Lactobacillus plantarum M3 Fermentation Enhances Mulberry Juice Antioxidant Capacity: Metabolomic Analysis

Abstract

Mulberry, a plant highly valued for medicinal–edible features, was fermented with Lactobacillus plantarum M3 to enhance its bioactive profile. This study conducted a comprehensive evaluation of the antioxidant activity of fermented mulberry juice (FMJ) and identified key metabolites through an integrated approach involving non-targeted metabolomics, network pharmacology, RT-qPCR, and molecular docking. Under optimized conditions (28 °C, pH 5.5, 12°Bx initial sugar content, 48 h and 5% inoculum), fermentation significantly bolstered the antioxidant capacity of MJ. Specifically, superoxide dismutase (SOD) activity increased from 62.41 ± 0.11 to 84.99 ± 0.07 U/mL, while total phenolic content (TPC) surged from 1108.98 ± 2.90 to 2494.17 ± 7.05 mg GAE/L; DPPH radical scavenging activities were improved by 63.09%. Non-targeted metabolomics identified 195 secondary metabolites, primarily comprising alkaloids, flavonoids, and phenolic acids. Among these, protocatechuic acid, Albanin A, and apigenin exhibited significant dynamic shifts, indicating that they may play a pivotal role in regulating antioxidant capacity. Integrated network pharmacology, RT-qPCR validation, and molecular docking further elucidated that Albanin A and Moracin Q likely drive these enhanced antioxidant effects by activating the Nrf2 pathway, suppressing the NF-κB pathway, and upregulating SOD1 expression. These findings provide a theoretical basis for the development of high-potency functional mulberry products.

1. Introduction

The mulberry is a large deciduous tree of the genus Morus (Moraceae), cultivated for its fruit, known as mori fructus. White mulberries (Morus alba L.), black mulberries (M. nigra L.), and red mulberries (M. rubra L.) are popular types that have long been consumed in China. Mori fructus, which is listed as a medicinal and edible homologous substance in China, contains numerous valuable components. Mulberries are high in sugars, such as fructose, glucose, and sucrose, with a total sugar content of 12% to 15%, according to recent research [1,2]. In addition to their high nutritional and therapeutic value, these fruits are utilized in various food products, including wine, juice, jam, canned goods, ice cream, and vinegar [3,4]. Additionally, mulberries are naturally rich in bioactive compounds including polyphenols, polysaccharides, anthocyanins, and flavonoids. These compounds contribute to a range of nutritional and health benefits, such as antioxidant [5,6], antibacterial [7], and anti-inflammatory effects [8], along with potential for lowering blood sugar and lipids, reducing blood pressure, combating cancer, protecting the liver [9], managing obesity, and supporting neuroprotection [10].

Phenolic compounds act as potent antioxidants by neutralizing free radicals through mechanisms such as hydrogen atom transfer or single electron transfer [11]. While plant phenolic compounds primarily exist in bound forms—as glycosides, esters, or polymers—which restrict their antioxidant efficacy, microbial fermentation can enhance their bioavailability. By producing specific enzymes such as β-glucosidase, cellulase, and tannase, fermentation releases bound flavonoids, thereby boosting their biological activity [12]. This increase in activity stems from the enhanced bioavailability of free phenolic compounds; specifically, the liberated aglycones possess a markedly higher antioxidant capacity [13,14].

The global functional beverage market is experiencing rapid expansion, propelled by heightened consumer health consciousness and a distinct preference for natural, nutrient-dense products [15]. Probiotics, particularly lactic acid bacteria (LAB), play a crucial role in this sector [16,17]. Lactic acid bacteria (LAB) are widely employed in juice fermentation to enhance nutritional value, extend shelf life, and refine sensory properties. Liquid-state fermentation is particularly advantageous, facilitating rapid biotransformation and metabolite release. Furthermore, LAB fermentation has been shown to boost both the antioxidant activity and bioavailability of phenolic compounds, alongside other bioactive and flavor-enhancing constituents [18]. A positive correlation was observed between prolonged fermentation time and increased phenolic content in the broth. This is exemplified by studies on spontaneous beetroot fermentation, where the process reduced bound phenolic acids while concurrently increasing free phenolic acids-such as sinapine, p-coumaric acid, and isoferulic acid by 2.64-fold [19]. Studies have demonstrated that LAB fermentation significantly enhances the total anthocyanin, phenolic, and flavonoid content in mulberry fruit. Consequently, the improved antioxidant activity of the fermented juice is strongly associated with the increased levels of these phenolic compounds [20]. Studies involving alternative matrices, such as fig or goji berry, suggest that the metabolic pathways utilized by LAB vary significantly depending on the initial sugar-to-acid ratio and the specific phenolic profile of the raw juice. This indicates that the nutritional enhancement achieved through fermentation is not a universal constant but rather a strain-specific metabolic event that remains poorly understood for many high-antioxidant substrates.

Despite extensive investigation into various fruit juices, a critical knowledge gap persists regarding the synergy between Lactobacillus plantarum and the unique flavonoids and phenols matrix of mulberry juice (MJ). Existing studies often focus on “before and after” snapshots of total phenolic content, failing to elucidate the specific shifts in metabolite profiles that occur during the fermentation kinetics of MJ. Furthermore, the correlation between these specific metabolite transformations and the resulting antioxidant capacity has not been sufficiently quantified for this specific pairing. To compensate for these shortcomings, this study aims to prove that the fermentation of plant lactobacilli can not only significantly enhance the antioxidant capacity of mulberry juice by increasing the total phenolic content but also improve antioxidant efficiency by regulating the specific metabolic profiles of phenolic and flavonoid substances—converting bound precursors into highly active free metabolites.

In response to increasing consumer interest in functional foods, the rising prevalence of lactose intolerance, and the limited shelf life of fresh mulberries, enhancing the bioactivity of mulberry juice (MJ) through lactic acid bacteria fermentation has emerged as a research focus. However, existing studies predominantly examine the macro-level effects of fermentation, with limited systematic elucidation of the dynamic metabolic patterns of flavonoids and polyphenols during fermentation, or the molecular mechanisms underpinning enhanced antioxidant properties. Building upon our previous identification of Lactobacillus plantarum M3 as a superior strain, this study investigates its capacity to efficiently convert complex mulberry constituents through specific metabolic pathways. By integrating network pharmacology, molecular docking, and RT-qPCR, we elucidate the mechanisms underlying enhanced antioxidant activity and delineate the metabolic pathways governing the conversion of key metabolites. These findings provide a robust scientific basis for the high-value utilization of mulberry resources and the precise development of functional beverages.

2. Materials and Methods

2.1. Materials and Reagents

‘Black Girl’ Sang (M. alba ‘Heiniu’) taken from the Xinjiang Jiamu Fruit Tree Science National Long term Research Base in Aksu Prefecture, Xinjiang (41°15′ N, 80°32′ E). L. plantarum M3 preserved by the China General Microbiological Culture Collection Center (CGMCC, No. 26065, Beijing, China).

A001-1 hydroxylamine method total superoxide dismutase (T-SOD), A015-3-1 FRAP method total antioxidant capacity (T-AOC) test kits were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, Jiangsu, China). Microplate method Malondialdehyde (MDA), Microplate method reduced glutathione assay kit(GSH): Boyan Bio; Yeast extract, beef extract, and soy peptone (BR) were procured from Beijing Aoboxing Biotechnology Co., Ltd. (Beijing, China). All other chemicals for extraction were of analytical grade from Tianjin Zhiyuan Chemical Reagent Co., Ltd. (Tianjin, China). Folin phenol 2,2′-azino-bis(3-ethyl-benzothiazoline)-6-sulphonic acid (ABTS) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). HPLC-grade acetonitrile and methanol were purchased from Thermo Fisher Scientific (Shanghai, China) Co., Ltd. BS350B CCK-8 reagent kit, Biosharp (Hefei, China); DMEM (High glucose) Complete Medium (10) (PM150210B), Procell system (Wuhan, China); D6470 DCFH-DA Reactive Oxygen Species (ROS) Kit, Solarbio (Beijing, China); Removal and cDNA Synthesis SuperMix: AT311-03 (Beijing, China); PerfectStart^® Green qPCR SuperMix: AQ601-02-V2 (Beijing, China); TransZol Up: ET111-01-V2 (Beijing, China).

2.2. Preparation of Fermentation Products

Select mature ‘Black Girl’ mulberries that were fresh, free from pests, and without mechanical damage. Mash the berries, filter the mash, adjust the pH to 5.5, and set the initial sugar content to 12 °Bx. Placed the prepared mixture in a UV disinfection cabinet for 30 min for sterilization before further use. Activate the strains using basic MRS liquid medium, incubate at 37 °C for 24 h, and pass them for subsequent use. Select the second-generation strains for inoculation. Measure an appropriate amount of MJ, add 12.5 g/kg of metabisulfite, mix thoroughly, and inoculate with 5% L. plantarum M3. Put the inoculated sample in a constant-temperature shaking incubator at 100 rpm for 10 min. Mix thoroughly at 28 °C for fermentation testing. Three parallel experiments were set at the beginning of the experiment. With 6 h as one fermentation stage, fermented samples were collected separately at 0–72 h (M0, M6, …, M72) and stored at −80 °C for subsequent analysis.

2.3. Study on Antioxidant Activity and Bioactive Components of FMJ

2.3.1. Determination of T-SOD

Accurately pipette 25 μL of FMJ and dilute it with 200 μL of chilled physiological saline. Sonicate the mixture in an ice-water bath at a power setting of 80 W for 20 min. Centrifuge the solution at 4 °C and 10,000 rpm for 5 min, then carefully collect the supernatant for further analysis. Sequentially add 50 μL of the prepared sample along with other required reagents into a 96 well plate and measure the absorbance at 550 nm using a microplate reader. The calculation formula is as follows:

$$\mathrm{S}\mathrm{O}\mathrm{D}(\mathrm{U}/\mathrm{m}\mathrm{L})=\frac{{\mathrm{A}}_0-{\mathrm{A}}_{\mathrm{i}}}{{\mathrm{A}}_0}÷50\%\times \mathrm{B}\times \mathrm{C}$$ (1)

In the formula, A₀ is the control OD value; A_i is for measuring OD value; B is the dilution factor of the reaction system; and C is the dilution factor before sample testing.

2.3.2. Determination of MDA

Take 25 μL of FMJ and dilute it to 200 μL with distilled water; add 200 μL of sample and 300 μL of working solution in sequence, mix well, take out and cool on ice in a 90 °C water bath for 30 min, centrifuge at 25 °C and 12,000 rpm for 10 min, take 200 μL of supernatant in a 96 well plate, and measure the absorbance values at 532 nm and 600 nm, respectively. The calculation formula is as follows:

$$\mathrm{M}\mathrm{D}\mathrm{A}(\mathrm{n}\mathrm{m}\mathrm{o}\mathrm{l}/\mathrm{m}\mathrm{L})=[∆\mathrm{A}\times {\mathrm{V}}_2÷(\mathrm{\epsilon }\times \mathrm{d})\times {10}^9]÷{\mathrm{V}}_1$$ (2)

In the formula, $∆$A = A_532nm − A_600nm; V₂ is the amount of sample added and the total reaction volume of the working solution, 5 × 10⁻⁴ L; d is the optical path, 0.5 cm; $\mathrm{\epsilon }$ is the molar extinction coefficient of MDA, 155 × 10³ L/mol/cm; and V₁ is the sample volume added to the reaction system, mL.

2.3.3. Determination of Reduced GSH

Add 20 μL of FMJ sample and other required reagents to a 96-well plate in sequence, mix immediately, let it stand for 5 min, and measure the absorbance at 412 nm. At the same time, prepare a control sample under the same conditions. The calculation formula is as follows:

$$\mathrm{G}\mathrm{S}\mathrm{H}(\mathsf{\mu }\mathrm{m}\mathrm{o}\mathrm{l}/\mathrm{m}\mathrm{L})=\left[\right(∆\mathrm{A}-0.006)÷0.9676\times {\mathrm{V}}_1]÷{\mathrm{V}}_1$$ (3)

In the formula, $∆\mathrm{A}={\mathrm{A}}_{\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{t}\mathrm{u}\mathrm{b}\mathrm{e}}-{\mathrm{A}}_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{c}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{o}\mathrm{f}}$; V₁ is the volume of supernatant added to the reaction system, mL.

2.3.4. DPPH Radical Scavenging Activity

Add 100 μL of FMJ sample solution and 100 μL of DPPH solution (150 μM DPPH methanol solution) to a 96-well plate, mix well, and react at room temperature in the dark for 30 min. Immediately measure their absorbance at 517 nm [21]; the calculation formula for DPPH radical scavenging rate is as follows:

$$\mathrm{D}\mathrm{P}\mathrm{P}\mathrm{H}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}/\%=(1-\frac{{\mathrm{A}}_{\mathrm{i}}-{\mathrm{A}}_{\mathrm{j}}}{{\mathrm{A}}_0})\times 100$$ (4)

In the formula, ${\mathrm{A}}_0$ is the absorbance value of 100 μL methanol + 100 μL DPPH solution; ${\mathrm{A}}_{\mathrm{i}}$ is the absorption value of 100 μL FMJ sample solution + 100 μL DPPH solution; and ${\mathrm{A}}_{\mathrm{j}}$ is the absorption value of 100 μL FMJ sample solution + 100 μL methanol.

2.3.5. Determination of Total Anthocyanin

Take 100 μL of FMJ, add anhydrous ethanol with a volume fraction of 70% at a ratio of 1:30, extract with ultrasound at room temperature at 80 Hz for 30 min, and centrifuge at 4 °C and 6000 rpm for 5 min. Collect the supernatant. Take 500 μL of the extraction solution and dilute it to 1 mL with pH 1.0 buffer and pH 4.5 buffer, respectively. Equilibrate at 30 °C for 20 min in a water bath and measure the absorbance at 510 nm and 700 nm, respectively. The calculation formula is as follows:

$$\mathrm{A}=({\mathrm{A}}_{510}-{\mathrm{A}}_{700})\mathrm{p}\mathrm{H}1.0-({\mathrm{A}}_{510}-{\mathrm{A}}_{700})\mathrm{p}\mathrm{H}4.5$$ (5)

$$\mathrm{C}(\mathrm{m}\mathrm{g}/\mathrm{L})=\left(\right(\mathrm{A}\times \mathrm{M}\times \mathrm{D}\mathrm{F}\times 1000)/\mathsf{\alpha }\times \mathrm{l})\times \mathrm{T}$$ (6)

In the formula, the absorbance value of A₅₁₀ FMJ sample at 510 nm; the absorbance value of A₇₀₀ FMJ sample at 700 nm; M is the molecular weight, 449.2 g/mol; DF is the dilution factor; α is the molar extinction coefficient of cyanidin-3-O-glucoside, 26,900 L/(mol·cm); l is the length of the optical path, cm; and T is the dilution factor.

2.3.6. ABTS Radical Scavenging Activity

Slightly modified according to the method proposed by Yao et al. [22], accurately weigh 0.0768 g of ABTS free radicals and 0.0268 g of potassium persulfate sample, dilute to 20 mL with pure water, and then mix evenly in a 1:1 ratio and store in the dark for 12 h to form ABTS⁺ working solution. Dilute with pure water in a 1:14 ratio before use. Add 100 μL of FMJ sample solution in sequence to a 96-well plate, mix well, and react in the dark at 37 °C for 10 min. Measure the absorbance at 734 nm. The calculation formula for ABTS cation radical scavenging rate is as follows:

$$\mathrm{A}\mathrm{B}\mathrm{T}\mathrm{S}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}/\%=(1-\frac{{\mathrm{A}}_{\mathrm{i}}-{\mathrm{A}}_{\mathrm{j}}}{{\mathrm{A}}_0})\times 100$$ (7)

In the formula, ${\mathrm{A}}_0$ is the absorbance value of 100 μL deionized water + 100 μL ABTS solution; ${\mathrm{A}}_{\mathrm{i}}$ is the absorption value of 100 μL FMJ sample solution + 100 μL ABTS solution; and ${\mathrm{A}}_{\mathrm{j}}$ is the absorption value of 100 μL FMJ sample solution + 100 μL deionized water.

2.3.7. Determination of Total Phenol

Accurately weigh 0.1 g of gallic acid, dissolve it in 50 mL of distilled water, and make it 100 mL to obtain a standard stock solution of gallic acid with a mass concentration of 1000 mg/mL. Take 0, 1.25, 2.5, 5, 10, 20, and 40 mL of the standard stock solution into 100 mL volumetric flasks, fill to the mark with distilled water, and prepare a series of standard solutions with mass concentrations of 0, 12.5, 25, 50, 100, 200, and 400 mg/L. Take 100 μL of FMJ sample, dilute to 1 mL with 80% ethanol, sonicate for 30 min in an ice bath at 80 Hz, 4 °C, and centrifuge at 6000 rpm for 5 min. Collect the supernatant and determine the total phenolic content using the Folin–Ciocalteu method [23]. Mix the extraction solution (0.2 mL) and distilled water (3 mL) in a 10 mL volumetric flask, then add Folin–Ciocalteu reagent (0.5 mL). After 1 min, add 1 mL of 20% Na₂CO₃ and mix well. After reacting at 75 °C for 10 min, measure the absorbance at 760 nm. Take 200 μL of the sample for testing.

$$\mathrm{X}(\mathrm{m}\mathrm{g}/\mathrm{L})=\mathrm{C}\times \mathrm{T}$$ (8)

In the formula, C is the concentration measured by the standard curve, mg/L; T is the dilution factor.

2.3.8. Determination of Total Flavonoid

Accurately weigh 200 μL of the sample, add a volume fraction of 60% ethanol solution in a solid–liquid ratio of 1:5, extract with ultrasound for 60 min, perform secondary extraction, and combine the filtrate. Accurately weigh 0.015 g of rutin and dilute it with 80% ethanol in a 50 mL volumetric flask. Take 0, 1.0, 2.0, 3.0, 4.0, 5.0, and 6.0 mL of the aforementioned rutin solution and dilute with 80% ethanol to a 50 mL volumetric flask for later use. The total flavonoid content of FMJ was measured at 418 nm with slight modifications based on the colorimetric method using aluminum trichloride acetic acid sodium acetate buffer solution.

$$\mathrm{X}(\mathrm{m}\mathrm{g}/\mathrm{L})=\mathrm{C}\times \mathrm{T}\times 1000$$ (9)

In the formula, C represents the concentration measured by the standard curve, mg/mL; T is the dilution factor; and 1000 is the conversion factor.

2.3.9. Determination of T-AOC

Determine the ability of the sample to reduce trivalent iron ions according to the protocol of the reagent kit. Take FMJ and sonicate under ice-water bath (power 80 W, 20 min), centrifuge at 4 °C, 10,000 rpm for 5 min, and take the supernatant for testing. And sequentially add 5 μL of sample solution and 180 μL of FRAP working solution to the 96-well plate. Incubate the plate at 37 °C for 5 min and measure the absorbance at 593 nm. Dilute the standard working solution with distilled water using FeSO₄·7H₂O as the standard to create the corresponding concentration gradient. Draw a standard curve with concentration as the x-axis and absorbance as the y-axis.

2.4. Broad Targeted Metabolomics Analysis

2.4.1. Metabolite Extraction

A 100 μL aliquot of the FMJ sample was measured and diluted to 2 mL with 70% methanol. The mixture was sonicated for 40 min and centrifuged at 10,000 rpm for 10 min at 4 °C. The resulting supernatant was collected and filtered through a 0.22 μm organic filter membrane. Each sample was prepared in triplicate as biological replicates. For monitoring instrument stability and analytical reproducibility, quality control (QC) samples were generated by pooling aliquots from all individual samples. These QC samples were analyzed together with the experimental samples. Additionally, a QC sample was inserted regularly after every five experimental samples during the analytical sequence.

2.4.2. UPLC-Q-TOF-MS Based Widely Targeted Metabolic Analysis

A UPLC-Q-TOF-MS system (I-CLASS/UPCC/XEVOG2-XS Waters, Milford, MA, USA) equipped with an ACQUITY UPLC^® BEH C18 column (1.7 μm, 2.1 mm × 100 mm) was used (Waters, Milford, MA, USA). The mobile phase consisted of acetonitrile (A) and 0.1% formic acid aqueous solution (B). Gradient elution conditions are detailed in Table 1. The flow rate was set at 0.4 mL/min, the column temperature at 45 °C, and the sample cell temperature at 4 °C. The injection volume was 2 μL; balance the column with the initial mobile phase for no less than 30 min prior to injection. Thoroughly clean the system before and after each injection. Before injection, flush three times with the methanol. After injection, sequentially flush with methanol, ultrapure water, and the sample solvent to ensure no cross-contamination occurs. After every 10 samples are injected, inject methanol to monitor residues. After testing is complete, clean the chromatographic column.

Table 1.

Gradient elution conditions.

Time/min	A/%	B/%
0	5	95
3	40	60
10	95	5
12	95	5
13	5	95
15	5	95

The ion source employed is an electrospray ionization (ESI) source, capable of operating in both positive and negative ion modes. Detection was performed in sensitivity mode using the MSE data acquisition method. The ion source temperature was maintained at 120 °C. Nitrogen was utilized as the desolvation gas with a flow rate of 800 L/h, while the cone gas flow rate was set to 50 L/h at 450 °C. The capillary voltage was adjusted to 3 kV for positive ions and 2.5 kV for negative ions, with the cone voltage fixed at 40 V. The mass scanning range covered from 50 to 1200 Da. During low-energy scans, the collision energy was set to 0 eV for both positive and negative ions, whereas it ranged from 15 to 40 eV during high-energy scans. Accurate mass calibration was achieved using a Leucine-Enkephalin (LE) solution at a concentration of 200 pg/mL, resulting in m/z values of 556.2771 in positive ion mode and 554.2615 in negative ion mode. Data collection was conducted without further calibration.

2.5. Network Pharmacology Analysis and Screening of Antioxidant Targets

Obtain target information of key antioxidant compounds identified in FMJ through the SwissTarget Prediction database (http://swisstargetprediction.ch/ (accessed on 30 October 2024)) and obtain antioxidant disease targets from the GeneCards database (https://www.genecards.org/ (accessed 1 November 2024)). Cross-target information between the composite target and antioxidant related targets was screened through Wayne analysis. GO and KEGG enrichment analysis of cross-targets was performed using the David database (https://davidbioinformatics.nih.gov/). Metabolite target activity network and protein–protein interaction network were constructed using Cytoscape 3.9.1 software. The Centiscape plugin conducted topology analysis and obtained Betweenness unDir, Degree, Closeness unDir. And select core antioxidant targets with high correlation based on the average values of these parameters.

2.6. Molecular Docking

The protein crystal structure of the core antioxidant target was obtained from the PDB protein database (https://www.rcsb.org/ (accessed 1 November 2024)), and the ligand structure was extracted from Pubchem (https://pubchem.ncbi.nlm.nih.gov/ (accessed 1 November 2024)). Through RCSB PDB, download the target PDB file from the homepage. After hydrogenation and dehydration of the core antioxidant target structure using Discover Studio 4.5 software, molecular docking and visualization were performed.

2.7. Protective Effect of Mulberry Juice on the BV2 Cells Oxidative Damage Model

2.7.1. Cell Viability Experiment

BV2 cells were provided by Wuhan SUNNCELL. Dilute the cell suspension of BV2 cells in logarithmic growth phase to 5 × 10⁴ cells/mL and inoculate 100 μL per well into a 96-well plate. After culturing the cells in a 37 °C, 5% CO₂ incubator for 24 h, a blank group and a MJ treatment group (dilute mulberry juice by 4, 8, 16, 32, 64, 128 times, respectively) were set up with six replicates per well. After treating the cells with mulberry samples for 24 h, remove the old culture medium and add 100 μL/mL (10 μL) of CCK-8 culture medium (100 μL/well). Incubate in a carbon dioxide incubator for 30 min and record the absorbance value at 450 nm.

Cell viability = [OD _(dosing) − OD _(blank)]/[OD _{(0 dosing)} − OD _(blank)] × 100% (10)

In the formula, OD _{(drug addition)}: absorbance value of cells + culture medium + CCK-8 + drug (sample); OD _(blank): Culture medium + CCK-8 absorbance value; and OD _{(0 dosing)}: Cell + culture medium + CCK-8 absorbance value.

2.7.2. Establishing a Cellular Oxidative Stress Model by Stimulating BV2 Cells with H₂O₂

BV2 cells were cultured in a 96-well plate, ensuring that the number of cells in each well was around 5 × 10⁴. Divide the cells into a blank group and an H₂O₂ group (0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7 mM) with six replicates in each group. Treat with H₂O₂ for 6 h, remove the old culture medium, add 100 μL/mL (10 μL) of CCK-8 culture medium (100 μL/well), incubate in a carbon dioxide incubator for 2 h, and record the absorbance value at 450 nm. The calculation method is the same as Formula 10.

2.7.3. The Effect of Fermented Mulberry Residue on the Survival Rate of BV2 Cells Induced by H₂O₂

Digest BV2 cells in logarithmic phase and inoculate them into a 96-well plate (5 × 10⁴/well). The experimental groups are as follows, with six replicates in each group—Control: blank control group, Model: H₂O₂ (300 μM), M (medium dose): 16 times + H₂O₂ (300 μM), H (high dose): 8 times + H₂O₂ (300 μM). After cell adhesion, the drug group was treated with samples of different concentrations for 24 h, and the old culture medium was removed. The model group and drug group were treated with 300 μM H₂O₂ for 6 h, and the cell survival rate was detected by CCK-8 method.

2.8. Flow Cytometry Analysis of ROS

Use a DCFH-DA probe-based ROS detection kit [24] to evaluate ROS production in cells through FCM. Dilute DCFH-DA with FBS free medium to a final concentration of 10 μM. BV2 cells (4 × 10⁵) were treated with the sample for 24 h, then treated with H₂O₂ for 6 h and washed twice with PBS. Then, add 1.5 mL of 10 μM DCFH-DA and incubate at 37 °C for 30 min (shaking once every 5 min), and wash twice with FBS free medium. Finally, ROS in the cells was detected and analyzed using a BD FACSAria III with an excitation wavelength of 525 nm.

2.9. RT-qPCR Validation

The total RNA sample used for RT-qPCR analysis is the same as the sample used for metabolomics and antioxidant analysis. Extract BV2 cells RNA using TransZol Up (ET111-01-V2, TransGen Biotech, Beijing, China) and perform reverse transcription using the Removal and cDNA Synthesis SuperMix kit (AT311-03, TransGen Biotech, China). Screening targets were based on PPI–protein interaction core targets and KEGG pathway enrichment results, using NCBI (https://www.ncbi.nlm.nih.gov/ (accessed 1 November 2024)). Primer pairs were designed with Shenggong Bioengineering (Shanghai) Co., Ltd. (Shanghai, China). β-actin mRNA is expressed at moderate abundance levels in most cell types and is a common reference gene for RT-qPCR [25]. Use ABI 7500 Fast real-time PCR System and PerfectStart^® Perform RT-qPCR using Green qPCR SuperMix (AQ601-02-V2, TransGen Biotech, China) in eight tubes to verify the specificity of the amplicons. Perform three replicates of RT-qPCR on each sample and calculate the relative expression level of each gene using the 2^−ΔΔCt method [26]. The list of gene primers is shown in Table 2.

Table 2.

Target genes and their primer sequences for RT-qPCR analysis.

Gene Name	Primer Sequence
β-actin-F	5′-tactgctctggctcctagca-3′
β-actin-R	5′-cggactcatcgtactcctgc-3′
SOD1-F	5′-gactgactgaaggcctgcat-3′
SOD1-R	5′-gtgaggacctgcactggtac-3′
Src-F	5′-tccgacttcgacaatgccaa-3′
Src-R	5′-gacatacggtagtgaggcgg-3′
Nfe2l2-F	5′-cagagtgatggttgcccact-3′
Nfe2l2-R	5′-cacacactttctgcgtgctc-3′

2.10. Statistical Analysis

All experiments and sample analyses were conducted in three biological replicates, with results expressed as mean ± standard deviation, and data were processed using GraphPad 10 software with one-way analysis of variance (ANOVA) at a significance level of 0.05 followed by multiple comparisons via Tukey’s multiple comparison test to identify specific differences between each pair of levels, while data visualization was performed using Origin 2024 and GraphPad 10. The metabolomic dataset was subjected to multivariate statistical analysis through the Metware Cloud Platform (https://cloud.metware.cn/#/user/login (accessed on 11 November 2024)) using established workflows for metabolic profiling.

3. Results and Discussion

3.1. Comparison of Antioxidant Activity of MJ

To systematically characterize the functional transformation of mulberry juice, a multi-dimensional experimental strategy was implemented. First, the dynamic changes in nutritional and bioactive components were quantified to establish a robust chemical baseline. Subsequently, the correlation between the evolution of phenolic profiles and in vitro antioxidant activities during fermentation was scrutinized via non-targeted metabolomics. Building on this, network pharmacology and molecular docking were employed to predict the specific antioxidant targets and bioactive drivers within the fermented mulberry juice (FMJ). The cytoprotective efficacy of FMJ in the H₂O₂-induced BV2 cell model is highly consistent with the fermentation-induced elevations in TPC, TFC, and SOD activity. This synergy underscores that the biotransformation of polyphenols not only enhances chemical radical scavenging (DPPH/ABTS) but also translates into a tangible biological response, effectively shielding cells from oxidative stress-mediated injury. Spanning comprehensive chemical analysis, metabolic tracking and cellular validation ensure a holistic evaluation of the fermentation-induced enhancement of antioxidant potency in MJ.

Chemical antioxidant assays represent one of the most common and straightforward approaches for evaluating antioxidant capacity. Given that different methods operate via distinct mechanisms, relying solely on a single assay may lead to incomplete or biased results. Therefore, it is essential to employ multiple complementary methods to obtain a comprehensive understanding of the antioxidant properties of FMJ [27]. The antioxidant activity of fermented mixed juice (FMJ) was assessed using a combination of nutritional component analysis—specifically total anthocyanin content (TAC), total phenolic content (TPC), and total flavonoid content (TFC)—alongside key antioxidant assays, including superoxide dismutase (SOD), DPPH, ABTS, and FRAP. This multi-faceted approach aimed to provide a holistic assessment of the changes in antioxidant capacity during fermentation (Figure 1). As depicted in Figure 1c–h, key antioxidant metrics—SOD activity, GSH content, DPPH and ABTS radical scavenging rates, and TAC—all demonstrated a consistent pattern of initial increase followed by a decrease after fermentation. This phenomenon, particularly the increase in SOD activity, can be attributed to the proliferation of LAB, which are known to produce SOD—a finding corroborated by previous studies on edible grass fermentation [28]. This finding is consistent with Akan et al., who reported a similar initial increase followed by a decrease in the DPPH radical scavenging rate during yogurt fermentation with L. casei over time. It is hypothesized that L. plantarum fermentation enhances the bioavailability and utilization efficiency of antioxidant compounds acting as proton donors, such as vitamin C. This hypothesis is supported by prior evidence indicating that LAB fermentation increases the availability of proton-donating substances, thereby enhancing DPPH radical scavenging capacity [29]. Furthermore, research has demonstrated that lactic acid fermentation significantly improves the ABTS radical scavenging activity of orange juice. Similarly, studies on L. plantarum-fermented quinoa beverages have shown that ABTS radical scavenging activity follows a trend comparable to that of DPPH radical scavenging [30]. During fermentation, the total phenolic content in the juice increases markedly. These polyphenolic compounds exhibit strong antioxidant properties and effectively neutralize free radicals in biological systems, thus reducing MDA formation [31].

Figure 1.

Changes in antioxidant substances during FMJ fermentation process: (a) TPC; (b) TFC; (c) TAC; (d) SOD enzyme activity; (e) MDA; (f) GSH; (g) ABTS; (h) DPPH and (i) FRAP. a–m indicate significant differences (p < 0.05) between consecutive time points (0 h vs. 6 h, 6 h vs. 12 h, …, 66 h vs. 72 h) and between 72 h vs. 0 h; ns = not significant. Standard curves for TPC, TFC, and FRAP are shown in Supplemental Figure S1.

As shown in Figure 1a,b,i, the total antioxidant capacity of FMJ—measured by TPC, TFC, and FRAP values—initially increased and then decreased. The significant increases in total phenolic content (TPC) and total flavonoid content (TFC) observed during the fermentation process (Figure 1a,b) are highly consistent with previous studies on LAB-fermented beverages, such as cactus pear juice, mulberry juice [32], and Lycium barbarum-Longan blended juice [33]. This enrichment of bioactive compounds is primarily attributed to the enzymatic activities of LAB (β-glucosidase and esterase), which catalyze the hydrolysis of complex or bound phenolics into their more bioavailable aglycone forms. Notably, our findings align with the synergistic effects reported in mulberry juice fermentation (co-culturing L. paracasei and L. helveticus), as the TPC and TFC in this study reached their peaks at 36–48 h, suggesting high bioconversion efficiency within the specific fruit matrix. A similar trend in TPC was reported in LAB-fermented sweet potato residue, where phenolic content significantly increased (90.51–97.35 mg/100 g DW, p < 0.05), likely due to the hydrolysis of phenolic compounds and release of bound phenolics during fermentation [34]. During fermentation with L. plantarum M3, TFC in MJ also showed an initial rise, peaking at 36 h before declining. This pattern aligns with reports on LAB-fermented citrus juice, where flavonoid levels increased significantly up to 48 h, indicating time-dependent variation [35]. Similar dynamics have been observed in other fermented substrates: for instance, LAB-fermented pumpkin juice showed a notable increase in TPC after fermentation [36], while the antioxidant activity of Bacillus subtilis-fermented millet hydrolysate first rose and then fell with prolonged fermentation [37]. The initial enhancement is attributed to enzymatic hydrolysis of macromolecules, releasing peptides, flavonoids, and polyphenols, whereas prolonged fermentation may lead to the degradation of these antioxidants, ultimately reducing overall antioxidant capacity. Furthermore, the strong positive correlation between the elevation in TPC/TFC and the enhanced antioxidant capacities (DPPH, ABTS, and FRAP) corroborates findings in fermented strawberry and kiwi juices [38], where the accumulation of phenolic metabolites directly drives the scavenging of free radicals. These results collectively demonstrate that the LAB strains not only preserve but significantly fortify the nutritional profile and functional value of the juice through optimized fermentation kinetics.

3.2. Analysis of Metabolic Product Differences Between Adjacent Samples During Fermentation

A comprehensive targeted metabolomics analysis was performed using UPLC-Q-TOF-MS to profile compositional changes during fermentation. Figure 2a,b shows the total ion chromatograms of MJ samples acquired in ES⁺ and ES⁻ modes, respectively. In these plots, metabolite ion intensity (y-axis) is displayed against retention time (x-axis), with different colored curves representing different fermentation time points. Potential metabolites were annotated by querying established databases—including CNKI, Google Scholar, and SciFinder—along with mass spectral data collected in both positive and negative ionization modes.

Figure 2.

(a) shows the peak ion chromatogram (BPI) of the FMJ fermentation process detected by UPLC-Q-TOF-MS in ES⁺ mode (positive electrospray ionization mode). (b) Under ES⁻ mode (negative electrospray ionization mode), UPLC-Q-TOF-MS detects the peak ion chromatogram (BPI). (c) Identification of compound quantities in ES⁺ and ES⁻ modes. (d) Classification of all identified metabolites during fermentation. (e) PCA diagram of the entire fermentation and QC, including all detected compounds. (f) OPLS-DA diagram of the entire fermentation and QC. (g) Compound clustering analysis heatmap. (h,i) show OPLS-DA response permutation test chart.

Spectral data were processed using QI software (Progenesis QI V2.0) to establish a putative metabolite database by aligning retention times and secondary fragmentation spectra. This approach enabled the identification of compounds along with their names, molecular formulas, and structures. A total of 195 chemical components were annotated, predominantly comprising 67 alkaloids, 36 flavonoids, 17 amino acids, and 17 phenolic compounds, along with smaller proportions of fatty acids, organic acids, peptides, and other minor constituents (Supplementary Table S1). As shown in Figure 2d, alkaloids and amino acids maintained high relative abundance throughout fermentation, while flavonoid content declined significantly in the early fermentation stage before stabilizing in later phases. UPLC-Q-TOF-MS results indicate that the hydrolysis of bound polyphenols, such as rutin, into free polyphenols, including gallic acid, occurred during this stage, thereby enhancing bioavailability and antioxidant efficacy, which resulted in a significant increase in antioxidant activity.

The initial increase in TFC following LAB fermentation is mainly attributed to the enzymatic conversion of bound flavonoids into their free forms. However, the inherent instability of these compounds, coupled with complex enzymatic dynamics, likely contributes to the subsequent decline, consistent with the trend observed in this study [39]. Similarly, total phenolic content also showed an initial rise in the early fermentation stage, which aligns with reports by Zhao et al., who ascribed such increases to fermentation-associated enzymatic activities [40].

PCA was applied to assess global metabolite differences and variation among FMJ samples. As shown in the score plot (Figure 2e), the first two principal components, PC1 and PC2, accounted for 57.58% and 15.59% of the total variance, respectively, with a cumulative contribution of 73.13. While the three biological replicates within each group clustered closely, indicating good reproducibility, clear separation was observed between the blank control (BY) and fermentation samples. Notably, pre-fermentation (M0–M24) and post-fermentation (M30–M72) samples formed distinct clusters, reflecting the substantial influence of L. plantarum M3 fermentation on the metabolite profile of FMJ. Additionally, QC samples clustered tightly together and were well separated from the FMJ samples, demonstrating the stability of the analytical process and the high reproducibility of the acquired data.

To further investigate compositional changes during fermentation, cluster analysis was performed and visualized using a heatmap (Figure 2g). The color gradient, ranging from green to pink, represents a decrease in relative metabolite abundance. The heatmap revealed three distinct clusters, one containing BY and M0, a second comprising M6–M24, and a third consisting of the later time points, indicating pronounced shifts in metabolite composition across fermentation stages. This visualization aligns with the PCA results, highlighting the distinct metabolic profiles between the early, middle, and late stages of fermentation.

To examine non-volatile metabolite variations between adjacent time points, OPLS-DA was applied to each sequential interval. As shown in Supplemental Figure S2, samples from consecutive time points were clearly separated along the first principal component, suggesting significant metabolic alterations. Moreover, high values of both model fit (R²Y) and predictability (Q²) obtained from permutation tests (Supplemental Figure S3) confirmed the robustness and reliability of the OPLS-DA models, supporting their use in identifying differential metabolites throughout fermentation.

The volcano plot (Supplementary Figure S4) clearly reveals significant differences in metabolite abundance between adjacent fermentation time points. Notably, flavonoids were substantially up- or down-regulated during this process, indicating that fermentation profoundly alters the metabolite profile of MJ.

The fermentation process was divided into five stages: I (M0–M6), II (M6–M30), III (M30–M42), IV (M42–M54), and V (M60–M72). Stages II and IV were identified as critical transition phases for metabolites, likely driven by rapid chemical transformations such as isomerization, oxidation, polymerization, and hydrolysis [41]. Existing studies indicate that phytochemicals such as phenols, flavonoids (e.g., apigenin, luteolin), and carotenoids exhibit strong antioxidant properties [42,43]. Visualization of flavonoid and polyphenolic metabolism (Supplementary Figure S5) revealed their predominant distribution in subclasses including benzofurans, phenolic acids, and diverse flavonoids. Previous studies have demonstrated that fermentation effectively promotes the release and transformation of phenolic compounds in fruit and vegetable juices. For instance, fermentation of strawberry juice increases the content of gallic acid, protocatechuic acid, caffeic acid, and p-coumaric acid [44], whereas fermentation of red beetroot juice significantly elevates the concentration of flavonoid compounds such as rutin, kaempferol, and (+)-catechin [45]. During the initial fermentation phase (0–48 h) of this study, the system pH progressively decreased as lactic acid accumulated. Concurrently, enzymes secreted by lactic acid bacteria, such as β-glucosidase, convert large-molecule conjugated polyphenols like rutin into small-molecule free polyphenols including gallic acid, ferulic acid, p-coumaric acid, and protocatechuic acid. This process markedly enhances the system’s bioavailability and antioxidant efficacy. However, as fermentation progressed, the rapid proliferation of microbial cells consumed substantial nutrients. When substrates in the fermentation broth became insufficient to sustain microbial growth, accumulated metabolites were reutilized. This led to a decline in phenolic compounds such as gallic acid after 48 h, a trend corroborated by antioxidant activity analyses. Although antioxidant components re-accumulated during the later fermentation stage, overall, mulberry juice exhibited optimal antioxidant substance enrichment levels and antioxidant activity at 48 h of fermentation. Furthermore, isopentenyl flavonoids predominantly exhibited upregulation. According to Mazimba [46] and Martins [47], their monomeric forms demonstrate DPPH radical scavenging activity and may enhance overall antioxidant efficacy through synergistic interactions with other active compounds.

The clustering heatmap of flavonoids and polyphenolic metabolites (Supplementary Figure S6) illustrates the temporal dynamics of metabolite profiles across adjacent fermentation time points. Studies have shown that the relative concentrations of protocatechuic acid, Albanin A, rutin, apigenin, and eriodictyol initially increase and then decrease during fermentation. This pattern can be attributed to the metabolic activity of LAB, which consumes glucose and oxygen from phenolic compounds. This process liberates free glycosidic ligands, often with increased hydroxyl group availability or reduced steric hindrance, ultimately altering the composition and levels of phenolic substances. Experimental findings demonstrate that LAB fermentation contributes to the increase in gallic acid in fruit juice through the enzymatic breakdown of gallic acid esters, thereby releasing free gallic acid with strong antioxidant properties. This study also observed a significant decrease in dihydroquercetin, whereas protocatechuic acid levels began to rise after 30 h—a pattern consistent with its known metabolic derivation from quercetin. Meanwhile, p-coumaric acid, a hydroxylated cinnamic acid derivative, can be hydrolyzed from its esterified forms or conjugated into acylated anthocyanins, leading to an initial decline. However, as the pH increases during fermentation, anthocyanin deacylation occurs, subsequently releasing p-coumaric acid and raising its concentration [48]. Trace amounts of phenolic acids, including caffeic acid and ferulic acid, were detected in the early fermentation stage, suggesting their potential formation through chalcone intermediates. This observation is further supported by previous studies demonstrating that naringin serves as a precursor to flavonoids such as apigenin and luteolin, as well as the flavanone naringenin. In the flavonoid biosynthesis pathway, flavonoid synthase (FNS) oxidoreductase catalyzes the oxidation of naringin to form apigenin by introducing additional double bonds. Subsequently, flavanone 3-hydroxylase (F3H) plays a critical role in hydroxylating the B-ring of apigenin to produce luteolin, thereby modifying the chemical structure of flavonoids. This structural alteration not only affects their biological activity but also influences plant color and aroma. Additionally, F3H can hydroxylate naringenin at the C3′ position of its B-ring to directly generate eriodictyol. Studies have confirmed that apigenin, luteolin, and eriodictyol exhibit strong antioxidant activity [49]. Rutin is first converted to kaempferol-3-rutinoside via dehydroxylation, which can be further hydrolyzed by enzymes such as α-rhamnosidase, β-glucosidase, or hesperidinase to yield kaempferol-3-glucoside or free kaempferol. The conversion of flavanols to their glycosidic forms has also been documented during fermentation processes involving lactic acid bacteria and yeast [50]. In fermented fruit and vegetable juices, phenols and flavonoids constitute the key functional components underpinning antioxidant activity. The liberation of bound polyphenols through hydrolysis enhances their content in mulberry juice, leading to improved antioxidant properties [51]. Numerous in vitro studies have confirmed the potent antioxidant activity of phenolic compounds. In alignment with this, our correlation analysis (Supplementary Figure S7) revealed that eriodictyol was significantly negatively correlated with MDA content while positively correlated with total flavonoids. Additionally, quinic acid showed a significant positive correlation with total phenolic content.

3.3. The Mechanism of Action of Key Antioxidant Substances in FMJ

3.3.1. Screening of Antioxidant Targets Based on Network Pharmacology

Following experimental optimization, which identified 48 h as the optimal fermentation duration for FMJ, differences in flavonoid and polyphenolic metabolite profiles were analyzed utilizing volcano plots (Figure 3a). The potential antioxidant mechanisms of these differential metabolites were further explored through network pharmacology (Figure 3). A total of 260 targets for 11 key antioxidant compounds in FMJ were identified from the SwissTarget Prediction database. Additionally, 5179 antioxidant-related targets were screened using the GeneCards database. By intersecting these datasets, 197 potential targets associated with antioxidant compounds in FMJ were pinpointed. To investigate the relationship between antioxidant compounds and their targets, a “metabolite-target-activity” network was constructed, comprising 270 nodes (1 sample node, 11 metabolite nodes, and 258 antioxidant targets) and 269 edges. This network reveals complex many-to-many interactions between antioxidant compounds and their targets, indicating a synergistic effect among these compounds. To identify core antioxidant targets, a protein–protein interaction (PPI) network was constructed using the STRING database and Cytoscape 3.10.0 software, analyzing the antioxidant targets of 260 active compounds in FMJ samples. The PPI network consists of 195 nodes and 2835 edges, suggesting that key antioxidant substances in FMJ exert their effects through multiple targets and pathways. Topological analysis using the CentiScaPe plugin revealed that the Betweenness unDir is 213.28, the average Degree is 29.08, and the average Closeness unDir is 0.0025. Thirty-seven antioxidant indicators exceed the average values of these three parameters (Supplementary Table S2).

Figure 3.

Network pharmacology analysis of FMJ core antioxidant compounds. (a) Volcanic diagram comparing the differences in flavonoids and polyphenolic compounds between M0 and M48 (FC > 2 or <0.5, p < 0.05, VIP > 1). (b) Comparison of differences in flavonoids and polyphenols clustering heatmap between M0 and M48. (c) Wayne diagram of core antioxidant compounds and antioxidant targets. (d) FMJ metabolite target activity network. (e) FMJ protein–protein interaction network. (f) GO enrichment analysis. (g) KEGG enrichment analysis.

Ten core antioxidant targets—such as AKT1, SRC, EGFR, PTGS2, STAT3, CTNNB1, ESR1, HSP90AA1, and CASP3—were identified based on network centrality (Betweenness unDir > 1000). Notably, SRC kinase, one of the key targets, has been reported to promote oxidative stress and suppress autophagy via PI3K/Akt/mTOR signaling in Duchenne muscular dystrophy models, where its inhibition alleviated oxidative damage and improved autophagy-lysosomal function [52]. As depicted in Figure 3, GO term enrichment analysis highlighted significant associations with biological processes including the positive regulation of superoxide anion generation and negative regulation of oxidative stress-induced apoptosis, molecular functions such as NAD(P)H-dependent oxidoreductase activity, and cellular components comprising the endoplasmic reticulum and mitochondrial outer membrane. Concurrently, KEGG pathway analysis identified 197 antioxidant targets as being significantly enriched in a range of pathways, with the SRC signaling pathway emerging among representative entries encompassing chemical carcinogenesis–reactive oxygen species, atherosclerosis-related processes, and neutrophil extracellular trap formation.

3.3.2. Molecular Docking Between Key Antioxidant Compounds and Antioxidant Core Targets

To further elucidate the mechanism of key antioxidants in FMJ, molecular docking simulations were conducted using Discovery Studio 4.5. The binding affinities between core antioxidant compounds and their potential targets were systematically evaluated through conformational sampling within a 10 Å radius. Comparative analysis indicated that the flavonoid Albanin A and the phenolic compound Moracin Q exhibited particularly strong binding potentials with three pivotal antioxidant targets: NFE2L2, SOD, and STAT3 (Supplementary Table S3).

Notably, these compounds demonstrated remarkable LibDock scores with the targets: Albanin A achieved scores of 136.055 (NFE2L2), 103.713 (SOD), and 111.641 (SRC), while Moracin Q showed scores of 129.741, 94.545, and 101.046, respectively. Molecular docking indicates that Albanin A may have critical binding interactions with NFE2L2, particularly through hydrogen bonding with ARG-69, ARG-515, and ARG-72 residues, as well as hydrophobic interactions involving LYS-516, LEU-519, and multiple cysteine residues (CYS-68, CYS-514) (Figure 4).

Figure 4.

Molecular docking of key antioxidants and antioxidant targets in FMJ. (a) Albanin A-NFE2L2; (b) Albanin A-SOD; (c) Albanin A-SRC; (d) Moracin Q-NFE2L2; (e) Moracin Q-SOD; (f) Moracin Q-SRC; (g) Ferulic acid-NFE2L2; (h) Ferulic acid-SOD; (i) Ferulic acid-SRC.

The computational findings identify Albanin A as a principal antioxidant component in FMJ, with NFE2L2, SOD, and STAT3 as its main molecular targets. The favorable binding interactions provide a structural basis for the observed antioxidant activity at molecular resolution, validating the importance of polyphenolic compounds in targeting key redox signaling pathways. It is worth noting that due to the inherent computational boundaries of molecular docking, these results serve as predictive support rather than definitive mechanistic validation.

3.4. Protective Effect of Mulberry Juice on BV2 Cell Oxidative Damage Model

Oxidative stress and inflammation are associated with the occurrence and progression of neurodegenerative diseases. Chronic activation of microglia leads to excessive production of nitric oxide and reactive oxygen species, which can induce neurodegeneration [53]. The CCK-8 method was used to evaluate the effect of MJ before and after fermentation on BV2 cell viability (Supplementary Figure S8a,b). After incubating cells with MJ before and after fermentation for 24 h, no significant effect was observed on cell viability, indicating that fermented mulberry juice by L. plantarum M3 has high biocompatibility and safety. It can be preliminarily concluded that fermented mulberry juice has the potential for in vivo research. At the same time, a concentration gradient of H₂O₂ was established to further explore the effect of H₂O₂ on cell viability. The results showed that when the H₂O₂ concentration was 0.3 mM, the survival rate of BV2 cells decreased to about 60%, and when it exceeded 0.3 mM, cell viability sharply decreased. Therefore, 0.3 mM was selected as the model concentration for subsequent cell experiments (Supplementary Figure S8c). To assess the effect of fermented mulberries on H₂O₂-induced BV2 cell survival, an oxidative stress experiment was conducted using BV2 cells, as shown in Supplementary Figure S8d. Compared with the blank control group, the BV2 cell survival rate was significantly reduced in the model group (p < 0.05), confirming that H₂O₂ successfully induced an oxidative stress damage model, leading to a decrease in cell viability. Compared with the model group, pre-treatment with fermented mulberries significantly increased the survival rate of H₂O₂-damaged BV2 cells (p < 0.05), indicating that fermented mulberries possess antioxidant stress protective effects. To further determine the inhibitory effects of BY and FMJ on oxidative stress induced by H₂O₂ in BV2 cells, the DCFH-DA probe was used to stain the cells, and the fluorescence intensity of ROS in the cells was detected by flow cytometry (Figure 5a). The fluorescence intensity of ROS in the H₂O₂ group was significantly higher than that in the control group, indicating the successful construction of the oxidative stress model. Compared with the H₂O₂ group, both BY and FMJ treatments reduced the fluorescence intensity of ROS, and FMJ had a better inhibitory effect than BY. The results showed that fermented mulberry juice by L. plantarum M3 significantly increased the antioxidant activity of MJ, and FMJ could resist oxidative stress at the cellular level and reduce cellular oxidative damage.

Figure 5.

(a) ROS levels of mulberry juice in H₂O₂-induced oxidative stress model of BV2 cells. (b–d) Validation of key antioxidant genes in FMJ by RT-qPCR. Using LSD and Duncan’s methods, a one-way analysis of variance was employed to compare differences between groups. (a–d) denote significant differences between different samples; p < 0.05.

3.5. Validation of Antioxidant Key Genes by RT-qPCR

Experimental validation via RT-qPCR provided evidence supporting the network pharmacology predictions regarding the potential antioxidant mechanisms of fermented mulberry juice (FMJ). Our results suggest that fermentation by L. plantarum M3 may enhance the protective effects of mulberry juice in a dose-dependent manner (Figure 5b–d). While H₂O₂ treatment was associated with suppressed Nfe2l2 expression, FMJ pretreatment—particularly in the Mulberry-H group—appeared to attenuate this downregulation (Figure 5b), potentially indicating that fermentation-derived metabolites may more effectively support the Nrf2 signaling axis. This observation was further linked to an increased expression of the downstream effector gene SOD1 (Figure 5c), which correlated with the observed reduction in intracellular ROS levels (Figure 5a), potentially contributing to strengthened cellular defense against oxidative stress. Furthermore, the observed decrease in Src gene expression (Figure 5d) in FMJ-treated groups suggests a possible dual-action mechanism; compared to the unfermented group (MJ), FMJ was associated with a more pronounced mitigation of the H₂O₂-induced elevation of Src, which may help in modulating oxidative stress-linked inflammatory cascades. Collectively, these findings provide molecular evidence, which still requires validation through direct protein-level functional assays, suggesting that Lactobacillus plantarum M3 fermentation can optimize the bioactivity of mulberry juice by simultaneously upregulating the Nfe2l2/SOD1 antioxidant defense and inhibiting pro-inflammatory Src signaling.

4. Conclusions

Research has established lactic acid bacteria (LAB) fermentation as a pivotal strategy for enhancing the processing quality of berries, including blueberries, cranberries, and blackberries. Functioning as an efficient biocatalytic platform, LAB drive the enzymatic bioconversion of macromolecular bound polyphenols into bioactive, low-molecular-weight free forms, thereby significantly bolstering radical scavenging capacity and bioavailability. This biotransformation mechanism not only optimizes the sensory profile of berry juice but also facilitates a qualitative leap from a conventional nutritional beverage to a high-activity functional food. To further unlock the nutritional potential of mulberry juice (MJ), this study systematically investigated the fermentation-mediated enhancement of its antioxidant properties through an integrated multi-omics approach. Using Lactobacillus plantarum M3-fermented MJ (FMJ) as the experimental model, we performed time-resolved metabolic profiling coupled with antioxidant capacity tracking. Correlation analysis and metabolomics revealed that the evolution of phenolic and flavonoid profiles during fermentation was strongly correlated with enhanced antioxidant potency. Potential mechanisms were predicted using network pharmacology and molecular docking, identifying Albanin A and Moracin Q as key bioactive constituents, while Nfe2l2, SOD1, and Src emerged as critical regulatory proteins. Furthermore, in a BV2 cell oxidative stress model, FMJ effectively inhibited H₂O₂-induced reactive oxygen species (ROS) generation. Experimental validation via RT-qPCR confirmed these predicted antioxidant targets, revealing that FMJ effectively alleviates oxidative damage possibly by activating the Nrf2 pathway and suppressing the NF-κB pathway, while concurrently upregulating the antioxidant enzyme SOD1. Collectively, this multi-dimensional analysis demonstrates that fermentation by L. plantarum M3 significantly amplifies the antioxidant capacity of MJ and mitigates oxidative stress. The significantly enhanced biological activity of fermented mulberry juice (FMJ) may be closely related to the biotransformation process driven by Lactobacillus plantarum M3, especially the conversion of bound polyphenols induced by enzymatic reactions to high-bioavailability free radical states.

This study systematically elucidates the metabolic evolution and biological potential of mulberry juice during L. plantarum fermentation by integrating non-targeted metabolomics with computational biology. While the findings demonstrate rigorous technical reproducibility, the inter-batch variability inherent in microbial processes warrants further assessment through multi-batch validation to ensure functional stability. Additionally, while this research highlights the holistic synergy of the fermented matrix, the independent contributions of pH reduction and organic acid remain to be experimentally decoupled. To profoundly elucidate the mechanism of action, subsequent research will establish a multi-batch fermentation model for mulberry juice. This will integrate protein-level validation with the analysis of oxidative stress markers, complemented by in vivo model validation. Such comprehensive validation is crucial for definitively clarifying the molecular regulatory mechanisms of fermented mulberry juice, thereby fully unlocking its application potential in functional foods and nutraceuticals.

Abbreviations

Acronym	Definition
MJ	Mulberry Juice (M0)
FMJ	Fermented Mulberry Juice
BY	Mulberry Juice
L. plantarum M3	Lactobacillus plantarum M3
LAB	Lactic Acid Bacteria
NFE2L2	NFE2 Like BZIP Transcription Factor 2
STAT3	Signal Transducer and Activator of Transcription 3
GSK3B	Glycogen Synthase Kinase 3 Beta
PRKCB	Protein Kinase C Beta
RELA	RELA Proto-Oncogene, NF-KB Subunit
SRC	SRC Proto-Oncogene, Non-Receptor Tyrosine Kinase
β-actin	Mus Musculus Actin, Beta
BV2 cells	Mouse Microglia Cells

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/foods15050906/s1, Figure S1: Antioxidant test standard curve; Figure S2: OPLS-DA analysis of all metabolites in the FMJ fermentation; Figure S3: OPLS-DA discriminant diagram; Figure S4: Volcanic diagram differential metabolite analysis during FMJ fermentation process; Figure S5: Volcanic diagram of FMJ fermentation process, and differential clustering heat of flavonoids and polyphenolic metabolites; Figure S6: Cluster heatmap analysis of flavonoids and polyphenolic metabolites during FMJ fermentation; Figure S7: Correlation analysis between phenolic and flavonoid compounds and antioxidant testing; Figure S8: H₂O₂-induced oxidative stress model in BV2 cells; Table S1: Metabolism of compounds during BY fermentation process; Table S2: PPI–protein interaction core targets; Table S3: Molecular docking of key antioxidants and antioxidant targets in FMJ.

Author Contributions

X.-S.Z.: Writing—Original Draft, Methodology, Conceptualization. S.-L.F. and J.-Y.Z.: Writing—Review and Editing. B.K. and Y.-N.D.: Writing—Review and Editing. L.Y. and L.S.: Writing—Review and Editing, Visualization, Methodology, Data Curation, Conceptualization. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Should any raw data files be needed in another format, they are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors have no conflicts of interest to declare.

Funding Statement

This work was supported by the Major Science and Technology Project of Xinjiang Uygur Autonomous Region, China (2023A02008-2).