Abstract The quality and shelf life of beef are significantly affected by storage conditions, emphasizing the urgent need for efficient preservation technologies to extend shelf life and reduce waste. This study evaluated the quality changes of beef during storage under low-voltage electrostatic field (LVEF) combined with partial freezing and under partial freezing alone. The results showed that during the 0–24 day storage period, LVEF effectively slowed the increases in pH, thiobarbituric acid reactive substances, total volatile basic nitrogen, and total viable counts, maintained tenderness, and delayed protein oxidation. Metabolomics analysis indicated that LVEF inhibited the formation of off-flavors such as hexanal, 1-octen-3-ol, and benzaldehyde, and reduced the loss of inosine monophosphate, thereby preserving beef quality. Metabolic network analysis further revealed that these metabolites were closely associated with the metabolism of fatty acids, phenylalanine, and proline. These findings suggest that LVEF can effectively enhance the storage quality of beef. Keywords: Beef, Low-voltage electrostatic field, Metabolomics, Flavor quality Graphical abstract [33]Unlabelled Image [34]Open in a new tab Highlights * • Low-voltage electrostatic field can extend the shelf life of beef. * • Using metabolomics to reveal quality changes in beef during storage. * • The main metabolic pathway during beef storage is arginine biosynthesis. * • Octanal and benzaldehyde are closely associated with beef quality. 1. Introduction Beef is high in protein, low in fat, and tasty, and is popular with consumers. However, beef is susceptible to microbial contamination during transportation, storage and marketing, leading to quality deterioration and meat spoilage ([35]Casaburi et al., 2015). Proteins will decompose under the action of microorganisms and enzymes to produce metabolites such as trimethylamine and sulfide with a spoilage odor ([36]Lin et al., 2022). Moderate fat oxidation helps to produce volatile compounds with characteristic aroma. However, excessive lipid oxidation produces odors such as rancidity. With the extension of storage time, metabolites continue to accumulate and further generate off-flavor volatile compounds, which significantly deteriorate beef quality. Studies have shown that compounds such as benzaldehyde, octanal, and nonanal are closely associated with meat spoilage ([37]Huang et al., 2022). According to statistics, the global waste rate of fresh meat products is about 20–30 %. Therefore, developing efficient preservation technologies to delay beef quality deterioration is of significant practical importance for reducing resource waste and ensuring meat quality. Freezing is a common preservation method that can significantly extend the shelf life of meat; however, ice crystals formed during the freezing process can damage muscle tissue, resulting in moisture loss and tissue softening, which in turn lead to a decline in product quality ([38]Tan et al., 2021). During frozen storage, intrusion of ice crystals into surimi tissues and mechanical damage can lead to thawing losses, affecting the oxidation and aggregation of surimi proteins, resulting in increased hardness and reduced palatability of surimi ([39]Yang, Jiang, et al., 2024). In addition, frozen storage may reduce the tenderness of pork, increase total exudate, and decrease the volatile flavor compounds of hand-pulled lamb([40]Luo et al., 2023; [41]Medić et al., 2018). Due to the limitations of conventional freezing storage, partial freezing storage—which involves storing samples at temperatures 1–2 °C below their biological freezing point—has attracted increasing attention for its ability to better preserve the original quality of food. [42]Cao et al. (2023) studied four low-temperature storage methods for beef and found that the shelf life of partial freezing beef was approximately 18 days. [43]Liu et al. (2013) found that partial freezing storage can slow down the changes in pH, total volatile basic nitrogen (TVB-N), and TCA-soluble peptides in fish fillets. [44]Qiu et al. (2020) found that partial freezing storage can delay protein oxidation and inhibit the activity of tissue proteases. However, partial freezing storage alone still faces the problem of a relatively short shelf life, which limits its widespread application. Low-voltage electrostatic field (LVEF) is an emerging non-thermal preservation technology characterized by low energy consumption, environmental friendliness, and operational safety, and has thus attracted widespread attention. This technology can resonate with water molecules inside cells to interfere with the nucleation and growth of ice crystals, affect biochemical reactions, and inhibit microbial growth, thereby extending the shelf life of food ([45]Jha et al., 2017). Currently, LVEF is mainly used for defrosting meat, e.g. beef, pork, etc. However, LVEF has also been used for meat storage and preservation. [46]Zhang et al. (2023) found that LVEF combined with partial freezing could inhibit the increase of TVB-N and total viable counts (TVC) and extend the shelf-life by 3 d. [47]Wu et al. (2024) investigated the effects of LVEF-assisted freezing at different temperatures on the physicochemical properties of pork and found that LVEF-assisted freezing could inhibit ice crystal formation, maintain the integrity of the muscle microstructure, and reduce protein denaturation caused by freezing. [48]Yang, Wu, et al. (2024) found that LVEF synergized at −12 °C could effectively inhibit the rise of carbonyl content and surface hydrophobicity, maintain the integrity of protein secondary and tertiary structures, and reduce the cross-linking aggregation of proteins. At present, research on the application of LVEF in meat preservation remains limited. Existing studies have primarily focused on its effects on the physicochemical properties of meat, with a lack of systematic analysis of metabolic changes in beef during storage. Therefore, this study employed a metabolomics approach to systematically evaluate the quality changes in beef under LVEF combined with partial freezing storage, with the aim of extending the shelf life of beef and reducing resource waste. This work not only helps to broaden market strategies and distribution channels for beef products, but also provides theoretical and technical support for the practical application of LVEF technology in meat preservation. 2. Materials and methods 2.1. Sample preparation In this study, six Guizhou native yellow cattle with similar body conditions, approximately 2 years of age and weighing around 300 kg, were selected from the Guizhou Yellow Cattle Industry Group Co., Ltd. To ensure consistency across experimental samples, all samples were taken from the hind leg muscles. The cattle were slaughtered at a certified slaughterhouse following standard operating procedures. Immediately after slaughter, the beef was placed in insulated containers with ice and transported to the laboratory. Under sterile conditions, the samples were cut into small pieces weighing 500 ± 50 g and sealed in sterile polyethylene bags. Subsequently, all samples were randomly divided into two groups, with 18 samples in each group. LVEF combined with partial freezing group (LP group): Samples were stored in a − 5 ± 1 °C cold storage equipped with a LVEF device (Model: DENBA+, Discharge plate model: DP-10, Waterproof discharge cloth model: DP-2, AGUA business corporation, Japan). In addition, multiple electrode plates were effectively installed around the cold storage room. The electrostatic field generator produces 2500 V of output voltage, 220 V of alternating voltage, 0.2 mA of current, and 50 Hz of frequency. The LVEF generator was installed outside the cold storage chamber, while the electrode plates were mounted in an opposing configuration on the interior side walls. Upon activation of the power supply, an electrostatic field environment was established between the electrodes. Partial freezing group (PS group): Samples were stored in cold storage at −5 ± 1 °C. On days 0, 6, 12, 18, and 24, samples were randomly collected from each experimental group. After the samples were chopped, visible connective tissue and fat were removed as much as possible, and the meat was then ground using a meat grinder. Relevant physicochemical parameters were measured subsequently. For each measurement, three portions of beef from the LP and PS groups were taken, and the average value was used as the representative measurement for that sample. 2.2. Determination of pH The pH was determined following the method described by [49]Cao et al. (2022), with minor modifications. A total of 5.0 g of beef was mixed with 45 mL of physiological saline and homogenized for 1 min, after which the pH was measured using a pH meter (PHS-3C, Shanghai Youping Scientific Instrument Co., Ltd.). The pH meter was calibrated with phosphate buffers of pH 4.00 and 6.86 prior to measurement. 2.3. Determination of TVB-N TVB-N was determined using a volatile basic nitrogen rapid detection kit (Xiamen Standao Scientific Instruments Co., Ltd.), following the same method as described in the study by [50]Wang et al. (2025). 2.4. Determination of TBARS TBARS was determined following the method described by [51]Jiang, Li, et al. (2024), with minor modifications. Briefly, 5.0 g of beef sample was homogenized with 25 mL of trichloroacetic acid solution, thoroughly mixed, and then incubated in a thermostatic shaker at 50 °C for 30 min. Then the mixture was filtered through qualitative filter paper. Subsequently, 5 mL of the filtrate was transferred into a test tube and mixed with 5 mL of thiobarbituric acid aqueous solution. The tube was sealed with a stopper and incubated in a water bath at 90 °C for 30 min. After cooling to room temperature, the absorbance was measured at 532 nm using a microplate reader (Multiskan SkyHigh, Thermo Fisher Scientific, USA). 2.5. Determination of total viable counts (TVC) TVC was determined following the method described by [52]Jiang, Liu, et al. (2024), with minor modifications. Briefly, 2.5 g of beef sample was aseptically weighed and mixed with 22.5 mL of sterile physiological saline (0.85 %), followed by homogenization for 1 min. The homogenate was then subjected to a series of tenfold serial dilutions using sterile saline solution. An aliquot of 1 mL from the appropriate dilution was inoculated onto plate count agar and incubated at 37 ± 1 °C for 48 ± 2 h in a constant-temperature and humidity-controlled incubator. TVC was expressed as lgCFU/g. 2.6. Determination of shear force Shear force was determined following the method described by [53]Cao et al. (2022), with minor modifications. The beef samples were cut into small cubes of 1 cm × 1 cm × 1 cm and measured along the vertical direction using a digital muscle tenderness tester (C-LM3B, Beijing Tianxiang Feidu Instrument Co., Ltd.). 2.7. Determination of total sulfhydryl and carbonyl The total sulfhydryl content was determined following the method described by [54]Lin et al. (2024) with minor modifications. A total of 10.0 g of beef was weighed and added to 40 mL of 0.1 mol/L sodium chloride solution, homogenized for 1 min (5000 g), and then centrifuged at freezing temperature for 15 min (5000 g, 4 °C). The supernatant was discarded, and the process was repeated three times. After the third centrifugation cycle, the supernatant was discarded, the mixture was homogenized with sodium chloride solution, and filtered through four layers of gauze. The precipitate obtained after centrifugation was the myofibrillar protein. It was then diluted with a phosphate buffer solution (pH = 6.5), stored at −80 °C, and set aside. After pipetting 0.5 mL of the protein solution, 2.5 mL of Tris-Glycine-8 M urea solution and 20 μL of Ellman's reagent were added. The mixture was thoroughly mixed and incubated at room temperature for 1 h. It was then centrifuged at 1000 g for 15 min, and the supernatant was collected. The absorbance was measured at 412 nm. Carbonyl content was determined using the instructions of Protein Carbonyl Content Assay Kit (Soleberg Biotechnology Co., Ltd.). 2.8. Determination of volatile flavor compounds Volatile flavor compounds were analyzed by a gas chromatography–mass spectrometry (GC–MS) system (Trace1300-TSQ8000, Thermo Fisher Scientific, USA). The volatile flavor compounds was determined following the method described by [55]Wang et al. (2025) with minor modifications. A total of 3.00 g of beef was weighed and mixed with 2 mL of saturated sodium chloride solution and 20 μL of the internal standard methyl octanoate (30 μg/mL). The mixture was then transferred into a 20 mL headspace vial, which was sealed with a polytetrafluoroethylene-silica gel septum. The sample was equilibrated at 60 °C for 20 min. Subsequently, the automatic extraction head coated with fibers was inserted into the solid-phase microextraction device. The device was then placed into the headspace vial and shaken at 60 °C for 40 min to extract and adsorb the volatile flavor compounds. The fibers were immediately transferred to the gas chromatography injection port, where the adsorbed compounds were thermally desorbed and analyzed at 250 °C for 5 min. Gas chromatography conditions: A CP-WAX 57 CB capillary column (50 m × 0.25 mm × 0.2 μm, Agilent Technologies, USA) was used. Helium was employed as the carrier gas at a flow rate of 1.0 mL/min in splitless mode. The chromatographic temperature program was initiated at 40 °C and held for 2 min, then ramped at 5 °C/min to 120 °C, followed by a 3-min hold. The temperature was then ramped at 5 °C/min to 180 °C and held for 3 min. Mass spectrometry was performed in electron ionization (EI) mode at an ionization energy of 70 eV. The ion source and transfer line temperatures were set to 250 °C and 230 °C, respectively. The full scan mass range was from 30 to 550 m/z. The obtained mass spectra were compared with the National Institute of Standards and Technology (NIST) mass spectral database, and only compounds with a matching score ≥ 700 were retained for further analysis. Quantitative analysis: 20 μL of 30 mg/mL methyl octanoate solution was used as the internal standard. Volatile compound content is calculated using the following formula: [MATH: Ci=Cis×Ai ×VAis×m :MATH] Where C[i] is the mass concentration of any volatile flavor compound (μg/kg); A[i] is the chromatographic peak area of a volatile flavor compound; A[is], C[is], and V represent the peak area of the internal standard, its concentration (μg/mL), and volume (μL), respectively; and m is the sample mass (g). 2.9. Non-volatile metabolites A total of 100 mg of ground beef sample was weighed and transferred into an EP tube, and then 500 μL of aqueous methanol (80 %) was added. The EP tube was vortexed, shaken, and then placed on ice for 5 min, followed by centrifugation for 20 min (1500 g, 4 °C). The supernatant was collected, diluted to 53 % methanol, and centrifuged again for 20 min (1500 g, 4 °C). The supernatant was then injected into the Liquid Chromatography Mass Spectrometry/Mass Spectrometry (LC-MS/MS) system for analysis. The liquid chromatography conditions were described by [56]Fu et al. (2024) with slight modifications. Liquid chromatographic conditions: The Hypersil Gold C18 column (Agilent Technologies, Santa Clara, CA, USA) was maintained at 40 °C. The flow rate was 0.2 mL/min, the mobile phase A was 0.1 % formic acid, and the mobile phase B was methanol. The chromatographic gradient elution program was as follows: 2 % B, 0 min–1.5 min; 2 %–85 % B, 1.5 min–3 min; 85 %–100 % B, 3 min–10 min; 100 %–2 % B, 10 min–12 min. Mass spectrometry conditions: scanning range from 100 to 1500 m/z, spray voltage of 3.5 kV, sheath gas flow rate of 35 L/min, auxiliary gas flow rate of 10 L/min, ion transfer tube temperature of 320 °C, RF level for ion introduction set to 60, auxiliary gas heater temperature of 350 °C, polarity set to both positive and negative modes, and MS/MS performed in data-dependent acquisition mode. 2.10. Statistical analysis Experimental data were performed in three independent experiments and were expressed as mean ± standard deviation. Data were analyzed using SPSS Statistics 26 (IBM Corporation, Armonk, NY, USA), and significance was determined using one-way analysis of variance (ANOVA) and Duncan-Watson test or two-tailed test. Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) were performed using SIMCA 14.1 software (Umetricus AB, Sweden). Heatmaps were generated using TBtools, and volcano plots were created via the online platform Bioinformatics ([57]https://www.bioinformatics.com.cn). Pathway enrichment analysis and bubble plots were performed using Metabo Analyst 6.0 ([58]https://www.metaboanalyst.ca). All other graphical visualizations were generated using Origin 2021 (Origin Lab, Northampton, MA, USA). 3. Results and discussion 3.1. Changes in pH, TVB-N, TBARS, and TVC during beef storage Changes in pH can reflect the process of muscle tissue breakdown. The pH of fresh meat (FM) was 5.35, as shown in [59]Fig. 1a. During storage, the pH of beef tended to decrease and then increase. After beef slaughter, anaerobic metabolism of glycogen breakdown in muscle produces lactic acid, which lead to a lower pH in early beef ([60]Lin et al., 2022). And microorganisms in beef break down proteins into alkaline substances such as ammonia and amino acids, which raises pH. At 24 d of storage, the pH of LP and PS groups increased to 5.63 and 5.73, respectively. Notably, the pH of LP group was significantly lower than that of PS group (p < 0.05), suggesting that LVEF effectively inhibited pH elevation during beef storage, consistent with the findings of [61]Ko et al. (2016). Fig. 1. [62]Fig. 1 [63]Open in a new tab Changes in pH (a), TVB-N (b), TBARS (c), and TVC (d) of beef during storage. Different capital letters (A–B) indicate significant differences among storage methods at the same storage time, while different lowercase letters (a–e) indicate significant differences among storage times under the same storage condition (p < 0.05), as below. TVB-N refers to the microbial decomposition of proteins into ammonia or amines in the case of spoilage of animal food, which reflects the freshness of the food to a certain extent. The Chinese National Food Safety Standard (GB 2707–2016) specifies a limit of 15 mg/100 g TVB-N for fresh beef ([64]Fu et al., 2024). The TVB-N value of FM was 2.33 mg/100 g as shown in [65]Fig. 1b. The TVB-N values of beef gradually increased with storage time. This was caused by microorganisms and protein hydrolyzing enzymes breaking down the proteins in meat and the gradual accumulation of alkaline nitrogenous substances ([66]Alirezalu et al., 2021). Notably, LP group rose more slowly and had a significantly lower TVB-N value of 10.37 mg/100 g at 24 d of storage than PS group (p < 0.05). The ozone generated by the electrostatic field inhibited the growth of microorganisms and the activity of enzymes, thereby suppressing the production of TVB-N ([67]Papachristodoulou et al., 2018). Lipid oxidation generates aldehydes and ketones, which are associated with unpleasant, sour odors and can diminish beef quality. As shown in [68]Fig. 1c, the TBARS value for FM was approximately 0.16 mg/kg. Throughout the storage period, the TBARS values of all beef samples exhibited an upward trend, with LP group consistently showing significantly lower TBARS levels compared to PS group (p < 0.05). Increased oxidation of lipids and proteins in meat with longer storage time, leading to increased levels of TBARS. At 24 d of storage, the TBARS values for LP and PS groups were 0.35 mg/kg and 0.5 mg/kg, respectively. This difference can be attributed to the electrostatic induction effect of LVEF, which charges the food surface, reducing its contact with oxygen and thereby slowing lipid oxidation ([69]Ko et al., 2016). TVC is a direct and effective indicator of meat freshness and shelf life. [70]Chen et al. (2019) considered 6.00 lgCFU/g as the upper acceptable limit for fresh beef. As shown in [71]Fig. 1d, the TVC values of beef increased with storage time. This was due to the fact that meat samples are rich in protein, which can be used as a medium for microbial reproduction. The initial TVC in beef was 3.96 lgCFU/g. At 18 d of storage, the TVC of LP group and PS groups were 4.87 and 6.07 lgCFU/g, respectively, and those of PS group exceeded the upper limit for fresh meat. At 24 d of storage, the TVC of LP group was 5.92 lgCFU/g, still within the fresh meat range. [72]Zhang et al. (2024) reported that high-voltage electric fields can slow the physicochemical and microbial growth rates in fish during storage, effectively preserving its original flavor and nutritional value. Overall, LVEF inhibited the rate of increase in pH, TVB-N, TBARS and TVC during storage, thereby effectively extending the shelf life of beef. 3.2. Changes in shear force during beef storage The tenderness of meat plays an important role in the consumer's perception of its palatability ([73]Lin et al., 2022). As shown in [74]Fig. 2, the shear force of the beef continued to decrease as the storage time increased. The shear force of FM was 65.72 N. At 24 d of storage, the shear force of LP and PS groups decreased by 45.80 % and 63.99 %, respectively, and the shear force of LP group beef was significantly higher than that of PS group (p < 0.05). LVEF accelerates the freezing rate, resulting in smaller and more uniform ice crystals, which minimizes damage to muscle tissue and effectively preserves the shear force of beef ([75]Xie et al., 2021). The results showed that LVEF storage delayed beef quality deterioration, thus maintaining the texture of fresh meat. Fig. 2. [76]Fig. 2 [77]Open in a new tab Variation of shear force during storage of beef. 3.3. Changes in total sulfhydryl and carbonyl during beef storage Total sulfhydryl is one of the most functional groups in proteins and can not only affect the enzymatic activity within the protein but is also closely linked to protein denaturation. As shown in [78]Fig. 3a, the total sulfhydryl content of myofibrillar proteins of FM was 104.79 μmol/g. The total sulfhydryl content of beef decreased significantly with increasing storage time. The initial carbonyl content in FM was 0.66 μmol/g. After 24 days of storage, the carbonyl content increased to 2.32 μmol/g in the LP group and 3.35 μmol/g in the PS group, with the LP group showing a significantly lower level than the PS group (p < 0.05). Previous studies have shown that LVEF can effectively inhibit sulfhydryl oxidation ([79]Yang et al., 2023). The formation of carbonyl derivatives and disulfide bonds can cause intramolecular and intermolecular cross-linking of proteins, which can adversely affect beef quality. As shown in [80]Fig. 3b, Carbonyl content in beef increased with storage time, indicating increased oxidation of beef myofibrillar proteins. The carbonyl content of FM was 0.66 μmol/g, and after 24 days of storage, the carbonyl content of beef from the LP and PS groups was 2.32 and 3.35 μmol/g, respectively. An electric field can accelerate the freezing rate of water and promote the formation of small ice crystals, thereby reducing damage to muscle fibers and inhibiting the denaturation of myofibrillar proteins. Meanwhile, because small ice crystals cause less damage to muscle tissue and cellular structures, they can delay the release of lysosomal enzymes and pro-oxidative factors within the cells, further suppressing protein oxidation. Therefore, LVEF can effectively inhibit the oxidation of beef myofibrillar proteins and extend their shelf life. Fig. 3. [81]Fig. 3 [82]Open in a new tab Changes in total sulfhydryl (a) and carbonyl content (b) of beef during storage. 3.4. Analysis of volatile flavor compounds during storage of beef 3.4.1. Analysis of the composition and content of volatile flavor compounds in beef During the storage period of beef samples, a total of 92 volatile flavor compounds were identified, and their chromatograms are shown in Fig. S1. These compounds included 22 alcohols, 22 aldehydes, 13 alkanes, 15 esters, 8 ketones, 3 acids, and 9 others. Among them, 74 compounds were identified in LP group, and 70 compounds were identified in PS group. These volatile flavor compounds are generated during storage through processes such as protein degradation, lipid and carbohydrate metabolism, and microbial decomposition. As shown in [83]Fig. 4a and b, a total of 40 volatile flavor compounds were detected in the FM, with a total content of 854.88 μg/kg. Among them, aldehydes and alcohols accounted for 47.21 % and 27.70 %, respectively, indicating that these two types of compounds play a significant role in the flavor of beef. This result is similar to the findings reported by [84]Ba et al. (2012). From 0 to 24 d, the total content of volatile flavor compounds in beef showed a trend of increasing and then decreasing. The total volatile flavor compounds reached a maximum of 3540.28 μg/kg at 18 d for LP group and 3021.51 μg/kg at 12 d for PS group. At 24 d of storage, the total content of volatile flavor compounds in the LP group and the PS group was 2767.30 μg/kg and 1754.80 μg/kg, respectively, with the LP group showing a significantly higher content than the PS group. This difference may be attributed to the effect of the low-voltage electrostatic field (LVEF) on the transmembrane potential and electron transfer of muscle cell membranes during beef oxidation, which in turn altered the formation pathways and rates of volatile flavor compounds, resulting in differences in their content between the two groups ([85]Jiang, Li, et al., 2024). Fig. 4. [86]Fig. 4 [87]Open in a new tab Changes in volatile flavor compounds during beef storage. a: Species; b: Total content. 3.4.2. Analysis of differential volatile flavor compounds PCA was used to analyze the differences in volatile flavor compound content. As shown in [88]Fig. 5a, the first two principal components (PC1 and PC2) account for 20.1 % and 14.9 % of the total variance, respectively. In the score plot, the nine samples show a clear separation, with the LP group clustered in the positive direction of PC1 and the PS group in the negative direction, suggesting that LVEF may have a significant influence on the volatile flavor compounds of beef. To further investigate the impact of LVEF on beef flavor, OPLS-DA was used to analyze volatile flavor compounds, and the reliability of the model was verified through 200 permutation tests, as shown in [89]Fig. 5b and c. Based on a VIP > 1 and p < 0.05 in the OPLS-DA analysis, 21 differential volatile flavor compounds were identified in the beef ([90]Fig. 5d). Odor activity value (OAV) is an important indicator used to evaluate the contribution of volatile compounds to the overall flavor. It is defined as the ratio of a compound's concentration in the sample to its odor threshold. When OAV > 1, it indicates that the concentration of the compound exceeds the human sensory threshold and may contribute perceptibly to the overall aroma of the sample. Therefore, we calculated the odor activity values (OAVs) of 21 differential volatile compounds and identified 14 key ones (OAV > 1). Fig. 5. [91]Fig. 5 [92]Open in a new tab Analysis of volatile flavor compounds during beef storage. a: PCA; b: OPLS-DA; c: substitution test; d: Thermogram of significantly different volatile flavor compounds. In this study, five significantly different alcohol compounds were screened from beef samples, including 1-octen-3-ol, 1-hexanol, 1-heptanol, 1-octanol, and trans-2-octen-1-ol. 1-Octen-3-ol is formed through lipid oxidation and imparts an undesirable mushroom-like odor to beef. During storage, the content of 1-octen-3-ol showed a trend of initially increasing and then decreasing. In FM, the concentration of 1-octen-3-ol was 120.14 μg/kg. After 12 days of storage, the concentration of 1-octen-3-ol in the PS group reached a maximum of 636.26 μg/kg (OAV: 636.26), which was significantly higher than that observed in the LP group (408.76 μg/kg, p < 0.05). These results indicate that LVEF may exert a delaying effect on the formation of this volatile compound. By day 24 of storage, the concentrations of 1-octen-3-ol in the LP and PS groups were 410.22 μg/kg and 109.58 μg/kg, respectively, with a significant decline observed in the PS group. This reduction may be associated with the gradual depletion of metabolic precursors within the beef matrix during storage. In contrast, the application of LVEF combined with micro-freezing storage effectively delayed the formation and accumulation of 1-octen-3-ol, resulting in significantly higher levels in the LP group at the later stage of storage compared to the PS group. 1-Hexanol has a grassy or fruity aroma and a low sensory threshold, which contributes to enhancing the fresh aroma of beef. In FM, the content of 1-hexanol was 24.15 μg/kg (OAV: 3.02). After 24 days of storage, the concentration of 1-hexanol in the LP group was 98.31 μg/kg (OAV: 12.29), which was significantly higher than that in the PS group at 27.81 μg/kg (OAV: 3.28). Furthermore, studies have shown that 1-hexanol exhibits antioxidant and antibacterial properties, which may help delay the decline in beef quality ([93]Lu et al., 2018). The 9 aldehyde differentiators were screened in beef and most of them originated from lipid oxidation in meat, such as octanal, hexanal and nonanal. High concentrations of octanal emit a pungent odor and are considered a marker of lipid oxidation ([94]Huang et al., 2023). The octanal content in FM was 39.68 μg/kg (OAV:56.67), and after 18 d of storage, the octanal level in PS group beef reached 369.08 μg/kg (OAV: 527.26), significantly higher than that in LP group (133.65 μg/kg), which may result in the generation of off-flavors. Hexanal is high in beef and produces fruity and fatty flavors and is the main volatile flavor compound in beef. During storage, the hexanal content showed an increasing and then decreasing trend. In LP group, hexanal increased from 115.24 μg/kg (0 d) to 965.24 μg/kg (18 d) and then decreased to 820.97 μg/kg (24 d, OAV:182.44). In PS group, hexanal increased from 115.24 μg/kg (0 d) to 527.36 μg/kg (6 d) and then decreased to 254.48 μg/kg (24 d, OAV,56.55). At 6 d of storage, the LP group hexanal content was 267.13 μg/kg, which was significantly lower than that of PS group (p < 0.05). Therefore, we speculate that LVEF treatment may affect the formation and transformation of hexanal. With the increase in storage time, the degree of lipid oxidation increases, leading to higher levels of nonanal ([95]Fan et al., 2023). At 24 d of storage, the nonanal content in LP group was 261.75 μg/kg, significantly lower than in PS group (370.26 μg/kg, p < 0.05). Benzaldehyde produces an unpleasant odor and can affect the aroma of beef. Throughout the storage period, benzaldehyde was detected only in the later storage stages (18–24 d) in the PS group, which could be related to the decline in beef quality. After 24 days of storage, the benzaldehyde content in the PS group was 19.12 μg/kg (OAV: 4.78). [96]Yu et al. (2025) found that benzaldehyde was associated with quality deterioration during frozen storage of meat. Esters are generally formed either through the reaction of acids produced by the degradation of fats or proteins with alcohols, or via transesterification between fatty acids in triglycerides and ethanol. They contribute desirable fruity aromas to meat products. In beef, there were 2 esters of significantly different substances. Ethyl esters can impart creamy, fruity, floral, and sweet aromas to beef. In FM, Caproic acid vinyl ester was detected at 29.44 μg/kg (OAV: 29.43). After storage, it was only detected on day 6 in the LP group at 93.82 μg/kg (OAV: 93.82). (E)-9-Octadecenoic acid ethyl ester was detected only in the LP group after 12 days of storage, with a concentration of 59.06 μg/kg. Ketone volatile flavor compounds are products of lipid oxidation and can interact with myofibrillar proteins in meat products, affecting the flavor of the meat. Acetoin and 2,3-octanedione have a butterscotch flavor. In the present study, LP group detected a higher total content of 3-hydroxy-2-butanone and 2,3-octanedione than PS group. In the LP group, the content of 2-pentylfuran increased with the prolongation of storage time, reaching 46.37 μg/kg (OAV: 7.73) after 24 days. However, in PS group, 2-pentylfuran showed an increasing and then decreasing trend from 10.65 μg/kg (FM) to 87.99 μg/kg (12 d) and then to 10.55 μg/kg (24 d, OAV: 1.76). 2-pentylfuran has a low threshold, exhibits a fruity flavor and contributes to roasted meat aroma, and is a typical oxidative breakdown product of linoleic acid. In summary, LVEF storage can increase certain volatile flavor compounds during beef storage and reduce the formation of off-flavors. 3.5. Analysis of nonvolatile metabolites during beef storage 3.5.1. Overview of non-volatile metabolomics analysis To investigate the changes in non-volatile flavor compounds during beef storage, the composition of metabolites was identified by LC-MS/MS, and the chromatograms are shown in Fig. S2. Retaining the metabolites that were full match in mzCloud and removing the metabolites that were not queried in the chemistry book ultimately resulted in the screening of 313 effective non-volatile metabolites, including 56 amino acids and their derivatives, 31 nucleosides, nucleotides, and analogs, 20 organic acids and their derivatives, 99 lipids and lipid-like molecules, 28 organic oxygen compounds, 20 organic heterocyclic compounds, 38 aromatic compounds, and 21 other metabolites ([97]Fig. 6a). To further compare the differences between the two groups of samples, the samples were analyzed using OPLS-DA, and differential metabolites (fold change (FC) > 2 or < 0.5, p < 0.05) were visualized as volcano plots, as shown in [98]Figs. 6b and [99]7. In the volcano plot, gray dots represent metabolites with no significant difference, while red and blue dots indicate metabolites that are significantly upregulated and downregulated, respectively. The results show that LP and PS groups were completely separated at each time point, indicating that LVEF treatment can cause differences in the levels of certain metabolites. In the comparisons of LP-6d vs. PS-6d, LP-12d vs. PS-12d, LP-18d vs. PS-18d, and LP-24d vs. PS-24d, 72, 8, 17, and 70 metabolites were significantly upregulated, while 27, 17, 17, and 51 metabolites were downregulated, respectively. At 24 d of storage, LP and PS groups exhibited the richest differential metabolites. Fig. 6. [100]Fig. 6 [101]Open in a new tab Proportion of non-volatile metabolite types (a) and OPLS-DA analysis (b). Fig. 7. [102]Fig. 7 [103]Open in a new tab Volcano plot of non-volatile metabolites. 3.5.2. Non-volatile differential metabolite analysis To identify representative metabolites, differential metabolites were further selected based on p < 0.05, FC > 2 or < 0.5, and VIP > 1 predicted by the OPLS-DA model. Most of the differential metabolites between LP and PS groups during storage were amino acids and their derivatives, lipids and lipid-like molecules. Non-volatile compounds in fresh meat, such as amino acids and their derivatives, are crucial for flavor. Proteins are easily degraded into peptide chains and amino acids during storage due to microbial action. At 24 d of storage, the number of significantly differential metabolites in the amino acids and their derivatives category increased. In LP group, the abundance of many amino acids and their derivatives, such as argininosuccinic acid, L-cysteine-glutathione gisulfide, and l-serine, was significantly lower than in PS group (p < 0.05). Higher levels of argininosuccinic acid can cause meat to have a bitter flavor ([104]Ramalingam et al., 2019). The abundance of L-hydroxyproline and L-asparagine showed an increasing trend with increasing storage time, and they may be potential markers of beef deterioration during storage. Furthermore, throughout the entire storage period, the carnosine content in the LP group remained consistently higher. At 24 d of storage, the carnosine content in the LP group was 12.65 %, significantly higher than the 3.83 % in the PS group (p < 0.05). Carnosine is widely present in the muscles of most animals and can inhibit lipid oxidation through a combination of free radical scavenging and metal chelation. Therefore, carnosine helps slow down lipid oxidation during beef storage ([105]Peiretti et al., 2012). Nucleosides and nucleotides are essential in many physiological processes. After 6 d of storage, the content of guanosine monophosphate (GMP) in LP group was significantly higher than in PS group. GMP has a distinctive mushroom flavor that interacts with disodium sarcosinate to bring out freshness and thus add flavor to beef. Adenosine triphosphate in beef is degraded to adenosine diphosphate, inosinic acid (IMP), and hypoxanthine by enzymes. During storage, N2-methylguanosine and xanthosine tended to increase, and IMP tended to decrease, but the IMP of LP group decreased more slowly. IMP is the main flavor component of meat, providing sweetness and pleasant flavor, unstable in muscle, can be further degraded to hypoxanthine and xanthine, associated with meat spoilage ([106]Yi et al., 2024). During storage, lipids degrade into aldehydes, ketones, lower fatty acids, and other compounds. In beef, a total of 29 differential lipid and lipid-like molecules were screened. Acetylcarnitine content in beef increased with storage time. At 24 d of storage, the LP and PS groups acetylcarnitine contents were 11.39 % and 29.78 %, respectively. [107]Jia et al. (2021) found that the accumulation of long-chain fatty acyl carnitines promotes β-oxidation of fatty acids, leading to poor meat quality. During the spoilage process of meat, changes in organic acids and their derivatives, benzene and its derivatives, as well as heterocyclic compounds, have a relatively minor impact on meat quality ([108]Fu et al., 2024). 5-hydroxyindole belongs to organic heterocyclic compounds, and the content of 5-hydroxyindole in beef increased significantly after 24 d of storage, and the 5-hydroxyindole in LP group was significantly lower than that in PS group (p < 0.05). Proteins are broken down by microorganisms into indoles, phenol, putrescine, and various nitrogenous acids and fatty acids ([109]Liu et al., 2022). This may be due to the fact that LVEF better inhibits microbial growth and reproduction, thus reducing the production of 5-hydroxyindole. 3.6. Metabolic pathway and metabolic network analysis To better visualize the intrinsic relationships among metabolites, the detected metabolites were subjected to enrichment and pathway analysis using Metabo Analyst 6.0. As shown in [110]Fig. 8a, a total of 58 metabolic pathways were enriched. Among them, pathways such as arginine biosynthesis, alanine-aspartate-glutamate metabolism, the pentose phosphate pathway, and the tricarboxylic acid (TCA) cycle exhibited high enrichment ratios and strong statistical significance. Furthermore, based on pathway impact values (PI>0.1) and statistical significance (p < 0.05), eight key metabolic pathways were identified ([111]Fig. 8b), including arginine biosynthesis, the pentose phosphate pathway, and the TCA cycle. These pathways are closely associated with quality changes in beef during storage. Fig. 8. [112]Fig. 8 [113]Open in a new tab Analysis of metabolic pathways during beef storage. a: Enrichment analysis plot; b: Metabolic pathway bubble plot; c: Metabolic network plot. To gain deeper insights into the quality changes of beef during storage, the metabolic network involved in flavor formation was predicted based on the KEGG database and relevant literature, as shown in [114]Fig. 8c. Proteins in beef are metabolized primarily by alanine, aspartic acid, and glutamic acid to produce intermediate metabolites that are then involved in the formation of other flavor compounds. Hydrolysis of proteins by endogenous and microbial enzymes produces peptides and free amino acids, which then produce alcohols and aldehydes by transaminases and α-ketoglutaric acid enzymes. For example, L-hydroxyproline can be generated through the metabolic pathway of proline. As storage time increases, the concentration of L-hydroxyproline in beef gradually accumulates. However, the L-hydroxyproline content in samples treated with LP group was significantly lower than PS group. LVEF treatment reduced proline metabolism, a result consistent with the findings of [115]Fei et al. (2024). L-glutamate, l-serine and d-glutamine are converted to pyruvic acid, α-ketoglutaric acid, oxaloacetic acid, etc., and then further converted to phosphoenolpyruvic acid, and finally glucose is synthesized via the gluconeogenesis pathway. Their contents changed dynamically during storage, but at the end of storage, LP group had lower L-glutamate, l-serine and d-glutamine contents than PS group. This suggests that LVEF treatment may slow down the degradation of proteins and the metabolism of amino acids, thereby helping to delay the deterioration of beef quality. Carbohydrates go through the glycolytic pathway to produce pyruvate, which in the production of acetyl-CoA, then enters the TCA cycle. Citric acid and fumaric acid are critical intermediates in the TCA cycle that promote lactic acid decomposition and have anti-aging, appetite and fatigue-relieving effects. At the end of storage, the citric and fumaric acid contents of LP group were significantly lower than those of PS group. This suggests that LVEF treatment may have reduced acetyl-CoA production, thereby affecting the TCA cycle. Malic acid is produced through the TCA cycle and subsequently converted into aspartic acid. Aspartic acid can then be converted into oxaloacetic acid through the metabolic pathways of alanine, aspartic acid, and glutamic acid, and further participate in the synthesis of various other metabolites. Lipids are hydrolyzed by lipases to produce free fatty acids, which are then further degraded into various metabolites. Unsaturated fatty acids can be converted into flavor compounds such as heptanol, octanol, (E,E)-2,4-nonenal, and 2-pentylfuran by the action of enzymes. The LVEF treatment increased the content of these flavor compounds resulting in better beef flavor during storage. Lipid oxidation can promote the breakdown of palmitic acid in meat, resulting in the formation of capric acid. In addition, oxidation of fatty acids to produce acetyl-CoA can further participate in fatty acid metabolism and contribute to the formation of other flavor compounds. For example, 1-octanol and 3-octanone can be produced by the catalytic action of acetyl-CoA carboxylase, β-ketoacyl synthase, and malonyl transferase ([116]Jiang et al., 2025). Whereas acetyl-CoA produces α-ketoglutarate via the TCA cycle, it then undergoes the mangiferolic acid pathway to produce phenylalanine, which is finally metabolized by phenylalanine to produce benzaldehyde. During storage, benzaldehyde was detected only at the late PS group stage, suggesting that LVEF treatment affected phenylalanine metabolism. In summary, during beef storage, intermediate metabolites of proteins, carbohydrates, and lipids can be interconverted through various metabolic pathways, leading to the formation of diverse flavor compounds. However, these metabolic activities are also accompanied by continuous energy release and nutrient depletion, ultimately affecting beef quality. LVEF treatment can partially slow down these metabolic processes and help delay flavor deterioration. The findings not only reveal the complex metabolic network involved in flavor formation during beef storage but also demonstrate that LVEF treatment plays a positive role in preserving beef quality and enhancing flavor during storage. 4. Conclusion This study demonstrated that LVEF treatment effectively slowed the changes in pH, TVB-N, TBARS, TVC, and shear force in beef, and inhibited protein oxidation. Ninety-two volatile flavor compounds were identified using HS-SPME-GC–MS; alcohols and aldehydes were the predominant compounds. LVEF promoted the formation of desirable volatiles such as 1-hexanol, 1-heptanol, and hexanal, while reducing undesirable compounds including 1-octen-3-ol, octanal, and benzaldehyde. Furthermore, 313 non-volatile metabolites were identified via LC-MS/MS. The results showed that LVEF significantly suppressed the increase in the levels of L-hydroxyproline and L-asparagine and mitigated the reduction in IMP during storage. Metabolic pathway analysis indicated that LVEF modulated multiple pathways, slowing the degradation of proteins, lipids, and carbohydrates, thereby enhancing the stability of beef quality. In conclusion, LVEF holds promise for delaying the deterioration of beef quality. Future research should aim to further optimize LVEF processing parameters and improve equipment stability. Additionally, the incorporation of sensory evaluation and microbiome analysis is recommended to systematically assess its impact on meat quality, thereby supporting its practical application in the field of preservation. The following are the supplementary data related to this article. Supplementary Fig. S1. [117]Supplementary Fig. S1 [118]Open in a new tab Chromatograms of GC-MS Supplementary Fig. S2. [119]Supplementary Fig. S2 [120]Open in a new tab Chromatograms of LC-MS/MS CRediT authorship contribution statement Yuxia Liu: Writing – original draft, Formal analysis, Data curation. Wei Su: Writing – review & editing, Funding acquisition. Yingchun Mu: Writing – review & editing, Supervision. Xiaomin Liu: Data curation, Conceptualization. Kangli Yang: Formal analysis. Rongrong Mu: Investigation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments