The findings come from a team led by Hsin-Yu Wang, Chun Ming How, Yong-Shan Li, Yuqing Mao, Thanh H. Nguyen and Chia-Cheng Wei, who set out to answer a question that has become increasingly urgent in food safety research: do non-antibiotic chemicals, particularly the fungicides sprayed widely on fruit and vegetable crops, push bacteria toward resistance or heightened virulence? Prior work has hinted at the danger. Azoxystrobin and carbendazim have been shown to enrich bacterial resistomes in nematode guts, tebuconazole can promote the spread of multidrug-resistant plasmids in soil bacteria, and chlorothalonil facilitates metabolic adaptation in soil microbial communities. But whether such exposure produces phenotypic resistance and increased pathogenicity in a major foodborne pathogen remained largely untested.

Bacillus cereus was an obvious candidate for scrutiny. The Gram-positive, spore-forming organism is found in 36 to 45 percent of dairy products, vegetables, beans and cereals, and it is capable of causing food poisoning, eye infections, anthrax-like progressive pneumonia, fulminant sepsis and central nervous system infections. Multidrug-resistant strains of the species have already emerged in hospital wastewater, and its versatility—including the ability to build biofilms of varied architecture—makes any shift in its behavior a serious public health concern.

The researchers first screened eight widely used fungicides against B. cereus: chlorothalonil (CHT), propineb (PRO), tebuconazole (TEB), azoxystrobin, propiconazole, mancozeb, carbendazim and triadimefon. Three of them—CHT at 8 micromolar, TEB at 500 micromolar and PRO at 175 micromolar—completely inhibited bacterial growth within 24 hours and were selected for long-term adaptation experiments. The design was demanding: every day for 30 days, the bacteria endured a three-hour fungicide challenge followed by recovery and regrowth in fresh medium. Survival trajectories differed by compound. Under chlorothalonil, survival dipped to about 70 percent on day one but rebounded within 24 hours. Tebuconazole initially halved the population before recovery stabilized around day eight. Propineb proved the harshest pressure, dropping survival below five percent on day six before the bacteria clawed back to stable levels by day twelve. The bacterium, in short, adapted to all three chemical regimes.

Whole-genome sequencing of the adapted lineages delivered a surprising verdict: no meaningful genetic mutations. Phylogenetic comparison against reference strains and variant-calling analyses found the treated bacteria essentially identical to their ancestors. Instead of classical, mutation-driven resistance, the adaptation appears to be physiological—a reversible, non-heritable tolerance state akin to the persister-cell and stress-response phenomena documented in bacteria subjected to repeated antibiotic cycles. Similar patterns have been reported when Listeria monocytogenes and uropathogenic Escherichia coli were exposed to disinfectants such as benzalkonium chloride and triclosan, with minimum inhibitory concentrations rising without stable genetic change.

The phenotypic consequences, however, were substantial. Biofilm formation—often a shield against both immune attack and antimicrobial agents—was initially suppressed during early exposure but rose significantly in tebuconazole-adapted bacteria from day ten onward and climbed markedly in propineb-exposed cells by day ten. Statistical testing confirmed significant effects of the fungicide treatment, the duration of exposure, and their interaction on biofilm output. Antibiotic challenge assays revealed a parallel erosion of susceptibility. Bacteria adapted to chlorothalonil grew significantly better than controls in gentamicin at 4, 6 and 8 micrograms per milliliter; tebuconazole- and propineb-adapted lineages also outgrew controls at key gentamicin doses, and chlorothalonil- and tebuconazole-adapted cells showed improved growth at 8 micrograms per milliliter of tetracycline. Because no growth occurred at concentrations of 16 micrograms per milliliter or above, the strains do not meet formal clinical criteria for resistance—but the shift toward tolerance was clear and reproducible.

The most dramatic result emerged in living hosts. Using the nematode Caenorhabditis elegans, a genetically tractable infection model whose intestinal epithelium provides a biologically meaningful readout of colonization and killing, the team measured how fungicide-adapted bacteria fared against unadapted controls. All three adapted lineages killed worms significantly faster than the parent strain, with log-rank tests showing p values below 0.001. Tebuconazole-adapted bacteria were especially aggressive: worm survival collapsed within two days, and by day three most of the animals were dead. Follow-up colonization assays showed that tebuconazole-adapted B. cereus also established significantly higher intestinal loads in the worms, indicating that the fungicide had promoted persistence within the host gut, not merely faster killing.

Transcriptional profiling of the tebuconazole-adapted lineage offers a mechanistic window into these changes. Quantitative real-time PCR revealed significantly elevated expression of genes encoding the non-hemolytic enterotoxin (nheC) and the hemolysin BL complex (hblA, hblC and hblD)—toxins that disrupt intestinal epithelial cells—alongside upregulation of purC and purL, which support purine biosynthesis and extracellular DNA release during early biofilm formation, and calY, a bifunctional matrix protein that promotes adhesion to host tissues. The efflux-pump gene smr was also induced, a plausible explanation for the reduced antibiotic susceptibility, and one that echoes efflux upregulation seen in stressed Mycobacterium tuberculosis. Importantly, the elevated virulence and resistance gene expression persisted even when the adapted bacteria were subsequently exposed to gentamicin, suggesting that the stress-adapted state complicates antibiotic treatment rather than simply surviving it.

The authors are careful to frame the work as hazard identification rather than a direct portrait of what happens on farms or in food. The experiments used a single reference strain, BCRC15850, and the exposure concentrations—particularly 500 micromolar tebuconazole and 175 micromolar propineb—exceed the residue levels typically reported on treated foods, although the chlorothalonil dose is of the same order of magnitude as residues found in some food commodities. Local bioavailable concentrations in soil and produce depend on moisture, adsorption, formulation and degradation, so the laboratory model of recurrent acute stress cannot be directly translated into field-level risk estimates. Nor should the transcriptional findings be generalized beyond the tebuconazole lineage without confirming that chlorothalonil- and propineb-adapted bacteria share the same regulatory program. Nonetheless, the study reveals an understudied scenario in which persistent sublethal chemical stress can harden a major foodborne pathogen—improving its resilience, deepening its virulence and weakening the grip of frontline antibiotics—without any mutational fingerprint. Whether such phenotypes persist after fungicide withdrawal, and whether they arise in the genetically diverse field isolates that actually contaminate the food supply, are the questions the team now hopes will drive the next round of research.

Subject of Research: Effects of long-term fungicide exposure on adaptation, antibiotic tolerance and virulence of the foodborne pathogen Bacillus cereus

Article Title: Long-term fungicide exposure promotes bacterial adaptation and increases virulence of foodborne pathogen Bacillus cereus in Caenorhabditis elegans

Article References: Wang, H.-Y., How, C. M., Li, Y.-S., Mao, Y., Nguyen, T. H., & Wei, C.-C. (2026). Long-term fungicide exposure promotes bacterial adaptation and increases virulence of foodborne pathogen Bacillus cereus in Caenorhabditis elegans. Current Research in Food Science, 13, Article 101572. https://doi.org/10.1016/j.crfs.2026.101572

Image Credits: AI Generated

DOI: 10.1016/j.crfs.2026.101572

Keywords: Bacillus cereus, fungicide exposure, antimicrobial tolerance, biofilm formation, Caenorhabditis elegans, tebuconazole, chlorothalonil, propineb, virulence genes, food safety, foodborne pathogen, efflux pumps

The study, led by Aline Pompermaier of the Federal University of Fronteira Sul in Brazil with collaborators at Purdue University, focused on a commercial premix fungicide containing two active ingredients: azoxystrobin, a strobilurin compound that blocks mitochondrial respiration in fungi, and tebuconazole, a triazole that inhibits sterol biosynthesis in fungal cell membranes. These dual mechanisms, belonging to Fungicide Resistance Action Committee Groups C3 and G1, make the product highly effective against a broad range of crop diseases, from soybean rust to wheat stripe rust. It is registered for use on more than two dozen crops, including cotton, rice, corn, coffee, sugarcane, and grapes, and is sold in countries including Brazil, Australia, Argentina, Canada, Spain, and Germany.

What makes the new work particularly significant is that the researchers tested the product exactly as it is sold—as a ready-made formulation containing 120 grams per liter of azoxystrobin, 200 grams per liter of tebuconazole, and a substantial fraction of undisclosed excipients. Most previous ecotoxicology studies have either tested single active ingredients or combined pure compounds in the laboratory. Commercial premixes, by contrast, enter waterways as complete formulations, and the inert ingredients they contain can modify how the active substances are absorbed and how toxic they become. The authors argue this distinction matters, and their results suggest it does.

The exposure regime was designed around environmental reality. Zebrafish embryos were placed in solutions containing 30, 100, or 180 micrograms per liter of the fungicide, with concentrations benchmarked against the highest dose of the mixture component, tebuconazole. The two lower levels correspond to tebuconazole concentrations actually measured in surface and groundwaters in Brazil and elsewhere, while the highest matches Brazil’s maximum allowable limit for tebuconazole in drinking water. Notably, all three levels fall well below concentrations of azoxystrobin recorded in some Brazilian agricultural watersheds, where average detections have reached 1,055 micrograms per liter and peaks of 6,672 micrograms per liter—values that dwarf the experimental doses.

Embryos were exposed from 3 to 120 hours post-fertilization, a window covering the critical organogenesis phase when the nervous system, heart, and other organs take shape. The zebrafish is an ideal model for such work: its embryos are transparent, develop externally, and within days display species-specific behaviors that can be quantified with simple video recording. The team monitored 110 embryos per group for survival and hatching, tracked spontaneous movement in 20 embryos per group under a stereomicroscope at 24 hours post-fertilization, and counted heartbeats at 72 hours. Later, they filmed shoaling behavior at five days and tested antipredator responses at seven days using a moving red sphere displayed on a monitor beneath the test plates.

The survival results were stark. Kaplan-Meier analysis revealed that larvae exposed to 100 and 180 micrograms per liter died at significantly higher rates than controls. In the 100 microgram per liter group, the hazard ratio indicated a 70 percent lower probability of survival compared with unexposed animals; at 180 micrograms per liter, survival probability dropped by 63 percent. These are not laboratory extremes—they are concentrations of tebuconazole documented in real waterways and legally permitted in Brazilian tap water. Hatching, remarkably, was unaffected, meaning the fungicide killed embryos after they emerged rather than preventing emergence itself. The authors point out that azoxystrobin, unlike tebuconazole, has no established regulatory limit in drinking water in Brazil, a regulatory gap they describe as a genuine threat to aquatic life.

Equally troubling were the sublethal effects on developing physiology. At the highest concentration, embryos showed significantly reduced spontaneous movement, the early wriggling behavior controlled by the nascent nervous system. Because this reflexive coiling is orchestrated by developing neural circuits, its suppression is widely interpreted as an early warning sign of neurotoxicity or broader developmental impairment. Heart rate told a stranger story: exposure to 30 micrograms per liter slowed larvae heartbeats, while 100 micrograms per liter sped them up. This non-monotonic, potentially biphasic pattern mirrors hormetic responses documented in fish embryos exposed to other contaminants, and the authors caution that the underlying mechanism—whether adaptive compensation, direct cardiotoxicity, or oxidative stress—remains unresolved and warrants dedicated investigation.

The behavioral findings may be the most ecologically provocative. In the shoaling test, larvae exposed to 100 and 180 micrograms per liter bunched significantly closer together than controls, with reduced average distance between individuals. Zebrafish are intensely social animals, and tight shoaling is their primary defense against predators. But abnormally heightened cohesion can signal an anxiety-like state—a baseline perception of danger that persists even when no threat exists. The researchers suggest the exposed larvae behaved as though a predator were always present. Intriguingly, when the team later confronted the animals with an actual simulated threat in the aversive stimulus test, the exposed larvae responded normally, dodging away from the moving sphere just like controls. This dissociation is telling: it indicates the fungicide did not paralyze the animals or destroy their motor capacity, but rather shifted their default behavioral state, plausibly through dysregulation of cholinergic neurotransmitter pathways that govern anxiety and sociability.

The mechanistic picture drawn from prior literature is consistent with this interpretation. Azoxystrobin has been shown to disturb acetylcholinesterase activity, elevate reactive oxygen species, and disrupt dopaminergic signaling in zebrafish brains, while tebuconazole is known to inhibit acetylcholinesterase in larvae, induce oxidative stress in carp and zebrafish, and cause neurochemical and behavioral alterations. A mixture combining a mitochondrial respiratory inhibitor with a sterol biosynthesis inhibitor may therefore impose a dual burden—neural and oxidative—on embryos whose defenses are still under construction, potentially overwhelming redox homeostasis and cardiac regulation simultaneously. The authors are careful to note that they did not measure biochemical biomarkers directly, framing these pathways as plausible hypotheses grounded in the literature rather than confirmed mechanisms, and they call for follow-up work measuring acetylcholinesterase activity, antioxidant enzymes, lipid peroxidation, and gene expression.

The ecological and regulatory implications are difficult to overstate. A fish larva locked in a perpetual state of vigilance burns energy that should fuel growth and reproduction, and an abnormally cohesive shoal may become easier for predators to spot or less efficient at finding food. Reduced larval survival at environmentally realistic concentrations, meanwhile, can ripple through food webs and disrupt trophic cascades. The study’s central message is that environmental risk assessment must evolve beyond testing isolated active ingredients and counting corpses. Sublethal endpoints—heart function, embryonic movement, social behavior—capture harms that mortality statistics miss entirely, and commercially formulated mixtures must be evaluated as the integrated chemical systems they are. As the authors conclude, premix fungicides entering the environment as ready-made cocktails may act additively or synergistically on aquatic fauna, and regulatory frameworks that ignore this reality leave ecosystems—and potentially drinking water supplies—protected by standards built for a simpler chemical age.

Subject of Research: Toxicological effects of a commercial azoxystrobin-tebuconazole fungicide mixture on zebrafish physiology and behavior

Article Title: Beyond mortality: A fungicide mixture alters physiology and causes impaired behavior in zebrafish

Article References: Beyond mortality: A fungicide mixture alters physiology and causes impaired behavior in zebrafish. (n.d.). https://doi.org/10.1007/s10646-026-03147-z

Image Credits: AI Generated

DOI: 10.1007/s10646-026-03147-z

Keywords: zebrafish, fungicide mixture, azoxystrobin, tebuconazole, ecotoxicology, pesticides, shoaling behavior, developmental toxicity, heart rate, neurotoxicity, water contamination, environmental risk assessment