The strict dependence of the (human) brain on glucose has always been puzzling to me. While ketones can substitute for a portion of the brain's energy needs, it cannot substitute completely: blood glucose levels below 2--3mM somewhere causes serious neurological problems and can lead to unconsciousness.

Other body tissues are not strictly dependent on glucose, but can oxidize amino acids as well, which is a good backup solution since there is always lots of protein around. But not the brain.

In terms of evolution, this strict glucose dependence must be a major drawback --- falling unconscious just because you don't get enough sugar is probably a bad thing out in the wilderness, when the lions are after you ... So there must be a very important reason.

So why is the brain so dependent on glucose?

  • 4
    $\begingroup$ Everything is bad when the lions are after you ;) $\endgroup$
    – canadianer
    May 11, 2015 at 6:33
  • $\begingroup$ Actually you can get sugar everywhere where there are plants. $\endgroup$
    – Anixx
    May 11, 2015 at 9:25
  • $\begingroup$ I don't have a clue, but just plausible ideas: (1) amino acid or lipid processing is more dangerous or hinders function. Neurons need to survive for a life time. (2) the rate at which neurons need to regulate their energy level requires fast modulation, and only glucose can do it. $\endgroup$
    – Memming
    May 11, 2015 at 14:02
  • $\begingroup$ @Memming, those sound like good starting points! For (1), amino acid oxidation does yield ammonia, which perhaps could cause a problems; I'm not aware of toxicity / risks with fatty acid oxidation. (2) sounds like an interesting angle. The brain oxidizes glucose completely though, so it's not related to rapid glycolysis I think. $\endgroup$
    – Roland
    May 15, 2015 at 6:14
  • 1
    $\begingroup$ This is a great question. Most answers one can find are shallow, along the lines neurons rely on glucose because they don't use protein/fat. The equivalent superficial answer for red blood cells is because they don't have mitochondria to do oxidative phosphorylation. But neither explains why neurons specialized in this way. This might well be an unexplored question! $\endgroup$
    – SeanJ
    May 31, 2017 at 9:01

4 Answers 4


I want to present another (possibly more practical) approach towards this phenomenon. Lets begin with amino acids as alternatives.

  1. Amino acids, apart from being a source of toxic ammonia, are also harmful to the brain in another way. The two (concerned) types of amino acids, aromatic and acidic ones, are a lot more than just energy sources for the brain. Aromatic amino acids: tyrosine, phenylalanine and tryptophan, are precursors for the biosynthesis of neurotransmitters like serotonin, melatonin, norepinephrine, dopamine, etc. The greater the amount of precursors, more the product formed. Talking about acidic amino acids, glutamate and aspartate are themselves neurotransmitters. So, same case could apply to them too. So, it might have been that neurons, in evolution, decided to just get rid of such sensitive and powerful energy sources which can disrupt their basic function. You can see Fernstrom, 1994 for a study on this.

  2. Fatty acids are definitely a nice candidate. But they have some side effects too. For a long time, it was thought that since fatty acids (or lipids in general) are attached to albumin while in blood, they cannot cross the blood-brain barrier (see Stryer for example). But, this is not the case, as now proven. Then why are fatty acids not a preferred energy source for brain? There might be quite a few reasons for this.

    • First, $\beta$-oxidation of fatty acids demands more oxygen as compared to glucose, making neurons more vulnerable to hypoxia. For example, the reaction for complete oxidation of glucose is (from Molecular Cell Biology):

    $\ce{C_6H_{12}O_6 + 6 O_2 + 36 ADP^{3-} + 36 P_i^{2-} + 36 H^+ \rightarrow 6 CO_2 + 36 ATP^{4-} + 42 H_2O}$

    On the other hand, the reaction for $\beta$-oxidation of palmitic acid is (from pharmaxchange):

    $\ce{C_{37}H_{66}N_7O_{17}P_3S (Palmitoyl-coA) + 23 O_2 + 108 P_i^{2-} + 108 ADP^{3-} \rightarrow C_{21}H_{36}N_7O_{16}P_3S (coenzyme-A) + 108 ATP^{4-} + 16 CO_2 + 23 H_2O}$

    where palmitic acid is $\ce{C_{16}H_{32}O_2}$. This clearly shows that glucose oxidation requires less oxygen (1 molecule per carbon) as compared to fatty acids (almost 3 molecules per 2 carbon).

    • Second, nonesterified fatty acids (NEFA) decrease membrane potential at inner mitochondrial membrane, causing collapse of electrochemical proton gradient.

    • NEFA also interfere with electron transport chain, stimulating generation of reactive oxygen species like superoxide ions (see Zhang et al, 2006 for example).

    • Apart from this, the rate of ATP generation from fatty acids is slower as compared to glucose, which makes glucose a better option for neurons (see Stryer).

    NEFA effects on mitochondria

    All these factors might have a played a role in making neurons choosy for glucose. This is also supported by the fact that one of the enzymes of $\beta$-oxidation, 3-ketoacyl coenzyme-A thiolase, has significantly lower activity in neurons as compared to other tissues (Yang et al, 1987). You can see Reiser et al, 2013 for a study on this.

PS: It has been shown that the brain can also use lactate along with ketone bodies as an energy source. See Wyss et al, 2011 to know more about this.

  • 1
    $\begingroup$ Very good points Homo Sapiens. $\endgroup$
    – SeanJ
    Jun 1, 2017 at 11:22
  • $\begingroup$ Thanks, if you really like it, you can give an upvote :P BTW if you want to reply to somebody, you need to put @ sign and then that person's name, without spaces (if there are spaces in name). For example, @another'Homosapien' to reply to me :) $\endgroup$ Jun 1, 2017 at 11:57
  • $\begingroup$ Thanks for the suggestions. (1) is interesting. Amino acids are already present in brain tissue though, and oxidizing amino acids might not increase their levels. Also, compartments are probably important here. (2) "beta-oxidation demands more oxygen" -- this is true per carbon atom (respiratory quotient), but the relevant parameter is the P/O ratio, which is very similar between the glucose and fat. I'm not sure NEFAs are relevant here since beta-oxidation handles fatty acids coupled to carnitine, but will think about it. Re: glucose oxidation is slower, see comment to @SeanJ $\endgroup$
    – Roland
    Jun 1, 2017 at 17:36
  • $\begingroup$ @roland (1) yeah amino acids are already present in brain "tissue" for their usual purpose (not oxidation), using them for oxidation might also create problems (2) I agree with you about P/O ratio. However, I don't understand your point about free fatty acids. That's why I put Reiser et al paper, they talk solely about NEFA. I'll see for glucose oxidation very soon :) $\endgroup$ Jun 1, 2017 at 17:43
  • 1
    $\begingroup$ Also, I don't think the Stryer page is talking about glycolysis vs beta oxidation (it does say glucose oxidation). Also, even if glycolysis is faster than beta oxidation, then since the later steps (krebs cycle) are same, we can still conclude that glucose oxidation is faster than fatty acid oxidation (both need to form acetyl coA first, glycolysis does it faster than beta oxidation). $\endgroup$ Jun 1, 2017 at 19:35

Glucose is the only fuel normally used by brain cells. Because neurons cannot store glucose, they depend on the bloodstream to deliver a constant supply of this fuel. Fatty acids do not serve as fuel for the brain, because they are bound to albumin in plasma and so do not traverse the blood-brain barrier. In starvation, ketone bodies generated by the liver partly replace glucose as fuel for the brain. (Biochemistry. 5th edition.Berg JM, Tymoczko JL, Stryer L.)

Also, amino acid catabolism is the process of using amino acids as an energy source. Turning amino acids into molecules that can be used in the Krebs cycle takes energy, which means that burning protein for fuel is not as efficient as burning carbohydrates. In addition, your body needs amino acids to make new proteins. When amino acids are used as an energy source, it reduces the reserves of amino acids that are available for protein synthesis.

  • 3
    $\begingroup$ Yes, the fact that glucose is the main fuel of the brain is well established -- the question is why. There seems to be plenty of evidence that the brain takes up albumin-bound fatty acids (but not lipoproteins), see for example ncbi.nlm.nih.gov/pubmed/17901540 . There is also evidence for fatty acid oxidation by the brain, for example ncbi.nlm.nih.gov/pubmed/3368026 . But still glucose is required. $\endgroup$
    – Roland
    May 11, 2015 at 16:45
  • 1
    $\begingroup$ Regarding amino acids, during fasting the glucose supplied to the brain from the liver derived from amino acids anyway, so I don't think the efficiency argument works. It rather seems like the amino acid metabolism is not possible in brain tissue, for some reason, and is "outsourced" to the liver. Maybe the brain has difficulties disposing of ammonia? $\endgroup$
    – Roland
    May 11, 2015 at 16:49
  • $\begingroup$ I think protein breakdown is the important factor in using amino acids. It seems more logical to use i.e. muscle mass than brain mass, as to not lose structures with a memory function. $\endgroup$
    – Sleepses
    May 12, 2015 at 13:49
  • 1
    $\begingroup$ @Sleepses, that sounds reasonable. There are plenty of amino acids in blood though (largely derived from muscle) that other tissues can take up and oxidize. $\endgroup$
    – Roland
    May 15, 2015 at 6:17
  • $\begingroup$ Well considering how import fats are in the construction of the brain, breaking down fats for energy probably would be counter productive. Our course it could all just be an evolutionary hold over, the blood brain barrier probably prevented the use of anything but sugars in the early origins of the central nervous system and we might just be stuck with the results. $\endgroup$
    – John
    May 31, 2017 at 10:06

This is something I have been researching on and off for many years. I studied nutrition back in the early 1980s and our lecturer always said that "fats burn in the flame of CHOs". He said that the citric acid cycle was dependent on oxaloacetic acid - the predominant source was from glucose (pyruvic acid). https://www.sciencedirect.com/topics/medicine-and-dentistry/oxaloacetic-acid

In nutrition, for patients with epilepsy, they are given a ketogenic diet as this is known to suppress brain signals. Clearly the brain functions sub-optimally when deprived of glucose. One of my concerns with the rise in dementia is the high protein/low carb diet proposed, and the claims that the body (and brain) can do without CHOs.

The other piece of research was from Professor Robert Horn (retired, Oslo) in his work for students. He compiled a wonderful PDF on glucose metabolism and here I quote: "The blood-brain-barrier is comprised of glia cells, primarily astrocytes. A small fraction of the glucose released from capillaries wanders directly to nerves and synapses. However, most is "trapped" in astrocytes and oxidized to lactate by these cells. Lactate goes further into the brain and nourishes the brain's neurons. This seems to be especially important for glutamatergic neurons which comprise much of the brain. The astrocytes also efficiently pick up released glutamate, convert this to glutamine, and send the product back to glutamatergic neurons where it continues to cycle as a neurotransmitter. The axons do not contain glycogen and are, therefore, completely dependent upon the lactate sent from astrocytes to maintain ATP levels. Astrocytes and other glia cells appear to have some glycogen which can serve as a very short-term source of lactate.So the answer to the preceding question is that much of the brain is dependent upon lactate from glia cells to provide substrate for aerobic energy production; ketone bodies cannot cover their substrate requirements. One can reduce glucose consumption and use ketone bodies during starvation. However, some neurons must have lactate and the brain must continue to use blood sugar."


The body uses 2 dedicated groups of fuel molecules (fats and sugars) and can break down its own structure molecules in case of emergency (proteins). The brain uses sugars, mostly glucose, and ketone bodies, a soluble form of acetyl derived from fat breakdown. I love this question, Roland, since you've probably hit on an open, poorly investigated question that goes to the fundamentals of why neurons evolved to be this specialized. So why does the neuron rely much less on fats and proteins than more flexible cells like those of the liver?

Protein digestion could indeed be dangerous for neurons because it produces ammonium (NH4+). Neither fats nor sugars contain significant amounts of nitrogen. Their final breakdown products, H2O and CO2, are harmless. Ammonium, on the other hand, is very similar to potassium (K+), an ion central to the neuronal membrane potential and thus its function. It has a similar size and charge (see image below) and is hypothesized to mess with potassium signaling in the brain. It might do this by interfering with K+ transporters (see this schematic by Eid). The potassium ion might be mimicked by ammonium in neuronal signaling.

This leaves us with fats. I have 2 hypotheses here. The final steps of fat breakdown in the mitochondrion involve the transfer of electrons onto oxygen to produce water. An unavoidable side reaction here is the creation of a small amounts of reactive oxygen species that have the potential to damage close-by molecules via radical reactions. Since neurons are so long lived, this is very dangerous for them as they accumulate damage little by little. Not so for liver cells which are readily replaced if damaged beyond recovery. Second, fat breakdown is a much slower way to energy/ATP than glucose breakdown, since it involves the sequential shortening of the fatty acid chain (beta oxidation), the generation of reduced intermediates (mostly NADH in the citrate cycle), and finally oxidative phosphorylation - several dozen steps in total. Glucose metabolism yields ATP quickly in just a few steps (glycolysis). Ketone bodies also fit, since they are already very short and thus much more quickly metabolized than long chain fatty acids. A combination of these 2 reasons might be why neurons prefer glucose to fats, but I believe this is a poorly investigated question that merits more research. PhD anybody?

  • $\begingroup$ So if you are offering a PhD, what is your proposal to test your hypotheses and solve a problem that people have been aware of for at least 50 years? It's one thing in science identifying a problem, the trick is to realize when either new concepts or new methods appear to make it possible to solve it. $\endgroup$
    – David
    May 31, 2017 at 12:59
  • $\begingroup$ Thanks for the suggestions. Some comments: (1) toxicity ammonia production is an interesting hypothesis (and it has been suggested before) but I haven't seen any evidence that the brain would be particularly sensitive to ammonia. (2) Regarding reactive oxygen species, consider that neurons already oxidize glucose carbon in mitochondria. (3) I don't think it's true that fat oxidation necessarily is slower (in terms of ATP synthesis rate) than glucose oxidation; and generally, the number of enzymatic steps has little to do with flux through a pathway. Just some critical feedback! $\endgroup$
    – Roland
    May 31, 2017 at 19:56
  • $\begingroup$ Hey David, one could try to slow or block flow through glycolysis and see how the shifted energy metabolism affects neurons. I don't find that it's a well identified problem. Most people (but not Roland), textbooks, and articles are perfectly happy with the shallow explanation that neurons have downregulated some enzymes and are thus more glucose dependent than other cells. $\endgroup$
    – SeanJ
    Jun 1, 2017 at 11:13
  • $\begingroup$ Hi Roland, 1) have you had a look at the paper by Eid. They present evidence as to why neurons might be more sensitive. 2) Correct. They might also derive some energy from glycolysis to pyruvate/lactate without oxygen. Does anybody know of data quantifying which energy pathway is used the most in neurons? 3) If it were as fast, why do we use glycolysis almost exclusively for bursts of muscle energy and not oxidative phosphorylation? I'd love to read more about steps versus flux if you could recommend a paper or a book. All the best $\endgroup$
    – SeanJ
    Jun 1, 2017 at 11:19
  • $\begingroup$ (1) Sorry, I missed the Eid et al paper link-- it's quite interesting, thanks! (3) Fast-twitch muscle relies on glycolysis to lactate, that's quite different; here the choice is between glucose oxidation or fat oxidation. Re: flux through pathways, this depends on a number of factors, such as metabolite concentrations, enzyme amounts and turnover rates, Gibb's energies of reactions, compartmentalization, and what not. Suffice it to say that there is no easy relation between the number of steps in a pathway and flux :) We might want to move this discussion to chat. $\endgroup$
    – Roland
    Jun 1, 2017 at 17:10

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .