Fermentation temperature acts as a key environmental selective factor driving microbial adaptation and succession during spontaneous sourdough back-slopping; however, its role in sourdough microecosystem stability and bread quality remains unclear. In this study, spontaneous sourdoughs were subjected to repeated back-slopping at different temperatures, and temperature-mediated microbial succession and its effects on bread properties were investigated. Sub-low temperature (8 °C) delayed sourdough maturation but promoted the establishment of temperature-specific microbial communities. After repeated back-slopping, Latilactobacillus replaced Pediococcus as the dominant bacterial genus and Wickerhamomyces was enriched in 8 °C-adapted sourdough, whereas Pediococcus remained dominant in 15 °C- and 25 °C-adapted sourdough. Bread incorporating 8 °C-adapted sourdough (bread T8 group) exhibited improved gluten network structure, the highest specific volume (4.45 mL/g), and sensory score (88.71). Conversely, bread T15 group optimized protein hydrolysis, showing the highest in vitro protein digestibility (85.39%) and total free amino acids (561.07 mg/kg). GC-MS identified 48 volatile compounds, and odor activity value analysis revealed that bread T8 group specifically accumulated key aroma-active esters and aldehydes ((E)-2-nonenal, methyl laurate, and ethyl dodecanoate), which positively correlated with Latilactobacillus and Wickerhamomyces. Overall, repeated temperature selection shaped microbial adaptation and succession, which correlated with differences in sourdough functionality and bread quality.

Escherichia coli O157:H7 is frequently associated with foodborne disease outbreaks, particularly on raw lettuce. In this study, an endolysin LysGPK derived from a Geobacillus phage was identified, recombinantly expressed and characterized for lysing Escherichia coli O157:H7. The antibacterial activity of LysGPK was evaluated using a turbidity reduction assay. Biochemical characterization revealed that LysGPK exhibited optimal lytic activity at 37 °C and pH 9.0, and remained stable at 70 °C. Notably, LysGPK exhibited broad-spectrum lytic activity against Gram-negative pathogens. Furthermore, the combination of LysGPK (15 μg/mL) and EDTA achieved a significant 99.4% reduction in Escherichia coli O157:H7 on lettuce within 2 h using the plate counting method. The subsequent molecular dynamics simulation displayed that the binding of LysGPK to peptidoglycan was stabilized by hydrogen bonds and hydrophobic interactions, and the MM-PBSA calculation showed residues Arg62 and Arg132 contributed critical binding energy. This study provides a theoretical basis for applying endolysins in controlling foodborne pathogens, and offers molecular insights into interactions between the LysGPK and bacterial peptidoglycan.

The recent approval of Weissella species as food ingredients in Korea has expanded their application as probiotic starter cultures for kimchi fermentation. Here, we evaluated Weissella cibaria EFEL4000 as a novel kimchi starter culture by investigating its fermentation performance, probiotic characteristics, safety, and anti-inflammatory activity. As a starter culture, EFEL4000 reached approximately 9.0 log CFU/mL and pH 4.42 after 3 days of fermentation at 10 °C and produced major sulfur-containing volatiles, including dimethyl disulfide, methyl propyl disulfide, and methyl 2-propenyl disulfide. As a probiotic, EFEL4000 showed a favorable strain-level safety profile and tolerance to acidic and bile salt conditions. It also exhibited a Caco-2 adhesion rate of 33% and maintained TEER at 74.11% following H2O2-induced oxidative stress. Additionally, heat-killed EFEL4000 significantly reduced NO production and the expression of IL-6, IL-1β, iNOS, and COX-2 in LPS-stimulated RAW 264.7 macrophages. Genome analysis identified genes associated with environmental adaptation and probiotic functions, while no biogenic amine-producing, acquired antibiotic resistance, or virulence genes were detected. Collectively, these findings provide scientific evidence for the industrial application of W. cibaria as a probiotic starter culture and support the development of functional plant-based fermented foods.

Four pre-chilling treatments, including flake ice-water mixture (FI), slurry ice (SI), chlorine dioxide-slurry ice (40 mg/L chlorine dioxide, SI-40), and ozone-slurry ice (3 mg/L ozone, SI-3), were comparatively evaluated for their effects on chicken meat quality, microbiota composition, and protein properties during refrigerated storage. SI exhibited superior chilling efficiency to FI. Among all treatments, SI-40 most effectively suppressed microbial proliferation, extending shelf life by approximately 3 days, likely due to rapid chilling combined with the chlorine dioxide-mediated antimicrobial activity. Notably, the relative abundance of Pseudomonas in SI-3 was significantly lower than that in the other treatments on day 12. Protein analyses further revealed that SI-40 more effectively inhibited cathepsin B and cathepsin B + L activities than SI-3, indicating better maintenance of muscle protein stability. Antimicrobial-coupled slurry ice therefore integrated rapid pre-chilling with microbial intervention. SI-40 provided the strongest comprehensive preservation effect, whereas SI-3 showed a distinct advantage in limiting late-stage Pseudomonas abundance, providing a basis for the targeted application of antimicrobial-coupled slurry ice in poultry preservation.

Psychrotrophic bacteria are important targets for microbiological control of cold-chain raw milk because of their ability to produce spoilage-associated enzymes and harbor antimicrobial resistance determinants. However, these two functional characteristics are often investigated separately, leaving their co-occurrence within the same isolate collection insufficiently characterized. To address this gap, 543 dominant culturable bacterial isolates were recovered from raw milk and dairy-farm-associated samples collected from six regions of China. Pseudomonas was the most prevalent genus, followed by Acinetobacter, Chryseobacterium, Serratia, and Psychrobacter. β-Lactam minimum inhibitory concentration (MIC) distributions varied substantially among genera, with Chryseobacterium exhibiting the highest proportion of β-lactamase-associated phenotype-positive isolates. Among 226 raw-milk isolates screened using ceftazidime, all 44 isolates showing inhibitor-responsive or carbapenemase-associated phenotypes were subjected to whole-genome sequencing. Genome-wide resistance-gene screening identified annotated β-lactamase genes predominantly in Chryseobacterium and Acinetobacter. In contrast, the resistance-related annotations retained in phenotype-positive Pseudomonas isolates were mainly associated with efflux systems and adaptive response mechanisms. In the integrated analysis, a relative MIC-enzyme risk-prioritization score ranked Serratia (0.88), Chryseobacterium (0.71), and Pseudomonas (0.67) as the three highest-ranking genera. The β-lactam MIC score showed weak positive correlations with the protease-associated score (Spearman's ρ = 0.297, P < 0.001) and with the lipase-associated score (ρ = 0.194, P < 0.001). These findings indicate that genus abundance alone is insufficient to characterize the functional heterogeneity of psychrotrophic bacteria in raw milk. Integrating β-lactam MIC patterns with spoilage-enzyme activity provides complementary information for genus-level prioritization within this isolate collection.

Psychrotrophic Gram-negative bacteria and thermophilic or thermoduric spore-forming bacteria are major causes of quality deterioration and shelf-life loss in milk and dairy products. Their impact is mediated primarily by heat-stable peptidases and lipases, heat-resistant spores and biofilms that persist in processing environments; with the notable exception of toxigenic members of the Bacillus cereus group, most of these organisms are spoilage rather than safety hazards. Culture-based methods remain the reference standard but are slow and provide single time-point information. This review critically evaluates molecular and sensing methods, including quantitative and digital PCR, isothermal amplification, CRISPR-Cas assays, biosensors, viability assays, spectroscopy and (meta)omics, with explicit attention to whether performance has been demonstrated in naturally contaminated dairy matrices rather than in buffer or spiked samples, and examines how their outputs could inform predictive microbiology and machine learning. A classification framework based on growth-temperature range, heat resistance, sporulation and dairy relevance is provided, and each detection-to-prediction pathway is graded as demonstrated in dairy, partly demonstrated or proposed. Validated dairy applications currently exist mainly for individual detection assays and for mechanistic or stochastic shelf-life models driven by culture-derived data; prospective, validated workflows in which molecular or biosensor outputs are used as model inputs to forecast spoilage have not yet been reported. Because heat-stable enzymes can spoil ultra-high-temperature products after the producing cells have been inactivated, viable counts alone are insufficient predictors of such defects. Key barriers include matrix effects, small and imbalanced datasets, domain shift, limited external validation and poorly defined intended use. Detection-to-prediction integration should therefore be regarded as an emerging research direction rather than an established industrial capability.

Listeria monocytogenes control in fresh-cut produce facilities relies heavily on hygienic zoning and physical segregation of processing lines, under the assumption that these measures limit microbiological connectivity between compartments. However, whether segregated lines within a single facility harbour genomically independent populations has rarely been assessed at high resolution. Here, we investigated the population structure and connectivity of L. monocytogenes in two physically and hygienically segregated processing lines (cut-fruit and cut-lettuce) within a fresh-cut produce facility. Whole-genome sequencing was performed on 100 isolates collected during three pre-sanitation sampling campaigns. High-quality assemblies were analysed using multilocus sequence typing (MLST), core-genome MLST (cgMLST) and high-quality SNP (hqSNP) approaches. The population was dominated by two sequence types, ST155 (81%) and ST6 (19%), both displaying limited intra-lineage diversity consistent with closely related clonal populations (median pairwise distances of 7 and 3 hqSNPs, respectively). Hygienic compartmentalization did not result in genomically independent populations. Within each lineage, processing line accounted for only a minor fraction of genomic variation, between-line genetic distances were comparable to within-line distances, and near-identical genotypes were recovered from both lines. Two distinct connectivity patterns were identified: a facility-wide distribution of ST155, with boot-washer and floor isolates clustering across lines, consistent with transfer through personnel-associated routes; and a more localized, drain-associated connectivity pattern involving ST6. Collectively, these findings demonstrate that hygienic segregation alone does not ensure microbiological independence between processing lines and highlight the importance of monitoring hygienic transition zones, including footwear sanitation stations, mobile equipment and drainage systems.

Maize is frequently contaminated by toxigenic fungi, mainly Fusarium spp. and Aspergillus section Flavi, which produce several mycotoxins and represent a major food safety concern. Climate change is expected to alter fungal communities in agroecosystems, potentially modifying mycotoxin contamination patterns in maize. This study compared mycotoxin accumulation and the mycotoxigenic potential of fungal isolates between two consecutive growing seasons characterized by contrasting climatic conditions (2023, hot/dry, and 2024, mild/wet) in maize samples from northeastern Spain. From 2023 and 2024, 853 Fusarium and 210 Aspergillus section Flavi isolates were obtained, and their toxigenic potential for fumonisins (FBs), deoxynivalenol (DON), zearalenone (ZEN), T-2/HT-2 toxins and aflatoxins (AFs) was assessed by HPLC, while natural mycotoxin contamination in the maize samples was determined by ELISA. The 2024 season was associated with a higher proportion of mycotoxin-producing Fusarium isolates and higher DON and ZEN concentrations in maize. DON and ZEN concentrations were positively correlated in both growing seasons, and the relative proportion of F. graminearum was positively associated with DON. F. proliferatum showed high FBs-producing potential in both seasons and a higher proportion of FBs-producing isolates than F. verticillioides. Although AFs-producing isolates were more frequent in 2023, AFs remained below the analytical LOD in all maize samples from both seasons. A. parasiticus showed high AF-producing capacity in vitro despite its relatively low occurrence. No significant correlations were detected between the in vitro toxigenic potential of fungal isolates and mycotoxin concentrations in the corresponding maize samples. Overall, contrasting seasonal conditions were associated with differences in toxigenic fungal profiles and mycotoxin accumulation in maize.

Postharvest disease caused by Alternaria alternata is one of the most serious threats to the production and food security of tomatoes worldwide. Biocontrol microbes offer a sustainable alternative of chemical fungicides without residuals on agricultural commodities. However, there is a lack of effective biocontrol microbes for the tomato Alternaria rot disease currently. In this study, we demonstrated that Bacillus velezensis P15 can efficiently inhibit A. alternata both in vitro and in vivo, with a prominently higher biocontrol efficacy on tomato than chemical fungicides mancozeb. The metabolites of B. velezensis P15 suppressed the growth of A. alternata on PDA in a concentration-dependent manner, causing hyphal surface deformation, ultrastructural damage to hyphal cells, and earlier sporulation. Multiple biosynthetic gene clusters responsible for producing various antimicrobial compounds, such as macrolactin H, difficidin, fengycin, bacillibactin and bacilysin, were further identified in the genome of strain P15. Beyond disease suppression, B. velezensis P15 also improved fruit quality and alleviated membrane lipid peroxidation of tomato by modulating its antioxidant enzyme activities during postharvest, showing superior performance than mancozeb. Time-series transcriptomic analysis revealed that both strain P15 and mancozeb can suppress disease development through the regulation of host transcriptional reprogramming. However, in addition to the host immune responses commonly induced by these two treatments, strain P15 specifically activated a set of plant defense-related genes, including antioxidant enzymes, phytohormones, and pathogenesis-related proteins. Collectively, these findings highlight B. velezensis P15 as a promising biocontrol agent against postharvest Alternaria rot in tomatoes.

Antagonistic microorganisms have gained increasing attention for the control of postharvest fruit diseases due to their safety, low risk of resistance development, and capacity to activate the fruit's innate defense responses. This study evaluated the biocontrol potential of Bullera alba against major tomato diseases and elucidated its underlying mechanisms. The results indicated that Bullera alba effectively reduced disease incidence and lesion diameter in tomato fruit and significantly inhibited conidial germination and germ tube elongation of Alternaria alternata and Botrytis cinerea. The strain rapidly colonized tomato wounds at both 25 °C and 4 °C. Furthermore, Bullera alba markedly enhanced the activities of POD, PPO, SOD, APX, and PAL, and significantly promoted the accumulation of total phenolics, flavonoids, and lignin. Transcriptomic analysis indicated that Bullera alba treatment significantly upregulated the expression of genes associated with several defense-related pathways in tomato fruit, including plant-pathogen interactions, phenylpropanoid biosynthesis, glutathione metabolism, and MAPK signaling pathways. Overall, these findings suggest that Bullera alba may enhance tomato fruit resistance to two major postharvest diseases by inducing the expression of defense-related genes, increasing defense enzyme activities, and promoting the accumulation of bioactive secondary metabolites.

This study characterizes genomic and phenotypic diversity among Vibrio parahaemolyticus isolates from shrimp aquaculture by integrating phenotypic profiling with pangenomic analysis. Growth was quantified under temperature, salinity, pH, nutrient, and ethanol stresses, together with motility, hemolysis, and antibiotic susceptibility. Core-genome SNP phylogeny identified genetic differentiation among representative isolates, with relationships not consistently associated with sampling location, sample type, or collection period. Comparative genomics showed variation in accessory-gene composition and isolate-specific distributions of blaCARB variants, whereas tet(34), tet(35), and several motility and chemotaxis genes were widely detected. Phenotypic assays showed isolate- and condition-dependent variation in stress responses, antimicrobial susceptibility, hemolysis, and motility. Neither tdh nor trh was detected by PCR or WGS-based screening. No clear hemolytic zones were observed on Columbia blood agar, whereas Ref, 1P1, 2P2, 3P2, 4 M2, and 4 N2 showed hemolytic activity on Wagatsuma agar, indicating a discrepancy between the Wagatsuma phenotype and molecular detection of these hemolysin genes. Isolate 1 M1 exhibited greater swimming without increased swarming activity. Accessory-genome principal coordinate analysis (PCoA) and phenotypic principal component analysis (PCA) showed partial correspondence between genomic and phenotypic relationships. The close accessory-genome relationship between 2P2 and 3P2 was not reflected phenotypically, whereas 2Q2 and 3P2 showed phenotypic similarity despite greater accessory-genome separation. These contrasting relationships indicate that accessory-gene presence and absence alone did not reproduce the measured multivariate phenotypic patterns. These findings indicate that genomic and phenotypic variation represent related but non-equivalent dimensions of strain-level diversity and support integrated genomic and phenotypic characterization for surveillance in aquaculture environments.

Nisin, a natural food preservative, is widely used in food preservation. However, it has the drawbacks of poor antibacterial stability and a restricted antibacterial spectrum. In this study, a novel ternary complex of nisin/rhamnolipid (RL)/chitosan (CS) was prepared based on complex coacervation, and its antibacterial effects and application in beef preservation was investigated. The results demonstrate that the particle size of the complex was uniform within 682.5-744.4 nm under pH 2-8. At the pHopt (pH = 4.5), the encapsulation efficiency of nisin was 59.17% and the loading capacity was 29.58%. In vitro release experiments showed the total release rate of nisin in nisin/RL/CS complex was only 39.55% with good sustained release performance at 72 h. Moreover, nisin/RL/CS complex had broad-spectrum antibacterial activity, where the minimum inhibitory concentration and minimum bactericidal concentration against Staphylococcus aureus were 0.5 and 1.0 mg/mL, and 0.25 and 0.5 mg/mL for Escherichia coli, respectively. Further beef preservation experiments confirmed that the nisin/RL/CS complex could effectively inhibit the proliferation of microorganisms on the surface of beef, reduce the rate of lipid oxidation, maintain pH and texture quality, and significantly prolong the shelf life. This study provides theoretical and technical support for the development of nisin-based broad-spectrum antibacterial preservatives and their application in food preservation.

Fermentation stagnation is a persistent challenge in industrial vinegar production, yet its microecological mechanisms remain poorly understood. Here, we conducted time series metagenomic analysis comparing normal and stagnant solid state acetic acid fermentation, and propose the Multilayer Defense Collapse (MDC) model as a unifying framework for stagnation mechanisms. We found fermentation stagnation represents a cascade collapse of four progressive defense layers rather than a single-point failure. Initial perturbation first impairs community network resilience, manifesting as community simplification with 72% reduction in species richness and synchronous viral diversity loss. Metabolic monopoly by core functional bacteria then weakens through triple reconfiguration involving genus-level replacement of lactic acid bacteria, reduced acetic acid bacterial contribution, and limited pathogen penetration. The acidic chemical barrier subsequently fails through both barrier weakening and quorum sensing-mediated active breakthrough. Antimicrobial peptide defense ultimately collapses with 74% AMP decline, driving the system into a dysregulated steady state characterized by broad-spectrum resistome expansion. Random forest analysis further showed that rare taxa, rather than core bacteria, serve as the most sensitive early warning indicators, with 20-50 sentinel species sufficient for high-accuracy classification. This study provides a mechanistic framework for fermentation stagnation and offers potential early-warning indicator and graded intervention.

In this study, we developed a Bayesian two-component Gaussian mixture model to predict coagulase-positive staphylococci (CPS) concentration in raw milk and estimate the posterior probability of belonging to latent low- and high-CPS contamination levels. Thirty-six raw milk samples were collected through longitudinal sampling of 12 raw-milk batches. CPS, total bacterial count (TBC), lactic acid bacteria (LAB), somatic cell count, milk protein and fat content, and pH were quantified, while rainfall, environmental temperature and relative humidity were recorded. Missing TBC and LAB values were handled using Bayesian model-based imputation. The final pH-extended mixture model showed satisfactory Markov Chain Monte Carlo diagnostics and adequately reproduced the observed non-unimodal CPS distribution, with 93.3% of posterior predictive replicated datasets showing at least two modes. TBC was positively associated with the low-CPS contamination component, whereas pH showed a weak negative tendency in the high-CPS contamination component. Prior sensitivity analysis indicated that the main findings were stable under stronger prior regularization. These findings demonstrate the feasibility of a regularized Bayesian mixture modelling framework for probabilistic CPS prediction in raw milk. External validation is required before the model can be considered operational. To support its application, a user-friendly web-based prediction tool has been developed.

Vibrio vulnificus is a highly virulent foodborne pathogen that threatens seafood safety. Although cold-chain storage is widely used for transportation and preservation, V. vulnificus can enter a viable but nonculturable (VBNC) state under low-temperature stress, while retaining pathogenicity and posing a hidden public health risk. Consequently, robust and safe antibacterial preservation strategies are urgently required. Lycopene, a tomato-derived lipophilic carotenoid, is a promising functional food additive; however, its potential to control V. vulnificus during seafood preservation remains unexplored. Therefore, this study aims to investigate the antibacterial efficacy and mode of action of lycopene against V. vulnificus, evaluating its potential as a natural seafood preservative. Here, lycopene exhibited potent antibacterial and bactericidal activity against V. vulnificus, with a minimum inhibitory concentration of 7.45 μM and a minimum bactericidal concentration of 14.9 μM. Mechanistic investigations showed that lycopene induced intracellular accumulation of reactive oxygen species (superoxide and hydroxyl radicals) and decreased GSH/GSSG ratio, thereby disrupting cellular redox homeostasis. This imbalance caused membrane depolarization, increased permeability, iron accumulation, and lipid peroxidation. Furthermore, pretreatment with N-acetyl-L-cysteine and ferrostatin-1 markedly attenuated these effects, indicating that lycopene triggers redox imbalance-mediated ferroptosis-like cell death. Collectively, these findings indicate the in vitro bactericidal activity of lycopene against V. vulnificus via ferroptosis-like cell death, supporting its potential as a natural seafood preservative.