Every week, the average person may swallow roughly five grams of microplastic, about the weight of a credit card, according to a 2019 analysis commissioned by WWF and conducted at the University of Newcastle. Most of those fragments pass through the gut, but a growing body of research suggests some do not. They lodge in intestinal tissue, cross into the bloodstream, and turn up in organs. A team of South Korean scientists now reports that a single bacterium isolated from kimchi can grab onto nanoplastic particles and help drag them out of the digestive tract before they settle in.
The strain, Leuconostoc mesenteroides CBA3656, was identified by researchers at the World Institute of Kimchi, a government-funded institute under South Korea’s Ministry of Science and ICT. In a study published in Bioresource Technology in March 2026, the team showed that CBA3656 adsorbed 87% of polystyrene nanoplastics in standard lab conditions, held onto 57% of them after passing through fluids designed to mimic the human intestine, and roughly doubled the nanoplastic content in the feces of germ-free mice compared with untreated controls.
What the experiments actually showed
The researchers began with a straightforward binding test. They exposed CBA3656 to polystyrene nanoplastics in a controlled dish and measured how much plastic stuck to the bacterial surface. The result, 87% adsorption, was high but not unique on its own. A second kimchi-derived strain, Latilactobacillus sakei CBA3608, performed almost as well at 85%.
The real separation came in the next phase. When both strains were run through a simulated gastrointestinal environment, complete with bile salts, digestive enzymes, and the pH swings of the stomach and small intestine, CBA3656 retained 57% of its bound nanoplastics. CBA3608 collapsed to just 3%. That 54-percentage-point gap points to something specific about CBA3656’s surface chemistry that resists the harsh conditions of digestion, though the precise molecular mechanism has not yet been identified.
The animal trial used germ-free mice, which are raised in sterile isolators and carry no gut bacteria of their own. This setup lets scientists observe the effects of a single introduced microbe without interference from the hundreds of species that normally inhabit the intestine. Mice that received CBA3656 excreted more than twice the nanoplastics in their feces compared with controls, according to the ScienceDirect record for the paper. First author Jisu Lee and corresponding authors Tae Woong Whon and Se Hee Lee led the work, which was funded by the World Institute of Kimchi, the National Research Foundation under the Ministry of Science and ICT, and the Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry.
Why nanoplastics in the gut matter
The study lands in a field that has moved quickly over the past several years. Researchers using advanced in vitro models of the adult gut have demonstrated that microplastic particles do not behave like inert sand grains passing through plumbing. Instead, they form aggregates with resident microbes and digestive enzymes, and they can alter intestinal barrier function and lipid digestion. One such study, published in Environmental Health Perspectives, found that plastic particles changed how nutrients and bile components moved across the intestinal lining.
Reviews of immunotoxicity data have added another layer of concern, showing that nano- and microplastics can shift the composition of gut microbiota and interfere with normal intestinal immune responses. None of this work involved kimchi bacteria, but it establishes the biological logic behind the South Korean team’s approach: if plastics routinely interact with gut contents, a bacterium selected for unusually strong adhesion to those particles could, in principle, act as a biological escort, binding the fragments and carrying them toward the exit.
The large gaps that remain
The distance between a germ-free mouse and a living human gut is enormous. Germ-free animals lack the trillions of competing microbes that populate a normal intestine. In a real digestive tract, CBA3656 would need to maintain its plastic-binding ability while surrounded by hundreds of resident bacterial species, all competing for space, nutrients, and attachment sites on the gut wall. The published data do not yet address whether the strain can hold its advantage in that crowded environment.
Key details of the mouse experiment also remain limited in publicly available summaries. The greater-than-twofold excretion figure is confirmed in the paper’s abstract, but exact animal numbers, dosing schedules, treatment duration, and full statistical outputs have not been laid out in press materials. Without those specifics, outside scientists cannot fully evaluate effect size, variability, or how reproducible the results might be. Dose-response relationships, sex differences among the animals, and whether CBA3656 persists in the gut or washes out after dosing stops are all open questions.
Perhaps the most important limitation is what the study did not measure. Tracking nanoplastics leaving the gut in feces is not the same as proving that fewer particles ended up in tissues. The experiments did not assess whether plastic had already crossed the intestinal barrier into the bloodstream, liver, kidneys, or brain. Whether CBA3656 can intercept nanoplastics before they translocate into deeper tissues would require longer studies with organ sampling, imaging, and biomarkers for inflammation or oxidative stress.
The simulated intestine results, while striking, carry their own caveats. In vitro gut models replicate chemical conditions like pH and enzyme concentrations but cannot fully reproduce peristalsis, mucus turnover, or immune signaling. The 57% binding figure may shift substantially in a living system where transit time, diet composition, and mucosal interactions add layers of complexity. Dietary fibers, emulsifiers, and bile acids could compete with nanoplastics for bacterial binding sites or change how the bacteria themselves adhere to the intestinal wall.
Safety is another unresolved question. Many Leuconostoc strains are used in food fermentation and are generally regarded as safe in that context. But deliberately administering large quantities as a targeted microplastic-removal intervention is a different proposition. Researchers would need to confirm that concentrated doses do not promote harmful biofilms, facilitate antibiotic resistance gene transfer, or cause opportunistic infections in people with weakened immune systems.
No other probiotic has reached this stage for microplastic removal
What makes the CBA3656 work notable is not just the binding numbers but the absence of competition. As of June 2026, no other probiotic strain has been tested through this full sequence: high adsorption in vitro, retention through simulated digestion, and increased excretion in a live animal model. Other research groups have explored whether certain bacteria or dietary fibers might reduce microplastic absorption, but none have published a comparable set of results in a peer-reviewed journal. That makes CBA3656 a first mover in a space that barely existed five years ago, though being first also means the findings have not yet been replicated by independent labs.
The study was published in Bioresource Technology, an established Elsevier journal with a strong track record in environmental and applied microbiology research. Peer review at that level provides a meaningful quality check, but it does not substitute for replication. A single paper from a single research group, however well-designed, is the starting line for scientific confidence, not the finish.
What would need to happen next
Moving CBA3656 from a laboratory curiosity to anything resembling a consumer product would require several distinct steps. Conventional (non-germ-free) mouse studies would need to show that the strain still increases plastic excretion in the presence of a full gut microbiome. Tissue-level analyses would need to demonstrate that enhanced fecal excretion actually translates to lower plastic burdens in organs. Toxicology and safety profiling would need to establish appropriate dosing ranges. And eventually, human clinical trials, likely starting with small pharmacokinetic studies, would need to confirm that the bacterium survives the human stomach, colonizes or at least transits the intestine in sufficient numbers, and binds plastics without causing adverse effects.
Each of those steps could take years. Regulatory pathways for a probiotic marketed with health claims vary by country but universally demand evidence well beyond what a single animal study provides.
For now, CBA3656 is best understood as a proof of concept: a microbe plucked from a jar of fermented cabbage that can latch onto synthetic particles and escort them out of a simplified gut system. The work highlights a genuinely novel intersection between traditional fermented foods and modern environmental contamination. It also serves as a reminder that promising mouse data, no matter how clean, are only the first chapter of a much longer scientific story.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
