To establish a multiplex digital polymerase chain reaction (dPCR) quantitative detection system for Cronobacter (Escherichia sakazakii), Staphylococcus aureus, Vibrio cholerae, Vibrio parahaemolyticus, Salmonella, Listeria monocytogenes and pathogenic Escherichia coli O157.

Species-specific genes were screened from the genomes of the 7 kinds of pathogenic bacteria in the National Center for Biotechnology Information (NCBI) genome database, and their conservation was analyzed. Primer-probes were designed based on the conserved regions. The primer specificity and probe structure were analyzed by using MFE Primer (V3.1) to alignment against the human genome, bacterial and viral genomes in the NCBI ftp database, followed by the synthesis of primers probes and pathogenic bacterial plasmids. A multiplex dPCR detection system for foodborne pathogens was established by using multicolor coding technology to encode the detection targets of the 7 kinds of pathogenic bacteria with different fluorescence channel combination modes, and the specificity, accuracy, linearity and repeatability of the system were verified.

The detection system was characterized by high specificity, accuracy and reproducibility, with linear ranges of 15.55–17336.02 copies/μL for Cronobacter (Escherichia sakazakii), 11.92–12755.55 copies/μL for Staphylococcus aureus, 9.47–11811.35 copies/μL for Vibrio cholerae, 9.86–11631.06 copies/μL for Vibrio parahaemolyticus, 12.20–14244.65 copies/μL for Salmonella, 16.70–15329.10 copies/μL for Listeria monocytogenes, and 28.22–26325.95 copies/μL for pathogenic Escherichia coli O157.

The construction of a multiplex dPCR detection system for various foodborne pathogens, utilizing multicolor coding technology, facilitate the simultaneous detection of 7 kinds of foodborne pathogens. This system offer a valuable technical reference for pathogen surveillance, contamination source tracing and emergency response in the fields of food, medicine, and the environment, demonstrating broad application potential. It will facilitate further exploration of a super-multiplex dPCR detection system for pathogens.

To achieve rapid and non-destructive detection of thiobarbituric acid reactive substances(TBARS) content in beef by leveraging deep learning and explainable artificial intelligence-driven visible-near infrared (Vis-NIR) spectroscopy.

Vis-NIR spectra were collected from 420 beef samples, and the TBARS values were determined experimentally. Different spectral pretreatment methods, feature selection methods and modeling algorithms were compared to realize the non-destructive and rapid detection of TBARS in beef.

It demonstrated that the standard normal variate preprocessing-successive projections algorithm for feature selection-long short-term memory network model achieved the best performance [R-square of prediction set(R²_P)=0.8378, root mean square error of prediction set (RMSEP)=0.0679 mg/kg, residual prediction deviation(RPD)=2.4833], outperforming partial least squares regression (PLSR) and extreme gradient boosting(XGBoost). Furthermore, through the interpretability analysis using Shapley additive explanations (SHAP)revealed the contribution mechanisms of critical wavelengths, such as 1080 nm and 956 nm, to TBARS prediction.

The proposed method offers an innovative strategy for rapid, non-destructive, accurate and explainable detection of TBARS in beef.

To analyze the etiology and traceability of foodborne pathogens in two suspected mixed infection food poisoning events in Shaoxing City in 2024.

The re-identified isolates were serotyped by slide agglutination, virulence genes (tlh+/tdh+/trh-) of Vibrio parahaemolyticus were identified by real-time fluorescence polymerase chain reaction, drug resistance was detected by broth microdilution method, and pulsed field gel electrophoresis (PFGE) was used to detect the isolates, and were performed for molecular typing. Acquired drug resistance genes were predicted by ResFinder and multilocus sequence typing (MLST) analysis by PubMLST database. core genome multilocus sequence typing (cgMLST) was used for molecular tracing.

Of the 2 incidents, both 2 isolates of Vibrio parahaemolyticus from incident 1 were serotype O10:K4, with virulence gene profile tlh+/tdh+/trh-, resistant to cefazolin, and exhibited identical PFGE banding patterns, indicating homologous strains. The 2 Salmonella agona isolates were susceptible to all 27 kinds of antibiotics, showed identical PFGE banding patterns, and were identified as homologous strains. The 4 Vibrio parahaemolyticus isolates from incident 2 were all serotype O3:K6, with virulence gene profile tlh+/tdh+/trh- and identical PFGE banding patterns, but exhibited two distinct antimicrobial resistance phenotypes: Cefazolin and ampicillin-cefazolin. The 2 Salmonella ruzizi II isolates showed different antimicrobial resistance phenotypes: Nalidixic‑tetrecycline and ampicillin‑streptomycin‑azithromycin‑trimethoprim/sulfamethoxazole‑gentamicin‑tetrecycline‑chloramphenicol‑florfenicol, with a PFGE band similarity of 51.90%. Whole-genome sequencing results showed that the 2 Vibrio parahaemolyticus isolates from incident 1 belonged to ST3 by MLST, and the β-lactam resistance gene bla_CARB-22 was predicted in each isolate. The Salmonella isolates were identified as ST13, and 4 kinds of antimicrobial resistance genes were predicted in total.The cgMLST clustering results showed high similarity between the isolates, respectively. The 4 Vibrio parahaemolyticus isolates from incident 2 belonged to ST3, and the bla_CARB-22 resistance gene was predicted in each isolate. The 2 Salmonella isolates were identified as ST516 and ST398, with 19 kinds of antimicrobial resistance genes predicted. cgMLST clustering revealed that the Vibrio parahaemolyticus isolates were highly homologous, whereas the Salmonella isolates were highly divergent.

PFGE typing, whole-genome sequencing-based MLST, and cgMLST results demonstrated that incident 1 is a food poisoning case caused by mixed infection with Vibrio parahaemolyticus and Salmonella, whereas incident 2 is attributed to Vibrio parahaemolyticus alone. The etiological characteristics of mixed-infection food poisoning incidents in Shaoxing City provide a theoretical basis for the prevention and control of such mixed infections. It is recommended that relevant departments strengthen food risk supervision capabilities and enhance public food safety awareness.

To understand the molecular characteristics of Salmonella in Qianxinan Buyi and Miao Autonomous Prefecture from 2023 to 2025, and explore the relationship between serotype and sequence typing (ST), drug resistance characteristics and virulence.

The drug susceptibility test was performed by trace broth dilution method, and the serotypes, ST, drug resistance genes and virulence factors of 82 Salmonella strains were analyzed by whole-genome sequencing and bioinformatics analysis techniques, and the correlation was analyzed by combining with phylogenetic tree.

A total of 22 serotypes were identified, with Salmonella enterica (17/82,20.73%) as the predominant type; 24 ST were identified, with ST11 as the predominant type, and serotypes were highly correlated with ST. The total drug resistance rate was 93.90% (77/82), and the multi-drug resistance rate was 74.39% (61/82). The highest rate of drug resistance was to tetracycline (78.05%) and no drug resistance to 4 kinds of drugs including amikacin. The 43 kinds of drug resistance genes were detected, with golS (75.61%), bla_TEM-1(42.68%), AAC(6')-Iaa (39.02%) and QnrS1 (32.93%) as the predominant genes, and the drug resistance phenotypes were largely coincident with the genotypes. A total of 40 virulence factors were detected, among which the carrying rate of virulence factors related to secretion systems was the highest.

Salmonella in Qianxinan Buyi and Miao Autonomous Prefecture shows high multi-drug resistance and complex drug resistance gene profile, and serotypes are closely related to their ST. Continuous monitoring of foodborne Salmonella needs to be strengthened to ensure food safety in China and guide the precise use of drugs in clinical practice.

To evaluate the applicability of the Soleris rapid microbial detection method for detecting molds and yeasts in food and its equivalence with the national standard method.

Validation experiments were performed in accordance with domestic and international standards, focusing on 5 core parameters: Inclusivity, exclusivity, sensitivity, relative limit of detection, robustness and inter-batch variability. A systematic equivalence comparison was conducted against the conventional reference method specified in GB 4789.15—2016 National food safety standard-Microbiological examination of food-Enumeration of molds and yeasts.

All 30 target mold and yeast strains were successfully detected, yielding an inclusivity rate of 100%; while none of the 31 non-target strains were detected, resulting in an exclusivity rate of 100%. The relative limit of detection values across diverse food matrices were consistently below the preset acceptance threshold of 1.5. Robustness and differences between reagent batches had no statistically significant impact on the detection results, demonstrating excellent repeatability. The detection time was shortened to within 48 hours.

The detection results of mold and yeast in food using the Soleris microbial rapid detection method are equivalent to the GB 4789.15—2016 method and can be used for the rapid detection of mold and yeast in food, providing efficient technical support for food safety quality control, supervision and sampling inspection, and emergency screening.

To analyze of the contamination status of pathogenic bacteria in ready-to-eat fruit and vegetable products in Gansu Province in 2024.

According to the principle of random sampling, a total of 241 ready-to-eat fruit and vegetable products were collected from the catering and distribution sectors across the province. According to the series of methods in the GB 4789—2016 National food safety standards-Food microbiological examination, tests were conducted on diarrheagenic Escherichia coli, Listeria monocytogenes, Salmonella and Staphylococcus aureus. The whole genome sequencing was performed on the detected Listeria monocytogenes and Staphylococcus aureus to obtain the multilocus sequence typing results. Drug sensitivity tests were conducted on the diarrheagenic Escherichia coli, Listeria monocytogenes and Salmonella species.

The overall detection rate of pathogenic bacteria in 241 ready-to-eat fruit and vegetable products was 5.81%. Among them: Listeria monocytogenes had the highest detection rate, at 3.73% (9/241); followed by diarrheagenic Escherichia coli and Staphylococcus aureus, with a detection rate of 0.83% (2/241); the detection rate of Salmonella was 0.41% (1/241). The 9 strains of Listeria monocytogenes belonged to 4 different sequence type (ST) types. Among them, 4 strains were ST87,2 strains were ST121, 2 strains were ST288 and 1 strain was ST378. The 2 Staphylococcus aureus strains belong to 2 ST types, one of which was ST1 and the other was ST59. One strain of diarrheagenic Escherichia coli was drug-resistant, and the drug resistance spectrum was ampicillin-amoxicillin/clavulanate-cefazolin-cefotiofur-streptomycin-florfenicol.The 4 strains of Listeria monocytogenes were resistant, with the resistance spectrum being tetracycline-erythromycin.One strain of Salmonella was resistant, and the resistance spectrum was ertapenem-meropenem-streptomycin.

In 2024, the contamination of pathogenic bacteria in ready-to-eat fruit and vegetable products in Gansu Province mainly original from cold dishes. The detection rate of Listeria monocytogenes is relatively high compared to the other 3 kinds of bacteria, which requires close attention.

To establish a method for the simultaneous determination of migration levels of 10 kinds of restricted substances [including bisphenol A, bisphenol S, bisphenol F, phthalic acid, terephthalic acid, isophthalic acid, trimellitic acid, caprolactam, 1,1’-sulfonylbis(4-chlorobenzene) and 4,4’-biphenol] in food plastic packaging products by high performance liquid chromatography.

Following the general principles of migration tests, plastic packaging products were subjected to migration testing using food simulants, the food simulants were separated by gradient elution with a mobile phase of 0.1% phosphoric acid aqueous solution-methanol on a Welch Ultimate AQ-C₁₈ chromatographic column (3.0 mm×100 mm, 3.5 μm), detected by a diode array detector, and quantified by the external standard method.

All 10 kinds of compounds exhibited good linear relationships in the concentration range of 0.10 to 10.00 mg/L (r²>0.999) in various food simulants, including 4% (volume fraction) acetic acid, 10% (volume fraction) ethanol, 50% (volume fraction) ethanol and 95% (volume fraction)ethanol. The limits of detection were in the range of 0.01–0.02 mg/L and the limits of quantitation were 0.05 mg/L.At 3 spiked concentration levels (low, medium and high), the recoveries of the target analytes in different food simulants ranged from 87.8% to 104.0%, with relative standard deviations (n=6) ranging from 0.33% to 7.58%.

This method is simple, efficient and accurate and is suitable for the determination of migration levels of 10 kinds of restricted substances in food plastic packaging products.

To establish a method for the determination of fenpicoxamid and florylpicoxamid residues (2 kinds of new types picolinamide pesticide) in vegetables and fruits by pipette-tip based modified QuEChERS-high performance liquid chromatography-tandem mass spectrometry.

The vegetable and fruit samples were extracted by shaking with acetonitrile assisted with QuEChERS salts, and purified using disposable tips packed with clean-up sorbents. The analytes were separated by a HPLC Shim-pack GIST C₁₈ column (100 mm×2.1 mm, 2 μm) with acetonitrile-0.1% formic acid in water as the mobile phase by gradient elution mode. The target compounds were analyzed by high performance liquid chromatography-tandem mass spectrometry in multiple reaction monitoring mode, and quantified by the external standard method with matrix-matched calibration curves.

The linear relationship between the 2 kinds of pesticides was good within the mass concentration range of 1–40 ng/mL (correlation coefficient r>0.994), and the limits of quantitation were 0.01 mg/kg, could meet the requirements of residue detection. Method validation was conducted in 5 kinds of matrices including Wogan mandarin, celery, apple, scallion and raisin, and the average recovery rates at the spiked levels of 0.01–0.10 mg/kg were 62.6% to 119.1% with the relative standard deviation of 1.98% to 13.10%.

The proposed method is simple, rapid and efficient, and is suitable for the rapid screening and accurate quantitation of 2 kinds of novel picolinamide pesticide residues in vegetables and fruits.