75% of Sweeteners Disrupted Gut Bacteria, Antidepressant Combo Was Worst

A University of Cambridge laboratory study published June 25, 2026 in Molecular Systems Biology tested 39 commercially used sweeteners against 25 gut bacterial species and found that about three-quarters of them disrupted microbial growth. The research, led by Dr. Sonja Blasche and Professor Kiran Patil at the MRC Toxicology Unit, also mapped how sweeteners interact with medications, caffeine, and food flavorings – turning up more than 100 distinct interaction effects. The most alarming finding was not a sweetener acting alone, but a specific pairing: a stevia-derived compound combined with a common antidepressant.

39 Sweeteners Tested: 75% Disrupted Gut Bacterial Growth in Lab

The study screened a broad range of sweeteners – artificial, low-calorie, and natural varieties – at concentrations relevant to commercial food and beverage use. Researchers introduced each compound to cultures of 25 gut bacterial species and measured bacterial growth rates along with the overall diversity of synthetic bacterial communities.

The result: approximately 75% of the 39 sweeteners altered the growth of at least one bacterial species. That is a striking proportion for compounds routinely marketed as metabolically neutral. The prevailing regulatory assumption is that because sweeteners pass through the body without providing calories, they are biologically inert. This study challenges that framing, at least in controlled lab conditions.

Dr. Blasche and Professor Patil's approach was deliberately systematic. Most prior research tested individual sweeteners against individual bacteria in isolation. By running 39 compounds across 25 bacterial species, this work maps how an entire class of food additive behaves in a microbial context rather than cherry-picking one compound for study.

Isosteviol Combined With Duloxetine Suppressed Two Key Gut Bacteria

The sharpest finding came from a combination experiment. When researchers paired isosteviol, a compound derived from the stevia plant used in commercial food and beverage products, with duloxetine (an antidepressant, brand name Cymbalta), two gut bacterial species were strongly suppressed: Roseburia intestinalis and Parabacteroides merdae. Both are bacteria linked to healthy digestion and blood sugar regulation, according to the study.

Their suppression in synthetic bacterial communities was accompanied by a reduction in overall microbial diversity, a metric broadly associated with gut resilience and immune function. A 2021 review in Frontiers in Cellular and Infection Microbiology described Roseburia intestinalis as a butyrate-producing bacterium that prevents intestinal inflammation and maintains energy homeostasis – making its suppression a finding worth watching in any subsequent human study.

The effects extended beyond bacteria. The same isosteviol-duloxetine combination altered inflammatory signaling in cultured intestinal cells, specifically reducing secretion of IL-6 and IL-8, two cytokines (signaling proteins involved in immune responses), while increasing cellular toxicity in the laboratory model.

One clarification the researchers make explicit: isosteviol is an intermediate compound produced during the processing of steviol glycosides. It is not the same as the "stevia" label on a food package. Whether consumer-level stevia products contain sufficient isosteviol to produce comparable effects in a living gut is a separate question the study does not address.

The antidepressant side of this finding also has prior context. A 2023 peer-reviewed study in Frontiers in Behavioral Neuroscience found that all six antidepressants tested, including duloxetine, showed inhibitory effects on gut microorganism growth in in vitro conditions. The Cambridge study suggests that pairing duloxetine with a sweetener compound may compound that disruption significantly.

How Researchers Screened 39 Sweeteners Across 25 Bacterial Species

Beyond single-compound tests, the Cambridge team ran combination experiments: sweeteners paired with caffeine, vanilla extract, medications, and other common dietary exposures. Across all combinations, they identified more than 100 interaction effects.

Of those, 34 combinations intensified microbial disruption while 68 combinations weakened it. That bidirectionality matters. Some pairings amplified harm; others appeared to dampen it. The direction of the effect depends on which compounds are combined and which bacteria are in the mix.

This matters because gut biology is not a single-compound problem. People consume multiple sweeteners, medications, caffeine, and flavorings simultaneously and continuously. Studying any one sweetener in isolation, as most prior research has done, may systematically miss the effects that emerge only in combination.

All Findings Are From Lab Cultures, Not Living Humans

Everything described above happened in test tubes and cell cultures. That distinction is load-bearing. The human gut hosts hundreds of bacterial species interacting continuously with fiber, nutrients, immune signals, medications, and one another. A synthetic community of 25 bacterial species in a laboratory vessel does not replicate that complexity.

Interactions that disrupt bacteria in culture may behave differently in a living system. They could be amplified, suppressed, or overridden entirely by other factors. The researchers are explicit: the clinical significance of these bacterial disruptions is unknown, and human studies are needed to determine whether similar interactions occur in actual digestive systems.

The study was funded by the EU Horizon 2020 programme and the UK Medical Research Council, with no commercial conflicts of interest identified.

Sweetener Research Gets a Roadmap, but Human Trials Are Still Needed

What this study provides is a prioritized target list. Rather than requiring clinical trials for every possible sweetener-drug combination – an impossibly large parameter space – the Cambridge screen identifies specific pairs and conditions that produced the largest in vitro disruptions. The isosteviol-duloxetine combination now has a scientific basis for follow-up in animal and human models.

The broader implication is methodological: if sweeteners' effects on gut bacteria change meaningfully depending on what else is present in the gut, then studying any single sweetener in isolation may miss the most important effects. A 2026 review in Current Nutrition Reports found that common sweeteners including steviol glycosides alter gut bacterial composition and downstream metabolic processes – an active but unresolved field that the Cambridge work now advances.

For now, the study's lead researchers note that sweeteners are often marketed as metabolically neutral, and that this lab evidence challenges that assumption. The next step is finding out whether those same interactions unfold inside a human body – and if they do, what they mean for health.

References

1. (2026). Think diet sweeteners are harmless? Here's how they alter gut-brain signalling. https://economictimes.com/news/india/think-diet-sweeteners-are-harmless-heres-how-they-alter-gut-brain-signalling/articleshow/132703025.cms
2. Kai Nie, Kejia Ma, Weiwei Luo, et al. (2021). Roseburia intestinalis: A Beneficial Gut Organism From the Discoveries in Genus and Species. Frontiers in Cellular and Infection Microbiology. https://doi.org/10.3389/fcimb.2021.757718
3. Grigory Rukavishnikov, Lubov Leonova, Evgeny Kasyanov, Vadim Leonov, Nikholay Neznanov, Galina Mazo (2023). Antimicrobial activity of antidepressants on normal gut microbiota: Results of the in vitro study. Frontiers in Behavioral Neuroscience. https://doi.org/10.3389/fnbeh.2023.1132127
4. Rabie Khattab (2026). Artificial Sweeteners and Gut Microbiota: Mechanistic Insights and Implications for Metabolic Health. Current Nutrition Reports. https://doi.org/10.1007/s13668-026-00768-y