How is NADH from glycolysis normally re-oxidised?

This suggests that NADH cannot cross the mitochondrial membrane.

If NADH does cannot cross the mitochondrial membrane, it cannot be re-oxidised via the electron transport chain. This means that we are left with the problem of how NADH produced in the cytosol can be re-oxidised.

In skeletal muscle lactate dehydrogenase acts to reduce pyruvate to lactate, in order to re-oxidise NADH. However, the form of lactate dehydrogenase in heart muscle acts preferentially in the opposite direction, oxidising lactate to pyruvate. Heart takes up lactate from the bloodstream and uses it as a metabolic fuel.

In the following experiments, isolated cardiomyocytes (heart muscle cells) were used to study the metabolism of lactate under various conditions.

Fluoropyruvate is a potent inhibitor of pyruvate dehydrogenase. While it might also be expected to inhibit lactate dehydrogenase acting in the direction of reduction of pyruvate to lactate, it would not be expected to inhibit the oxidation of lactate to pyruvate, especially in cardiomyocytes, whose lactate dehydrogenase has a very much higher affinity for lactate than for pyruvate.

Cardiomyocytes were incubated with 100 mmol /L lactate and 1 mmol /L fluoropyruvate in an oxygen electrode. There was consumption of oxygen that was dependent on the addition of ADP, and a P:O ratio of ~2.5 was observed.

After 30 minutes the reaction mixture was pipetted out into a vial and frozen in liquid nitrogen. The cells were disrupted by repeated cycles of thawing and freezing, then centrifuged to remove mitochondria.

There was a significant accumulation of pyruvate in the cytosol.

These results suggest that lactate has been oxidised to pyruvate, with the reduction of NAD to NADH, and that the resultant NADH has been re-oxidised in the mitochondria. However, we know that NADH does not cross the mitochondrial membrane.

In further studies with cardiomyocytes incubated with lactate and fluoropyruvate, a variety of inhibitors of electron transport or oxidative phosphorylation were used. The following results were obtained:

These results confirm the suggestion that the NADH is being re-oxidised in the mitochondria, since inhibition of cytochrome oxidase (by cyanide), inhibition of complex I (by rotenone or antimycin A) and inhibition of ATP synthase (by oligomycin) all lead to inhibition of oxygen consumption and pyruvate formation. If there were any other way in which the NADH formed by lactate dehydrogenase could be re-oxidised then inhibition of the electron transport chain or ATP synthase would not inhibit pyruvate formation under these conditions.

In the next series of experiments the cells were incubated with [14C]lactate, with and without the addition of fluoropyruvate, and the cytosol preparation was subjected to high pressure liquid chromatography linked to a scintillation counter to permit determination of radioactivity in metabolites. Apart from lactate, four compounds were found to be labelled: aspartate, oxaloacetate and malate.

The results were as follows (figures show dpm in the sample):

The results in the presence of fluoropyruvate confirm what we have already seen, that lactate can be oxidised to pyruvate, which accumulates in the cytosol if it cannot undergo reaction with pyruvate dehydrogenase in the mitochondria.

The results in the absence of fluoropyruvate are more interesting. They show that much of the pyruvate has now entered the mitochondria, and a number of compounds in the cytosol have become labelled. This means that they must have been formed in the mitochondria from the acetyl CoA produced by pyruvate dehydrogenase. (You will see in later exercises that malate and oxaloacetate are intermediates in the citric acid cycle, and would be expected to become labelled if [14C]pyruvate is provided.

Oxaloacetate is a product of oxidation of malate - the enzyme malate dehydrogenase catalyses the following reaction shown on the right.

Oxaloacetate is a product of oxidation of malate - the enzyme malate dehydrogenase catalyses the following reaction shown on the right.

What is the relationship between oxaloacetate and aspartate?

Oxaloacetate is the keto-acid corresponding to aspartate.

Aspartate transaminase catalyses the reaction shown on the left.

In a further series of experiments, isolated mitochondria were incubated in an oxygen electrode with pyruvate, oxaloacetate, malate or aspartate, and the following results were obtained:

We already know from the exercise on Overheating after overdosing on E - and slimming by taking dinitrophenol that malate can enter mitochondria and be oxidised, linked to the electron transport chain and phosphorylation of ADP to ATP, and previous results in this exercise have shown that pyruvate also enter the mitochondria. These results conform that, and also show that aspartate can cross the mitochondrial membrane. However, oxaloacetate cannot cross the mitochondrial membrane - there is no consumption of oxygen when the mitochondria are provided with oxaloacetate

In the next series of experiments isolated heart muscle cells were incubated with lactate in the oxygen electrode, with and without the addition of methylene aspartate, a specific inhibitor of aspartate transaminase.

The following results were obtained:

Transamination of aspartate / oxaloacetate seems to be essential for the oxidation of lactate, and therefore presumably for the transfer into the mitochondria of the reducing equivalents on the NADH formed by oxidation of lactate.

From the information you have deduced to date, can you propose a pathway for transfer of reducing equivalents from cytosolic NADH into the mitochondrion?

We know that oxaloacetate can be reduced to malate in the cytosol, and that malate enters the mitochondria.

We also know that aspartate can cross the mitochondrial membrane, while oxaloacetate cannot, and that transamination of aspartate / oxaloacetate is essential for transfer of reducing equivalents into the mitochondria.

Therefore it seems likely that the sequence of events is as shown on the right. This pathway is known as the malate-aspartate shuttle, because it involves exchange of mitochondrial aspartate for cytosolic malate:

• aspartate from the mitochondria crosses to the cytosol.
• aspartate is transaminated (at the expense of ketoglutarate) to yield oxaloacetate
• oxaloacetate is reduced to malate by cytosolic NADH
• malate enters the mitochondrion and is oxidised to oxaloacetate, yielding NADH inside the mitochondrion
• oxaloacetate is transaminated to aspartate, which leaves the mitochondrion.

Overall there is transport into the mitochondrion of reducing equivalents from cytosolic NADH to yield NADH inside the mitochondrion.

It is also known, although not shown in the diagram above, that aspartate can only leave the mitochondria in exchange for glutamate entering, so there is a continuing supply of glutamate in the mitochondria to act as the amino donor to oxaloacetate. In turn, the ketoglutarate formed by transamination of glutamate leaves the mitochondria in exchange for malate entering.

Since the mitochondrial glutamate-aspartate exchanger is a key pert of the malate-aspartate shuttle, you would expect some-one who lacked the protein, or in whom it had reduced activity, to be unable to transfer reducing equivalents from cytosolic NADH into the mitochondrion. This would lead to lactic acidosis, as muscle and other tissues would be unable to re-oxidise cytosolic NADH except for forming lactate. This is the same as occurs in muscle in vigorous exercise, when oxygen is limiting - see the exercise on Breathless after sprinting. The liver would be able to utilise the lactate for gluconeogenesis as normal, since gluconeogenesis from pyruvate will re-oxidise the NADH produced by lactate dehydrogenase.

Surprisingly, people who lack the mitochondrial glutamate-aspartate exchanger protein do not develop lactic acidosis. They do develop citrullinaemia, and sometimes also hyperammonaemia, as a result of inadequate export of aspartate from mitochondria to the cytosol for urea synthesis. You will consider this in a later exercise.

There must be an alternative pathway that permits transfer of the reducing equivalents form cytosolic NADH into mitochondria.

Note that to date the experiments we have considered in this exercise have used heart muscle cells or isolated mitochondria. Heart is unusual among tissues in that it takes up lactate from the circulation as a metabolic fuel. Liver takes up lactate mainly for gluconeogenesis, although, as we saw in the exercise on Breathless after sprinting, liver can also metabolise some of that lactate as metabolic fuel to provide the ATP needed for gluconeogenesis. Muscle puts out lactate under conditions of maximum exertion, but does not take it up from the bloodstream.

You will see in a later exercise that the esterification of fatty acids to form diacylglycerol phosphate occurs on the outer mitochondrial membrane and the endoplasmic reticulum, and that the final step in triacylglycerol synthesis, the reaction between diacylglycerol phosphate and fatty acyl CoA occurs on the endoplasmic reticulum. The starting point for triacylglycerol synthesis is glycerol 3-phosphate, which is synthesised in the cytosol from dihydroxyacetone phosphate, a glycolytic intermediate, by glycerol 3-phosphate dehydrogenase, as shown on the right.

What is surprising is that there is another isoenzyme of glycerol 3-phosphate dehydrogenase in the mitochondrial matrix, which uses FADH2 rather than NADH.

Assuming that the reaction of glycerol 3-phosphate can enter mitochondria, and that the reaction of glycerol 3-phosphate dehydrogenase is reversible, there would be consumption of oxygen and a P:O ratio of ~1.5 as a result of re-oxidation of the FADH2 formed by oxidising glycerol 3-phosphate to dihydroxyacetone phosphate.

Isolated mitochondria were incubated with [14C]glycerol 3-phosphate, and the following results were obtained (figures show dpm in each metabolite):

Glycerol 3-phosphate obviously can enter mitochondria, and be oxidised to dihydroxyacetone phosphate. However, it does not undergo complete oxidation to carbon dioxide and water.

Although there is some label in dihydroxyacetone phosphate inside the mitochondria, most of the label is found in the supernatant. This cannot be the result of the action of cytosolic glycerol 3-phosphate dehydrogenase, since we are working with isolated mitochondria that have been washed in preparation and there are no cytosolic enzymes present. Therefore it seems most likely that dihydroxyacetone phosphate formed in the mitochondria has been exported out into the incubation medium.

It is possible that dihydroxyacetone phosphate is reduced to glycerol 3-phosphate in the cytosol, which enters the mitochondrion and is oxidised to dihydroxyacetone phosphate at the expense of FAD being reduced to FADH2. The dihydroxyacetone phosphate then leaves the mitochondrion again.

You could incubate isolated mitochondria in an oxygen electrode with NADH, ADP and dihydroxyacetone phosphate. You would also have to add purified cytosolic glycerol 3-phosphate dehydrogenase.

If your hypothesis is correct then you would see consumption of oxygen, with as P:O ratio ~ 1.5.

No consumption of oxygen, because none of the extra-mitochondrial dihydroxyacetone phosphate will be reduced to glycerol 3-phosphate dehydrogenase.

We now have two shuttles for transferring reducing equivalents from cytosolic NADH into the mitochondrion: the glycerol 3-phosphate shuttle shown above and the malate-aspartate shuttle shown below:

The malate-aspartate shuttle leads to production of NADH inside the mitochondrion; re-oxidation of NADH in the mitochondrial electron transport chain results in the formation of ~2.5 x ATP.

The glycerol 3-phosphate shuttle leads to the production of FADH2 inside the mitochondrion (but from cytosolic NADH). Re-oxidation of FADH2 in the mitochondrial electron transport chain results in the formation of ~1.5 x ATP.

This means that there is a loss of ~1 x ATP for each cytosolic NADH re-oxidised via the glycerol 3-phosphate rather than the malate-aspartate shuttle.

However, the capacity of the malate-aspartate shuttle is limited, and it cannot operate fast enough to meet the need to re-oxidise all of the NADH formed in tissues with a high rate of glycolysis, so there is a need for the less efficient, but faster, glycerol 3-phosphate shuttle.

Key points from this exercise:

• NADH formed in the cytosol cannot enter mitochondria to undergo re-oxidation in the electron transport chain.
• There are two substrate shuttles that transfer reducing equivalents from cytosolic NADH into the mitochondrion:
• The malate-aspartate shuttle, in which :
  • aspartate from the mitochondria crosses to the cytosol.
  • aspartate is transaminated (at the expense of ketoglutarate) to yield oxaloacetate
  • oxaloacetate is reduced to malate by cytosolic NADH
  • malate enters the mitochondrion and is oxidised to oxaloacetate, yielding NADH inside the mitochondrion
  • oxaloacetate is transaminated to aspartate, which leaves the mitochondrion.
• The glycerol phosphate shuttle, in which
  • dihydroxyacetone phosphate (a glycolytic intermediate) is reduced to glycerol 3-phosphate in the cytosol, and enters the mitochondria
  • in the mitochondrion glycerol 3-phosphate is oxidised to dihydroxyacetone phosphate, at the expense of reducing FAD to FADH2. Dihydroxyacetone phosphate then leaves the mitochondrion, back into the cytosol.
• The glycerol phosphate shuttle is energetically less efficient than the malate-aspartate shuttle, because the FADH2 formed inside the mitochondrion yields only ~1.5 x ATP in the electron transport chain, while the NADH formed in the malate-aspartate shuttle yields ~2.5 x ATP in the electron transport chain.
• The capacity of the malate-aspartate shuttle is limited, and it cannot operate fast enough to meet the need to re-oxidise all of the NADH formed in tissues with a high rate of glycolysis, so there is a need for the less efficient, but faster, glycerol 3-phosphate shuttle.