The accumulation mechanisms of aflatoxin B1 (AFB1) and deoxynivalenol (DON) in fungal-infected maize kernels were comparatively investigated using hyperspectral imaging (HSI) and scanning electron microscopy (SEM). The moldy maize kernels were prepared by inoculation with Aspergillus flavus (A. flavus), Fusarium graminearum (F. graminearum), and natural molding (control group). A distinct difference in spatial distribution patterns between the mycotoxins was revealed by the combined HSI and HPLC analysis. AFB1 was predominantly concentrated in the internal tissues (especially in the A. flavus-inoculated group), whereas DON was mainly located on the kernel surface (particularly in the F. graminearum-inoculated group). Intermediate and variable toxin distribution was exhibited in the control group influenced by indigenous fungi. These findings were further supported by SEM analysis. An AFB1 quantification model based on a back-propagation neural network (BPNN) was established with high predictive accuracy (R2v = 0.95, RPD = 4.55), highlighting the significant potential of non-destructive detection methods. Conversely, the predictive accuracy for DON was relatively lower (R2v = 0.80, RPD = 2.22) due to its weaker spectral response and increased spatial heterogeneity. The integration of HSI and SEM provides an effective multi-scale strategy to elucidate the mechanisms of mycotoxin accumulation under different infection scenarios, thereby establishing a crucial foundation for future rapid and non-destructive detection technologies aimed at ensuring grain quality and safety.

Background: In recent years, with the rapid growth of global turtle farming, turtle shell, as a major by-product, has been used in traditional medicine and health foods, gaining increased attention for its nutritional and bioactive value. Scope and approach: This review systematically summarized the classification and global utilization of turtle shells, as well as the regulatory policies on its use. From a "component-activity-product" perspective, its key nutrients, health benefits, and application prospects in relation to health food development were reviewed. Key findings and conclusions: Global research on turtle nutrition and health has shown regional variations, shaped by conservation needs and international regulations. While Western countries like the United States, Brazil, and Australia emphasized the nutritional value of turtle fat, China has focused on the health benefits of turtle shell as a by-product. Moreover, turtle shells contain proteins, amino acids, minerals, fatty acids/esters, steroids, peptides, and polysaccharides, providing anti-inflammatory, antioxidative, immunomodulatory, hematopoietic, and blood pressure- and lipid-regulating effects. Moreover, they can potentially treat osteoporosis, Parkinson's disease, Alzheimer's disease, and liver cancer. They have also been used in products like beverages, turtle jelly, gelatin, starch-based foods, and medicinal wines. However, future research should investigate its health benefits based on composition or determine the active compounds from bioactivities, supporting its development into functional foods and therapeutics.

Decellularization is an effective strategy to eliminate the immunogenicity of biomaterials to the host. The impact of raw material thickness on decellularization process was assessed. Three different thicknesses materials were selected and decellularized using an ultrasound-assisted method combined with tributyl phosphate. The results showed that Alaska pollock skin (0.23±0.06mm) was severely damage structurally after decellularization. Tilapia skin (0.64±0.12mm) exhibited higher decellularization efficiency and a relatively intact collagen fibre structure. Porcine skin (1.27±0.15mm), however, required additional decellularization treatment due to its dense structure. Based on these results, tilapia skin and porcine skin were selected for subsequent performance analysis. The results showed that the porosity (78.38%±4.06%) and swelling ratio (473.16%±15.62%) of tilapia skin were both higher than those of porcine skin. Hydroxyproline content and FTIR spectra confirmed preserved collagen triple-helix structure. Thermal stability increased after decellularization, with tilapia skin denaturation temperature reaching 56.2°C. Cytotoxicity testing revealed that decellularized tilapia skin extracts were non-cytotoxic, while porcine skin extracts showed mild cytotoxicity at higher concentrations. In conclusion, tilapia skin offers better decellularization efficiency, structural integrity, and biosafety under ultrasound-assisted treatment, making it a promising candidate for tissue engineering and wound repair.

Fermented soybean paste, a traditional high-salt condiment, faces challenges in standardization and quality control due to its reliance on natural fermentation. This study systematically evaluated the effects of a defined starter culture, Tetragenococcus halophilus CICC 10286, on soybean paste fermentation by comparing natural fermentation (NF) and fortified fermentation (FF). Compared with NF, FF maintained a higher moisture in the later stage (NF-LS: 50.30%; FF-LS: 60.08%) and lower total acid levels in the middle and later stages (NF-MS: 1.58 g/100 g; FF-MS: 0.96 g/100 g; NF-LS: 2.23 g/100 g; FF-LS: 1.11 g/100 g). Although protein degradation was more pronounced in the FF group at the midpoint (p < 0.0001), the lower accumulation of amino acid nitrogen suggests a potential shift in nitrogen metabolism, possibly toward enhanced transamination or deamination processes. Free amino acid profiling indicated that FF facilitated earlier accumulation of umami and sweet amino acids, but the total free amino acid content in the later stage was lower. Specifically, Glu and Asp reached 724.47 nmol/L and 305.52 nmol/L, respectively, in NF-LS, whereas the corresponding values in FF-LS were 397.16 nmol/L and 275.46 nmol/L. Meanwhile, Pro reached 337.81 nmol/L in FF-MS, indicating earlier accumulation of some amino acids under FF. Notably, the proportion of bitter amino acids in the FF group was reduced in the later stage. Microbial community analysis showed that FF promoted the enrichment of Tetragenococcus and halotolerant bacteria, such as Halomonas, at the midpoint, and increased the relative abundance of the aroma-producing yeast Zygosaccharomyces (NF-MS: 37.73%; FF-MS: 65.11%). Functional prediction based on PICRUSt2 suggested a higher predicted abundance of genes involved in pyruvate metabolism and branched-chain amino acid degradation in the FF group. These findings demonstrate that T. halophilus CICC 10286, as a starter culture, can effectively modulate the fermentation of soybean paste, providing a scientific basis for developing standardized and quality-controlled fermentation processes.

The effects of magnetic field-assisted freezing pretreatments, including static magnetic fields (SMF) and alternating magnetic fields (AMF) with intensities of 0.5, 1.0, 1.5, and 2.0 mT, on the freezing and quality characteristics of freeze-dried kiwifruit slices were investigated. The results showed that the magnetic field-assisted freezing technique has unique advantages in improving the quality of freeze-dried kiwifruit crisps. The microstructure showed that by shortening the freezing time, the finest and uniform pore structure was formed in the 1.5 mT SMF-treated group, and the hardness (729.46+56.37 g) was significantly increased compared with the control group (509.17+45.10 g). The relative antioxidant capacity (RACI) index of 1.5 SMF (0.73) and 1.0 AMF (0.53) treatment groups was higher than control group. The total sensory evaluation scores were elevated in all magnetic field treatment groups, and the electronic nose analysis confirmed the reduced sulfide response value. However, the key challenge remains that the magnetic field treatment did not improve the color and rehydration of the crisps. This study confirms that magnetic field-assisted freezing can optimize the freezing process and product quality of kiwifruit slices, but different parameters need to be balanced for product characteristics to obtain the best overall quality.

To design a novel emulsifier capable of enhancing the bioavailability of curcumin (Cur)-loaded emulsions in the gastrointestinal tract, soy protein-based ternary composite nanoparticles (SEPn) were fabricated by transacylation reaction. The results showed that SEPn was formed by the covalent binding of the carboxyl groups in PGA to the amino groups in SEC through multiple forces. SEPn-1:1 was determined to be the optimal condition for preparing Cur-loaded emulsions. Additionally, SEPn-1:1 had superior emulsifying capacity as formed plastic-state emulsion gel with φ as low as 0.5. Moreover, the rise in oil content promoted the development of gel, thus increasing the apparent viscosity, gel strength, and stability of Cur-loaded emulsions. Furthermore, SEPn-1:1 emulsion exhibited excellent gastric stability and higher free fatty acid (FAA) release rates in the small intestine phase compared with that of SECcon (SEC control sample) and Mixture emulsion, thus leading to the highest bioavailability of Cur (28.57 ± 1.91 %).

The effect of ultrasound and plasma pretreatment on freeze-dried kiwifruit crisps was investigated in this study. Using unpretreated kiwifruit as a control group (CG), the effects of ultrasound (US), plasma-activated water (PAW), ultrasound combined with plasma-activated water (UPAW), plasma-jet (PJ), and ultrasound combined with plasma-jet (UPJ) on the quality of vacuum freeze-dried kiwifruit were investigated. The results showed that all the pretreatments could change the microstructure of the crisps. The microstructure of dried kiwifruit after pretreatment showed more porous structures with different number and size compared to the CG group. The largest pore structure was observed in the UPAW group which had the highest crispness. The activity of water (Aw) of all pretreatment samples was significantly lower than the CG group (P < 0.05). In addition, the UPAW group had the lowest moisture content (4.85 %) and the highest rehydration ratio (288.03 %), indicating the better drying characteristics. Furthermore, the UPAW pretreatment sample showed good appearance with the highest brightness and the lowest color difference (Delta E). The total sugars and total phenolics of the UPAW pretreatment sample were mostly retained, and its flavor was the closest to the CG group. The combination of US and PAW promoted the formation of a larger cavity structure and improved the drying characteristics and physicochemical properties of dried kiwifruit crips. However, all the pretreatments resulted in a decrease in antioxidant capacity, with the least decreasing of the US group and the most decreasing of the UPAW group. Correlation analysis showed that the chlorophyll and vitamin C were the major antioxidants in dried kiwifruit crips. The mechanism of decrease in antioxidant activity of pretreatment, especially UPAW, should be discussed and the effective measure to reduce the change in chlorophyll and vitamin C should be taken in future research.

Glucosamine (GlcN) is a high-value compound with significant health applications. GlcN is widely used in the food and health industry as a food additive or functional food. The development of a green, efficient, and safe method for GlcN production is of great significance due to the complexity of traditional production methods, environmental pollution, and sensitization of raw materials. In this study, Saccharomyces cerevisiae genes PFK1, PDB1, GNA1, ISR1, and PCM1 were knocked out using the Clustered Regularly Interspaced Short Palindromic Repeats Cas9 (CRISPR-Cas9) method. In addition, three key enzyme genes, glucosamine-6-phosphate deaminase GlmD, glucosamine-6-phosphate phosphatase GlmP, and ammonium transporter AMT1, were introduced to construct engineered strains for GlcN synthesis in the presence of high-concentration inorganic ammonium ions. The results indicated that S. cerevisiae HPG5 with GlmD, GlmP, and AMT1 integration and simultaneous deletion of PFK1, PDB1, GNA1, PCM1, and ISR1 achieved the highest GlcN yield (1.95 ± 0.02 g/L) during fermentation with 10 g/L (NH4)2SO4, which was 2.47-fold higher than the control. The conversion rate of glucose to GlcN in HPG5 was 9.75% in liquid YPD medium containing 20 g/L of glucose and 10 g/L of (NH4)2SO4. Thus, the results indicated that S. cerevisiae HPG5 could effectively produce GlcN in the presence of high-concentration ammonium sulphate. This study provides a promising alternative, S. cerevisiae HPG5, for GlcN production.

Biogenic amines (BAs) are critical indicators of spoilage in aquatic products, but conventional detection methods are destructive and inefficient. This study proposes a nondestructive approach combining near-infrared (NIR) spectroscopy with deep learning to predict BAs and classify quality grades in Chinese mitten crabs. Six BAs were identified during cold storage, among which putrescine, cadaverine, histamine, and tryptamine were excellently predicted by a hybrid convolutional neural network-long short-term memory-squeeze-and-excitation (CNN-LSTM-SE) model (R2 > 0.84, RMSE <2.0). Based on a quality index, crabs were graded as fresh, acceptable, and spoiled, with those stored >24 h post-mortem at 8 ± 1 °C deemed unfit for consumption. The CNN-LSTM-SE model achieved 100% accuracy in identifying spoiled crabs, outperforming convolutional neural networks, long short-term memory, and traditional machine learning models. This study demonstrates the potential of NIR spectroscopy and deep learning for assessing crab freshness, providing valuable tools for intelligent monitoring of aquatic products.

The whiteness (W), total volatile basic nitrogen (TVB-N) and total viable count (TVC) are three important indicators to evaluate the freshness of crayfish. This work aimed to develop robust models for non-destructive detecting multiple freshness indicators of crayfish during cold storage based on near-infrared (NIR) spectroscopy. After collecting NIR spectra of crayfish, the W, TVB-N and TVC were measured by traditional methods. Partial least squares regression (PLSR) models along with multiple preprocessing and variable selection methods and one-dimensional convolutional neural network (1D-CNN) models based on raw spectra were constructed. The results demonstrated that the wavelet threshold denoising (WTD) and competitive adaptive reweighted sampling (CARS) could improve the performance of PLSR models. Compared with other models, the 1D-CNN model showed the best performance in predicting TVB-N and TVC, with Rp2 of 0.9397 and 0.9318, and RPD of 2.8279 and 2.7560, respectively, indicating outstanding advantages of CNN in NIR spectral analysis. To sum up, the overall results suggested that NIR spectroscopy combined with deep learning would be a feasible approach for detecting the freshness of crayfish.

Effective hemostasis is a critical step in the treatment of injuries. The chitosan-tortoiseshell collagen composite sponge cross-linked by glutaraldehyde (C-G-TCS) was prepared. The potential of it as a biomedical material was explored through comparative analysis. The results showed that the C-G-TCS had better thermal stability and lower enzymolysis degree compared to pure tortoiseshell collagen sponge (TCS), chitosan-tortoiseshell collagen composite sponge (C-TCS) and chitosan-walleye pollock skin collagen composite sponge cross-linked by glutaraldehyde (C-G-PCS). The cytotoxicity of C-G-TCS and C-G-PCS was evaluated as non-cytotoxic for cell growth. Although the blood absorption capacity of C-G-TCS and C-G-PCS was lower than that of iRE, a commercial collagen sponge, they had a lower blood clotting index and a similar clotting time compared to iRE. The C-G-TCS treatment was observed to demonstrate optimal adhesion of whole blood cells to the sponges. The prothrombin time and activated partial thromboplastin time results indicated that the C-G-TCS and C-G-PCS might promote activation of the extrinsic coagulation pathway. The C-G-TCS and C-G-PCS treatments for the tail amputation mouse exhibited a reduced hemostasis time in comparison to the control group, and there was a similar blood loss amount with iRE treatment.

This study explores the application of near-infrared (NIR) spectroscopy combined with machine learning for the non-destructive detection of aflatoxin in peanuts contaminated by Aspergillus flavus (A. flavus). The key innovation lies in the development of an optimized spectral processing pipeline that effectively overcomes moisture interference while maintaining high sensitivity to low aflatoxin concentrations. NIR spectra were collected from peanut samples at different incubation times within the spectral range of 950 to 1650 nm. Spectral data were preprocessed, and Competitive Adaptive Reweighted Sampling (CARS) selected ten characteristic bands. Correlation analysis was performed to examine the relationships between physicochemical properties, characteristic bands, and aflatoxin content. Three machine learning models—Backpropagation Neural Network (BPNN), Support Vector Machine (SVM), and Random Forest (RF)—were used to predict aflatoxin levels. The SNV-SVM model demonstrated superior performance, achieving calibration metrics (R2C = 0.9945, RMSEC = 9.92, RPDC = 14.59) and prediction metrics (R2P = 0.9528, RMSEP = 19.58, RPDP = 7.01), along with leave-one-out cross-validation (LOOCV) results (R2 = 0.9834, RMSE = 11.20). The results demonstrate that NIR spectroscopy combined with machine learning offers a rapid, non-destructive approach for aflatoxin detection in peanuts, with significant implications for food safety and agricultural quality control.

To make the emulsion more stable under complex environments, a novel emulsifier, soy protein isolates (SPI)-propylene glycol alginate (PGA) binary nanocomposite particles (SPIPn), was synthesized using a transacylation reaction. The findings revealed that SPIPn formed when the carboxyl groups in PGA and the amino groups in SPI were covalently bonded, facilitated by various forces. SPIPn-2:3 was identified as the most suitable emulsifier for the subsequent preparation of curcumin-loaded emulsions and could emulsify emulsions with an φ as low as 0.5 to form emulsion gels. Moreover, an increase in the φ value strengthened the gel network structure, leading to enhanced macroscopic viscosity, elasticity strength, and creaming stability. Furthermore, the emulsion with an φ of 0.6 retained up to 96 % and 99 % of curcumin after storage for 30 days and irradiation for 720 min, respectively. This study could provide a simple and convenient method to improve food product stability.

The nutritional differences between the carapace and plastron of turtle shells, and their grinding challenges, are not well understood. This study aimed to evaluate the nutritional differences between a turtle shell’s carapace and plastron, and the effects of high-pressure steam (HP) and vinegar stir-frying (VS) pretreatments. The carapace showed higher crude protein and fat but lower ash than the plastron. Both pretreatments reduced grinding energy and increased ash content while decreasing protein and amino acids levels. HP resulted in more water-soluble extract (15.13%), protein (17.26%), and peptide (20.77%), whereas total free amino acids (FAAs: 545.62mg/100g) were increased by VS. Nutritional value was comparable between carapace and plastron. Additionally, in contrast to the conventional VS pretreatment approach, the HP method not only enhances the grinding process of turtle shells but also promotes the release of their water-soluble constituents. This advancement sets a solid foundation for future comprehensive studies aimed at exploring the nutritional potential of turtle shells.

Aflatoxin B1 is a highly carcinogenic and teratogenic substance mainly produced by toxin-producing strains such as Aspergillus flavus and Aspergillus parasitic. The efficient decomposition of aflatoxin is an important means to reduce its harm to humans and livestock. In this study, Trametes versicolor aflatoxin B1-degrading enzyme (TV-AFB1D) was recombinantly expressed in Bacillus subtilis (B. subtilis) 168. MMT-CTAB-AFB1D complex was prepared by the immobilization of TV-AFB1D and montmorillonite (MMT) by cross-linking glutaraldehyde. The results indicated that TV-AFB1D could recombinantly express in engineered B. subtilis 168 with a size of approximately 77 kDa. The immobilization efficiency of MMT-CTAB-AFB1D reached 98.63

Curcumin (Cur) and (-)-Epigallocatechin gallate (EGCG) have synergistic bioactivity in several terms, but their polar differences and environmental sensitivity make co-utilization challenging. Soy protein fibrils (UF) have sparked broad attention as outstanding carriers for bioactive substances. In this study, EGCG was applied to remodel the structure of Cur-loaded UF complex (UFC) to form novel UF-based hydrogels co-encapsulated Cur and EGCG (UFCE). The fluorescence spectroscopy, ITC, and FTIR analysis demonstrated that EGCG absorbed and deposited on the surface of UFC through hydrophobic and electrostatic interactions. The binding of EGCG led to the formation of UFCE, which constructed a stable and uniform gel network during heat treatment. UFCE gels formed with varying EGCG/protein concentration ([E]/[F]) ratios showed different structural properties with different encapsulation effects. At 2.5% protein concentration, UFCE gels formed at [E]/[F] ratio of 0.08 (UFCE0.08) exhibited better gelling properties. The encapsulation efficiency of Cur and EGCG in UFCE0.08 gel was up to 97.71% and 91.02%, respectively. Moreover, UFCE0.08 gel exhibited significantly enhanced stability of Cur (95.27%) and EGCG (70.27%) than that of free Cur (53.13%) and free EGCG (13.32%) after thermal treatment at 85 degrees C for 120 min. UFCE0.08 gel was also effective in improving the UV light stability of Cur (88.70%) and EGCG (72.96%). Furthermore, the binding of EGCG had pronouncedly boosted the anticancer activity of Cur in the UFCE0.08 gel against Caco-2 cells. This study provides new perspectives into the fibril-polyphenol hydrogel systems for the co-encapsulation of EGCG and Cur, which has substantial potential for health-promoting applications.

The acid-soluble collagens (ASC) extracted from the skins of walleye pollock and silver carp were compared with ASCs from skins of duck and pig in physicochemical properties. The dry weight extraction rates of ASCs from walleye pollock and silver carp were 14.10% and 10.30%, which were significantly higher than those of pig and duck (4.53% and 3.14%). It was found that the ASCs from skins of pig, walleye pollock and silver carp were all type I collagen with intact triple helix structure, while ASC from duck skin was a mixture of type I and type II collagen.The PI of the four ASCs was in the weakly acidic range (5.49-6.84), but ASCs from two fish skins were less thermally stable than that from pig and duck. The dynamic frequency scanning showed that collagen from porcine skin always maintained elastic behavior, while the other three ASCs changed from viscous behavior to elastic behavior with the increasing of frequency. The collagen from two fish skins had a less dense microstructure. The high extraction rate proved that two fish skins were high-quality raw materials for collagen, and the differences in physicochemical properties allowed fish skin collagen to have different applications than terrestrial animal collagen.

The effect of static and alternating magnetic fields assisted freezing with intensity of 1, 2, and 3 mT on the microstructure and protein properties of channel catfish fillet were investigated. The results showed that the magnetic field treatment shortened the phase transition time of freezing, and significantly reduced the size of the formed ice crystals. The changes of trichloroacetic acid-soluble peptide, Ca2+-ATPase activity, particle size, and Zeta potential, which represented solubility, denaturation and aggregation of protein, indicated that magnetic field treatment could improve the protein stability. The chemical force analysis, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and Fourier transform infrared spectroscopy (FTIR) results proved that the magnetic field could change the structure of protein. Furthermore, these changes had effects on the thermal stability of catfish meat protein which reflected by increasing of the transition temperature and enthalpy. However, the waveform and intensity of magnetic field affected the stability of protein structure.

Arabinoxylan (AX) can form stable covalent bonds with protein to improve gel properties. We aimed to prepare a conjugate between soymilk protein (SMP) and AX by peroxidase, followed by the addition of transglutaminase (TG) to prepare tofu gels. The conjugate's properties and their effects on the mechanical properties, rheological properties, and microstructure of tofu gels were evaluated. Results revealed that the & alpha;-helix content decreased, the & beta;-sheet content increased, and the surface hydrophobicity reduced from 1.60 x 105 to 1.27 x 105. The optimal amount of AX required to improve the properties of tofu gel was 1.0%. The tofu gel showed better hardness (118.44 g), water holding capacity (WHC) (86.17%), and higher storage modulus (G & PRIME;) and loss modulus (G & DPRIME;). Low-Field NMR (LF-NMR) showed that the water was evenly distributed. Scanning electron microscopy (SEM) revealed a denser and more regular three-dimensional gel network.