Macrophage cytotoxicity: role for L-arginine deiminase and imino nitrogen oxidation to nitrite.

Previous studies have shown that cytotoxic activated macrophages cause inhibition of DNA synthesis, of mitochondrial respiration, and of aconitase activity in tumor target cells. An L-arginine-dependent biochemical pathway synthesizing L-citrulline and nitrite, coupled to an effector mechanism, is now shown to cause this pattern of metabolic inhibition. Murine cytotoxic activated macrophages synthesize L-citrulline and nitrite in the presence of L-arginine but not D-arginine. L-Citrulline and nitrite biosynthesis by cytotoxic activated macrophages is inhibited by NG-monomethyl-L-arginine, which also inhibits this cytotoxic effector mechanism. This activated macrophage cytotoxic effector system is associated with L-arginine deiminase activity, and the imino nitrogen removed from the guanido group of L-arginine by the deiminase reaction subsequently undergoes oxidation to nitrite. L-Homoarginine, an alternative substrate for this deiminase, is converted to L-homocitrulline with concurrent nitrite synthesis and similar biologic effects.

Ammonia production in muscle: the purine nucleotide cycle.

Experiments are reported which throw new light on the problem of ammonia production by muscle and, probably, by other tissues.

Ammonia production by isolated mouse proximal tubules perfused in vitro. Effect of metabolic acidosis.

We examined the effects of metabolic acidosis in vivo and reduced bath and luminal pH in vitro on total NH3 (NH3 + NH+4) production rates by isolated mouse proximal tubule segments. Midproximal tubule segments were obtained from mice with NH4Cl-induced metabolic acidosis and from nonacidotic controls. The segments were perfused with modified Krebs-Ringer bicarbonate (KRB) buffer, incubated in KRB buffer containing 0.5 mM L-glutamine and 1.0 mM sodium acetate, and gassed with 95% O2 and 5% CO2. Isolated unperfused and perfused proximal tubules from acidotic mice produced total NH3 at higher rates than corresponding tubules from nonacidotic mice. Perfusion of the tubular lumen stimulated total NH3 production by tubules from both acidotic and nonacidotic mice. In contrast, lowering the bath pH to 7.0 by lowering the HCO3- concentration increased total NH3 production rates by tubules from nonacidotic mice but not by tubules from acidotic mice. Reducing the HCO3- concentration of the bath buffer to 10 mM while maintaining a pH of 7.4 had no significant effect on total NH3 production by tubules from nonacidotic mice. Lowering the luminal fluid pH by reducing the perfusate HCO-3 from 25 mM to 10, 5, or 1.2 mM while maintaining a bath pH of 7.4 lowered collected luminal fluid pH but had no effect on total NH3 production by proximal tubules from nonacidotic mice. These observations demonstrated that metabolic acidosis in vivo stimulated total NH3 production in isolated mouse proximal tubule segments and that low peritubular pH and HCO-3 stimulated total NH3 production by proximal tubule segments from nonacidotic mice in vitro.

Regulation of renal cortex ammoniagenesis. I. Stimulation of renal cortex ammoniagenesis in vitro by plasma isolated from acutely acidotic rats.

We studied the acute renal metabolic response in rats made acidotic by a single oral dose of ammonium chloride. Cortical slices from acutely (2-h) acidotic rats utilized more glutamine and produced more ammonia and glucose from glutamine than slices from normal animals. When cortical slices from normal rats were pretreated in vitro with plasma isolated from acutely acidotic rats, they achieved similar increases in glutamine utilization, ammonia formation, and gluconeogenesis from glutamine. We did not observe such stimulation in normal cortical slices pretreated in a low pH-low bicarbonate medium. Our data show that a nondialysable factor is present in plasma from acutely acidotic rats that may be responsible for the early increase in the urinary ammonia observed in such animals.

Regulation of ammonia production by mouse proximal tubules perfused in vitro. Effect of luminal perfusion.

To investigate factors regulating ammonia (NH3) production by isolated defined proximal tubule segments, we examined the rates of total NH3 (NH3 + NH+4) production by individual proximal tubule segments perfused in vitro under a variety of perfusion conditions. Segments consisting of late convoluted and early straight portions of superficial proximal tubules were incubated at 37 degrees C in Krebs-Ringer bicarbonate (KRB) buffer containing 0.5 mM L-glutamine and 1.0 mM sodium acetate, pH 7.4. The rate of total ammonia production was calculated from the rate of accumulation of total NH3 in the bath. The total ammonia production rate by unperfused proximal segments was 6.0 +/- 0.2 (+/- SE) pmol/mm per minute, which was significantly lower than segments perfused at a flow rate of 22.7 +/- 3.4 nl/min with KRB buffer (21.5 +/- 1.4 pmol/mm per minute; P less than 0.001) or with KRB buffer containing 0.5 mM L-glutamine (31.9 +/- 2.5; P less than 0.001). The rate of NH3 production was higher in segments perfused with glutamine than in segments perfused without glutamine (P less than 0.01). The perfusion-associated stimulation of NH3 production was characterized further. Analysis of collected luminal fluid samples revealed that the luminal fluid total NH3 leaving the distal end of the perfused proximal segment accounted for 91% of the increment in NH3 production observed with perfusion. Increasing the perfusion flow rate from 3.7 +/- 0.1 to 22.7 +/- 3.4 nl/min by raising the perfusion pressure resulted in an increased rate of total NH3 production in the presence or absence of perfusate glutamine (mean rise in rate of total NH3 production was 14.9 +/- 3.7 pmol/mm per minute in segments perfused with glutamine and 7.8 +/- 0.9 in those perfused without glutamine). In addition, increasing the perfusion flow rate at a constant perfusion pressure increased the rate of luminal output of NH3. Total NH3 production was not affected by reducing perfusate sodium concentration to 25 mM and adding 1.0 mM amiloride to the perfusate, a condition that was shown to inhibit proximal tubule fluid reabsorption. These observations demonstrate that the rate of total NH3 production by the mouse proximal tubule is accelerated by perfusion of the lumen of the segment, by the presence of glutamine in the perfusate, and by increased perfusion flow rates. The increased rate of NH3 production with perfusion seems not to depend upon normal rates of sodium reabsorption. The mechanism underlying the stimulation of NH3 production by luminal flow is unknown and requires further study.

Luminal secretion of ammonia in the mouse proximal tubule perfused in vitro.

A major portion of the total ammonia (tNH3 = NH3 + NH+4) produced by the isolated perfused mouse proximal tubule is secreted into the luminal fluid. To assess the role of Na+-H+ exchange in net tNH3 secretion, rates of net tNH3 secretion and tNH3 production were measured in proximal tubule segments perfused with control pH 7.4 Krebs-Ringer bicarbonate (KRB) buffer or with modified KRB buffers containing 10 mM sodium and 0.1 mM amiloride. Net tNH3 secretion was inhibited by 90% in proximal tubule segments perfused with the pH 7.4 modified KRB buffer while tNH3 production remained unaffected. The inhibition of net tNH3 secretion by perfusion with the modified KRB buffer was only partially reversed by acidifying the modified KRB luminal perfusate from 7.4 to as low as 6.2. These data indicate that the Na+-H+ exchanger facilitates a major portion of net tNH3 secretion by the proximal tubule and that luminal acidification may play only a partial role in the mechanism by which the Na+-H+ exchanger mediates net tNH3 secretion.

Renal metabolism of alanine.

In the acidotic dog, alanine is extracted from plasma and utilized as a precursor of ammonia. Simultaneously, it is formed de novo within tubular cells and added to renal venous blood. When plasma concentration is within a normal range, production of alanine greatly exceeds utilization. Increasing the plasma concentration reduces production and increases utilization of plasma alanine. The infusion of glutamine increases the renal production of alanine without appreciable change in utilization of plasma alanine. These results are consonant with the view that alanine is metabolized by transamination with alpha-ketoglutarate to form glutamate, which is subsequently deaminated oxidatively to liberate ammonia. Conversely, alanine is formed by transamination of pyruvate with either glutamate or glutamine and is added to renal venous blood. The balance between production and utilization is dependent, at least in part, on the concentrations of the reactants.

Renal hemodynamics and ammoniagenesis. Characteristics of the antiluminal site for glutamine extraction.

Renal production of ammonia by the left kidney was studied in 31 acidotic dogs (NH(4)Cl) after acute constriction of the renal artery. Renal ammoniagenesis fell in direct proportion with the reduction in glomerular filtration rate and renal plasma flow. The renal extraction of glutamine by the experimental kidney fell in direct proportion with the reduction in renal hemodynamics. Extracted glutamine remained greater than filtered glutamine indicating that both the luminal and antiluminal transport sites were operative. The relationship between renal extraction of glutamine and ammoniagenesis observed during control was maintained after renal artery constriction (1.7 mumol NH(3) produced for each mumol of glutamine extracted). Systemic venous or renal intra-arterial infusion of glutamine during arterial constriction increased renal production of ammonia to or above control values. These observations indicate that the mechanisms responsible for glutamine extraction and ammonia production were operating normally despite reduced hemodynamics. When measured immediately after arterial clamping, the renal venous pNH(3) was found to rise significantly decreasing progressively thereafter towards control values. The extracted fraction of total glutamine delivered to the kidney (31%) did not change after acute reduction of the glutamine load. Thus, the antiluminal extraction site was incapable of lowering renal venous plasma glutamine concentration below 0.33 muM/ml. In a second series of experiments, the properties of the antiluminal site of transport for glutamine were studied after complete occlusion of the left ureter in acidotic and nonacidotic animals. Under these circumstances, it was demonstrated that the antiluminal site is capable of extracting sufficient glutamine to maintain total ammonia production at 60% or more of control. In acidotic animals, changes in cellular pNH(3) appeared to play a key role on the antiluminal extraction of glutamine since the significant rise in renal blood flow often observed after ureteral occlusion prevented the rise in pNH(3) noted when blood flow remained constant. Thus, when renal blood flow rose glutamine extraction and ammonia production were maintained at control values. In these acidotic animals, glutamine infusion failed to influence ammonia production until luminal transport was restored by release of ureteral clamp and resumption of glomerular filtration. The latter observation establishes that reabsorbed glutamine is utilized at least in part for ammonia production.

Plasma and urinary amino acids in primary gout, with special reference to glutamine.

Measurement of the plasma free amino acids by column chromatography (AutoAnalyzer) in 32 patients with primary gout showed statistically significant increases or decreases in several components when compared with the spectrum in 18 control subjects, but the absolute amounts involved were small and the mean total plasma amino acid concentrations in both groups were the same. In the urine all major amino acid components, notably glutamine, serine, threonine, and leucine, were lower in our gouty than in our nongouty subjects, as were also the corresponding renal clearance ratios. These deficits could be reproduced by restricting dietary protein, so appear to be due largely to the some-what reduced mean dietary protein intake of our gouty subjects. However, the low renal clearance of glutamine, the most striking and consistent of the deficits in urinary amino acids noted, could not be accounted for by dietary or other relevant factors, and is interpreted as indicating increased tubular reabsorption of glutamine in primary gout. This interpretation was supported by the results of glutamine loading. The possible compensatory relationship of the abnormality in renal handling of glutamine to the deficiency in renal production of ammonia previously reported is discussed.

Pathways of glutamine and organic acid metabolism in renal cortex in chronic metabolic acidosis.

The metabolism of labeled glutamine and of several labeled organic acid anions was compared in tissue slices of renal cortex from chronically acidotic and alkalotic littermate dogs. (15)NH(3) formation and (15)N-amideglutamine utilization were significantly increased by slices from acidotic animals providing further evidence for the similarity of the metabolic responses seen in the tissue slice system and the physiologic effects produced by chronic metabolic acidosis on renal metabolism in the intact animal. Slices from acidotic dogs formed more (14)CO(2) and glucose-(14)C than did slices from alkalotic animals when labeled glutamine, citrate, or malate was used as substrate but (14)CO(2) production from pyruvate-1-(14)C was slightly reduced in acidotic tissue. With most of the substrates used glucose-(14)C formation was small compared with (14)CO(2) formation. Using the amount of glucose-(14)C formed, the expected (14)CO(2) production was calculated based on the hypothesis that the primary site of action of metabolic acidosis is on a cytoplasmic step in gluconeogenesis. The actual difference in (14)CO(2) production between slices from acidotic and alkalotic animals always greatly exceeded this predicted amount, indicating that stimulation of gluconeogenesis represents a minor metabolic response to chronic metabolic acidosis. Evidence from experiments with citrate labeled in various positions showed that metabolic acidosis has its principal effect on an early step in substrate metabolism which must be intramitochondrial in location.

Relation of renal cortical gluconeogenesis, glutamate content, and production of ammonia.

Glutamate is an inhibitor of phosphate dependent glutaminase (PDG), and renal cortical glutamate is decreased in metabolic acidosis. It has been postulated previously that the rise in renal production of ammonia from glutamine in metabolic acidosis is due primarily to activation of cortical PDG as a consequence of the fall in glutamate. The decrease in cortical glutamate has been attributed to the increase in the capacity of cortex to convert glutamate to glucose in acidosis. In the present study, administration of ammonium chloride to rats in an amount inadequate to decrease cortical glutamate increased the capacity of cortex to produce ammonia from glutamine in vitro and increased cortical PDG. Similarly, cortex from potassium-depleted rats had an increased capacity to produce ammonia and an increase in PDG, but glutamate content was normal. The glutamate content of cortical slices incubated at pH 7.1 was decreased, and that at 7.7 was increased, compared to slices incubated at 7.4, yet ammonia production was the same at all three pH levels. These observations suggest that cortical glutamate concentration is not the major determinant of ammonia production. In potassium-depleted rats there was a 90% increase in the capacity of cortex to convert glutamate to glucose, yet cortical glutamate was not decreased. In vitro, calcium more than doubled conversion of glutamate to glucose by cortical slices without affecting the glutamate content of the slices, and theophylline suppressed conversion of glutamate to glucose yet decreased glutamate content. These observations indicate that the rate of cortical gluconeogenesis is not the sole determinant of cortical glutamate concentration. The increase in cortical gluconeogenesis in acidosis and potassium depletion probably is not the primary cause of the increase in ammonia production in these states, but the rise in gluconeogenesis may contribute importantly to the maintenance of increased ammoniagenesis by accelerating removal of the products of glutamine degradation.

Renal metabolic response to acid-base changes. II. The early effects of metabolic acidosis on renal metabolism in the rat.

The early renal metabolic response was studied in rats made acidotic by oral feeding of ammonium chloride. 2 hr after feeding of ammonium chloride there was already significant acidosis. Urinary ammonia also increased after ammonium chloride ingestion and at 1(1/2) hr was significantly elevated. In vitro gluconeogenesis by renal cortical slices was increased at 2 hr and thereafter increased steadily. Ammonia production by the same slices was also increased at 2 hr, but thereafter fell and at 6 hr had decreased to levels which, although higher than those of the control, were lower than those obtained from the rats acidotic for only 2 hr. There was no correlation between in vitro gluconeogenesis and ammonia production by kidney slices from rats during the first 6 hr of acidosis, but after 48 hr of ammonium chloride feeding, these two processes were significantly correlated. The early increase in renal gluconeogenesis was demonstrable with both glutamine and succinate as substrates. The activity of the enzyme phosphoenolpyruvate carboxykinase was increased after 4-6 hr of acidosis. During this time there was a decrease in renal RNA synthesis as shown by decreased uptake of orotic acid-(5)H into RNA. Metabolic intermediates were also measured in quick-frozen kidneys at varying times after induction of acidosis. There was an immediate rise in aspartate and a fall in alpha-ketoglutarate and malate levels. There was never any difference in pyruvate or lactate levels or lactate:pyruvate ratios between control and acidotic rats. Phosphoenolpyruvate rose significantly after 6 hr of acidosis. All the data indicate that increased gluconeogenesis is an early response to metabolic acidosis and will facilitate ammonia production by utilization of glutamate which inhibits the glutaminase I enzyme. The pattern of change in metabolic intermediates can also be interpreted as showing that there is not only enhanced gluconeogenesis, but also that there may be significant increase of activity of glutaminase II as part of the very early response to metabolic acidosis.

Effects of glutamine deamination on glutamine deamidation in rat kidney slices.

Glutamate is known to inhibit the activity of isolated glutaminase I; however, its actual physiologic importance in regulating renal ammoniagenesis has not been established. To determine the regulatory role of glutamate on the metabolism of glutamine by rat kidney slices, we followed the effects on glutamine (2 mM) deamidation of increased removal of glutamate via augmented deamination. Three agents (malonate, 2,4-dinitrophenol, and methylene blue) were known to and shown here to hasten exogenous glutamate deamination. In slices from 10 control rats, 21.5+/-1.7 (SEM) mumol/g of ammonia were formed from amide nitrogen and 9.3+/-0.5 (SEM) mumol/g from the amino nitrogen of glutamine in vitro. Over 90% of the glutamine deamidated formed glutamate at one point in its catabolism. After addition of malonate (10 mM), 2,4-dinitrophenol (0.1 mM), or methylene blue (0.5 mM), the production of ammonia from the amino group rose to 29.3+/-6.0 (SEM) mumol/g, 20.0+/-1.8 (SEM) mumol/g, and 15.5+/-4.2 (SEM) mumol/g, respectively; ammonia production from the amide nitrogen rose also, 45.1+/-7.3 (SEM) mumol/g, 39.7+/-2.6 (SEM) mumol/g, and 41.9+/-3.7 (SEM) mumol/g. In the case of the former two, a minimum of 99% and 75% of the glutamine catabolized formed glutamate. Despite increased glutamine catabolism, there was no build up of glutamate in the media. A correlation between the formation of ammonia from the amino and amide nitrogen was apparent. Since none of the three agents selected affected phosphate activated glutaminase I activity directly or appeared to affect glutamine transport, we interpret the increase in deamidation as an expression of deinhibition of glutaminase I activity secondary to lowered glutamate concentrations at the deamidating sites through more rapid removal of glutamate via hastened deamination. Interestingly, slices removed from acidotic rats produced more ammonia from both the amino 29.1+/-3.8 (SEM) and amide nitrogens 45.9+/-4.3 (SEM) of glutamine, without a buildup of glutamate in the medium. At least 90% of the glutamine deamidated formed glutamate. A common mechanism is proposed to explain these results and the previous ones.

Effect of adenosine 3',5'-monophosphate on production of glucose and ammonia by renal cortex.

In studies employing rat renal cortical slices, the addition of adenosine 3',5'-monophosphate (cyclic AMP) to the incubation medium caused an increase in production of glucose from glutamine, glutamate, alpha-ketoglutarate, fumarate, malate, and oxalacetate, but not from glycerol and fructose. These observations suggest that cyclic AMP accelerates a rate-limiting gluconeogenic reaction between oxalacetate and the triose phosphates. The addition to the medium of parathyroid hormone, which is known to increase renal cortical cyclic AMP, also stimulated glucose production from glutamine. When renal cortical slices were incubated in the presence of glutamine, the addition of cyclic AMP caused a fall in tissue glutamate concentration and a rise in ammonia production, as well as an increase in gluconeogenesis. These changes are similar to those observed in renal cortex of rats with induced metabolic acidosis. The present observations are consistent with a previously advanced hypothesis that cortical gluconeogenesis, ammonia production, and glutamate concentration may be interdependent.

Glutamine transport in rat kidney mitochondria in metabolic acidosis.

In order to study factors regulating renal ammoniagenesis, the transport and metabolism of L-glutamine were studied in mitochondria from kidneys of control and acidotic rats. On incubation in 1 mM [(14)C]glutamine, there was production and accumulation of [(14)C]glutamate within the matrix space. However no [(14)C]glutamine was detected in the matrix space, even with 10 mM [(14)C]glutamine as substrate or with inhibition of glutamine deamidation (low temperature, p-chloromercuribenzoate, mersalyl). These results suggest that glutamine crosses the inner membrane by a carrier-mediated step and that this step is rate-limiting in glutamine deamidation. In chronic acidosis there is a fourfold increase in the uptake of radioactivity from [(14)C]glutamine, but not from alpha-ketoglutarate, glutamate, or acetate. In 3-h acidosis, before any increase in extracted glutaminase levels, there is a significant and reproducible increase (39+/-3.8%, n = 25) in matrix uptake of radioactivity from [(14)C]glutamine and also an increased ammonia production (17+/-3.7%, n = 12). Administration of furosemide produces a similar degree of potassium depletion and a greater degree of sodium depletion over 3 h when compared to a 3-h acidosis. However, it produces no change in mitochondrial uptake of radioactivity. These results show that the adaptation of renal glutamine metabolism observed in acidosis is due to the acidosis and is demonstrable in isolated rat kidney mitochondria. The site of adaptation is in the carrier system, which transports glutamine across the inner membrane. The increased transport in acidosis delivers more glutamine to glutaminase, which results in the increased renal ammonia production.

The effect of ketone bodies on renal ammoniogenesis.

Infusion of ketone bodies to ammonium chloride-loaded acidotic dogs was found to induce significant reduction in urinary excretion of ammonia. This effect could not be attributed to urinary pH variations. Total ammonia production by the left kidney was measured in 25 animals infused during 90 min with the sodium salt of D,L-beta-hydroxybutyric acid adjusted to pH 6.0 or 4.2. Ketonemia averaged 4.5 mM/liter. In all experiments the ammonia content of both urine and renal venous blood fell markedly so that ammoniogenesis was depressed by 60% or more within 60 min after the onset of infusion. Administration of equimolar quantities of sodium acetoacetate adjusted to pH 6.0 resulted in a 50% decrease in renal ammonia production. Infusion of ketone bodies adjusted to pH 6.0 is usually accompanied by a small increase in extracellular bicarbonate (3.7 mM/liter). However infusion of D,L-sodium lactate or sodium bicarbonate in amounts sufficient to induce a similar rise in plasma bicarbonate resulted in only a slight decrement in ammonia production (15%). The continuous infusion of 5% mannitol alone during 90-150 min failed to influence renal ammoniogenesis. Infusion of pure sodium-free beta-hydroxybutyric acid prepared by ion exchange (pH 2.2) resulted in a 50% decrease in renal ammoniogenesis in spite of the fact that both urinary pH and plasma bicarbonate fell significantly. During all experiments where ketones were infused, the renal extraction of glutamine became negligible as the renal glutamine arteriovenous difference was abolished. Renal hemodynamics did not vary significantly. Infusion of beta-hydroxybutyrate into the left renal artery resulted in a rapid decrease in ammoniogenesis by the perfused kidney. The present study indicates that ketone bodies exert their inhibitory influence within the renal tubular cell. Since their effect is independent of urinary or systemic acid-base changes, it is suggested that they depress renal ammoniogenesis by preventing the transformation of glutamine and glutamate into alpha-ketoglutarate in the mitochondria of the renal tubular cell.

Protein-catabolic stage of isolated rat hepatocytes.

Isolated rat hepatocytes in suspension are in a protein-catabolic state (negative nitrogen balance), as measured by the continuous release of nitrogen in the form of amino acids and urea. The nitrogen loss corresponds to a protein degradation rate of 3-4% per h, while the rate of protein synthesis is negligible. Cells prepared from fasted, fed to regenerating livers are all highly protein-catabolic. The nitrogen balance is unaffected by insulin or amino acids (physiological mixture), and various metabolites and sera have only moderate effects. However, incubation of the cells for 2-4 h in a tissue culture medium (Dulbecco's) reduces the nitrogen loss dramatically, suggesting the formation of an anticatabolic factor under these conditions.

Ammonia production by the small intestine of the rat.

1. Slices of duodenum and jejunum produce ammonia from glutamine in vitro. 2. Ammoniagenesis does not increase in response to acidosis or potassium deficiency, two conditions known to cause enhanced ammoniagenesis in the kidney. 3. Gut contains glutaminase 1 as well as gamma-glutamyl transpeptidase. 4. These enzymes do not show any increase during starvation.

Free fatty acids in an animal model of Reye's syndrome.

Recent studies have indicated that viral infections, aspirin treatment and hyperammonemia are associated with Reye's syndrome. It has also been reported that free fatty acids in serum and total lipids in the liver of Reye's syndrome patients are elevated during illness. The role of the lipid changes in the development of the disorder cannot be optimally studied in human patients, because infection and aspirin ingestion occur prior to the earliest symptoms of Reye's syndrome. Effects of influenza B infection, aspirin treatment and hyperammonemia on the level of free fatty acids, total lipids and triacylglycerols in serum and liver of an animal model of Reye's syndrome are reported here. Hyperammonemia was produced in young, male ferrets either by feeding them small amounts of an arginine-deficient diet after overnight fasting or by an intraperitoneal injection of jackbean urease. The ferret model resembled Reye's syndrome in developing increased levels of individual and total serum free fatty acids, liver triacylglycerol and total lipids. The results also indicate that influenza infection or aspirin treatment, or both, while increasing the severity of encephalopathy in the deficient ferrets, did not cause a significant change in the level of serum free fatty acids. Other results suggest that elevation of serum ammonia, serum free fatty acid or liver lipids, either singly or in various combinations, does not provide conditions that can explain the rapidly developing encephalopathy in the arginine-deficient ferrets.

The role of amino acid catabolism in the formation of the tricarboxylic acid cycle intermediates and ammonia in anoxic rat heart.

Amino acid catabolism, the tricarboxylic acid cycle intermediates and ammonia formation were studied in isolated perfused rat heart under anoxia. The total net anaplerosis due to amino acid degradation in anoxia was equal to that in oxygenation (6.29 and 6.09 mumol/g dry weight per h, respectively) as a result of the increased transamination of glutamic and aspartic acids. During anoxic perfusion, the rate of catabolism of glutamic and aspartic acids was 1.5-times higher than in normoxia, while depletion of branched-chain amino acids, lysine, proline, arginine and methionine, was inhibited. Alanine was the product of excessive degradation of glutamic and aspartic acids. Under anaerobic conditions, in spite of inhibition of amino acid deamination, ammonia formation was increased 2.7-fold as compared to oxygenation. The principal amount of ammonia (96%) was produced at degradation of adenine nucleotides. A 2.5-fold increase in the pool of the tricarboxylic acid cycle intermediates under anoxia was associated mainly with accumulation of succinate. The data suggest that the coupling of alanine- and aspartate amino transferases is a mechanism controlling the tricarboxylic acid cycle pool size in anoxic heart.