The research, led by Unnikrishnan Kannan and corresponding author Saji George at McGill University, together with colleagues at Manipal Academy of Higher Education, the Canadian Light Source, and Christ University, set out to test a fundamental assumption in the field. Toxicity studies of microplastics and nanoplastics have largely depended on primary plastic particles with smooth surfaces, yet the particles to which humans are predominantly exposed in food and environmental matrices are secondary particles, generated by fragmentation of larger plastic items. If those real-world particles behave differently from laboratory standards, then much of the existing hazard literature may need re-examination.

To create realistic test materials, the team mechanically fragmented two of the world’s most common plastics: high-density polyethylene, widely used in containers and packaging, and polyethylene terephthalate, the polymer of beverage bottles. From this process they generated microplastic particles in the 100 to 350 micrometer range and nanoplastic particles below 1000 nanometers. Molecular fingerprinting confirmed a crucial point: the parent polymer identity was retained after fragmentation. In other words, the fragments were still recognizably HDPE or PET. What changed was everything on and just beneath the surface.

Using an array of physicochemical characterization techniques, the researchers found that smaller particles showed greater surface roughness, altered surface charge, reduced crystallinity, chemical heterogeneity, and enhanced surface oxidation. These are not cosmetic differences. Crystallinity reflects how orderly the polymer chains are packed; mechanical grinding shatters that order, leaving disordered, strained, and chemically unsatisfied structures at the newly created surfaces. Those defect-rich surfaces are precisely where reactive chemistry happens. The two polymers diverged in instructive ways: HDPE nanoplastics remained mainly hydrocarbon rich but structurally disordered, whereas PET nanoplastics became oxygen enriched, with increased contributions from carbon-oxygen single bonds and ester groups of the O-C=O type.

The functional consequences of these surface changes were assessed with two complementary approaches. Electron paramagnetic resonance, a technique capable of detecting unpaired electrons and therefore free radicals, revealed stronger radical-associated activity in nanoplastics than in microplastics. Chemical probe assays told the same story from another angle, showing enhanced abiotic generation of reactive oxygen species, the aggressive oxygen-containing molecules that can damage lipids, proteins, and DNA. Reactive oxygen species produced by the particles themselves, without any biological catalyst, represent a form of intrinsic chemical hazard that travels with the particle wherever it goes.

The team then asked whether this heightened oxidative activity translates into biological damage. In experiments with erythrocyte membranes, the red blood cell membranes served as a model target, nanoplastics induced greater membrane lysis than their microplastic counterparts. Membrane lysis, the rupture of the lipid bilayer that encloses cells, is a direct physical-and-chemical assault that depends heavily on the particle surface contacting the membrane. Rougher, more oxidized, more reactive surfaces appear to make that contact more destructive.

Cellular studies reinforced the pattern. When Caco-2 cells, a widely used human intestinal epithelial model, were exposed to the particles, the nanoplastics triggered greater intracellular oxidative stress. Because Caco-2 cells model the gut lining, the first biological barrier that ingested plastic particles encounter, this finding is particularly relevant to dietary exposure. The result links the physical process of fragmentation, through altered surface chemistry, to a plausible mechanism of harm inside human-relevant cells.

One of the most nuanced findings of the study concerns the protein corona. When the researchers allowed bovine serum albumin, a common blood protein, to coat the particles before exposure, the cellular oxidation was attenuated. This indicates that the accessible particle surface, not merely the particle’s size or polymer type, drives biological reactivity. A protein layer acts as a buffer, masking the reactive sites that would otherwise generate radicals and attack membranes. The implication is double-edged: in protein-rich environments such as blood or digestive fluids, some of the particles’ oxidative aggression may be dampened, but the underlying reactive capacity remains, ready to manifest wherever the corona is absent or stripped away.

Taken together, the results carry a clear message for hazard assessment. The smaller a plastic particle becomes through mechanical fragmentation, the more reactive its surface grows, and the greater its potential to generate reactive oxygen species, lyse membranes, and stress cells. This means that size is not merely a matter of where particles can travel in the body, the usual focus of nanoplastic concern, but also of what those particles do chemically once they arrive. Secondary nanoplastics formed in the environment may be intrinsically more hazardous than the smooth primary particles that dominate laboratory test kits, meaning current risk assessments built on model beads could underestimate real-world toxicity.

The study also highlights the value of methodological rigor. By combining molecular fingerprinting, electron paramagnetic resonance, chemical probe assays, membrane lysis tests, and cell-based oxidative stress measurements, the researchers built a chain of evidence connecting a physical process, mechanical size reduction, to altered surface chemistry and finally to biological effect. The work, conducted with facilities including the Canadian Light Source and McGill’s electron microscopy and imaging platforms and supported by NSERC and Canada Foundation for Innovation funding, offers a template for how future toxicity studies might evaluate realistic secondary particles rather than idealized ones. As plastic pollution continues to fragment across ecosystems, from ocean surfaces to kitchen countertops, understanding that every crack and grind marks a fresh, reactive surface may reshape how scientists, regulators, and the public judge the true hazard potential of the plastic particles we all encounter daily.

Subject of Research: Surface reactivity and toxicity of mechanically fragmented secondary microplastics and nanoplastics

Article Title: Surface reactivity of plastic particles increases with size reduction during mechanical fragmentation: Implications for hazard potential

Article References: Kannan, U., Sangeeth, K., Wang, J., Lukose, J., Vinod, T. P., & George, S. (2026). Surface reactivity of plastic particles increases with size reduction during mechanical fragmentation: Implications for hazard potential. Microplastics and Nanoplastics. https://doi.org/10.1186/s43591-026-00231-3

Image Credits: AI Generated

DOI: 10.1186/s43591-026-00231-3

Keywords: microplastics, nanoplastics, mechanical fragmentation, surface reactivity, reactive oxygen species, free radicals, electron paramagnetic resonance, membrane lysis, Caco-2 cells, HDPE, PET, oxidative stress

The importance of this question lies in the peculiar vulnerability of marine products. Fish muscle is dominated by myofibrillar proteins, the structural workhorses that determine whether a fillet holds together, retains its juices, and forms a proper gel in products like surimi. At the same time, marine lipids are unusually rich in polyunsaturated fatty acids, which are notoriously easy for oxygen and free radicals to attack. Sturgeon, an increasingly important aquaculture species with annual global production exceeding 100,000 tonnes, is a case in point: polyunsaturated fatty acids account for roughly 40 percent of the total fatty acids in its meat, and that meat makes up 40 to 50 percent of the animal’s body weight. When lipid oxidation and protein oxidation occur in the same package, they do not proceed independently. Lipid hydroperoxides and their breakdown products, such as malondialdehyde and 4-hydroxynonenal, chemically attack amino acid side chains on proteins, triggering carbonylation, cross-linking, and aggregation. In turn, protein-derived radicals can kick off lipid peroxidation chain reactions. The two processes feed each other in a mutually destructive spiral.

What had been missing, the authors argue, was a systematic map of how this co-oxidation behaves across a genuinely broad moisture range. Previous studies had tended to focus either on dry powders below 30 percent moisture or on dilute emulsions above 90 percent, with only narrow windows in between. Yet the commercial products most at risk, including semi-moist restructured fish items, ready-to-eat marine protein snacks, and prepared gel products, typically sit in the intermediate zone. To fill the gap, the team prepared five ternary systems containing 10 percent lipid and moisture contents of 10, 30, 50, 70, and 90 percent. All samples were sterilized with a 10 kGy dose of gamma irradiation to eliminate microbial activity, ensuring that any changes observed were purely physicochemical, and then stored at 4 degrees Celsius, the temperature of real cold-chain distribution, for 35 days.

To capture the full arc of oxidation, the researchers went beyond single time-point measurements. Because several oxidation indicators rise and then fall during storage, they calculated integrated indices, the protein oxidation index and lipid oxidation index, from the area under the curve of standardized change rates. These confirmed what the raw data hinted at: both indices climbed steadily from 10 percent moisture to a maximum at 70 percent, where the protein index reached 1.035 and the lipid index 8.168, before dropping when moisture rose to 90 percent. The lipid index was always far higher than the protein index, underscoring just how much more vulnerable unsaturated fats are to oxidative attack, but the ratio between the two also peaked at 70 percent moisture, indicating that under intermediate-moisture conditions protein oxidation responded most strongly to the lipid oxidation occurring around it.

The protein-side evidence was written into the molecules themselves. Total and reactive sulfhydryl contents, markers of oxidation-sensitive cysteine residues, fell in every sample, but the losses were steepest at high moisture: total sulfhydryl content dropped by 61 percent at 70 percent moisture and 49 percent at 90 percent, compared with only 25 percent in the driest system. Carbonyl content, a signature of amino acid side-chain damage, peaked highest at 70 percent moisture. Raman spectroscopy revealed that the ordered alpha-helical structure of the proteins disintegrated most severely in the 70 percent system, collapsing from 64.07 percent to 40.24 percent over the storage period while beta-sheet and random coil fractions swelled. Intrinsic fluorescence showed tryptophan residues, normally buried in the protein interior, becoming exposed to the polar aqueous environment as the tertiary structure loosened, with the emission maximum shifting from about 355 to roughly 380 nanometers in all samples.

Free radical measurements tied the structural damage directly to reactive chemistry. Using electron paramagnetic resonance spectroscopy with spin-trapping agents, the team tracked superoxide anion radicals and hydroxyl radicals. Superoxide rose and then declined, peaking at 21 days in the 50 and 70 percent systems, with the 70 percent sample reaching the highest peak of approximately 2.37 micromoles per milligram, consistent with its role as a transient intermediate that converts into other reactive species. Hydroxyl radicals, by contrast, accumulated continuously, and at the end of storage their levels followed the order 70, 50, 90, 30, then 10 percent moisture. Intermediate water content, it appeared, created ideal conditions for generating and propagating the very radicals that shred proteins and peroxidize lipids.

The explanation lies in water’s double-edged character. Water activity measurements showed the familiar pattern: values of about 0.70 at 10 percent moisture and 0.92 at 30 percent, then a plateau near 0.96 for everything wetter. Nuclear magnetic resonance added the crucial detail about water’s physical state. Free water content rose with total moisture, exceeding 80 percent in the 70 and 90 percent systems, while bound and immobile water fractions shrank. At low moisture, restricted molecular mobility and limited hydration keep oxidation-sensitive residues hidden and reaction partners apart, slowing the chemistry. At moderate moisture, water acts as a plasticizer, lubricating protein chains, opening up buried residues, and giving oxygen, radicals, and lipid oxidation products the mobility they need to find their targets. But at 90 percent moisture, the tide turns: excess water dilutes the reactants, oxygen must diffuse through an ever-thicker aqueous layer to reach the lipid and protein interfaces, and the probability of productive oxidative collisions drops. Oxidation at high moisture is blunted, not by stopping the chemistry, but by starving it of concentration.

Multivariate statistics sharpened this mechanistic picture. Correlation analysis showed that moisture-related parameters, including water activity and the proportions of free, immobile, and bound water, correlated more strongly with oxidation and structural indicators than moisture content itself, suggesting that water’s influence operates through the state it adopts rather than its mere quantity. Principal component analysis separated the five systems cleanly along the first principal axis, which explained 48.6 percent of the variance, with the 10 and 30 percent systems clustering with bound water and preserved protein structure, the 90 percent system clustering with free water and high water activity but only weakly with oxidation markers, and the 50 and 70 percent systems sitting squarely in the oxidation zone. The authors suggest a transition mechanism, and possibly a critical threshold, lies somewhere in the 40 to 60 percent moisture range.

A final validation brought the model back to reality. The team compared the 70 percent model system against native minced sturgeon meat and a composition-matched sturgeon meat sample adjusted to the same protein, lipid, and moisture contents. Both real matrices oxidized more severely than the model, with native meat posting a lipid oxidation index of 21.242 against the model’s 8.168, a difference the researchers attribute to endogenous pro-oxidants such as heme proteins, non-heme iron, oxidative enzymes, and membrane phospholipids that the simplified ternary system lacks. Yet the temporal fingerprint, the sequence and rhythm of rises and falls across sulfhydryl loss, carbonyl formation, peroxide buildup, malondialdehyde accumulation, and radical generation, matched closely between model and meat. That consistency matters, because it means the clean ternary system can serve as a legitimate platform for studying co-oxidation dynamics without the confounding complexity of whole tissue.

The practical implications reach well beyond sturgeon. For manufacturers of surimi, fish protein powders, restructured seafood, and intermediate-moisture meat snacks, the message is that formulation moisture is not a passive variable but an active lever on oxidative stability. Products formulated in the intermediate range may be sitting in the worst possible chemical neighborhood, where just enough water mobilizes radicals and oxidation products without enough to dilute them. Targeted moisture adjustment, alongside conventional antioxidant strategies, could therefore extend refrigerated shelf life in ways that freezing alone cannot. More broadly, the study adds a compelling chapter to a growing recognition that the water in food is never inert: it is a solvent, a plasticizer, a diffusion highway, and a diluent all at once, and knowing which of those roles it is playing at any given moisture level may be the key to keeping protein-rich foods fresh.

Subject of Research: Effect of moisture content on protein-lipid co-oxidation in sturgeon muscle model systems

Article Title: Effects of moisture content on the co-oxidation behavior of sturgeon myofibrillar protein and sturgeon oil

Article References: Fang, Y., Yin, Z., Yu, J., Huang, G., & Li, J. (2026). Effects of moisture content on the co-oxidation behavior of sturgeon myofibrillar protein and sturgeon oil. Food Chemistry: X, 39, Article 104481. https://doi.org/10.1016/j.fochx.2026.104481

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104481

Keywords: sturgeon, protein oxidation, lipid oxidation, co-oxidation, moisture content, water activity, free radicals, food chemistry, myofibrillar protein, shelf life, marine food, refrigerated storage