More Evidence William Davis Was Right: Wheat Does Increase Obesity According to New Study

Last Updated on June 11, 2019 by Afifah Hamilton
Read Time: 6 min

SCOOP: new trial demonstrates that gluten increases weight gain independent of calories via suppression of thermogenesis!

Mouse in the wheat -
A new study found that when gluten was added to the diets of mice they had increased weight gain, even though they consumed the same calories as the mice without gluten in the diet.

Cardiologist William Davis has been accused of stirring up unnecessary fears over gluten when, in his New York Times bestseller Wheat Belly, he claimed that wheat causes obesity, heart disease and a host of other metabolic and digestive problems.

The Washington post, for example, recently challenged Davis ideas in an article entitled The truth about gluten-free, paleo and other diet books. They quoted Lawrence Cheskin, director of the Johns Hopkins Weight Management Center and associate professor of health, behavior and society at the Johns Hopkins Bloomberg School of Public Health, who said “The problem with [Davis’] premise is that there’s little evidence to support it.”

Well Cheskin and other detractors may have to eat their hats as a study about to be published in the prestigious International Journal of Obesity – the leading obesity journal in the Nutrition & Dietetics category – provides direct evidence that gluten does indeed induce weight gain independent of its calorie content.

Cardiologist William Davis claims that wheat causes obesity and inflammation.

The study by Brazilian researchers at Federal University of the Minas Gerais, Department of Biochemistry and Immunology, builds on earlier work in which the group showed that mice fed a gluten-free diet had less weight gain than mice eating the same caloric quantity of gluten-containing food.

It their latest investigation they went  further, investigating the metabolic and lipid effects of gluten-containing diets to uncover how gluten exerts its obesogenic effects

In their study they used four groups of mice. Two groups were fed either standard rodent chow, or a high-fat (obesogenic) diet. Groups three and four were given the same two diets, but with 4.5% of the calories replaced with pure gluten.

In both cases the gluten-containing diets led to greater weight gain than their gluten-free counterparts, and this was not due to differences in the quantity of food they consumed as all four diets contained the same number of calories. The only difference was the presence or absence of wheat gluten.

In recent years the dogma that “a calorie is a calorie” has started to be challenged, with critics arguing that some food calories contribute to obesity more than others. Whereas, adherents, including most national health authorities and heart disease and obesity charities, have stuck to the claim that weight gain is simply a consequence of consuming too many calories and their source is immaterial. This study, however, clearly demonstrates that calories from gluten-containing diets do indeed produce more weight gain than those from equal energy gluten-free diets.

To understand why the mice were responding to gluten like this the researchers examined their subcutaneous and brown fat deposits at the end of the experiment. Whilst subcutaneous fat’s primary role is for energy storage, brown fat’s main function is heat production (thermogenesis). Brown fat is associated with lower body weight because it has the ability to turn calories directly into heat. White, subcutaneous fat can do this too, but at a much lower rate than brown fat.

What the researchers found was that thermogenesis was increased in the gluten-free groups, but down-regulated in the gluten fed groups. Put simply: gluten caused more calories to be stored as fat and fewer to be burned off as heat. Gluten changed the metabolic fate of the food calories.

Whilst both the standard and high-fat diets produced more weight gain when gluten was included, other effects of the gluten varied between the two diets.

In the high fat diet, gluten led to reduced adiponectin, peroxisome proliferator-activated receptor (PPAR) α and PPARγ and hormone-sensitive lipase – indicative of a metabolism geared for higher levels of fat storage rather than fat burning. To illustrate the role of these markers of fat metabolism, consider the fat hormone adiponectin, lower levels of which are associated with insulin resistance, dyslipidemia, and atherosclerosis in humans [ref].

By contrast, in the standard (relatively high-carb) diet, the addition of gluten increased levels of the cytokines interleukin 6 (IL-6) and tumor necrosis factor alpha, (TNFα) – both involved in inflammation. Interestingly, TNFα is released during invasion by a pathogen, and there are many similarities between gluten and pathogens (see Bethune and Khosla’s paper)

Finally, the researchers fed the mice radio-labelled gluten, to trace where it was ending up in the body. They found deposits in the blood, liver and visceral adipose tissue – sites far from the intestines where gluten is usually thought to act.

It is worth noting that these results were identified in mice consuming just 4.5% of calories from gluten – about the same as found in a normal western diet where gluten consumption is typically 10-40 g per day.

[For the maths nerds: being a protein, gluten provides roughly 4 calories (4 kCal) per gram. 4.5% of a typical 2000 calorie human diet is about 90 kCal, which equates to 22.5 g gluten per day – just below the middle of the normal gluten consumption range].

So when Davis says that wheat gluten causes obesity and inflammation, don’t let anyone tell you there is no evidence.

Reference: Freire RH et al. Wheat gluten intake increases weight gain and adiposity associated with reduced thermogenesis and energy expenditure in an animal model of obesity. International Journal of Obesity (7 October 2015); PubMed,

* * * * *

Having co-opted Steinbeck’s classic for my introductory image “Of (fat) Mice and Men” I’ll finish with some bastardised quotes from his work, which are surprisingly fitting…


“Trouble with mice is you always kill ’em. (Then examine their adipose tissue)”
― John Steinbeck, Of Fat Mice and Men

“We could live offa the (gluten-free) fatta the lan’.”
― John Steinbeck, Of Fat Mice and Men

“(Gluten) don’t need no sense to be a nice fella. Seems to me sometimes it jus’ works the other way around. Take a real smart guy (like Gluten) and he ain’t hardly ever a nice fella.”
― John Steinbeck, Of Fat Mice and Men

“His ear heard more than what was said to him, and his slow speech had overtones not of thought, but of understanding beyond thought. ‘Jeeze.. ya mean it’ was tha gluten all ‘long?…”
― John Steinbeck, Of Fat Mice and Men

* * * * *

EDIT 14/10/2015

Based on some of the comments (see below) I felt it would be helpful to provide some more details from the paper. In the comments, I stated that the mice were over-fed, i.e. forced to be obese. In fact mice were free to eat as much as they wanted. Those on the standard chow diet (CD) gained normal weight, as illustrated by the lower two lines of this graph from the paper.

NOTE: the mice were 8 weeks old at the start of the study, i.e. juveniles, and were still growing. Healthy pet mice might be expected to reach 26g weight by 16 weeks old, pretty much  in line with those on the standard chow (CD) diet above.  – R H Freire et al. International journal of obesity. Preview, 7 October 2015.

A ‘high fat’ rodent diet is actually high fat and sugar – a particularly yummy mix that mice cannot stop eating, hence it is classed as ‘obesogenic’. Fed this diet, the mice’s body weight skyrocketed, as illustrated by the top two lines (HFD).

The interesting point is the clear increase in weight gain (dark lines CD-G and HFD-G) when gluten was included. This occurred in both diets. The implication is, therefore, that the presence of gluten increased weight gain whether the mice were over-fed or not.

This graph shows the energy intake of each group of mice:

Mice in the high fat diet groups (HFD) consumed 50% more calories per day than those in the standard chow groups (CD) – R H Freire et al. International journal of obesity. Preview, 7 October 2015.

The presence of gluten in the diets (CD-g and HFD-G) didn’t make the mice eat more, so the additional weight gain was not due to additional calories consumed, but rather to down regulation of the basal metabolic rate. The authors conclude:

Despite having had the same energy intake, weight gain was approximately 20% higher in mice fed gluten-enriched diets compared to the respective gluten-free groups. Adipose tissue depots were increased by approximately 30% in CD-G and HFD-G groups. No other controlled studies using animal models or clinical trials have evaluated the obesogenic effects of gluten intake.

Our study supports the hypothesis that gluten intake (4.5% of diet) can contribute to an acceleration of body weight gain that results from a reduction in energetic expenditure. The present results open a field of study that will help us to understand the association between gluten or gluten peptides and fat metabolism.
We conclude that gluten or its peptides could reach extra-intestinal organs, such as liver and [visceral adipose tissue], supporting a direct effect on those sites. Moreover, the inclusion of wheat gluten in isocaloric diets decreases thermogenesis, browning and energy expenditure, and accelerates weight gain and adiposity in CD-G and, mainly, in HFD-G mice.

12 thoughts on “More Evidence William Davis Was Right: Wheat Does Increase Obesity According to New Study”

  1. Interesting study. Although, this study cannot explain the French, who have one of the highest levels of wheat consumption in the world (40% greater than the US according to FAOSTAT) and have one of the lowest obesity rates of any developed nation.

    • I might also point out that the “hydrolyzed wheat gluten” they used in the study is not like the gluten would obtain from real bread.

      Hydrolyzed wheat gluten is basically MSG, and the effects of feeding MSG to mice have the same exact effects that were shown in this study.

      • The paper only refers to hydrolysed gluten in relation to the radio labelled step. Natural digestion of course leads to a degree of hydrolysis, but you make a fair point – how representative are the gluten peptides used in such studies compared to those a normal human is exposed to when eating bread?

        I’ll put your point to the study authors to see if I can get a comment from them. I’ll post it below if they get back to me!

      • I’d also point out that some cultures—including 6th century Chinese—intentionally added wheat gluten to food.

        Wheat gluten has been documented in China since the 6th century. It was widely consumed by the Chinese as a substitute for meat, especially among adherents of Buddhism. The oldest reference to wheat gluten appears in the Qimin Yaoshu, a Chinese agricultural encyclopedia written by Jia Sixie in 535. The encyclopedia mentions noodles prepared from wheat gluten called bo duo. Wheat gluten was known as mian jin by the Song dynasty (960–1279). Wheat gluten arrived in the West by the 18th century. De Frumento, an Italian treatise on wheat from 1745, describes the process of washing wheat flour dough in order to extract the gluten. John Imison wrote an English-language definition of wheat gluten in his Elements of Science and Art published in 1803. By the 1830s, Western doctors were recommending wheat gluten in diets for diabetics. In the United States, the Seventh-day Adventists promoted the consumption of wheat gluten beginning in the late 19th century. Sanitarium Foods, a company affiliated with John Harvey Kellogg’s Battle Creek Sanitarium, advertised wheat gluten in 1882.

        I think if one is going to demonize gluten, one needs to be able to explain exactly how gluten was considered to be a health food up, right until the Industrial Revolution. And one needs to be able to explain why (non-fortified) European countries don’t have the non-celiac gluten sensitivities and obesity that we have in the US, despite their often higher intakes of wheat.

    • Thanks for the comment Ian,

      Of course this was an overfeeding study – so is probably indicative of how gluten affects the body’s response to hyper-caloric intake. In other words, in the already-obese it might well help to cut out gluten as per Davis.

      The French are famed for their love of real food, cooking from scratch and eating a very diverse range of foods from crèche upwards. Pamela Druckerman’s book ‘French children don’t throw food’ describes their marvellous and meticulous gustatory education system, which, I suspect, is a significant contributor to protecting the French From the obesity epidemic.

      • “Of course this was an overfeeding study”

        Nice honesty.

        Someone emailed me this morning about this post and I put it out to some folks to chew on, but that was my knee-jerk about the thing. I appreciate futzing with variables to tease out clues, just so long as we don’t go crazy.

        In other words, overfeeding that’s long and regular enough to show weight gain is a confounder built and baked into the cake.

        Thing is, there’s just too many falsifications (such as Ian recounts even in modern times) to really buy the proposition that eating gluten causes weight gain per se (perhaps in some people, unless it’s the pastries).

        Second, I’ve been on a reasonably high gluten diet for some months now, but exclusively from artsy whole grains, non-fortified. I have found them, converse to all I had known and experienced growing up with white enriched bread, fortified cereals, very satiating. It’s rather remarkable.

        • Thanks for your thoughts Richard, they stimulated me to look more closely at some of the details in the full paper. Consequently, I have to correct myself – only the ‘high fat’ diet group were over fed – they were given free access to an obesogenic diet (actually high fat and sugar) – pretty well those pastries you referenced! The standard chow group, however ate normal healthy quantities of food, and put on normal weight. What is significant is that in both diets the ‘lil micies put on more weight with gluten supplementation.

          I’ve checked on the kind of gluten they were given, and it seems it was from a bakery supplier, not some hydrolyzed wheat protein that Ian suggested (they only used that in the final radio tracer stage). I’ve emailed the authors for their comment.

          To clarify, I have added an additional section at the end of the post showing some of the graphs from the full paper. Take a look and let me know your thoughts.

    • Yes, and Sweden and Finland both fortified their flour from the 1970s up until the mid-1990s. Incidentally, Sweden’s famous “celiac epidemic” ended in 1995—immediately after fortification ended. Could be a coincidence. Or not.

  2. @Ian
    Re: 6th century China, adding wheat gluten to foods. “I think if one is going to demonize gluten, one needs to be able to explain exactly how gluten was considered to be a health food up, right until the Industrial Revolution.”

    Celiac disease has almost certainly been present since the dawn of agriculture, yet the link to gluten was only identified in the mid 20th century*. So the fact it was eaten a ‘long time ago’ does not make it a health food as they were not aware of its disadvantages. Furthermore, wheat (gluten) was a cheap source of protein, and for most of human history protein availability would have been a population limiting factor, so the advantages of wheat consumption would outweigh the disadvantages. In the modern world there are no longer any advantages to wheat over the alternative sources of nutrition.

    *Reference: “The first description of a patient affected by [Celiac Disease] is ascribed to Areteus of Cappadocia, who in the 2nd century AD reported a case of chronic diarrhea and malabsorption. However, only at the end of the 19th century, Samuel Gee scientifically described the celiac syndrome in childhood and in the 1950s, Willem Dicke identified gluten as the environmental agent of CD” Sreebny LM. WJG, (2015)


Leave a Reply