In my post about my own experience with going gluten-free, I describe how removing wheat from my diet felt like breaking an addiction. Less than a year after my experience, Dr. William Davis published his book Wheat Belly. One of the arguments he uses for why cutting out wheat is a good idea is that it is addictive in the same way as opioid drugs like heroin. According to him, wheat breaks down into heroin-like molecules during digestion, which are able to enter your brain and cause cravings for even more wheat, leading to overeating and weight gain. Are these things really true? Sounds pretty wild, right? Well, let's examine his argument in detail, and check out his references.
Dr. Davis begins his argument in Chapter 4 of Wheat Belly by referencing a paper that showed that opioid peptides (more info here) can be derived from digested food proteins and that those opioids are capable of binding to opioid receptors in the brain. The 1979 paper he cites is by Zioudrou et al. (full text) and showed that opioid peptides can be produced during digestion from both wheat gluten and α-casein (one of the types of protein found in cow's milk). These food-derived opioid peptides are called exorphins, since they come from an exogenous source (outside of the body, in contrast to endogenous, produced from within the body) and have morphine-like (opioid) activity.
The authors simulate digestion in the laboratory by treating wheat gluten and α-casein with the digestive enzyme pepsin. Then, they do a lot of fancy labwork to purify and isolate the opioid peptides from the resulting mixture. Here's how Dr. Davis describes what happens next in the study (emphasis mine):
"The dominant polypeptides were then isolated and administered to laboratory rats. These polypeptides were discovered to have the peculiar ability to penetrate the blood-brain barrier that separates the bloodstream from the brain."
Hmm. If he read the methods section, he'd see that the authors actually used homogenized rat brains to determine if these peptides could bind to brain opioid receptors. No live lab rats were used in the study. Also, it's impossible to study the blood-brain barrier in a petri dish. You have to have a live animal. In the discussion section of the paper the authors even state (emphasis mine):
"The results presented here have shown that peptides with morphine-like activities, which we call exorphins, can be isolated from some food proteins after treatment with the stomach proteinase, pepsin. Since the opiate assays which we have used so far are performed with cell homogenates or isolated organ preparations, it seems pertinent to ask whether these findings have physiological significance. ... [A]n unequivocal answer to this question must await the results of whole animal experiments."
Dr. Davis's isn't getting off to a good start here. His statements about the results of this study are unapologetically false. The only conclusions that can be drawn from this paper are: 1.) that it is plausible that gluten exorphins are formed in the gut as a result of digestion, and 2.) that if those gluten exorphins are able to exit the gut, enter the bloodstream, and get past the tricky blood-brain barrier, that they are able to bind to opioid receptors somewhere in the brain. I know I said in my last post I wasn't aiming to try to discredit anyone, but making such incredibly misleading arguments and false statements while pretending to be providing valid scientific information is totally unacceptable, and risks making going wheat-free look like a silly made up fad (which it isn't). But this is only the tip of the iceberg. Let's look at where he goes next.
He discusses 2 studies that examined the effects of the drug naloxone on eating behaviors in humans. Before I get into the details of papers, I want to give some background info about opioid receptors, naloxone, and competitive antagonists, so that we are all on the same page.
Receptors, Agonists, Antagonists
Opioid drugs like heroin are able to exert the effects that they do on humans because they are able to bind to and activate the opioid receptors in our bodies and brains (which means crossing the blood-brain barrier). This makes them opioid receptor agonists. In contrast, the drug naloxone is an opioid receptor antagonist, meaning it is able to bind to our opioid receptors but does NOT activate them. It just sits there and takes up space, doing nothing. It is also able to cross the blood-brain barrier. Naloxone is not just any antagonist, it is a competitive antagonoist, which means if you've taken an opioid agonist like heroin, naloxone comes along and kicks the heroin off of your opioid receptors in your body and brain, thereby reversing the effects of the drug. That's why they give naloxone to people who have overdosed on opioids as an emergency measure to save their lives.
So how does Dr. Davis relate naloxone to wheat?
"In lab animals, administration of naloxone blocks the binding of wheat exorphins to the morphine receptor of brain cells. Yes, opiate-blocking naloxone prevents the binding of wheat-derived exorphins to the brain. The very same drug that turns off the heroin in a drug-abusing addict also blocks the effects of wheat exorphins."
He does not provide a citation for this statement, so I'm not sure which article he got this idea from, but I have come across a few studies where researchers have taken isolated gluten exorphins and injected them directly into the brains (or ventricles) of live lab animals, observed the effect, and then used naloxone also directly injected into the brain/ventricles to reverse the observed effect (thereby confirming that the effect was due to the opioid activity of the exorphins. An example is this study, looking at prolactin secretion from the pituitary).
He's just making things up at this point, but let's see where he goes next:
"So this is your brain on wheat: Digestion yields morphinelike compounds that bind to the brain’s opiate receptors. It induces a form of reward, a mild euphoria. When the effect is blocked or no exorphin-yielding foods are consumed, some people experience a distinctly unpleasant withdrawal."
He then asks his readers: "what happens if normal humans are given opiate-blocking drugs?" And then he cites two studies:
- Cohen MR, et al. Naloxone reduces food intake in humans. Psychosom Med. 1985;47(2):132-8 (full text available free here)
- Drewnowski A, et al. Naloxone, an opiate blocker, reduces the consumption of sweet high-fat foods in obese and lean female binge eaters. Am J Clin Nutr. 1995;61(6):1206-12. (full text available free here).
Regarding the first study, he says "...wheat-consuming participants given naloxone consumed 33 percent fewer calories at lunch and 23 percent fewer calories at dinner (a total of approximately 400 calories less over the two meals) than participants given a placebo."
Okay, that's not so bad. He's truthful in reporting the results, but it really is a non sequitur, since the authors were evaluating the endogenous opioid system's effects on eating behavior. This paper has absolutely nothing to do with wheat or gluten exorphins. His description of the participants as "wheat-consuming" is a red herring.
The study is pretty interesting, so here's a quick summary: the authors examined the effect of giving either naloxone or placebo in the morning after an overnight fast, and then measured food intake at lunch (2.75 hours after administration) and dinner (6.75 hours after administration). It's a double-blind crossover design, meaning each subject acted as his or her own control (one day receiving placebo, one day receiving naloxone) and that both the subjects and the researchers were blind to treatment assignment, to avoid potential bias in data collection or analysis and to prevent altered behavior from the experimental subjects. The authors report the following in the results section:
- naloxone significantly reduces total food intake compared to placebo; it reduces it by 28%
- naloxone significantly reduced total food intake at both lunch (33% reduction) and dinner (23% reduction) compared to placebo
- naloxone significantly reduces total fat and total protein intake compared to placebo
- naloxone does not significantly reduce total carbohydrate intake compared to placebo. This is important, since wheat is a carbohydrate! Funny how Dr. Davis conveniently leaves this part out. Probably because it disagrees with his argument that eating wheat makes you want to eat more wheat (and the converse argument, that taking naloxone prevents you from craving wheat, thereby reducing your consumption).
What about the second paper? He says that "binge eaters were confined to a room filled with food for one hour. ... Participants consumed 28 percent less wheat crackers, bread sticks, and pretzels with administration of naloxone."
This statistic is completely, totally, 100% MADE UP! Gah!
Just like the first study, this one is pretty interesting, so here's a brief overview: subjects were female binge eaters or non-bingers, who were further categorized by body type: lean or obese. The researchers examined the effects of naloxone as well as an opioid agonist, butorphanol, on eating behavior compared to saline (placebo). Just like the other study, each subject served as her own control, meaning each received naloxone, butorphanol, and saline on different days and it was double-blind, so neither the subjects nor the researchers knew treatment assignment. Subjects were allowed to freely consume as much food as they desired for a 1 hour period that began 1.5 hours after drug/placebo infusion.
Foods were broken down into 4 categories: low sugar/low fat; low sugar/high fat; high sugar, low fat; and high sugar/high fat. Table 2 below shows the specific foods that were in each category.
Table 4 below shows the results. Naloxone administration lowers consumption of high sugar/high fat foods in binge eaters compared to saline. There was no difference in any other category.
You know what else is really interesting? That butorphanol, an exogenous opioid agonist (just like gluten exorphins!) did NOT increase food intake. Just like the first paper, the authors here state that they are interested in the endogenous opioid system's effects on food consumption. This paper has nothing to do with wheat or gluten exorphins!
The authors also included Figure 1 below, to show the average intake of each specific food type after naloxone infusion, expressed as a percentage of how much of the same food was consumed after placebo. So, if you ate 10 crackers after placebo and ate 10 crackers after naloxone, that would be shown as 100% in this figure. If you had 15 after naloxone, that'd be 150%. Make sense?
Remember how Dr. Davis summed up these results? 28% less crackers, breadsticks, and pretzels? Um, actually, breadstick consumption increased by 41% with naloxone. Yes, cracker consumption decreased by 17% and pretzels by 40%, bu the non-wheat items also decreased (corn chips, potato chips, cream cheese, jelly beans, marshmallows). The authors describe Figure 1 the following way: "naloxone reduced the consumption of foods that were rich in fat, sugar, or both."
There's absolutely nothing wheat-specific about these results!
Dr. Davis concludes his argument by saying (emphasis mine):
"When administered to normal people or people with uncontrollable appetite, opiate-blocking drugs yield reductions in appetite, cravings, and calorie intake, as well as dampen mood, and the effect seems particularly specific to wheat-containing products."
At this point, I just want to do a face-palm and shake my head. He has failed to provide any evidence to support these claims.
So What Do Gluten Exorphins Do, Then?
Dr. Davis has done a fine job of providing zero information about what is actually known about gluten exorphins. How about we examine what we do know about them.
There are five known gluten exorphins. The first four, known as A5, A4, B5, and B4, were identified in 1992; a year later, the same group identified gluten exorphin C (papers here and here). Unfortunately, very little research has been conducted to characterize the biological/physiological activity of the gluten exorphins, and so not much is actually known about them.
- In mice, gluten exorphin A5 (GE-A5) seems to facilitate part of the learning/memory process, reduce anxiety, and, when given directly in the ventricles of the brain (instead of orally), also produced analgesia (source).
- Both GE-A5 and GE-B5 were shown in rats to increase postprandial insulin secretion in rats after oral administration, but the dose required for GE-B5 to produce a significant effect was 10 times higher than the GE-A5 dose (source).
- GE-B5 increases levels of the pituitary hormone prolactin, and it has been confirmed that this is due to binding to opioid receptors located outside of the blood-brain brain (source). GE-B4 had no effect on prolactin levels.
- GE-C actually produces hyperalgesia (increased pain experience) and also increases anxiety in rats (source).
- I couldn't find any articles that had studied the effects of GE-A4. If I missed something, please let me know.
Other Food-Derived Exorphins
Something that I think is really fascinating, and that Dr. Davis decides to leave out of his argument, is that wheat is NOT the only food whose proteins break down into exorphins during digestion. The way Dr. Davis writes about wheat, you'd think that it is unique in this respect. Well, it isn't. Milk, rice, corn, blood (from meat), and spinach are all part of the same club (full text source). Researchers have tried to find exorphins in oats, rye, soy, and eggs, too, but been unable to find any. It may turn out that there are more foods whose proteins break down into exorphins during digestion than there are foods whose proteins don't. This research is still pretty young, and I'm excited to see what discoveries come to light in the next few years.
Conclusion: The Evidence Does Not Support the Claim
As you can see, we don't really know much at all about gluten exorphins. Based on the current body of scientific knowledge, we don't have any evidence to support Dr. Davis's claim that wheat is addictive due to the formation of gluten exorphins during digestion. Does that mean wheat isn't addictive for some other reason yet to be discovered? Well, we don't know. Like I described, my own wheat-free experience felt like breaking an addiction. And I am not alone in describing it this way; just take a look at any paleo/primal/low-carb/gluten-free forum and read people's stories.
Do I still think eliminating wheat is a good idea? YES! For many reasons (that I plan to discuss on my site), but for now you can read this, this, this, and this and also the info about non-celiac gluten sensitivity at the bottom of my post here.
Am I glad that Dr. Davis published Wheat Belly? YES! I just wish he hadn't twisted the science around and literally made things up to support his arguments.
The success stories on the Wheat Belly Blog site are compelling, and should not be dismissed because there's some faulty arguments and references in the book. I realize that the book was written to appeal to a wide audience, and it's amazing and wonderful that it has reached so many people, and has helped so many see improvements in health. Anything that makes people stop and think about what they are putting in their bodies, and strive to make changes in order to improve health is a good thing.
But, I also think it is unforgivable to provide bogus citations to support your arguments and falsely represent the conclusions of scientific studies. Thanks to Wheat Belly, the myth that wheat is addictive like heroin has spread far and wide, and been re-written on tons of other websites that are just trying to spread the good word.
What do you think? Have you read Wheat Belly? Have you already benefited from going wheat/gluten-free? Let me know in the comments! And, as always, thanks for reading!