LC Blogger Answers the Question: Are Alcohol and Carbs Fattening? Part I

I have to give credit where credit is due. In his zeal to dismiss the role of calories in weight (fat) gain or loss, Ian Lane makes a pretty good case for why alcohol is not likely to be the carbon source for accumulated fat.  In doing so, however, he accidentally all but exonerates the lowly carbohydrate.  I don't think he meant to do that ...  The particular post in question was the subject of a couple of blog posts recently, see here and here.   It would appear that it has now been officially (in the title line) "revised" to eliminate all of the overt scientific inaccuracies, and focuses on ethanol, which makes it easier to direct readers of this post to just the part in question!  Ian writes:

The molecule I'd like to highlight, which, from my perspective, cannot realistically be stored as body fat, is ethyl alcohol, or ethanol. (That’s not to say that ethanol can’t possibly aid in fat accumulation somehow — see below — rather, the calories in ethanol, themselves, are not likely serving to create new fat.)

After now anonymously calling out some "nutrition professional" (rather than the previous identification of said person) Ian writes:

Again, my point is not that “alcohol consumption isn’t associated with increased body fatness,” per se. (Although, as shown above, this depends on the study you choose to look at.) My contention is with the assumption — purported as fact — that the “overconsumption of alcohol calories [themselves] cause increased body fatness.”

What's with this "themselves" nonsense?  How many "SAD" eaters are consuming even the majority of their calories from one macronutrient?  Carbs you say?  Actually on average we barely reach 50%, so not even those.  So I'm not sure what Ian's obsession with the "calories themselves" from any particular source is, because one could extend that quite easily to the foods themselves being the cause of accumulated fat.    But of the macronutrients, here are the ones least likely to cause obesity "by themselves" in the order of likelihood:  1. alcohol, 2. protein, 3. carbs, and 4. fat.    You'd poison yourself before obtaining sufficient calories from alcohol, and the same goes for protein to a slightly lesser extent.  Just by themselves, carbs are almost as unlikely as alcohol to find their way into body fat (though they can), which leaves fat.  Interestingly, dietary fat is incredibly efficient at becoming body fat, most especially because it doesn't need to be transformed in order to do so!  It is also interesting that the above list is exactly the hierarchy of macronutrient priority in metabolism when consuming the more realistic "mixed meal".

So after dissing (perhaps rightly, perhaps prematurely) observational studies, Ian turns his attention to sifting through  

Google Scholar or the PubMed database in search of controlled studies on ethanol consumption and obesity, many interesting articles pop up, but only a few caught my attention.[3-5]

Are there any controlled studies on ethanol consumption and obesity?  Have they taken a group of people, randomized them, and then asked half of them to now consume 20% of calories as alcohol for a year?  I'm not aware of any such study ... it would fail to meet the ethics challenges given what we know about alcohol.  So here are the three studies that caught Lane's attention (he needs to revise once more as he subsequently refers to "the next three" studies, meaning these. )

Reference #3 - Suter, Paolo M., Yves Schutz, and Eric Jequier. “The effect of ethanol on fat storage in healthy subjects.”

I include the authors, because this one bears the name of Eric Jequier whose work has focused on the metabolic fate of macronutrients.  I wonder if Lane ever stumbled across Jequier's 2002 paper in Nature entitled Pathways to Obesity.  You see, here is where the cherry picking gets rather ripe.  In Pathways, Jequier concludes:

2. Postingestive fuel selection favours the oxidation of dietary proteins and carbohydrates, whereas dietary fats are preferentially stored as triacylglycerol in adipose tissue. Alcohol, by inhibiting lipid oxidation, indirectly favours the storage of dietary fats.

Of course calories are part of the remaining conclusions, but this is the part that goes to Ian's point.  These biochemical interactions of molecules with the cells' essence of life or somethingorother.  In Ian's reference, written a decade prior to Pathways, Jequier concluded:

Ethanol, either added to the diet or substituted for other foods, increases 24-hour energy expenditure and decreases lipid oxidation.  Habitual consumption of ethanol in excess of energy needs probably favors lipid storage and weight gain.

Reference #4 - Shelmet, John J. many more, and Guenther Boden.  “Ethanol causes acute inhibition of carbohydrate, fat, and protein oxidation and insulin resistance.”

ABSTRACT:

To study the mechanism of the diabetogenic action of ethanol, ethanol (0.75 g/kg over 30 min) and then glucose (0.5 g/kg over 5 min) were infused intravenously into six normal males. During the 4-h study, 21.8 +/- 2.1 g of ethanol was metabolized and oxidized to CO2 and H2O. Ethanol decreased total body fat oxidation by 79% and protein oxidation by 39%, and almost completely abolished the 249% rise in carbohydrate (CHO) oxidation seen in controls after glucose infusion. Ethanol decreased the basal rate of glucose appearance (GRa) by 30% and the basal rate of glucose disappearance (GRd) by 38%, potentiated glucose-stimulated insulin release by 54%, and had no effect on glucose tolerance. In hyperinsulinemic-euglycemic clamp studies, ethanol caused a 36% decrease in glucose disposal. We conclude that ethanol was a preferred fuel preventing fat, and to lesser degrees, CHO and protein, from being oxidized. It also caused acute insulin resistance which was compensated for by hypersecretion of insulin.

So I included Boden in the author list because he's a pretty major researcher in the area of my "obsession" -- that being the role of free fatty acids, or my preferred acronym NEFA, in the etiology of diabetes.  See for example Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and beta-cell dysfunction  (full text unfortunately not free).  

What's this hypersecretion of insulin?  And that ethanol shut down that fat burning beast like crazy!  Surely this in vivo biomolecular chemical cell metabolism interaction like stuff about alcohol -- mentioned in just the abstract -- caught Ian's attention?  Apparently not long enough to do more than list the citation ...  Lastly, 

Reference #5 - Siler, Scott Q., Richard A. Neese, and Marc K. Hellerstein. “De novo lipogenesis, lipid kinetics, and whole-body lipid balances in humans after acute alcohol consumption.” 

Results: The fractional contribution from DNL to VLDL-triacylglycerol palmitate rose after alcohol consumption from 2 ± 1% to 30 ± 8 %; nevertheless, the absolute rate of DNL (0.8 g/6 h) represented < 5% of the ingested alcohol dose; 77 ± 13% of the alcohol cleared from plasma was converted directly to acetate entering plasma. Acetate flux increased 2.5-fold after alcohol consumption. Adipose release of nonesterified fatty acids into plasma decreased by 53% and whole-body lipid oxidation decreased by 73%.

Conclusions: We conclude that the consumption of 24 g alcohol activates the hepatic DNL pathway modestly, but acetate produced in the liver and released into plasma inhibits lipolysis, alters tissue fuel selection, and represents the major quantitative fate of ingested ethanol.

Ahhh Marc Hellerstein.  I've blogged on his work looking at the role of de novo lipogenesis from carbohydrate numerous times.  More on that at the end of this post.  

Summary of References 3 through 5, ethanol:

• Stimulates energy expenditure
• Inhibits fatty acid oxidation dramatically
• Inhibits lipolysis, reduces circulating NEFA
• Indirectly stimulates insulin
• Dramatically stimulates rates of DNL
• Is probably a factor in increasing body fat when total calories are in excess

Ian Lane's Commentary:  

The next[These] three studies all seemed to show that, when ethanol is metabolized, it causes a decrease in glucose and fatty acid oxidation. Why? Well, probably because the body is too busy ridding itself of this poison to bother breaking down carbohydrates and fat,[6-9] and so inhibits lipolysis temporarily to deal with the most important substrate, first.

This actually serves as interesting evidence on behalf of my hypothesis that the body preferentially oxidizes certain substrates over others, depending upon their inherent toxicity.

My feeling is that if fatty acids, proteins, glucose and ethanol are floating around the bloodstream, at the same time, the body will first oxidize the ethanol and excrete its metabolites (while storing carbohydrates, fats and proteins), since it is the most toxic to us out of the four. The next most important fuel to get rid of would be the glucose; evidenced by the fact that carbohydrate metabolism causes a direct decrease in lipolysis (fatty acid mobilization), due to the presence of insulin, and the immediacy of regulating blood glucose concentrations is imminent, so as to avoid tissue damage.[10]

ASIDE:  It seems that on some level Ian is listening to criticism so I'm going to insert a wee bit more here.  The above wording and referencing structure implies that the macronutrient heirarchy is discussed in references 6-9.  I fixed the first sentence to indicate that it was 3-5 that discussed this.  The other problem is remedied if he were to move the [6-9] to just after the word "poison", as each of those references deals with the toxicity of ethanol.  

Please see my last post, Is Glucose Toxic to Cells? Answers from the LIRKO Mouse and More ..., for why I'm not about to break out the Barry Manilow for Lane.    There is little in the way of evidence that there is any immediacy or imminence to clearing glucose due to inherent toxicity.   Furthermore, rather than clearing something toxic, the metabolism of amino acids (which too prevails over fats) actually creates a mildly toxic byproduct.  But let's move on ...

Is Alcohol Fattening?

Ian goes through what happens to ethanol in the body and presents a pretty good case for why alcohol, as a molecule, is unlikely to be a source of body fat.  First he writes:  

As an aside, according to this study: http://www.ncbi.nlm.nih.gov/pubmed/8184963 ~17% of the calories in ethanol are thrown off as heat due to the thermic effect of feeding, during digestion. Just under 20%? That’s a rather high number, in my opinion.

At this point I'm beginning to wonder if Ian is toying with his readers with his revisionism, as he had made quite the big deal over that Tweeter saying that the thermogenic factor was 15% rather than ~17%.  Last I checked 17 is closer to "just above 15%" than the 20% he remarks as being rather high. Feelings ... whoa, whoa, whoa, feelings .....

Moving on.  Here is how ethanol is metabolized (from Marks' Medical Biochemistry:

Fig. 25.1 from Marks' Basic Medical Biochemistry, 2nd Edition, p. 458
ADH = alcohol dehydrogenase , ALDH = acetaldehyde dehydrogenase ,
ACS = acetyl-CoA synthase

In most, non-to-moderate drinkers, ethanol is metabolized through this metabolic pathway.  It is first broken down to acetaldehyde, a molecule that is indeed toxic to human cells at higher levels.   This first step also generates one NADH "reducing equivalent" per ethanol molecule.  Sigh.  I know it must seem like I'm really picking on Ian now, but c'mon man, ya gotta get the terminology right if you want to even appear to know what you're talking about.  Ian wrote:

When ethanol is metabolized in the body, it is oxidized to form two byproducts: acetaldehyde and NADH.

This makes it sound as if ethanol is broken down into these two molecules.  NADH is technically a byproduct of ethanol metabolism, so I'll give him a pass on that one even though that term implies incidental production of something (e.g. since ammonia is a minor "waste" in amino acid metabolism, byproduct is appropriate).   NADH is anything but some incidental "byproduct" in biochemistry and is usually described as being produced or generated in the reaction as a short hand.  In actuality even that is misleading because we're really talking about altering the redox state of a molecule that is otherwise unchange and neither created nor consumed in the process).  Still, if NADH is a byproduct, acetaldehyde is a product of the first oxidation step (and I just caught and corrected myself for calling a main product of a reaction pathway a byproduct, so there ya go, and I appreciate if anyone points out my doing so elsewhere.)    The terminology needs tidying up!

The next step is for the acetaldehyde to be converted (oxidized) to acetate.  Another NADH is generated in this step.  Now here is where Ian plays fast and loose with thought experimental factoids.  After first reminding the reader that NADH is not converted to body fat (here is where byproduct or product imply that somehow ethanol has added this entire molecular species to the body, it has not), he writes:

Most of the acetaldehyde produced is excreted in the urine, as this chemical is purported to be responsible for many of the toxic effects associated with alcohol consumption, and so the body tends to “want” to get rid of it as fast as possible.[20-22]

I'm beginning to believe there's a need for a "referencing checker" app similar to grammar checkers.  This structure would seem to imply that references 20-22 support the claims made in this single sentence.  When, in fact, they do not.  Let's rewrite that sentence to be suitable for Wikipedia, shall we?

Most of the acetaldehyde produced is excreted in the urine[citation needed], as this chemical is purported to be responsible for many of the toxic effects[20-22] associated with alcohol consumption, and so the body tends to “want” to get rid of it as fast as possible.[citation needed]

I'll fix the link to reference 22, as I did go check for myself if that blue bolded statement was made there.  I did ask Ian for his source.  Not surprisingly he didn't get back to me.

I am not sure how many are familiar with the drug Disulfiram.  This drug is given to alcoholics to engender an adverse reaction when they consume alcohol, in the hopes of presenting a further deterrent to drinking.  What disulfiram inhibits ALDH resulting in a buildup of acetaldehyde in the blood.  This causes nausea and a general "sick" feeling.  So I ask you, if humans were able to merely excrete most of the acetaldehyde produced in the first step, how would disulfiram work?   I knew of this drug but not of its mechanism of action until several days ago when I went looking for more substantial quantification of Ian's claims re: acetaldehyde excretion.  I found Wikipedia to be helpful in pointing out that it is excreted in rats, but wanted further details and some idea of a rough percentage for "most of".  Well, according to Marks' Biochem (p. 460):

If it is not removed by metabolism, acetaldehyde exerts toxic actions in the liver and can enter the blood and exert toxic effects in other tissues.  Approximately 90% of the acetaldehyde that is generated is further metabolized to acetate in the liver.

OK ... I'm beginning to think a Lego character is in order here ;-)  Perhaps to complement the pool noodler we can use this one!  Perhaps Ian should crack open a few biochemistry texts before sharing his knowledge opinions and feelings further.  His writing style conveys a position of authority that is simply not backed up by what the scientific literature has to say ... about some pretty basic and uncontroversial topics, I might add.

Ian makes his first CICO-based argument for why alcohol isn't fattening with what followed his above, erroneous, claim:

So, although acetaldehyde itself also has a certain number of calories in its chemical bonds, most of it is excreted from the body before our cells can make any use out of it.

If this were indeed true, this would not do anything to turn CICO on its head.  Rather, the excretion of "energy containing" molecules, e.g. urea, has already been factored into Atwater factors.   If, say 75 of acetaldehyde were indeed excreted, and this actually varied from 60 to 90%, then again, the case is made through CICO -- it would just be unpredictable in establishing a "net caloric value" for each individual.  But, since this isn't even the case, we can scrap this argument regarding alcohol.  Rather, according to Marks':

Acetate, which has no toxic effects, may be activated to acetyl CoA in the liver (where it can enter either the TCA cycle or the pathway for fatty acid synthesis). However, most of the acetate that is generated enters the blood and is activated to acetyl CoA in skeletal muscles and other tissues (see Fig. 25.1). Acetate is generally considered nontoxic and is a normal constituent of the diet.

Thus it would appear that our bodies can access the energy in alcohol molecules quite handily, but ultimately limited by the capacity of the liver.  In Ian's scenario, there is scant little acetyl CoA to be made from alcohol.

Trace amounts of acetyl-CoA could, theoretically, be processed in a lipogenic fashion, and produce fatty acids that may end up stored in the adipocyte, temporarily, later on down the road — but, maybe not, depending upon the circumstances, and, of course, how much alcohol was consumed. (I suspect it would have to be quite a lot.) And remember, the acetyl-CoA left over from the acetaldehyde metabolite of ethanol is but a fraction of what it would have been, if most of it wasn’t wasted in one way or another. It’s also far less than there would be in the metabolism of carbohydrates or fats, comparatively.

Which has now been shown to be a total fairy tale.  If one reads on in Marks', on page 461:

Not only is most acetaldehyde converted to acetate, but most acetate is metabolized in heart and skeletal muscle through the TCA (aka Krebs) cycle -- to ultimately generate energy in the form of ATP.   Ethanol oxidized in this manner  generates a net 13 molecules of ATP (Marks' p. 463).

Fig. 25.3  
MEOS pathway in the endoplasmic reticulum

You may have noticed -- or not! -- that my page numbers skipped there.  That's because, there is an alternate pathway for ethanol metabolism called the microsomal ethanol oxidation system, or MEOS.

The affinity of the MEOS enzyme for ethanol is much lower than that of ADH, so that this pathway is only significantly activated in heavy drinkers.  According to Marks' only 10-20% of alcohol is metabolized by MEOS in moderate drinkers (p.460).  While NADP+ is a cofactor for lipogenic enzymes, the lack of NADH production in this pathway reduces the net ATP production per alcohol molecule from 13 down to only 8.  I'd love to delve into this more thoroughly, but just don't have the time.  The MEOS pathway is activated in a metabolic milieu of up and downregulation of various enzymes leading to fatty liver and such so that it is difficult to point fingers at one part.  I will, however, plop this diagram (source) here for discussion!

I mean if we're going to get all down and dirty about every enzyme and pathway, we might as well cover it all {grin}.  I'll leave you with one last pearl of "wisdom" from Ian:

The idea that the little acetyl-CoA derived from ethanol will go directly to producing new free fatty acids and then subsequently store them all in the adipocytes is silly and overly simplistic, in my opinion.

Aww, he lets me down here.  Not because he didn't fact check this whole acetaldehyde-pee calorie drain, but because he changed his wording so I can't ding him once more.  I guess someone conveyed to him that his original wording of "borderline retarded" might be offensive and he has replaced that with "silly and overly simplistic".   Oh wait ... I just dinged him anyway!

click to
enlarge
in browser

Hang with me, I'm getting to the good part where I have praise in store for Mr. Lane!  There was a reason for my linking up and excerpting from Lane's References 3-5 above, number 5 in particular.  In this reference, radiolabels were used to track the source of fatty acids secreted by the liver.  Hellerstein's research group has used this method extensively to elucidate the contribution of de novo lipogenesis from carbs to triglycerides (see: Where do Triglycerides Come From - series).  Or, the graphic on the right, direct link for easier viewing, authored by Hellerstein, provides a nice lay-friendly summary of the whole "are carbs fattening" debate.

The bottom line here is that for carbohydrates, despite seemingly dramatic increases in DNL activity under overfeeding circumstances, the absolute amount of fatty acids produced by this metabolic route only accounts for a few grams per day.  Furthermore, these fatty acids are usually subsequently oxidized so that there is little chance they can contribute significantly to body fat.

This is the exact same argument -- correct at last! -- that Ian makes regarding alcohol.  Even though far more acetaldehyde is converted to acetate and eventually acetyl CoA than Ian would like you to believe, less than 5% of the alcohol is converted to fatty acids in the liver by DNL.  In the Hellerstein study, out of 24g of alcohol (1 shot = 14 grams), only 0.8 grams of fat was produced in 6 hours.  It is unlikely that this would find its way into adipocytes and get trapped there!   Alcohol doesn't turn to body fat.

Summary of Part I

Lane on Alcohol:

• Incorrect:  large quantities of ingested ethanol calories are lost when acetaldehyde is excreted
• Correct:  only small quantities of ingested ethanol molecules can be the carbon source for fatty acids

Conclusion:  Ethanol calories are not fattening.  Ethanol might be fattening by causing you to eat more foods that biochemically interact to form and trap body fat because they are fattening, but not ethanol.

Hellerstein on Carbs:

• Only small quantities of ingested carbohydrate molecules are converted to fatty acids 

Conclusion:  Dietary carbohydrate is an unlikely source of body fat, but when carbohydrate calories contribute to a caloric surplus, the excess calories, mostly from fat, will be stored.

---

In Part II I'm going to discuss some of the other parallels between Ian's case for why alcohol isn't fattening (per se of course) that further exonerate carbohydrates. 

In the mean time, I made reference to the pool noodler who is, of course, Adam "Gatewing" Kosloff.  See if you can guess who the third character in my low carb "Brat Pack" is?  

---

Are Carbs Fattening graphic taken from Discovering Nutrition Insel, Turner & Ross, copyright 2010  (Amazon search to all versions, older editions that are more economical are highly recommended!) .  I have the 2001 edition (original?) of Nutrition with evil deadly wheat on the cover.  This book really presents biochemistry at an understandable level, without sacrificing accuracy for simplicity.