Monday, July 18, 2011

Dietary protein does not become body fat if you are on a low carbohydrate diet

By definition LC is about dietary carbohydrate restriction. If you are reducing carbohydrates, your proportional intake of protein or fat, or both, will go up. While I don’t think there is anything wrong with a high fat diet, it seems to me that the true advantage of LC may be in how protein is allocated, which seems to contribute to a better body composition.

LC with more animal protein and less fat makes particularly good sense to me. Eating a variety of unprocessed animal foods, as opposed to only muscle meat from grain-fed cattle, will get you that. In simple terms, LC with more protein, achieved in a natural way with unprocessed foods, means more of the following in one's diet: lean meats, seafood and vegetables. Possibly with lean meats and seafood making up more than half of one’s protein intake. Generally speaking, large predatory fish species (e.g., various shark species, including dogfish) are better avoided to reduce exposure to toxic metals.

Organ meats such as beef liver are also high in protein and low in fat, but should be consumed in moderation due to the risk of hypervitaminosis; particularly hypervitaminosis A. Our ancestors ate the animal whole, and organ mass makes up about 10-20 percent of total mass in ruminants. Eating organ meats once a week places you approximately within that range.

In LC liver glycogen is regularly depleted, so the amino acids resulting from the digestion of protein will be primarily used to replenish liver glycogen, to replenish the albumin pool, for oxidation, and various other processes (e.g., tissue repair, hormone production). If you do some moderate weight training, some of those amino acids will be used for muscle growth.

In this sense, the true “metabolic advantage” of LC, so to speak, comes from protein and not fat. “Calories in” still counts, but you get better allocation of nutrients. Moreover, in LC, the calorie value of protein goes down a bit, because your body is using it as a “jack of all trades”, and thus in a less efficient way. This renders protein the least calorie-dense macronutrient, yielding fewer calories per gram than carbohydrates; and significantly fewer calories per gram when compared with dietary fat and alcohol.

Dietary fat is easily stored as body fat after digestion. In LC, it is difficult for the body to store amino acids as body fat. The only path would be conversion to glucose and uptake by body fat cells, but in LC the liver will typically be starving and want all the extra glucose for itself, so that it can feed its ultimate master – the brain. The liver glycogen depletion induced by LC creates a hormonal mix that places the body in fat release mode, making it difficult for fat cells to take up glucose via the GLUT4 transporter protein.

Excess amino acids are oxidized for energy. This may be why many people feel a slight surge of energy after a high-protein meal. (A related effect is associated with alcohol consumption, which is often masked by the relaxing effect also associated with alcohol consumption.) Amino acid oxidation is not associated with cancer. Neither is fat oxidation. But glucose oxidation is; this is known as the Warburg effect.

A high-protein LC approach will not work very well for athletes who deplete major amounts of muscle glycogen as part of their daily training regimens. These folks will invariably need more carbohydrates to keep their performance levels up. Ultimately this is a numbers game. The protein-to-glucose conversion rate is about 2-to-1. If an athlete depletes 300 g of muscle glycogen per day, he or she will need about 600 g of protein to replenish that based only on protein. This is too high an intake of protein by any standard.

A recreational exerciser who depletes 60 g of glycogen 3 times per week can easily replenish that muscle glycogen with dietary protein. Someone who exercises with weights for 40 minutes 3 times per week will deplete about that much glycogen each time. Contrary to popular belief, muscle glycogen is only minimally replenished postprandially (i.e., after meals) based on dietary sources. Liver glycogen replenishment is prioritized postprandially. Muscle glycogen is replenished over several days, primarily based on liver glycogen. It is one fast-filling tank replenishing another slow-filling one.

Recreational exercisers who are normoglycemic and who do LC intermittently tend to increase the size of their liver glycogen tank over time, via compensatory adaptation, and also use more fat (and ketones, which are byproducts of fat metabolism) as sources of energy. Somewhat paradoxically, these folks benefit from regular high carbohydrate intake days (e.g., once a week, or on exercise days), since their liver glycogen tanks will typically store more glycogen. If they keep their liver and muscle glycogen tanks half empty all the time, compensatory adaptation suggests that both their liver and muscle glycogen tanks will over time become smaller, and that their muscles will store more fat.

One way or another, with the exception of those with major liver insulin resistance, dietary protein does not become body fat if you are on a LC diet.


gallier2 said...

Warburg effect is about glycolysis, not about glucose oxidation, isn't it?
Or did you mean that glucose oxidation damages (in the long run) the mitochondria which leads the cell to glycolysis for its sole ATP source, disabling apoptosis as a side effect? (Peter from hyperlipid hypothesized once in that diection)

Alex said...

Hi Ned, What range of carbs would you consider it to be LC? less than 100, 150?

quality said...

yes low carbohydrate diet not allow your body to became fat.


sustenir said...

What happens if one consumes a large amount of fructose postprandially along with other carbs? It seems this would inhibit hepatic glycogenesis, perhaps leading the muscle glycogen stores to be refilled more quickly. Or maybe the carbs would just get stored as fat?

Ned Kock said...

Hi gallier2. That is a very good question, one that I grappled with for quite some time after I first heard about the Warburg effect. From what I understand, Warburg effect refers to “aerobic” glycolisys, where oxygen is used and free radicals of a particular type are produced: reactive oxygen species. Aerobic glycolisis is so strongly associated with tumor growth that it is used as a marker of tumor growth in diagnostic tests.

Ned Kock said...

Hi Alex. The intake of carbs should be low enough at least to induce some of the responses associated with LC. For example, one of the hallmarks of LC is ketosis at rest. (Ketosis usually happens after intense exercise even if you are on a high carb diet.) Different people have different thresholds, and those vary at different points in time. For me now, it seems that fewer than 100 g/d lead to ketosis at rest. In the past, it used to be less than that, as far as I remember.

Ned Kock said...

Hi sustenir. Fructose actually accelerates liver glycolgenesis, a lot. But this happens only until the liver tank is filled. After that, fructose will be converted into fat and secreted in VLDL particles. (At least this is the current consensus.)
At this point, what you said may happen – liver glycogenesis will be halted, and blood glucose levels will go up and stay up for longer. As glucose will stick around longer it may be taken up in higher quantities by muscle, but it will also be oxidized for energy.
If glucose is around in the blood the body will prioritize it for oxidation over almost anything else, including amino acids. The only exception as far as I know is ethanol, which is given higher priority, and, like amino acids, is not converted to fat either.

David Isaak said...

Hi, Ned--

The tank-refilling aspect of a high-carb day, or days, is employed in Jeff Volek's "TNT Diet" ( If I follow a "diet" these days, it's probably that one.

His diet plan is in stages, and high-carb days are not added until most fat loss goals are nearly achieved.

I was interested by your comment that muscle glycogen is replenished only slowly, and that mainly from liver stores. Can you elaborate?

shooflypie said...

I don't understand how someone can "keep their liver and muscle glycogen tanks half empty all the time." Unless you're on a starvation diet, won't your body just turn whatever you're eating, be it protein or carbohydrates or fat, into glycogen?

Or wait, you can't turn fat into glycogen. So you could have plenty of energy (e.g. from ketones), but not be replacing your glycogen if you're eating a lot of fat and not enough protein/carb?

But why would chronically not having enough glycogen mean that your reserve capacity would get smaller in response? On the contrary, I would think that keeping your glycogen reserves constantly full and never using them up would mean that you would compensate to store less, since in that case you don't seem to need as much storage.

I'm confused!

Amber said...

I don't understand your statement that 600g of protein is too much by any standard. It seems like any other standard is not taking this situation into account, and wouldn't apply. If all the protein is getting used for glycogen, then why would it be too much? Is there some reason to minimize excess protein that is independent of how it is being used? It seems to me that in this case, none of the protein involved would be "excess", and so those standards are inappropriate.

Amber said...
This comment has been removed by the author.
Ned Kock said...

Hi David. Much of the postprandial substrates for glycogenesis (glucose and fructose) never make it past the liver, in part due to the proximity to the portal vein and the fact that the liver glycogen tank is bigger than most people think it is.

Muscle gets the scraps, so to speak. Because of that, glycogen synthase activity remains elevated for much longer in muscle, often for days. In between meals, part of the glucose secreted by the liver to feed the brain is taken up by muscle tissue.

This is one of the reasons why anaerobic exercise increases insulin sensitivity for days after the exercise session.

Ned Kock said...

Hi shooflypie. If you keep them full all the time, the body should reduce their size because the signal will be that storage is unnecessary (since they are full all the time because of constant input). If you keep them empty all the time, the body should reduce their size because the signal will be that there is not enough to be stored.

The way to get them to increase in size is to fully deplete them (in a manner of speaking, because they cannot be fully depleted), and then fully replenish them, periodically. This way, the signal to the body will be that it really needs large glycogen tanks.

Ned Kock said...

Hi Amber. There is a limit on the amount of byproducts of protein digestion and metabolism that the body can handle, which is reflected in a limit on protein consumption. Usually this limit is underestimated, in my opinion. But 600 g/d of protein is probably more than the body can handle.

In excess, even water kills.

BigWhiskey said...

Ned? An animation of these processes would be great!! Maybe, I am just a "Draw me a picture" type. Any recommended sources? I know this is a labor of love for you and this is provided to us at no cost; I thank you very much for the way it is.

David Moss said...

This is an interesting, but very counter-intuitive hypothesis Ned. I say counter-intuitive, because prima facie, it seems odd that it would be the allocation of the smallest portion of one's calories (from protein) that makes the metabolic difference. I would have thought that low carbers are typically eating around 15-20% of calories from protein. At 70kg and doing intense exercise, I couldn't eat more than 20% for more than a couple of days without feeling sick.

The least plausible explanation for your conclusion seems to me to be the decreased “caloric value” of protein. Assuming I'm getting slightly under 500 calories from protein, there doesn't seem much scope for calories to be lost. Nor does this mechanism seem to capture what most people experience when they increase protein- substantially increased satiety and not only that, but a definite urge to not eat any more protein, after a surprisingly short while.

Relatedly, 'caloric density' doesn't seem like a compelling explanation for LC's successes. Even assuming I eat 25% of calories from protein (more akin to atkins) and that protein's caloric density decreases to 2 calorie per g, I still only need to eat 520g of food on LC (in the utterly unrealistic world where the caloric density of a food is wholly determined by the caloric density of its macronutrients) compared to 625g of food on high carb + 25% protein.

I'm also confused by this passage:
“In LC, it is difficult for the body to store amino acids as body fat. The only path would be conversion to glucose and uptake by body fat cells” [which is the same pathway as in HC]
“but in LC the liver will ... want all the extra glucose for itself, so that it can feed ... the brain.”
[So of course in LC my peripheral tissue will be insulin resistant and thus not use any excess glucose generated from protein. So instead, the 40g or so excess protein I consume indirectly feed the brain, rather than fat tissue. But I don't see why this makes LC+protein more efficient. It seems to be begging the question against HC to say that those 40g would otherwise have turned to fat, rather than fuelling body and brain as glucose.]
“The liver glycogen depletion induced by LC creates a hormonal mix that places the body in fat release mode, making it difficult for fat cells to take up glucose via the GLUT4 transporter protein.”
[So glycogen depletion and LC places the body in fat release mode, gotcha, but where's the [dietary] protein in this?]

Ned Kock said...

Hi BigWhiskey. Even a simulator would be useful. I don’t know of any for this.

Ned Kock said...

Hi David. Let me just be clear about something here: you do need dietary fat in your diet; it is part of too many tissues in the body that need constant repair. How much fat you need depends on your caloric expenditure and whether you are growing or not, among other factors.

LC with more dietary protein and less fat seems to be particularly useful for those folks for whom body fat loss has stalled on a high fat LC diet.

malpaz said...

some clarity, i am very confused at what you mean. amino acids and protein cant be stored as fat in a LC environment?? would that be why protein spike insulin so when it needs to be stored as fat it can be? or why people get hypo-symptoms on PSMF yet have high BS readings?

Ned Kock said...

Hi Mal. The insulin secreted in response to dietary protein promotes dietary fat storage in adipocytes, and also liver glycogenesis. Since LC causes liver glycogen depletion, liver glycogen synthase activity will also be increased. So you have two factors that strongly promote liver glycogenesis based on whatever glucose results from amino acid metabolism.

When a person doing LC eats a fatty steak, it is the fat in the steak, not the protein, which will quickly become body fat. That is driven by the insulin secreted in response to the protein. That is fine up to a point; we are designed to store body fat in adipocytes for later use as energy.

At the same time, the insulin secreted in response to dietary protein promotes muscle protein accretion, as long as there is also a stimulus at the muscle-tissue level. That stimulus comes from, interestingly enough, glycogen depleting exercise (and thus glycogen depletion at the muscle-tissue level).

So, LC with more dietary protein and less fat, together with weight training, has the potential to promote muscle gain and body fat loss at the same time. That is, as long as calorie intake is not too low. If calorie intake is too low the body will catabolize muscle, in addition to body fat, to supply its energy needs.

This is a difficult balance, for which the body relies in part on hunger. We can help by making some dietary choices, and I’m not talking about protein powders with zero fat. For example, shrimp’s high protein-to-fat ratio makes it a natural alternative to protein powders. The n3-n6 ratio is also pretty good. The same applies to crab, octopus, and mussels.

malpaz said...

oh ok i get it, gracias!!

David Moss said...

Ned, I'm not sure what in my comment made you think either that I believe you don't need fat in your diet or that I believe that you believe you don't need fat in your diet.

I grant that increased protein might be good for fat loss, but would think that would be due to increased satiety, not from 'protein not becoming fat' or any such.

My query was about the claim that "the true advantage of LC may be in how protein is allocated." I don't understand any of your reasons for thinking that. You suggest that the excess protein (once turned to glucose) won't turn to fat, because the body will send it all to the brain. Maybe so, but then the point is only that: 'limited amounts of glucose don't turn to fat on a low carb/glucose diet [so long as it's so low that all glucose is scavenged by the brain].' And I don't see what that claim, or what you said about glycogen depletion and fat-burning have to do with dietary protein.

Ned Kock said...

Hi David.

> I don't see what that claim, or what you said about glycogen depletion and fat-burning have to do with dietary protein.

The answer can be found through a simple exercise. Go through all of the macronutrients, and you’ll see that the points made on this post apply only to protein. Take alcohol for example. Like dietary protein, it doesn’t become body fat, but it does not become glycogen (in liver or muscle tissue) either.

Btw, I am not trying to convince you by repeatedly re-stating what I said before in different ways; I never try to do that here. If you disagree with me, I am fine with that. No problem. I may well be wrong.

Stargazey said...

Ned, anecdotally, a group of low-to-no-carbers noticed that eating too much protein led to high blood glucose. I discussed a possible mechanism for it here.

Even if the conversion of protein to fat is metabolically unlikely, the consumption of excess protein can present other health risks that need to be considered.

malpaz said...

i have gained weight in the past on high protein low carb... but 2 things. i included dairy which assumes a higher than needed fat content, and i put on a lot of muscle in the process.

about the brain getting excess amino/protein energy... i find a high protein diet keep me level mentally but also keeps me 'going' mentally at night. if the excess protein doesnt spike BS/insulin/glucose will it lead to longer than needed periods of elevated BS?

i think the main point is anything in excess is going to be stored as fat, but fat being WHAT is stored macro wise. i think you would die before you gained weight eating a fat free all protein diet. or become to dry skinned and constipated before diarrhea from rabbit starvation... i dont see the 'point' in the post because most people know if you eat in excess its the dietary fat thats stored, very effectively

Ned Kock said...

Hi Stargazey. Thanks for commenting and for the links; I read both posts with interest. I also went to Jenny’s calculator. I don’t know if you realized this, maybe you did - the calculator’s recommendations are very consistent with this post.

In fact, after I entered my data it recommended a high protein (200 g/d) and relatively low fat (100 g/d) intake to “avoid catabolizing muscle” at a 50 g/d carb intake. One key factor in the calculation seems to be the amount of exercise I do.

I think that 200 g/d of protein is more than what I currently consume. Moderate weight training tends to contribute to placing the body in positive nitrogen balance. I’m not sure the calculator differentiates between this type of exercise and aerobic exercise, which actually pushes you into negative nitrogen balance territory.

For others interested, the calculator is here:

Ned Kock said...

Hi Mal. The brain is fueled primarily by glucose and ketones.

I assume you are talking about insomnia. The only connection I can see is with calorie restriction. Protein is not very calorie-dense and severe calorie restriction often leads to mental disturbances, including psychosis.

If one eats 200 g/d of protein (not a low intake), and nothing else more, chances are that’ll be severe calorie restriction territory.

Stargazey said...

Ned, thanks for reading and considering my posts. I realize that you are mostly discussing dietary protein in the context of an exercise program. However, it's easy for us sedentary types (especially those of us over 50) to take your conclusions and use them to substitute protein for carbs and then wonder why our blood sugar is much higher than we're used to. Caveat consumptor!

I don't see Malpaz suggesting a protein-only diet. Kimkins taught us about the dangers of rabbit starvation. However, when you eat lots of protein plus extra fat or carbs but do not consider total daily caloric expenditure, if the protein doesn't go to fat, then the fat or carbs will. Because of that, people are able to increase their body fat on a zero-carb diet. (Sorry, Malpaz, if I misinterpreted your comment.)

Ned Kock said...

Hi Stargazey. I think you are generally correct, and this is something that deserves a separate post. In fact, for those who don’t exercise (or exercise very little), a high fat and moderate protein approach may not be a bad idea at all. This is something more in line with the approach that our good friend Peter follows and talks about over at Hyperlipid. (He may also consider starting another blog – Hypermover.)

Btw, I suspect that the results you’ve seen in your survey are in part due to insulin resistance. IR is positively associated with age, and also with body fat mass. Body fat mass might have gone up for several of those folks as they increased their protein intake without significantly decreasing their fat intake.

I am guessing here, but I think it is a reasonable guess based on the results you got.

David Isaak said...

As a follow-on to Stargazey's comments, many people have reported blood-sugar spikes from particular forms of protein--especially rapidly absorbed forms such as whey.

The case of whey is especially interesting, as two groups seem to use it in large volumes--bodybuilders on the one hand, people on quick weight-loss programs like the Eades 6-Week Cure.

Bodybuilders tend to report muscle gain and leaning out.

On the other hand, many folks on the 6-Week Cure reported huge spikes in blood glucose, and a number of those people reported no weight loss (even though it is a very calorie restricted diet.

The people who reported those blood-sugar spikes tended to be diabetic (who else measures their blood sugar that often?) and relatively sedentary (the two often go together).

So, metabolically, I think what happens with excess protein depends at the minimum on 1) how rapidly absorbed it is, and 2) exercise levels, glycogen tanks, and the rest, and 3) half a dozen other factors I probably haven't thought of.

David Isaak said...

BTW, I'm reading an interesting eBook on protein by Brad Pilon. He points out something I hadn't really thoght about before:

"Here is some more little known information about protein. If you were to eat 50 grams of protein on any given day, you would actually be digesting between 100 and 150 grams of protein! This is because your body also digests an additional 50 to 100 grams of endogenous protein every single day. Endogenous protein is the protein that is secreted into your gastrointestinal tract by your body
(endogenous means that your body produced it without getting it from your diet).

These proteins come from saliva, gastric juice, pancreatic enzymes and other secretions as well as intestinal cells and proteins that have leaked into the intestine from the blood. We basically have a built-in protein recycling operation, utilizing ‘scrap’ proteins that are no longer needed by the body to build new proteins that are ‘in demand’."

He then goes on to discuss the new idea that we may also absorb some proteins produced by gut bacteria. (Their role in providing short-chain fatty acids is well-known.)

I'd never really thought of protein intake before in the context of the body's overall protein flux...

Stargazey said...

Ned, I would agree that high blood sugars in response to excess protein are in part due to insulin resistance. Dr. Richard Bernstein makes a point of telling diabetics to control their protein intake as well as their carb intake, but he is one of the few who realize that protein intake provokes an insulin response.

As David Isaak said, the Eades' 6-Week Cure for the Middle-Aged Middle calls for lots of whey shakes and produces high blood sugars in some people. That suggests that the shakes may increase blood sugar in everybody, but for those with an intact insulin response system, insulin covers it and the effect is not seen if it is measured with a glucometer alone.

malpaz said...

on the protein excess...many, even lots to argue bodybuilders who eat insatiable amount of protein go on to have diabetes...

Ned Kock said...
This comment has been removed by the author.
Ned Kock said...

That idea of endogenous protein digestion is very interesting David, but the amount (50-100 g) sounds quite high!

Anyway, there is also the recycling of amino acids that goes on at the cell level, related to autophagocytosis by the cell’s cleaning up crew (lysosomes).

Ned Kock said...

Mal, based on what I’ve read before as a basis for some posts here, I think that bodybuilders often develop diabetes because they become obese as they get older. Not because of the high protein intake earlier. I may be wrong though.

Ned Kock said...

Protein normally leads to a significant insulin response, of the same order as natural carb-rich foods, and less than refined carb-rich foods.

The link below is for a post summarizing Holt et al’s study on the insulin responses to carbs and protein, which is a classic, and apparently fairly reliable:

But typically the glucose response to protein ingestion is insignificant. The main reason is that glucagon and insulin are secreted together by the pancreas.

A couple of possible reasons for the glucose response to protein are: (a) mildly impaired beta-cell function combined with non-adapted alpha-cell function; and (b) impaired beta-cell function combined with adapted alpha-cell function.

If someone has mildly impaired beta-cell function, and non-adapted alpha-cell function, I can see the possibility of the glucagon-to-insulin ratio going up immediately after a protein meal. The glucose response would be due glycogenolysis in the liver, not production of glucose based on amino acids. I would expect this to be observed in folks with early stage impaired beta-cell function.

If someone has impaired beta-cell function, and adapted (or matched) alpha-cell function, the glucagon-to-insulin ratio may not go up after a protein meal, but there may be no insulin to deal with the glucose produced later based on glucogenic amino acids (alanine, glycine etc.). I would expect this to be observed in folks who’ve had impaired beta-cell function for a while.

Stargazey said...

Ned, your explanations of alpha and beta cell function sound right to me. Thanks for bringing up the alternative scenario, which I tend to forget for some reason.

malpaz said...

is there a post on beta cell functioning, that's a new one for me. I am confused again :)

Stargazey said...

I don't have a post, Malpaz, but here's an article: Five Stages of Evolving Beta-Cell Dysfunction During Progression to Diabetes.

David Isaak said...

"That idea of endogenous protein digestion is very interesting David, but the amount (50-100 g) sounds quite high!"

Yes, the number surprised me, too, but when I think about the rate at which gastrointestinal cells are replaced, I'm more inclined to consider it. (200 square meters, replaced every 5-7 who knows how many digestive enzymes?)

I think we tend to represent our bodies as a series of boxes that things are put into or taken out of, when in fact when we are healthy there is a good deal more flux.

Ned Kock said...

Hi Mal. In simple terms, the pancreas talks to the liver via two main hormones – insulin and glucagon. Insulin tells the liver to store sugar (as glycogen), and glucagon tells the liver to release sugar. Insulin is secreted by what are known as beta cells in the pancreas; glucagon by alpha cells.

In humans, diabetes is a disease of impaired beta cell function, where insulin is produced in insufficient quantities (type 2 diabetes) or not at all (type 1 diabetes), with many levels in between. In birds, diabetes is (usually) a disease of impaired alpha cell function, where glucagon secretion is messed up.

The main function of the hormones insulin and glucagon is to allow the pancreas to talk to the liver.

BigWhiskey said...

Ned, thanks for having the patience to break ideas and processes down to "in other words" level for the lay-folks such as yours truly.

garymar said...

I didn't know birds got diabetes.

More fun bird facts: in chickens it's the female who has the Y chromosome, the reverse of mammals. Actually they call it a W chromosome: hens have ZW chromosomes but roosters have ZZ.

Ned Kock said...

You’re welcome BigWhiskey. It helps me too.

Ned Kock said...

Thanks garymar, very interesting.

David Isaak said...

Lyle McDonald makes a good case for what you have been arguing: