Tannin Rich-Extracts: Natural Solutions for Preserving the Physicochemical, Oxidative, and Microbiological Quality of Beef Patties During Cold Storage

Abstract

Increasing consumer demand for healthier and clean-label meat products, together with health concerns over synthetic antioxidants, has driven interest in natural alternatives. In this context, tannin-rich extracts were evaluated as functional ingredients to improve the quality and shelf life of beef patties. The effect of two different tannin-rich extracts, each tested at three different concentrations (0.005%, 0.02%, and 0.04% w/w), was investigated in beef patties. Obtained results were compared with a commercial rosemary extract (0.2% w/w) and an untreated control. Natural antioxidant supplementation significantly reduced lipid oxidation during refrigerated storage, limiting malondialdehyde (MDA) formation. At the end of the 7-day storage period, the control sample exhibited the highest thiobarbituric acid reactive substance (TBARS) value of 2.99 ± 0.01 mg MDA/kg, whereas treated samples showed markedly reduced oxidation (0.34–0.97 mg MDA/kg), with tannin-rich extracts presenting greater antioxidant efficacy with respect to rosemary extract (p < 0.001). The natural compounds also effectively inhibited hexanal formation and delayed the accumulation of 1-octen-3-ol compared with the control (p < 0.001). Moreover, all extracts enhanced meat redness, as indicated by higher CIELAB a* values, while no significant effects (p > 0.05) were observed on texture, microbial growth, or overall sensory acceptance. These results highlight tannin-rich extracts as promising natural antioxidants for improving oxidative stability and extending the shelf life of beef patties.

1. Introduction

Meat plays a crucial role in the global food market due to its economic value and nutritional contribution [1]. Worldwide demand continues to increase, exceeding 300 million tonnes annually, mainly driven by pork, poultry, and beef, while ground meat products are experiencing the most pronounced growth [2]. However, fresh ground meat and obtained patties are extremely susceptible to lipid oxidation and microbial spoilage, which represent the main causes of quality deterioration and reduced consumer acceptability. In detail, the grinding process facilitates the breakdown of the muscle cell membrane structure, exposing an increased meat surface area to oxidation [3,4]. In addition, a high concentration of unsaturated fats, pro-oxidant compounds, including salt and metals, high oxygen availability, light, and the storage temperature contribute to the initiation and propagation of lipid oxidation, strongly affecting the safety and nutritional quality of meat products [5]. The lipid oxidation process mainly involves the decomposition of lipid in primary oxidation products, such as free radicals and hydroperoxides, and later, in secondary oxidation products, as well as volatile organic compounds (VOCs), including aldehydes, ketones, epoxides, hydroxyl compounds, oligomers, and polymers [6]. Loss of nutrients, unpleasant colour changing, rancidity, off-flavour and off-odour, gas production, texture deterioration, and sensory alteration are examples of changes caused by microbial alteration and lipid oxidation. To prevent lipid oxidation process, synthetic additives, such as butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), and tertbutyl hydroquinone (TBHQ), ascorbic acid, tocopherols, gallates, and isoascorbate, are commonly used, while nitrite and phosphate control microbial growth [7]. However, scientific evidence has confirmed potential health risks associated with their use, ranging from mild toxicity to cancer [4], leading to strict regulations on their inclusion in food formulations. Consequently, research studies are focusing on natural, greener alternatives, which may offer beneficial health effects and improved environmental sustainability. Essential oils, extracts, and oleoresins, obtained from natural sources, are examples of current alternatives to synthetic antioxidants [8]. Specifically, rosemary extract has been extensively studied for meat product preservation due its bioactive compounds, including carnosol and carnosic acid, terpenes, and flavonoids, able to counteract lipid oxidation and microbial growth [8,9]. Nevertheless, growing interest is being directed toward other natural polyphenol-rich extracts, among which tannin-rich extracts have emerged as promising candidates for sustainable meat preservation. Tannins are a heterogeneous class of water-soluble polyphenolic compounds, traditionally classified according to their chemical structure into hydrolysable, condensed, and complex tannins. They are currently employed in multiple sectors, including leather, agriculture, animal nutrition, food packaging, cosmetics, and pharmaceutical fields [10,11]. Within the food sector, commercial tannin extracts are primarily applied in oenology and brewing to enhance the colour and texture of the beverages [12]. In addition to these functional abilities, studies have been reported that tannin-rich extracts exhibit in vitro and in vivo antioxidant, antiviral, and antimicrobial activity, coupled with anti-inflammatory, cardioprotective, and anticancer activity, and microbiota modulators [13,14,15]. As a result of these promising properties, tannin-rich extracts have been largely applied through dietary supplementation of animals to improve meat quality [16,17]. In contrast, their direct application in meat products has only recently begun to be explored. Such studies are often limited to partial evaluations, focusing on a restricted number of parameters rather than on a comprehensive assessment of the final product. However, technological properties, sensory acceptability, and shelf-life stability should be considered together for a complete evaluation. To address this gap, the present study provides a comprehensive investigation of two chemically distinct tannin-rich extracts, each tested at three different concentrations, directly applied to beef patties. A commercial rosemary extract, widely used as a natural antioxidant in the meat industry, was included as a benchmark to assess the performance of the tannin-rich extracts in comparison. Multiple aspects of meat quality were examined, ranging from lipid oxidation and microbial stability to physical characteristics, including colour and texture, as well as sensory properties. By adopting this holistic approach, our work highlights the ability of tannin-rich extracts to enhance oxidative stability and delay discoloration while maintaining overall product quality, thereby offering new insights into their applicability in meat systems.

2. Materials and Methods

2.1. Materials and Chemicals

Beef meat from the Longissimus thoracis et lumborum (LTL) muscle of 24-month-old Bos taurus steers was purchased from three different local butcher’s shops (Turin, Italy). The meat was coarsely ground, packaged in food-grade butcher’s kraft paper, and immediately transported to the laboratory under refrigerated conditions. A commercial food-grade household table salt (Italkali, Italy) was acquired from a local market (Turin, Italy). Two powder extracts rich in hydrolysable tannins, differing in their chemical composition, were kindly provided by Silvateam S.p.A. (San Michele Mondovì, Italy). Specifically, tannin-rich extract A (TA), previously characterized by Radebe et al., was distinguished by the presence of ellagic acid derivatives, whereas tannin-rich extract B (TB) was rich in gallic acid esters [18]. In detail, TA had a tannin content of 70% on dry matter (OIV method) and 40% total polyphenols (gallic acid equivalents). In contrast, TB exhibited a tannin content of 96% on dry matter (OIV method) and 70% polyphenols (gallic acid equivalents). A commercial rosemary extract, dispersed in sunflower oil as the carrier system, was supplied by Camlin Fine Sciences Ltd. (Mumbai, India). Moreover, rosemary extract contained 3.2–3.8% of carnosic acid and 3.4–4.2% of phenolic diterpenes (sum of carnosol and carnosic acid). All the solvents and chemicals were of analytical grade. Sodium phosphate monobasic dihydrate, sodium phosphate dibasic anhydrous, ethylenediaminetetraacetic acid (EDTA), thiobarbituric acid (TBA), trichloroacetic acid (TCA), heptahydrate, 2,2′-diphenyl-1-picrylhydrazyl (DPPH), Trolox, butylated hydroxytoluene (BHT), and sodium ascorbate were supplied by Merck (Darmstadt, Germany), while ethanol absolute anhydrous was from Carlo Erba (Milan, Italy). 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ), ferric chloride hexahydrate, sodium acetate, and hydrochloric acid were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Oxoid Ringer solution tablets were acquired from Thermo Fisher Scientific (Waltham, MA, USA); Plate Count Agar (PCA) from VWR Chemicals (Leuven, Belgium); Violet Red Bile Glucose Agar (VRBG) from Liofilchem (Teramo, Italy); Lactobacilli MRS Broth from Neogen (Lansing, MI, USA); and Agar from Neogen (Paisley, UK). Ultrapure water was purified with a Milli-Q filter system (Millipore, Milan, Italy), and n°1 filters (70-mm diameter) were purchased from Whatman (Cytiva, Maidstone, UK).

2.2. Evaluation of the Antioxidant Properties of Tannin-Rich Extracts

The antioxidant activity of the two chemically distinct tannin-rich extracts, TA and TB, was evaluated using in vitro assays to determine the radical scavenging activity (DPPH) and ferric reducing antioxidant power (FRAP). For both assays, five concentrations of each extract (1, 10, 25, 50, and 100 μg/mL) were investigated, and all measurements were performed in triplicate (n = 3). Butylated hydroxytoluene (BHT) and sodium ascorbate were included as reference antioxidants at the same concentrations. The DPPH radical scavenging activity was determined according to the method described by Cantele et al. [19], and results were expressed as the percentage of inhibition (%I). The FRAP assay was carried out following the procedure reported by Molino et al. [20], and results were expressed as μmol Trolox equivalents (TE)/g extract.

2.3. Preparation, Packaging, and Storage of Beef Meat Patties

In order to better elucidate the antioxidant effect of the extracts, three ground beef meat batches (1 kg each), acquired from three different local markets, were pooled to obtain a single homogenised batch to ensure consistency across treatments. Meat was then processed using a home-type meat grinder equipped with a KitchenAid 5KSMMGA attachment (Benton Harbor, MI, USA). The grinding was performed in two consecutive steps: first using an 8 mm coarse stainless-steel grinding plate, followed by a second grinding with a 4.5 mm stainless-steel plate. The fat content was below 15%, and no fat removal was performed. Beef patties were obtained by manually mixing and kneading minced beef (95.5% w/w), cold deionized water (3%, w/w), food-grade salt (1.5%, w/w), and, on top of the base formulation, the extracts, testing three different concentrations for tannin-rich extracts TA and TB (0.005%; 0.02% and 0.04%, w/w) and 0.02% (w/w) for rosemary extract. In detail, mixing was carried out for 60 s to ensure a uniform distribution of all ingredients. The resulting mixture was then portioned into 30 g patties and shaped into a circular disk with a diameter of 4 cm and a height of 3 cm. Patty shaping was performed manually by pressing the patties into a cylindrical mould to obtain a uniform size and geometry. Then, the patties (30 g) were placed in food-grade PET clamshell trays (hinged-lip snap closure; non-barrier; tray dimensions 14.2 cm × 12.3 cm × 6.8 cm) and packaged under atmospheric air conditions. Each tray contained three patties, and a total of three trays per treatment were prepared and stored at 4 °C in the dark. Three sampling days were considered: day 0 (T0, after four hours of patties preparation), day 3 (T1), and day 7 (T3). Each experiment was repeated in triplicate. Samples were labelled as TA50, TA200, and TA400 for patties enriched with tannin-rich extract A (TA); TB50, TB200, and TB400 for tannin-rich extract B (TB), respectively; and RE2000 for rosemary extract. In addition, a control sample (CTR) was prepared without the addition of extracts.

2.4. Thiobarbituric Acid Reactive Substance (TBARS) Determination

TBARS were determined according to Tarladgis et al. [21]. Briefly, 2 g of each meat sample was weighed and mixed with 8 mL of an aqueous phosphate buffer solution (pH 7.0) containing 0.1% (w/v) EDTA. The mixture was homogenized using a T-25 Ultra-Turrax (IKA-Werke GmbH & Co. KG, Staufen, Germany) at 21,500 rpm for 30 s [13]. Subsequently, 2 mL of trichloroacetic acid solution (30%, w/v) was added, and the mixture was further homogenized at 17,500 rpm for 30 s. The solution was filtrated with a paper filter (Whatman No. 1), and 5 mL of filtrate was collected in a 20 mL Sorvitel tube, mixed with 5 mL of TBA aqueous solution (0.02 M), put in a water bath (Julabo SW 20, Savatec; Turin, Italy) for 20 min at 90 °C, refrigerated for at least 30 min at 4 °C, and sonicated for 10 min. Then, the absorbance of the complex was read at 530 nm with a UV-1800 spectrophotometer (Shimadzu, Kyoto, Japan). The quantitative determination was carried out using a 1,1,3,3-tetramethoxypropane standard calibration curve, ranging from 0.03 to 2.26 µg/mL, and the results were expressed as milligrams of malondialdehyde (MDA) per kilogram of meat (mg MDA/kg meat). All the measurements were carried out in triplicate.

2.5. Volatile Organic Compounds

According to Cantele et al. [3], the volatile organic compounds (VOCs) were investigated using headspace solid-phase microextraction (HS-SPME) coupled with GC/MS (QP-2010 Plus, Shimadzu, Kyoto, Japan). Two grams of each sample were introduced into a 20 mL glass vial and sealed with an aluminium cap with a PTFE silicone septum. The vials were equilibrated for 15 min at 40 °C in a Combi Pal system (CTC Analytics, AG, Zwingen, Switzerland), and a DVB/CAR/PDMS-coated fused silica fibber (10 mm length, df 50/30 µm; Supelco, Bellafonte, PA, USA) was exposed for 30 min at 40 °C in the headspace of the vial. Desorption to the GC/MS inlet at 260 °C was set for 5 min, with a 1:25 split ratio. An RTX-5 fused silica capillary column (20 m × 0.10 mm × 0.10 µm; Restek, Bellafonte, PA, USA) and helium as the carrier gas with a constant linear velocity of 34.7 cm/s were selected. The oven temperature was set at 40 °C (held for 4 min) and then increased up to 220 °C with a rate of 4 °C/min. Then, the temperature was further increased to 260 °C at a rate of 20 °C/min and held constant for 5 min. Ions were scanned within the range of 33–350 (m/z) at a scan rate of 1111 amu/s. The VOCs were recognized by comparison with the mass spectra reported in the NIST08s (National Institute of Standards and Technology, Gaithersburg, MD, USA) library. Results are expressed as area/g of sample (AU/g). The analyses were carried out in triplicate.

2.6. Microbiological Analyses

To investigate microbial dynamics during meat storage, 5 g of each sample, selected from a random part of the patties, was placed into an aseptic stomacher bag with 45 mL of Ringer’s solution (ten-fold dilution). After 90 s of homogenization with a BigMixer 400 stomacher (Interscience, Breda, The Netherlands), decimal dilutions were prepared in quarter-strength Ringer’s solution. Aliquots of homogenized samples were inoculated into appropriate media to investigate the three following microbial groups: total viable count (TVC) for mesophilic bacteria on Plate Count Agar (PCA), for 48 h of incubation at 30 °C; Enterobacteriaceae on Violet Red Bile Glucose Agar (VRBGA) for 24 h of incubation at 37 °C, and lactic acid bacteria (LAB) on De Man, Rogosa, and Sharpe (MRS) agar for 48 h of incubation at 30 °C in anaerobic condition. The obtained results were expressed as the logarithm of the total colony-forming units per gram of meat (log CFU/g) and reported as mean ± standard deviation. The analyses were carried out in triplicate. For the T0 analysis, only the CTR sample was evaluated, as the initial conditions of the minced meat were the same for all samples. Otherwise, both at three days (T1) and at the end of the storage period (T2), each sample was evaluated to assess the effect of natural extracts within the patties with respect to the control sample.

2.7. pH Determination

The pH value was monitored throughout the whole storage period by homogenizing 5 g of meat in distilled water (1:10). The pH value was measured using a Mettler Toledo FiveEasy F20 pH-meter (Greifensee, Switzerland). At the beginning of each day of analysis, the instrument was calibrated using standard buffer solutions at pH 4.00, 7.00, and 10.00. Before each measurement, the temperature of the samples was monitored and allowed to reach 22 ± 1 °C. The instrument was used in combination with an electrode equipped with an integrated temperature sensor, enabling automatic temperature compensation (ATC) during the analysis.

2.8. Colour Measurement

The changing of raw meat patties’ colour was measured through a CM-5 spectrocolorimeter (Konica Minolta, Tokyo, Japan), with D65 illuminant, a slit diameter of 8 mm, and a 10° standard observer angle. The analysis was performed in SCE (Specular Component Excluded) mode, and the CIELab colour parameters (L*: lightness, a*: redness, b*: yellowness) were determined after a blooming time of 30 min at room temperature for the patties. Following the calibration of the instrument with a standardized white tile, three replicates were analysed, with six measurements for each patty, and the results are expressed as the mean and standard error of the mean of the replicates.

2.9. Texture Measurements

Empirical-imitative rheological analyses were performed by using a Texture Analyser “TA-XT2i HDi500” (Stable Micro System, Vienna Court, UK), equipped with a load cell of 5 kg. A penetration and a compression test was performed using a 6 mm diameter stainless-steel probe and a cylindrical stainless-steel probe with a 5 cm diameter, respectively, by employing a test speed of 5 mm/s. From the penetration test, the maximum peak of the curve was used as the hardness value (N), while from the compression test, carried out by compressing samples to 40% of their initial height, the consistency parameter (N*s) was extrapolated as the area under the obtained curve.

2.10. Sensory Evaluation

The study adhered to the Institute of Food Science & Technology Guidelines for Ethical and Professional Practices in the Sensory Analysis of Foods [22]. It followed a prior low-risk assessment that evaluated allergenic, microbiological, chemical, and physical hazards. To assess the effect of tannin extract on the sensory profiling of beef patties, the evaluation was conducted immediately after preparation, during a single session on the same day (T0). The patties were cooked for 3 min on each side in a non-stick pan preheated to 180 °C on an electric hot plate, without adding any fat to avoid flavour interference. A laboratory glass-stem probe thermometer was used to monitor the core temperature, ensuring it exceeded 75 °C for each patty. After cooking, the patties were placed on a plate, covered with aluminium foil, and allowed to rest at room temperature (approximately 22 °C) for 2 min before serving. For each session, 10 g samples of each treatment was randomly assigned a three-digit code for evaluation. Fifteen untrained panellists (balanced by gender and aged 25–50 years) participated in the sensory evaluation. These panellists had previously undergone a brief training session to familiarize themselves with the evaluation attributes, scoring system, and typical characteristics of fresh beef patties. A hedonic test using a 7-point scale was applied to assess colour (1 = very light/pale; 7 = reddish/black), odour, flavour, aroma (1 = very poor; 7 = excellent), tenderness (1 = extremely hard; 7 = very soft), juiciness (1 = very dry; 7 = succulent), and overall acceptability (1 = extremely unacceptable; 7 = highly acceptable). Scores with means ranging from 4 to 7 were considered acceptable.

2.11. Statistical Analysis

Results are reported as the means and standard deviation of three independent experiments (n = 3). Data obtained from sensory test are reported as mean values. The statistical analysis of all data was carried out by SPSS Statistics software (version 29.0.1; IBM; Chicago, IL, USA). For measured parameters (e.g., TBARS, VOCs, microbial growth, pH, colour, texture, and sensory attributes), a one-way analysis of variance (ANOVA) was carried out to assess the main effects of treatment (A) and storage time (T); and two-way analysis of variance (ANOVA) to estimate their interaction (A × T; fixed effect). Replicates were independent experimental units. Tukey’s post hoc test was applied for multiple comparisons. A 95% confidence level was used, and statistical significance was set at p < 0.05. The sensory analysis was performed in one single session. Figures were generated using R software (version 4.5.0; R Core Team; Vienna, Austria).

3. Results

3.1. Antioxidant Properties

The in vitro antioxidant activity of the tannin-rich extracts was compared with that of BHT and sodium ascorbate, and results are reported in Figure 1.

The DPPH radical scavenging activity (Figure 1a) was strongly concentration-dependent, and both tannin-rich extracts, TA and TB, exhibited a progressive increase in radical inhibition, with increasing concentrations (p < 0.001). Specifically, TA achieved 42.99% and 72.07% inhibition at 50 and 100 μg/mL, respectively, whereas TB showed a comparable trend with slightly higher activity, reaching 58.60% at 50 μg/mL and 75.29% at 100 μg/mL. In contrast, the reference antioxidants BHT and sodium ascorbate displayed a more moderate scavenging capacity across the tested concentration range. At 100 μg/mL, BHT and sodium ascorbate reached inhibition values of 43.59% and 42.21%, respectively, which were markedly lower than those observed for both tannin-rich extracts at the same concentration. For the FRAP assay (Figure 1b), all extracts exhibited concentration-dependent reducing activity. Both tannin-rich extracts, TA and TB, showed better activity than BHT at all considered concentrations (p < 0.001). At lower concentrations (1 and 10 μg/mL), both extracts exhibited a greater reducing capacity than sodium ascorbate, whereas at higher concentrations (25–100 μg/mL), sodium ascorbate demonstrated superior FRAP values.

3.2. Thiobarbituric Acid Reactive Substances (TBARS)

The effect of different treatments on beef patties was assessed by monitoring the formation and evolution of thiobarbituric acid reactive substances (TBARS), representative of lipid oxidation. The corresponding results are presented in the heatmap (Figure 2), where the colour intensity represents the relative TBARS values at each time point (T0, T1, and T2) for each treatment. Darker colours correspond to higher TBARS values, indicating greater lipid oxidation, while lighter colours represent lower TBARS, indicating a stronger antioxidant effect.

At T0, after 4 h of storage, the control sample (CTR) presented an initial TBARS value of 0.78 mg MDA/kg of meat. At the same storage time, all samples containing tannin-rich extracts (TA and TB) and rosemary extract (RE) exhibited significantly lower TBARS compared to the control (p < 0.001), indicating an early antioxidant effect. Among the treatments, TB samples showed the greatest inhibition of lipid oxidation, resulting in TBARS reduction up to 71% at the highest concentration tested (0.04%), followed by TA and RE, with maximum reductions of 67% and 51%, respectively. After three days of storage (T1), lipid oxidation markedly increased in the control, reaching 1.97 mg MDA/kg of meat. In contrast, all treated samples maintained significantly lower TBARS levels compared to the control (p < 0.001). Both tannin-rich extracts were effective in limiting oxidation, with TB showing a stronger antioxidant activity than TA and RE. A similar pattern was observed after seven days of storage (T2). The control sample exhibited the highest TBARS value (2.99 mg MDA/kg of meat), whereas all samples enriched with tannin-rich extracts or RE showed significantly lower levels of lipid oxidation (p < 0.001). In particular, TB-treated samples consistently displayed the lowest TBARS values (0.34 mg MDA/kg of meat), followed by TA (0.36 mg MDA/kg of meat) and RE (0.97 mg MDA/kg of meat). Overall, at each storage time, both tannin-rich extracts demonstrated a higher antioxidant efficacy than rosemary extract (p < 0.001). Detailed quantitative data and statistical analysis are reported in Table S1 of the Supplementary Materials.

3.3. Volatile Organic Compounds (VOC) Determination

Secondary volatile organic compounds (aldehydes, ketones, hydrocarbons, and alcohols) are formed as a result of lipid oxidation and microbial spoilage, which have a deleterious impact on meat quality. During the experiment, twenty-seven volatile organic molecules were recognized to monitor lipid oxidation or microbial degradation. Hexanal and 1-octen-3-ol were chosen as lipid oxidation markers, as well as ethanol and acetoin as microbial degradation indicators, as clarified in Figure 3 and Table S2 of the Supplementary Materials.

Hexanal was identified solely in the control sample (CTR) as early as 4 h of storage (T0), and its level rose further throughout storage (p < 0.001). At the same storage durations, all samples containing natural extracts efficiently suppressed hexanal formation, with the exception of TB50, which was detectable in trace amounts at T2 (Figure 3a). As demonstrated in Figure 3b, 1-octen-3-ol was found in the CTR sample at T0 and grew gradually throughout storage. Natural extracts significantly reduced 1-octen-3-ol levels at all storage times (p < 0.05), suggesting an inhibitory action on lipid oxidation. A concentration-dependent response was observed, with larger extract concentrations resulting in stronger inhibition. The highest concentration of TA, TB, and RE entirely prevented 1-octen-3-ol formation at T0 (p < 0.001), whereas lesser amounts were less effective but still considerably decreased it compared to the control (p < 0.001). The same pattern was detected at T1 and T2, when all extracts reduced 1-octen-3-ol accumulation when compared to the control group. Tannin-rich extracts had a stronger inhibitory impact than rosemary extract, particularly after extended storage (p < 0.001; Table S2). Regarding microbial spoilage markers, no significant differences were observed among samples in ethanol (Figure 3c) and acetoin (Figure 3d) formation at the same storage time (p > 0.05). However, the levels of both compounds increased over time in all samples from the beginning to the end of storage (p < 0.001).

3.4. Antimicrobial Effects of Natural Extracts

The ability of tannin-rich extracts and rosemary extract to counteract microbial growth was tested at the beginning (T0), after three days of storage (T1), and at the end of the storage period (T2) of beef patties, as reported in

Table 1. Changes in total viable count (TVC), lactic acid bacteria (LAB), and Enterobacteriaceae during the storage period of beef patties.

Log CFU/g
	Treatment	TVC	LAB	Enterobacteriaceae	p-Value
T0	CTR	4.48 ± 0.43	4.48 ± 0.33	1.45 ± 0.21	A	T	A × T
T1	CTR	4.78 ± 0.18	4.83 ± 0.16	1.75 ± 0.32	>0.05	>0.05	>0.05
	TA50	4.72 ± 0.25	4.76 ± 0.23	1.78 ± 0.12
	TA200	4.76 ± 0.13	4.71 ± 0.19	1.76 ± 0.28
	TA400	4.66 ± 0.25	4.72 ± 0.32	1.81 ± 0.23
	TB50	4.62 ± 0.07	4.66 ± 0.25	1.68 ± 0.37
	TB200	4.39 ± 0.31	4.43 ± 0.36	1.72 ± 0.08
	TB400	4.83 ± 0.17	4.64 ± 0.31	1.75 ± 0.17
	RE2000	4.74 ± 0.29	4.78 ± 0.15	1.82 ± 0.29
T2	CTR	4.89 ± 0.09	5.16 ± 0.31	<2
	TA50	5.01 ± 0.16	5.26 ± 0.56	<2
	TA200	4.83 ± 0.56	4.59 ± 0.61	<2
	TA400	4.81 ± 0.11	4.67 ± 0.75	<2
	TB50	4.52 ± 0.68	4.56 ± 0.63	<2
	TB200	4.10 ± 0.10	4.26 ± 0.26	<2
	TB400	5.35 ± 1.91	5.37 ± 1.90	<2
	RE2000	4.93 ± 0.87	4.85 ± 0.54	<2
	Sig.	N.S.	N.S.	N.S.

Results are expressed as the mean ± standard deviation of three independent replicates (n = 3), and results of two-way ANOVA and Tukey’s post hoc test are also reported. The corresponding significance levels (p-value) are provided. “A” indicates the treatments effect, “T” indicates storage time, and “A × T” represents the interaction between treatment and storage time. Abbreviations: Sig. = statistical significance; N.S. = not significant; CTR: control; TA50 sample added with 0.005% of tannin-rich extract A; TA200 sample added with 0.020% of tannin-rich extract A; TA400 sample added with 0.040% of tannin-rich extract A; TB50 sample added with 0.005% of tannin-rich extract B; TB200 sample added with 0.020% of tannin-rich extract B; TB400 sample added with 0.040% of tannin-rich extract B; RE2000 sample added with 0.200% of rosemary extract.

.

Over the one-week storage, a limited bacterial growth was observed across all considered microbial indicators (i.e., Enterobacteriaceae, LAB, and TVC) in both the control and samples enriched with natural extracts. Notably, the final levels of Enterobacteriaceae were lower than 2 CFU/g in each sample. Overall, the results showed that one week of storage did not compromise meat quality in terms of microbial spoilage; however, none of the tested extracts, independently of the investigated concentration, exhibited any antimicrobial activity. For Enterobacteriaceae, LAB, and TVC, no significant differences (p > 0.05) were detected compared to the control (Table 1).

3.5. pH

The pH values measured throughout the entire storage period are reported in

Table 2. pH changes in beef patties during cold storage.

pH
Treatment	Time of Storage				p-Value
	T0	T1	T2	Sig.	A	T	A × T
CTR	5.67 ± 0.01 B	5.74 ± 0.02 A	5.52 ± 0.03 C	***	0.092	0.029	0.103
TA50	5.66 ± 0.01 B	5.75 ± 0.02 A	5.50 ± 0.02 C	***
TA200	5.66 ± 0.00 B	5.76 ± 0.00 A	5.48 ± 0.03 C	***
TA400	5.65 ± 0.01 B	5.74 ± 0.01 A	5.54 ± 0.01 C	***
TB50	5.67 ± 0.01 B	5.74 ± 0.00 A	5.47 ± 0.01 C	***
TB200	5.69 ± 0.01 B	5.75 ± 0.01 A	5.43 ± 0.01 C	***
TB400	5.62 ± 0.03 B	5.71 ± 0.03 A	5.41 ± 0.04 C	***
RE2000	5.64 ± 0.02 B	5.73 ± 0.01 A	5.43 ± 0.02 C	***
Sig.	N.S.	N.S.	N.S.

Results are expressed as the mean ± standard deviation of three independent replicates (n = 3), and results of two-way ANOVA with Tukey’s post hoc test are also reported. The corresponding significance levels (p-value) are provided. “A” indicates the treatments effect, “T” indicates storage time, and “A × T” represents the interaction between treatment and storage time. Different uppercase letters indicate statistically significant differences (p < 0.05) between different days of storage within the same treatment. Abbreviations: Sig. = statistical significance; N.S. = not significant; *** = p < 0.001; CTR: control; TA50: sample added with 0.005% of tannin-rich extract A; TA200: sample added with 0.020% of tannin-rich extract A; TA400: sample added with 0.040% of tannin-rich extract A; TB50: sample added with 0.005% of tannin-rich extract B; TB200: sample added with 0.020% of tannin-rich extract B; TB400: sample added with 0.040% of tannin-rich extract B; RE2000: sample added with 0.200% of rosemary extract.

.

All samples initially presented pH values ranging from 5.62 to 5.69. At T1, a significant (p < 0.05) increase was observed across all samples, with values ranging from 5.71 to 5.76 (p < 0.001); however, no differences (p > 0.05) were detected between samples. Subsequently, at T2, a significant pH decrease was recorded, ranging from 5.41 to 5.54 (p < 0.001). Again, no differences (p > 0.05) were revealed within the samples and the control. Overall, these results showed that the extracts had no impact on the pH of meat patties.

3.6. Colour

CIELAB L*, a*, and b* values were analysed to investigate colour changes in samples during storage. Results are as reported in

Table 3. CIELAB values L*, a*, and b* during the storage period of beef patties.

Colour
	Treatment	Time of Storage				p-Value
L* value		T0	T1	T2	Sig.	A	T	A × T
	CTR	30.85 ± 3.94 B	33.08 ± 4.13 AB	35.62 ± 1.81 A	*	0.084	<0.001	0.415
	TA50	30.71 ± 3.42	31.71 ± 3.03	32.34 ± 4.80	N.S.
	TA200	28.94 ± 1.07 B	30.38 ± 2.60 B	35.47 ± 4.52 A	***
	TA400	29.63 ± 3.37	31.98 ± 3.60	31.78 ± 5.33	N.S.
	TB50	27.97 ± 1.82 B	29.33 ± 1.99 B	33.50 ± 1.28 A	***
	TB200	28.33 ± 1.26 B	29.30 ± 3.82 B	39.93 ± 2.21 A	**
	TB400	28.89 ± 1.82 B	30.36 ± 3.48 B	35.37 ± 5.48 A	**
	RE2000	28.84 ± 1.49 B	31.19 ± 2.52 B	34.48 ± 3.60 A	***
	Sig.	N.S.	N.S.	N.S.
a * value		T0	T1	T2	Sig.	A	T	A × T
	CTR	7.74 ± 2.17 bA	5.06 ± 1.43 bB	3.37 ± 0.91 B	***	<0.001	<0.001	<0.001
	TA50	12.44 ± 2.02 aA	6.83 ± 0.91 aB	3.62 ± 1.35 C	***
	TA200	10.94 ± 1.37 aA	5.56 ± 1.82 aB	3.62 ± 1.09 C	***
	TA400	11.17 ± 2.50 aA	7.45 ± 0.49 aB	4.19 ± 0.76 C	***
	TB50	8.79 ± 1.06 bA	6.09 ± 0.74 bB	3.73 ± 0.57 C	***
	TB200	7.82 ± 0.90 bA	6.37 ± 0.56 abB	3.67 ± 0.38 C	***
	TB400	7.82 ± 1.49 bA	6.69 ± 0.68 abA	3.56 ± 0.70 B	***
	RE2000	7.72 ± 2.07 bA	6.71 ± 1.22 aA	4.16 ± 1.39 B	***
	Sig.	***	**	N.S.
b* value		T0	T1	T2	Sig.	A	T	A × T
	CTR	8.94 ± 1.78 B	8.64 ± 0.78 B	10.81 ± 1.34 A	*	0.009	<0.001	0.202
	TA50	9.01 ± 1.82	8.87 ± 1.50	9.99 ± 1.25	N.S.
	TA200	8.79 ± 1.06 B	8.66 ± 0.88 B	10.94 ± 1.37 A	***
	TA400	8.07 ± 0.65 B	8.86 ± 0.71 AB	9.13 ± 1.16 A	*
	TB50	8.91 ± 2.30	8.19 ± 0.66	9.01 ± 1.03	N.S.
	TB200	8.25 ± 0.93 B	9.57 ± 1.76 AB	10.53 ± 1.65 A	*
	TB400	8.43 ± 1.55	9.53 ± 0.97	9.88 ± 1.67	N.S.
	RE2000	8.22 ± 0.54 C	9.31 ± 0.64 B	10.54 ± 1.25 A	***
	Sig.	N.S.	N.S.	N.S.

Results are expressed as the mean ± standard deviation of three independent replicates (n = 3) and results of two-way ANOVA with Tukey’s post hoc test are reported. The corresponding significance levels (p-value) are provided. “A” indicates the treatments effect, “T” indicates storage time, and “A × T” represents the interaction between treatment and storage time. Different lowercase letters indicate means statistically different (p < 0.05) within the same day of analysis between the different treatments, while different uppercase letters indicate means statistically different (p < 0.05) between different days of storage within the same treatment. Abbreviations: Sig. = statistical significance; N.S. = not significant; * = p < 0.05; ** = p < 0.01; *** = p < 0.001; CTR: control; TA50 sample added with 0.005% of tannin-rich extract A; TA200 sample added with 0.020% of tannin-rich extract A; TA400 sample added with 0.040% of tannin-rich extract A; TB50 sample added with 0.005% of tannin-rich extract B; TB200 sample added with 0.020% of tannin-rich extract B; TB400 sample added with 0.040% of tannin-rich extract B; RE2000 sample added with 0.200% of rosemary extract.

.

Regarding the L* value, no significant differences (p > 0.05) were observed between the CTR and the enriched samples, exhibiting an increase in lightness over time (p < 0.001). The addition of the natural extracts resulted in no change in L* values during the whole experiment. About the a* value, bright red colour naturally decreased during the entire storage period; however, a significant interaction between treatment and time (p < 0.05) was highlighted, demonstrating the antioxidant ability to protect patties’ red colour. The CTR presented at T0 a value of 7.74 and then significantly decreased at T1 (5.06, p < 0.001). At T2, the value settled at 3.37, not significantly lower than the T1 value. At T0, TA, at all concentrations evaluated, presented significantly higher values compared with the control, presenting a* values of 12.44 (p < 0.001), 10.94 (p < 0.001), and 11.17 (p < 0.001), when TA50, TA200, and TA400 were tested. At T1, all the a* values of TA decreased compared to T0 (p < 0.001); however, they displayed a brighter red colour with respect to CTR (p < 0.01), with a* values of 6.83, 5.56, and 7.45, respectively. However, at T2, the differences were no longer significant, and the a* values of each TA concentration were comparable to that one of the control. On the other hand, TB did not display significant differences (p > 0.05) throughout the entire storage period compared to the control. Similarly, RE at T0 showed no variation from the control. However, as reported for TA, at T1, RE exhibited a redder colour with respect to the control (p < 0.01), presenting an a* value of 6.71 compared to 5.06 for the control. Similar to L*, the b* value showed an increase during the storage time for all the evaluated samples, and neither the tannin-rich extracts nor RE could influence the yellowness in the enriched meat patties (Table 3).

3.7. Texture Analysis

Table 4. Texture evaluated as consistency and hardness during the storage period of beef patties.

Hardness (N)
	Treatment	Time of Storage				p-Value
		T0	T1	T2	Sig.	A	T	A × T
	CTR	11.44 ± 1.43	12.01 ± 2.21	13.87 ± 3.81	N.S.	0.665	0.475	0.237
	TA50	11.40 ± 2.02	13.01 ± 2.02	12.64 ± 1.61	N.S.
	TA200	11.65 ± 1.69	13.54 ± 2.33	14.20 ± 3.53	N.S.
	TA400	11.35 ± 2.47	13.64 ± 2.78	12.92 ± 1.89	N.S.
	TB50	11.30 ± 0.78	12.38 ± 1.67	11.80 ± 1.94	N.S.
	TB200	11.85 ± 1.52	12.99 ± 1.81	12.34 ± 2.03	N.S.
	TB400	11.39 ± 1.84	12.74 ± 1.11	12.31 ± 1.47	N.S.
	RE2000	11.27 ± 1.69	12.73 ± 1.75	12.42 ± 1.86	N.S.
	Sig.	N.S.	N.S.	N.S.
Consistency (N*s)
		T0	T1	T2	Sig.	A	T	A × T
	CTR	1403.57 ± 308.11 B	1833.22 ± 375.83 A	2043.81 ± 171.28 A	***	0.837	<0.001	<0.001
	TA50	1461.93 ± 184.26 B	1841.56 ± 218.49 A	2128.43 ± 253.62 A	***
	TA200	1538.11 ± 236.72 B	1928.93 ± 154.66 A	2184.44 ± 203.81 A	***
	TA400	1605.29 ± 205.41 C	1875.96 ± 246.31 B	2249.37 ± 193.87 A	***
	TB50	1354.43 ± 216.94 C	1783.46 ± 245.19 B	2068.82 ± 184.75 A	***
	TB200	1498.43 ± 301.57 C	1876.29 ± 183.24 B	2179.72 ± 205.49 A	***
	TB400	1467.41 ± 177.28 C	1937.82 ± 174.35 B	2583.21 ± 210.92 A	***
	RE2000	1743.75 ± 351.54	1868.76 ± 338.03	1823.32 ± 173.44	N.S.
	Sig.	N.S.	N.S.	N.S.

Results are expressed as the mean ± standard deviation of three independent replicates (n = 3) and results of two-way ANOVA with Tukey’s post hoc are reported. The corresponding significance levels (p-value) are provided. “A” indicates the treatments effect, “T” indicates storage time, and “A × T” represents the interaction between treatment and storage time. Different uppercase letters indicate means statistically different (p < 0.05) between different days of storage within the same treatment. Abbreviations: Sig. = statistical significance; N.S. = not significant; *** = p < 0.001; CTR: control; TA50 sample added with 0.005% of tannin-rich extract A; TA200 sample added with 0.020% of tannin-rich extract A; TA400 sample added with 0.040% of tannin-rich extract A; TB50 sample added with 0.005% of tannin-rich extract B; TB200 sample added with 0.020% of tannin-rich extract B; TB400 sample added with 0.040% of tannin-rich extract B; RE2000 sample added with 0.200% of rosemary extract.

reports the results of the texture attributes evaluated, namely consistency and hardness, which are fundamental to examine the impact on the acceptability of beef patties by consumers.

These results suggest that the presence of extracts did not influence in any way the texture profile of patties. In more detail, each sample showed no significant differences (p > 0.05) for hardness over the storage period, indicating a certain degree of meat stability. In addition, at each timepoint evaluated (i.e., T0, T1, T2), no differences were found between the enriched meat samples and CTR. Similar results were obtained for the consistency attribute. Indeed, no differences were observed between the samples enriched with natural extracts and the control. However, a significant increase was observed for all the samples over time (p < 0.001).

3.8. Sensory Evaluation

Parameters such as colour, odour, flavour, aroma, tenderness, and juiciness, along with overall acceptability, were considered to perform the sensory evaluation of cooked beef patties, enriched or not with natural extracts. For tannin-rich extracts, only the highest concentrations (namely TA400 and TB400) were selected for sensory testing, as these conditions were considered the most likely to show differences compared with the control. As depicted in Figure 4, the addition of all the natural extracts did not affect the consumer perception during sensory evaluation; indeed, most of the evaluated attributes did not exhibit significant differences (p > 0.05) compared to CTR. However, when the patties were enriched with TA400, the tenderness was perceived as lower (p < 0.05) than in the CTR (detailed results are reported in Table S3 of Supplementary Materials). Nevertheless, the overall linking was not negatively influenced by this alteration, since similar scores were obtained among all samples, presenting values ranging from 4.75 to 6. Although no significant differences were found, the most liked samples were in the order CTR > TA400 > TB400 > RE2000 (Figure 4).

4. Discussion

The present study aimed to evaluate the potential of two distinct tannin-rich extracts as natural antioxidants in beef patties, comparing them with a commercial rosemary extract. A broad assessment of meat quality was carried out, including lipid oxidation, microbial stability, physical characteristics, and sensory properties.

Lipid oxidation, alongside microbial spoilage, is one of the most critical factors affecting meat quality, as it directly influences products’ shelf life and consumer acceptability. In beef patties, which contain a high fat content, lipid oxidation leads to quality deterioration, resulting in nutrient loss, the development of off-odours and off-flavours, the formation of toxic compounds, and economic losses [6]. Both the tannin-rich extracts (ellagitannins TA and gallotannins TB) and rosemary extract (RE) evaluated in this study were effective in delaying lipid oxidation compared to the control. The varying performance of the ellagitannins (TA) and gallotannins (TB) can be attributed to their distinct chemical structures and reactivity.

Gallotannins are biosynthesized by esterifying gallic acid with β-D-glucose to generate 1,2,3,4,6-pentagalloylglucose (PGG), which is a highly reactive structure. Enzymatic galloylation of PGG results in the creation of polymers with two galloyl groups covalently bound together by ester (depsidic) linkages. Gallotannins may efficiently scavenge free radicals and chelate transition metals due to the high density of free galloyl groups in these structures, neutralizing reactive oxygen species and preventing the start of lipid peroxidation [14]. On the other hand, the oxidation of PGG both inside and between molecules produces ellagitannins. This process results in a more complex molecular structure, with hexahydroxydiphenoyl (HHDP) biaryl units, where the phenolic groups become less accessible due to steric hindrance. Moreover, while the presence of multiple hydroxyl groups in the ortho position of ellagitannins enhances their antioxidant activity by facilitating hydrogen atom donation and neutralizing free radicals, the presence of sugar moieties in the structure reduces the accessibility of these hydroxyl groups [23]. As a result, ellagitannins exhibit slightly lower lipid oxidation inhibition compared to gallotannins, as confirmed by DPPH results. On the other hand, rosemary extract (RE) is rich in carnosol, carnosic acid, and rosmarinic acid. Differently from tannins, which predominantly act during the oxidation phase of oxidation, RE exerts its effect mainly during the propagation phase. The lipophilic diterpenes carnosic acid and carnosol partition into the lipid compartments of meat, intercepting peroxyl radicals (ROO•) and terminating chain reactions, thus preventing the accumulation of secondary lipid oxidation products [24]. Moreover, carnosic acid and carnosol exhibit regenerative antioxidant activity, whereby carnosic acid oxidizes to carnosol, which can be partially converted back, sustaining radical scavenging over time [25]. Thanks to their chemical composition, all extracts effectively kept MDA levels below 1 mg/kg, the threshold where rancidity becomes detectable by consumers [26]. Moreover, in the present study, hydrolysable tannins TA and TB outperformed other tannin-rich extracts, such as wine pomace and pomegranate peel, which required higher concentrations (2% and 1%, respectively) to achieve similar antioxidant effects in meat products [5,27]. Interestingly, tannin-rich extracts also showed a superior antioxidant activity compared to synthetic ascorbic acid (0.05% w/w) in beef patties [28].

The antioxidant effect of the natural extracts was further supported by the analysis of volatile organic compounds (VOCs). Hexanal and 1-octen-3-ol were selected as markers of lipid oxidation, as they result from the decomposition of alkyl hydroperoxides of linoleic acid and intermolecular cyclic rearrangement and cleavage of linoleic acid esters [3,29]. Specifically, gallotannins (TB) were most effective in reducing VOC accumulation, followed by ellagitannins (TA) and rosemary extract, highlighting the influence of both the chemical structure and phase-specific antioxidant mechanisms [18,24]. Indeed, TA and TB worked as hydroxyl radical scavengers and metal ion chelators in the initiation phase [30,31], while RE worked as chain-breaking antioxidants, terminating free radicals, scavenging peroxyl radicals, and inhibiting lipid peroxidation during the propagation phase [24].

Differently from the confirmed antioxidant capability, no microbial inhibition was observed, in contrast to many in vitro reports [32,33]. A possible explanation lies in the complexity of meat matrices. Indeed, the proteins can readily interact with tannins, limiting their availability to exert antimicrobial effects. This hypothesis is further supported by recent findings showing that the interaction between hydrolysable tannins and proteins does not occur through direct binding but rather through hydrogen-bonding networks mediated by water molecules. It has been demonstrated that phenolic hydroxyl groups of hydrolysable tannins preferentially interact with proteins via water bridges rather than through direct protein–tannin contacts [34]. Under the conditions of the present study, the addition of water to the meat matrix may have further enhanced these water-mediated interactions, increasing tannin–protein association and consequently reducing the fraction of free tannins available to interact with microbial cells. It should also be considered that the extracts could be more pronounced against specific microbial species not investigated in this study or effective at higher concentrations, as 1% (w/w) was reported to be effective [35].

In addition to lipid oxidation and microbial changes, colour represents a key quality attribute influencing consumers’ purchasing decisions, as it is closely associated with the perceived freshness of meat. During storage, the a* value naturally declined in all samples as a consequence of myoglobin oxidation, promoted by oxygen exposure and the presence of pro-oxidant metals, leading to the formation of metmyoglobin with its characteristic brown colour [36,37]. Nevertheless, all the examined natural extracts were able to retard this process, contributing to the preservation of myoglobin in its reduced ferrous state. This protective effect can be rationalized on the basis of the distinct chemical structures and reactivity of the extracts. In particular, red colour stability is strongly dependent on the ability to limit heme iron oxidation. Ellagitannins (TA), owing to their more rigid and partially oxidized structure characterized by hexahydroxydiphenoyl (HHDP) units, may interact more effectively with the heme environment or surrounding proteins, thereby limiting iron oxidation and delaying metmyoglobin formation. In contrast, gallotannins (TB), despite their superior efficacy in inhibiting lipid oxidation, appeared less effective in stabilizing the ferrous heme, likely due to their preferential reactivity with free radicals and metal ions in the lipid phase rather than direct interaction with the myoglobin iron centre. Rosemary extract (RE) also contributed to slowing colour deterioration, mainly by reducing the overall oxidative pressure through radical scavenging during the propagation phase, although this effect was insufficient to fully prevent redness loss over prolonged storage [38]. Interestingly, these results differ from those reported by Fruet et al. [39], who found that quebracho extract, rich in condensed tannins, did not preserve meat redness. This discrepancy is likely attributable to fundamental differences in tannin structure and reactivity. Condensed tannins are characterized by highly polymerized flavan-3-ol units, which tend to form strong, often irreversible complexes with proteins, limiting their mobility and accessibility within the meat matrix. As a result, their ability to chelate heme iron or to interfere with myoglobin oxidation may be reduced. In contrast, hydrolysable gallotannins, such as those predominant in TA and TB, with a lower degree of polymerization, chelate pro-oxidant iron species at early oxidation stages. This structural advantage determines their superior effectiveness in delaying metmyoglobin formation and preserving red colour. Another notable finding is that, unlike the observations reported by Fruet et al. [39], the natural coloration of all extracts, ranging from yellow to brownish tones, did not negatively affect the appearance of the patties, even at high concentrations. The changes in the red a* value, associated with myoglobin oxidation, were the main indicator of colour variation in the samples.

Alongside colour, texture is a key attribute influencing meat acceptability, encompassing parameters such as tenderness, chewiness, juiciness, springiness, and cohesiveness. Textural changes reflected protein degradation, moisture loss, and the intrinsic properties of connective tissue and collagen, leading to a reduction in tenderness [40,41,42]. In this study, none of the natural extracts significantly affected the hardness or overall texture of the patties. This contrasts with previous reports, in which rosemary, grape seed, or green tea extracts, containing different polyphenol compositions, produced measurable negative effects on texture [43]. This lack of impact can be rationalized by considering that the limited concentration of extracts determined few interactions between protein and tannins. Observed minor increases in hardness during storage were more likely driven by natural protein breakdown and moisture loss rather than by the addition of the extracts [41,42]. Regarding consistency, the observed increase could be correlated to the meat composition, as the presence of connective tissue and collagen contributes to reduced tenderness because of their ability to bind muscle fibres [40].

Based on the sensory analysis, no significant differences were highlighted among samples for most evaluated attributes, except for tenderness, for which the control was perceived as slightly softer than the samples containing extracts (p < 0.05). Literature findings support this trend; indeed, authors reported that the addition of grape pomace at 2% or 4% (w/w) reduced the tenderness and the overall liking of beef patties [41]; similar effects were observed by Fruet et al. [39], who noted a decrease in desirability when 0.5% (w/w) quebracho tannins were incorporated. Conversely, Al-Hijazeen et al. reported that tannic acid has improved the overall acceptability of patties [44]. Collectively, these studies suggest that natural compounds can modulate the sensory profile of meat. Nevertheless, in the present work, all samples received positive overall liking scores, indicating that the selected concentration of natural extract did not negatively influence consumer perception or acceptability.

5. Conclusions

The findings of the present study highlighted the strong antioxidant potential of both tannin-rich extracts (ellagitannins, TA and gallotannins, TB) compared to rosemary extract (RE) in beef patties. TA, TB, and RE effectively reduced lipid oxidation, preserving the red colour and maintaining overall sensory quality, without detrimental effects on texture, even at the highest concentrations. Tannin-rich formulations outperformed rosemary in controlling lipid oxidation and maintaining colour, with TB showing the strongest antioxidant activity and TA being most effective in stabilizing the red a* value. These results emphasize the importance of the chemical structure and antioxidant mechanisms in determining efficacy. Unfortunately, this study was limited by the concentration range tested and the short shelf life evaluated, as only a single tasting trial was conducted. Future research should address these limitations by testing a wider range of concentrations, exploring potential synergistic effects between different extracts, and evaluating antimicrobial activity against an increased range of microorganisms. Additionally, extending the shelf-life evaluation, testing different storage conditions, such as Modified Atmosphere Packaging (MAP) or vacuum, and repeating the sensory tests at the end of the preservation period would provide further insights into the long-term effects of these natural extracts.