# Feeding the Axon: Glia as Metabolic Partners

> Whoever wraps an axon helps feed it, and the firmest evidence is for oligodendrocytes, the myelinating glia of the brain and spinal cord. An axon can reach a metre from its cell body, spends energy without pause to pump ions, and is mostly covered by myelin that limits its access to the fluid around it. This lesson asks how such an axon is fed; Schwann cells in nerves have had fewer direct tests. The lesson follows two experiments, one that switches off these glia's respiration and one that removes the transporter they are thought to export fuel through, then weighs the evidence for and against the idea that astrocytes hand lactate to neurons, which is still debated. Afterwards you can design and read the experiments that test whether a glial cell feeds an axon.

Course: Developmental Neuroscience (Glia). Level: advanced university and graduate. About 28 minutes.
Lesson page (link this one when citing): https://stemhop.com/lessons/glia-metabolic-partners/

Builds on: Myelination: Wrapping the Axon So the Signal Jumps · Where Glia Come From

## What you'll learn

- Explain why axons have a supply problem, using the cost of ion pumping, the distance from the cell body and the coverage of the axon by myelin, and state the main answer: glia that wrap axons help supply them.
- State the astrocyte-neuron lactate shuttle hypothesis and weigh one line of evidence for it against one line of evidence against it.
- Describe the evidence that oligodendrocytes supply lactate or pyruvate to axons through monocarboxylate transporters, and predict what happens when that supply fails.
- Discuss, with appropriate hedges, how metabolic changes in glia after nerve injury could help or harm the axons they surround.

**Your brain is about 2 percent of your body weight and uses about 20 percent of the energy your body spends at rest.** Much of that energy is spent in axons and synapses, and the axon is the part of a neuron farthest from where most of its supplies are made. A motor axon running from your spinal cord to your foot is about a metre long. Along most of that length it is wrapped in myelin, a membrane coat made by glial cells that limits the axon’s direct access to the fluid outside. So how is the far end of an axon fed?

> **The one idea to carry through this lesson.** Whoever wraps the axon helps feed it. Myelin covers almost all of a long, fully myelinated axon, and along each myelinated stretch the glial cell that made the myelin is the cell pressed against it. That cell can live on glycolysis (shown for oligodendrocytes that have already made their myelin, not for Schwann cells that are still myelinating), and the evidence fits it passing lactate or pyruvate across to the axon. When it loses the transporter thought to carry that fuel out, axons in ageing mice degenerate.

**What the rule covers.**

- **Which cells.** Myelinating glia, most clearly oligodendrocytes, which myelinate axons in the brain and spinal cord. Schwann cells, which myelinate axons in peripheral nerves such as the one to your foot, have had fewer direct tests. Whether astrocytes feed neurons at synapses is a separate, contested question.
- **Which job.** Long-term upkeep of the axon, because that is what the experiments test. Glial fuel for each action potential as it fires has been proposed; the experiments do not measure it and do not rule it out.
- **Help, not sole supply.** Axons still take up glucose and still have their own mitochondria.

This is **metabolic support**: fuel from the cell alongside the axon, not the target-derived survival factor of [the neurotrophic hypothesis](https://stemhop.com/lessons/neurotrophic-hypothesis/).

## An axon has a supply problem

Three facts make the problem: what the axon spends, how far it is from the cell body, and how much of it is covered.

**Cost.** Every action potential lets sodium in and potassium out, and excitatory synapses do much the same at a smaller scale. The sodium-potassium pump (Na+/K+-ATPase) restores the gradients, spending one ATP to move three sodium ions out and two potassium ions in. Energy budgets for rodent grey matter, beginning with Attwell and Laughlin in 2001 and revised since, find that most of the energy used for signalling goes to these pumps. The pumps never get a day off.

**Distance.** A neuron’s three parts have different supply problems.

- **The soma** has dense mitochondria, makes most of the cell’s proteins and has its biosynthetic machinery close at hand.
- **The axon** is long and thin. The enzymes and mitochondria it uses are mostly made in the soma and carried down by motor proteins that also spend ATP.
- **The synapse** has a high, fluctuating demand from recycling vesicles and pumping ions after each release.

**Predict.** The soma is rich in mitochondria. Why can it not simply make ATP and let it diffuse down the axon? Guess how long ATP would take to diffuse one micrometre, one millimetre, one centimetre and one metre.

> **Check your prediction.** Diffusion time grows with the square of the distance: double the distance and the time goes up four times. These rough estimates use time ≈ L²/2D, with a diffusion coefficient D for ATP in cytoplasm of about 150 to 300 µm² per second. They ignore that ATP is used up along the way, which makes the far end harder still to reach.
>
> | Distance | Time for ATP to diffuse that far |
> | --- | --- |
> | 1 micrometre | a few milliseconds |
> | 1 millimetre | about half an hour to an hour |
> | 1 centimetre | about 2 to 4 days |
> | 1 metre | about 50 to 100 years |
>
> So ATP spreads easily within one short stretch of axon; the problem is the distance from the soma to the far end, not that ATP diffuses poorly. An axon’s ATP has to be made locally, by axonal mitochondria and axonal glycolysis.

Many proteins come from the soma by the slow form of axonal transport, a few millimetres a day, so they take months to more than a year to cover a metre.

**Cover.** In a myelinated axon, compact myelin covers the long stretches between the nodes of Ranvier, the short gaps where the axon membrane is bare. A node is about a micrometre long. The myelinated stretch between two nodes, the internode, is about 50 to 100 micrometres in many CNS axons and about a millimetre in a large peripheral axon. So along a continuously myelinated stretch, myelin covers about 98 to 99 percent of a CNS axon’s length and about 99.9 percent of a large peripheral one. Real axons vary, and some CNS axons are myelinated only in part.

That cover is not a seal. A thin space remains between the axon and its myelin, but the axon membrane meets the extracellular fluid mainly at the nodes, so the insulation that speeds conduction also limits the axon’s direct access to glucose. Whether the glucose that enters at the nodes and along the space under the myelin would be enough on its own is not known. So the geometry makes glial supply plausible; it does not prove the axon needs it. The genetic experiments below test that.

**Who is in contact.** Each segment of myelin is a sheet of one glial cell’s membrane, wound around the axon, with the cell’s cytoplasm still connected to it. In the CNS one oligodendrocyte makes many segments, on many different axons. In a peripheral nerve each myelinating Schwann cell makes one. So along each myelinated stretch, the cell pressed against the axon is the cell that made its myelin, which makes myelinating glia candidate suppliers.

*Figure (AI-generated illustration).* Each part of a neuron has a different supply problem: the soma has mitochondria close at hand, the long axon must make ATP locally, and the synapse has a high, fluctuating demand. Compact myelin limits direct access to surrounding fluid mainly to the nodes, but a thin extracellular space remains beneath myelin. Myelinating glial processes lie close to the axon at individual segments, making these cells candidate suppliers. Image: a neuron with mitochondria in its soma, a long myelinated axon containing mitochondria, and a synaptic terminal. Separate gold myelinating glial cells connect to individual myelin segments. An enlarged node shows exposed axonal membrane between adjacent sheaths and a thin extracellular space beneath myelin; direct fluid access is mainly at nodes, not exclusively there.

## Just enough chemistry

[Glycolysis](https://stemhop.com/lessons/glycolysis/) splits one glucose into two pyruvate in the cytosol, with a net gain of two ATP, and turns NAD+ into NADH along the way. If the cell is to keep going, it must turn that NADH back into NAD+. Pyruvate then has two routes.

1. **Into the mitochondrion.** Pyruvate goes into the citric acid cycle and [oxidative phosphorylation](https://stemhop.com/lessons/oxidative-phosphorylation/), which yields far more ATP. The NADH made in glycolysis passes its electrons into the mitochondria through shuttles such as the malate-aspartate shuttle.
2. **To lactate.** Lactate dehydrogenase reduces pyruvate to lactate, regenerating NAD+ in the cytosol. Glycolysis can then keep running without mitochondria.

The lactate dehydrogenase reaction runs in both directions. A cell that takes up lactate can turn it back into pyruvate and burn it in its own mitochondria. Lactate and pyruvate cross membranes on monocarboxylate transporters (MCTs).

Now the arithmetic that makes lactate worth passing on. Burned all the way, one glucose can yield about 30 to 32 ATP, depending on which shuttle carries the NADH from glycolysis. A cell that stops at lactate keeps 2 ATP and passes on the rest. The two lactates still carry about 28 to 30 of those 30 to 32 ATP, more than nine-tenths. Lactate is a portable fuel.

## Myelin that looked fine, axons that did not

**Film transcript: A Glial Cell That Stops Respiring**

### Film transcript: A Glial Cell That Stops Respiring

If a glial cell stopped burning fuel in its mitochondria, would the axon it wraps starve? Start with a puzzle. Mice lacking the myelin protein PLP, or the enzyme CNP, made close to normal myelin. Yet as they aged, their axons swelled and degenerated. The glial cell seems to give the axon more than insulation. Fuel is one candidate. Glycolysis turns glucose into pyruvate. Pyruvate can enter the mitochondrion, for far more ATP.

Or it can become lactate. That turns NADH back into NAD+, +, so glycolysis keeps running without mitochondria. Another cell can turn lactate back and burn it. A cell that stops at lactate keeps only 2 ATP. Its lactates carry more than nine-tenths of what the glucose could yield. Now the test. One laboratory deleted the gene Cox10 in myelinating cells. Without it, they can make ATP only by glycolysis. If oligodendrocytes must respire to keep axons healthy, what should happen in the adult brain over the following months?

If they had to respire, myelin and axons should fail. Instead, the oligodendrocytes survived. Myelin and axons were largely preserved for months. Oligodendrocytes that have made their myelin can live on glycolysis alone. Pyruvate and lactate they cannot burn must leave or be left over. Leaving would fit glia feeding axons, but this does not show it. Peripheral nerves differed. Schwann cells without Cox10 survived but failed to myelinate properly. The mice developed a severe neuropathy. Those cells were still building myelin, a costly stage. They survived, but needed respiration to myelinate. The stage may explain part of the difference. One more clue. A scan that measures brain chemicals found extra lactate in the mutants, but only under anaesthesia.

Within an hour after the anaesthetic ended, it fell below detection. Why would it show up only under anaesthesia? One reading: glycolytic glia make more lactate than they burn. Awake, something uses it as fast as it is made. Under anaesthesia, that use slows. The axons are an obvious candidate user. But the experiment never traced where the lactate went. So switch off the glial cell's respiration, and the axon in the brain is spared, over the months studied.

From the film

- Mice lacking PLP or CNP made close to normal myelin yet had axons that swelled and degenerated with age, so glia seem to give axons more than insulation.
- Oligodendrocytes that have made their myelin can live on glycolysis alone, and adult brain axons were spared over the months studied.
- The lactate seen under anesthesia fits glia making more than they burn, but the experiment never traced where it went.

The case for myelinating glia began with a puzzle from [myelination](https://stemhop.com/lessons/myelination/) research. Mice lacking the myelin protein PLP (Griffiths and colleagues, 1998), and later mice lacking the myelin enzyme CNP (Lappe-Siefke and colleagues, 2003), made myelin that looked close to normal under the [electron microscope](https://stemhop.com/methods/electron-microscopy/). Yet as the mice aged, their axons swelled and degenerated.

Normal-looking myelin does not prove the insulation worked normally, but it makes failed insulation an unlikely full explanation. The simplest reading is that the glial cell gives the axon something besides insulation. Fuel is one candidate, and two experiments test it.

### Switch off the glial cell’s respiration

Ursula Fünfschilling, Klaus-Armin Nave and colleagues deleted *Cox10* only in myelinating cells with a [conditional knockout](https://stemhop.com/methods/cre-lox/): the gene is cut out only in cells that make the enzyme Cre, here a Cre line active in myelinating cells. Cox10 is needed to assemble complex IV, the last step of the respiratory chain. Once the cells’ existing complex IV wears out, they can make ATP only by glycolysis.

**Predict.** Suppose oligodendrocytes must respire to keep their axons healthy. What would you expect to see in the adult brain over the following months? And what would you expect instead if they can live on glycolysis and pass its products to the axon?

> **Check your prediction.** If the glia had to respire, they or the axons would fail: over months, thinning or loss of myelin, swollen and degenerating axons, and perhaps the death of neurons whose axons run through white matter.
>
> The 2012 result was the reverse. In the adult CNS, once myelination was complete, the oligodendrocytes survived, and myelin and axons were largely preserved over the months the mice were followed. Oligodendrocytes that have made their myelin can live on glycolysis alone, and the axons did not need their oligodendrocytes to respire. With complex IV gone, the glia cannot burn pyruvate in their own mitochondria, so any pyruvate and lactate they make by glycolysis would have to leave the cell or be left over. Leaving would fit glia feeding axons, but the result does not by itself show the feeding.
>
> The peripheral nerves told a different story, and it sets a limit on the rule. Schwann cells lacking Cox10 failed to myelinate properly, and the mice developed a severe neuropathy. The Schwann cells survived, but they lost respiration while still dividing and myelinating, an energy-costly stage, and could not myelinate without it. Part of the difference may be the stage of myelination, not only the cell type. So the rule, as tested here, describes oligodendrocytes that have already made their myelin.

The mutant brains held one more clue. Magnetic resonance spectroscopy, which measures chemicals such as lactate in the living brain, found more lactate in mutant brains than in controls, but only while the mice were under isoflurane, an inhaled anaesthetic. Within an hour after the anaesthetic ended, the lactate fell below detection.

**Predict.** Why would extra lactate show up only under anaesthesia?

> **Check your prediction.** One reading: glia that rely on glycolysis make more lactate than they can burn. In an awake animal, something uses that lactate as fast as it is made; under anaesthesia, that use slows and the excess builds up. The axons the glia wrap are an obvious candidate user, but the experiment never traced where the lactate went.

## Cut the export line

**Film transcript: Cut the Export Line**

### Film transcript: Cut the Export Line

If glia feed axons, the fuel must leave the glial cell. What happens if you block the exit? Lactate leaves on transporters. In 2012, Jeffrey Rothstein's group showed that one, MCT1, is concentrated in oligodendrocytes. In spinal cord slices, they withheld glucose briefly. Some slices also got a drug that blocks MCT1. Some of those also got lactate. What happens to the motor neurons in each condition?

With glucose withheld, they survived. Add the blocker, and they died. Add lactate with the blocker still on, and they survived. That is a rescue. The best reading is that the neurons lacked lactate, which glia are thought to supply. Other drug effects are not ruled out. In living mice, lowering MCT1 in the spinal cord made axons swell and degenerate. Motor neurons were lost. A cleaner test came in 2021. Deleting MCT1 only in oligodendrocyte lineage cells left young mice normal. As the mice aged, axons degenerated and myelin thinned. Side by side: with glial respiration switched off, adult brain axons were spared over the months studied. Remove MCT1, the proposed export route, and axons degenerate with age.

One caution: the two sets of mice were not followed over matched ages, and myelin also thinned late. So the picture that fits is this. Oligodendrocytes run glycolysis and export lactate or pyruvate through MCT1. The axon turns any lactate back into pyruvate and burns it in its own mitochondria. Still unknown: the exact route, and how much of an axon's ATP this supplies. These experiments infer the transfer; none of them traces it.

Astrocytes are a second, contested candidate. One sensor study in mouse cortex reported a lactate gradient from astrocytes to neurons. Another saw stimulated neurons raise their own glycolysis, with no sign of lactate import. So the astrocyte shuttle remains a hypothesis with partial support. For the axon under myelin, the evidence fits one picture. Whoever wraps the axon helps feed it.

From the film

- Blocking MCT1 killed motor neurons in glucose-starved spinal cord slices, and adding lactate rescued them.
- Deleting MCT1 only in oligodendrocyte lineage cells left young mice normal, but axons degenerated and myelin thinned as they aged.
- These experiments infer the transfer of fuel and none traces it, and the astrocyte shuttle remains a hypothesis with partial support.

If glycolytic glia feed axons, the product has to leave the glial cell on a monocarboxylate transporter. In 2012 Youngjin Lee, Jeffrey Rothstein and colleagues used reporter mice, engineered so that cells making monocarboxylate transporter 1 (MCT1) carry a visible label, to show that one of them, MCT1, is concentrated in oligodendrocytes. They then lowered MCT1 and watched the neurons.

**Predict.** In cultured slices of spinal cord, the group withheld glucose for a short time, a stress the slices normally survive, so that any need for another fuel would show. In some slices they also blocked MCT1 with an inhibitor, and to some of those they added lactate. If oligodendrocytes feed neurons through MCT1, what should happen to the motor neurons in each condition?

> **Check your prediction.** | Slice culture condition | Motor neurons |
> | --- | --- |
> | Glucose withheld briefly | survive |
> | Glucose withheld, MCT1 blocked | die |
> | Glucose withheld, MCT1 blocked, lactate added | survive |
>
> Adding back the missing product while the transporter is still blocked is a **rescue**. In the terms of [the neurotrophic hypothesis lesson](https://stemhop.com/lessons/neurotrophic-hypothesis/), blocking MCT1 asks whether it is necessary, and adding lactate asks whether lactate is enough to replace what the block took away. Because lactate rescued the neurons, the best reading is that they were missing lactate, a monocarboxylate that MCT1 carries and that glia are thought to supply. That reading does not rule out every other effect of the drug. The drug also acts on every cell in the slice (and inhibitors of this kind can also reach MCT2 in neurons), so it does not show which cell’s transporter mattered or the direction of transfer.

Lowering MCT1 itself in cultured slices of spinal cord also killed motor neurons. When the group lowered MCT1 in the spinal cord of living mice, axons swelled and degenerated, and motor neurons were lost. That [knockdown](https://stemhop.com/methods/rnai-knockdown/), too, was not limited to oligodendrocytes. They also found less MCT1 in tissue from people with amyotrophic lateral sclerosis (ALS) and from an ALS mouse model.

A cleaner test came later from the same group. Philips and colleagues (2021) deleted MCT1 only in cells of the oligodendrocyte lineage. Young mice were normal. Only as the mice aged did axons degenerate, and myelin thinned.

### What the experiments show together

Switch off oligodendrocytes’ respiration, and adult CNS axons are spared over the months followed. Remove the transporter they are thought to export fuel through, and axons degenerate as the mice age. One caution: the two sets of mice were not followed over matched ages, and MCT1 loss also thinned myelin late.

Put together, the picture that fits is this. Oligodendrocytes take up glucose, run glycolysis, and export lactate or pyruvate through MCT1. The axon takes it in, turns any lactate back into pyruvate, and burns the pyruvate in its own mitochondria.

*Figure (AI-generated illustration).* The picture that fits the experiments: an oligodendrocyte runs glycolysis and exports lactate or pyruvate through MCT1, and the axon burns it in its own mitochondria. The exact route from glial cytoplasm to axon, and how much of the axon's ATP depends on it, are still unknown. Image: an oligodendrocyte takes up glucose from a blood vessel and makes lactate or pyruvate by glycolysis; the fuel travels down its process to the myelin around an axon, leaves through the transporter MCT1, and crosses into the axon by a dashed route marked with a question mark, where it enters a mitochondrion.

Two things are still unknown: the exact route from glial cytoplasm to axon, and how much of an axon’s ATP, in any tract of a living animal, depends on this supply.

## Astrocytes: a contested second candidate

Astrocytes wrap synapses, take glutamate back up after release, and put end-feet on blood vessels, so they sit between the blood’s glucose and the neuron.

In 1994 Luc Pellerin and Pierre Magistretti gave glutamate to cultured mouse astrocytes. The cells took up more [2-deoxyglucose](https://stemhop.com/methods/pet-imaging/), a glucose analogue, and released more lactate. D-aspartate, an analogue taken up by the same transporters, had the same effect, and the authors concluded that receptor activation was not responsible. Blocking the sodium pump with ouabain blocked the response. Their explanation had three steps. Glutamate transporters bring sodium in with each glutamate. The sodium pump has to remove that sodium. The pump’s demand for ATP drives glycolysis. From this they proposed the astrocyte-neuron lactate shuttle: an active synapse makes its astrocyte burn glucose to lactate, and the lactate goes to the neuron as fuel.

The model is attractive because it couples supply to use and would explain why glucose uptake rises where synapses are active. Some later in vivo work supports it. In 2016 Philipp Mächler, Bruno Weber and colleagues used a fluorescent lactate sensor in mouse cortex and reported a lactate gradient running from astrocytes to neurons. A gradient fits transfer but does not by itself measure net flow.

The objections are serious. Critics, including Gerald Dienel, have argued that culture results do not transfer cleanly to the brain, and that because lactate is in near-equilibrium with pyruvate through lactate dehydrogenase, net flow between cells is hard to infer from where the enzymes and transporters are. In 2017 Gary Yellen’s laboratory imaged sensors of NADH and other metabolites in single neurons of mouse brain slices and live mice. Stimulation made neurons raise their own glycolysis, and the group saw no sign of lactate import during the stimulus.

Both sides now agree that neurons can take up and use glucose directly; they disagree about how much of a neuron’s fuel comes through astrocytes, when, and in which brain region. So treat the shuttle as a hypothesis with partial support. The answer may differ between resting and active states, between cortex and other regions, and between preparations.

*Figure (AI-generated illustration).* Two readings of how an active synapse is fuelled. In the proposed astrocyte-neuron lactate shuttle, glutamate uptake drives the astrocyte to burn glucose to lactate, which it passes to the neuron; in the alternative, stimulated neurons raise their own glycolysis. Neurons can use glucose directly; how much of their fuel comes through astrocytes is still contested. Image: two panels of a synapse beside an astrocyte whose end-foot touches a blood vessel. Left, the proposed lactate shuttle: the astrocyte takes up glutamate from the synapse, burns glucose to lactate, and a dashed arrow with a question mark passes lactate to the neuron. Right, the neuron takes up glucose from the blood directly and raises its own glycolysis, with no arrow from the astrocyte.

## After injury: when glial metabolism changes

After a peripheral nerve is cut, the pieces of axon cut off from the cell body degenerate, a process covered in [injury and regeneration](https://stemhop.com/lessons/injury-and-regeneration/). Schwann cells respond to the injury with metabolic programs of their own.

**Predict.** After injury, mouse Schwann cells shift toward glycolysis. This goes beyond the rule, which is about intact axons. Would you expect the shift to help or harm the cut-off piece of axon, and what would have to be true for it to help?

> **Check your prediction.** If the extra lactate or pyruvate leaves the Schwann cell and the injured axon uses it, the shift should help, at least for a while; more glycolysis alone does not guarantee that. In 2020 Elisabetta Babetto, Keit Men Wong and Bogdan Beirowski, who reported the shift, found that it slows the degeneration of the injured axons.

The big question is how these metabolic changes in glia might be beneficial or detrimental to neuronal survival and function. A glial cell that makes more lactate may tide a stressed axon over, and a glial cell that makes lipids may supply membrane for regrowth. Glial metabolism can also go wrong in ways that harm axons (see Going deeper). Whether the net effect on the axon is a benefit or a harm probably depends on the cell, the injury and the timing, and that is still being measured.

## Putting it together

| Experiment | Animal or preparation | Observed result | What it supports | What it does not establish |
| --- | --- | --- | --- | --- |
| PLP-null, CNP-null | mouse | myelin close to normal; axons swell and degenerate with age | glia do more for axons than insulate them | what else they provide |
| Cox10 deleted in myelinating cells | adult mouse CNS | oligodendrocytes survive; myelin and axons largely preserved; extra lactate only under anaesthesia | oligodendrocytes can live on glycolysis; axons do not need glial respiration | that the glia feed the axons |
| Cox10 deleted in myelinating cells | mouse peripheral nerve | Schwann cells survive but fail to myelinate properly; severe neuropathy | developing Schwann cells need respiration to myelinate | whether cell type or stage explains the difference |
| MCT1 blocked, glucose withheld, with or without lactate | spinal cord slices | motor neurons die; lactate prevents the loss | the neurons were missing a monocarboxylate that MCT1 carries | that the drug had no other effect; which cell’s MCT1 mattered |
| MCT1 lowered | living mouse spinal cord | axons degenerate; motor neurons lost | MCT1 is needed for axons and neurons | which cell’s MCT1 mattered; the route of transfer |
| MCT1 deleted in oligodendrocyte lineage | mouse, young and old | normal when young; axons degenerate and myelin thins with age | oligodendroglial MCT1 is needed for long-term integrity | that export of fuel is the reason; how much of the axon’s ATP it supplies; whether thinner myelin adds to the damage |
| Glutamate given to astrocytes (1994) | mouse astrocyte culture | more glucose uptake and lactate release | a mechanism for the lactate shuttle | that it happens in the brain |
| Lactate sensor (2016) | mouse cortex | lactate gradient from astrocytes to neurons | a lactate gradient consistent with astrocyte-to-neuron transfer | net flux, or how much of a neuron’s fuel it supplies |
| NADH sensors (2017) | mouse slices and live mice | stimulated neurons raise their own glycolysis; no sign of lactate import | neurons can raise their own glycolysis when stimulated | that neurons never use astrocyte lactate |
| Glycolytic shift after injury (2020) | injured mouse nerve | Schwann cells shift toward glycolysis; injured axons degenerate more slowly | glial metabolism can help an injured axon | how the shift is controlled, or whether it is always helpful |

The guiding idea survives in its “helps” form only, with oligodendrocytes as the clearest case: the glia are not shown to be the axon’s only supply, and the transfer has been inferred, not traced.

### Design problem: trace the transfer

You want to show that lactate made in oligodendrocytes is burned in the axon, not just that it is needed. Design the experiment.

> **One way to do it.** The Cox10 experiment showed that glial respiration is not necessary for the axon, at least over the months studied. The MCT1 experiments showed that glial MCT1 is necessary as mice age. Neither follows the fuel. Two approaches, which can be combined:
>
> - **Follow a labelled fuel.** Get the label into the glial cell first: for example, give labelled glucose to a compartment holding oligodendrocyte cell bodies whose processes myelinate axons in a second compartment. Measure labelled lactate released by the glia and labelled pyruvate and citric acid cycle products in the axon. Control for tracer leaking into the axon compartment, for the axon taking up glucose directly, and for loss of glial MCT1, which should lower the axon’s label. Such a system is hard to build.
> - **Watch the axon’s lactate.** Express a lactate sensor in neurons and image an axon in white matter while glial MCT1 is deleted. A fall would fit a lost glial supply, but the axon could make up the loss with its own glycolysis, so a steady signal would not rule one out. An NADH sensor alone cannot tell the sources apart, since glycolysis and lactate conversion both make NADH.
>
> Either way, the hard part is the axon’s own glucose uptake, which would otherwise look like a glial supply.

## Going deeper

### What steers the glycolytic shift after injury

The signals that steer such shifts are ones you know from [mTOR and AMPK](https://stemhop.com/lessons/mtor-and-ampk/). AMP-activated protein kinase (AMPK) senses a falling energy charge and turns on catabolism and autophagy. mTOR drives growth and lipid synthesis when nutrients and growth factors are plentiful. HIF1-alpha, a transcription factor stabilized at low oxygen, turns on glycolytic genes. In injured mouse nerve, mTORC1 and HIF1-alpha were reported to drive the glycolytic shift in Schwann cells; the role of AMPK there is less clear, and how they divide the work in a repairing nerve is not settled.

### Satellite glia and lipids for regrowth

Satellite glia, which surround neuron cell bodies in sensory ganglia, change too. In 2020 the laboratory of Valeria Cavalli reported that after peripheral nerve injury in mice, satellite glia turn on genes for fatty acid synthesis, and that this program helps the sensory axons regrow.

### When glial metabolism harms axons

In 2011 the laboratory of Jeffrey Milbrandt reported that mice whose Schwann cells cannot maintain their mitochondrial DNA develop a neuropathy in which axons degenerate. That looks at first like a missing fuel. A follow-up from the same laboratory (2013) reported that the metabolically abnormal Schwann cells release acylcarnitines, lipid metabolites that damage axons, and proposed this as one contributor to the neuropathy. If that account holds, glial metabolism here harms axons, not only fails to supply them.

### Astrocytes: glycogen and transporters

Astrocytes do take part in brain energy supply, for example by holding glycogen, a stored form of glucose that in the brain is held mostly in astrocytes. In the shuttle model, astrocytes were reported to export lactate mainly through monocarboxylate transporter 1 (MCT1) and MCT4, and neurons to take it up through MCT2.

### Energy budgets in white matter

The budgets in the main text are for grey matter. Harris and Attwell (2012) estimated that white matter uses less energy per volume than grey matter, with a larger share going to upkeep that is not signalling. A budget like this does not say which demands glial fuel meets.

### Background numbers

In rat liver the free NAD+-to-NADH ratio in the cytosol has been estimated at several hundred to one, which keeps glycolysis running in the forward direction; values in neurons depend on the cell and the measurement.

How many glial cells are there to do this work? Older textbooks said glia outnumber neurons by ten to one. Counts of whole human brains by the [isotropic fractionator](https://stemhop.com/methods/cell-counting-in-tissue/) method (Azevedo and colleagues, 2009) found roughly as many non-neuronal cells, most of them glia, as neurons overall, with glia outnumbering neurons in the cerebral cortex and far outnumbered by neurons in the cerebellum.

### Possible costs of glial glycolysis (hypotheses)

Two possible costs of a large glial glycolytic response have been proposed, and neither is established in injured nerve. Lactate comes with protons, so a large glycolytic output can acidify the tissue, and a glial cell that diverts its resources to its own repair program may give less to the axon.

---

From STEMhop (https://stemhop.com), free lessons by Christopher Deppmann. Lesson text © the authors. Practice questions: https://stemhop.com/lessons/glia-metabolic-partners/
