GCBC Reference Check ~ Part III of ? ~ Is glycerol phosphate rate-limiting?

In his most recent interview with Jimmy Moore, Gary Taubes did a bit of a mea culpa on the notion that dietary carbs are required to store fat.   He offers up a rather weak description of how he got it wrong for so long in his lectures ("skewed") and claims this wasn't something from the book but rather the lectures. Originally my post here stated that strictly speaking this was true, but upon rereading those sections of GCBC it is quite obviously not.  One could even go so far as to say it is the lynchpin of his hypothesis, but that is open to interpretation.  Still, Taubes repeated in the interview that G3P is "rate limiting" in the esterification process, a claim he made unequivocally  in GCBC.  Here's the relevant paragraph:
A single molecule plays the pivotal role in the system.  It goes by a number of names, the simplest being glycerol phosphate.  This glycerol-phosphate molecule is produced from glucose when it is used for fuel in the fat cells and the liver, and it, too, can be burned as fuel in the cells.  But glycerol phosphate is also an essential component of the process that binds three fatty acids into a triglyceride.  It provides the glycerol molecule that links the fatty acids together. †116  In other words, a product of carbohydrate metabolism -- i.e., burning glucose for fuel -- is an essential component in the regulation of fat metabolism: storing fat in the fat tissue.  In fact, the rate at which fatty acids are assembled into triglycerides, and so the rate at which fat accumulates in the fat tissue, depend primarily on the availability of glycerol phosphate.  The more glucose that is transported into the fat cells and used to generate energy, the more glycerol phosphate will be produced.  And the more glycerol phosphate produced, the more fatty acids will be assembled into triglycerides.  Thus, anything that works to transport more glucose into the fat cells -- insulin, for example, or rising blood sugar -- will lead to the conversion of more fatty acids into triglycerides, and the storage of more calories as fat.
In his footnotes, Taubes cites the following book:  Regulation in Metabolism by Newsholme & Start, 1973.  Here are two of the citations because I think their wording would indicate that Taubes feels this text is authoritative, and, presumably if he's recommending it, he has read it himself. 
An excellent review of the regulation of fat metabolism and adipose tissue is Newsholme and Start 1973:  195-246.
Glycerol phosphate:  For a review of the role of this molecule, the triglyceride/fatty-acid cycle, and the glucose/fatty acid cycle, see Newsholme and Start 1973:214-34.
OK, so I did go "see" for myself.  I've been somewhat sitting on this one for a while because I think it is particularly damning given specific wording of the excerpts from this text I'm about to share.   I wanted to share it first in my interview with Jimmy (even though that won't air for a few months).

What I transcribe below is the section in Chapter 5 Adipose Tissue and Fat Metabolism entitled "The control of esterification" that deals with this theory:  {} are my comments ;-) and I'll number the points in red so you can scan back to them as I discuss below.
In the case of the hypothetical, primitive animal it has been shown how large changes in blood glucose concentration could automatically cause reciprocal changes in plasma fatty acid concentration.  However, in the rat the blood glucose concentration does not fall by more than 30% during two days of starvation and this would appear to be insufficient to cause the marked increase in fatty acid mobilization that occurs under these conditions.  During the same period the plasma insulin level is decreased by about 80% (ref 41, see Table 7.3) and this hormone is known to regulate esterification.  Insulin stimulates the membrane transport of glucose into the adipose tissue cell and increases the rate of glycolysis and the glycerol phosphate concentration.  During starvation the reduction in the plasma insulin level results in a decreased rate of glycolysis and a lowering of the glycerol phosphate concentration.  This restricts esterification and consequently fatty acid mobilization is stimulated.   {I know what you're thinking ... so far so good right?} 
This theory is supported by the fact that the content of glycerol phosphate in adipose tissue is decreased during starvation and it is markedly increased when adipose tissue from a starved animal is incubated with glucose and insulin.  {Here is where I believe, having heard all he needed to hear, Taubes stopped reading ... but read on dear readers}   There are a number of problems associated with the simple idea that esterification is controlled by the membrane transport of glucose. [1] First, there is no evidence that the concentration of glycerol phosphate is limiting for the process of esterification (the Km for glycerol phosphate of the first enzyme in the pathway is not known).  [2]  Second, since glycerol phosphate dehydrogenase catalyses a reaction which is close to equilibrium, the concentration of glycerol phosphate can be controlled by both the cytoplasmic [NAD+]/[NADH] ratio and the concentration of dihydroxyacetone phosphate. (ref 42)
Dihydroxyacetone phosphate + NADH ↔ Glycerol phosphate + NAD+   
These quantities may vary independently of the glycolytic rate. (ref 17) [3] Third, the addition of adrenaline, fatty acids or acetate to the incubated fat pad preparation stimulates esterification but does not increase the content of glycerol phosphate. (ref 42)  This experiment suggests that factors other than the glycerol phosphate concentration can regulate esterification.  Such factors remain unidentified at present.  Nonetheless, it must be emphasized that a marked decrease in the concentration of glycerol phosphate could limit the rate of esterification and therefore variations in its concentration must always remain a potential mechanism of control.

The entire first paragraph essentially lays out a hypothesis, and evidence consistent with that hypothesis is included in the opening sentence of the second.  But the gist of the remainder of this paragraph is to lay out the case AGAINST it.

Point 1

I don't know how much more bluntly Newsholme and Start could have stated the first point.  To repeat:  

"there is no evidence that the concentration of glycerol phosphate is limiting for the process of esterification."

Just how does one go from that to:
In fact, the rate at which fatty acids are assembled into triglycerides, and so the rate at which fat accumulates in the fat tissue, depend primarily on the availability of glycerol phosphate.
???????????????????   You'll note I'm picking at Taubes' wording here too.  He uses that word fact.  Not probably.  Not "could".  Fact.  There is no ambiguity in his statement or any indication that there is even question at all.  

Point 2

The second point is really the biggie here.  Let me try to explain this to those without a background in chemistry.  All chemical reactions can theoretically proceed in either direction, so that if A + B → C + D, then C + D → A + B.  Many chemical reactions are considered irreversible because one direction is much more energetically favorable than the other.  The energy requirement to go in reverse is simply too high.  Combustion of hydrocarbons is a good example (e.g. burning propane).  But other reactions can go in both directions, we call these reversible reactions, and use the double arrow to indicate this (actually it's usually two arrows in opposite directions on top of each other, often with the length of the arrows reflecting the preferred direction).     If we put compounds A&B, C&D, or three or all four of them into a beaker, initially one reaction occurs at a faster rate but both are occurring until the system reaches a state of equilibrium.  At equilibrium, the concentrations of A, B, C, and D remain constant, and we can define an equilibrium constant (Keq) that represents a ratio of these concentrations.  This makes the system appear static, but it is actually a dynamic equilibrium with the rates of the forward and reverse reactions being equal.

There is a principle called LeChatelier's Principle that applies to systems in equilibrium.  In short, if we do anything to displace the system from its equilibrium state (in chemistry we usually referred to this as perturbing the equilibrium or applying a stress), the reaction rates will temporarily "shift" in order to re-establish the equilibrium.  One way we can do this is to add one or more of the compounds to the system.  The equilibrium constant Keq is called a constant for just that reason.  If I add more A to the system above that is at equilibrium, I have increased the concentration of A and thus altered K (which is a ratio of the concentrations).  The system will react so as to lessen the concentration of A so as to bring K back to Keq.  This would involve a "shift to the right" being an increase in the reaction rate of the forward reaction.  (If you're interested in the math involved, here's one example).

So what they are saying in the second point is that


is close to equilibrium and therefore should behave in the manner described above.  If we add DHAP then GP formation should increase.  If there's more NADH, it should as well, and if there's more NAD+ then the reverse reaction should "consume" GP and reduce its concentration.  But the concentrations of these other three components of the equilibrium system vary in ways that are not correlated with how much glycolysis is going on.  Were this system controlled simply by LeChatelier's Principle, then increases in DHAP (an intermediary in glycolysis) would exert predictable changes in the other three components of this system.  Rather, the conversion of DHAP to GP is determined by the activity of an enzyme.

Point 3

This part directly addresses whether it is the G3P level (however manipulated) that is rate limiting in esterification.  The addition of adrenaline, fatty acids OR acetate alone can increase esterification rate without a corresponding increase in the G3P level in the cell.   The text states that the mechanisms of this were not yet known, but it is not necessary to know them.  Other things -- hormones and/or metabolic substrate -- stimulate esterification irrespective of G3P levels.   The lack of change in G3P level indicates two possibilities: either basal G3P levels are in sufficient excess of needs that whatever stimulates esterification only depletes the G3P slightly as it is incorporated into triglycerides, or fatty acid uptake (or adrenaline or acetate) stimulates G3P production to meet the needs to esterify FA's but this additional G3P is consumed in the esterification process so that the level remains unchanged ... or both.

Bottom line:  

This reference DIRECTLY contradicts that which Taubes cites in GCBC and implies is discussed in this text.  At a minimum, Taubes should have done what he says he does  and follow the research forward to present day.  That there were factors unknown in 1973 should have led him to further research and no doubt to discover ASP.  But he clearly didn't want to.  
Throughout this process, I necessarily made judgments about the quality of the research and about the researchers themselves.  I tried to do so using what I consider the fundamental requirement of good science:  a relentless honesty in describing precisely what was done in any particular work, and a similar honesty in interpreting the results without distorting them to reflect preconceived opinions or personal preferences […]  I hope that I, too, will be judged by the same standard.    
~ Gary Taubes in Good Calories, Bad Calories

Just doing my part here.  Taubes' continued silence on these matters "speaks volumes".

If this comment remains here, it means that to date that you are reading this post, Gary Taubes has not formally addressed this very real criticism and issue with his writings.  He has also failed to set the record straight in any meaningful forum which is why you still have people claiming, to this day, that you can't store fat or get fat without dietary carbohydrate.

Last updated (mostly formatting changes) 9/23/14 .....


Melchior Meijer said…

One question keeps nagging in my head. How relevant is all this in real life? I mean, isn’t insulin an important driver of fat accumulation after all? Even if it’s not the only factor – which it isn’t, as you clearly demonstrate – don’t you admit it’s damned hard to gain fat mass with ‘normal’ or low (Kitava style) insulin levels? Or that it’s damned hard to lose fat with elevated (‘Western’ style) insulin levels? Insulin is at least necessary. Why is it easier to fatten a pig on corn and potatoes than on fat? Or are the farmers who claim this (like my grandfather) spreading an urban myth?

Don’t get me wrong. I love your work. Please keep poking. Just have to ask.
Sanjeev said…
>> damned hard to gain fat mass with ‘normal’
>> or low (Kitava style) insulin levels?

This is not Taubes's claim, exactly.

Folks have been doing carefully controlled studies on themselves - the careful control raises these above normal anectode, but still below formally controlled multiple-subject studies. Some have reported gaining fat mass on low to zero carb.

In the "real world", Millions of people went off Atkins. Some low-fatters have "I told you so"-ing and self-congratulatority, derisively chortling about this. Even if we grant that the majority were not doing Atkins correctly (too much cheating on carbs), do you thing every single one who gained weight on Atkins and quit it was, in fact, cheating? Every single one?

Until recently Taubes basically denied that it happens based on his biochemistry, NOT based on observation.

Next up, Alan Aragon did a research review on part of this question (paid paper & I don't have it but I know the conclusion) and concluded Insulin is not very strong at preventing fat burning.

And there are the studies showing that zero insulin is needed for ASP (although insulin accelerates the effects of ASP).

And I myself am one of those that got to the greatest fat mass of my life around 5 years ago on very low carb.
Sanjeev said…
James Krieger's insulin myths articles are wonderfully informative too.

here's one to get you started

For actual diet advice I HIGHLY recommend Lyle McDonald, Alan Aragon, Martin Berkhan and Leigh Peele.

And Mr. Krieger, (when he finally puts out advice for general consumption)
CarbSane said…
Agreed, which is why I have linked to them.

My thoughts on insulin have evolved from not giving it much thought (pre-discovering the LC community on the web, Taubes, Eades, etc.) to trying to reconcile the theories with losing weight just fine on low fat/high carb diets, to viewing insulin more as a *life force* hormone. To function optimally metabolically, we want to maximize our sensitivity to insulin and maintain a relatively normal body weight (for most of us that can mean 10,20 or even 50+ pounds of extra weight depending on how we carry it). In this state, we function efficiently, have less inflammation and oxidative stress, preservation of lean mass and store fat where it's supposed to be stored!
Melchior Meijer said…

It suddenly occured to me that insulinoma (rare insulin secreting neoplasms) would be a nice model of the isolated effects of hyperinsulinism. A quick look (due to connection problems) seems to reveal that weight gain is at least not universal (I saw a figure which said 20 to 40 % of cases become obese).

And guess what, insulinoma patients do not typically have hypertension.


Sigh... I guess you're right ;-).

The mean lesson seems to be: stay insulin sensitive, keep up the signalling. Eat less, move more, don't stress ;-).
CarbSane said…
The mean lesson seems to be: stay insulin sensitive, keep up the signalling. Eat less, move more, don't stress ;-).


I do believe that it is Taubes' misrepresentation of the direction of causality of insulin resistance and obesity that is most damaging in the long run.

Given all of the protective functions of insulin, I do not see it as sensible to merely strive to lower insulin levels. Postprandial insulin in a normal person does NOT cause chronic hyperinsulinemia. There's no evidence for this.

VLC has great utility in reversing IR initially. But as time passes (as reflected by my eating a bit more carb regularly these days) both anecdotal and study evidence seems to point towards VLC eating inducing insulin resistance in normal people. Whether this is sufficient to progress to T2 and related CVD implications is really unknown. We don't have long term studies on VLC eating. When I see low carbers mentioning that they can tolerate fewer and fewer carbs or they regain, and then seemingly regain anyway because they can't get satiated on an almost ZC diet, Houston ... we have a problem!
CarbSane said…
(Apologies in advance for some repetition in this comment)

Melchior, the more and more I learn about all of this the greater issue I have with this whole notion that insulin is a "fattening" hormone per se. Insulin is not required for fat accumulation in the body, it is required for the accumulation of excesses where they ought be (adipose tissue) and not where they don't belong in excessive amounts (leading to lipotoxicity and all sorts of problems). Insulin seems to play a secondary role in fat deposition, ASP is the main actor here and presumably other factors control the triglyceride forming enzymes within the cells themselves.

Taubes points the arrow of causation in the wrong direction in terms of the relationship between obesity and hyperinsulinemia. Chronic hyperinsulinemia is caused by increased adiposity, not the other way around.

Where this applies to real life in terms of battling obesity to begin with and reversing MetS or diabetes is understanding the relationship between obesity and insulin levels. One can decrease postprandial insulin on low carb controlled protein plans, but this does little to influence basal insulin independently of weight loss.

In normal people (which most of us are to begin with) insulin is secreted in response to diet in appropriate amounts so as to maintain circulating levels within a relatively narrow window. I've come across a lot of research demonstrating how these responses can be different even for the same foods depending on time of day eaten or whether it's a first, second or evening meal. Also, the insulin response can be different for the same quantities of food depending on what else it is consumed with (notably, 50g glucose + 50g protein generally elicits more insulin than the sum of the responses to either eaten separately). Also of interest, and perhaps why liquid calories are not well sensed, is that insulin response to infused glucose is lower than to solid ingested food.

I've got a pig post coming up ;)
Anonymous said…
Excellent beat ! I would like to apprentice while you amend
your web site, how could i subscribe for a blog website?

The account aided me a acceptable deal. I had been a little bit acquainted of this your broadcast provided
bright clear concept

my blog ... diet plans that work