Urea synthesis in the liver, and potentially fatal hyperammonaemia in a child

The dietary intake of protein is the main factor that determines how much urea is synthesised and secreted, since protein is the main source of nitrogenous compounds in the diet. Remember that for an adult in nitrogen balance the excretion of nitrogenous compounds (mainly urea in the urine) is equal to the dietary intake of nitrogenous compounds. See the exercise on Nitrogen balance and protein requirements for more on this topic.

When isolated hepatocytes with increasing concentrations of ammonium, there is a steady increase in the formation of urea at low concentrations of ammonium, levelling off as the pathway for urea formation becomes saturated. If such studies are performed with isotopically labelled ammonium (15N, a stable isotope) only one of the two N atoms in urea is labelled.

The sources of ammonium ions in the liver

There are two ways in which ammonium is formed in the liver: from glutamine by the action of glutaminase, and from adenosine, by the action of adenosine deaminase. Each provides about half the ammonium that is incorporated into urea directly.

Glutamine is formed from ammonium in peripheral tissues, as a way of transporting ammonium arising from amino acid and amine metabolism to the liver.

One possible source of urea is the reaction of arginase, which catalyses hydrolysis of the amino acid arginine to yield urea and ornithine, as shown on the right.

Isolated hepatocytes were incubated with varying concentrations of ammonium chloride and 0, 2.5, 5 or 10 mmol /L arginine. The reaction was stopped after 30 min by addition of trichloroacetic acid, and denatured protein was removed by centrifugation. The amount of urea in the supernatant form each incubation was measured by reaction with diacetyl monoxime, ferric ions and thiosemicarbazide to form a red colour that was measured at 540nm.

The results are shown in the table below and the graphs on the right.

Urea formed (mmol /L) in the presence of different concentrations of arginine, with no added ammonium or a saturating amount (100 mmol /L)

Since the product of the reaction catalysed by arginase is ornithine, it will be interesting to see the effect of adding ornithine to the incubations. In this experiment, isolated hepatocytes were incubated with varying concentrations of ammonium chloride and 0, 2.5, 5 or 10 mmol /L ornithine, for 30 minutes. The reaction was stopped and the amount of urea formed was measured as described for experiment 1.

The results are shown in the table below and the graphs on the right.

Urea formed (mmol /L) in the presence of different concentrations of ornithine, with no added ammonium or a saturating amount (100 mmol /L)

There is no formation of urea when no ammonium is provided. This is as you would expect from the chemistry of ornithine; there is no way in which it can be a source of urea directly.

However, in the presence of ammonium, there is considerably more urea formed when ornithine is added than when no ornithine is added. It therefore seems possible that both arginine and ornithine are intermediates in the pathway of urea synthesis, and increasing the amount of either will increase the rate of urea formation.

Experiment 3:The metabolism of ornithine

Isolated hepatocytes were incubated with 100 mmol /L ammonium chloride and 10 mmol /L [14C-2]ornithine at a specific radioactivity of 0.1 µCi /mmol for 30 minutes. The reaction was stopped by the addition of trichloroacetic acid, followed by centrifugation to precipitate denatured protein. The supernatant was neutralised and metabolites were measured by liquid chromatography and scintillation counting to determine radioactivity.

In addition to ornithine, three compounds were found to be labelled: citrulline, arginine and one that was eventually identified as argininosuccinate. The asterisk shows the labelled carbon atom, as determined by mass fragmentation studies.

It now seems likely that arginine, which we know is hydrolysed to yield urea and ornithine, is synthesised from ornithine, since label from [14C]ornithine appears in arginine. Citrulline and argininosuccinate seem to be intermediates in the pathway.

Experiment 4: The effect of citrulline on urea synthesis

In this experiment, isolated hepatocytes were incubated with varying concentrations of ammonium chloride and 0, 2.5, 5 or 10 mmol /L citrulline, for 30 minutes. The reaction was stopped and the amount of urea formed was measured as described for experiment 1.

The results are shown in the table below and the graphs on the right.

Urea formed (mmol /L) in the presence of different concentrations of ornithine, with no added ammonium or a saturating amount (100 mmol /L)

These results are similar to those seen when arginine was added to the incubation. With no added ammonium there is stoichiometric formation of urea from citrulline. 1 mol of urea is formed for each mol of citrulline added.

However, in the presence of a saturating amount of ammonium there is considerably more urea formed per mol of citrulline than can be accounted for by the amount of citrulline added.

This suggests that the nitrogen added when citrulline is formed from ornithine has come from ammonium.

The catalytic effect of adding arginine, ornithine or citrulline on urea synthesis from ammonium suggests that there is a cyclic pathway, with ornithine, citrulline, argininosuccinate and arginine as intermediates.

Experiment 5: The incorporation of ammonium

The reaction to form citrulline from ornithine involves incorporation of both nitrogen and carbon, as shown in the diagram on the right.

In this experiment, isolated hepatocytes were incubated with 100 mmol /L ammonium chloride, 10 mmol /L ornithine and [14C] sodium bicarbonate at a specific radioactivity of 0.1 µCi /mmol for 30 minutes. The reaction was stopped by the addition of trichloroacetic acid, followed by centrifugation to precipitate denatured protein. The supernatant was neutralised and metabolites were measured by liquid chromatography and scintillation counting to determine radioactivity.

Radioactivity was found in citrulline, argininosuccinate, arginine and urea, and a metabolite that was identified as carbamoyl phosphate (shown on the left).

It seems likely that ammonium is incorporated into carbamoyl phosphate, and then into citrulline.

Carbamoyl phosphate is synthesised from ammonium, carbon dioxide and ATP in the reaction catalysed by carbamoyl phosphate synthetase, shown on the left.

There are two isoenzymes of carbamoyl phosphate synthetase in liver cells:

carbamoyl phosphate synthetase 1 is a mitochondrial enzyme; it is induced by feeding a high-protein diet.

carbamoyl phosphate synthetase 2 is a cytosolic enzyme; it is inhibited by pyrimidine nucleotides. (Carbamoyl phosphate is the starting substrate for pyrimidine synthesis).

Carbamoyl phosphate synthetase 1, the mitochondrial enzyme that is induced by feeding a high protein diet. (This isoenzyme uses ammonium as the source of nitrogen; the cytosolic enzyme that is involved in pyrimidine synthesis uses glutamine directly as its substrate.)

Hall and coworkers (1958) reported that synthesis of citrulline from ornithine by liver preparations required not only ATP, ammonium and bicarbonate, but also a soluble factor that could be isolated from liver, and which they identified as N-acetylglutamate (shown in the inset box in the graph on the right).

The graph on the right shows the effect of adding N-acetylglutamate on the formation of citrulline.
[From data reported by Hall, LM et al, Journal of Biological Chemistry 230: 1013, 1958]

At the end of the incubations the same amount of N-acetylglutamate remained in the incubations as had been added - i.e. it was not consumed in the reaction. Subsequent studies with [14C]N-acetylglutamate showed that no other compounds became labelled during the incubation.

When carbamoyl phosphate was provided, instead of ammonium, bicarbonate and ATP, citrulline was formed without the need for any added N-acetylglutamate, and adding N-acetylglutamate had no effect on the amounts of citrulline formed.

What conclusions can you draw from these results?

The amount of citrulline formed is ~5-fold higher than the amounts of N-acetylglutamate added, and N-acetylglutamate is not consumed, not is it incorporated into any other compound. This means that it is not an intermediate in the reaction pathway.

N-Acetylglutamate is an obligatory activator of carbamoyl phosphate synthetase.

It was noted above that when [15N]ammonium is used, only one of the nitrogen atoms in urea is labelled. This raises the question of the source of the second nitrogen atom in urea.

The second nitrogen of urea comes from aspartate; the reactions of argininosuccinate synthetase and argininosuccinase are shown below:

Experiments using [14C]aspartate showed that as well as fumarate, malate and oxaloacetate became labelled.

Experiments using [14C] fumarate showed that as well as malate and oxaloacetate, aspartate also became labelled.

Fumarate is an intermediate in the citric acid cycle, forming oxaloacetate. In turn, oxaloacetate is the substrate for a variety of transaminases, forming aspartate. This means that only a catalytic amount of aspartate is required for urea synthesis; it is regenerated by citric acid cycle activity and transamination.

We can now put together the full pathway for urea synthesis from ammonium:

Click here to download a printable version of this pathway.

There is a cost of 2 x ATP per mol of carbamoyl phosphate synthesised, and the equivalent of 2 x ATP per mol of argininosuccinate synthesised, since AMP, rather than ADP, is formed in this reaction.

Against this cost of 4 x ATP, there is a gain of ~2.5 x ATP from re-oxidation in the mitochondrial electron transport chain of the NADH formed by malate dehydrogenase.

The net cost of urea synthesis from ammonium is therefore 1.5 x ATP per mol.

Potentially fatal hyperammonaemia in a child: argininosuccinic aciduria

A rare genetic defect of the enzyme argininosuccinase leads to the condition of argininosuccinic aciduria; the affected infants excrete large amounts of argininosuccinic acid in their urine.

More importantly, affected infants suffer from hyperammonaemia, and after even a moderate intake of protein can become comatose, and may suffer convulsions, leading to brain damage, or may die.

Normally, only catalytic amounts of arginine and the other intermediates of the urea cycle are required, because arginine is regenerated form ornithine. However, in argininosuccinic aciduria, large amounts of argininosuccinate are excreted in the urine. This rapidly depletes the liver content of arginine (and the other intermediates), resulting in insufficient ornithine to take up carbamoyl phosphate, and therefore accumulation of ammonium that cannot be metabolised.

Brusilow & Batshaw (1979) reported the successful treatment of a child with argininosuccinic aciduria using supplements of arginine

[Brusilow, S W &. Batshaw, ML (1979). Arginine therapy of argininosuccinase deficiency. Lancet 313 (issue 8108): 124-7].

If the child is provided with relatively large amounts of arginine, then the two nitrogen atoms that would have been excreted as urea can be excreted as argininosuccinate. There is now a linear pathway from arginine to argininosuccinate:

Disappearing label from urea

A group of young men who were in in N balance were given a slow intravenous infusion of 20 mmol [13C,15N]urea (i.e. urea in which both nitrogen atoms were the stable isotope 15N and the carbon atom was the stable isotope 13C. Their urine was collected for 24 hours.

You would expect to recover 40 mmol 15N and 20 mmol 13C in urine over 24 hours.

In fact, the results are shown in the table below:

Only 85% of the 15N label has been recovered, and of that, only 85% is in urea, the remainder is in other nitrogenous compounds.

Only 2.5% of the 13C label has been recovered, and of that, very little is in urea.

This suggests that urea has been hydrolysed to carbon dioxide and ammonium, a reaction catalysed by urease. The resultant ammonium has been used largely to resynthesise urea, but a significant proportion has been incorporated into other compounds (probably mainly amino acids), many of which have been retained in the body.

The labelled carbon dioxide will be diluted by the much larger amount of carbon dioxide produced in respiration, and most of it will be exhaled.

Urease is well known in plants and bacteria, and indeed the first enzyme to be crystallised was urease form the jack bean. However, there is no mammalian urease.

The infusion of labelled urea was repeated after the volunteers had received the antibiotic neomycin for four days. Neomycin is a broad spectrum antibiotic that is not absorbed from the gut, but kills essentially all intestinal bacteria.

The results are shown in the table below:

This suggests that the hydrolysis of urea is the result of intestinal bacterial action.

Urea can diffuse across the intestinal wall, where it is a substrate for bacterial urease. The bacteria use some of the resultant ammonium for amino acid synthesis, and then bacterial protein synthesis. Some of the ammonium is absorbed from the intestinal lumen into the bloodstream, and then into the liver, where it is used mainly for urea synthesis.

Blood samples were taken from the volunteers on the day after the first experiment; plasma proteins were hydrolysed and 15N label in amino acids was determined by mass spectrometry. As expected, there was significant labelling of glutamate and glutamine, as well as a number of other amino acids. Labelling of non-essential amino acids is not surprising, but the unexpected finding was that all of the essential amino acids also contained 15N.

As intestinal bacteria die, some the proteins that they have synthesised are hydrolysed, and the amino acids that they have synthesised using the ammonium form 15N urea are released and absorbed into the bloodstream, where they can be used for protein synthesis. Bacteria are capable of synthesising all the amino acids they require, including those that are dietary essentials for human beings.

This entero-hepatic circulation of urea is shown below

Key points from this exercise:

• The pathway of urea synthesis is cyclic. Arginine is synthesised stepwise form ornithine, and is then hydrolysed to release urea and reform ornithine. This means that adding any of the intermediates of the cycle will increase the rate at which urea is synthesised form ammonium.
• One of the nitrogen atoms in urea comes from ammonium, and the other from aspartate.
• The ammonium nitrogen comes from either glutamine or adenosine, and is incorporated as carbamoyl phosphate. Carbamoyl phosphate synthetase has an absolute requirement for N-acetylglutamate as an activator.
• Aspartate reacts with citrulline to form argininosuccinic acid, which is then cleaved to release fumarate and arginine. Fumarate is metabolised to oxaloacetate via the citric acid cycle, and oxaloacetate is the substrate for transamination to form aspartate. This means that only a catalytic amount of aspartate is required.
• Infants who lack argininosuccinase excreted large amounts of argininosuccinic acid in the urine, so depleting their available pool of ornithine to take up ammonium from carbamoyl phosphate. They therefore become hyperammonaemic after a moderately high protein meal. Provision of supplements of arginine provides a linear pathway from ornithine to argininosuccinate for the excretion of ammonium, so relieving the hyperammonaemia.
• Urea diffuses form the bloodstream into the intestinal lumen, where it is a substrate for bacterial urease. The ammonium liberated may either be used by bacteria for amino acid and protein synthesis or may be absorbed into the hepatic portal vein and used for synthesis or urea and non-essential amino acids in the liver. Some essential amino acids formed by intestinal bacteria may be absorbed.