent NO formation, deactivation of GS and dawn-regulation of glutamate uptake. However, in glutamatergic areas, the distribution of both glial glutamate receptors and glial glutamate transporters parallels the GS location, suggesting a functional coupling between glutamate uptake and degradation by glutamate transporters and GS to attenuate brain injury in these areas. In hyperammonemia, the astroglial cells located in proximity to blood-vessels in glutamatergic areas show increased GS protein content in their perivascular processes. Since ammonia freely crosses the blood-brain barrier (BBB) and astrocytes are responsible for maintaining the BBB, the presence of GS in the perivascular processes could produce a rapid glutamine synthesis to be released into blood. It could, therefore, prevent the entry of high amounts of ammonia from circulation to attenuate neurotoxicity. The changes in the distribution of this critical enzyme suggests that the glutamate-glutamine cycle may be differentially impaired in hyperammonemic states.

Effect of acute exposure to ammonia on glutamate transport in glial cells isolated from the salamander retina

A rise of brain ammonia level, as occurs in liver failure, initially increases glutamate accumulation in neurons and glial cells. We investigated the effect of acute exposure to ammonia on glutamate transporter currents in whole cell clamped glial cells from the salamander retina. Ammonia potentiated the current evoked by a saturating concentration of L-glutamate, and decreased the apparent affinity of the transporter for glutamate. The potentiation had a Michaelis-Menten dependence on ammonia concentration, with a K(m) of 1.4 mM and a maximum potentiation of 31%. Ammonia also potentiated the transporter current produced by D-aspartate. Potentiation of the glutamate transport current was seen even with glutamine synthetase inhibited, so ammonia does not act by speeding glutamine synthesis, contrary to a suggestion in the literature. The potentiation was unchanged in the absence of Cl(-) ions, showing that it is not an effect on the anion current gated by the glutamate transporter. Ammonium ions were unable to substitute for Na+ in driving glutamate transport. Although they can partially substitute for K+ at the cation counter-transport site of the transporter, their occupancy of these sites would produce a potentiation of < 1%. Ammonium, and the weak bases methylamine and trimethylamine, increased the intracellular pH by similar amounts, and intracellular alkalinization is known to increase glutamate uptake. Methylamine and trimethylamine potentiated the uptake current by the amount expected from the known pH dependence of uptake, but ammonia gave a potentiation that was larger than could be explained by the pH change, and some potentiation of uptake by ammonia was still seen when the internal pH was 8.8, at which pH further alkalinization does not increase uptake. These data suggest that ammonia speeds glutamate uptake both by increasing cytoplasmic pH and by a separate effect on the glutamate transporter. Approximately two-thirds of the speeding is due to the pH change.

Preclinical models of hepatic encephalopathy

Hepatic encephalopathy is a multifactorial neuropsychiatric syndrome accompanying acute or chronic liver failure. Techniques for developing animal models of hepatic encephalopathy associated with acute or chronic liver failure, or vascular shunting are illustrated. In addition, the behavioral and biochemical characteristics of these models are described.

A quantitative evaluation of the permeability of the blood brain barrier of portacaval shunted rats

The integrity of the blood-brain barrier (BBB) was measured in male Sprague Dawley rats subjected to 16 weeks of portacaval shunting (PCS), the optimal time required for the cerebral changes to develop, by using an in situ brain perfusion technique. The penetration of a vascular space marker 14C mannitol, and labelled amino acids 3H-phenylalanine or 3H-glutamate were measured in brain and cerebrospinal fluid (CSF) using an in situ brain perfusion technique, over 2 or 20 minutes. The patency of the surgical shunt was confirmed by measurement of significantly increased plasma ammonia (131.5 +/- 14.8 micromol x l(-1)) and AST (159.5 +/- 19.9 IU x l(-1)) concentrations compared to controls 39.9 +/- 3.7*, and 82.5 +/- 6.6* respectively. Brain and CSF 14C-mannitol space (ml x 100g(-1)), was not increased by PCS where brain space was 1.31 +/- 0.27 mL x 100g(-1) compared to control 1.19 +/- 0.49 mL x 100g(-1), and CSF was 0.14 +/- 0.06 mL x 100g(-1) compared to control 0.15 +/- 0.05 (PCS n=10, control n=8). The uptake for 3H-glutamate, which is required for cerebral ammonia detoxification, was also unchanged in both brain and CSF. However, brain uptake of 3H-phenylalanine was significantly reduced from 871 +/- 80 microL x min(-1) x g(-1) to 356 +/- 154* microl x min(-1) x g(-1) (n=4), although there was no change in CSF uptake. These data suggest that there is no generalized breakdown of the blood-brain or blood-CSF barriers during PCS as assessed by mannitol penetration. The reduction in phenylalanine uptake into the brain may help stabilize high cerebral aromatic amino acid levels. *P<0.05, Two-tailed, Student's unpaired t-test.

Modulation of glutamate transporters (GLAST, GLT-1 and EAAC1) in the rat cerebellum following portocaval anastomosis

Glutamate transporters have the important function of removing glutamate released from synapses and keeping extracellular glutamate concentrations below excitotoxic levels. Extracellular glutamate increases in portocaval anastomosis (PCA), so we used a portacaval anastomosis model in rats to analyze the expression of glutamate transporters (GLAST, GLT-1 and EAAC1) in rat cerebellum, 1 and 6 months after PCA, using immunohistochemical methods. In controls, EAAC1 immunoreactivity in Purkinje cells and glial GLAST and GLT-1 immunoreactivities in the molecular layer (ML) increased from young to old rats. One month after PCA, Purkinje cell bodies were not immunostained for neuronal EAAC1 glutamate transporter, whereas glial glutamate transporter expressions (GLAST and GLT-1) were decreased when compared to young controls. In rats with long-term PCA (6 months post-PCA), neuronal and glial glutamate transporter expressions were increased. The expression of the neuronal glutamate transporter EAAC1 was less intense than old controls, whereas glial glutamate transporters (GLAST and GLT-1) increased more than their controls. Since the level of the neuronal glutamate transporter (EAAC1) in long-term PCA did not reach that of the controls, GLAST and GLT-1 glutamate transporters seemed to be required to ensure the glutamate uptake in this type of encephalopathy. EAAC1 immunoreactivity also was expressed by Bergmann glial processes in long-term PCA, but this increase did not suffice to reverse the alterations caused at the early stage. The present findings provide evidence that transitory alteration of glutamate transporter expressions could be a significant factor in the accumulation of excess glutamate in the extracellular space in PCA, which probably makes Purkinje cells more vulnerable to glutamate effect.

Effects of hyperammonaemia on brain function

Neuropsychiatric symptoms of hyperammonaemia include alterations of mood and personality, cognitive impairment, ataxia, convulsions and coma. The nature and severity of CNS dysfunction depend upon the aetiology and degree of hyperammonaemia, its acuteness of onset and the age of the patient. Neuropathological studies reveal Alzheimer type II astrocytosis in the adult hyperammonaemic patient, whereas hyperammonaemia in the infant resulting from congenital urea cycle disorders or Reye syndrome is accompanied by cerebral atrophy, neuronal loss and cerebral oedema. Several electrophysiological and biochemical mechanisms have been proposed to explain the deleterious effects of ammonia on CNS function. Such mechanisms include direct effects of the ammonium ion on excitatory and inhibitory neurotransmission and a deficit in cerebral energy metabolism due to ammonia-induced inhibition of alpha-ketoglutarate dehydrogenase. In addition, ammonia has been shown to interfere with normal processes of uptake, storage and release of various neurotransmitters. Ammonia disrupts monoamine storage, inhibits the high-affinity uptake of glutamate by both astrocytic and neuronal elements and activates 'peripheral-type' benzodiazepine receptors leading to the potential synthesis of neuroactive steroids in brain. On the basis of these actions, it has been proposed that ammonia disrupts neuron-astrocyte trafficking of amino acids and monoamines in brain. The increased formation of brain glutamine in hyperammonaemic syndromes could be responsible for the phenomenon of brain oedema in these disorders. Therapies aimed at either decreasing ammonia production in the gastrointestinal tract or increasing ammonia removal by liver or skeletal muscle are the mainstay in the prevention and treatment of the CNS consequences of hyperammonaemia. New therapeutic approaches aimed at correction of the neurotransmitter and cerebral energy deficits in these syndromes could hold promise for the future.

Modulation of AMPA receptor subunits GluR1 and GluR2/3 in the rat cerebellum in an experimental hepatic encephalopathy model

The immunohistochemical expression and distribution of the AMPA-selective receptor subunits GluR1 and GluR2/3 were investigated in the rat cerebellum following portocaval anastomosis (PCA) at 1 and 6 months. With respect to controls, GluR1 and GluR2/3 immunoreactivities increased over 1 to 6 months following PCA, although immunolabelling patterns for both antibodies were different at the two analysed times. GluR1 immunoreactivity was expressed by Bergmann glial cells, which showed immunoreactive glial processes crossing the molecular layer at 6 months following PCA. The GluR2/3 subunit was expressed by Purkinje neurons and moderately expressed by neurons of the granule cell layer. Immunoreactivity for GluR2/3 was detectable in cell bodies and dendrites of Purkinje cells in young control cerebella, whereas GluR2/3 immunoreactivity was scarce 1 month post PCA. However, despite a lack of immunoreactivity in the Purkinje somata and main processes of adult control rats, GluR2/3 immunoreactivity was strongly enhanced in Purkinje neurons following long-term PCA. These findings suggest that the localization of the GluR2/3 subunit in Purkinje cells undergoes an alteration and/or reorganization as a consequence of long-term PCA. The combination of enhanced GluR immunoreactivity in long-term PCA, both in Bergmann glial cells and in Purkinje neurons, suggests some degree of neuro-glial interaction, possibly through glutamate receptors, in this type of encephalopathy.

Increased neuronal nitric oxide synthase expression in brain following portacaval anastomosis

It has previously been suggested that increases of L-arginine uptake into brain following portacaval shunting may result in increased activities of constitutive neuronal nitric oxide synthase (nNOS). In order to further address this issue, nNOS protein and gene expression were studied by Western blot analysis using a monoclonal nNOS antibody and RT-PCR respectively in the brains of rats following portacaval shunting or sham operation. Portacaval shunting resulted in a 2-fold increase (P < 0.01) in nNOS protein and a concomitant 2.4-fold increase (P < 0.01) in nNOS mRNA. Increased nNOS activity in brain and the resulting increase in nitric oxide production could contribute to the increased cerebral blood flow and to the pathogenesis of hepatic encephalopathy in chronic liver disease.

New aspects on etiology, biochemistry, and therapy of portal systemic encephalopathy: a critical survey

There is scientific agreement that portal systemic encephalopathy (PSE) is caused morphologically by portal systemic shunts and biochemically by constituents of the portal venous blood. Ammonium has a key role in the pathogenesis of PSE. Direct correlations with the degree of PSE have been established exclusively with glutamine, i.e. the terminal product of the peripheral detoxification of ammonium. In PSE, ammonium is probably responsible for damage to astrocytic and neuronal cells. Ammonium's toxic effect is due to the intracerebral glutamine synthesis. After several metabolic steps, which will be discussed in detail, brain cell damage is caused directly or indirectly (exitotoxically) by energy deficiency. Hyperammonemia and PSE are each well defined though different forms of disturbance. Therefore, ammonium is not the sole decisive factor in the pathogenesis of PSE. We performed a detailed and critical analysis of all studies on amino acid therapy of PSE, especially those that were randomized and controlled. This analysis revealed a close and direct correlation between qualitative and quantitative dosages of amino acids on one hand, and parallel improvements of amino acid imbalance (essentially associated with PSE) and degree of PSE on the other. A close and direct dose/efficacy correlation must be assumed. Disturbed plasmatic amino acid homeostasis and cerebral monoaminergic neurotransmission are probably important pathogenic factors of PSE. A fundamental cofactor in the efficacy of each adequate amino acid therapy might be a substantial decrease of endogenous ammonium production. Physiologic benzodiazepines may also have an important function in the pathogenesis of PSE: not so, however, the glutamate-ergic and GABA-ergic neurotransmission, which are disturbed principally in PSE. In close correlation to pathogenesis, established and proposed therapies of PSE are critically discussed.

Long-term changes in glial fibrillary acidic protein and glutamine synthetase immunoreactivities in the supraoptic nucleus of portacaval shunted rats

The present study was undertaken to ascertain whether, and to what extent, glial fibrillary acidic protein (GFAP) and glutamine synthetase (GS) expressions in the supraoptic nucleus (SON) could be modulated after one month and six months of portacaval shunting (PCS) in rats. GFAP and GS immunoreactivities were significantly higher in PCS rats than in control rats at one and six months. The increased GFAP and GS immunoreactivities observed in the SON astrocytes were directly related to the duration of PCS. In PCS rats, the number and length of both GFAP and GS immunopositive astroglial processes increased not only in the hypothalamic nucleus but in the perinuclear zone, where glutamatergic pathways have been described, whereas GFAP and GS expressions decreased in the ventral glial lamina. Since GS is one of the glutamate metabolizing enzymes and the SON is one of the areas of glutamatergic activity, our results show that astrocytes respond differentially to glutamate toxicity. This suggests that overexpression of GFAP and GS immunoreactivities could be associated with glutamatergic neurotransmission disorders.

Neuroactive amino acids in hepatic encephalopathy

There is abundant evidence to suggest that alterations of excitatory and inhibitory amino acids play a significant role in the pathogenesis of hepatic encephalopathy (HE) in both acute and chronic liver diseases. Brain glutamate concentrations are reduced in patients who died in hepatic coma as well as in experimental HE, astrocytic reuptake of glutamate is compromised in liver failure and postsynaptic glutamate receptors (both NMDA and non-NMDA subclasses) are concomitantly reduced in density. Recent studies in experimental acute liver failure suggest reduced capacity of the astrocytic glutamate transporter in this condition. Together, this data suggests that neuron-astrocytic trafficking of glutamate is impared in HE. Other significant alterations of neuroactive amino acids in HE include a loss of taurine from brain cells to extracellular space, a phenomenon which could relate both to HE and to brain edema in acute liver failure. Increased concentrations of benzodiazepine-like compounds have been reported in human and experimental HE. Clinical trials with the benzodiazepine antagonist flumazenil reveal a beneficial effect in some patients with HE; the mechanism responsible for this effect, however, remains to be determined.

Ammonia and glutamine metabolism during liver insufficiency: the role of kidney and brain in interorgan nitrogen exchange

BACKGROUND

During liver failure, urea synthesis capacity is impaired. In this situation the most important alternative pathway for ammonia detoxification is the formation of glutamine from ammonia and glutamate. Information is lacking about the quantitative and qualitative role of kidney and brain in ammonia detoxification during liver failure.

METHODS

This review is based on own experiments considered against literature data.

RESULTS AND CONCLUSIONS

Brain detoxifies ammonia during liver failure by ammonia uptake from the blood, glutamine synthesis and subsequent glutamine release into the blood. Although quantitatively unimportant, this may be qualitatively important, because it may influence metabolic and/or neurotransmitter glutamate concentrations. The kidney plays an important role in adaptation to hyperammonaemia by reversing the ratio of ammonia excreted in the urine versus ammonia released into the blood from 0.5 to 2. Thus, the kidney changes into an organ that netto removes ammonia from the body as opposed to the normal situation in which it adds ammonia to the body pools.

Pathophysiology of alcoholic brain damage: synergistic effects of ethanol, thiamine deficiency and alcoholic liver disease

Chronic alcoholism results in brain damage and dysfunction leading to a constellation of neuropsychiatric symptoms including cognitive dysfunction, the Wernicke-Korsakoff Syndrome, alcoholic cerebellar degeneration and alcoholic dementia. That these clinically-defined entities result from independent pathophysiologic mechanisms is unlikely. Alcohol and its metabolite acetaldehyde are directly neurotoxic. Alcoholics are thiamine deficient as a result of poor diet, gastrointestinal disorders and liver disease. In addition, both alcohol and acetaldehyde have direct toxic effects on thiamine-related enzymes in liver and brain. Alcoholics frequently develop severe liver disease and liver disease per se results in altered thiamine homeostasis, in cognitive dysfunction and in neuropathologic damage to astrocytes. The latter may result in the loss of neuron-astrocytic trafficking of neuroactive amino acids and thiamine esters, essential to CNS function. The present review article proposes mechanisms whereby the effects of alcohol, thiamine deficiency and liver disease combine synergistically to contribute to the phenomena of cognitive dysfunction and "alcoholic brain damage".

Choline acetyltransferase and acetylcholinesterase activities are unchanged in brain in human and experimental portal-systemic encephalopathy

Activities of choline acetyltransferase, acetylcholinesterase and butyrylcholinesterase were studied in the frontal cortex, temporal cortex, cerebellum and caudate nucleus obtained at autopsy from eight alcoholic cirrhotic patients who died in hepatic coma and from an equal number of age-matched subjects free from hepatic, neurological or psychiatric disorders. Activities of these enzymes were unaltered in the brains of cirrhotics compared to controls. Choline acetyltransferase and cholinesterase activities were also studied in the cerebral cortex, cerebellum, brain stem and striatum of rats four weeks following portacaval anastomosis and their sham-operated controls. Portacaval-shunting did not cause any statistically significant differences in the activities of choline acetyltransferase, acetyl or butyrylcholinesterases. These results argue against a presynaptic cholinergic lesion in human and experimental portal-systemic encephalopathy.

Ammonia added in vitro, but not moderate hyperammonemia in vivo, stimulates glutamate uptake and H(+)-ATPase activity in synaptic vesicles of the rat brain

The uptake of radiolabelled neurotransmitters: glutamate (GLU), GABA, and dopamine (DA) and the activity of the vacuolar type H(+)-pumping ATPase (H(+)-ATPase), were measured in crude synaptic vesicles treated in vitro with a neurotoxic (3 mM) dose of NH4+ (acetate or chloride), or isolated from rats with a moderate increase of brain ammonia (to approximately 0.6 mM) induced by i.p. administration of ammonium acetate (HA rats) or a hepatotoxin-thioacetamide (HE rats). In vitro treatment with ammonium salts increased the sodium-independent, chloride-dependent uptake of GLU but did not stimulate the uptake of GABA or DA. The in vitro treatment also stimulated the H(+)-ATPase activity. Since H(+)-ATPase generates the electrochemical gradient driving synaptic vesicular neurotransmitter transport, its stimulation by ammonia may have facilitated GLU uptake. However the GLU specificity of the effect must be related to other factors differentially affecting GLU uptake and the uptake of other neurotransmitters. Enhanced GLU accumulation in the synaptic vesicles may contribute to the increase of synaptic GLU exocytosis previously reported to accompany acute increases of brain ammonia to toxic levels. However, GLU uptake and H(+)-ATPase activity, but also the uptake of GABA and DA, were unchanged in synaptic vesicles prepared from rats with HA or HE. This indicates that changes in GLU and/or GABA release reported for moderate hyperammonemic conditions must be elicited by factors unrelated to the synaptic vesicular transport of the amino acids.

Molecular mechanism of acute ammonia toxicity and of its prevention by L-carnitine

In summary, we propose that acute ammonia intoxication leads to increased extracellular concentration of glutamate in brain and results in activation of the NMDA receptor. Activation of this receptor mediates ATP depletion and ammonia toxicity since blocking the NMDA receptor with MK-801 prevents both phenomena. Ammonia-induced metabolic alterations (in glycogen, glucose, pyruvate, lactate, glutamine, glutamate, etc) are not prevented by MK-801 and, therefore, it seems that they do not play a direct role in ammonia-induced ATP depletion nor in the molecular mechanism of acute ammonia toxicity. The above results suggest that ammonia-induced ATP depletion is due to activation of Na+/K(+)-ATPase, which, in turn, is a consequence of decreased phosphorylation by protein kinase C. This can be due to decreased activity of PKC or to increased activity of a protein phosphatase. We also show that L-carnitine prevents glutamate toxicity in primary neuronal cultures. The results shown indicate that carnitine increases the affinity of glutamate for the quisqualate type (including metabotropic) of glutamate receptors. Also, blocking the metabotropic receptor with AP-3 prevents the protective effect of L-carnitine, indicating that activation of this receptor mediates the protective effect of carnitine. We suggest that the protective effect of carnitine against acute ammonia toxicity in animals is due to the protection against glutamate neurotoxicity according to the above mechanisms.

Effects of ammonium ions on synaptic transmission and on responses to quisqualate and N-methyl-D-aspartate in hippocampal CA1 pyramidal neurons in vitro

Effects of NH4Cl on CA1 pyramidal neurons and synaptic transmission were investigated with intracellular recording in fully submerged rat hippocampal slices. Superfusion with 1-4 mM NH4Cl reversibly depolarized the membrane by 15.1 +/- 1.4 mV, reduced the amplitude and broadened the duration of action potentials due to a slower rate of repolarization, without significant change in membrane conductance. When membrane potential was returned to control level by the injection of a steady outward current, action potential amplitude recovered but repolarization remained slow. The extent of depolarization was not dependent on the concentration of NH4Cl between 1 and 4 mM. NH4Cl greatly depressed orthodromic transmission evoked by the stimulation of Schaffer collateral/commissural fibers several minutes after depolarizing the CA1 neuron. Interruption of transmission began with a decrease in excitatory postsynaptic potential (EPSP) amplitude and eventually EPSPs were almost eliminated. When NH4Cl was removed, it took 2-3 min for membrane potential and 10-15 min for transmission to recover. Inward currents induced by bath application of quisqualate acting on alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors were also depressed. In contrast, NH4Cl enhanced N-methyl-D-aspartate (NMDA)-induced currents. This potentiation disappeared in the absence of added Mg2+. A reduction in quisqualate-induced responses provided a possible explanation for the inhibition of excitatory transmission by NH4Cl.

Region-selective reductions in activities of glutamine synthetase in rat brain following portacaval anastomosis

Portacaval anastomosis in the rat results in liver atrophy, sustained hyperammonemia and mild encephalopathy. Previous studies have demonstrated region-selective alterations of glutamine and other ammonia-related amino acids in brain following portacaval anastomosis. Ammonia removal by brain relies on glutamine synthesis and the enzyme responsible, glutamine synthetase, has an almost exclusively astrocytic localization. Glutamine synthetase activities were measured using a radioenzymatic assay in homogenates of seven brain regions of rats four weeks after end-to-side portacaval anastomosis. Enzyme activities were significantly reduced in hippocampus (by 25%, p < 0.01), in cerebellum (by 29%, p < 0.01) and in cerebral cortex (by 14%, p < 0.05). Enzyme activities in other brain regions were within normal limits. Region-selective reductions of glutamine synthetase following portacaval anastomosis could result in disruption of neuron-glial metabolic interactions and in a deficit of glutamatergic synaptic regulation. Similar mechanisms could be implicated in the pathogenesis of hepatic encephalopathy accompanying chronic liver disease in humans.

Transport and metabolism of glutamate by rat cerebellar mitochondria during ammonia toxicity

Pathophysiological concentrations of ammonia, both in vivo and in vitro, suppressed the oxidation of glutamate by rat cerebellar mitochondria. The transport of glutamate into mitochondria was either unaltered or enhanced during hyperammonemic states. Activities of mitochondrial enzymes, aspartate aminotransferase, alanine aminotransferase, glutamate dehydrogenase, glutaminase, and GABA-transaminase were suppressed during hyperammonemic states. Suppression of 14CO2 production with (aminooxy)acetic acid but not with glutamic acid diethyl ester indicated that transamination but not oxidative deamination of glutamate plays a major role in glutamate oxidation during normal and hyperammonemic states.

Cerebral cortex ammonia and glutamine metabolism in two rat models of chronic liver insufficiency-induced hyperammonemia: influence of pair-feeding

Enhanced cerebral cortex ammonia uptake, subsequent glutamine synthesis, and glutamine release into the bloodstream have been hypothesized to deplete cerebral cortex glutamate pools. We investigated this hypothesis in rats with chronic liver insufficiency-induced hyperammonemia and in pair-fed controls to rule out effects of differences in food intake. Cerebral cortex plasma flow and venous-arterial concentration differences of ammonia and amino acids, as well as cerebral cortex tissue concentrations, were studied 7 and 14 days after surgery in portacaval-shunted/bile duct-ligated, portacaval-shunted, and sham-operated rats, while the latter two were pair-fed to the first group, and in normal unoperated ad libitum-fed control rats. At both time points, arterial ammonia was elevated in the chronic liver insufficiency groups and arterial glutamine was elevated in portacaval shunt/biliary obstruction rats compared to the other groups. In the chronic liver insufficiency groups net cerebral cortex ammonia uptake was observed at both time points and was accompanied by net glutamine release. Also in these groups, cerebral cortex tissue glutamine, many other amino acid, and ammonia levels were elevated. Tissue glutamate levels were decreased to a similar level in all operated groups compared with normal unoperated rats, irrespective of plasma and tissue ammonia and glutamine levels. These results demonstrate that during chronic liver insufficiency-induced hyperammonemia, the rat cerebral cortex enhances net ammonia uptake and glutamine release. However, the decrease in tissue glutamate concentrations in these chronic liver insufficiency models seems to be related primarily to nutritional status and/or surgical trauma.

Cerebral cortex ammonia and glutamine metabolism during liver insufficiency-induced hyperammonemia in the rat

Hyperammonemia has been suggested to induce enhanced cerebral cortex ammonia uptake, subsequent glutamine synthesis and accumulation, and finally net glutamine release into the blood stream, but this has never been confirmed in liver insufficiency models. Therefore, cerebral cortex ammonia- and glutamine-related metabolism was studied during liver insufficiency-induced hyperammonemia by measuring plasma flow and venous-arterial concentration differences of ammonia and amino acids across the cerebral cortex (enabling estimation of net metabolite exchange), 1 day after portacaval shunting and 2, 4, and 6 h after hepatic artery ligation (or in controls). The intra-organ effects were investigated by measuring cerebral cortex tissue ammonia and amino acids 6 h after liver ischemia induction or in controls. Arterial ammonia and glutamine increased in portacaval-shunted rats versus controls, and further increased during liver ischemia. Cerebral cortex net ammonia uptake, observed in portacaval-shunted rats, increased progressively during liver ischemia, but net glutamine release was only observed after 6 h of liver ischemia. Cerebral cortex tissue glutamine, gamma-aminobutyric acid, most other amino acids, and ammonia levels were increased during liver ischemia. Glutamate was equally decreased in portacaval-shunted and liver-ischemia rats. The observed net cerebral cortex ammonia uptake, cerebral cortex tissue ammonia and glutamine accumulation, and finally glutamine release into the blood suggest that the rat cerebral cortex initially contributes to net ammonia removal from the blood during liver insufficiency-induced hyperammonemia by augmenting tissue glutamine and ammonia pools, and later by net glutamine release into the blood. The changes in cerebral cortex glutamate and gamma-aminobutyric acid could be related to altered ammonia metabolism.

Changes in brain ECF amino acids in rats with experimentally induced hyperammonemia

Using microdialysis, we studied brain extracellular fluid (ECF) amino acid metabolism in rats with experimentally induced hyperammonemia and regional elevation of brain ECF ammonia levels. The total brain ECF amino acid level was increased by an elevation of the blood ammonia level. Hyperammonemia elevated brain ECF aromatic amino acids and reduced arterial blood branched chain amino acids. When rats with hyperammonemia were intravenously administered norleucine, the brain ECF norleucine level rose markedly, suggesting increased permeability of the blood-brain barrier. When rats with hyperammonemia were infused with a branched chain amino acid-rich preparation, the elevated brain ECF aromatic amino acids level was not suppressed. Following local intracerebral ammonia infusion, only glutamate levels showed a marked elevation. These results suggest that impairment of the blood-brain barrier related to hyperammonemia increases the inflow of low molecular weight substances including amino acids. Furthermore, the ammonia-induced increase of glutamate in the cerebral ECF suggests that high ammonia levels may increase the excitability of the brain. Thus, ammonia may serve as a key factor in the onset of hepatic encephalopathy.

Ammonia-induced alterations in the metabolism of glutamate and aspartate in neuronal perikarya and synaptosomes of rat cerebellum

The effect of subacute and acute doses of ammonium acetate was studied on the production of 14CO2 from 14C-labeled glutamate and aspartate by neuronal perikarya and synaptosomes isolated from rat cerebellum. Studies with inhibitors for aminotransferases (aminooxy acetic acid) and glutamate dehydrogenase (glutamic acid diethyl ester) indicated that transamination reactions play a major role in this process. There was a suppression in this process in hyperammonemic states. Activities of the enzymes, aspartate aminotransferase, alanine aminotransferase, glutamate dehydrogenase and glutaminase were decreased in both preparations in hyperammonemic states. Activity of glutamine synthetase was unaltered.

Decreased potassium-stimulated release of [3H]D-aspartate from hippocampal slices distinguishes encephalopathy related to acute liver failure from that induced by simple hyperammonemia

The calcium-dependent, high (65 mM) potassium-evoked release of the L-glutamate analogue [3H]D-aspartate (D-Asp) was measured in hippocampal slices derived from rats with (a) hepatic encephalopathy (HE) induced with a hepatotoxin, thioacetamide, (b) hyperammonemia produced by i.p. administration of ammonium acetate, and (c) in normal slices preincubated for 30 min with 1 mM ammonium acetate. HE (variant a) inhibited the release by about 30%, which was interpreted to indicate depressed exocytosis of synaptic glutamate. This phenomenon is likely to lead to a decrease of glutamate-mediated neural excitation, which in turn could contribute to the neural inhibition typical of HE. By contrast, and in agreement with earlier reports, hyperammonemia (variant b) did not affect D-Asp release, whereas in vitro treatment of the slices with ammonium acetate (variant c) resulted in a 60% increase of the release. Hence, impairment of synaptic glutamate exocytosis is the phenomenon that distinguishes HE related to toxic liver failure from simple hyperammonemia. This result emphasizes the role of other factors than ammonia in the pathophysiological mechanism of HE.

Aspartate aminotransferase, malate dehydrogenase, and pyruvate carboxylase activities in rat cerebral synaptic and nonsynaptic mitochondria: effects of in vitro treatment with ammonia, hyperammonemia and hepatic encephalopathy

The effects of in vitro treatment with ammonium chloride, hepatic encephalopathy (HE) due to thioacetamide (TAA) induced liver failure and chronic hyperammonemia produced by i.p. administration of ammonium acetate on the activity of the two malate-aspartate shuttle enzymes: aspartate aminotransferase (AAT), malate dehydrogenase (MDH), and on the pyruvate carboxylase (PC) activity were examined in synaptic and nonsynaptic mitochondria from rat brain. With regard to the shuttle enzymes the response to ammonium ions in vitro (3mM NH4Cl) was observed in nonsynaptic mitochondria only, and was manifested by a 27% decrease of AAT activity and a 16% decrease in MDH activity. By contrast, both in vivo conditions primarily affected the synaptic mitochondrial enzymes: TAA-induced HE produced a 26% decrease of synaptic mitochondrial AAT and a 50% decrease of synaptic mitochondrial MDH. Hyperammonemia inhibited synaptic mitochondrial AAT by 30% and synaptic mitochondrial MDH by 45%. HE produced no effect at all in nonsynaptic mitochondria while hyperammonemia produced a 30% increase in the AAT activity, but no changes in MDH. All the experimental conditions affected the nonsynaptic mitochondria PC: ammonium chloride in vitro produced a 20% decrease, TAA-induced HE--a 30% decrease, whereas hyperammonemia inhibited the enzyme by 53%. The PC activity in synaptic mitochondria was very low (about 2% of that measured in nonsynaptic mitochondria), which is consistent with the primarily astrocytic localization of the enzyme.

Intracellular pH rises and astrocytes swell after portacaval anastomosis in rats

The basis for astrocytic swelling after the early period after portacaval anastomosis (PCA) is poorly defined. In other eukaryotic cells intracellular pH (pHi) and volume are determined, in part, by the same general mechanisms, yet how astrocytic pHi varies with enlargement of these cells after PCA is unknown. Therefore, direct measurements of pHi in astrocytes were made and compared with pericapillary astrocytic area as determined from electron micrographs in rats 5-8 days after PCA. Astrocytic area (n = 14 measurements for each group) was found to be significantly (P less than 0.0009) greater in PCA animals (n = 3) than in sham-operated control animals. (n = 3). Double-barrel pH-sensitive microelectrodes were used to measure pHi in neocortical cells defined by electrophysiological criteria to be astrocytic. Astrocytes (n = 25) from PCA animals (n = 5) had a resting membrane potential of 72 +/- 5 mV (mean +/- SD) and an pHi of 7.11 +/- 0.11 while comparable cells (n = 12) from sham-operated controls (n = 2) had a membrane potential of 81 +/- 6 mV and an pHi of 7.00 +/- 0.10. Astrocytes from PCA animals were significantly more depolarized (P less than 0.001) and alkaline (P less than 0.009), at a time when they were also significantly larger than those from sham-operated controls. Astrocytes are known to become more alkaline when they are activated by brief depolarizing stimuli. However, this is the first demonstration that such an interrelationship can also exist for steady-state conditions of these cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Cerebrospinal fluid amino acids in relation to neurological status in experimental portal-systemic encephalopathy

Using an indwelling cisterna magna catheter technique, serial CSF samples were analyzed for amino acid content in rats at various stages of portal-systemic encephalopathy resulting from ammonium acetate administration following portacaval anastomosis. Anastomosis alone resulted in increased CSF concentrations of glutamine, tyrosine, phenylalanine, glutamate and alanine. GABA levels, on the other hand were not significantly changed. Onset of severe neurological symptoms following ammonium acetate administration resulted in selectively increased CSF alanine. Other amino acids were not further increased at severe stages of encephalopathy. Increased CSF alanine probably results from increased glutamine transamination in the brains of portacaval shunted rats.