She is slightly hypoglycaemic, and her non-esterified fatty acids are slightly higher than would be expected for an overnight fast. The lack of detectable ketone bodies is interesting and will bear further investigation. Her plasma concentrations of insulin and glucagon are appropriate for her mild hypoglycaemia. Her plasma ammonium concentration is somewhat higher than normal.

The elevated activities of aspartate and alanine transaminases, alkaline phosphatase and gamma-glutamyl transpeptidase are all suggestive of liver disease or damage.

The elevated activity of creatine kinase is suggestive of muscle damage - either skeletal or cardiac muscle, but the high normal activity of lactate dehydrogenase suggests that there has not been any recent death of cardiac muscle cells.

The abnormally low serum concentration of carnitine is interesting.

Carnitine is both synthesised in the body and also obtained the diet. This means that dietary carnitine deficiency is unlikely, and indeed strict vegetarians, who have negligible dietary sources of carnitine are still able to maintain a normal plasma concentration.

A urine sample showed that she excreted ??-times more carnitine than children of her age, and most of this was as free carnitine, not fatty acid esters of carnitine (acyl carnitine).

Her renal clearance of carnitine (both free and esterified) was almost the same as that of creatinine, whereas normally the renal clearance of free carnitine is only about 2% of that of creatinine (although esterified carnitine has a higher renal clearance than free carnitine).

Carnitine is a small, water-soluble compound, and will therefore be more or less completely filtered at the glomerulus. The fact that the normal renal clearance of carnitine is only about 2% of that of creatinine suggests that there must be active reabsorption of carnitine in the renal tubules. If CUD's renal clearance of carnitine is more or less the same as her creatinine clearance then it seems likely that she has a defect of the carnitine transporter that is responsible for the reabsorption of carnitine in the renal tubules. Her very low plasma carnitine is therefore most likely the result of excessive loss in the urine rather than a defect in carnitine synthesis or a dietary deficiency.

While CUD was in hospital she developed a viral infection that led to a moderate fever. The next morning she was deeply unresponsive and could not be woken up. A blood sample taken at this time gave the following results:

Her coma is obviously due to both profound hypoglycaemia, and the significantly elevated ammonium concentration. (You will see in a later exercise how even a modest increase in plasma ammonium can lead to a disturbance of consciousness).

It is noteworthy that despite her profound hypoglycaemia, while plasma non-esterified fatty acids are higher than normal (suggesting that they are being mobilised from adipose tissue triacylglycerol), again she has no detectable ketone bodies. This suggests a defect in the synthesis of ketone bodies from fatty acids.

This is presumably the result of gluconeogenesis from amino acids, in an attempt to maintain an adequate plasma concentration of ammonium. Normally the ammonium released in amino acid catabolism is used for the synthesis of urea, which is excreted in the urine. Urea synthesis occurs in the liver, and we already have evidence from the enzyme data when she was first admitted to hospital that she has liver damage or disease.

Since her main problem is profound hypoglycaemia, the obvious immediate treatment would be intravenous glucose. She responded well, and regained consciousness.

There is an increase in basal metabolic rate of about 13% for each degree C of fever - so a fever will increase her metabolic rate, and hence the rate at which she metabolises glucose - effectively the only fuel available to her tissues.

In order to investigate her muscle weakness and liver damage, biopsy samples were taken.

As shown on the right, her liver showed evidence of fatty infiltration, with an abnormal number of droplets of lipid in the liver, appearing in this slide as empty vacuoles.

Her muscle biopsy also showed deposition of abnormal amounts of lipid between the fibres.

It seems likely that although she can mobilise non-esterified fatty acids from adipose tissue in the fasting state, neither her muscle nor her liver can metabolise them adequately, so that they accumulate as triacylglycerol, disrupting liver and muscle function.

Carnitine was measured in the liver and muscle biopsy samples. The results were as follows:

The normal intracellular concentrations of carnitine are very much higher than that in plasma - muscle carnitine is some 50-fold higher than plasma carnitine, and in liver the concentration is more than 10,000-times that in plasma.

We can assume that the very high concentration of carnitine in liver results from both dietary intake (the liver is likely to buffer the concentration that is made available to the rest of the body), and also the fact that carnitine is synthesised in the liver.

The fact that muscle accumulates carnitine to a concentration that is so much higher than that in plasma suggests that there must be active transport of carnitine into muscle.

We already know that CUD is carnitine deficient because of excessive urinary losses, and this may explain the very low concentrations in her muscle and liver.

Cultured fibroblasts and myoblasts accumulate carnitine in a saturable manner. This uptake is inhibited by incubation under anaerobic conditions, the addition of cyanide (a respiratory poison), 2,4-dinitrophenol (an uncoupler of electron transport) and oligomycin (an inhibitor of ATP synthase), or incubation under conditions of sodium ion depletion. (See the exercise on Overheating after overdosing on E - and slimming by taking dinitrophenol to revise the actions of these compounds).

This confirms that muscle and fibroblasts take up carnitine by active transport; the inhibition by sodium ion depletion suggests that the carnitine transporter is sodium-dependent.

The graphs on the right show uptake of radioactive carnitine into cultured fibroblasts and myoblasts from control subjects and CUD.

From data reported by Pons et al. Pediatric Research 42: 583 1997

This suggests that CUD's problem is in the carnitine transporter, not only in the renal tubules, but also in muscle and other tissues. The kidney transporter prevents loss of carnitine in the urine; the transporter in other tissues permits uptake of carnitine to a considerably higher concentration than that in plasma.

The liver does not have the same high affinity carnitine transporter as kidney, muscle and other tissues, but has a low affinity transporter that will permit it to take up or secrete carnitine.

It is highly unlikely that she would have defects in two different carnitine transporters. Why do you think her liver carnitine was so low?

This presumably reflects her very low plasma concentration of carnitine as a result of the continual loss in her urine.

Similar problems of early fatigue, muscle weakness and profound fasting hypoglycaemia with undetectable ketone bodies occur in patients treated for prolonged periods with with the antibiotic pivampicillin.

Pivampicillin is the antibiotic ampicillin (a penicillin-like beta-lactam antibiotic) that has been conjugated with pivalic acid to increase its absorption and bio-availability when given orally. In the liver the conjugate is hydrolysed, releasing active ampicillin and pivalic acid, as shown on the right.

Like other non-esterified fatty acids, the pivalic acid is esterified with coenzyme A, but cannot be metabolised; there is no pathway for pivalic acid or pivaloyl CoA metabolism in human tissues.

We saw in the exercise on Poisoned by unripe ackee fruit that there is only a small amount of CoA in tissues, and it turns over rapidly. The problem with unripe ackee fruit was that the metabolite of hypoglycine esterified to coenzyme A depleted tissue levels of CoA to such an extent that fatty acid metabolism and oxidative metabolism of pyruvate arising from glycolysis were severely impaired, leading to profound hypoglycaemia.

Here the problem does not seem to be sequestration of CoA, since the pivaloyl moiety can be transferred from CoA onto carnitine, forming pivaloyl carnitine, which is excreted in the urine. However, this can lead to depletion of tissue reserves of carnitine, and eventually to carnitine deficiency.

This provides an easy way of investigating carnitine deficiency, and hence carnitine function, in experimental animals. In the following studies, rats were fed a carnitine-free diet and dosed with pivalic acid to induce carnitine deficiency. After the animals were killed, studies were performed using the gastrocnemius muscle and isolated liver cells.

In the first set of experiments portions of gastrocnemius muscle were incubated with three different fatty acids, all labelled with [14C] in carbon-1 at a specific radioactivity of 1 µCi /µmol, with and without the addition of carnitine. The tissue samples were incubated in centre well vials, and after 20 min, 1 mL of methoxymethylamine was injected through the seal into the outer compartment of the incubation vial, to trap carbon dioxide, and 0.5 mL of 1 mol /L perchloric acid was injected into the incubation mixture in the centre well, to precipitate proteins, and drive carbon dioxide out of the solution.

The flasks were shaken for a further 60 min, then the methoxymethylamine (containing carbon dioxide) was washed out with scintillator fluid and transferred to a scintillation counter vial.

During the incubation the muscle was stimulated electrically, in order to increase ATP utilisation, and hence oxidative metabolism. The results were as follows (figures show dpm in carbon dioxide /gram tissue for 3 x replicate incubations):

The addition of carnitine has no effect on the oxidation of the short-chain fatty acid butyrate, nor on the medium-chain fatty acid caprate. However, in the absence of carnitine there is very little oxidation of the long-chain fatty acid palmitate, but the addition of carnitine leads to an increase in palmitate oxidation to the same level as that of butyrate or caprate.

This suggests that carnitine is essential for the oxidation of long-chain fatty acids, but not short- and medium-chain fatty acids.

In the second set of experiments isolated hepatocytes were incubated [U-14C]palmitate (i.e. palmitate labelled in all 16 carbon atoms with 14C) or [U-14C]caprate (i.e. caprate labelled with 14C in all 10 carbon atoms. As well as measuring radioactivity in carbon dioxide, the acidified incubation medium was extracted with a mixture of chloroform and methanol (to remove any unmetabolised fatty acid) and the radioactivity in water-soluble compounds (the ketone bodies, acetoacetate, beta-hydroxybutyrate and acetone) was measured. The results were as follows (figures show dpm /gram tissue for 3 x replicate incubations):

Again carnitine is essential for oxidation of the long-chain fatty acid, but not the medium-chain fatty acid. In the absence of carnitine there is no formation of ketone bodies from palmitate, although there is from caprate. You will see in a later exercise how ketone bodies are synthesised - for now it is enough to say that they are synthesised from the product of fatty acid oxidation.

She has an impaired ability to oxidise long-chain fatty acids, which are the main fatty acids found in adipose tissue reserves. In the fasting state she continues to release non-esterified fatty acids from adipose tissue. They cannot be metabolised to any significant extent, and so accumulate in liver and muscle, where they are esterified to form triacylglycerol, which form the lipid droplets seen in the liver and muscle histology. We have no direct evidence, but it is probable that there is also fatty infiltration of her heart muscle, since heart will also take up fatty acids, but be unable to metabolise them.

Since she cannot oxidise long-chain fatty acids, she cannot synthesise ketone bodies in the fasting state. As a result, even after a normal overnight fast she is more profoundly hypoglycaemic than would be expected, since there is no alternative to glucose as a fuel for her tissues.

Why do you think that impaired fatty acid metabolism might lead to easy fatigue and muscle weakness?

Muscle only uses significant amounts of carbohydrate under two conditions:

• in maximum exertion, when it metabolises mainly its glycogen reserves, and to a considerable extent metabolises anaerobically (see the exercise on Breathless after sprinting)
• in the fed state, when insulin stimulates the uptake of glucose into muscle, both for glycogen synthesis and also for metabolism as the main fuel of muscle

At other times the main, and indeed preferred fuel of muscle is fatty acids, which may arise from:

• plasma lipoproteins (chylomicrons and very low density lipoprotein) in the fed state)
• non-esterified fatty acids released from adipose tissue (in the fasting state)
• fatty acids arising from triacylglycerol between muscle fibres

However, fatty acid metabolism cannot meet all of muscle's energy needs, and in the fasting state it also needs a supply of ketone bodies from the liver.

The impaired fatty acid oxidation, and possibly also fatty infiltration, will also lead to cardiomyopathy, and heart failure.

Although her problem is due to defective transport of carnitine into muscle and other tissues, it is likely that relatively large supplements of carnitine may be helpful. The studies of myoblast and fibroblast carnitine uptake above showed that there was some very slight residual activity of the transporter, so if her plasma concentration of carnitine can be maintained relatively high this may permit modest uptake into tissues. Of course, she will still excrete a very large amount of carnitine, and the supplement should be given in divided doses or as a sustained release preparation.

Case reports of patients like CUD show that within a few weeks of starting high dose carnitine supplementation the heart abnormalities begin to reverse, and there is recovery of muscle strength.

In addition, it would be sensible to limit her intake of long-chain fatty acids, and perhaps provide a source of triacylglycerols containing medium-chain fatty acids, since these seem to be oxidised more or less normally in carnitine deficiency.

We now need to consider how carnitine is involved in the oxidation of fatty acids.

In the fed state fatty acids are available to enter tissues as a result of lipoprotein lipase acting at at the cell surface on chylomicrons and very low density lipoprotein (See the exercise on Children with fatty diarrhoea). In the fasting state non-esterified fatty acids are released from adipose tissue and circulate in the bloodstream bound to serum albumin. They readily enter cells, but at no time is there any significant amount of non-esterified fatty acid in cells.

If isolated hepatocytes are incubated with [14C]palmitate then what is found in the cells is a mixture of small amounts of palmitoyl CoA and rather more palmitoyl carnitine. There is an enzyme that catalyses the esterification of fatty acids with CoASH to form fatty acyl CoA, but no enzyme that will esterify non-esterified fatty acids to carnitine.

However, as shown on the right, there is an enzyme that will transfer a fatty acyl group from CoA onto carnitine.

We can now conduct a further experiment using the liver from the carnitine deficient rats. This time, instead of incubating isolated hepatocytes, the liver has been homogenised in buffer, and subjected to differential centrifugation to prepare isolated mitochondria.

In the first set of experiments isolated liver mitochondria from carnitine deficient rats were incubated with [14C]palmitate or [14C]palmitoyl CoA in centre well vials, and radioactivity in carbon dioxide and ketone bodies was determined, as described above. The results were as follows (figures show dpm /mg protein for 3 x replicate incubations):

The first conclusion is that fatty acid oxidation and ketone body formation occur in mitochondria, and not in the cytosol.

Regardless of whether or not carnitine is added, palmitate is not oxidised. However, palmitoyl CoA is oxidised, but only when carnitine is present. This still does not answer the question of what is the role of carnitine in fatty acid oxidation.

In the next set of experiments, isolated mitochondria were incubated with [14C]palmitoyl CoA, palmitoyl [35S]CoA, [14C]palmitoyl carnitine (i.e. with the fatty acid labelled) or palmitoyl [14C]carnitine (i.e. with the carnitine labelled). In each case the specific radioactivity of the substrate was 1 µCi / mmol. After 20 minutes the mitochondria were centrifuged out, and washed by resuspending in ice-cold buffer and recentrifuging three times. Radioactivity in the washed mitochondria was then measured. The results were as follows (figures show dpm /mg protein for 3 x replicate incubations):

When carnitine is provided the 14C label from the palmitoyl moiety of palmitoyl CoA enters the mitochondria. However, the 35S label from the CoA does not enter the mitochondria. This suggests that it palmitoyl carnitine that enters the mitochondria, and that at the outer face of the mitochondria carnitine acyltransferase acts to transfer the fatty acid from CoA onto carnitine.

This is confirmed by the fact that when palmitoyl carnitine is provided, the label from both the fatty acid and the carnitine is found inside the mitochondria. NOw ge addition of carnitine make no difference to the amount of label that accumulates inside the mitochondria. This is further evidence that it is palmitoyl carnitine that enters the mitochondria.

The amount of label that accumulates in the mitochondria from the labelled carnitine is significantly less than the accumulation of label from the fatty acid.

Since the same amount of 14C was added as either [14C]palmitoyl carnitine or palmitoyl [14C]carnitine, the fact that less of the label from carnitine has accumulated inside the mitochondria than from fatty acid suggests that free carnitine is re-exported from the mitochondria.

There are two possible tests.

• A quick and simple experiment would be look for radioactive free carnitine and palmitate in the incubation medium after incubating with [14C]palmitoyl carnitine and palmitoyl [14C]carnitine. If there is hydrolysis of palmitoyl carnitine at the outer surface of the mitochondrion then there should be equal amounts of both. On the other hand, if palmitoyl carnitine enters the mitochondria, and free carnitine is exported, then only radioactive carnitine, and not radioactive palmitate will be detected.
• A more tedious experiment would be to replete the carnitine deficient rats with [14C]carnitine before killing them and preparing mitochondria as described above. If these mitochondria are incubated with non-radioactive palmitoyl carnitine and radioactive carnitine is found in the incubation medium then you will have confirmed that carnitine is indeed exported from the mitochondria.

If you were to perform these experiments, you would indeed find that free carnitine is exported from mitochondria - but only when there is acyl carnitine outside the mitochondria to be transported in. A single protein transports acyl carnitine into the mitochondria in exchange for free carnitine being transported out.

Carnitine acyltransferase can be solubilised from the outer face of the mitochondrion using detergents, so that the kinetic properties of the enzyme can be studied. One such set of experiments showed that malonyl CoA (which is an intermediate in fatty acid synthesis - a cytosolic pathway) is a potent inhibitor of carnitine acyltransferase.

This provides an excellent route for metabolic regulation. If the cells are synthesising fatty acids in the cytosol (an energy-dependent process) then it would be pointless for the newly synthesised fatty acids to be taken into the mitochondria to be oxidised. Therefore, as malonyl CoA accumulates for fatty acid synthesis, it inhibits the transport of fatty acids into the mitochondria, and so inhibits fatty acid oxidation.

So far our studies have shown that fatty acids are esterified to form fatty acyl CoA in the cytosol, and at the outer face of the mitochondrion the fatty acid is transferred to carnitine. Acyl carnitine enters the mitochondria in exchange for free carnitine leaving, and the fatty acids that have been transported into the mitochondria undergo oxidation, either to carbon dioxide and water or to form ketone bodies.

The next set of experiments was conducted using disrupted mitochondria. The preparation was incubated with [14C]palmitoyl CoA or [14C]palmitoyl carnitine, with and without the addition of CoASH, and carbon dioxide was measured as described above. The results were as follows (figures show dpm /mg protein for 3 x replicate incubations):

Palmitoyl carnitine is not substrate for oxidation, but palmitoyl CoA is. When CoASH is added to the incubations, palmitate provided as palmitoyl carnitine is oxidised. This suggests that there is another carnitine acyltransferase at the inner face of the inner mitochondrial membrane that catalyses the transfer of the acyl group from acyl carnitine onto CoA, forming fatty acyl CoA inside the mitochondrion.

Studies with isolated mitochondrial inner and outer membrane preparations confirm this. There is one isoenzyme of carnitine acyltransferase at t the outer face of the outer mitochondrial membrane, and a separate isoenzyme at the inner face of the inner mitochondrial membrane.

The overall process of fatty acid transport into the mitochondria is as shown in the diagram on the right (CAT = carnitine acyltransferase)

It was noted above that short- and medium-chain fatty acids do not require carnitine for uptake into mitochondria. All the available evidence is that they are not esterified with CoA in the cytosol, but are transported into mitochondria as free fatty acids, and inside the mitochondrion are esterified with CoA ready to undergo oxidation.

Key points from this exercise:

• Carnitine is a small water-soluble molecule that is filtered at the glomerulus; normally about 98% of filtered free carnitine is reabsorbed by active transport in the renal tubules.
• Carnitine deficiency can result from a defect in this transport protein, which is found not only in the kidney, but also in tissues such as muscle that take up carnitine from the circulation.
• Most acyl carnitine is not reabsorbed in the kidneys, but is excreted. This means that abnormal organic acids that are esterified to carnitine can also cause carnitine deficiency.
• Fatty acid oxidation is severely impaired by carnitine deficiency, leading to fasting hypoglycaemia with negligible production of ketone bodies (which are formed as a product of fatty acid oxidation in the liver).
• Fatty acids are still mobilised from adipose tissue in the fasting state in carnitine deficiency, and are taken up by liver and muscle. They cannot be metabolised, and are esterified to triacylglycerol, leading to fatty infiltration of liver and muscle. This leads to metabolic impairment, early fatigue and muscle weakness, and cardiomyopathy, which may lead to heart failure.
• Fatty acids that enter tissues are rapidly esterified with CoA. Acyl CoA does not cross the mitochondrial membrane.
• At the outer face of the outer mitochondrial membrane the fatty acyl group is transferred from CoA onto carnitine by carnitine acyltransferase 1. Acyl carnitine is then transported into the mitochondrial matrix by an acyl carnitine / carnitine transporter that takes in acyl carnitine only in exchange for efflux of free carnitine from the mitochondrial matrix.
• At the inner face of the inner mitochondrial membrane carnitine acyltransferase 1 transfers the fatty acyl group from carnitine onto CoA. The resultant fatty acyl CoA is a substrate for mitochondrial oxidation.
• Only long-chain fatty acids require carnitine for mitochondrial uptake. Short- and medium-chain fatty acids can enter the mitochondria without the need for carnitine.