Comparative Evaluation of Green Extraction Technologies for Phenolic Compounds from Algerian Blackthorn (Prunus spinosa L.): Antioxidant, Antimicrobial, and Phytochemical Insights

TL;DR

This study compares green extraction methods to recover phenolic compounds from Algerian blackthorn fruits, finding pressurized liquid extraction most effective for food and nutraceutical use.

Contribution

Systematic comparison of green extraction technologies for phenolic recovery from Algerian blackthorn, identifying pressurized liquid extraction as the most effective method.

Findings

Pressurized liquid extraction yielded highest total phenolic and flavonoid content from blackthorn fruits.

Microwave-assisted extraction showed highest anthocyanin recovery and comparable antioxidant capacity.

Extracts exhibited antimicrobial activity against MRSA and other bacteria, with flavonoids and hydroxycinnamic acids as major compounds.

Abstract

Blackthorn (Prunus spinosa L.) is an underutilized Mediterranean wild fruit recognized as a valuable source of bioactive phenolic compounds with potential applications in food and nutraceutical formulations. Despite growing interest in sustainable extraction approaches, systematic comparisons of green extraction technologies for blackthorn phenolic recovery remain limited, particularly for North African ecotypes. In this study, four non-conventional green extraction techniques, pressurized liquid extraction, microwave-assisted extraction, ultrasound-assisted extraction, and Ultra-Turrax-assisted extraction, were compared for the recovery of phenolic compounds from Algerian blackthorn fruits under method-specific controlled conditions. Total phenolic compounds, flavonoids, anthocyanins, and condensed tannins were quantified, together with antioxidant capacity evaluated using multiple complementary assays. Pressurized liquid extraction yielded the highest total phenolic compound content (21.89 mg gallic acid equivalents/g dry weight) and flavonoid content (8.18 mg catechin equivalents/g dry weight), while microwave-assisted extraction showed the highest anthocyanin recovery (3.19 mg cyanidin-3-glucoside equivalents/g dry weight). Antioxidant capacity values obtained by different assays showed comparable trends for pressurized liquid extraction and microwave-assisted extraction. The extract obtained by the most effective method was further evaluated for antimicrobial activity, showing selective inhibitory effects against methicillin-resistant Staphylococcus aureus, Acinetobacter baumannii, and Bacillus subtilis. Chemical characterization by liquid chromatography–tandem mass spectrometry revealed a profile dominated by flavonoids, mainly quercetin derivatives, and hydroxycinnamic acids. Overall, these findings highlight pressurized liquid extraction as a promising green technology for the valorization of blackthorn fruits as sources of phenolic compounds for food-related applications.

Figures

Full text

1. Introduction

Phenolic compounds represent one of the most structurally diverse classes of plant secondary metabolites, encompassing more than 8000 molecules that differ widely in polarity, molecular weight, and chemical reactivity [1]. These compounds range from simple phenolic acids to highly polymerized tannins and are unevenly distributed across plant tissues, frequently occurring in bound forms that limit their extractability. In plants, phenolics play essential physiological roles, contributing to pigmentation, defense against biotic stressors, and key sensory attributes such as bitterness and astringency, as exemplified by procyanidins in beverages like wine and tea [2,3]. Beyond their ecological functions, phenolic compounds have attracted increasing interest in food and nutrition research due to their strong antioxidant capacity and their association with protective effects against oxidative stress–related disorders, including cardiovascular, neurodegenerative, and metabolic diseases [4]. Prunus spinosa L., commonly known as blackthorn or sloe, is a thorny wild shrub belonging to the Rosaceae family that can reach up to 4 m in height. It is widely distributed across Europe, western Asia, and the Mediterranean basin [5]. In Algeria, blackthorn is traditionally referred to as “Barkouk lemaiz” in Arabic and “Lvarquq tɣeten” in Kabyle Berber. During autumn, the plant produces small black drupes covered with a pale violet bloom [6]. These fruits are characterized by an extremely astringent and acidic taste due to their high tannin content, which limits their direct consumption and favors their processing into beverages, preserves, and other derived products [7]. Historically, sloe berries have been used in traditional medicine since the nineteenth century, mainly for the treatment of coughs, and have also been reported as anti-inflammatory, diuretic, laxative, and antispasmodic agents [8]. Phytochemical investigations have revealed that blackthorn fruits are rich in bioactive compounds, including flavonoids, phenolic acids, and anthocyanins, which contribute to their pronounced antioxidant activity and ability to mitigate oxidative stress [5]. Moreover, several studies have demonstrated antibacterial activity of Prunus spinosa fruit extracts against both Gram-negative bacteria, such as Escherichia coli and Pseudomonas aeruginosa, and Gram-positive bacteria, including Clostridium perfringens and Bacillus subtilis, with these effects largely attributed to their phenolic composition [8]. The extraction of phenolic compounds represents a critical step in the valorization of plant-based bioresources, particularly in the context of functional foods and nutraceutical development [9]. Extraction techniques are commonly classified into conventional and non-conventional approaches. Traditional methods, such as maceration, Soxhlet extraction, percolation, and decoction, remain widely used due to their simplicity and reproducibility; however, they are often associated with long extraction times, high solvent consumption, and elevated energy demands [10]. In response to these limitations, a variety of innovative extraction technologies have been developed to enhance efficiency while improving environmental sustainability. Among non-conventional techniques, ultrasound-assisted extraction (UAE) relies on acoustic cavitation to disrupt plant cell walls and enhance mass transfer, thereby significantly reducing extraction time [11,12]. microwave-assisted extraction (MAE) exploits dielectric heating to rapidly rupture cellular structures, enabling the selective recovery of polar compounds within minutes [9], whereas high-shear devices such as Ultra-Turrax-assisted extraction (UTAE) promote intracellular compound release through intense mechanical forces. Pressurized liquid extraction PLE combines elevated temperature and pressure to improve solvent penetration and solute diffusion, resulting in shorter extraction times and enhanced yields [13,14]. PLE has proven particularly effective for extracting polyphenols from medicinal plants such as Laurus nobilis leaves [15], bioactive compounds from Moringa oleifera [16], and various thermolabile compounds from plant matrices and agro-industrial waste. The technique’s ability to operate at subcritical conditions with green solvents such as water and ethanol while maintaining high extraction efficiency makes it a sustainable alternative for the recovery of bioactive compounds [17].

The selection of an appropriate extraction technique depends on multiple factors, including the nature of the plant matrix, the thermal sensitivity of target compounds, desired extraction yields, and economic and environmental considerations [18]. While UAE, MAE, UTAE, and PLE each offer distinct advantages through different physical mechanisms, comparative data are needed to guide method selection for specific matrices such as blackthorn fruits. Based on the distinct physical mechanisms underlying each extraction technique, we hypothesized that extraction efficiency and selectivity would vary among methods depending on their mode of action. Specifically, we expected that techniques combining thermal and pressure effects would differ in performance from purely mechanical approaches, and that individual methods might show preferential recovery of specific phenolic classes.

To test these hypotheses, the present study aims to: (i) comparatively evaluate four non-conventional extraction methods (PLE, MAE, UAE, UTAE) for the recovery of total phenolic content (TPC), total flavonoid content (TFC), total anthocyanin content (TAC), condensed tannin content (CTC), and antioxidant activity (DPPH, FRP, and ABTS); (ii) assess the antimicrobial activity of the extract obtained by the most efficient technique against selected Gram-positive and Gram-negative bacteria; and (iii) characterize the phenolic profile of Algerian blackthorn fruits using liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis. Conventional extraction methods (maceration, Soxhlet, and percolation) were not included in the present comparison, as their performance and limitations, such as long extraction times, high solvent consumption, and the possible thermal degradation of thermolabile phenolic compounds, have already been extensively documented for fruit matrices [19]. Accordingly, this study focuses on the comparative evaluation of selected green extraction technologies, aiming to assess their relative performance under method-specific conditions.

2. Materials and Methods

2.1. Chemicals and Reagents

All chemicals and reagents used in this study were of analytical or LC–MS grade. Ethanol, methanol, acetonitrile, formic acid, Folin–Ciocalteu reagent, sodium carbonate (Na_2_CO_3_), aluminum chloride (AlCl_3_), sodium nitrite (NaNO_2_), sodium hydroxide (NaOH), potassium chloride (KCl), sodium acetate, ammonium ferric sulfate, n-butanol, hydrochloric acid, potassium ferricyanide (K_3_Fe(CN)6), ferric chloride (FeCl_3_), trichloroacetic acid (TCA), and potassium persulfate (K_2_S_2_O_8_) were purchased from standard commercial suppliers and used without further purification.

Gallic acid, catechin, cyanidin-3-glucoside, Trolox, and ascorbic acid were used as reference standards for calibration curves and antioxidant assays. DPPH (2,2-diphenyl-1-picrylhydrazyl) and ABTS (2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)) reagents were employed for radical scavenging assays. Ultrapure water was obtained using a Milli-Q purification system and used throughout all experiments. All solvents employed for LC–MS/MS analysis were of HPLC or LC–MS grade.

2.2. Preparation of Prunus spinosa L. Fruit Powder

Ripe blackthorn (Prunus spinosa L.) fruits were collected in September 2023 Minar Zareza, Mila region, eastern Algeria (geographical coordinates: 36°31′33.7″ N 5°52′38.4″ E). Specialized agents from the Forest Conservation Department of Mila, who accompanied the authors during sampling, botanically identified the plant material as Prunus spinosa L. (Rosaceae) based on morphological characteristics (including spiny branches, small dark fruits, leaf shape and arrangement, and overall growth habit) combined with their expert knowledge of the species’ natural distribution, allowing reliable field identification. Due to the seasonal timing of collection, a herbarium voucher specimen could not be prepared. The sampling sites were located in semi-arid Mediterranean climate zones. The blackthorn shrubs grew under full sun exposure in natural, uncultivated conditions. The freshly harvested fruits were initially air-dried at ambient conditions (30 ± 5 °C, relative humidity 40 ± 5%) for 20 days. After manual removal of the pits, the whole fruit, including the peel, was further dried in a ventilated oven (Memmert UF55, Schwabach, Germany; forced air circulation, 100% fan speed) at 40 °C for three days. The moisture content was determined gravimetrically on ground samples dried at 105 °C until constant mass. The final moisture content was 5.39 ± 0.21%. The dried fruits were then finely ground and passed through a 250 µm mesh sieve to obtain a homogeneous powder. The powder was stored at 4 °C in airtight containers, protected from light and moisture, and used for extraction after one year. This extended storage period represents a limitation of the study, as some degradation of phenolic compounds and anthocyanins may have occurred despite the protective conditions. However, all extraction methods were applied to the same batch under identical conditions, ensuring the validity of the comparative evaluation.

2.3. Extraction Procedure

Four non-conventional extraction techniques were evaluated: UAE, MAE, UTAE, and PLE. Each extraction used 5 g of dried fruit powder with 100 mL of 70% ethanol, maintaining the 1:20 (w/v) ratio. For PLE, the solid-to-solvent ratio was maintained by loading 0.55 g of powdered fruit mixed with 6 g of marine sand as a dispersing agent into an 11 mL stainless steel extraction cell, with 70% ethanol as the extraction solvent.

Extraction parameters, including extraction time, temperature, power, stirring speed, and pressure, were selected based on representative and widely validated conditions reported in the literature for Prunus spinosa, other species of the Rosaceae family, or comparable Mediterranean fruit matrices processed using similar extraction technologies. It should be noted that the extraction methods were conducted at different temperatures and mechanical mechanisms (under typical literature conditions), so the comparison reflects commonly reported literature conditions rather than strictly standardized thermal loads.

As comprehensive optimization studies specifically focused on blackthorn fruits are still limited, the present work was intentionally designed as a standardized comparative screening of extraction technologies under literature-based operating conditions, rather than a full multivariate optimization study. This approach allows a robust comparison of extraction efficiency and selectivity while maintaining experimental feasibility and industrial relevance.

UAE was performed in an ultrasonic bath (Branson 2510EDTH, Brookfield, CT, USA, 40 kHz, 2.8 L) at 80 °C for 30 min. Extraction vessels were tightly sealed to prevent ethanol evaporation. Bath temperature was continuously monitored and maintained at 80 °C throughout the extraction period. MAE was carried out using a domestic microwave oven (MW8123ST; Samsung Electronics, Klang, Malaysia) modified with a condenser to enable reflux and prevent solvent loss at 500 W for 4 min. Temperature was not monitored during this study, which represents a limitation in the thermal characterization of this method. Consequently, mechanistic interpretations regarding MAE thermal selectivity should be considered preliminary and require further investigation; therefore, extraction power and time were selected based on literature data to limit excessive thermal effects. PLE was conducted using an ASE 200 system (Dionex Corp., Sunnyvale, USA) in static mode at 120 °C and 10 MPa (1500 psi) for 10 min. The extraction cycle included an 8 min preheat phase, 5 min static extraction, a 20 vol flush, and a 60 s purge with nitrogen. One extraction cycle was performed per sample, yielding a collected volume of approximately 19 mL. Before use in extraction, each solvent was degassed using an ultrasound bath to prevent oxidation. UTAE was performed using an ULTRA-TURRAX homogenizer (IKA-Werke, Staufen, Germany) with a rotor-stator head at 15,000 rpm for 15 min at room temperature with temperature monitoring to prevent overheating.

The detailed operating conditions applied for each extraction method are summarized in Table 1.

2.4. Total Phenolic Content (TPC)

Total phenolic content was determined using the Folin–Ciocalteu colorimetric method according to Ali et al. [24]. Briefly, 200 µL of each extract was mixed with 1 mL of diluted Folin–Ciocalteu reagent, followed by the addition of 800 µL of 7.5% (w/v) sodium carbonate solution. The reaction mixture was incubated at room temperature for 30 min in the dark, and absorbance was measured at 765 nm using a UV–Vis spectrophotometer. TPC was quantified using a gallic acid calibration curve (0–1 mg/mL) and expressed as milligrams of gallic acid equivalents per gram of dry weight (mg GAE/g DW). Briefly, the absorbance values were converted to gallic acid equivalents (C_GAE_ in mg/mL) using the calibration curve Formula (1):

$$[eqn]$$

and TPC was calculated using the Formula (2):

$$[eqn]$$

where V is the total extraction volume (mL), m (g) is the dry sample mass, and DF = 20 is the dilution factor applied before measurement. It should be noted that the Folin–Ciocalteu assay reports the overall reducing capacity and can be influenced by non-phenolic reductants present in the matrix.

2.5. Total Flavonoid Content (TFC)

Total flavonoid content was measured following the aluminum chloride colorimetric method [24]. An aliquot of 400 µL of extract, standard, or distilled water (blank) was mixed with 120 µL of 5% (w/v) NaNO_2_ solution. After 5 min, 120 µL of 10% (w/v) AlCl_3_ was added, and after an additional 6 min, 800 µL of 1 M NaOH was introduced. The absorbance was recorded at 510 nm against the blank. Quantification was performed using a catechin calibration curve (0–1 mg/mL), and results were expressed as milligrams of catechin equivalents per gram of dry weight (mg CE/g DW). Briefly, the absorbance values were converted to catechin equivalents (C_CE_ mg/mL) using the calibration curve Formula (3):

$$[eqn]$$

and TFC was calculated using the Formula (4):

$$[eqn]$$

where V (mL) is the total extraction volume, m is the dry sample mass, and DF = 10 is the dilution factor applied before measurement.

2.6. Condensed Tannin Content (CTC)

Condensed tannin content was determined using the butanol–HCl assay as described by Vermerris et al. [25]. Briefly, 250 µL of extract was mixed with 2.5 mL of acidified ferric reagent prepared by dissolving 77 mg of ammonium ferric sulfate [NH_4_Fe(SO_4_)2] in 500 mL of n-butanol/HCl (3:2, v/v). The mixture was heated at 95 °C for 50 min, and then cooled to room temperature. The absorbance was measured at 530 nm.

Condensed tannin concentration was calculated using the following Equation (5):

$$[eqn]$$

where A_530_ is the absorbance at 530 nm, DF is the dilution factor (DF = 3), MW is the molecular weight of cyanidin (287 g/mol), V(L) is the final volume of the reaction mixture, ε is the molar extinction coefficient (34,700 L·mol^−1^·cm^−1^), L is the optical path length of the cuvette (1 cm), m (g) the mass of dry sample corresponding to the extract volume used, and 1000 is the conversion factor from g/g to mg/g. Results were expressed as milligrams of cyanidin equivalents per gram of dry weight (mg CYE/g DW).

2.7. Total Anthocyanin Content (TAC)

Total anthocyanin content was determined using the pH differential method as described by Oğuz et al. [26]. Two dilutions of each extract were prepared using potassium chloride buffer (pH 1.0; 1.86 g/L) and sodium acetate buffer (pH 4.5; 54.43 g/L). After equilibration for 15 min at room temperature, absorbance was measured at 510 and 700 nm.

TAC was calculated according to the following Equation (6):

$$[eqn]$$

where A = (A_510_ − A_700_)pH 1.0 − (A_510_ − A_700_)pH 4.5 is the absorbance difference, DF is the dilution factor (DF = 10), M is the molecular weight of cyanidin-3-glucoside (449.2 g/mol), V(L) is the total volume of extraction solvent (L), K is the molar absorptivity coefficient (26,900 L·mol^−1^·cm^−1^), L is the path length (1 cm), m is the dry weight of the sample (g), and 1000 is the conversion factor from g to mg. Results were expressed as milligrams of cyanidin-3-glucoside equivalents per g of dry weight.

2.8. DPPH Radical Scavenging Activity

DPPH radical scavenging activity was determined according to the method described by Salem et al. [27]. Briefly, 50 µL of Prunus spinosa L. extract or ascorbic acid (used as reference standard) was mixed with 1.45 mL of a 0.06 mM DPPH ethanolic solution. The reaction mixture was incubated for 30 min in the dark at room temperature, and absorbance was measured at 515 nm using ethanol as a blank. DPPH inhibition was calculated as (Equation (7)):

$$[eqn]$$

where A_0_ and A_1_ are the negative control and test sample absorbances, and converted to mg ascorbic acid equivalents per gram dry weight (mg AAE/g DW) using an ascorbic acid calibration curve (50–250 µg/mL, y = 0.346x − 0.2887, R^2^ = 0.99). The final activity was calculated as (Equation (8)):

$$[eqn]$$

where C is the concentration obtained from the curve (µg/mL), DF is the dilution factor (20), V is extraction volume (mL), m is sample dry weight (g), and 1000 converts µg to mg. All measurements were performed in triplicate.

2.9. ABTS Radical Cation Decolorization Assay

The ABTS radical cation scavenging activity was evaluated following the method reported by Salem et al. [27]. The ABTS•^+^ radical was generated by reacting a 7 mM ABTS aqueous solution with 2.45 mM potassium persulfate (K_2_S_2_O_8_) in the dark at room temperature (20 ± 5 °C) for 16 h. Prior to analysis, the radical solution was diluted with ethanol to obtain an absorbance of 0.70 ± 0.02 at 734 nm. An aliquot of 1 mL of ABTS•^+^ solution was mixed with 10 µL of extract or Trolox standard and incubated for 6 min at 30 °C. Absorbance was then recorded at 734 nm. Antioxidant activity was calculated using a Trolox calibration curve (0–1.5 mM) and expressed as millimoles of Trolox equivalents (TE) per gram of dry weight (DW). ABTS inhibition was calculated as (Equation (9)):

$$[eqn]$$

where Ac and As represent the absorbance values of the negative control and the tested sample, respectively. Antioxidant activity was expressed as millimoles of Trolox equivalents per gram of dry weight (mmol TE/g DW) using a Trolox calibration curve ranging from 0 to 2.5 mM. The calibration curve was described by: y = 44.472x (R^2^ = 0.98). The final antioxidant activity was calculated according to the following Equation (10):

$$[eqn]$$

where C is the Trolox concentration from the calibration curve (mM), DF is the dilution factor (DF = 5), V is the extraction volume (L), and m is the sample dry weight (g).

2.10. Ferric Reducing Power (FRP) Assay

Ferric reducing power was determined according to the method described by Oğuz et al. [26], with minor adaptations. Briefly, 1 mL of extract was mixed with 2.5 mL of phosphate buffer (0.2 M, pH 6.6) and 2.5 mL of 1% (w/v) potassium ferricyanide (K_3_Fe(CN)6). The mixture was incubated at 50 °C for 20 min, after which 2.5 mL of 10% (w/v) trichloroacetic acid (TCA) was added, and the mixture was centrifuged at 1500× g for 10 min. Subsequently, 2.5 mL of the supernatant was combined with 2.5 mL of distilled water and 0.5 mL of 0.1% (w/v) ferric chloride (FeCl_3_). Absorbance was measured at 700 nm. Antioxidant capacity was quantified using a gallic acid calibration curve: y = 12.978x + 0.0181 (R^2^ = 0.99) and expressed as gallic acid equivalents (GAE). Ferric reducing power was calculated as (Equation (11)):

$$[eqn]$$

where C is the gallic acid concentration from the calibration curve (mg/mL), DF is the dilution factor (DF = 20), V is the extraction volume (mL), and m is the sample dry weight (g).

This approach was adopted to maintain consistency with total phenolic content determination and to facilitate the interpretation of electron-transfer–based reducing capacity in relation to phenolic concentration. It should be noted that this assay corresponds to a ferric reducing power (FRP) method and not to the classical FRAP (TPTZ/Fe^3+^) assay; therefore, results are intended for intra-study comparison among extraction methods rather than direct inter-study FRAP benchmarking.

2.11. Antimicrobial Activity

Based on the comparative extraction study, the extract obtained by the method yielding the highest phytochemical content and antioxidant activity (PLE) was selected for antimicrobial evaluation. The antimicrobial activity of Algerian blackthorn extract was evaluated using the agar well diffusion method, a semi-quantitative preliminary screening assay. The extract was tested at 20 mg/mL, a concentration chosen based on solubility and literature reports [8,28]. The dry extract was dissolved in 80% (w/v) methanol prior to testing.

A total of eight bacterial strains were tested, including three Gram-positive bacteria: Bacillus subtilis (ATCC 6633), Enterococcus faecalis (ATCC 29212), and methicillin-resistant Staphylococcus aureus (MRSA) (ATCC 43300); and five Gram-negative bacteria: Escherichia coli (ATCC 25922), Klebsiella pneumoniae (ATCC 13883), Acinetobacter baumannii (ATCC 610), Pseudomonas aeruginosa (ATCC 6633), and Salmonella enterica (ATCC 14028).

Bacterial inocula were prepared by culturing fresh colonies on nutrient agar at 37 °C for 24 h, followed by suspension in sterile physiologic water and adjustment to 0.5 McFarland turbidity standard (approximately 1.5 × 10^8^ CFU/mL). The standardized suspensions were uniformly spread onto Mueller–Hinton agar plates with an agar depth of approximately 3.5–4 mm. Wells were then aseptically punched into the agar (6 mm diameter), and each well was filled with 50 µL of the blackthorn extract solution.

Amoxicillin (500 mg/5mL oral suspension) was used as a positive control at a volume of 50 µL per well, while 80% methanol served as the negative control. Plates were then kept at 4 °C for 2 h to allow initial diffusion before incubation at 37 °C for 24 h, after which antimicrobial activity was assessed by measuring the total diameter of inhibition zones (mm), including the well diameter, surrounding each well using a digital caliper. Given the non-standardized nature of the positive control and the qualitative nature of the assay, results are presented as a preliminary indication of antimicrobial potential rather than for quantitative comparison.

2.12. LC-MS/MS Analysis

The blackthorn extract obtained by pressurized liquid extraction was analyzed by LC–MS/MS for the identification of phenolic compounds, using an Agilent 1100 liquid chromatograph with a diode array detector (DAD) (Agilent Technologies, Santa Clara, CA, USA) coupled to a Bruker Esquire 2000 Ion Trap mass spectrometer (Bruker Corp., Bremen, Germany), with electrospray ionization (ESI) operated in both positive and negative modes, but only the negative mode was used for compound identification.

Prior to analysis, the dry extract was dissolved in methanol at a concentration of 10 mg/mL and filtered through 0.45 µm nylon syringe filters. This concentration and the 20 µL injection volume were optimized to ensure adequate signal intensity without column overload or ion suppression, considering the extract’s low phenolic content (21.89 mg GAE/g DW). Chromatographic separation was achieved on an Eclipse XDB-C18 column (4.6 × 150 mm, 5 µm; Agilent Technologies), maintained at room temperature (23 °C). The injection volume was set at 20 µL, and the flow rate was fixed at 0.8 mL/min.

The mobile phase consisted of solvent A (0.1% v/v formic acid in ultrapure water) and solvent B (0.1% v/v formic acid in acetonitrile). A gradient elution program was applied as follows: 5% B at 0 min, 15% B at 25 min, 25% B at 35 min, 35% B at 40 min, 50% B at 47 min, 95% B at 55 min, and 100% B at 60 min. UV–Vis spectra were recorded by the diode array detector at 254, 280, 320, 360, and 520 nm.

MS/MS spectra were acquired in Auto MS/MS mode, with the ESI source set at a capillary voltage of 3500 V, drying gas (N_2_) flow of 10 L/min, drying gas temperature of 350 °C, and nebulizer pressure of 40 psi. A post-run re-equilibration of 8 min was set at 5% B. Compounds were tentatively identified based on their retention times, UV–Vis spectra, molecular ions, and MS/MS fragmentation patterns, as no authentic standards were used.

2.13. Statistical Analysis

All experiments were performed in triplicate, and results are expressed as mean ± standard deviation. For each extraction method, three independent extractions (biological replicates) were performed, with each extract analyzed in technical triplicates. Statistical analyses used mean values of technical triplicates (n = 12: 4 methods × 3 biological replicates). Statistical analyses were conducted using R software (version 4.3.3). Prior to parametric testing, statistical assumptions were verified: normality was assessed within each treatment group using the Shapiro–Wilk test, and homogeneity of variances was confirmed using Levene’s test (“car” package). All parameters satisfied both assumptions (p > 0.05), validating the use of parametric analysis. Data were subjected to one-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD) post hoc test (“agricolae” package) to evaluate differences among extraction methods. Pearson correlation analysis was applied to assess relationships between phytochemical contents and antioxidant activities. Prior to PCA and HCA, data were mean-centered and scaled to unit variance. Multivariate analyses, including principal component analysis (PCA) (“FactoMineR” and “factoextra” packages) and hierarchical cluster analysis (HCA) (“ggdendro” package), were performed on mean-centered and unit-variance–scaled data to explore patterns and similarities among extraction methods. Graphical representations were generated using “ggplot2”, “corrplot”, “RColorBrewer”, “reshape2”, “gridExtra”, and “dplyr” packages. Statistical significance was set at p < 0.05.

3. Results and Discussion

3.1. Comparative Analysis of Extraction Methods

Four non-conventional extraction technologies, pressurized liquid extraction, microwave-assisted extraction, ultrasound-assisted extraction, and Ultra-Turrax-assisted extraction, were comparatively evaluated for their efficiency in recovering bioactive compounds from Prunus spinosa fruits. One-way ANOVA revealed highly significant differences among extraction methods for all phytochemical and antioxidant parameters (p < 0.0001; Table 2), confirming that extraction technology is a critical determinant of bioactive compound recovery from blackthorn berries [29]. The low coefficients of variation (0.35–6.19%) demonstrate excellent reproducibility at the laboratory scale. Among the tested techniques, PLE exhibited the highest extraction efficiency, yielding the greatest amounts of total phenol compounds (21.89 mg GAE/g DW) and total flavonoids (8.18 mg CE/g DW).

This superior performance can be attributed to the combined effects of elevated temperature and pressure, which enhance solvent penetration, decrease solvent viscosity and surface tension, and facilitate the disruption of interactions between phytochemicals and the plant matrix. Consequently, PLE extracts displayed high antioxidant activity in the DPPH assay (32.34 mg AAE/g DW), which was statistically comparable to UTAE (31.91 mg AAE/g DW) and significantly higher than UAE and MAE. For FRP activity was highest for MAE (28.43 mg GAE/g DW), and ABTS values were not significantly different between MAE (0.68 mmol TE/g DW) and PLE (0.71 mmol TE/g DW). These results support a close relationship between phenolic content and antioxidant capacity [30,31].

From a cost-effectiveness perspective, MAE offers lower equipment and energy costs, shorter extraction times, a smaller footprint, and proven industrial scalability, while maintaining good antioxidant activity and superior anthocyanin recovery. PLE, in contrast, provides the highest polyphenol and flavonoid yields but requires higher investment, more energy, and presents challenges for industrial-scale implementation [17].

The application of PLE to blackthorn fruits remains scarcely documented in the literature. Comparable results obtained by Pavez et al. [32] for olive pomace and by Gómez-López et al. [33] for Opuntia stricta var. Dillenii further corroborate the effectiveness of PLE for the recovery of phenolic compounds from plant matrices. Interestingly, MAE demonstrated the highest selectivity toward total anthocyanins (3.19 mg C3GE/g DW), even surpassing PLE. Although anthocyanins are generally considered thermolabile, this enhanced recovery could potentially be related to the rapid dielectric heating associated with microwave irradiation, which can induce instantaneous cell rupture and promote anthocyanin release before extensive thermal degradation occurs [34]. However, since the MAE temperature was not monitored in this study, mechanistic interpretations regarding thermal selectivity remain preliminary and require further investigation with controlled temperature measurements. In contrast, UAE and UTAE exhibited statistically comparable performances, with generally lower yields across most phytochemical and antioxidant parameters. These results suggest that mechanical or ultrasonic energy alone, without the synergistic contribution of thermal or pressure-assisted effects, may be insufficient to efficiently disrupt cell walls and release tightly bound phenolic compounds [11]. Condensed tannins (CTC) showed maximum values in PLE extracts (1.99 mg CYE/g DW), but differences among extraction methods were less pronounced compared to other phenolic classes. This relatively uniform extractability may be related to the polymeric nature and structural complexity of proanthocyanidins, which allow multiple solvent interactions regardless of the extraction technique employed [35].

Overall, no single extraction method consistently outperformed the others across all parameters. For total phytochemical content (TPC, TFC, TAC, and CTC), the efficiency followed the order: PLE > MAE > UTAE ≈ UAE. This trend was not consistent for antioxidant assays. FRP activity ranked as MAE > PLE > UTAE ≈ UAE, DPPH activity as UTAE ≈ PLE > MAE > UAE, and ABTS activity as PLE ≈ MAE > UTAE > UAE. These results highlight that extraction method selection should depend on the target compounds and intended application. While PLE generally gave the highest phytochemical yields, MAE excelled in ferric-reducing power, and UTAE achieved comparable DPPH scavenging under milder conditions. Significant F-values (182.2–1956.01, p < 0.0001) indicate that the extraction method strongly affects all measured parameters.

The Pearson correlation matrix (Figure 1), calculated using mean values of technical triplicates (n = 12: 4 methods × 3 biological replicates), reveals strong positive relationships between phytochemical contents and antioxidant capacities across all extraction methods. TPC, TFC, TAC, and CTC showed high intercorrelations (r = 0.864–0.995, p < 0.01), consistent with Muflihah et al. [36] who reported similar correlations (r = 0.913) between TPC and TFC in indigenous herbs, indicating that these phenolic classes co-extract together. TPC exhibited the strongest correlations with FRP (r = 0.935) and ABTS (r = 0.983), confirming that total phenolics are the primary driver of antioxidant capacity due to their hydroxyl groups that neutralize free radicals [37]. Similarly, TFC, TAC, and CTC showed very strong correlations with FRP (r = 0.966, 0.992, 0.906) and ABTS (r = 0.997, 0.940, 0.980), demonstrating their significant contributions to antioxidant activity. In contrast, DPPH exhibited notably weaker correlations with all phytochemical parameters (r = 0.159–0.401), which can be attributed to differences in reaction mechanisms and structural sensitivity. DPPH predominantly involves hydrogen atom transfer (HAT) with potential electron transfer (ET) contributions, while FRP and ABTS are primarily ET-based, though mixed mechanisms may occur under certain conditions. The lower DPPH correlations likely reflect its greater sensitivity to steric hindrance, whereas ET-dominant assays are less affected by molecular size and structural constraints [38].

The PCA biplot (Figure 2) shows that PC1 and PC2 explain 79.9% and 14.3% of total variance, respectively, giving 94.2% cumulative variance. PC1 mainly reflects differences among the extraction methods in terms of their measured responses. PLE is on the positive side of PC1 and is associated with TFC, CTC, ABTS, TAC, and FRP vectors. UAE and UTAE are on the negative side of PC1, indicating comparatively lower values for these variables. MAE is positioned between PLE and UAE/UTAE on PC1, suggesting an intermediate positioning based on the evaluated parameters. The TAC vector points toward MAE, indicating a closer association with anthocyanin-related variables compared to the other methods. The DPPH vector points in a different direction along PC2 compared to FRP and ABTS, suggesting a distinct contribution of this assay to the overall variance [37]. The vectors for ABTS, TFC, TAC, and CTC point in similar directions, consistent with the correlations observed in the correlation analysis. The separation observed among the four extraction methods suggests that the methods yield extracts with distinguishable phytochemical profiles, within the limits of the small number of samples [39].

The hierarchical clustering dendrogram (Figure 3) groups the extraction methods into two clusters based on phytochemical and antioxidant parameters. The first cluster includes UAE and UTAE, which merge at a height of 1.5, indicating similar extract compositions and antioxidant activities. This is consistent with their comparable performance observed in previous analyses. The second cluster contains MAE and PLE, which merge at a height of 1.9, suggesting that these methods also produce relatively similar phytochemical profiles. The two main clusters merge at height 6, demonstrating clear differences between the low-efficiency methods (UAE and UTAE) and the high-efficiency methods (MAE and PLE). This clustering pattern supports the extraction efficiency hierarchy established through statistical analysis, though these results should be interpreted as descriptive, given the limited number of extraction methods examined.

Overall, multivariate analyses consistently confirm the robustness of the extraction efficiency ranking and support the use of complementary antioxidant assays to capture different reaction mechanisms.

Several recent studies have investigated optimized extraction strategies for Prunus spinosa phenolic compounds. For example, Hourani et al. [40] optimized natural deep eutectic solvent extractions of blackthorn pomace and reported enhanced total phenolic content and antioxidant activity under tailored solvent compositions, which is consistent with the high phenolic yields observed with pressurized liquid extraction in the present study. Likewise, microwave-assisted extraction has been shown to effectively recover condensed tannins from P. spinosa branches, confirming the potential of energy-assisted methods for targeted phenolic recovery [41]. Additionally, the effects of classical solvent extraction parameters on total phenolic content and antioxidant profiles in blackthorn fruits have been documented, further demonstrating that process variables such as temperature and extraction time strongly influence phenolic accessibility, paralleling the methodological contrasts observed in this work [42]. Comparable trends have also been reported for other berry matrices, where combinations of green extraction solvents and techniques enhanced phenolic yields and antioxidant activities, reinforcing the broader applicability of sustainable extraction strategies beyond the subject species. Sallustio et al. [43] demonstrated that NADES-based green extraction of Rosa canina and Prunus spinosa resulted in higher phenolic contents and antioxidant activities compared to conventional solvents, supporting the effectiveness of green extraction media across different fruit matrices. Similar behavior has been observed in aronia (Aronia melanocarpa), where deep eutectic solvent systems selectively enhanced the extraction of phenolic acids and anthocyanins, with strong correlations between phenolic content and antioxidant capacity [44]. Overall, although direct head-to-head comparisons of green extraction technologies for Prunus spinosa remain limited, the present results are fully consistent with both species-specific and cross-species evidence, reinforcing the suitability of pressurized and energy-assisted green extraction approaches for maximizing phenolic recovery from blackthorn fruits.

3.2. Antimicrobial Efficacy of PLE Extracts

The antimicrobial evaluation of the blackthorn extract obtained by PLE (Table 3) revealed a selective antibacterial profile, with inhibitory activity observed against three of the eight tested strains. The highest efficacy was detected against methicillin-resistant Staphylococcus aureus (inhibition zone: 12.69 mm), followed by Bacillus subtilis (9.17 mm) and Acinetobacter baumannii (11.33 mm) (Figure 4). The observed activity was strain-specific rather than Gram-classification dependent. In contrast, no detectable inhibition was observed against Enterococcus faecalis, Klebsiella pneumoniae, Salmonella enterica, Escherichia coli, and Pseudomonas aeruginosa. Although previous studies have reported antibacterial activity of Prunus spinosa extracts against some Gram-negative strains [45,46,47], such variability is well documented in plant-derived antimicrobials and is strongly influenced by extraction technology, solvent polarity, phytochemical composition, and concentration of active metabolites. The reduced susceptibility of Gram-negative bacteria observed in the present study is consistent with their intrinsic structural resistance, particularly the presence of an outer lipopolysaccharide membrane that limits the penetration of many phenolic compounds, especially at moderate concentrations [48]. In this context, the selective activity of the PLE extract may reflect a phytochemical profile enriched in compounds more effective against Gram-positive targets, such as flavonols and phenolic acids, rather than a general loss of antimicrobial potential. Considering the qualitative nature of the agar diffusion assay and the non-standardized positive control, the results provide an initial assessment of antimicrobial activity rather than a definitive quantitative evaluation. Overall, these findings indicate that the PLE blackthorn extract does not exhibit broad-spectrum antibacterial activity, but rather a target-specific antimicrobial behavior, which is increasingly recognized as advantageous for food-related applications. Selective inhibition of pathogenic bacteria such as methicillin-resistant Staphylococcus aureus, while minimizing non-specific antimicrobial effects, may represent a desirable feature for the development of functional food ingredients or natural preservatives. Targeting specific contamination risks, such applications would require validation through quantitative antimicrobial endpoints (e.g., MIC, MBC, time-kill assays), stability assessments under relevant storage and processing conditions, and formulation compatibility in food matrices. It should be noted that the present study did not evaluate extract performance in real food systems or potential sensory impacts, which remain important considerations for practical implementation. While agar diffusion provides valid initial screening, quantitative assays would complement these results for a more comprehensive assessment of antimicrobial potency. Future studies incorporating concentration-response analyses could further characterize the antimicrobial potential of blackthorn extracts.

Values are expressed as mean ± standard deviation (n = 3). ND: not detected (no inhibition zone observed). MRSA: methicillin-resistant Staphylococcus aureus. PLE: pressurized liquid extraction. The positive control was amoxicillin, and the negative control was 80% (v/v) methanol.

3.3. Identification of Phenolic Compounds by LC-MS/MS

LC–MS/MS analysis of the blackthorn extract obtained by PLE enabled the identification of fifteen compounds (Figure 5), which were tentatively characterized based on their retention times, molecular ions [M–H]^−^, fragmentation patterns, and UV–Vis absorption maxima (Table 4). The phenolic profile was clearly dominated by flavonoids, particularly quercetin-derived glycosides, followed by phenolic acids, confirming the characteristic composition of Prunus spinosa fruits reported in previous studies [49]. Among phenolic acids, chlorogenic acid was identified as a major component, exhibiting a molecular ion at m/z 353 and characteristic MS/MS fragments at m/z 191, 179, and 173 (Figure 6), in agreement with literature data [50,51]. Chlorogenic acid is widely recognized as one of the most abundant hydroxycinnamic acid esters in fruits and contributes significantly to antioxidant activity [52]. The detection of quinic acid (m/z 191) and caffeic acid (m/z 179) as distinct chromatographic peaks further suggests partial hydrolysis of chlorogenic acid or the coexistence of free and esterified forms, a phenomenon previously observed in blackthorn and other Rosaceae species [50,53].

Other hydroxycinnamic derivatives, including ferulic acid and caffeoylshikimic acid, were also identified based on their characteristic fragmentation patterns [54,55]. These compounds are known for their antioxidant potential and contribute to the overall redox capacity of fruit extracts [52]. Although quinic acid is not a phenolic compound per se, its presence is relevant due to its role as a structural backbone for esterified hydroxycinnamates and its frequent co-occurrence with chlorogenic acid derivatives in plant matrices [56].

Flavonoids represented the predominant class of identified compounds, with quercetin and its glycosylated derivatives accounting for the majority of the flavonoid fraction. Two quercetin-O-rutinoside isomers (peaks 8 and 9) were among the most prominent compounds, displaying a molecular ion at m/z 609 and a characteristic fragment at m/z 301 corresponding to the quercetin aglycone after loss of the rutinoside moiety [57]. The chromatographic separation of these quercetin-O-rutinoside isomers (retention times 31.9 and 32.2 min) likely reflects positional or anomeric variations in the rhamnose-glucose linkage. The presence of other quercetin glycosides, including quercetin-O-glucoside, two quercetin-O-arabinoside isomers, quercetin-O-rhamnoside, and isorhamnetin-O-rutinoside, highlights the structural diversity of flavonol conjugates in blackthorn fruits [57,58,59,60]. The two quercetin arabinoside isomers, eluting at different retention times (34.6 and 35.3 min) but sharing the same molecular ion (m/z 433), likely reflect structural variations (positional, anomeric, or configurational) glycosylation, a common feature of flavonoid metabolism in fruits [61].

Free quercetin was also detected at m/z 301, showing typical MS/MS fragments at m/z 179 and 151 derived from retro-Diels–Alder cleavage [62]. Although present at lower relative abundance compared to its glycosides, quercetin aglycone is known to contribute substantially to antioxidant and biological activities due to its high redox potential [49,63].

Although anthocyanins were quantified spectrophotometrically, they were not detected by LC–MS/MS, because only the negative ionization mode was used for identification, which favors flavonols and phenolic acids; thus, the reported compounds represent a tentative identification, not the complete phenolic profile.

Overall, LC–MS/MS profiling revealed that the phenolic composition of Algerian blackthorn fruits is largely dominated by flavonoids, mainly quercetin-based glycosides, followed by phenolic acids, predominantly hydroxycinnamic derivatives.

It should be noted that these proportions refer to the distribution of tentatively identified compounds by chemical class, and do not represent quantitative abundance, as differences in ionization efficiency and response factors among phenolic subclasses may affect signal intensity. This compositional pattern supports the strong antioxidant performance observed for the PLE extract and suggests a selective enrichment of flavonols and phenolic acid esters by pressurized liquid extraction, reinforcing its suitability for the valorization of blackthorn fruits as functional food ingredients.

4. Conclusions

This study demonstrates that extraction technology significantly influences the phytochemical profile and biological properties of Prunus spinosa fruit extracts. Pressurized liquid extraction proved most effective for overall phenolic recovery and antioxidant capacity, while microwave-assisted extraction showed superior toward anthocyanins (3.19 mg C3GE/g DW), though the mechanistic basis for this selectivity requires further investigation under controlled thermal conditions. LC–MS/MS profiling revealed a composition dominated by flavonoids, mainly quercetin glycosides, followed by phenolic acids, predominantly hydroxycinnamic derivatives. The PLE extract exhibited selective antimicrobial activity against methicillin-resistant Staphylococcus aureus, Acinetobacter baumannii, and Bacillus subtilis. These findings highlight Algerian blackthorn as a promising underutilized fruit and identify PLE as a suitable green extraction technology for its valorization. Further studies should assess extract stability, performance in food matrices, and quantitative antimicrobial potency for potential applications in functional foods and food preservation.