Kurt G. Harris MD

PāNu means paleonutrition. The "paleo" here signifies "old" and not necessarily paleolithic. The PāNu approach to nutrition is grounded on clinical medicine and basic sciences disciplined by knowledge of evolutionary biology and paleoanthropology. The best evidence from multiple disciplines supports eating a pastoral (animal-based) diet rather than a grain-based agricultural one, while avoiding what I call the neolithic agents of disease - wheat, excess fructose and excess linoleic acid.

Support PāNu

PāNu is ad-free, completely independent and has no outside sponsorship. If you value PāNu, now you can support it. Read this for more information.

In addition to buying from the book list, you can also support PāNu by making all of your Amazon purchases for any item through the Amazon Portal below

Amazon Portal

« The PāNu Mission | Main | Optimal Nutrition by Kwasniewski »

Carbohydrates: No dietary requirement but metabolically critical 

Reader Mallory has asked some questions, which I will use as a framework for talking about the metabolic need for carbohydrates.

In response to "what are the Triglyceride/fatty acid/Krebs cycle and the Randle cycle and can you comment on Taubes description of these in GCBC?"

For starters, the triglyceride/ fatty acid cycle (not really a metabolic cycle in the sense of metabolism of substrates, just a feedback mechanism really) as described by Taubes has nothing whatever to do with the Krebs cycle. That is just his term for the equilibrium between esterification and lipolysis and the transport of Triglycerides back to adipocytes from the liver.

The Kreb's cycle is totally different, occurring inside mitochondria and is the keystone of aerobic respiration. That is what you are using when you burn fat or glucose aerobically and amino acids and ketone bodies can feed into it as well. The Kreb's cycle is very efficient, and is employed as long as there are enough substrates and oxygen available in most circumstances. It is not determined by diet, really, it is like your furnace if you live in wisconsin in the winter. It is always on to some degree when it has substrates in a functioning cell with mitochondria.

The Randle cycle also is not really a cycle of metabolic substrates the way the TCA or Krebs cycle is. The Randle cycle is just mutual feedback between glucose and NEFA (non-esterified fatty acids) or ffa (free fatty acids, same thing) that determines which is the predominant energy substrate. When glucose in the blood is high, glucose is the preferred energy substrate and glycerol phosphate, a metabolite of glucose, rises in the cell. This is the backbone of triglyceride (fat) and fat storage is thereby stimulated and lipolysis inhibited. Glycolysis is stimulated so more glucose can be burned (Usually in the Kreb's cycle)

(Important: note that inhibited does not mean stopped! It means the balance is towards fat storage but there is always lioplysis and esterification happening at the same time! Think of patrons passing by a crowded bar. Some enter and some leave. The change over time in the number of patrons in the bar is determined by the net difference between those leaving and those entering over time. That's integral calculus, basically. The point is, you don't have to stop people from leaving to get a bigger crowd, just shift the balance so more come in than go. )

Conversely, when NEFAs are high, the balance between lipolysis and esterification is shifted so there is net release of NEFA for use a substrate (usually burned in the Kreb's cycle again) and glycolysis is inhibited (again, not stopped, but inhibited) Part of this mutual feedback is being mediated by insulin, which is responding to glucose levels, among other things.

So the Randle cycle is just the mechanism whereby your body "knows" how to shift fuel sources based on fuel availability. Makes sense, as multicellular life was founded first on metabolism of glucose. The ability to store and burn fat (much more efficient than glucose) evolved later. You need mitochondria to bun fat. Almost any living cell can burn glucose, with our without mitochondria. It makes perfect sense that we are evolved to be able to use either glucose or fat for fuel.

PaNu dogma is that we are healthier when spending more time in fat-burning mode. This dogma does not, however require us to fantasize that we are somehow unable to metabolize glucose or that glucose is useless as an internal fuel source. It is adaptative that humans can survive well for long periods, especially when calorie restricted, with carbohydrate as the predominant fuel source. It is entirely consistent that fatty acids are the best fuel when available for most processes, but that glucose remains a backup fuel both on a dietary and internal metabolic basis. How long would our ancestors have survived if they were unable to eat plants?

Let's stipulate that there is no absolute dietary need for carbohydrate. Is there a metabolic need?


As most of you know, fatty acids cannot cross the blood/brain barrier. Neurons can shift about half their metabolism to ketone bodies, but still require a gradient of glucose that can diffuse into the brain. If your blood glucose level drops too low, you are in a coma. Also, red blood cells require glucose.

Finally, apart from these special situations, there is another very significant situation in which it makes perfect sense that we have retained the ability to burn glucose generally.

That is called anaerobic glycolysis.

When you need to perform a lot of work very quickly, like running from a predator, or spending twenty minutes lifting weights, you are not burning fatty acids. Aerobic glycolysis and lipolysis are very efficient but too slow for very high intensity tasks.

Any time you need high intensity work done by muscles faster than can be accommodated by aerobic respiration, you are absolutely, positively, using glucose for anaerobic glycolysis.

I don't care if you have not eaten of the plant world in eons, that is what is happening.

The question was asked: "does low insulin keep muscles from burning glucose"

For as long as there is aerobic work, with low insulin levels, fatty acids will be preferred. However, the instant anaerobic work is done, you are absolutely using glucose to do it because you cannot burn NEFAs anaerobically. If I lift weights or sprint on a 15 hour fast, my insulin levels are very low, yet I can instantaneously burn glucose anaerobically.

How about: "When insulin is low the liver does not give up the glucose."

Generally False. It's just the opposite. When fasting or just between meals, you are constantly burning a small amount of glucose in your brain (hopefully) and you need to constantly maintain your serum BG level. If you did not replenish your serum BG from somewhere, it would drop to dangerous levels. In order to make up for the glucose consumed, and in response to falling glucose, hormones like glucagon and epinephrine are released, which promote gycolysis of liver glycogen, which is then released into the blood.

Good thing, too, or I would quickly be in a coma with my daily 15 hour fasts.

Finally, I need to make sure everyone understands what is happening with glycogen in the liver if you eat no carbohydrates (or less than replacement levels, like I often do on VLC) .

It is still there. Really. You have lots of glycogen stored in your liver as the glucose storage depot to keep your blood glucose stable. If you don't eat enough carbohydrate, this is where gluconeogenisis comes in. GNG is induced when your liver glycogen has been depleted to a certain critical level. As discussed in my previous post, amino acids will be used to make new glucose (hence, gluco- neo-genesis) and the glycogen will be restocked.

The big difference between VLC and high carb eaters is in the turnover of liver glycogen. Th high carb eater is using it as his main fuel, so his "gas tank" is being both filled and emptied simultaneously at a higher rate.

Everyone has glycogen in their liver, and everyone uses it.

The Final Question asked: "Is excess protein just peed out?"

No. Unless your kidneys are diseased, there are no protein or amino acids in your urine. Excess protein not used for new protein synthesis may be burned (efficiently or inefficiently) or stored as fat. If metabolised, urea shows up in the urine as a byproduct, but amino acids are not just spilling out like your body doesn't know what to do with them.

To repeat from my prior post, there is no particular fate for any macronutrient in your diet, and any excess substrate with caloric value can be stored as fat or burned. If you can eat excess protein or fat or for that matter carbohydrate and not gain weight, that is only proving that your diet has not deviated from favorable insulin levels, your caloric intake is low, or whatever. It is not because that macronutrient "can't be" turned into fat.

Reader Comments (23)

great stuff Dr.

thank you,

August 28, 2009 | Unregistered Commenterd. cosa

Amazing stuff!

August 28, 2009 | Unregistered CommenterRoberto

Hi Kurt,

First off, great stuff as usual! When counting carbs if we in fact do, does fiber count? Does soluble/insoluble make a difference? I don't think so but I wanted to make sure. Thank you for a very informative blog.

August 28, 2009 | Unregistered CommenterMark

Great article!

I have a question - does eating protein cause the pancreas to release insulin before or after amino acids are converted to glucose? In other words, can the pancreas detect a rise in amino acids in the blood just as it detects a rise in blood glucose and respond with a release of insulin? I hope my question makes sense!

August 28, 2009 | Unregistered CommenterJoselyn

haha loved the post...my name included an everything. thanks for clearing it up. i cannot wait for my grassfed beef to get here!!!

August 28, 2009 | Unregistered Commentermallory


If I recall correctly, free amino acids themselves can stimulate insulin secretion.

August 28, 2009 | Unregistered CommenterRoberto

I would take integral calculus and differential equations over biochemistry any day! It's nice to have all of this important information spelled out in layman's terms. Thanks again for your ability and willingness to do this, Kurt.

What do you think about raising the anaerobic threshold through training? Is it worth the effort through HIIT or similar protocols to work at increasing the time in the aerobic zone (that term shows my age right there, lol) with increasing intensity?

As a martial artist, performance is important to me. I had no problem performing my art at regular proficiency while on a near zc regime. However, my muscles (legs in particular) progressively felt -at all times - as if in an increasing state of (what felt like) semi-cramping over the course of the 3 months I severely limited all plant foods. Now that I have added a few vegetables back and switched to a pretty strict 19/5 fasting/eating schedule, I find my health feels pretty similar to zc, but without any cramping.

I am definitely more interested in what works for me than following a preconceived assumption of some ideal diet. Your 12 steps seem very reasonable and flexible in this regard. Good stuff.

August 28, 2009 | Unregistered CommenterSatya

Thanks for creating this excellent blog.

Maybe at some point you'd care to comment on the "fourth macronutrient," ethanol. I've read that it, in fact cannot be converted to fat. Is this so?

August 28, 2009 | Unregistered CommenterTodd

Thanks Roberto. That's what I was thinking but I wasn't sure.

August 28, 2009 | Unregistered CommenterJoselyn

depends on how you look at it. Chemically ethanol carbon skeleton will rarely be incorporated into a fatty acid save for some special circumstances (majority of caloric intake from ethanol as with heavy drinkers). Nonetheless ethanol dehydration and subsequent acetaldehyde dehydration causes an overabundance of NADH and ATP. Both of them act together by inhibiting some reactions of the TCA cycle which results in citrate accumulation in the cytosol and fatty acid synthase induction.

August 29, 2009 | Unregistered CommenterAdrian

Thanks for that reply about ethanol. It confirms something I read, about an unhealthy "trick" used by supermodels to avoid weight gain: living on bacon and vodka.

Getting back on-topic, another perennial question has to do with insulin response to protein. There seems to be very little basic science on this. I recall Barry Sears claiming, in his first "Zone" book, that if one consumes more than about 35g of protein at a single sitting, the insulin response to that protein starts to resemble that of CHO. Or maybe it wasn't in his first book. In any case, no source was given. At any rate, this is why his diet is all about small meals and snacks scattered through the day. There is a well-known study of the "insulin index of common foods" by Jenny Brand Miller et al., which includes some meats, but this study doesn't plot the insulin response to various amounts of meat. It does suggest that the insulin reponse to meat is comparable to that of some fruits. But it was a small study, and additional information would be welcome.

August 29, 2009 | Unregistered CommenterTodd

From what I understand, that study, showing insulin response to meat, was done using EXTREMELY lean meat; and that fattier cuts would not have had the same response.

August 29, 2009 | Unregistered CommenterDaryl


"There is a well-known study of the "insulin index of common foods" by Jenny Brand Miller et al...."

I find this study fairly difficult to comprehend regarding insight into diet; for example, the foods were all provided in isocaloric doses (i.e. 1/3 lb of steak was compared to 3/4 lb of fish, to 1.3 lbs of oranges), which, while nice in one respect (easy comparison between foods), it doesn't give me any idea what goes on with a common meal portion. For example, would the insulin score be different if 100 Cal were consumed instead of 240? How does eating 2 lbs of oranges compare to 1 lb, compared to 1/2 lb?

Also, the foods were grouped in misleading ways; e.g. the "snacks" group contains a relatively large gradient, from peanuts (20+/-5) at the low end to Jellybeans (120+/-16) at the high end. The "protein" group also contained baked beans, which had around twice the score (120+/-19) of the next highest food in the group. As such, the study doesn't really make it easy to create a generalization, as comparing between groups is problematic.

Finally, because insulin can be released in anticipation of food, consider that a subject habituated to eating (e.g.) steak with soda, or fish with chips may dramatically throw off the reading. Although they tried to control for this (serving food from a covered box), I question the relevance of an "insulin score" measurement when the taste buds expect 1/2 lb of steak to be followed by fries and a coke.

August 29, 2009 | Unregistered CommenterChandler


Above you state:

"If you don't eat enough carbohydrate, this is where gluconeogenisis comes in. GNG is induced when your liver glycogen has been depleted to a certain critical level. As discussed in my previous post, amino acids will be used to make new glucose (hence, gluco- neo-genesis) and the glycogen will be restocked."

Does the same go for depleted muscle glycogen? I am a avid CrossFitter following a VLC paleo diet very close to what you have laid out on this site. At times I experience both light-headedness (brain fog?) as well as a reduced capacity to perform the most taxing (metcon-type) workouts. I'm thinking that the load that I am placing on my body is just too much for gluconeogenisis to keep up with and that I need to raise my carbohydrate intake somewhat, possibly with sweet potatoes or something else that minimizes insulin response. Any thoughts?

THANK YOU for what you are doing here! This site is having a huge positive impact on the health of both myself and my family.

August 29, 2009 | Unregistered CommenterRob


You are welcome

I am going to ask my brother-in-law, who is a Navy Seal and serious cross-fitter, to write a guest post for the site.

We both do workouts only in the fasting state with great results. We are both about 10% carbs.


More to come on protein and insulin.

August 29, 2009 | Registered CommenterKurt G. Harris MD

Hi Kurt,

I'm really excited for the post by your borther-in-law! Hopefully he can find the time to do it.

As always, thank you for what you do. I'm working real hard on not counting calories or carbs really, just trying to eat PaNu. I battle everyday with dairy because Cordain doesn't like it. I try to keep my mind from counting Zone blocks at every mea, thanks to Crossfit. I try to stop from obsessing over how to lose fat and just focus on eating PaNu. In an effort to do everything right with food, I think it's common for many around here to have a mild case of Orthorexia nervosa (denoting an eating disorder characterized by excessive focus on eating healthy foods). While I'm not diagnosing, I think I might qualify. It's just so hard to stop thinking about insulin storing fats versus eating high fat and therefore fat already being there and able to be stored easier than carbs. This is the kind of stuff that runs through my head all day every day.

Thanks again!


August 31, 2009 | Unregistered CommenterMark


post about orthorexia coming up - was writing it last night and my browser crashed!

August 31, 2009 | Registered CommenterKurt G. Harris MD


That would be great if your brother-in-law could write a guest post. If/when you talk to him, you can tell him that one of your dedicated readers absolutely died (unexpectedly) on 'Barbara' after three rounds. I was about 14 hours fasted and had been limiting carbs even more than usual. I am assuming that my body just isn't capable of replenishing my glycogen stores fast enough through gluconeogenesis, for this level of activity.

For those unfamiliar with CrossFit, 'Barbara' is one of the workouts and consists of the following:

20 Pullups, 30 Pushups, 40 Situps, 50 Squats
3 min rest between each of 5 rounds

August 31, 2009 | Unregistered CommenterRob


You mentioned your intake of pastured cream and butter; would you be kind enough to provide the nutritional breakdown of these items (Calories, Pro, Cho, Fat (SAT, MUFA, PUFA)) as I'm confused in my local stores about which is the better option per 100ml / 100g. I see cream such as Clotted, Double, Single, Whipping, Fermented etc. and butter as salted, unsalted, organic and so on. As much as grass fed, pastured red meat is preferred I'm confused on the dairy fat options to purchase.



August 31, 2009 | Unregistered Commenterwinalot


read the labels and count the usual 4 kcal.g for pro and carbs and 9 for fat

ignore the carton description, but

double, whipping, and whole are generally good

As far as MUFA PUFA, etc try wikipedia for estimates based on corn fed

I have a paper on this somewhere - perhaps a future post.

August 31, 2009 | Unregistered CommenterKurt G Harris MD

Great post.

Yesterday at Carnivore Health there was a supposed metabolic Eureka! moment and they stated there was another hormone that could create fat:

"An enzyme called ASP (Acylation Stimulating Protein). This little jewel has the ability to directly store fat in the fat cells completely bypassing the glucose and insulin pathways."

Is this the major insight they are claiming? It seems unlikely that such a simple end-around had been overlooked by Taubes.

September 1, 2009 | Unregistered Commentermarjorie


Just to clarify carnivorehealth.com is run by a 24 year old layman (me). Taubes didn't overlook anything, if you read the post again ASP and HSL are still controlled by almighty insulin.

It was a eureka moment because I originally thought, after Lex Rooker's experiment, that a high fat diet would create too much glycerol phosphate and ultimately lead to weight gain.

The weblog is just a outlet for me to share experiences and articles while I'm on my way to perfect health.

September 1, 2009 | Unregistered CommenterDanny Roddy

usually i'm vlc or zc, though sometimes in order not to give my body forget how to handle carbs, i eat a meal of considerable carb content (sweet potato, pumpkin or something); i have read that eating fat alongside with carbs reduces insulin spike associated with the latter; so, when you eat carbs, should you add some fat (lard or chicken fat) to them?

September 8, 2009 | Unregistered Commentergn
Comments for this entry have been disabled. Additional comments may not be added to this entry at this time.