Saturday, 22 December 2012

Evolutionary Trade-Offs: Fast Versus Famine

This is a sister post to Primal North's "Keto Adaptation vs Low Carb Limbo".

Gluconeogenesis (GNG) is a big topic in ketogenic dieting, and most people think it means eating too much protein knocks you out of ketosis as they think excess protein increases GNG which then increases blood glucose and thus insulin. This is completely wrong. GNG changes little between a high-carb and low-carb diet (link), which of course means Jaminet's idea of eating 'safe starches' to "less the burden on the liver" is nonsense, but it actually has many implications that no-one seems to be addressing: glycogen depletion from exercise in keto-adapted individuals. To many the mere idea of glycogen being using when keto-adapted, let alone it being depleted, is seer heresy but it can and does happen, and a lot more often than people think it does.

We have three major fuel tanks, as it were, in our body: glycogen, fatty acids, and ketones. We burn a mix of these at all times. There are three basic 'modes' that our body uses:
  • Carb mode: most energy is being derived from glycogen, and a small amount from fatty acids.
  • Fat mode: most of the energy comes from fatty acids, and of the rest most comes from glycogen, and a small amount from ketones.
  • Ketone mode: about half of the energy comes from fatty acids still, but of the rest most comes from ketones and a small amount from glycogen.

We burn glycogen all the time, whether we eat high-carb, low-carb, ketogenic, zero carb, nutritional ketosis; it doesn't matter what diet you eat, we always burn some glycogen. While exercising energy requirement increases and so too does need for glycogen.

The reason I describe them as 'modes' is because the body doesn't slowly change what fuel mix it uses, it's like a switch, the body will suddenly change from say fat mode to ketone mode. This study (link) on sled dogs revealed that "[d]uring the first few days of racing, sled dogs draw energy from glycogen stored inside muscle cells. But instead of depleting glycogen stores and tiring the muscles, the animals suddenly switch to a glycogen-sparing metabolism." (Emphasis mine)

But what's interesting is that after racing 100 miles for 5 days, their muscle glycogen was slightly higher than when they started (link). Even if they're burning a smaller amount of glycogen something has to be refilling it for it to end up higher. Most people think GNG means turning protein into glucose that then raises your blood sugar, this is wrong, GNG is the turning of protein into glycogen. All carnivorous animals have high levels of GNG in their liver, and they need it to be high, as they need glucose to simply live (all animals die if blood glucose levels drop to 0) but also to fuel their high intensity exercise aka chasing down prey.

Dogs and cats, which are domestic versions of wild wolves and big cats, have much higher levels of GNG in their liver, and only can tolerate higher levels of protein intake than us before toxicity aka rabbit starvation. The sled dogs in this study ate a lot of protein: "[e]ach 50-pound canine consumes about 12,000 calories daily (typically 60 percent fat and 40 percent carbohydrate and protein)", this works out to 800g of fat, and up to 1200g of protein and this is for an animal a third our size!

The reason these animals can tolerate a much larger amount of protein then we can (anything over ~200g of protein for a human starts to become toxic), is because they have a higher level of GNG. Dietary protein has three major metabolic pathways it can take: muscle/protein synthesis, GNG, and breakdown to urea. Only so much goes into muscle/protein synthesis, as much is needed but only so much proteins are needed and even rapid muscle growth works out to only a handful of grams a day; the rest either gets turned into glycogen via GNG, or is broken down in urea and then excreted in our urine. So when certain leaders of certain zero carb groups say that excess protein isn't a problem at all and won't raise blood sugars and drive up GNG, they're partly correct. GNG is limited by the size of our liver, and doesn't change much depending on diet, the excess protein is indeed broken down in urea and excreted as they say. But we also have a limited capacity to excrete urea and excess protein leads to a build up of urea in the blood which is the mechanism by which we get protein toxicity aka rabbit starvation.

Why is the level of GNG important? Because it limits our ability to refuel glycogen when we don't eat carbs (or not enough). If our glycogen fuel tank becomes empty, doesn't matter what 'mode' we're in, then we will 'bonk out', 'hit the wall', call it what you want but it means you're not going to finish that race, if keto-adapted you may be able to struggle through it. Once the glycogen tank is empty, it can take a long time to refill it and during this time many of the common effects of low carb limbo are seen such as hypoglycaemia and increased risk of infection. Our ability to refill glycogen limits our exercise capacity, in animals such as dogs they can eat huge amounts of protein which undergoes GNG and refills that glycogen such that they can have more glycogen after 500 miles than before they started. In tests on humans running while consuming low-carb diets, we see that glycogen is depleted during exercise, and while it is depleted much slower than someone eating a high-carb diet, it needs to be refuelled between exercises. But because GNG is governed by our livers, and eating more protein doesn't increase it, this means that we can only refill so much of our glycogen tank before we have to exercise again. So while a sled dog can refill that glycogen tank completely say overnight, we need longer than that to get our glycogen back up.

So if one is only eating no/little carbs, they need to ensure that there is adequate time between heavy exercises such that our glycogen can be refilled sufficiently. Repeated heavy exercise too close together results not only in 'bonking out' but even illness. My good friend Danny Albers of Primal North has experienced this first hand, if he doesn't 'carb back-load  (to refill glycogen), consume 'superstarch' (more on that later), or wait long enough between exercising, he gets a cold or infection, sometimes very bad.

So then how do people like Jimmy Moore avoid this? Because Jimmy Moore measures his blood ketones, and Danny doesn't. Jimmy Moore reports that if his blood ketones are over 1.5mmol/L then he can exercise fine, if it's lower then he will bonk out (link) and instead chooses not to exercise that day. Your blood ketones need to reach a certain level for you to be in ketone mode, aka keto-adapted, below this level you are only in fat mode and will still be burning quite a lot of glycogen.Jimmy's diet is 85% fat, 30g carbs, the rest protein, but even he isn't in ketone mode all the time and must measure his blood ketones before exercising. If you aren't measuring your blood ketones before you exercise then you have no way of telling if you're keto-adapted or not, blood ketones mean nothing. Take the time to read Danny's brilliant post on what true keto-adaptation is (link). Everyone will have a different level of blood ketones at which they will switch to ketone mode though, so you will need to test each time and not at which level you don't bonk. Those who have never been obese, and have been on VLC/ZC diets for a very long time, will have a lower threshold for keto-adaptation, this applies to us and animals (in the zero carb community they actually stress not exercising for 6 months, likely in part to attain this ideal before you do). The wolf who has eaten a carnivorous diet his whole life will be able to access that ketone mode much easier than a previously morbidly obese person with metabolic syndrome, as evidenced that the sled dogs switched to ketone mode only after a couple days of running, while it can take somebody months and months to keto-adapt.

But even those who have been eating a zero carb diet for a very long time, and have never been obese will find there is a limit to how much they can exercise because of this lower GNG level compared to other carnivorous animals; a certain leader of a certain zero carb group only runs half marathons and finds he bonks out if he attempts to run a full marathon. But wait, I hear you cry, what about that guy who set a new world record in the Western States 100, wasn't he low carb? (link) Yes he was, but he consumed about 2,000 calories worth of glucose over the course of the race, AND he was taking a bee pollen supplement that enhances fatty acid oxidation. 2,000 calories of glucose may sound a lot, but compared to the roughly 10,000 calories needed to run the race it's actually quite small (the high-carb runners have to consume 10,000 calories of glucose over the race minus what they can carbo-load the night before), the rest of coming from fatty acids and ketones. So even he had to refill that glycogen or risk bonking, but he needed much much less as he was burning it at a slower rate and using mostly his own body fat for fuel (as a mix of fatty acids and ketones).

So why do we have a lower level of GNG, especially if we're meant to be pure carnivores as some claim? Well let's take a look at cats, cats have the highest level of GNG in the animal kingdom and thus the highest dietary protein requirements. What's really interesting is if you fast a house cat for more than a day or so, it can develop fatal fatty liver as it starts to burn it's own muscle tissue for GNG. So if you're an animal with a very high level of GNG, you can't fast even for a few days, but you can refill the glycogen that's used up in heavy exercise hunting down prey to ensure you're not fasting. But instead if you're a carnivore with a low level of GNG then you may not be able to run marathons every day such as to hunt prey, but you can happily survive those periods in between hunts, even if they're weeks apart, such as if they're not much prey.

Our lower level of GNG compared to other carnivorous animals is an adaptation to famine!

So we have an evolutionary trade-off, while other animals such as cats are adapted for hunting more frequently and faster (just look at a cheetah!), we humans selected for famine over fast.

So what do we do if you want/need to do heavy exercise more frequently than can be supplied by GNG? Here are the options:
  • Backload carbs: refill the glycogen after exercising.
  • Cycle carbs: eat high-carb on workout days and low-carb on rest days, or carb-up once a week.
But what if you don't want to or can't eat carbs for whatever reason? There may be a couple ways to 'cheat'...

Talking to many zero-carbers it seems that animal based carbs (such as from dairy or shellfish) don't raise their blood sugar as the same amount of carbs from plants would. Are animal carbs somehow different from plant carbs? They may well be, in chapter 9 of "How To Prevent Heart Attacks" by Ben Sandler (link) he talks about something called 'gamma-glucose'. The basic idea is that there is a third kind of glucose (normal glucose comes in two kinds, alpha and beta) which is different, it's unstable and made by the liver. When we eat carbs or otherwise increase insulin then we make less 'gamma-glucose', and increase our production of 'gamma-glucose' after a meal with only protein and/or fat. So carbs from animal sources may be in the form of this 'gamma-glucose' and so animal carbs may be good for refilling glycogen without triggering insulin production.

Another 'cheat' is super starch (said I'd get to it). Super starch is a special kind of carbohydrate that is designed to refill glycogen without increasing blood glucose or insulin. Many athletes use it because it's doesn't upset the stomach like other carbs (the main reason ultra-marathoners are turning to low-carb is because the frequent carb-ups over the course of the race upset the stomach to the point where it won't except any more and they simply throw everything up, then run out of glycogen and bonk). Volek, of The Art And Science Of Low-Carb Living/Performance has written several articles/papers on super starch (link). Peter Attia uses super starch to help him exercise efficiently (link and link). My good friend Danny Albers has personally tried out a similar product to super starch (again designed to refill glycogen without increasing glucose/insulin, just cheaper) with great results in his exercise (link).


  1. what a great article, wish i had read this 12 months ago when i started lowering carbs. also thank 2 jimmy moore 4 tweeting this.


  2. Excess protein is not converted into urea and excreted, at least not in the way you’re implying. Urea is created only from the alpha-amino group of an amino acid that is cleaved from the molecule and ultimately combined with carbon dioxide. The remaining portion of the amino acid, called the ‘carbon skeleton’, has various fates depending on its structure and the needs of the body. It can be used to synthesize glucose or lipid, or it can be oxidized in the citric acid cycle to produce energy plus CO2. The carbon skeleton is not so easily gotten rid of. I don't think you know biochemistry/physiology as well as you think you do. This is basic, first-year knowledge.

  3. Jaminet's idea of eating 'safe starches' to "less the burden on the liver" is nonsense

    Of course it is--otherwise, cats and dogs would start dropping in the streets from liver failure! As it is, it isn't often that a carnivorous pet suffers from liver failure--kidney or heart failure, yes, but not liver failure.

  4. 1. "Yes he was, but he consumed about 2,000 calories worth of glucose over the course of the race, AND he was taking a bee pollen supplement that enhances fatty acid oxidation." You apparently have a source in addition to the link you provided, because the only thing Phinney says when asked what the runner ate is that it's "confidential research information".

    2. "What's really interesting is if you fast a house cat for more than a day or so, it can develop fatal fatty liver as it starts to burn it's own muscle tissue for GNG." Not quite. It's more likely in an obese cat. "Several hypotheses have been proposed as to why some cats, particularly obese cats, develop such severe lipidosis in association with anorexia. Cats of normal size are apparently able to tolerate long periods of starvation without developing this severe complication."