Enhanced Protective Potential of a Pea Protein‐Fortified Fermented Almond Beverage Against Paraquat‐Induced Oxidative Stress

ABSTRACT

Oxidative stress, arising from an imbalance between reactive oxygen species and antioxidant defenses, drives numerous diseases. Paraquat (PQ) induces oxidative damage through free radical generation. Developing antioxidant‐rich products from natural sources may mitigate these effects. We evaluated the antioxidant content and protective effects of a plant‐based fermented beverage (PBFB) based on almond milk and plant protein. Methanolic extracts were analyzed for total phenolic (TPC) and flavonoid (TFC) content, as well as DPPH and ABTS assays. Fermentation enhanced antioxidant properties, increasing phenolic content (80.20 ± 0.81 vs. 119.60 ± 1.43 mg GAE/mg), total flavonoid content (9.32 ± 1.32 vs. 22.35 ± 1.52 mg CAE/mg), DPPH (4.97 ± 0.17% vs. 10.43 ± 0.15%), and ABTS (23.57 ± 1.46% vs. 44.30 ± 0.51%) radical scavenging capacity (p < 0.05). In vitro, PQ increased catalase (CAT) (16.36 ± 3.33 vs. 36.70 ± 5.52 units/min/mg, p < 0.05) and decreased glutathione S‐transferase (GST) activities (39.42 ± 2.92 vs. 18.27 ± 2.37 µmol/min/mL, p < 0.05) in CHO‐K1 cells. PBFB treatment partially restored GST activity (26.04 ± 2.30 µmol/min/mL, p < 0.05) and normalized CAT levels (21.25 ± 2.95 units/min/mg, p < 0.05). Cell viability remained unchanged. These findings highlight the synergistic effect of fermentation and plant protein enrichment, yielding a functional food with cytoprotective potential.

Fermenting pea protein‐fortified almond milk with probiotics increases its antioxidant content. In CHO‐K1 cells, the beverage enhanced protective enzymes, helping them cope with paraquat‐induced oxidative stress, highlighting its potential as a functional food that supports cellular defenses.

Abbreviations

ABTS

2,2′‐azino‐bis(3‐ethylbenzothiazoline‐6‐sulfonic acid)

BCA

Bicinchoninic acid assay

CAT

Catalase

CDNB

1‐chloro‐2,4‐dinitrobenzene

DPPH

2,2‐diphenyl‐1‐picrylhydrazyl

GSH

Reduced glutathione

GST

Glutathione S‐transferase

PBFB

Plant‐based fermented beverage

PQ

Paraquat

TFC

Total flavonoid content

TPC

Total phenolic content

1. Introduction

Reactive oxygen species (ROS) are inherent byproducts of normal aerobic metabolism. When not properly balanced, these species can react with proteins, lipids, and DNA, compromising cellular integrity. To prevent this damage, cells rely on endogenous enzymatic antioxidants (e.g., superoxide dismutase, catalase, glutathione‐related enzymes) and low‐molecular‐weight antioxidants (e.g., glutathione, melatonin, coenzyme Q10). Catalase (CAT) is a key antioxidant enzyme that decomposes hydrogen peroxide into oxygen and water, thereby limiting oxidative damage. Catalase downregulation promotes hydrogen peroxide (H₂O₂) accumulation and cancer progression [1, 2, 3]. Glutathione S‐transferases (GSTs) are phase II detoxification enzymes that catalyze the conjugation of reduced glutathione (GSH) to a wide range of xenobiotics, including herbicides. Reduced GST expression and genetic polymorphisms are associated with paraquat toxicity susceptibility and Parkinson's disease development [4, 5].

Oxidative stress, arising from an imbalance between free radicals and antioxidants, serves as a primary and secondary pathogenic mechanism in numerous diseases [6]. Herbicide exposure via dermal, oral, or respiratory routes induces biological damage largely through oxidative pathways [7, 8, 9]. Paraquat (PQ), a widely used herbicide, acts as a bipyridylium redox cycler that promotes superoxide anion formation. This is subsequently converted by superoxide dismutase (SOD) into H₂O₂ and hydroxyl radicals. PQ exposure is associated with renal and pulmonary injury, increased risk of Parkinson's disease, and lymphocyte damage [10, 11, 12, 13]. Due to the oxidative nature of PQ toxicity, antioxidant‐based strategies have gained increasing attention. However, because humans cannot synthesize key antioxidants de novo, compounds including bioactive peptides and phenolics must be obtained from food, highlighting the need to develop antioxidant‐enriched products [14, 15]. Almonds are a nutrient‐dense source of proteins, vitamins, minerals, and polyphenols, which are associated with managing diabetes, cardiovascular disease, and gut barrier dysfunction [16, 17, 18]. Regular almond consumption has been shown to improve lipid profiles, reduce inflammation in spinal cord injury or colitis, and mitigate drug‐induced hepatotoxicity [18, 19, 20, 21]. Thus, integrating almonds into plant‐based formulations significantly enhances their nutritional and functional value.

The growing demand for health‐promoting and sustainable diets has stimulated the development of functional foods based on plant‐derived ingredients. Functional foods provide physiological benefits beyond their basic nutritional value due to the presence of bioactive compounds such as phenolics, peptides, and antioxidants [22]. Within this context, plant‐based functional foods have gained increasing attention as potential alternatives to traditional animal‐derived protein sources, as plant proteins from legumes, cereals, seeds, and nuts provide essential amino acids and act as precursors of bioactive peptides generated during processing or fermentation [23]. In addition, plant‐derived matrices contain antioxidants and other compounds capable of modulating metabolic and physiological processes associated with oxidative stress and chronic diseases [14, 15, 16, 17]. Fermentation and probiotic incorporation may further enhance these functional properties by improving the bioavailability of bioactive compounds and promoting beneficial interactions with the gut microbiota, particularly when involving probiotic genera such as Lactobacillus, Lacticaseibacillus, and Bifidobacterium [24, 25]. Consequently, fermented plant‐based beverages have attracted growing attention due to their antioxidant potential and their alignment with ethical, environmental, and dietary demands [26, 27]. These beverages represent suitable alternatives for individuals with lactose intolerance, milk‐protein allergies, hypercholesterolemia, or vegan lifestyles [26, 27, 28, 29, 30], and their functional potential can be further enhanced by incorporating plant proteins such as pea protein, which act as substrates for the release of bioactive peptides during fermentation, improving the nutritional, functional, and antioxidant properties of the final product [31].

While research often focuses on isolated compounds, assessing the integrated biological activity of complex matrices like fermented beverages remains essential. Consequently, this study evaluated the antioxidant content of a novel almond‐milk and vegetable‐protein fermented beverage and its in vitro protective effects against PQ‐induced oxidative stress using the Chinese hamster ovary (CHO) cell line. This adherent epithelial‐like mammalian model was selected due to its well‐characterized sensitivity to xenobiotics and its reliability in assessing oxidative damage and cytoprotective mechanisms in response to environmental stressors such as PQ [32]. To our knowledge, this study is the first to demonstrate the protective effects of a legume protein‐enriched plant beverage against herbicide‐induced stress. Using an integrated biochemical and cellular approach, our findings reveal the protective potential of plant‐based fermented formulations and support their development as sustainable, non‐dairy functional foods for effective oxidative stress mitigation.

2. Materials and Methods

2.1. Sample Preparation

The formulation of the PBFB was standardized using 88.6% (w/w) commercial almond milk (10% almond, water, and sea salt; Green Food Makers) as the continuous liquid phase. This base was supplemented with 1.8% (w/w) pea protein isolate (VITESSENCE Pulse 1803, Ingredion) to enhance the nutritional profile and provide the necessary nitrogen source for lactic acid bacteria, while also serving as a substrate for the release of bioactive peptides. To ensure the desired texture and physical stability, 2.5% (w/w) modified corn starch (Snowflake 6420, Ingredion) was added as a thickening agent, and 0.1% (w/w) carob bean gum was included as a stabilizer to prevent syneresis. Additionally, 7.0% (w/w) sucrose was incorporated as a sweetener. The mixture was homogenized (150/50 bar at 60°C) and pasteurized (90°C for 20 min). After cooling, the base was inoculated with 0.02% (w/v) of the starter culture (F‐DVS YoFlex YF‐L02 DA, Novonesis), containing Streptococcus. thermophilus, L. bulgaricus, L. acidophilus, L. paracasei, and Bifidobacterium. Fermentation proceeded at 42°C for 5 h until a pH of 4.6 was achieved, resulting in a stable matrix with 2.4% protein and 3.9% fat (Figure 1).

FIGURE 1.

Flow diagram of the plant‐based fermented beverage process and sampling stages. Almond milk was mixed with starch, pea protein isolate, sugar, and gum, and then pasteurized. Next, the pasteurized base was fermented with F‐DVS YoFlex YF‐L02 DA (Novonesis) and cooled to obtain a plant‐based fermented beverage (PBFB). Samples from almond milk, pasteurized base, and the PBFB were taken for antioxidant content and capacity determinations and for in vitro assays.

2.2. Antioxidant Extraction at Different Stages of Production

Antioxidants were extracted following a procedure modified from Limón et al. [33]. Briefly, 10 g of almond milk, pasteurized base, or plant‐based fermented beverage (PBFB) samples were mixed with 20 mL of 80% (v/v) methanol, vortexed for 10 s (s), sonicated in a cold (4°C) ultrasonic bath for 20 min, and centrifuged at 13 000 rpm for 10 min at 4°C. The supernatant of each sample was filtered through a 0.45 µm nylon filter (25 mm diameter) using a 5 mL syringe, and aliquots were stored at −20°C until analysis.

2.3. Total Phenolic Content Estimation

Total phenolic content (TPC) was determined using the Folin–Ciocalteu (FC) method, adapted from the protocol described by Singleton and Rossi [34]. For this, 300 µL of gallic acid standard (1.5–24 µg/mL) or sample was mixed with 1.5 mL of FC reagent (F9252, Sigma–Aldrich; 1:10 in distilled water), vortexed, and 1.2 mL of a 75 g/L sodium carbonate (NaHCO₃) solution was added. After 2 h in the dark at room temperature, absorbance was measured at 765 nm against a reagent blank using a Shimadzu UV‐1280 UV–VIS Spectrophotometer (Shimadzu Corp., Kyoto, Japan). Results were expressed as milligrams of gallic acid equivalents (GAE) per milligram of sample (mg GAE/mg). All samples were evaluated in triplicate.

2.4. Total Flavonoid Content Estimation

The total flavonoid content (TFC) was quantified using the aluminum chloride colorimetric assay, following a modified version of the method described by Yan et al. [35]. In brief, an aliquot of 0.15 mL of the extract (or catechin standard, 0.2–2 mg/mL, Sigma–Aldrich C1251) was mixed with 0.6 mL of distilled water and vortexed for 15 s. At time zero, 0.045 mL of 5% (w/v) NaNO₂ was added and vortexed again. After a 5 min incubation, 0.045 mL of 10% (w/v) AlCl₃ was added to the mixture. After exactly 6 min, the reaction was stabilized by adding 0.3 mL of 1 M NaOH and 0.36 mL of distilled water, followed by a final vortex step (15 s). The absorbance was measured at 510 nm against a reagent blank using a Shimadzu UV‐1280 UV–VIS Spectrophotometer (Shimadzu Corp., Kyoto, Japan) equipped with microcuvettes. Results were expressed as mg of catechin equivalent per mg of sample (mg CAE/mg). All samples were evaluated in triplicate.

2.5. Antioxidant Capacity Determination

Radical scavenging activity was evaluated via 2,′‐azino‐bis(3‐ethylbenzothiazoline‐6‐sulfonic acid) (ABTS) and 2,2‐diphenyl‐1‐picrylhydrazyl (DPPH) assays, following adapted versions of the methods by Re et al. [36] and Brand‐Williams et al. [37], respectively. For the ABTS assay, the radical was generated 24 h prior by dissolving a 10 mg ABTS tablet (A9941, Sigma–Aldrich) in 2.6 mL distilled water, followed by the addition of 46 µL of 145 mM ammonium persulfate and incubation in the dark for 24 h. This solution was then diluted with 5 mM phosphate‐buffered saline (PBS) to an absorbance of 0.8 ± 0.02 at 734 nm. In a glass tube, 50 µL of sample (or methanol as a control) was mixed with 2 mL of the ABTS radical solution, vortexed, and incubated in the dark. Reduction of absorbance was monitored at 734 nm for 30 min. ABTS inhibition percentages were calculated as follows:

$$\begin{array}{ccc}& & \mathrm{Percentage}\mathrm{of}\mathrm{ABTS}\mathrm{inhibition}\left(\%\mathrm{Inh}\right)\hfill \\ & & =\left(\frac{{\mathrm{A}}_{\mathrm{control}}-{\mathrm{A}}_{\mathrm{sample}}}{{\mathrm{A}}_{\mathrm{control}}}\right)\times 100.\hfill \end{array}$$

For the DPPH assay, 0.025 g of DPPH (D9132, Sigma–Aldrich) was dissolved in 50 mL of methanol and adjusted to an absorbance of 0.8–0.9 at 517 nm. As in the ABTS assay, 50 µL of sample or methanol and 1.2 mL of DPPH radical solution were mixed in a glass tube, vortexed, and incubated for 30 min in the dark. The decrease in the absorbance was monitored at 517 nm for 60 min. DPPH inhibition percentages were calculated as follows:

$$\begin{array}{ccc}& & \mathrm{Percentage}\mathrm{of}\mathrm{DPPH}\mathrm{inhibition}\left(\%\mathrm{Inh}\right)\hfill \\ & & =\left(\frac{{\mathrm{A}}_{\mathrm{control}}-{\mathrm{A}}_{\mathrm{sample}}}{{\mathrm{A}}_{\mathrm{control}}}\right)\times 100.\hfill \end{array}$$

For both assays, all samples were evaluated in triplicate.

2.6. Cell Culture and Treatments

Chinese hamster ovary cells subclone K1 (CHO‐K1) were selected as a mammalian model for studying paraquat (PQ)‐induced oxidative stress, toxicity, and evaluating the protective efficacy of plant‐based extracts. This selection was based on their reproducible oxidative stress response to redox‐cycling challenges, high sensitivity, and well‐characterized profiles for PQ‐induced redox cycling and ROS generation. These characteristics make them a reliable platform for investigating oxidative stress mechanisms, screening xenobiotic‐induced damage, and assessing the modulatory effects of food‐derived antioxidants on cell viability and antioxidant enzyme activities [32, 38, 39, 40, 41, 42]. Cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM, 12500096, Invitrogen) supplemented with 10% (v/v) fetal bovine serum (FBS, Natocor) and penicillin/streptomycin (15240062, Gibco). At ∼70% confluency, cells were trypsinized (0.05% (v/v) trypsin–ethylenediaminetetraacetic acid [EDTA], 15400054, Gibco), counted, seeded at 1 × 10⁵ cells/well in 24‐well plates, and grown for 24 h at 37°C under 5% carbon dioxide (CO₂).

To evaluate the cytoprotective effect, cells were distributed into four experimental groups: Control, 100 µM PQ, PBFB, and 100 µM PQ + PBFB. For the PBFB and Control groups, cells were initially incubated in fresh culture medium for 24 h. For the 100 µM PQ and 100 µM PQ + PBFB groups, oxidative stress was induced with 100 µM PQ (36541, Sigma) in culture medium for the same period. Subsequently, the medium in all groups was replaced for a final 2 h incubation as follows: the Control and 100 µM PQ groups received fresh culture medium, while the PBFB and 100 µM PQ + PBFB groups received a 1:10 dilution of the PBFB methanolic extract in culture medium. The extract concentration was selected as the highest non‐cytotoxic dose based on preliminary dose‐response assays (Figure S1). After the treatment period, cell viability and enzymatic activity were immediately assessed. All treatments were performed in triplicate across three independent biological experiments (n = 3).

2.7. Protein Isolation

After treatments, cells were washed with 1X PBS, harvested with 0.05% trypsin‐EDTA for 2 min at 37°C and 5% CO₂, and inactivated with culture medium. Cells were centrifuged (1000 rpm, 5 min) and resuspended in 1 mL cold hypotonic lysis buffer (10 mM 2‐[4‐(2‐Hydroxyethyl)piperazin‐1‐yl]ethane‐1‐sulfonic acid [HEPES] pH 7.9, 1.5 mM magnesium chloride [MgCl₂], 10 mM potassium chloride [KCl], 0.5 mM dithiothreitol [DTT], 0.2 mM phenylmethylsulfonyl fluoride [PMSF]; 5 × packed cell volume), incubated on ice 15 min, and centrifuged (420 g, 5 min, 4°C). The pellet was resuspended in 400 µL lysis buffer (2 × packed cell volume), lysed by five passages through a 27‐gauge needle, and checked microscopically. Lysates were centrifuged for 20 min at 10000 g, 4°C; 75 µL of the supernatant was used for protein determination by the bicinchoninic acid assay (Pierce BCA Protein Assay Kit, 23227, Thermo Scientific) and the remainder for enzyme assays. Aliquots were stored at −80°C.

2.8. Catalase Activity Assay

The catalase activity was determined by spectrophotometric detection of H₂O₂ breakdown by catalase [43, 44, 45]. Briefly, 10 µg of protein extract was mixed with a solution of 19 mM H₂O₂ in 50 mM potassium phosphate buffer (PKB), up to a final volume of 1 mL. The solution was read at 240 nm for 60 s in a quartz cuvette, employing a spectrophotometer connected to a thermostatic bath that pumped water at 25°C. This ensured that the optimum enzymatic activity remained constant throughout the reading. Blank consisted of 50 mM PKB. Specific activity values for samples were determined with the following formula:

where ΔAmin_{240 nm} is the difference between the absorbance at time (t) = 0 s and t = 60 s, ɛ _H2O2 is the molar extinction coefficient of H₂O₂ (43.6 M⁻¹ cm⁻1), and Vr is the reaction volume. Results were expressed as units of enzyme per minute per milligram of total protein (U/min/mg).

2.9. Glutathione‐S‐transferase (GST) Activity Assay

Glutathione S‐transferase (GST) activity was analyzed by the CDNB‐GSH method [46]. The reaction mixture contained 1.96 mL of 0.1 M PKB, 20 ul of 100 mM 1‐chloro‐2,4‐dinitrobenzene ethanolic solution (CDNB, 138630, Sigma–Aldrich), and 20 uL of 100 mM reduced glutathione (GSH, G4251, Sigma–Aldrich) dissolved in distilled water. Then, 1.80 mL of this solution was mixed with 200 uL of protein extract in a quartz cuvette, and absorbance at 340 nm was recorded every minute for 5 min using the same equipment as for the catalase assay. Blank consisted of PKB. Total GST activity was calculated using the following formula:

$$\begin{array}{cc}& \text{GST activity}(\mu \text{mol/min/mL})\hfill \\ & =\frac{\mathrm{\Delta }{\mathrm{A}}_{340\text{nm}}/\min \times Vr\times \text{dilution factor}}{{\epsilon }_{\text{CDNB}}\times Vs}\hfill \end{array}$$

where ΔA _340 nm is the difference between the absorbance at time (t) = 0 s and t = 300 s, Vr is the reaction volume (2 mL), ɛ _CDNB is the molar extinction coefficient of CDNB (0.0096 µM⁻¹ cm⁻¹), and Vs is the sample volume (0.2 mL). Results were expressed as units of micromole (µmol) of GS‐CDNB conjugate per minute per milliliter (µmol/min/mL).

2.10. Cell Viability Assessment

Cell viability was determined using the Trypan Blue Exclusion Assay. For this, 100,000 cells per well were plated in a 24‐well plate and incubated for 24 h with supplemented DMEM, 100 uM PQ, 1:10 dilution (in culture medium) of the beverage methanolic extract, or 100 uM PQ and beverage methanolic extract (for 2 h). Cells were then harvested with 0.05% trypsin‐EDTA, resuspended in fresh DMEM, and mixed 1:1 with 0.4% Trypan Blue (15250061, Gibco) for 2 min. An aliquot (10 µL) was counted in a Neubauer chamber to determine live (unstained) and dead (blue) cells, and viability was expressed as the percentage of viable cells per milliliter.

2.11. Statistical Analysis

Values are expressed as mean ± SEM. Statistical analysis was performed using a one‐way analysis of variance (ANOVA) followed by Tukey's multiple‐comparison post hoc test. GraphPad Prism Version 8 (GraphPad Software Inc., San Diego, CA, USA) was used for statistical analysis. Differences were considered significant when the probability value was less than 0.05 (p < 0.05). All the experiments were performed in triplicate.

3. Results

3.1. Antioxidant Molecules Content and Capacity of the Plant‐Based Beverage

Methanolic extracts of each stage of the product manufacturing were assayed with the Folin–Ciocalteu reagent to estimate the relative content of phenolic antioxidants. Significant differences were observed in TPC levels among the samples. The pasteurized vegetable base exhibited the highest content (129.80 ± 1.77 mg GAE/mg), representing a 1.62‐fold increase relative to the almond milk (Figure 2A). Moreover, the phenolic content in PBFB (119.60 ± 1.43 mg GAE/mg) was significantly higher than in almond milk (80.20 ± 0.81 mg GAE/mg, p < 0.05), though slightly lower than in the pasteurized base. Total flavonoid content (TFC) was also determined across the different manufacturing stages. Consistent with the TPC results, statistically significant differences were identified among the samples. The pasteurized vegetable base exhibited the highest flavonoid concentration (33.11 ± 1.36 mg CAE/mg), representing a 3.55‐fold increase relative to almond milk (Figure 2B). Notably, the flavonoid content in PBFB (22.35 ± 1.52 mg CAE/mg) was significantly higher than in almond milk (9.32 ± 1.32 mg CAE/mg, p < 0.05), following a similar pattern to the one observed for total phenolics.

FIGURE 2.

Antioxidant content and capacity of the different stages of a plant‐based fermented beverage (PBFB). (A) Total phenolic content (TPC) measured by the Folin–Ciocalteu (FC) method. (B) Total flavonoid content (TFC) determined by the aluminum chloride colorimetric assay. (C) Antioxidant activity measured by 2,2‐diphenyl‐1‐picrylhydrazyl (DPPH) test after 60 min. (D) Antioxidant activity measured by 2,2'‐azino‐bis(3‐ethylbenzothiazoline‐6‐sulfonic acid) (ABTS) test after 60 min. All analyses were performed on methanolic extracts of the samples at each stage. Bars represent mean ± SEM. Different letters indicate significant differences among groups (p < 0.05) determined by ANOVA followed by Tukey's Multiple Comparison test.

Antioxidant potential was evaluated with the DPPH and ABTS assays. Regarding DPPH, almond milk showed significantly higher inhibition than the pasteurized base (4.97 ± 0.17% vs. 2.73 ± 0.26%), while fermentation nearly doubled this activity (10.43 ± 0.15%), representing the highest capacity among samples (p < 0.05; Figure 2C). ABTS results followed the same pattern as TPC trends (Figure 2D), with the pasteurized base (47.00 ± 4.68%) and fermented product (44.30 ± 0.51%) exhibiting significantly higher inhibition than almond milk (23.57 ± 1.46%, p < 0.05) yet showing no significant difference between them.

3.2. Assessment of Viability in Paraquat and PBFB‐Treated Cells

Cell viability was assessed with the Trypan blue exclusion assay to evaluate the effects of the different treatments. Incubation with 100 µM PQ did not significantly affect cell survival, as the percentage of cell viability remained unchanged compared to Control (Figure 3A). Furthermore, neither PBFB treatment nor sequential incubation with 100 µM PQ followed by the antioxidant extract significantly affected cell viability compared to the other groups.

FIGURE 3.

Effects of a plant‐based fermented beverage (PBFB) on CHO‐K1 cells under 100 µM paraquat (PQ)‐induced oxidative stress. (A) Cell viability was estimated by the Trypan blue exclusion assay, as a percentage of live cells. (B) CAT activity (U/min/mg total protein) estimated by spectrophotometric measurement of H₂O₂ decomposition over time at 240 nm. (C) GST activity (µmol/min/ml) estimated by spectrophotometric measurement of the conjugation rate of CDNB with GSH at 340 nm. Bars represent mean ± SEM. Different letters indicate significant differences among groups, with p < 0.05 determined by ANOVA followed by Tukey's Multiple Comparison test.

3.3. Catalase Activity Modulation by the Beverage Antioxidants in Paraquat‐Induced Oxidative Stress

Catalase (CAT) activity was spectrophotometrically measured in cytosolic fractions from CHO‐K1 cells exposed to 100 µM PQ, a methanolic extract of the beverage, or a sequential treatment with PQ followed by the extract. Treatment with PQ alone significantly increased CAT activity compared to the control group (p < 0.05; Figure 3B). Cells incubated with the beverage extract showed no differences in CAT activity relative to the control. Notably, post‐treatment with PBFB after PQ exposure led to a partial recovery of CAT activity, with values significantly higher than those observed in the PQ‐treated group, tending towards baseline control levels (p < 0.05; Figure 3B).

3.4. Effects of the Beverage Extract on GST in Cells Under Oxidative Stress

GST activity in cytosolic extracts of treated CHO‐K1 cells was measured spectrophotometrically by GSH–CDNB conjugation. PQ exposure significantly reduced GST activity by ∼50% versus control (p < 0.05; Figure 3C), while PBFB alone had no effect. Sequential PQ‐PBFB treatment also decreased GST activity, but to intermediate levels between PQ‐ and PBFB‐treated cells (p < 0.05; Figure 3C).

4. Discussion

The food industry is increasingly addressing specialized dietary requirements, including lactose‐free, low‐cholesterol, and vegan diets. Plant‐based foods are key alternatives to animal products, providing antioxidants that reduce ageing and disease development. Fermentation further enhances the antioxidant capacity of matrices such as oat, kidney bean, and soybean [33, 47, 48, 49, 50]. Accordingly, this study evaluates the impact of probiotic addition to an almond‐based fermented beverage on its antioxidant content and cellular response under oxidative stress.

The observed higher TPC values in the PBFB compared to the non‐fermented almond milk are consistent with reports on fermented walnut and cashew products, where fermentation facilitated phenolics release or transformation [51, 52]. As a nutrient‐dense substrate, almonds facilitate fermentation‐induced enrichment via microbial metabolism and enzymatic cell wall degradation, which release simple phenolics and oligosaccharides [53, 54]. Additionally, LAB can cleave glucose moieties from glycosylated polyphenols, generating free aglycones with superior antioxidant activity, potentially strengthening cellular defenses against oxidative stress [55, 56, 57]. A similar mechanism may underlie the changes observed in total flavonoid content (TFC), as flavonoids represent a major subclass of phenolic compounds and are also subject to fermentation‐driven transformations. In almond‐ and legume‐based fermented systems, lactic acid bacteria produce enzymes such as β‐glucosidases and esterases that hydrolyze flavonoid glycosides, releasing bound forms but also converting them into aglycones or other derivatives, thereby modifying their chemical structure and analytical detectability [58, 59]. Experimental studies in fermented plant‐based beverages, including soy and almond matrices, have shown that flavonoid content may increase due to enhanced release at early stages and subsequently decrease or stabilize as a result of microbial biotransformation and metabolic activity [50, 60]. Therefore, the changes observed in TFC are consistent with the dynamic nature of flavonoid metabolism during fermentation, rather than reflecting a simple accumulation of these compounds.

The PBFB exhibited the highest DPPH scavenging activity, consistent with reports of increased DPPH inhibition and TPC in almond milk fermented with Bifidobacterium strains, L. plantarum, or L. rhamnosus [55, 56]. These findings support probiotic supplementation as an effective strategy to enhance antioxidant capacity in plant‐based foods. In our study, the discrepancy between DPPH and ABTS results likely reflects differences in compound solubility. DPPH preferentially detects lipophilic antioxidants in organic solvents, whereas ABTS is more sensitive to hydrophilic compounds in aqueous media. Accordingly, utilizing both assays provides a comprehensive assessment of antioxidant potential in complex extracts [57].

Phenolic stability under thermal treatment varies, as some compounds degrade while others become more bioavailable, affecting the antioxidant capacity. Pasteurization may increase TPC in plant‐based products by releasing phenolics through cell wall softening [61, 62], particularly hydrophilic compounds such as phenolic acids and flavonoids [63]. Hydrophobic phenolics may require stronger processing or emulsification, potentially explaining the reduced DPPH inhibition and increased ABTS values observed in the pasteurized base relative to almond milk.

Legume protein isolates, such as pea protein, enhance the protein content, rheological properties, and antioxidant capacity of fermented matrices through naturally occurring flavonoids and phenolic acids [64, 65]. Pea protein incorporation improves antioxidant stability in fermented dairy products and increases TPC and antioxidant activity in plant‐based matrices, including oat milk and chickpea‐based yogurts. These findings highlight legume‐derived isolates as functional ingredients that enhance nutritional, physicochemical, and cellular antioxidant profiles [66, 67, 68, 69, 70]. Bioactive peptides (2–20 amino acids) released during efficient protein hydrolysis by LAB in plant substrates exert antioxidant, antihypertensive, and immunomodulatory effects [71]. Specifically, LAB fermented almond and soy milks have demonstrated protective effects against oxidative DNA damage and inflammation in cellular models [72, 73]. These results support the use of pea protein to improve the functional value of plant‐based fermented yogurts. Our results suggest that vegetable‐derived antioxidants exert protective effects against herbicide‐induced toxicity. Studies in rats have shown that soy protein‐supplemented diets reduced paraquat‐induced damage, including lung enlargement, body weight loss, and lipid and glutathione oxidation [73, 74]. Overall, the functional properties of non‐dairy fermented products represent an emerging value for people with health limitations, and a promising solution for PQ‐induced damage.

PQ toxicity is mediated by redox cycling and superoxide production, requiring coordinated neutralization by SOD and CAT enzymes, an interplay well‐documented in CHO cells [41] and rats [74]. Deficiency in antioxidant enzyme expression, or activity might compromise the cellular response to PQ‐induced ROS overproduction. In this regard, CAT inhibition by aminotriazole in human fibroblasts (Hs27) increased intracellular H₂O₂ levels and macromolecular oxidative damage [75]. Similarly, CAT suppression by RNAi or chemical inhibition in Drosophila increased oxidative stress and sensitivity to PQ toxicity [76]. Conversely, high catalase expression and activity have been reported in mouse keratinocytes [77] and PC12 cells [78] following PQ exposure. In the present study, CAT activity increased while GST activity decreased, consistent with findings in rat hippocampus after intraperitoneal PQ exposure [79]. PQ also reduced GST activity in heart tissue [80] and rat leukocytes [81], including CDNB‐related activity in polymorphonuclear cells. As in our study, total GST activity was assessed, whereas isoform‐specific analyses yielded divergent results, which the authors attributed to the contribution of GST isoforms with higher affinity for CDNB, such as GST‐pi. In contrast, increased GST activity has been reported in Vero and SH‐SY5Y cells [82, 83]. As GSTs are not first responders to ROS, this pattern may reflect an early catalase‐mediated response with concurrent GST isoform shifting or oxidative inactivation, which can be prevented by catalase [4, 84]. In vivo, cardiac‐specific catalase overexpression attenuated PQ‐induced cardiac dysfunction and apoptosis [85] but did not extend lifespan in mice [86] or Drosophila [87]. Together, these findings suggest that PQ sensitivity depends on CAT levels and is cell‐ and tissue‐specific.

The plant‐based fermented beverage evaluated in this study exhibited high antioxidant potential and modulated the enzymatic response to paraquat‐induced oxidative stress in CHO‐K1 cells. It contributed to CAT normalization and partial GST recovery without affecting cell viability, suggesting protective effects associated with fermentation‐enhanced antioxidant capacity. To evaluate the integrated biological response of the fermented beverage extract, global antioxidant assays (TPC, TFC, DPPH, ABTS), widely accepted for complex food matrices, were employed, and the extract was analyzed as a whole. Future studies incorporating chromatographic profiling, additional paraquat‐sensitive cell types, and in vivo models would strengthen the translational relevance of these findings. Nevertheless, this work provides an initial framework for investigating food‐derived antioxidants in functional fermented matrices under oxidative stress conditions.

5. Conclusion

This work demonstrates that the fermentation of an almond‐based matrix supplemented with vegetable proteins enhances its antioxidant potential and modulates key detoxifying enzymes (CAT and GST) in cells exposed to paraquat‐induced oxidative stress. These findings support the potential of plant‐based fermented beverages as functional food candidates capable of partially mitigating cellular responses to oxidative challenges. By linking nutritional innovation with cellular protection mechanisms, this work contributes to a broader understanding of how plant‐based biotechnology can promote health resilience through natural antioxidant pathways.

Author Contributions

S.E.C. conceived and designed the study, curated and analyzed the data, obtained funding, conducted the investigation, developed the methodology, managed the project, provided resources, supervised and validated the experiments, prepared the visualizations, and wrote and revised the manuscript. M.S.G. contributed to the conceptualization and design of the study, curated and analyzed the data, secured funding, performed the investigation, developed the methodology, managed the project, provided resources, supervised and validated the experiments, prepared the visualizations, and participated in writing and revising the manuscript. B.E.B. curated and analyzed the data. G.R. curated and analyzed the data. V.B.D. contributed to the writing and critical revision of the manuscript.

Funding

This work was supported by Universidad Argentina de la Empresa (UADE) (Projects P21T05: Evaluation of the antioxidant capacity of a plant‐based milk ferment in eukaryotic models and A19T07: Development of prototypes of non‐dairy fermented foods).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: mnfr70519‐sup‐0001‐SuppMat.tif.

Data Availability Statement

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.