For decades, non-nutritive sweeteners have been sold to the public as a safe, calorie-free escape from the metabolic risks of sugar. Millions of people reach for sucralose-sweetened beverages and stevia-laced foods under the assumption that these compounds pass through the body without consequence. A new study published in Frontiers in Nutrition complicates that assumption in a significant way — suggesting that the effects of consuming these sweeteners may not stop with the person who eats them, but could propagate through at least two generations of unexposed offspring.
The research, led by Francisca Concha Celume and Martin Gotteland at the University of Chile, tracked three generations of mice in which only the founding generation (F0) consumed sucralose or stevia. Their children and grandchildren — who never received the sweeteners directly — still showed measurable alterations in gut microbiota composition, short-chain fatty acid (SCFA) production, and the expression of genes tied to inflammation and metabolism. The study was published April 10, 2026.
The scale of non-nutritive sweetener consumption makes the question an urgent one. More than 140 million Americans used some form of sugar substitute in 2020 alone, and the trend has accelerated across all age groups, including women of childbearing age.¹ In 2023, the World Health Organization issued a guideline questioning whether long-term NNS use offers any meaningful benefit for weight control, and flagged possible associations with increased risk of type 2 diabetes and cardiovascular disease.²
“These results challenge the long-standing assumption that non-nutritive sweeteners are metabolically inert and underscore their potential to influence offspring health through microbial and molecular pathways.” — Concha Celume et al.
The experimental design was straightforward but the stakes it raised were not. Forty-seven male and female C57BL/6J mice were divided into three groups: a control group receiving plain water, a sucralose group, and a stevia group. The sweetener doses — 0.1 mg/ml — were calibrated to approximate the FDA’s acceptable daily intake for humans. After 16 weeks on this regimen, the F0 mice were bred to produce the F1 generation, and F1 pairs were bred to produce F2. Neither F1 nor F2 animals received any sweeteners at any point.
What the researchers found in those unexposed generations was, in some respects, striking. In the sucralose lineage, male offspring in both the F1 and F2 generations showed mild but statistically significant disruptions in glucose tolerance — the body’s ability to process sugar efficiently. These changes persisted even as the sweetener itself was absent from the diet for two full generations. Stevia produced less pronounced effects: alterations appeared in the F1 generation but largely normalized by F2.
The gut microbiota data reinforced the divergence between the two sweeteners. Sucralose altered the composition of the fecal microbiome more dramatically than stevia across all three generations, affecting core taxa — the bacterial genera consistently present in all animals — at a level stevia never reached. Changes in beta-diversity, a measure of how microbial communities differ between individuals, remained statistically significant through the F2 generation in the sucralose group.
Perhaps the most biologically provocative finding concerned short-chain fatty acids, the molecules produced when gut bacteria ferment dietary fiber. SCFAs like butyrate, propionate, and acetate are not mere metabolic byproducts — they regulate inflammation, support the intestinal lining, inhibit certain enzymes involved in gene expression, and help maintain insulin sensitivity. In the F0 generation, both sucralose and stevia groups showed significantly reduced total SCFA concentrations compared to controls. That reduction persisted in F1 and F2, with the sucralose group in particular showing broader declines across individual SCFA species in successive generations.
“The intergenerational persistence of SCFA and BSCFA alterations highlights the need for further investigation into their long-term physiological impact.” — Concha Celume et al.
The researchers also tracked expression of several genes in the intestines and liver. Sucralose consumption in F0 mice was associated with elevated expression of Tlr4 and Tnf — genes involved in recognizing bacterial components and triggering inflammatory responses — as well as reduced expression of Srebp1, a liver gene that regulates fat synthesis and glucose metabolism. The inflammatory gene changes persisted into F1 in both the sucralose and stevia groups but largely normalized by F2. The Srebp1 suppression in the sucralose lineage, however, was more durable, remaining statistically significant through three generations.
The authors propose that the mechanism linking parental sweetener exposure to offspring outcomes runs through the microbiome itself. A 2020 study in Gut Microbes by Dai and colleagues found that maternal sucralose intake could alter the gut microbiota of offspring and worsen liver fat accumulation.³ A landmark 2014 paper in Nature by Suez and colleagues demonstrated that artificial sweeteners induce glucose intolerance in mice precisely by altering microbial community structure.⁴ The Chilean team builds on this body of evidence, arguing that dysbiotic microbiota passed from mothers to pups — alongside epigenetic changes in histone modification triggered by reduced butyrate — may be the primary transmission vehicle.
Outside experts reacted to the findings with a mixture of interest and methodological caution. Prof. Jules Griffin, Director of the Rowett Institute at the University of Aberdeen, acknowledged that the study “provides evidence that microbiome changes induced by artificial and natural non-nutritive sweeteners can occur across generations in mice,” while noting that mice are coprophagic — meaning they routinely consume their own feces — which provides an unusually efficient route for microbial transmission from parent to offspring not present in humans.
“The current paper also needs to be interpreted carefully in terms of its relevance for human health. These results are in mice and may not be translatable to humans.” — Prof. Jules Griffin, University of Aberdeen
Prof. Parveen Yaqoob of the University of Reading flagged the contested terrain the study enters when invoking transgenerational epigenetic inheritance. “The concept of epigenetic inheritance is highly contested,” she noted, adding that apparent second-generation effects in mice can sometimes reflect changes in the maternal environment or early-life microbial colonization rather than true genetic reprogramming. She also characterized the glucose tolerance differences as “subtle” and falling short of demonstrating clear clinical relevance.
Prof. Gunter Kuhnle, also of Reading, pointed to a structural challenge in the study: with a large number of measured endpoints, there is an elevated risk of false positives even with statistical corrections applied. He also noted that interpreting microbiome changes as inherently harmful remains genuinely difficult — many beneficial dietary interventions also shift microbial composition. Both EFSA and the UK’s Food Standards Agency have recently reaffirmed the safety of sucralose and stevia at approved intake levels.
The study’s authors are candid about their own limitations, acknowledging that their design cannot disentangle the effects of gestational exposure from perinatal ones, and that future research should isolate the contributions of each parent independently. The sample sizes, while consistent with comparable mouse studies, are small. Translation to human populations remains speculative.
But the significance of the study lies less in what it proves than in what it demands. The DOHaD framework — Developmental Origins of Health and Disease — has long argued that early environmental insults can echo across generations through epigenetic and microbial channels. If NNS exposure fits that framework even partially, the implications for dietary guidelines are substantial. Over 140 million Americans currently consume these compounds, many during pregnancy and early development, precisely because they are assumed to be inert. The Chilean team’s findings, taken alongside a growing body of animal research, suggest that assumption deserves continued and rigorous scrutiny.
Endnotes
1. Statista. “U.S. Population: Do You Use Sugar Substitutes?” (2020).
2. World Health Organization. “Use of Non-sugar Sweeteners: WHO Guideline.” (2023). https://www.who.int/publications/i/item/9789240073616
3. Dai X, Guo Z, Chen D, et al. “Maternal sucralose intake alters gut microbiota of offspring and exacerbates hepatic steatosis in adulthood.” Gut Microbes. (2020) 11:1043–63.
4. Suez J, Korem T, Zeevi D, et al. “Artificial sweeteners induce glucose intolerance by altering the gut microbiota.” Nature. (2014) 514:181–86.
5. Concha Celume F, Perez-Bravo F, Magne F, Olivares R, Gotteland M. “Artificial and natural non-nutritive sweeteners drive divergent gut and genetic responses across generations.” Frontiers in Nutrition. (2026) 13:1694149. DOI: 10.3389/fnut.2026.1694149
6. Science Media Centre. “Expert reaction to study looking at the sweeteners sucralose and stevia in mice and measures of gut microbiome, glucose oral tolerance and gene expression.” April 10, 2026. https://www.sciencemediacentre.org
7. Azad MB, Archibald A, Tomczyk MM, et al. “Nonnutritive sweetener consumption during pregnancy, adiposity, and adipocyte differentiation in offspring.” International Journal of Obesity. (2020) 44:2137–48.
8. Concha Celume F, Pérez-Bravo F, Gotteland M. “Sucralose and stevia consumption leads to intergenerational alterations in body weight and intestinal expression of histone deacetylase 3.” Nutrition. (2024) 125:112465.
Leave a Reply