Académie royale de Médecine de Belgique


Texte Santiago Grisolia, correspondant étranger

(Séance du samedi 26 mai 1984)


par Santiago GRISOLIA (Instituto de investigaciones citológicas de la Caja de Ahorros de Valencia, Espagne), correspondant étranger, et J.E. O’CONNOR et M. Costell, collaborateurs.

Since long time it is known that ammonia is an extremely toxic substance which deleterious effects occur on the central nervous system.  Also, ammonia is a key intermediate in nitrogen metabolism and its levels in the organism are kept within narrow limits by homeostatic mechanisms which integrate the ammonia releasing reactions with the ammonia consuming processes.

A number of evidences indicate that ammonia is toxic to the brain.  Hyperammonemia accompanies coma in some inborn errors of the urea cycle. Reye’s syndrome and severe liver disease.  Acute hyperammonemia provokes convulsions and coma in experimental animals.  Subactue or chronic hyperammonemia induces the growth and proliferation of astrocites, the glia cells which contain great amounts of glutamic dehydrogenase and glutamine synthetase.  It has been postulated that this is the compartment involved in ammonia detoxication in brain (for a general review on hyperammonemic syndromes syndromes, see Walser, 1982 and Duffy and Plum, 1983).

The mechanism by which raised concentrations of ammonia in brain cause their effects is still unknown, as the large number of hypotheses attests.  A number of them postulate an ammonia-induced depletion in the brain energy stores.  Marked hyperam-monemia induces significant changes in the energy metabolism in brain.  Hepatic encephalopathy in man and animals is associated with a decrease in brain oxygen consumption.  Also, high concentrations of ammonia inhibit pyruvate oxidation in brain slices.  The ammonia intoxication in rats has been shown to decrease the levels of ATP and phosphocreatine in different regions of the brain.  The ammonia-induced coma in “porta-cava shunted rats” also induces a decrease in the energy stores in brain.  When such animals are administered moderate amounts of ammonium salts the oxygen consumption in brain decreases, lactate and pyruvate cumulate and the levels of ATP and phosphocreatine decrease.  Glutamate and aspartate fall and glutamine and alanine increase.  These changes suggest a block in the oxydation of pyruvate through the Krebs cycle and are consistent with an interference of ammonia on the “malate-aspartate shuttle”, the mechanism for introducing reduction équivalents into mitrochrondria from cytosol (for a review on the mechanism of ammonia toxicity see Conn and Lieberthal, 1979 and Duffy and Plum, 1983).

Recently we have found that ethanol protects mice against acute ammonia intoxication (O’Connor et al., 1982a, 1982b).  Our results were compatible with the postulation that ammonia impairs brain energy compatible with the postulation that ammonia impairs brain energy metabolism and they also support the action of ammonia on the “malate-aspartate shuttle” and possibly indicated that ethanol raised the levels of acetyl CoA.  The essential role of L-carnitine, as a carrier for the entry of fatty acids into mitochondria where they undergo beta-oxidation and yield acetyl Coa and reducing equivalents, has been extensively documented (Frenkel and Mc Garry, 1980).  Therefore, we thought that administration of L-carnitine, by facilitating the mitochondrial uptake of fatty acyl groups, would increase the intramitochondrial content of the acetyl CoA and the availaible metabolic energy, thus restoring the ATP levels decreased ed by ammonia.  We show here that L-carnitine protects mice against acute ammonia toxicity.


Male Swiss albino mice weighing 25 to 30 g, fed a standard diet, were used. Acute ammonia intoxication was produced by a single intraperitoneal injection of ammonium acetate (12 mmoles/kg body weight, as a 0,4 M solution)K  The animals developed clear symptoms of ammonia toxicity; hyperexcitability and drowsiness appeared first, followed by noise-induced seizures and coma.  Clonic and tonic convulsions preceded death which occurred in all cases ten to fifteen minutes after the injection.  In another group, the mice were injected intraperitoneally with increasing doses of L-carnitine (as 20 % W/V solution in saline) before receiving the ammonium acetate.

In a second set of experiments, mice were given 16 mmoles L-carnitine/kg body weight and then injection thirty minutes later with ammonium acetate in the following doses, 12 mmoles/kg (LD 100), 9,5 mmoles/kg (LD10) and 6 mmoles/kg (non-lethal dose).

The experimental details and described elsewhere (O’Connor et al., 1984).  

Protective effect of L-carnitine on acute ammnonia intoxication in mice


























The animals received the indicated doses of L-carnitine in a single intraperitoneal injection.  Thirty minutes later they were given intraperitoneally 12 mmoles ammonium acetate/kg body weight.

Results and discussion

Table I shows that administration of L-carnitine entirely prevented death in mice given a lethal dose of ammonium acetate.  Symptoms of ammonia toxicity such as convulsions were either abolished or significantly delayed.

The protective effect of L-carnitine was accompanied by marked decrease in blood and brain ammonia.  This decrease did not result in a concomitant increase of urea production over and above that seen in the animals given ammonia only.  It should be noted that in the absence of L-carnitine the animals died in a few minutes at the high doses of ammonia used.  However, with L-carnitine, long-term effects of high ammonia concentration on urea synthesis could be tested.  Thus, when lower doses of ammonium acetate were injected, as shown, it was clear that treatment with L-carnitine enhanced ureogenesis in ammonia-intoxicated mice.  As can be seen, blood urea rose in animals with or without L-carnitine, but this effect was more transient in animals given ammonia only.  Also, the stimulation of urea production by L-carnitine seems to be somewhat delayed.

An impairing action of ammonia in the transport of reducing equivalents into mitochondria has been proposed.  Since in previous work with ethanol which also protected against ammonia toxicity our results were consistent with this hypothesis and suggested that ethanol improved the function of the “malate-aspartate shuttle”, we estimated the changes in the redox state in brain cytosol and mitrochondria, by measuring the cytosol and mitochondrial NADH/NAD+ ration.  As seen in Table II, ammonium acetate raised this quotient in the cytosol, while decreasing it in the mitochondria.  L-carnitine treatment shifted these altered ratios towards normal values.

We have shown that L-carnitine is an excellent protective agent against acute ammonia intoxication in mice, reducing mortality and preventing the appearance of inherent manifestations of toxicity.

At the time of death, unprotected animals had essentially the same urea levels in blood as the animals given L-carnitine.  This rules out an early increase in urea synthesis as a major factor in the protection observed.  These results extend those obtained previously with ethanol, which protects against acute ammonia intoxication without enhancing urea synthesis.  Most important seems to be the increase in glutamate in brain seen in animals protected with L-carnitine which may reflect and enhanced ammonia utilization via glutamate dehydrogenase.  A fall in the level of glutamate, which appears to be the major excitatory neurotransmitter in brain, could result in neurological manifestations, especially coma, following ammonia intoxication.