The research, led by scientists at the Korea Research Institute of Bioscience and Biotechnology together with colleagues at Hanyang University ERICA and Dong-A University, set out to answer a deceptively simple question: how much of the plant’s nutritional and antioxidant value depends on when it is picked and where it is grown? To find out, the team cultivated G. affine under two different cultivation environments and sampled it at multiple growth stages, then subjected every sample to ultra-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry, a technique abbreviated as UPLC-QTOF/MS. This platform allows researchers to detect and tentatively identify hundreds of small molecules in a single run, generating a chemical fingerprint of each sample that can be compared statistically across conditions.

In total, the team annotated fifty metabolites, and from that pool they isolated eight compounds through successive chromatographic separations, determining their structures using nuclear magnetic resonance spectroscopy and mass spectrometry. This is a crucial step that many metabolomics studies skip. Untargeted profiling can suggest what a compound might be, but only isolation and structural elucidation provide certainty about identity. By purifying these eight molecules, the researchers created their own reference standards, which they then used to build a targeted high-performance liquid chromatography method with diode-array detection, or HPLC-DAD, capable of precisely quantifying the most important constituents in any sample of the plant.

The analytical performance of that quantification method was rigorous. Calibration curves showed excellent linearity, with coefficients of determination of at least 0.999 across the tested ranges, and detection limits fell between 0.671 and 1.088 micrograms per milliliter. In practical terms, this means the method can reliably detect and measure even trace amounts of the key bioactive compounds, making it suitable for quality control in food and nutraceutical manufacturing, where batch-to-batch consistency is essential and regulatory standards demand validated analytical procedures.

To make sense of the enormous datasets generated by untargeted metabolomics, the researchers turned to chemometrics, the statistical analysis of chemical data. Principal component analysis, or PCA, provided an unbiased overview of the variation, and it revealed that the metabolite profiles of G. affine were shaped by both growth stage and cultivation environment. Samples clustered according to how mature the plants were and where they had been grown, indicating that these two factors leave distinct chemical signatures in the plant tissue. A second, supervised method, partial least squares discriminant analysis, or PLS-DA, sharpened the picture further, showing particularly clear separation between samples from the two cultivation environments. Supervised models like PLS-DA actively search for the variables that best distinguish predefined groups, so the strong discrimination confirmed that growing conditions exert a powerful and consistent influence on the plant’s secondary chemistry.

The compounds driving that discrimination belonged overwhelmingly to two chemical families: phenylpropanoids and flavonoids. Within these broad classes, three subgroups stood out as the main discriminatory metabolites. Caffeoylquinic acid derivatives, which arise when caffeic acid is esterified with quinic acid, are well-known phenolic compounds with documented antioxidant and anti-inflammatory activities in the food science literature. Quercetin glycosides and luteolin glycosides, meanwhile, are sugar-conjugated forms of two of the most studied plant flavonoids, molecules that plants typically produce as protective agents against ultraviolet radiation, herbivores, and oxidative stress. That these particular compounds dominated the statistical separation suggests that the plant’s defensive chemistry is highly sensitive to both its developmental program and its growing environment.

The functional consequences of these chemical shifts were confirmed by classical assays. The researchers measured total phenolic content and total flavonoid content across all samples, alongside antioxidant capacity tests. Both the phytochemical contents and the antioxidant activities varied significantly across growth stages and between the two cultivation environments, mirroring the metabolomic patterns. Most notably, samples grown in the open field showed increased phenolic contents and enhanced ABTS radical-scavenging activity at later growth stages under the conditions examined. The ABTS assay works by measuring a sample’s ability to neutralize a synthetic green-colored radical cation, providing a rapid proxy for antioxidant capacity. The finding that open-field plants became chemically richer as they matured is consistent with a well-established principle in plant physiology: environmental stresses such as full sunlight exposure tend to stimulate the phenylpropanoid and flavonoid pathways, prompting plants to accumulate protective phenolic compounds.

For the food and nutraceutical industries, these results carry immediate practical weight. G. affine is already consumed as a food and medicinal plant valued for its nutritional and bioactive properties, but like many botanical raw materials, its chemical quality can fluctuate dramatically depending on agronomic choices. The study demonstrates that harvest timing and cultivation system are not minor variables but primary determinants of the plant’s functional value. A processor seeking maximum phenolic content and antioxidant activity, for example, now has evidence-based guidance pointing toward open-field cultivation and later growth stages under the conditions tested. Conversely, the detailed metabolite maps allow manufacturers to verify that incoming plant material meets expected chemical specifications, using the validated HPLC-DAD method as a quality gate.

Scientifically, the work also illustrates the power of combining untargeted and targeted approaches. Untargeted metabolomics excels at discovery, revealing unexpected patterns and flagging candidate biomarkers, but it produces tentative identifications that require confirmation. By following up with isolation, structural elucidation, and targeted quantification using purified reference standards, the Korean team closed the loop from pattern to proof. The fifty annotated metabolites and eight fully characterized compounds now constitute a chemical library for G. affine that future researchers can build upon, whether they are investigating the plant’s bioactivity, breeding improved cultivars, or optimizing cultivation protocols for industrial production.

The broader significance extends beyond a single species. As consumer demand for plant-based functional foods grows, the industry increasingly needs rigorous, systems-level data on how agricultural practices shape phytochemical quality. This study offers a template: profile broadly, discriminate statistically, isolate the key players, quantify them precisely, and connect the chemistry to measurable biological function. Applied to G. affine, that template transformed a traditional spring vegetable into a chemically characterized functional food candidate, and the authors conclude that their findings support the plant as a promising resource for future nutraceutical applications. For a plant that has quietly flavored festival rice cakes for centuries, modern metabolomics has finally revealed just how much its chemistry depends on the sun, the soil, and the season.

Subject of Research: Metabolomic profiling of secondary metabolites in Gnaphalium affine across growth stages and cultivation systems

Article Title: Comparative metabolomic characterization of secondary metabolites in Gnaphalium affine D. Don considering the influence of growth stages and cultivation systems

Article References: Hong, S., Lee, H. G., Oh, S. M., Kim, H.-G., An, Y. M., Lee, S.-Y., Kim, D.-Y., Oh, S.-R., Kim, C. Y., Lee, J. H., & Ryu, H. W. (2026). Comparative metabolomic characterization of secondary metabolites in Gnaphalium affine D. Don considering the influence of growth stages and cultivation systems. npj Science of Food. https://doi.org/10.1038/s41538-026-01194-8

Image Credits: AI Generated

DOI: 10.1038/s41538-026-01194-8

Keywords: Gnaphalium affine, metabolomics, secondary metabolites, flavonoids, phenylpropanoids, caffeoylquinic acid, quercetin glycosides, luteolin glycosides, antioxidant activity, cultivation systems, UPLC-QTOF/MS, functional food

The target of the protection is a group of molecules that most consumers never think about but every meat eater experiences: myofibrillar proteins. These proteins, which make up roughly 60 to 70 percent of the total protein in muscle, govern the texture, water-holding capacity, and gel-forming behavior of meat products. Unfortunately, they are also exquisitely vulnerable to oxidation. During processing and storage, reactive oxygen species generated by lipid peroxidation, Fenton chemistry involving iron and hydrogen peroxide, and enzymatic pathways attack amino acid side chains. The result is protein carbonylation, the loss of sulfhydryl groups, structural disruption, aggregation, moisture loss, and the dreaded textural hardening that ruins otherwise good meat. Synthetic antioxidants can slow this decay, but consumer demand for clean-label products has pushed the industry toward plant-derived alternatives, and that is where fish mint enters the story.

Houttuynia cordata is a perennial herb distributed across East and Southeast Asia with a long history in folk medicine and functional foods, boasting antioxidant, antimicrobial, anti-inflammatory, and immunomodulatory activities. In China it is officially recognized as a substance traditionally used as both food and medicine, and recent work has suggested it can enhance the antioxidant capacity of muscle tissue in living animals. Yet its direct use in food preservation has been hampered by real problems: a pungent fishy odor from volatile compounds, astringent off-flavors from tannins, and poor bioavailability because a large share of its polyphenols are locked in glycosylated or cell wall-bound forms that the body and the food matrix cannot easily access. The research team hypothesized that the right microbe could unlock that trapped potential.

The microbe they chose was Lactiplantibacillus paraplantarum L-ZS9, a genome-sequenced strain originally isolated from fermented meat and deposited in the China Center of Industrial Culture Collection. Lactic acid bacteria are renowned bioconverters of plant phenolics. They carry glycoside hydrolases such as beta-glucosidases and beta-galactosidases that clip sugar groups off phenolic glycosides, releasing free aglycones with stronger antioxidant activity and better bioavailability. Their fermentation also produces organic acids that lower pH and trigger acid-catalyzed hydrolysis of cell wall polysaccharides, liberating even more entrapped phenolics. Microbial biotransformation through deglycosylation, hydroxylation, and ring cleavage can generate entirely novel bioactive metabolites along the way.

The experimental setup was elegantly simple. Fresh fish mint leaves from a Chengdu market were washed, homogenized with sterile water, and fermented at 36 degrees Celsius with L-ZS9 for six days, with three independent batches serving as biological replicates. The fermentation behaved exactly as a healthy lactic acid process should. Viable bacterial counts surged from 6.0 to 8.2 log CFU/mL within the first 24 hours, while pH dropped stepwise from 3.95 to 3.70 over three days as lactic acid accumulated, then plateaued as carbohydrates were exhausted. Total sugar fell from 5.2 to 2.6 mg/mL in two days, and reducing sugars plummeted from 1.6 to 0.35 mg/mL within a single day before partially recovering, revealing a dynamic tug-of-war between microbial consumption and acid- and enzyme-mediated release of sugars from the plant matrix.

The headline result was a striking enrichment of phenolic compounds. Total polyphenol content climbed progressively from 1.02 to 2.50 mg GAE/mL over the six days, a 2.45-fold increase, with the fastest accumulation coinciding with peak bacterial growth and acidification. Total flavonoids followed a slightly delayed trajectory, peaking at 1.12 mg RE/mL on day five. Heatmap profiling of individual compounds showed the enrichment was driven largely by three quercetin glycosides: isoquercitrin, rutin, and quercitrin, all of which rose markedly by day six, presumably liberated from cell wall-bound stores by acid and enzymatic disruption. Meanwhile, hyperoside declined, consistent with beta-galactosidase cleavage of its galactosidic bond and a concomitant rise in free quercetin, and the flavan-3-ols catechin and epicatechin also dropped significantly, possibly through ring cleavage reactions previously documented in the closely related Lactiplantibacillus plantarum.

Did the chemical changes translate into real antioxidant power? Four complementary radical scavenging assays said yes. ABTS radical scavenging capacity approximately doubled from 75 to 150 micromol Trolox equivalents per liter, while superoxide anion scavenging rose 68 percent, DPPH scavenging 63 percent, and hydroxyl radical scavenging 61 percent, all statistically significant. Pearson correlation analysis revealed that isoquercitrin and rutin were the compounds most strongly and positively correlated with antioxidant activity across all four assays, with correlation coefficients reaching as high as 0.98, strongly suggesting these enriched quercetin glycosides were the workhorses behind the observed boost.

The critical test came next: could the fermented extract actually protect meat proteins? The team extracted myofibrillar proteins from fresh pork loin and subjected them to a Fenton-like oxidation system containing iron chloride, ascorbic acid, and hydrogen peroxide for 24 hours at 4 degrees Celsius. Oxidation roughly doubled protein carbonyl content, from 1.89 to 3.70 nmol/mg, and slashed sulfhydryl levels from 46.8 to 25.9 nmol/mg, confirming severe damage. Adding the six-day fermented extract at just 10 percent by volume reduced carbonyls to 2.20 nmol/mg, statistically indistinguishable from the unoxidized blank, while the unfermented extract at the same dose left carbonyls significantly higher at 2.50 nmol/mg. Fermentation, in other words, measurably upgraded the protective capacity of the herb.

The structural evidence was equally compelling. Fenton oxidation collapsed protein solubility from 93.2 percent to 43.8 percent, but the fermented extract restored solubility in a dose-dependent manner, reaching 90.1 percent at 25 percent addition, a level that actually surpassed the 1 mg/mL Trolox positive control. Surface hydrophobicity, a marker of protein unfolding, rose with oxidation but fell back below even native levels with high-dose fermented extract, likely because hydrophilic sugar moieties on the enriched quercetin glycosides interact with the protein surface and shield hydrophobic patches. Intrinsic tryptophan fluorescence, quenched dramatically by oxidation, was recovered to near-native levels at the highest dose, and second-derivative UV spectroscopy showed the fermented extract, unlike the unfermented one, significantly preserved the tertiary structure around aromatic residues. Zeta potential measurements completed the picture: oxidation shifted the protein surface charge from minus 12.4 toward minus 7.0 millivolts, but the fermented extract restored it to minus 12.3 millivolts, restoring the electrostatic repulsion that keeps proteins dispersed.

One intriguing wrinkle deserves attention. Rather than restoring depleted sulfhydryl groups, the fermented extract further lowered detectable thiols in a dose-dependent manner. The authors propose a fascinating explanation: the catechol-containing phenolics enriched by fermentation can be oxidized to reactive ortho-quinones under Fenton conditions, and these electrophiles are known to form covalent Michael addition adducts with cysteine thiol groups. Such phenolic-protein interactions, rather than being purely destructive, may help anchor protective phenolic hydroxyl groups onto the protein surface, contributing to the observed charge recovery and electrostatic stabilization. The authors caution that this mechanism remains circumstantial and that the specific glycosidases involved and the exact covalent adducts still require direct verification. Organic acids and other microbial metabolites may also contribute. The team now plans to evaluate the fermented broth in real meat products during processing and storage, assessing stability, sensory impact, and scalability. If those trials succeed, a pungent herb once dismissed for its smell, tamed by a bacterium from fermented meat, could become a genuinely clean-label answer to one of the meat industry’s oldest chemical problems.

Subject of Research: Lactic acid bacterial fermentation of Houttuynia cordata to enhance polyphenols and protect myofibrillar proteins from oxidation

Article Title: Lactiplantibacillus paraplantarum L-ZS9 fermentation enriches Houttuynia cordata polyphenols and protects myofibrillar proteins from oxidative damage

Article References: Yang, L., Zhang, L., Bi, J., Li, Y., Li, Y., Zhang, D., Su, C., Li, H., & Rao, Y. (2026). Lactiplantibacillus paraplantarum L-ZS9 fermentation enriches Houttuynia cordata polyphenols and protects myofibrillar proteins from oxidative damage. Food Chemistry: X, Article 104570. https://doi.org/10.1016/j.fochx.2026.104570

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104570

Keywords: Lactiplantibacillus paraplantarum, Houttuynia cordata, lactic acid fermentation, myofibrillar proteins, protein oxidation, polyphenols, quercetin glycosides, antioxidant activity, clean-label food additives, meat quality, Fenton oxidation, food chemistry