I decided to begin our journey into feeding the hungry brain by looking at the wonderful world of the ketone molecule. This is partly because of my research into the potential benefits of ketones on cerebral energetics but mainly because of the almost cultish status that ketones diets and supplements have acquired over the past few years. In fact, its almost impossible to open a popular magazine or watch a health program on TV without coming across a reference to the benefits of ketones on everything from fitness to weight loss and mental function. Ketones have been advocated for treating various cancers and degenerative neurological diseases as well as improving longevity. The question I want to answer is, how much of this all just hype or is there some evidence to support what often seems to be outrages claims? This is what we intend to explore further.
SO, TO BEGIN WITH, WHAT ARE KETONES ANYWAY AND WHY DO WE NEED THEM?
As with all topics we intend to cover, we will begin with some background information. Although it is possible to skip ahead, I would suggest some basic information may be useful to the reader to help inform choices once we delve into more detail later on. As always, I will attempt to keep things as simple as possible without losing too much content.
The recent surge in interest in “all things ketone” is fascinating to me as for most of the 20th century, ketones were considered to be bad for your health. In fact, ketones were designated by the author Vanltallie, as “metabolism’s ugly duckling”. This is largely based on the high ketone concentrations found in patients with diabetic keto-acidosis or diabetic coma as its sometimes referred to. And this is true, DKA is a serious disease which can lead to death if not adequately treated but the high ketone levels are a result of hormone deficiency (insulin) and not the cause.
Ketones in fact are fundamental to the development of humans. In fact, evolutionary biologists suggest humans as an advanced species became a possibility with a change in our biology that allowed for ketone production (more on that later). Certainly, the development of the human brain, which requires large amounts of energy to function, was partly due to our progress in ketone production. Ketones provide a means of mobilizing stored energy and allow us to survive for long periods without food. This was critical when it came to leaving our birthplace in Africa and roaming around the planet (where food sources were few and far between). Ketones are pretty important.
Let’s dig a bit deeper …
But first, a little history;
HISTORY OF KETONES
In the mid-19th century, ketones were first discovered in the urine of diabetic patients. They were originally believed to result from incomplete combustion of fats and therefore of little positive clinical significance. That and the finding of ketones in diabetic coma patients meant that for many decades, ketones were negatively viewed by the medical establishment. However, by the 1920’s things began to change. It was noted that children with resistant epilepsy improved significantly with fasting. And that the observed ketosis may play a role in this finding. Because of the limitations of ongoing starvation to induce ketosis, the use of a ketogenic diet to mimic the effects of starvation was proposed in 1921 by researchers at the Mayo Clinic. And in 1924, Peterman reported clinical effectiveness of the regimen. (original 4:1) diet.
Although apparently effective, this approach was largely overlooked by most clinicians and tended to be limited to certain specialist units, one of which was located at Johns Hopkins. In 1967, it was noted that ketones could replace glucose as the major fuel source during starvation. And eventually, in the 1990’s benefits of ketogenic diets in managing refractory epilepsy in children was confirmed in well run clinical trials. Over the past 2 decades, ketone research has exploded into many aspects of metabolism and cerebral energetics. The difficulty as always, is separating the facts from the hype.
SO WHAT ARE KETONES ANYWAY?
We’ll come back to clinical applications of ketones in later sections. But now let’s explore the basic science behind these molecules. What are they? Where do they come from and why do we need them? And why do bears, which can fast for many weeks, not need them?
Firstly, ketones are produced from the breakdown of fatty acids and provide a vital alternative metafuel for all domains of life including eukaryota, archaea and bacteria.
At this point I need to clarify some terminology. I’ve been using the term ketone loosely to describe ketone bodies. Ketones strictly speaking consist of a functional group with the structure containing a carbonyl group (C=O) and a variety of carbon containing substituents.
Now don’t panic, its fairly simple and looks like this – where R = carbon containing entities
Not to be confused with aldehydes (as in formaldehyde)
Which looks very similar.
Ketone bodies (KB) on the other hand, are the ketone containing molecules which are produced in the liver from fatty acids during periods of starvation. I will be using the terms interchangeably for our discussions.
The 3 principle ketones produced by the human liver are, acetoacetate (AcAc), beta-hydroxybutyrate (BHB) and acetone.
I’ll dispense with acetone quickly as it is thought not be important in human metabolism (maybe). It is a volatile substance that can be excreted via the lungs and is responsible for the fruity smell noted on the breath of diabetics in diabetic coma. Although controversial, there is some evidence that acetone can be converted to lactate in the liver and may be able to enter the Krebs cycle.
For those few astute individuals …… although included in the list of ketone bodies, because of its chemical structure, BHB is technically not a ketone. However, it is considered a ketone for practical purposes.
KETONE METABOLISM
Where it all started ….
Most of our understanding of ketone metabolism in humans comes from work undertaken in the 1960’s by Cahill and Owen who measured ketones and other physiological changes in obese individuals during a prolonged fast (approx. 5-6 weeks). From these experiments they were able to determine the course and metabolic changes which occur in protein, fat and glucose metabolism in humans during starvation. Much of the following discussion and data is based on their original study results.
What is profoundly clear, is that without the ability to produce ketones, death would occur with 2 weeks of starvation rather than the 57–73 days noted in hunger strikers. Many will remember Bobby Sands, the Irish hunger striker who lived 66 days before succumbing. So why is it that humans, almost exclusively in the animal kingdom, need ketones to survive while others live happily off their glucose stores/metabolism?
The answer lies in the fact that the human brain requires significant energy to function (for more info., check out the previous section on brain metabolism). As noted, 20% of the basal metabolic rate in humans goes to the brain. Compare this to a bear which can hibernate for several months while maintaining virtually normal ketone concentration. This is because the bear brain requires less then 5% BMR to tick along nicely and as such, is able to maintain sufficient blood glucose from body stores of fat (prod. from glycerol). In fact, few animals increase KB concentrations above 1 mmol/l. This compares with 4-7 mmol/l after 2 weeks starvation for adult humans.
Furthermore, brain size does not necessarily correlate with function or BMR. For example, the brain of a 40-tonne whale (7.5 kg) only utilizes 2-3 times as much oxygen as compared with a human brain (1.5 kg) despite being significantly bigger. And it appears that the ability of the human brain to metabolize KB correlates with the increase in the cerebral metabolic rate which suggests that is was one of the defining moments in human evolution (more on this later).
So, given the importance of ketone metabolism in human development, lets look at what happens during starvation. As previously mention, in 1967, Owen followed the course of 3 obese patients who underwent 5-6 weeks of starvation for health reasons. During these weeks, a number of tests were undertaken to try and understand the metabolic changes which were taking place. Their results included the following;
Max concentration of KB during starvation;
AcAc 2 – 4 mmol/l
BHB 5 – 12 mmol/l
Acetone 3 – 5 mmol/l
They were also able to document the various metabolic changes of prolonged starvation;
12 – 24 hrs
Mild ketogenesis, lipolysis, gluconeogenesis and proteolysis
Marked glycogenolysis
Glucose and free fatty acids (FFA) supply approx. 40 % BMR each
Amino acids (AA) oxidation – 20% BMR
Ketones – 2-3% BMR
2-3 days
Maximum ketogenesis, lipolysis, gluconeogenesis and proteolysis
Glycogen stores depleted
FFA, KB and glycerol supply 90% BMR
KB – 30-40% BMR
AA oxidation supply the rest
> 3 days
Decrease thermogenesis
Decrease glucose
Decrease number of hormones including thyroid, cortisol, insulin and catecholamines
The maximum rate of ketone consumption was found to be around 129 g/d
There is an associated metabolic acidosis initially due to increase H+ production but pH stabilized eventually around 7.35
Although the study results apply to obese patients, similar processes occur in patients with a normal body mass index.
Figure 1
Figure 1: The five metabolic stages between the postabsorptive state and the near-steady
state of prolonged starvation.
(Ruderman NB, Aoki TT, Cahill GF. 1976. Gluconeogenesis and its disorders in man. In Gluconeogenesis: Its Regulation in Mammalian Species, ed. RW Hanson, MA Mehlman, pp 515–30. New York: Wiley)
Figure 3
Figure 3: Circulating concentrations of, BHB, glucose, free fatty acids and acetoacetate
in obese but otherwise normal man fasting for 40 days.
(Cahill Jr, George F., and Richard L. Veech. “Ketoacids? Good Medicine?” Transactions of the American Clinical and Climatological Association 114 (2003): 149.)
Before we look into specifics of ketone metabolism, lets explore why the above changes are necessary to keep humans alive and well when food is in short supply.
THE BRAIN AND ENERGY PRODUCTION
The human liver is very efficient at producing ketones and can rapidly ramp up production to 185g KB/d. After an overnight fast, 2-6% of BMR is provided by KB. This rapidly increases to 30-40% BMR after 3 days.
But why is this necessary? Couldn’t we simply make glucose from other substances such as proteins and AA as most of the animal kingdom.
Let’s look into this in a bit more detail.
Ketone production is highly sensitive to presence of glucose and as little as 50g/d greatly reduces production. Furthermore, other substrates may similarly affect ketone metabolism implying that a low carbohydrate (CHO) diet may not necessarily induce ketogenesis. This was demonstrated in the 1920’s in Baffin Island Eskimo’s who’s only CHO intake consisted of glycogen consumed from eating seal meat. Testing demonstrated only small quantities of ketonuria. Ketosis is therefore not only affected by CHO intake, but protein as well. This is because more then 50% of AA in diet are glucogenic. For every 2g protein consumed, approx. 1g of CHO is produced.
Aside: Interestingly, pure protein diets are particularly dangerous as has been documented on several expeditions where the only available food was rabbit meat, which is particularly lean (only 8% fat). Those individuals developed extreme fat-hunger (also known as rabbit-starvation) which eventually led to death.
If glucose was the only fuel available, the brain would require approx. 100 – 150g of glucose per day to provide enough energy to match the basal metabolic rate. While in the well-fed state, this is easily achieved but in times of prolonged starvation, between 172–259g of protein would have to be metabolised to provide enough glucose to keep the brain going. At this rate, death would be expected within 2 weeks, not the 57-73 days previously noted. Furthermore, it is clear this does not occur in-vivo as a study of a fasting man demonstrated an average rate of body protein catabolism of 55g per day after 31 days. 32g of this would be available for brain fuel (in the form of glucose) which works out to approx. 27% of brain needs. This is markedly insufficient to supply daily brain energy requirements.
KETONES SAVE THE DAY!
Biochemically, ketones are the ideal fuel. The average 70 kg male’s energy stores consist of; adipose tissue including triacylglycerol ~ 12 kg, muscle protein ~ 6 kg and glycogen ~ 500 g.
During starvation, ketone production is useful because;
Glycogen stores are low
KB limits protein breakdown
Large store of fats available to provide substrate
Long chain FFA’s are unable to cross BBB whereas ketones can
Furthermore, ketones are more energy efficient then glucose. Studies in 1940’s demonstrated that KB added to sperm decreased oxygen consumption while also increasing motility. Later Veech found adding KB to a heart perfusate led to a 28% increase in hydraulic work of a rat heart while decreasing requirements for oxygen. The reasons why metabolism of KB leads to increase energy compared to glucose is complicated and I refer to Veech’s original paper for the interested reader (see below).
Ketones also play a critical role in human development. Pregnancy tends towards a ketogenic state and may lead to 2-3 x increase production. KB can freely cross the placenta. In fact, humans are born ketotic. Following birth, neonatal blood glucose concentrations fall rapidly and BHB may rise to 2-3 mmol/l. The newborn brain consumes 60-70% BMR at birth, nearly half energy is provided by KB’s. Fats are critical for brain development. Colostrum (from breast milk) consists of protein and fat with minimal lactose (essentially a ketogenic diet). Over the next 3 days lactose production increases and ketone concentrations decrease rapidly.
Figure 2
Figure 2: Levels of β-hydroxybutyrate in starving subjects of different ages.
(Cahill, George F. “Fuel Metabolism in Starvation.” Annual Review of Nutrition 26, no. 1 (August 2006): 1–22. https://doi.org/10.1146/annurev.nutr.26.061505.111258.)
In fact, it turns out that children can increase their ketone concentrations much more rapidly then adults. Within as little as 30 hours of starvation, KB will have increased to around 5 mmol/l. Compare this with an adult which may take as long as 2 weeks to reach comparable concentrations. This is an important adaptive mechanism as children tend to have less glycogen stores and higher cerebral energy requirements compared to adults as a result of their larger brain:body mass ratio. This has major therapeutic implications and is the basis for the ketogenic diet utilized in children with poorly controlled epilepsy. It’s much more difficult to induce ketogenesis with a ketogenic diet in adults.
Figure 4
Figure 4: Brain substrate utilization in three fasting obese volunteers after several
weeks of starvation.
(Cahill, George F. “Fuel Metabolism in Starvation.” Annual Review of Nutrition 26, no. 1 (August 2006): 1–22. https://doi.org/10.1146/annurev.nutr.26.061505.111258.)
Ok, that sounds great. But what controls ketone production?
REGULATION OF KETONES
Right (for those of you who are still with me), we have explored the ketone metabolism and its implications fairly extensively. The final piece of the puzzle involves understanding the regulation of ketone metabolism. This is a very complicated process and potentially of limited interest to the average reader. It is however, critical to our understanding and is necessary when attempting to look at artificially increasing ketone concentrations with supplements. Clearly, the easiest way to induce ketosis is to stop eating. This of course doesn’t appeal to most of the population and we therefore need to look into other potential avenues.
We’ve noted the maximum concentrations of ketones after starvation above. There are a number of mechanisms which regulate this. These include;
Availability of free fatty acids converted from adipose tissue
Hormones – insulin is the main hormone. Low insulin levels are critical for KB to be produced. (Just think of what happens to diabetics who stop their insulin.) Other hormones include glucagon and adrenalin
Energy requirements of the body
Excretion in the kidneys
KB are produced almost exclusively in the liver although small quantities can come from tumours, astrocytes (in the brain), kidney’s and skeletal muscle. Max. production is around 185 g/d (although this figure may vary depending on population). The liver lacks the enzymes required to metabolise ketones and therefore most is exported to peripheral tissues in the form of BHB. Once taken up by the tissues (we’ll discuss brain in a moment), BHB is transported across the mitochondrial membrane (mitochondria are responsible for producing energy in cells) where it is converted back to AcAc, eventually entering the Krebs cycle via Acetyl-CoA leading to the generation of ATP.
Ok, take a deep breath, I realise this is a lot to take in, but this process has some very important implications;
The biggest users (organs) of ketones in the fed state are the heart and the kidneys (based on mass). The brain becomes an important user in starvation.
As ketones enter the Kreb cycle via Acetyl-CoA, they are able to bypass pyruvate (glycolysis pathway) and any blockages in that pathway providing readily available energy
Ketones are a very efficient form of energy as discussed above in the rat heart experiments
Ketones have both pro and anti-inflammatory properties which may decrease cell damage and death
Ketones provide substrates for the production of other molecules including cholesterol, FFA, AA and others
Figure 7
FIGURE 7: Entry of ketone bodies into the citric acid cycle. Ketone bodies are used as fuel by the brain during prolonged starvation.
(Laffel, Lori. “Ketone Bodies: A Review of Physiology, Pathophysiology and Application of Monitoring to Diabetes.” Diabetes/Metabolism Research and Reviews 15, no. 6 (November 1999): 412–26. https://doi.org/10.1002/(SICI)1520-7560(199911/12)15:6<412::AID-DMRR72>3.0.CO;2-8.)
KETONES METABOLISM AND THE BRAIN
So finally, after all the basic science, we’ve arrived at our primary destination.
As the brain is only capable of producing small quantities of ketones, the transportation of KB into the brain is of paramount importance. As previously discussed, the BBB is impervious to most molecules. Even glucose requires a transport system to get across (GLUT 1). Ketones in turn rely on the abundance of monocarboxylic acid transporters (MCT1) to regulate their transfer across the BBB. MCT is preserved with age.
Studies have shown that AcAc crosses more readily than BHB. Furthermore, it appears that the rate of transfer is related to the concentration of blood ketones and duration of ketosis. As such, rapid short-lived increases in blood ketones leads to a small increase in cerebral concentrations. However, prolonged raised ketone concentrations such as occurs in starvation, results in much higher brain concentrations.
Studies have demonstrated the relationship between KB concentrations and brain energy;
Ketone bodies Proportion of brain energy
0.3-0.5 mmol/l (12-24h fast) 3-5%
1.5 mmol/l (2-3 day fast) 18%
5 mmol/l (8 day fast) 60%
7 mmol/l (> 20 day fast) > 60%
KB have been utilized to treat a number of cerebral conditions associated with brain energy deficits. These will be discussed in later chapters but suffice it to say that part of the explanation for the benefit of KBs is that they may be a better energy source then glucose (see the description of the rat heart experiments above). This would appear to result from changes that are induced in the mitochondrial ATP production. (remember, ATP is the major energy containing molecule produced by mitochondria in cells). Essentially, there is an inherently higher heat of combustion in BHB then in pyruvate (from glucose metabolism). The upshot of all this is that ketones can provide the brain with greater energy per weight then glucose.
From the above it is clear that a number of disease states exist which may benefit from ketone supplementation. The ideal plasma ketone concentration is not known but maximum transfer across BBB occurs at KB concentrations of approx. 5 mmol/l. Whether this is clinically important is not known.
Well done. You’ve made it to the end.
We have completed our basic tour through the world of ketones. It is now time to put our knowledge to clinical use to ascertain whether these theoretical benefits translate into improved clinical outcomes. The following chapters will explore ‘all things ketone in health and disease’ with an emphasis on evidenced based outcomes or where lacking, expert opinion based on physiological evidence.
REFERENCES
Veech, Richard L. “The Therapeutic Implications of Ketone Bodies: The Effects of Ketone Bodies in Pathological Conditions: Ketosis, Ketogenic Diet, Redox States, Insulin Resistance, and Mitochondrial Metabolism.” Prostaglandins, Leukotrienes and Essential Fatty Acids 70, no. 3 (March 2004): 309–19. https://doi.org/10.1016/j.plefa.2003.09.007.
VanItallie, Theodore B., and Thomas H. Nufert. “Ketones: Metabolism’s Ugly Duckling.” Nutrition Reviews 61, no. 10 (2003): 327–341.
Owen, O. E., A. P. Morgan, H. G. Kemp, J. M. Sullivan, M. G. Herrera, and G. F. Cahill. “Brain Metabolism during Fasting*.” Journal of Clinical Investigation 46, no. 10 (October 1, 1967): 1589–95. https://doi.org/10.1172/JCI105650.
Owen, O. E., S. Caprio, G. A. Reichard, M. A. Mozzoli, G. Boden, and R. S. Owen. “Ketosis of Starvation: A Revisit and New Perspectives.” Clinics in Endocrinology and Metabolism 12, no. 2 (July 1983): 359–79.
Morris, A. A. M. “Cerebral Ketone Body Metabolism.” Journal of Inherited Metabolic Disease 28, no. 2 (2005): 109–121.
Hashim, Sami A., and Theodore B. VanItallie. “Ketone Body Therapy: From the Ketogenic Diet to the Oral Administration of Ketone Ester.” Journal of Lipid Research 55, no. 9 (September 2014): 1818–26. https://doi.org/10.1194/jlr.R046599.
Cahill, George F. “Fuel Metabolism in Starvation.” Annual Review of Nutrition 26, no. 1 (August 2006): 1–22. https://doi.org/10.1146/annurev.nutr.26.061505.111258.
Cahill Jr, George F., and Richard L. Veech. “Ketoacids? Good Medicine?” Transactions of the American Clinical and Climatological Association 114 (2003): 149.
Neal, Elizabeth G., Hannah Chaffe, Ruby H. Schwartz, Margaret S. Lawson, Nicole Edwards, Geogianna Fitzsimmons, Andrea Whitney, and J. Helen Cross. “The Ketogenic Diet for the Treatment of Childhood Epilepsy: A Randomised Controlled Trial.” The Lancet. Neurology 7, no. 6 (June 2008): 500–506. https://doi.org/10.1016/S1474-4422(08)70092-9.
Comments