Gluten is a major dietary component of wheat, consisting of large peptides such as gliadins and glutenins. It is difficult for human intestinal proteases to break down these large peptides, so they escape digestion and induce changes within the gut microbiota. Gluten has been associated with some diseases, including non-coeliac gluten sensitivity, celiac disease, and gluten ataxia. Individuals who have adopted a gluten-free lifestyle have reported digestive comfort, improved weight management and overall well-being. However, evidence for these health benefits in healthy individuals remains lacking, and gluten avoidance can also have nutritional and metabolic risks for those without gluten-related disorders.
One study showed that after one year of adopting a low-gluten diet (LGD) or gluten-free diet (GFD), celiac disease patients were at a greater risk of developing metabolic syndrome. The higher glycaemic index of many gluten-free foods could drive this. Such risks warrant long-term follow-up, as diet-driven shifts in gut microbiota can contribute to adverse metabolic outcomes. A study published in Nutrients assessed whether and how a low-gluten diet, over prolonged periods, impacts gut microbiota function and composition in healthy adults. This randomised controlled trial assessed the effects of sustained exposure to LGD on the composition and metabolic activity of the gut microbiota in a sample of 40 healthy adults in France.
The study sample comprised men and women who consumed an average of 160 g of bread and pasta daily, corresponding to about 14 to 15 g of gluten from these foods. The volunteers switched to an LGD from their usual high-gluten diet (HGD) for two successive periods of 8 weeks. The study design entailed the collection of 40 fecal samples at baseline (M0) and after 8 weeks of LGD (M2). Furthermore, 20 fecal samples were collected after 16 weeks of LGD (M4). During the LGD periods, the richness in α-diversity of the gut microbial community significantly declined, and this decline was greater after 16 weeks than after 8 weeks, showing a time-dependent worsening effect.
Regarding differences in β-diversity at M0, M2, and M4, an apparent clustering of microbial populations was noted in the LGD periods. Microbial communities at M2 and M4 formed similar clusters, significantly different from M0 regarding β-diversity. At the phylum level, the relative abundance of Verrucomicrobiota and Actinomycetota was significantly reduced at M4, relative to M0. Concurrently, Bacteroidota and Bacillota levels increased considerably. However, the ratio of Bacillota to Bacteroidota remained unchanged, which the authors note as an important nuance. At the family level, Veillonellaceae significantly increased in the same period, while Akkermansiaceae reduced significantly.
Despite high variability, Bifidobacterium significantly decreased, causing some comparisons to fall short of statistical significance in sequencing data. Molecular quantification analyses did not reveal any changes in the bacterial species E. coli and Faecalibacterium prausnitzii and the Lactobacillus-Pediococcus group. Following LGD, the Bacteroidia, Verrucomicrobiae, and Clostridia classes were observed in differential abundance at the species level. Akkermansia muciniphila was significantly decreased at M4. Lachnobacterium bovis, a lactate-producing species, also declined. At the same time, some butyrate producers (Roseburia and Faecalibacterium) increased, which the authors note helped to maintain stable butyrate levels despite community shifts.
Fibre-degrading species, including Ruminococcus callidus and R. champanellensis, were also significantly affected at M4. Eubacterium spp. and Blautia caecimuris (Lachnospiraceae family), were reduced at M4. The Lachnospiraceae family consists of many butyrate-producing species. Following LGD, the Enterobacteriaceae population increased 10-fold, while the level of total anaerobes in the population remained similar to M0. On assessing the gluten-degrading community level, a 10-fold decline was noted at M2. Enterobacteriaceae, which can include ethanol-producing species like E. coli, may contribute to inflammatory processes if overgrown.
There was no significant difference in the concentration of faecal fermentative metabolites between M2 and M4. At M2, there was a slight reduction in the proportion of acetate, in favour of propionate. The proportion of ethanol increased by more than three times at M2 and M4. Ethanol accumulation is a key metabolic red flag, as excess ethanol production is associated with metabolic syndrome and gut inflammation. This contrasts with the significant decline in isobutyrate at M4, but not M2. Despite microbial shifts, total acetate, propionate, and butyrate levels stayed largely stable, which the authors attribute to redundant butyrate-producing capacity among different bacterial taxa.
Most of the main gluten-degrading strains belonged to the class of Clostridia. Additionally, one isolate belonged to Actinomycetota, three to Erysipelotricha and two to Gammaproteobacteria. Five strains belonged to the Lachnospiraceae family among the class of Clostridia. An isolate from the family Oscillospiraceae was identified as Flavonifractor plautii. In three subjects, strains belonging to the Erysipelotrichaceae family were noted. Therefore, a 16-week LGD altered gut composition and metabolic activity in a sample of healthy French adults, leading to a dysbiotic shift. How a low- or gluten-free diet might affect the abundance of A. muciniphila remains however to be investigated.
One hypothesis is that the reduction in A. muciniphila abundance is due to changes in dietary fibre content in the low-gluten diet, particularly the fermentable oligo-, di, monosaccharides, and polyols (FODMAPs) commonly found in wheat. Indeed, oligofructose (i.e.inulin fibers) administration has been reported to restore the A. muciniphila population in diet-induced obese mice, suggesting that reducing the amount of FODMAPs may induce a reduction in the A. muciniphila population. Another hypothesis is related to the reduced polyphenol content that characterises gluten-free products compared to regular gluten-containing foods. Polyphenols are important micronutrients that play a key role in human health.
They strengthen the intestinal mucosal barrier through increased mucin production and modulation of gut microbiota. The reduced levels of polyphenols in low- or gluten-free diets may therefore weaken the mucus barrier. Many gluten-free foods, finally, are ultra-processed industrial formulations containing additives such as emulsifiers, stabiliser and sweeteners to improve sensory scores and extend shelf life. Emulsifiers, which are detergent-like molecules, are ubiquitous components of processed foods that may disrupt mucus-bacterial interactions and have the potential to promote diseases associated with gut inflammation.
Chassaing et al. demonstrated that the administration of two emulsifiers to a mouse model caused alterations in the composition of the gut microbiota, including a reduction in microbial diversity and increased levels of several mucin-degrading species, such as Ruminococcus gnavus and Akkermansia muciniphila, as well as an enrichment in mucosa-associated, inflammation-promoting Proteobacteria. Therefore, it could be hypothesised that some of the alterations to the gut microbiota observed in healthy subjects on a low-gluten diet may be due to an increased intake of food additives in the gluten-free products supplied.
Longer-term studies could further explore its impact on immunity, physiology, and metabolism. Still, the findings suggest that sustained LGD in healthy individuals can progressively impair gut microbiota balance and raise ethanol levels, potentially posing metabolic risks.
- Edited by Dr. Gianfrancesco Cormaci, PhD, specialist in Clinical Biochemistry.
Scientific references
Delmas E et al. Nutrients. 2025; 17(15):2389.
Bueno C, Thys R, Tischer B. Foods 2023; 12:4415.
Caio G. Lungaro L et al. Nutrients 2020; 12:1832.
Chassaing B, Koren O et al. Nature 2015; 519:92.
