Developing bioactive-rich hybrid beverages: A metabolomic study of grape must co-fermentation with orange and apple juices☆ Authors: S Parra-Cadenas, MC Díaz-Maroto, MS Pérez-Coello, PM Izquierdo-Cañas, E García-Romero Published: 2026 ABSTRACT AbstractThe bioactive profile of hybrid beverages produced by co-fermentation of white grape must with orange and/or apple juices was investigated, focusing on organic acids, polyphenols, amino acids and biogenic amines. Two strategies were tested, sequential inoculation with Torulaspora delbrueckii and Saccharomyces cerevisiae, and single inoculation with S. cerevisiae. Hybrid beverages showed higher levels of organic acids, mainly citric, ascorbic, and isocitric (orange) and malic and fumaric (apple), enhancing freshness, stability, and health-related properties. Acetic acid remained below spoilage thresholds (<0.36 g/L) regardless of yeast strategy. Amino acids were lower than in control wines, suggesting improved microbial stability. Histamine and putrescine stayed within safe limits, with variations by fruit and yeast. Orange-based hybrids contained more total polyphenols (between 335.43 and 420.61 mg GAE/L), flavonols, and antioxidant activity, while apple-based hybrids were enriched in flavan-3-ols and chlorogenic acid. Overall, fruit co-fermentation generated stable, bioactive-rich and lower-alcohol beverages with functional potential.Keywords: Co-fermentation, Metabolomics, Polyphenols, Yest-co-culture, Fruit wine Introduction The growing demand for natural, healthy, and sustainable products has driven innovation across the agri-food sector, including the wine industry (Tymchenko, 2025). One such innovation is the co-fermentation of grapes with other fruits, a still scarcely explored practice that offers numerous benefits and opportunities. By combining the chemical compounds of grapes and other fruits, this technique creates hybrid beverages or co-ferments with unique sensory profiles and enhanced nutritional diversity (Fracassetti et al., 2019; Martín-Gómez et al., 2021). Moreover, the use of non-viticultural fruits helps repurpose agricultural surpluses, seasonal or local produce, and less marketable fruit, supporting a circular economy by minimizing waste and maximizing raw material utilization. Additionally, this approach can reduce the use of additives and adapt to emerging markets, promoting more sustainable and healthier products. Recent research on functional fermented matrices (Aher et al., 2025; Asadzadeh et al., 2021), has highlighted the potential of fermentation to enhance bioactive compounds, however, fruit-based co-fermentations that integrate the use of different oenological yeasts remain largely unexplored. When co-fermenting grapes with other fruits, there are several important parameters to control, including the type of fruit and the fermentation conditions. In the first case, fruit selection is essential, considering aspects such as its chemical composition and availability. In Spain, two of the most widely produced fruits are oranges and apples. Oranges lead the citrus category, with an estimated production of 2.975 million tons for the 2024–2025 season, while apples are prominent among pome fruits, with a production of 527,680 tons in the 2023–2024 season. The chemical composition of fruit plays a crucial role in fermentation dynamics, as well as in the chemical profile and stability of the final product. Key factors to consider include sugar content, total acidity, and pH, as they influence the balance between alcohol and acidity. Both, oranges and apples are naturally high in acidity and low in sugar, which can help offset grape ripening challenges caused by climate change. Additionally, their high concentration of phenolic compounds provides strong antioxidant and antimicrobial properties (Altunayar-Unsalan et al., 2022; Díaz-García et al., 2013). As a result, co-fermenting grapes with oranges or apples can enhance antioxidant compounds levels (such as vitamin C and flavonoids) and may contribute to health benefits, such as reducing oxidative stress. Regarding fermentation conditions, controlling temperature and selecting the appropriate yeast species are among the most critical factors influencing the quality of the final product. Temperature regulation directly affects yeast metabolism, fermentation rate, and the development of desirable sensory profiles (Izquierdo-Cañas et al., 2020). Equally important is the choice of yeast species, as different strains can significantly alter the chemical composition of the wine. In wine industry, traditional wine yeasts such as Saccharomyces cerevisiae are commonly used, but there is growing interest in non-Saccharomyces strains for their potential to improve specific characteristics. Recent studies with yeasts like Torulaspora delbrueckii, Lachancea thermotolerans, and Metschnikowia pulcherrima have demonstrated their ability to reduce alcohol levels and pH, and enhance aromatic complexity, making them valuable tools for modern winemaking (Izquierdo-Cañas et al., 2025; Martín-Gómez et al., 2021; Varela et al., 2017). In particular, T. delbrueckii is associated with lower ethanol yields and positive modulation of fruity and floral aroma compounds, but it is often unable to complete fermentation on its own. Therefore, its use in sequential inoculation followed by S. cerevisiae allows exploitation of its metabolic traits while ensuring fermentation completion and microbiological stability, representing an innovative strategy for developing low-alcohol, bioactive fruit-based beverages. The objective of this study was to investigate the effects of different juice blends and fermentation strategies on the organic acids, phenolic profiles and nitrogen compounds of white hybrid beverages, produced by co-fermenting grape must with orange and/or apple juices. The fermentation process involved two approaches: simple inoculation with S. cerevisiae and sequential inoculation with T. delbrueckii followed by S. cerevisiae. This work therefore introduces a novel fermentation design that integrates mixed-fruit matrices with oenological yeast succession to modulate alcohol content, chemical stability, and bioactive compound retention in hybrid beverages. Balancing chemical composition and fermentation conditions is fundamental for achieving an appropriate co-fermentation that enhances both complexity and nutritional diversity, while maintaining product stability. Materials and methods Samples Grape must was obtained from Vitis vinifera L. Cv. Cayetana white grapes at the “Centro de Investigación de la Vid y el Vino de Castilla-La Mancha” (IVICAM). The grapes were harvested at optimal maturity, destemmed, crushed, and pressed to extract the must, which was treated with 3 g of SO₂ per 100 kg of grapes and 30 g of glutathione per hL of must. A pectolytic enzyme (Lallizyme, Lallemand Inc., Montreal, Canada) was then added at a dosage of 0.5 g/hL. The orange juice was obtained from Citrus sinensis Navelina oranges using an automatic juicer. The juice was protected with 20 mg/L of SO₂. Apple juice was prepared from Granny Smith apples, that were washed with sulfited water (40 mg/L), dried, and cut before juice extraction using an Amzchef Slow Juicer SJ-036 (Amzchef Co. Ltd., New York, USA). The apple juice was treated with 0.2 g/hL of Lallizyme and 40 mg/L of SO₂. All juices, including the grape must, were desludged under static conditions at 5 °C for 24 h. Once prepared, the juices were mixed in the following proportions: 1:1 for must/apple and must/orange, and 2:1:1 for must/apple/orange blends. Control wines (grape must only) and co-ferments were obtained under two conditions: (1) using S. cerevisiae (SC) (Safoeno UCLM S235, Lesaffre Ibérica, Valladolid, Spain), and (2) through sequential inoculation with T. delbrueckii (TD) (Zymaflore Alpha TD, Laffort, France) followed by S. cerevisiae. Both yeasts were obtained as commercial preparations and were handled according to the supplier's instructions for rehydration and inoculation. Active dry cultures were first reconstituted in water at 25–30 °C. In the single-inoculum fermentations, S. cerevisiae was introduced at a rate of 20 g/hL. In the sequential fermentations, T. delbrueckii was initially added at 30 g/hL, and after 72 h of fermentation, S. cerevisiae was introduced at 20 g/hL to complete the process. Alcoholic fermentation was carried out at 18 °C in 3 L vessels, monitored daily by weight loss, and deemed complete when the weight stabilized. Control wines and co-ferments were then decanted at 18 °C, and glucose and fructose concentrations were enzymatically determined. All samples were subsequently filtered through 0.45 μm membranes and bottled. The experimental samples were identified using the following coding system: CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. Physicochemical analysis Official analytical methods established by the International Vine and Wine Organization (OIV) were used to determine conventional parameters as ° Brix, pH, total acidity, yeast assimilable nitrogen (YAN), residual sugars (glucose and fructose (G + F)), alcoholic graduation, glycerol and volatile acidity. Total phenolic content (TPC) and CIELAB parameters TPC in wines was determined by the Folin-Ciocalteau assay. An UV–Vis spectrophotometer (Helios, Thermo Spectronic, Cambridge, UK) was used to measure the absorbance at 765 nm. Results were expressed as mg of gallic acid equivalent per liter (mg GAE/L). Chromatic characteristics were determined by following the OIV official method according to CIELAB parameters. An Agilent 8453 diode array spectrophotometer (Agilent Technologies, Santa Clara CA, USA) was used with a home-made program for spectra treatment. Measurements were made at 420, 450, 520, 570, 620 and 630 nm. Antioxidant capacity The antioxidant capacity of control wines and co-ferments was evaluated by the DPPH assay, which relies on the stable free radical 1,1-diphenyl-2-picrylhydrazyl (DPPH) to measure the ability of antioxidants to donate hydrogen atoms or electrons, leading to a decrease in absorbance at 517 nm (Brand-Williams et al., 1995). The results were expressed as mg of gallic acid equivalents per milliliter (mg GAE/mL). Organic acid analysis Organic acids as malic, citric, tartaric, succinic and lactic, were determined using the official analytical method established by the OIV. Other organic acids as acetic, quinic, isocitric, ascorbic, fumaric and shikimic as well as proline were first isolated by solid phase extraction (SPE) in Sep-Pack Plus C18 cartridges (Sep-Pack Plus C18 Short Cartridge, 360 mg; 55–105 μm; Waters Co., Milford, Massachusetts, USA). These cartridges allowed the elimination of phenolic compounds. Then, an HPLC Agilent 1200 series system equipped with a diode array detector (DAD, Agilent Technologies, Inc., Santa Clara, California, USA) coupled to a Hypersil GOLD column (150 mm × 4.6 mm, 3 μm particle; Thermo Fisher Scientific, Waltham, Massachusetts, USA) connected to a Luna C18 (250 mm × 4.6 mm, 3 μm particle, Phenomenex, Torrance, California, USA) at room temperature (25 °C) was used. The analysis was achieved with an injection volume of 10 μL and a flow rate of 0.4 mL/min. Phenolic compounds analysis Flavan-3-ols and stilbenes were analyzed by a HPLC Agilent 1200 series system equipped with a diode array detector (DAD, Agilent Technologies) and coupled to a mass spectrometry system AB Sciex 3200 Q TRAP (Applied Biosystems, Foster City, California, USA) with an hybrid triple quadrupole-linear ion trap (LIP) designed for qualitative and quantitative purposes and a turbo spray ionization (electrospray assisted by a thermonebulization). Chromatographic separation was achieved on an Ascentis-C18 column (4.6 mm × 150 mm; 2.7 μm particle; Supelco, Deisenhofen, Germany), thermostated at 16 °C, with a flow rate of 0.3 mL/min and with an initial pressure of 127 bar. Injection volume was 20 μL. The analysis of flavonols, hydroxycinnamic acids and benzoic acids was carried on the same previously described chromatographic system for flavan-3-ols. An Agilent Eclipse XDB reversed-phase column (250 mm × 4.6 mm; 5 μm particle; Agilent Technologies) thermostatized at 40 °C was used. The separation was achieved by using a ternary mobile phase, with an injection volume of 40 μL and a flow rate of 0.63 mL/min (Castillo-Muñoz et al., 2009). The analyses were carried out in the negative Multiple Reaction Monitoring (MRM) ion mode. Electrospray source parameters were, curtain gas (CUR) 15.0 psi, collision gas (CAD) high, ion spray voltage (IS) -4000 V, and source temperature (TEM) 425 °C. The identification was mainly based on spectroscopic data (UV–Vis and tandem mass spectrometry (MS/MS)), obtained from authentic standards or previously reported. Quantitative analyses were performed using the external standard method, and calibration curves were realized over the range 0.05–50 mg/L. The standards used for the calibration curves were (+)-catechin for flavan-3-ols, and trans-resveratrol-glucoside for stilbenes, both from Sigma Aldrich. Flavonol concentrations were expressed as equivalents of rutin (Extrasynthese, Genay, Auvergne-Rhône-Alpes, France, (mg/L)), hydroxycinnamic acids were expressed as equivalents of caffeic acid (Extrasynthese (mg/L)) and benzoic acids were expressed as equivalents of gallic acid (Sigma Aldrich Chemie, Deisenhofen, Germany). Aminoacids and biogenic amines analysis The presence of aminoacids and biogenic amines was determined by the method described by Gómez-Alonso et al. (2007), with some modifications. First, the samples were derivatized by mixing 1 mL of each wine with 1.75 mL of 1 M borate buffer (pH = 9), 750 μL of methanol and 30 μL of diethylethoxymethylenemalonate (DEEMM) in a screw cap test tube for 30 min in an ultrasound bath. Then, the mixture temperature was elevated to 70 °C and maintained for 2 h to achieve the complete degradation of the excess of DEEMM and reagent by-products. After derivatization, an HPLC equipment with a diode array detector (Agilent, Model 1110; Agilent Technologies) was used for analysis. Chromatographic separation was performed on an ACE HPLC column (5 C18-HL), particle size of 5 μm (250 mm × 2.1 mm), with three eluents (phase A: 25 mM acetate buffer, pH = 5.8 with 0.02% sodium azide; phase B: methanol; phase C: acetonitrile) and a flow rate of 0.9 mL/min. Lastly, a photodiode array detector was used monitoring at 280 and 269 nm. Compounds of interest were identified and quantified using the corresponding standards (Sigma-Aldrich Chemie, Deisenhofen, Germany). Statistical analysis Statistical analysis was achieved with SPSS software 24.0. A Student-Newman-Keuls test (SNK) was used to find out if significant differences existed between samples. Results and discussion Prior to a detailed analysis of organic acid, phenolic compounds and amino acid profiles, it is important to highlight those conventional oenological parameters of the juices and fermented beverages that showed expected variations depending on the type of fruit added. In previous studies using the same base formulations (Parra-Cadenas et al., 2026), the incorporation of apple and/or orange juice increased total acidity and decreased both the pH and the initial fermentable sugar content of the grape must, as the objective was to develop juice blends suitable for high-quality co-fermentation. Blending grape must with orange and apple juices adjusted the pH to 3.2 across all mixtures. Total acidity and sugar content varied: the grape-apple blend had moderate acidity and sugar (5.2 g/L total acidity and 15.3°Brix), the grape-orange blend had higher acidity and sugar (8.2 g/L and 15.8°Brix), and the triple blend showed more balanced values (6.5 g/L and 15.2°Brix). These physicochemical adjustments were not only aimed at optimizing fermentation performance, but also at achieving beverages with sensory profiles acceptable to consumers. Apple juice is commonly used in fermented drinks and is generally associated with fruity and sweet aromatic notes, while orange juice provides a distinctive citrus character that can enhance freshness when bitterness and acidity are properly controlled. In this context, the volatile composition and sensory properties of the same hybrid beverages were recently evaluated in a complementary study by our group, which demonstrated high aroma complexity and overall sensory quality across the formulations, particularly highlighting the harmonious integration of citrus, apple, and floral notes in the grape-apple-orange co-fermented beverage (Parra-Cadenas et al., 2026). These differences affected not only the standard chemical composition (lower alcohol content and higher total acidity, with low volatile acidity and nearly complete consumption of fermentable sugars) and the sensory quality of the hybrid beverages (Parra-Cadenas et al., 2026), but also the final composition of other secondary metabolites, as described below. Organic acids Organic acids are among the most important compounds in grape and non-grape wines. These compounds not only play a significant role in their organoleptic characteristics, but they also possess important bioactive properties, although their role in human health is not well studied (Nelson et al., 2016). Table 1 shows the organic acids identified in the control wines and hybrid beverages. Co-fermentations with orange and/or apple juices resulted in beverages with higher concentrations of organic acids, mainly due to the contribution of citric acid from the orange juice (12.18 g/L) and malic acid from the apple juice (6.68 g/L). This led to a lower pH and higher total acidity in hybrid beverages, which improved their freshness, and chemical and microbiological stability. After six months of refrigerated storage, the hybrid beverages showed no evidence of tartrate precipitation, mainly due to their lower tartaric acid content, nor any signs of oxidation, despite the absence of SO2 during bottling. In contrast, these signs of instability were observed in the control wines. This reduced risk of oxidation can be explained by the specific acid and polyphenolic profile of the hybrid beverages, which includes important antioxidant compounds derived from orange and apple juices. Several studies have shown that compounds present in orange juices, such as citric and ascorbic acids and flavonoids (hesperidin, naringin), exhibit antioxidant activity that regulates oxidative stress and the formation of reactive oxygen species (ROS), thereby may contribute to product stability during storage (Miles & Calder, 2021). In addition, flavan-3-ols, such as catechin and epicatechin, found at higher concentrations in apple juices, not only donate electrons to free radicals but can also chelate pro-oxidant metals, delaying oxidative browning processes (Moon et al., 2020). The stability of these compounds during storage has been documented in juices, supporting their role in reducing non-enzymatic oxidation and maintaining the sensory qualities of the beverage (van der Sluis et al., 2005). Table 1Concentration of the main organic acids identified in control wines and co-ferments (n = 3).CSCTSOSOTSASATSMIXSMIXTSMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMalic (g/L)0.697a0.0640.673a0.0310.737a,b0.1210.833b0.0213.777f0.0593.607e0.0252.040c0.0922.340d0.020Citric (g/L)0.140a0.0140.149a0.0166.363c0.0996.334c0.0800.148a0.0170.134a0.0013.198b0.1483.255b0.064Tartaric (g/L)1.561c0.0931.498c0.0571.051b0.0161.073b0.0500.856a0.0420.859a0.0070.993b0.0581.018b0.053Succinic (g/L)0.307a0.0250.523c0.0550.280a0.0200.280a0.0100.303a0.0120.397b0.0420.293a0.0060.410b0.050Lactic (g/L)0.045b0.0160.072c0.0080.019a0.0090.027a0.0100.026a0.0040.019a0.0020.012a0.0010.008a0.001Acetic (g/L)0.2430.0120.3230.0230.2700.2100.0730.0060.1070.0210.1130.0150.2130.0740.3600.347Quinic (g/L)0.271b0.0330.163a0.0200.524c0.1020.473c0.0160.162a0.0160.106a0.0020.518c0.0190.339b0.087Isocitric (g/L)ndnd0.184b0.0370.350c0.046ndnd0.068a0.0170.095a0.010Ascorbic (g/L)ndnd0.089c0.0140.264d0.005ndnd0.017a0.0050.058b0.036Fumaric (mg/L)0.951a0.1011.310a0.0810.728a0.0192.881b.c1.1823.717c0.0973.347c0.5271.713a.b0.1372.361a.b.c1.361Shikimic (mg/L)4.873a0.2584.624a0.70220.032c2.93618.278c1.3755.777a0.3685.716a0.1978.658b0.50210.194b0.230Total acids (g/L)3.268a0.1663.408a0.0739.538e0.1949.729e0.0635.389b0.0215.243b0.0057.364c0.2957.897d0.326Different superscripts (a-f) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. Concentration of the main organic acids identified in control wines and co-ferments (n = 3). Different superscripts (a-f) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. In addition, malic and citric acids have been associated with improved iron absorption and with cardioprotective effects on ischemic heart disease (Bortz & Kirschner, 2009; Tang et al., 2013). Besides to a higher concentration of malic acid, the apple-based hybrid beverages also presented an increased content of fumaric acid. Although fresh apple juice may naturally contain trace amounts of fumaric acid (Kadakal & Nas, 2002), the juice used in this study did not contain detectable levels of this compound. Therefore, the presence of fumaric acid in the fermented beverages is attributed to the dehydration of malic acid during alcoholic fermentation. Orange juice contributed not only citric acid to the hybrid beverages, but also significant levels of isocitric and ascorbic acids, which were absent in both the control wines and the grape-apple co-ferments. Citrus fruits are an important source of ascorbic acid, leading to fermented beverages enriched in this compound (Kelebek et al., 2009). Ascorbic acid, a key water-soluble vitamin (vitamin C), plays an essential role in human health (Alberts et al., 2025). Additionally, it is a potent antioxidant, and its addition to white wines is permitted by the OIV at concentrations up to 200 mg/L. All samples had acetic acid levels below 0.36 g/L. Acetic acid is the main component of volatile acidity, which, at elevated concentrations, imparts an undesirable vinegar-like flavour and is often linked to microbial spoilage. According to previous studies, T. delbrueckii tends to produce lower amounts of acetic acid compared to S. cerevisiae (Tondini et al., 2019), although the differences were not statistically significant. In the present hybrid beverages, however, the strong contribution of the fruit juices appears to dominate the final organic acid profile, potentially masking strain-dependent effects on major acids. Thus, the role of T. delbrueckii is mainly reflected in the modulation of specific metabolic pathways, such as those related to acetic and succinic acid production, rather than in broad changes in total acidity or primary acids. A significant increase in succinic acid, characterized by a bitter-salty flavour, in wines produced with Torulaspora have been previously reported (Puertas et al., 2018). These results support the idea that, in this hybrid fruit–grape system, T. delbrueckii acts more as a fine modulator of specific metabolites rather than as a driver of the overall chemical profile, which is largely governed by the fruit juice composition. Chromatic characteristics, total phenolic content and antioxidant activity Colour is a critical attribute in the quality assessment of wine and other fermented beverages, as it influences consumer acceptance and provides relevant information regarding stability, safety, and biological activity. All the samples showed similar chromatic characteristics (Table 2). They were characterized by high lightness (L*), with a negative influence from the red-green component (a*). Only the orange-containing samples showed a slight increase in the yellow-blue component (b*), resulting in hybrid beverages with a lower hue value. None of the beverages showed evidence of oxidative browning, with absorbance at 420 nm below 0.1 Absorbance Units (AU). Moreover, although yeasts may influence colour development through the formation of metabolites from raw material pigment precursors, no differences were observed between the beverages obtained by single inoculation and those produced by sequential inoculation. Table 2Chromatic characteristics, total phenolic content (TPC) and antioxidant activity (DPPH) in control wines and co-ferments (n = 3).CSCTSOSOTSASATSMIXSMIXTSMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDL*99.17b,c,d0.1299.47d0.1298.93b,c,d0.0698.70b,c,d0.1098.63b,c0.0699.33c,d0.0697.10a0.7698.43b0.42C*5.57b0.275.22b0.005.68b0.296.32c0.035.35b0.184.56a0.156.57c0.475.65b0.60h*103.17b,c0.65105.07c0.75103.93b,c0.49103.50b,c0.20101.45a,b1.58101.87a,b1.10100.41a1.27103.87b,c1.17a*-1.27a,b0.10-1.36a,b0.07-1.37a,b0.10-1.47a0.03-1.06c,d0.12−0.94d0.06−1.18b,c0.07−1.35a,b0.03b*5.43b0.275.04b0.025.51b0.286.15c0.025.24b0.204.46a0.166.46c0.495.48b0.41TPC (mg GAE/L)184.13a1.43190.52a9.98420.61c9.08420.06c15.92320.24b15.60321.26b6.77336.07b3.47335.43b9.32DPPH (mg GAE/mL)33.96a10.1225.85a5.4196.75b24.0598.87b11.9871.21b14.0074.86b9.6670.44b7.3566.48b4.17Different superscripts (a-d) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. Chromatic characteristics, total phenolic content (TPC) and antioxidant activity (DPPH) in control wines and co-ferments (n = 3). Different superscripts (a-d) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. The hybrid beverages obtained from blends with orange juice were characterized by a high total polyphenol content, which was significantly higher in grape-orange co-ferments (OS and OTS) (420.61 ± 9.08 and 420.06 ± 15.92 mg GAE/L, respectively, group c) than in apple-based (AS and ATS) (320.24 ± 15.60 and 321.26 ± 6.77 mg GAE/L, respectively, group b) or control samples (CS and CTS) (184.13 ± 1.43 and 190.52 ± 9.98 mg GAE/L, respectively, group a), and this increase was positively correlated with their elevated antioxidant activity (DPPH). Compared with apple and white grape juices, orange juice generally exhibits a higher total polyphenol content, although this parameter is strongly dependent on the cultivar (García-Infante et al., 2024). While the antioxidant capacity of oranges has traditionally been attributed to their high vitamin C content, more recent studies have highlighted a greater contribution from their phenolic composition (Rapisarda et al., 2022). Polyphenols represent the principal bioactive constituents of fruit wines and are largely responsible for their antioxidant activity (Yuan et al., 2024). Phenolic compounds Table 3 presents a detailed list of the phenolic compounds identified in the control wines and hybrid beverages. For flavan-3-ols, it was not possible to distinguish the different dimers, trimers, and glycosides; therefore, they are numbered according to their chromatographic retention times. Table 3Phenolic compounds (mg/L) identified in control wines and co-ferments (n = 3).CSCTSOSOTSASATSMIXSMIXTSMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDGallic1.00d0.011.10e0.030.37a0.010.43a0.010.60b,c0.060.64c0.060.64c0.040.54b0.01Protocathetic3.63e0.123.24d0.101.93b0.111.61a0.022.41c0.122.07b,c0.242.14b,c0.242.21b,c0.15Total Bi benzoics4.63c0.144.34c0.132.30a0.112.04a0.033.02b0.172.71b0.292.77b0.262.75b0.14GRP1.18e0.021.06d0.010.50a0.020.64b0.010.66b0.030.64b0.020.68b0.030.64b0.03Cafeic6.14e0.575.34b0.091.51a0.071.87a0.062.97b0.142.46b0.112.55b0.062.95b0.11Chlorogenicndnd0.06a0.040.14a0.0712.67c0.2712.57c0.076.63b0.196.23b0.10Caftaric8.47e0.237.69d0.284.38a0.125.20b0.105.07b0.085.39b0.185.69c0.165.10b0.04t-Cutaric0.87f0.030.76e0.030.54a0.040.69c,d0.020.65b,c0.020.72d,e0.020.76e0.030.64b0.03c-Cutaric0.81e0.010.72d0.030.36a0.010.41b,c0.010.38a,b0.020.41b,c0.020.45c0.030.42b,c0.02Coumaric0.05c0.000.04c0.010.02a0.010.01a0.010.04c0.000.04c0.000.04c0.010.03c0.00Fertaric0.93e0.030.81d0.020.43a0.010.51b0.010.54b0.010.52b0.010.57c0.020.53b0.02Total Hydroxycinamics18.45e0.7916.41c0.457.80a0.279.46b0.2422.98f0.4522.76f0.2117.38d0.4716.56c0.14Phlorizinndndndnd3.86d0.113.31c0.061.83a0.021.96b0.03Quercetinndnd3.27a0.283.25a0.19ndnd1.03b0.170.79b0.08Isorhamnetinndnd5.13c0.125.29c0.20ndnd1.64b0.181.17a0.22Hesperidinndnd293.0b19.07302.6b2.13ndnd188.4a3.18188.3a1.34Dydiminndnd28.29c0.5326.30b1.77ndnd15.71a0.6515.54a0.52Naringinndnd115.9b2.47112.9b2.74ndnd59.5a1.0660.4a1.00Total Flavonols & Dihydrochalconesndnd445.6c22.1450.3c6.743.86a0.113.31a0.06268.1b3.45268.2b1.00Catechin2.84b0.153.06b0.171.51a0.021.29a0.065.01d0.214.73d0.323.81c0.173.65c0.22Epicatechin1.07a0.101.06a0.070.56a0.060.53a0.1016.84e0.4216.31d0.2010.59b0.3811.15c0.30Gallocatechin1.14c0.071.36d0.080.68a,b0.020.60a0.090.66a, b0.050.65a,b0.030.79b0.050.74a,b0.07Epigallocatechin0.12a0.030.12a0.020.04b0.020.05b0.010.04b0.010.06b0.020.06b0.030.06b0.01Sum of monomers5.17b0.195.61b0.312.79a0.032.46a0.1822.54e0.6021.75d0.2315.25c0.2815.59c0.53(Epi)cat-(Epi)cat-12.51b0.212.97c0.131.31a0.101.39a0.033.44d0.123.69d0.352.66b,c0.162.68b,c0.21(Epi)cat-(Epi)cat-214.84c0.8916.54c0.557.89a0.939.66b0.2165.42e1.2467.57f1.4542.81d0.9242.39d1.21(Epi)cat-(Epi)cat-32.32b0.042.77c0.141.20a0.071.38a0.063.38d0.16343d0.102.77c0.152.94c0.12(Epi)cat-(Epi)cat-42.42a0.142.50a0.131.45a0.311.91a0.91201.2e1.37167.80d3.15103.67b0.68129.32c3.30(Epi)cat-(Epi)cat-50.44b0.040.48b0.010.23a0.010.24a0.021.72d0.181.77d0.101.05c0.091.08c0.03(Epi)cat-(Epi)cat −6ndndndnd4.63b0.174.73b0.082.56a0.322.48a0.15(Epi)cat-(Epi)gallocat 11.65d0.191.66d0.100.64a0.050.82a,b0.050.93b,c0.111.17c0.121.03b,c0.040.95b,c0.07Sum of dimers (diOH)222.54b1.3025.25b0.9012.09a1.1514.58a1.10279.8f1.67249.00e4.78155.52c0.82180.90d2.93(Epi)cat-(Epi)cat-(Epi)cat-10.36b0.060.46b0.030.18a0.030.20a0.011.81d0.152.19e0.081.13c0.071.05c0.06(Epi)cat-(Epi)cat-(Epi)cat-20.14a0.020.17a0.020.10a0.020.09a0.030.90c0.120.99c0.030.55b0.090.54b0.04(Epi)cat-(Epi)cat-(Epi)cat-30.10a0.010.13a0.010.09a0.020.09a0.0711.32d0.1812.63e0.606.86c0.076.06b0.11Sum of trimers (diOH)30.61a0.090.76a0.050.37a0.010.38a0.0514.04d0.2015.81e0.628.55c0.167.65b0.12(Epi)cat-Glucoside-1nd0.06a0.010.35b0.060.33b0.040.45c0.030.45c0.050.34b0.020.31b0.03(Epi)cat-Glycoside-20.08a0.010.06a0.020.04a0.01nd0.90d0.110.67c0.090.47b0.040.53b0.04(Epi)cat-Glycoside-3ndndndnd0.40c0.090.28b0.030.19a0.020.20a0.01(Epi)cat-Glycoside-4ndndndnd0.81c0.040.72b0.050.46a0.040.47a0.02(Epi)cat-Glycoside-50.05a0.020.05a0.020.04a0.010.06a0.010.13b0.020.12b0.020.08a,b0.050.06a0.02(Epi)cat-Glycoside-6ndndndnd0.25b0.040.23b0.040.17a0.010.17a0.03Sum of glycosides0.13a0.020.17a0.030.43b0.060.39b0.052.95e0.182.49d0.101.71c0.081.74c0.08Total Flavan-3-ols28.45b1.3231.79b1.2315.68a1.1017.81a1.31319.3e2.00289.05f5.32181.0c0.98205.88d3.53t-Glc-resveratrol0.38b0.040.14a0.020.16a0.150.30a,b0.060.33b0.030.14a0.040.26a,b0.040.35b0.06c-Glc-resveratrol6.10c0.265.41c0.094.20b0.223.55b0.083.98b0.233.02a0.214.02b0.264.18b0.25t-Resveratrol0.10b0.030.15c0.030.05a0.020.06a,b0.020.07a,b0.02ndndndc-Resveratrol1.14d0.051.19d0.110.28a,b0.050.20a0.020.36b,c0.030.42c0.050.47c0.010.27a,b0.08Piceatannol0.52c0.090.84d0.050.41a,b,c0.080.27a0.040.31a,b0.020.47b,c0.020.54c0.130.38a,b,c0.04Total Stilbenes8.24d0.377.73c0.105.11b0.284.39a0.125.05b0.254.05a0.265.29b0.405.17b0.40Total phenols59.77a0.7160.28a0.95476.50d23.47484.0d,e5.17354.2c2.07321.88b5.93474.58d4.19498.59e4.46Different superscripts (a-f) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. Phenolic compounds (mg/L) identified in control wines and co-ferments (n = 3). Different superscripts (a-f) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. The main differences were observed among the different types of fruits tested, with the hybrid beverages showing a substantial and selective enrichment in polyphenols. Other authors have reported an increase in the flavonoid content of strawberry wines produced through the sequential inoculation of two non-Saccharomyces yeasts (Hanseniaspora uvarum and Kurtzmaniella quercitrusa) followed by S. cerevisiae (Yang et al., 2024). In contrast, sequential fermentation with T. delbrueckii and S. cerevisiae does not appear to be an effective strategy for enhancing flavonoid levels. Control wines exhibited lower total polyphenol levels and were only distinguished by slightly higher contents of benzoic acids and stilbenes, mainly resveratrol and its derivatives. This pattern is typical of white vinification, which involve minimal skin and seed contact, thereby limiting the extraction of flavonoids (flavonols and flavan-3-ols). In contrast, hydroxycinnamic acids were relatively more abundant in control wines than in orange-based hybrids, with caffeic and caftaric acids being particularly prominent. The latter plays a pivotal role in oxidative stability, since its reaction with glutathione generates the grape reaction product (GRP), which prevents ortho-diphenol oxidation to quinones (Singleton et al., 1985). Although apples and oranges are recognized sources of glutathione (Al-Temimi et al., 2023), GRP has not been detected in these matrices or in cider, due to the absence of the precursors required. Tartaric acid, required for the formation of caftaric acid from caffeic acid, is absent in apples and oranges, being essentially restricted to grapes and certain fermented or processed derivatives containing grape-derived ingredients. Resveratrol and its derivatives have drawn interest due to their broad chemopreventive effects against various diseases and their potential therapeutic applications (Rauf et al., 2018). Orange-based hybrid beverages had the lowest concentrations of flavanols, benzoic acids, and hydroxycinnamic acids. However, they were the only beverages in which flavonols were detected at significant concentrations. Flavonols are notable for their health-promoting effects, including antioxidant activity, protection against oxidative stress, and mitigation of chronic diseases. These properties also contribute to improved food and wine stability and shelf life, making flavonols valuable as natural preservatives (Shahidi & Ambigaipalan, 2015). In orange-based beverages, five flavonols were identified, notably hesperidin, naringin, and dydimin. Hesperidin and naringin are the main flavonoids in citrus fruits, and their different biological activities, such as inhibition of lipid peroxidation and antiatherogenic effects, have been widely studied (Altunayar-Unsalan et al., 2022; Benavente-García & Castillo, 2008), whereas dydimin is a bioactive flavonoid glycoside with important pharmacological properties found in orange and other citrus fruits (Morelli et al., 2014). Moreover, previous studies have suggested a potential synergistic effect between bioactive compounds and the moderate alcohol content of fermented orange beverages, resulting in greater protection against cardiovascular risk factors than orange juice in healthy mice (Escudero-López et al., 2015). Co-fermentation with apple juice resulted in hybrid beverages with slightly higher concentrations of hydroxycinnamic acids compared to control wines, with apple being the only fruit with substantial amounts of chlorogenic acid. This compound occurs naturally in apples and has been reported to show important antioxidant and pharmacological activities (Boyer & Liu, 2004). Within the flavonol and dihydrochalcone group, only the dihydrochalcone phlorizin was detected. This metabolite, unique to apples and commonly used as a marker of apple-derived products, has been associated with potential benefits in reducing type 2 diabetes (Bondonno et al., 2017). Apple-based beverages were also characterized by their markedly higher flavan-3-ol content, approximately 10-fold higher than in control wines and up to 20-fold higher than in grape-orange co-ferments. Unlike the other beverages, apple samples contained three times more epicatechin than catechin. For trihydroxylated monomers, gallocatechin predominated over epigallocatechin across all matrices. Regarding dihydroxylated dimers [(Epi)catechin-(Epi)catechin], dimer 2 dominated in control wines and grape-orange hybrids, whereas dimer 4 was predominant in apple beverages, followed by dimer 2; notably, dimer 6 was exclusively detected in apple-based beverages. Recent research highlights the potent biological effects of flavan-3-ols, with increasing evidence indicating that dimers may display unique and even superior activities compared to monomers (Zou et al., 2025). However, their antioxidant potential is not clear, as has also been demonstrated in our study. Regarding dihydroxylated trimers, those were predominated in apple beverages. With respect to glycosides, apple-based beverages contained all seven identified compounds, with total glycoside levels exceeding those of both control wines and grape-orange hybrids. An in vivo study on rats showed that intestinal absorption of catechin may be increased for catechin glucosides (Raab et al., 2010). However, although their intestinal absorption depends on chemical structure, evidence suggests that flavan-3-ol absorption may not be essential for bioactivity, since these compounds could provide health benefits through modulation of the gut microbiota (Bondonno et al., 2017). Aminoacids and biogenic amines Table 4 shows the amino acid and biogenic amine content in both control wines and hybrid beverages. As expected, the predominant amino acid in all fermented beverages was proline, which is not metabolized by S. cerevisiae (Valero et al., 2003). This compound originates mainly from grape must, which exhibited an initial average concentration of 2.35 g/L. Apple juice contained 0.15 g/L, whereas proline was not detected in orange juice. Although no significant differences were observed in the hybrid beverages, the proline concentrations in control wines suggest partial metabolism of this amino acid by Torulaspora. Table 4Aminoacids and biogenic amines (mg/L) identified in control wines and co-ferments (n = 3).CSCTSOSOTSASATSMIXSMIXTSMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDMeanSDProline (g/L)2.38c0.801.50b0.100.74a0.080.79a0.050.71a0.040.68a0.020.68a0.040.75a0.11Aspartic acid0.55a,b0.100.34a0.030.87b0.411.46c0.131.20c0.272.15d0.140.49a.b0.010.31a0.04Glutamic acid2.75b0.721.83a,b0.492.17a,b,c0.962.91b0.252.48b0.513.38b0.101.62a,b0.141.13a0.05OH-Proline0.40c0.070.32b0.040.19a0.060.19a0.040.18a0.050.25a,b0.010.18a0.040.18a0.03Asparagine+Glutamine0.41a0.050.36a0.080.64a0.221.02b0.140.60a0.144.07c0.350.37a0.020.25a0.03Serine0.94a0.050.43a0.041.45a0.682.54b0.261.61b0.303.09b1.080.95a0.100.62a0.04Histidine5.29c0.574.92c0.110.87a,b0.261.32b0.290.89a,b0.090.55a0.060.60a0.010.37a0.02Glycine4.10b,c0.183.93b,c0.134.53c1.582.95b0.151.91a0.261.72a0.021.18a0.131.48a0.04Threonine0.30a0.020.20a0.020.92c0.421.53d0.180.79b,c0.210.56a,b,c0.010.44a,b0.020.25a0.04Arginine+β-alanine0.06a0.000.06a,b0.000.13d0.010.12d0.000.07b0.010.10c0.000.08b0.000.10c0.00α-Alanine35.65e1.1740.52f0.816.41b2.5916.55d1.034.31a0.7812.68c0.472.76a0.3217.13d0.43GABA2.44a0.542.19a0.283.01a,b1.464.27c0.372.95a,b0.613.92c0.111.75a0.311.38a0.09Tyrosine+ammonium0.54a,b0.050.53a,b0.050.55a,b0.120.62b0.070.44a,b0.060.85c0.060.37a0.110.55a,b0.06Valine1.38a,b0.541.64b0.141.32a,b0.191.01a,b0.351.21a,b0.110.78a0.201.10a,b0.341.60b0.08Methionine0.98a0.250.67a0.041.85b0.532.69c0.341.80b0.520.32a0.020.58a0.020.38a0.17Cysteine0.33a0.050.18a0.031.22b0.671.81c0.191.10b0.300.17a0.000.51a0.020.19a0.07Isoleucine0.10e0.03nd0.09d,e0.010.06c,d0.010.03ab0.01nd0.04b,c0.01ndTryptophan1.31d0.221.25d0.080.57b,c0.260.74c0.090.39a,b0.080.27a0.030.21a0.010.20a0.03Leucine0.49a0.010.33a0.021.16b0.371.44b0.251.13b0.350.24a0.000.36a0.040.32a0.32Phenylalanine0.69a0.200.21a0.031.28b0.611.97c0.231.24b0.290.19a0.000.58a0.030.30a0.02Lysine1.48a,b0.190.79a0.113.07c1.104.89d0.423.46c0.540.79a0.021.97b0.100.94a0.05Ornithine3.37b0.192.25a,b0.042.34a,b1.802.47a,b0.190.91a0.231.20a,b0.022.47a,b1.111.64a,b0.02Total aminoacids (without proline)65.94d2.8964.44d1.2835.37b13.1453.36c4.3429.43 a,b5.5037.93b1.6319.31a2.1630.09a,b1.10Histamine2.09c0.083.47d0.071.89b,c0.792.75d0.161.29a,b0.213.15d0.060.85a0.546.22e0.10Putrescine0.23a0.010.26a0.022.18c0.122.23c0.150.26a0.000.22a0.001.14b0.071.13b0.03Total amines2.33b0.083.73c0.094.07c0.794.98d0.171.55a0.213.37c0.061.99a,b0.507.35e0.11Different superscripts (a-f) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. Aminoacids and biogenic amines (mg/L) identified in control wines and co-ferments (n = 3). Different superscripts (a-f) in the same row mean significant differences (α = 0.05) according to the test of Student–Newman–Keuls. CS: control wine S. cerevisiae; CTS: control wine T. delbruekii + S. cerevisiae; OS: grape-orange co-ferment S. cerevisiae; OTS: grape-orange co-ferment T. delbruekii + S. cerevisiae; AS: grape-apple co-ferment S. cerevisiae; ATS: grape-apple co-ferment T. delbruekii + S. cerevisiae; MIXS: grape-apple-orange co-ferment S. cerevisiae; MIXTS: grape-apple-orange co-ferment T. delbruekii + S. cerevisiae. Excluding proline, the total amino acid concentration in control wines was approximately 60 mg/L. Although hybrid beverages exhibited lower values, this characteristic could be advantageous for enhancing microbial stability. Amino acids are vital for yeast growth and metabolism during fermentation. However, when they remain at elevated concentrations in the final product, they can favour the growth of undesirable microorganisms and the formation of compounds such as biogenic amines and ethyl carbamate, both associated with negative health effects (Vincenzini et al., 2017). This reduction could be due to a dilution effect due to the addition of fruit juices but also may reflect differences in yeast assimilation patterns in the hybrid matrices. The presence of readily fermentable sugars and alternative nitrogen sources in apple and orange juices may have promoted a more efficient uptake of free amino acids by the yeasts, resulting in lower residual levels in the final beverages. No consistent pattern was observed regarding individual amino acid consumption or release by the yeasts used, except for alanine, a sweet-tasting amino acid (Solms, 1969), whose concentration was higher in fermentations carried out with Torulaspora. Previous studies have reported that certain non-Saccharomyces yeasts, such as Pichia kluyveri and Metschnikowia pulcherrima, can release greater amounts of amino acids during sequential fermentations with S. cerevisiae (Benito et al., 2015). This behaviour supports the idea that non-Saccharomyces strains can actively modulate nitrogen metabolism, even if the overall amino acid profile in the present hybrid system is influenced by the fruit matrix. Biogenic amines are predominantly generated through microbial decarboxylation of amino acids, as well as amination and transamination reactions of aldehydes and ketones (Peña-Gallego et al., 2012). Two biogenic diamines were identified, histamine and putrescine. Histamine concentrations were higher in beverages fermented with Torulaspora, reaching a maximum of 6.22 mg/L in grape-apple-orange co-ferments (MIXTS, group e) compared to a minimum of 1.29 mg/L observed in grape-apple co-ferments obtained by single S. cerevisiae inoculation (AS, groups a and b). In contrast, putrescine levels were slightly elevated in hybrid beverages containing orange juice, ranging from 2.18 to 2.23 mg/L in grape-orange co-ferments produced by single S. cerevisiae inoculation and sequential T. delbrueckii + S. cerevisiae inoculation (OS and OTS, group c). Although all values remained well below typical maximum levels reported in wines (Peña-Gallego et al., 2012), these minor but significant differences suggest that both yeast species and the fruit matrix can modulate biogenic amine formation. Putrescine is one of the products of arginine decarboxylation; however, no major differences were observed in the arginine content of orange-based beverages, suggesting that this amine originates directly from orange juice (Vieira et al., 2007). Furthermore, previous studies have shown a significant reduction of histamine and tyramine, and an increase in spermidine and tryptamine in red wines produced by sequential inoculation of T. delbrueckii and S. cerevisiae (Wang et al., 2021). Our observations are consistent with the ability of Torulaspora to colonize the medium and limit the growth of other microorganisms, although this effect appears context-dependent and may vary with the fruit matrix and fermentation conditions. Conclusion Co-fermentation of grape must with orange and/or apple juices resulted in a selective enrichment of hybrid beverages in bioactive compounds, mainly organic acids and polyphenols. Each fruit contributed a distinct chemical profile, with orange enhancing citric acid and flavonols, apple increasing malic acid and flavanols, and grape must rich in tartaric acid and hydroxycinnamic acids; combining all three produced a balanced profile likely to enhance overall bioactivity. This approach demonstrates that fruit co-fermentation is a sustainable strategy for valorising surplus or highly perishable raw materials while preserving functional compounds. The resulting high-stable, lower-alcohol and bioactive-enriched beverages have clear potential to meet consumer demand for healthier, innovative drinks and offer industry opportunities for functional product diversification. CRediT authorship contribution statement S. Parra-Cadenas: Methodology, Investigation, Formal analysis. M.C. Díaz-Maroto: Writing – review & editing, Writing – original draft, Project administration, Methodology, Investigation, Funding acquisition. M.S. Pérez-Coello: Writing – review & editing, Project administration, Investigation, Funding acquisition. P.M. Izquierdo-Cañas: Resources, Methodology, Investigation. E. García-Romero: Writing – review & editing, Supervision, Methodology, Investigation, Funding acquisition, Formal analysis. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments This research was supported by the European Union (EU) through the ERDF and by the Junta de Comunidades de Castilla-La Mancha (JCCM) through INNOCAM (JCCM, Spain), Grant SBPLY/23/180225/000035, and the University of Castilla-La Mancha (UCLM), Grant 2025-GRIN-38446. Footnotes This article is part of a Special issue entitled: ‘Wine Chemistry’ published in Food Chemistry: X. Data availability Data will be made available on request. References Associated Data Data Availability Statement Data will be made available on request.