Enhancement of Phenolic Recovery by Probe-Type Ultrasound-Assisted Extraction of Acerola By-Product and Evaluation of Antioxidant and Antibacterial Activities

Abstract

This study aimed to optimize the total phenolic content (TPC) and antioxidant activity (AA) of acerola (Malpighia emarginata) by-product extracts obtained by probe-type ultrasound-assisted extraction and assess the composition and antibacterial activity of the extract obtained under optimized conditions. A Box–Behnken experimental design was applied to evaluate the effects of ultrasonic power (350 to 650 W), ethanol concentration (20% to 80% v v⁻¹), and extraction time (20 to 60 min) on TPC and AA. The optimal extraction conditions were 650 W, 50% (v v⁻¹) ethanol, and 20 min, which yielded the highest values of TPC (3.36 g gallic acid equivalent 100 g⁻¹) and AA through the DPPH radical scavenging method (4.97 mM Trolox equivalents 100 g⁻¹) and a ferric reducing antioxidant power assay (11.35 mM Trolox equivalents 100 g⁻¹). Organic acids, phenolic acids, flavonoids, and alkaloids were identified in the optimized extract, including malic acid, protocatechuic acid, resorcylic acid, and rutin. The optimized extract (2.89–11.32 mg mL⁻¹) inhibited the growth of Listeria monocytogenes, Shigella sonnei, Salmonella enterica subsp. enterica Typhi, Staphylococcus aureus, Pseudomonas aeruginosa, and Escherichia coli. Acerola by-products represent a promising source of extracts with the potential to replace synthetic additives, contributing to the circular economy of agroindustries.

1. Introduction

Acerola (Malpighia emarginata DC.) is a fruit renowned for its high vitamin C content and presence of phenolic compounds and carotenoids. Acerola cultivation holds great economic importance in Brazil [1,2,3]. The fruit is commonly processed into juices and pulps, generating large amounts of waste, which can account for up to 40% of the processed volume. These materials are often used as animal feed or soil amendments. However, such applications fail to harness the potential of acerola by-products as raw materials and may pose environmental risks [4,5].

Studies have shown that acerola by-products contain several bioactive compounds [6,7,8,9,10]. Silva et al. [11] reported that an acerola by-product extract had a total anthocyanin content 70% higher than the pulp. Regarding total phenolic content, the authors reported higher levels in acerola by-products than in other fruit by-products, with concentrations approximately 1.1, 2.6, 3.7, 9.3, 16.1, and 19.3 times greater than those found in cashew, pineapple, guava, papaya, passion fruit, and mango by-products, respectively.

Extraction methods exert a direct influence on the yield of bioactive compounds extracted from natural sources [6,12]. Green extraction technologies are particularly attractive because they use smaller volumes of solvents, require shorter processing times, and often result in higher yields of target compounds [13,14,15]. Among these, ultrasound-assisted extraction (UAE) stands out for its effectiveness in extracting phenolic compounds from plant by-products, offering advantages in terms of yield, efficiency, and selectivity. These benefits largely stem from acoustic cavitation caused by ultrasound, which disrupts the cellular structure of the plant matrix, improves solvent penetration, and, consequently, enhances the release of bioactive compounds [15,16].

UAE can be carried out using either plate-type (pl-UAE) or probe-type (pr-UAE) systems. Typically performed in an ultrasonic bath, pl-UAE is more accessible and easier to operate. However, it tends to be less efficient and less reproducible, as the ultrasound is transmitted indirectly through a metal plate connected to a transducer. In contrast, pr-UAE delivers ultrasound directly into the sample via a metallic probe, concentrating the energy in a specific area. This approach results in higher intensity and greater extraction efficiency, making it more suitable for the extraction of bioactive compounds [15,17,18].

Studies have explored different methods for extracting phenolics from acerola by-products, including subcritical water extraction, orbital shaking, and ultrasonic bath methods [7,12,19,20,21]. The parameters used in these extraction methods affect the yield of phenolic compounds from acerola waste. As an example, Borges [21] reported that time and the liquid–solid ratio affected the phenolic extraction by hydrothermal treatment. Santos [22] achieved the highest content of bioactive compounds at 110 min, 20 °C, and 34% ethanol. Silva [7] performed pl-UAE of an acerola by-product and showed that a 67.5% ethanol concentration, temperature of 80.9 °C, liquid/solid ratio of 59.8 mL/g, and extraction time of 13.6 min promoted the highest yield of total phenolic content. However, to date, no study has investigated the pr-UAE of phenolic compounds from an acerola by-product, especially regarding the combination of ultrasonic power, ethanol concentration, and extraction time.

Bioactive compounds extracted from acerola by-products may possess antioxidant and antimicrobial properties [6,8,9,10], which are valuable for preserving processed products, particularly in the food industry. This application is in line with the increasing interest of consumers in healthier lifestyles, leading to changes in dietary habits and greater attention to nutritional value and food composition. Synthetic antimicrobials and antioxidants such as sodium nitrate, sodium nitrite, butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), tert-butylhydro-quinone (TBHQ), and propyl gallate are among the most widely used additives in packaged foods. However, their excessive and prolonged consumption has been associated with adverse health effects, including allergic reactions, gastrointestinal disorders, and an increased risk of cancer [23,24,25,26]. Given this scenario, there is growing interest in replacing synthetic compounds with natural, safer, and healthier alternatives.

In light of these considerations, this study aimed to optimize the extraction of antioxidant compounds from acerola by-products via pr-UAE and assess the composition and antibacterial activity of the extract obtained under optimized conditions.

2. Materials and Methods

2.1. Preparation of Acerola By-Product

Acerola fruits were collected in Umuarama, Paraná, Brazil (23°45′57″ S, 53°19′30″ O). The acerola by-products (seeds and shells) were obtained after pulp remotion (Fruit pulper, Macanuda, Joinville, Brazil). The waste was dried in an oven with forced air circulation (TE-39413, Tecnal, Piracicaba, Brazil) at 50 °C for 48 h. The dried material was ground in a knife mill (A11, Ika, Campinas, Brazil) and subjected to particle size classification on Tyler-type shaking sieves (Bertel, São Paulo, Brazil). The material retained on a 0.15 mm Tyler was stored in hermetic bags at −18 °C, until the time of analysis.

2.2. Extraction Procedure

The dehydrated acerola by-product was subjected to pr-UAE (UCD-950, BIOBASE, Jinan, China), using a proportion of 1 g of sample per 10 mL of solvent, as defined in preliminary tests. The pulse ON and OFF were two and four seconds, respectively. To evaluate the effect of ultrasound power (X₁), ethanol concentration (X₂), and time (X₃) on the extraction of the compounds of interest, a Box–Behnken factorial experimental design was used, through the Statistica 10.0 software (StatSoft TM, Inc., Tulsa, OK, USA), with three replicates at the central point, testing three levels of the variables, according to

Table 1. Variables and levels used in the Box–Behnken experimental design for probe-type ultrasound-assisted extraction of acerola by-product.

Variables	Levels
	−1	0	1
Power (W)—X1	350	500	650
Ethanol concentration (%, v/v)—X2	20	50	80
Time (min)–X3	20	40	60

, totaling 15 experiments. Ethanol concentration and the time of extraction were selected based on previous works [19,20,21], while the power range was defined in preliminary tests and considering the heating of the samples (>40 °C) at power levels above 650 W.

After each extraction, the samples were centrifuged (MTD III PLUS; Metroterm, São Paulo, Brazil) at 3000 rpm for 10 min. This extract (5 mL) was purified with 0.5 mL of barium hydroxide (0.1 M) and 0.5 mL of zinc sulfate (5%), and after a rest of 30 min at 25 °C, the sample was centrifuged again, and stored at −18 °C, until analysis. The response variables of the experimental design were the total phenolic compound (TPC) content and antioxidant activity, determined by the free radical scavenging method—DPPH (2,2-diphenyl-1-picrylhydrazyl)—and ferric reducing antioxidant power (FRAP), as described in Section 2.3 and Section 2.4.

2.3. Total Phenolic Compounds

The TPC content was measured by the Folin–Ciocalteau method, by mixing the samples with 10% Folin–Ciocalteau reagent and 7.5% sodium carbonate. The absorbance was measured at 760 nm in a spectrophotometer (UV-1900, Shimadzu^®, Kyoto, Japan). The TPC content was estimated using an analytical curve, with concentrations between 0.1 and 0.5 mM of gallic acid, and a regression coefficient (R²) of 0.99. The results were expressed in gallic acid equivalent (GAE) per 100 g of raw material [27].

2.4. Antioxidant Activity

The DPPH method, an ethanolic solution of DPPH, with an absorbance of 0.8 ± 0.02 nm, was mixed with aliquots of the sample, and after 30 min of reaction, the absorbance was read at 517 nm [28]. In the FRAP method, the samples were added to a reagent composed of 0.3 mM acetate buffer pH 3.6, 10 mM TPTZ (2,4,6-tri(2-pyridyl)-1,3,5-triazine), dissolved in 40 mM HCl, and 20 mM ferric chloride. After incubation at 37 °C for 30 min, the absorbance was measured at 595 nm [29]. For these analyses, analytical curves (R² ≥ 0.99) were prepared with Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) solutions, with concentrations between 0.10 and 0.80 mM, and the results were expressed in Trolox equivalent (TE) per 100 g of raw material.

2.5. Characterization of the Extract Obtained Under Optimized Conditions

The extract obtained under optimized conditions by pr-UAE (650 W, 50% (v v⁻¹) ethanolic solution, for 20 min) was characterized by ultra-high-performance liquid chromatography coupled to mass spectrometry (UHPLC-MS/MS), and for antibacterial activity, as described below.

2.5.1. Composition Determined by UHPLC-MS/MS

For the UHPLC-MS/MS (8050 MS, Nexera X2 HPLC, Shimadzu^®, Kyoto, Japan) analysis, we used a C18 column (5 µm, 150 × 4.6 mm, Shimadzu^®, Kyoto, Japan) that was maintained at 40 °C in a column oven. At a flow rate of 0.5 mL min-1, a gradient was used with different percentages of methanol (MS grade, Merck^®, Darmstadt, Germany) over 1–9 min (20%), 10–15 min (40%), and 16–30 min (10% methanol), using Milli-Q water acidified with 0.1% formic acid (MS grade, Sigma-Aldrich, St. Louis, MO, USA) to complete 100% of the solvent volume. The filtered sample (0.45 μm pore, Filtrilo, PDVF, Colombo, Brazil) was injected (1 µL) and analyzed by an MS detector, using a negative electrospray ionization (EIS) source, which was configured for multiple reaction monitoring (MRM). This setup allowed for a scanning speed of up to 555 MRM events per second. The collision gas was composed of argon (maximum pressure of 20 mPa) and the collision energies were 30 V (negative) and −15 V (positive). The instrument’s gas flow rates were set at 3 L/min for the nebulizing gas and 10 L/min for the drying gas, while the interface voltage was 3 kV, the current was 7 µA, and the temperature was 300° C. Calibration curves (R² ≥ 0.99) were obtained by preparing solutions (10–250 µg L⁻¹) of the analytical standards (Sigma Aldrich): isovanillin, morin, coumarin, catechol, hydroxybenzaldehyde, catechin, rutin, luteolin, quercetin, naringenin, epicatechin, baicalin, chrysin, syringaldehyde, ferulic acid, chlorogenic acid, protocatechuic acid, vanillic acid, salicylic acid, vanillin, gallic acid, syringic acid, sinapic acid, p-hydroxybenzoic acid, naringin, p-coumaric acid, caffeic acid, coniferyl aldehyde, syringaldazine, sinapaldehyde, theobromine, fumaric acid, nicotinic acid, quinic acid, kaempferol, caffeine, malic acid, and resorcylic acid. Quantification was performed in Insight Software version 5.123 (Shimadzu^®, Kyoto, Japan) [30].

2.5.2. Antibacterial Activity Evaluation

The bacterial species evaluated in this study were Escherichia coli ATCC 25922, Listeria monocytogenes ATCC 7644, Shigella sonnei ATCC 25931, Staphylococcus aureus ATCC 25923, Pseudomonas aeruginosa ATCC 27853, and Salmonella enterica subs. enterica Typhi ATCC 19214.

The inoculum was prepared from a 24 h culture. For each microorganism, a standardized suspension was prepared in 0.85% saline solution, adjusted to the 0.5 McFarland standard (1.5 × 10⁸ CFU mL⁻¹) and verified using a spectrophotometer at 625 nm. Subsequently, the suspension was diluted 1:10 in Mueller Hinton Broth, resulting in an inoculum concentration of 1.5 × 10⁷ CFU mL⁻¹.

Analysis of the MIC (minimum inhibitory concentration) was performed using the broth microdilution method [31], adapted for natural products. The extract was lyophilized, for ethanol remotion, and diluted (0.10 to 100.00 mg mL⁻¹) in sterile distilled water. Sodium nitrite (prepared in sterile distilled water) was prepared and used as a positive control. For testing, a total volume of 100 μL was used, consisting of the culture medium and the extract solution. After serial two-fold dilutions, the inoculum (50 μL) was added to each well and incubated at 35 °C for 24 h. The results were evaluated by adding 20 μL of 2,3,5-triphenyltetrazolium chloride (Reatec^®, Neuhausen am Rheinfall, Switzerland) (1.0%) to each well, and incubation at 35 °C for 20 min. The MIC was established as the lowest concentration of the extract that inhibited visible microbial growth. To determine the minimum bactericidal concentration (MBC), a 10 μL aliquot from each well was subcultured onto Mueller Hinton agar plates and incubated at 35 °C for 24 h.

2.6. Statistical Analysis

All analyses were performed in triplicate. The results obtained in the Box–Behnken experimental design were evaluated using the Statistica^® 10.0 software (StatSoft TM, Inc., Tulsa, OK, USA), for analysis of variance (ANOVA), assessment of the significance (p = 0.05), and determination of R². A desirability profile was used to define the condition of maximum TPC and antioxidant activity extraction. Student’s t-test and ANOVA, followed by Tukey’s test, were also performed to compare the results obtained between two or more samples, respectively, both considering p = 0.05.

3. Results and Discussion

3.1. Total Phenolic Content and Antioxidant Activity

Table 2. Experimental conditions used on probe-type ultrasound-assisted extraction of acerola by-product and the results obtained for total phenolic compounds and antioxidant activity.

Experiment	Variables			Total Phenolic Content (g GAE 100 g−1 d.b.) *	Antioxidant Activity
	Power (W)	Ethanol Concentration (%, v/v)	Time (min)		DPPH (mM TE 100 g−1 d.b.) *	FRAP (mM TE 100 g−1 d.b.) *
1	−1 (350)	−1 (20)	0 (40)	1.74 ± 0.02 c,d	1.68 ± 0.03 a,b	6.59 ± 0.09 d
2	1 (650)	−1 (20)	0 (40)	1.60 ± 0.05 a,b,c	1.49 ± 0.03 a	5.70 ± 0.05 b,c
3	−1 (350)	1 (80)	0 (40)	1.62 ± 0.02 a,b,c	1.97 ± 0.01 c	5.20 ± 0.19 a,b
4	1 (650)	1 (80)	0 (40)	1.65 ± 0.01 b,c,d	2.00 ± 0.05 c	5.15 ± 0.05 a
5	−1 (350)	0 (50)	−1 (20)	2.53 ± 0.01 e	3.12 ± 0.08 e	8.72 ± 0.09 e
6	1 (650)	0 (50)	−1 (20)	3.36 ± 0.04 h	4.97 ± 0.24 h	11.35 ± 0.21 g
7	−1 (350)	0 (50)	1 (60)	3.08 ± 0.16 f	3.91 ± 0.05 f	9.76 ± 0.02 f
8	1 (650)	0 (50)	1 (60)	3.14 ± 0.10 f,g	4.40 ± 0.13 g	10.70 ± 0.03 g
9	0 (500)	−1 (20)	−1 (20)	1.63 ± 0.03 b,c,d	1.75 ± 0.01 a,b,c	6.05 ± 0.13 c,d
10	0 (500)	1 (80)	−1 (20)	1.44 ± 0.04 a	1.94 ± 0.01 b,c	5.32 ± 0.13 a,b
11	0 (500)	−1 (20)	1 (60)	1.50 ± 0.03 a,b	1.52 ± 0.12 a	5.42 ± 0.14 a,b
12	0 (500)	1 (80)	1 (60)	1.79 ± 0.02 d	2.60 ± 0.11 d	6.03 ± 0.08 c
13	0 (500)	0 (50)	0 (40)	3.28 ± 0.05 g,h	3.89 ± 0.02 f	11.06 ± 0.18 g
14	0 (500)	0 (50)	0 (40)	3.13 ± 0.07 f,g	3.93 ± 0.10 f	10.81 ± 0.52 g
15	0 (500)	0 (50)	0 (40)	3.17 ± 0.02 f,g,h	3.91± 0.05 f	10.80 ± 0.05 g

* Mean values with different letters within the same column are significantly different according to the Tukey test at p ≤ 0.05. GAE—gallic acid equivalent, TE—Trolox equivalent, d.b.—dry basis.

describes the experimental design matrix and the values of independent and response variables. TPC ranged from 1.44 to 3.36 g GAE 100 g⁻¹. Antioxidant activity determined by the DPPH method ranged from 1.49 to 4.97 mM TE 100 g⁻¹, whereas, when using the FRAP method, values ranged from 5.15 to 11.35 mM TE 100 g⁻¹. Center points (runs 13 to 15) showed little variation, indicating good repeatability. The highest values of TPC and antioxidant activity determined by the DPPH and FRAP methods were recorded in run 6, which was performed using an ultrasonic power of 650 W, an ethanol concentration of 50% (v v⁻¹), and an extraction time of 20 min. The TPC value of run 6 did not differ from that of runs 13 to 15 (center points), and the FRAP value of run 6 did not differ from those of runs 8 and 13 to 15.

Notably, the TPC values observed in this study were higher than those reported by Rezende et al. [6], who detected 1.07 g GAE 100 g⁻¹ in an acerola by-product extract obtained by pl-UAE under optimized conditions (46.5% v/v ethanol, solvent volume-to-sample weight ratio of 8.7 mL g⁻¹, and extraction time of 49.3 min). Santos et al. [22] and Stafussa et al. [32] also reported lower TPC levels when using orbital agitation extraction, reaching, respectively, 0.90 g GAE 100 g⁻¹ in acerola by-product (110 min of extraction, 20 °C, and 34% ethanol) and 0.59 g GAE 100 g⁻¹ in acerola pulp (1:10 sample weight/solvent volume ratio, 40% v/v ethanol, 130 rpm, 120 min, and 25 °C). It is also noteworthy that the DPPH values recorded in runs 5 to 8 and 12 to 15 were higher than those obtained by Gualberto et al. [33], which ranged from 2.21 to 2.31 mM TE 100 g⁻¹ in ethanolic, methanolic, and acetonic extracts of acerola by-products extracted by ultrasound-assisted or shaker extraction. Besides the extraction conditions, the values of TPC and antioxidant activity can be affected by the variety, cultivation conditions, and maturation stage [11,16]. The increase in antioxidant activity found in the present work is interesting, since free radical scavenging can help to prevent chronic diseases [34].

3.2. Effect of Extraction Variables on TPC and Antioxidant Activity

Table 3. Effects, significance, and coefficients of the variables used in the Box–Behnken experimental design on probe-type ultrasound-assisted extraction of acerola by-product.

Total Phenolic Compounds
Variables	Effect a	p b	Coefficient c
X1-Power (linear)	0.195	0.065	0.098
X1-Power (quadratic)	0.051	0.313	0.026
X2-Ethanol concentration (linear)	0.013	0.827	0.006
X2-Ethanol concentration (quadratic)	1.492	<0.010	−1.480
X3-Time (linear)	0.135	0.124	0.067
X3-Time (quadratic)	0.111	0.102	0.056
X1 (linear) ∗ X2 (linear)	0.089	0.354	0.044
X1 (linear) ∗ X3 (linear)	−0.387	0.035	−0.194
X2 (linear) ∗ X3 (linear)	0.238	0.085	0.119
Antioxidant activity determined by DPPH method
X1-Power (linear)	0.545	<0.010	0.272
X1-Power (quadratic)	−0.013	0.314	−0.006
X2-Ethanol concentration (linear)	0.515	<0.010	0.258
X2-Ethanol concentration (quadratic)	2.138	<0.010	−2.139
X3-Time (linear)	0.164	<0.010	0.082
X3-Time (quadratic)	−0.176	<0.010	0.175
X1 (linear) ∗ X2 (linear)	0.113	0.025	0.056
X1 (linear) ∗ X3 (linear)	−0.684	<0.010	−0.342
X2 (linear) ∗ X3 (linear)	0.443	<0.010	0.222
Antioxidant activity determined by FRAP method
X1-Power (linear)	0.657	0.037	0.328
X1-Power (quadratic)	0.384	0.057	0.192
X2-Ethanol concentration (linear)	−0.514	0.058	−0.257
X2-Ethanol concentration (quadratic)	4.812	<0.010	−4.761
X3-Time (linear)	0.116	0.466	0.058
X3-Time (quadratic)	0.338	0.072	0.169
X1 (linear) ∗ X2 (linear)	0.422	0.148	0.211
X1 (linear) ∗ X3 (linear)	−0.847	0.044	−0.424
X2 (linear) ∗ X3 (linear)	0.670	0.068	0.335

^a Effect of the independent variable on the dependent variable, ^b statistical significance, ^c coefficient of the quadratic model of the variables that presented a significant effect. Red letters indicate significant effects (p < 0.05).

presents the effects of extraction variables, their significance, and the coefficients of the mathematical model describing TPC and antioxidant activity by the DPPH and FRAP assays.

Ultrasonic power had a positive linear effect on antioxidant activity, indicating that higher power values result in enhanced extraction of compounds responsible for antioxidant activity, as measured by both methods (DPPH and FRAP). This finding is clearly illustrated by the results of runs 5 and 6: when the ultrasonic power was increased from 350 to 650 W and the other variables remained unchanged, antioxidant activity increased by 59.29% and 30.16% in the DPPH and FRAP assays, respectively. Similarly, More and Arya [35] observed a 36.13% increase in the antioxidant activity of pomegranate peel extracts when raising the ultrasonic power from 70 to 210 W. Sharmila et al. [36] observed a 27.43% increase in antioxidant activity in Cassia auriculata leaf extract when increasing the ultrasonic power from 30 to 50 W. Applying a higher ultrasonic power enhances solvent diffusion into plant cells by promoting the formation of cavitation bubbles and providing a greater input of energy into the system, thereby facilitating the extraction of antioxidant compounds [15].

According to Silva et al. [37], the mechanical effects of ultrasonic cavitation promote the rupture of cell walls, enhancing the release and dissolution of bioactive compounds into the medium, thereby increasing extraction efficiency. However, with prolonged treatment, the intense mechanical forces may lead to compound degradation, particularly of those responsible for antioxidant activity. This phenomenon explains the negative interaction between ultrasonic power and extraction time and its impact on TPC levels and antioxidant capacity. Therefore, it is likely that prolonged extraction at high ultrasound intensities contributed to the degradation of bioactive compounds [15,16]. Most of the compounds present in ruptured cells are released within the first minutes of extraction, owing to the thermal and mechanical effects of ultrasonication [37,38]. With the passage of time and the increase in system energy levels, the number of cavitation bubbles also increases, causing collisions, deformations, and non-spherical collapses. As such, the effect of their implosion is diminished, causing the degradation of antioxidant compounds and compromising their stability and activity [38].

The ethanol concentration exhibited a positive quadratic effect on all evaluated parameters, as well as a positive linear effect on antioxidant activity measured by the DPPH method. This finding suggests that the ethanol concentration plays a crucial role in the extraction of bioactive compounds, likely due to the polarity of the compounds present in the extract. Therefore, evaluating this parameter is essential during UAE. Santos et al. [22] extracted acerola by-products with ethanol concentrations close to 100% and observed a 12.66% reduction in TPC and a 21.09% reduction in antioxidant activity compared with the use of 13% ethanol. In a similar vein, Rodríguez-Martínez et al. [39] suggested that using a single solvent may be insufficient for effectively extracting TPC from avocado peel. The authors found that using a mixture of 40% ethanol and 60% water enhanced the solubility of these compounds, whereas increasing the ethanol concentration from 40% to 80% led to a reduction of 56%, 64%, and 51% in TPC, DPPH, and FRAP values, respectively.

Extraction time showed a positive linear and a negative quadratic effect on antioxidant activity measured by the DPPH method, suggesting that long extraction times may negatively affect this property. Similarly, Iftikhar et al. [40] reported a 20.37% decrease in polyphenol recovery from rye bran after increasing the pl-UAE time from 20 to 40 min. Airouyuwa et al. [41] found that the TPC and antioxidant activity of date seed extracts decreased by 5.61% and 17.48%, respectively, when the pl-UAE time increased from 10 to 30 min. According to these authors, a prolonged period of sonication resulted in the decomposition of phenolic constituents [40,41], corroborating the data found in this work.

The effects of ultrasonic power × ethanol concentration and ethanol concentration × time were significant and positively impacted antioxidant activity as determined by the DPPH method. For instance, in runs 1 and 4, the increase in power and ethanol concentration from the lowest to the highest level (−1 to 1) led to a 19.05% increase in antioxidant activity. In runs 9 and 12, where the ethanol concentration and time increased from levels −1 to 1, the DPPH value increased by 48.57%.

3.3. Development of the Mathematical Model

Statistical analysis of experimental data revealed that TPC and antioxidant capacity correlated significantly with some experimental variables, as shown in Equations (1)–(3), respectively.

$$\mathrm{T}\mathrm{P}\mathrm{C} \left(\mathrm{g} \mathrm{G}\mathrm{A}\mathrm{E} 100 {\mathrm{g}}^{-1}\right)=3.10050-1.48009X_2^2-0.19356X_1{X}_3$$

(1)

$$\begin{array}{c}\mathrm{D}\mathrm{P}\mathrm{P}\mathrm{H} \left(\mathrm{m}\mathrm{M} \mathrm{T}\mathrm{E} 100 {\mathrm{g}}^{-1}\right)\hfill \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}=3.91872+0.27256X_1+0.25760X_2-2.13901X_2^2+0.08223X_3\hfill \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+0.17545X_3^2+0.05648X_1{X}_2-0.34213X_1{X}_3+0.22155X_2{X}_3\hfill \end{array}$$

(2)

$$\mathrm{F}\mathrm{R}\mathrm{A}\mathrm{P} \left(\mathrm{m}\mathrm{M} \mathrm{T}\mathrm{E} 100 {\mathrm{g}}^{-1}\right)=10.44143+0.32836X_1-4.76064X_2^2-0.42351X_1{X}_3$$

(3)

where X₁ is the ultrasonic power (W), X₂ is the ethanol concentration (% v/v), and X₃ is the extraction time (min).

According to ANOVA, the F-value and F-critical were 135.40 and 3.88, respectively, for TPC; 12.82 and 4.14, respectively, for the DPPH value; and 61.80 and 3.59, respectively, for the FRAP value. Thus, the prediction model was deemed valid, as the F-value was greater than the F-critical, demonstrating a satisfactory fit to experimental data.

The coefficient of determination (R²) was 0.96 for TPC, 0.96 for the DPPH value, and 0.94 for the FRAP value. The F-test showed that the model satisfactorily represented the experimental data under the evaluated conditions, enabling the prediction of maximum values.

For the desirability profile X₁ = 1 (power = 650 W), X₂ = 0 (ethanol concentration = 50% v/v), and X₃ = −1 (time = 20 min), Equations (1)–(3) return the following values: TPC of 3.29 g GAE 100 g⁻¹, DPPH value of 4.63 mM TE 100 g⁻¹, and FRAP value of 11.19 mM TE 100 g⁻¹, respectively. These results were similar to experimental values (run 6): 3.36 ± 0.04 g GAE 100 g⁻¹ for TPC, 4.97 ± 0.24 mM TE 100 g⁻¹ for DPPH value, and 11.35 ± 0.21 mM TE 100 g⁻¹ for FRAP value. For all variables, no significant differences (p < 0.05) were observed between experimental and predicted values, evidence of the validity of the model. Similarly to these findings Samaram et al. [42] obtained papaya seed oil extracts with higher antioxidant activity when using a high ultrasonic power (700 W). Guandalini et al. [43], in investigating the phenolic contents of mango peel extracts obtained via ultrasonication, achieved higher yields when using a 50:50 ethanol/water solution, as also reported by Montero-Calderón et al. [44], who extracted bioactive compounds from orange (Citrus sinensis) peel. Singh et al. [45] found that the optimal extraction time was 25 min for the UAE of bioactive compounds from pomegranate peel.

3.4. Extract Composition by UHPLC-MS/MS

The acerola by-product extract obtained under optimal conditions (650 W, 50% v/v ethanol, and 20 min) was found to contain 29 compounds, of which 4 were organic acids, 11 were phenolic acids, 8 were flavonoids, 5 were phenolic aldehydes, and 1 was an alkaloid (

Table 4. Compounds detected in the extract obtained in the optimized conditions of probe-type ultrasound-assisted extraction of acerola by-product.

Class	Compound	Content (µg g−1 d.b.) *
Organic acid	Malic acid	825.71 ± 12.82
	Fumaric acid	5.29 ± 0.72
	Nicotinic acid	0.95 ± 0.00
	Quinic acid	0.33 ± 0.02
∑ Organic acids		832.28 ± 13.56
Phenolic acid	Protocatechuic acid	28.34 ± 0.45
	Resorcilic acid	25.94 ± 1.33
	p-hydroxybenzoic acid	9.98 ± 0.13
	p-coumaric acid	7.12 ± 0.11
	Vanillic acid	1.57 ± 0.06
	Salicylic acid	0.83 ± 0.17
	Caffeic acid	0.73 ± 0.00
	Syringic acid	0.64 ± 0.04
	Gallic acid	0.45 ± 0.01
	Ferulic acid	0.15 ± 0.10
	Chlorogenic acid	0.11 ± 0.02
∑ Phenolic acids		75.86 ± 2.42
Flavonoid	Rutin	11.17 ± 0.37
	Morin	2.04 ± 0.04
	Catechin	1.12 ± 0.08
	Epicatechin	1.10 ± 0.08
	Kaempferol	0.91 ± 0.05
	Quercetin	0.61 ± 0.17
	Naringenin	0.53 ± 0.03
	Crisin	0.01 ± 0.00
∑Flavonoids		17.49 ± 0.82
Phenolic aldehyde	Coniferyl aldehyde	4.43 ± 1.30
	Hydroxybenzaldehyde	0.99 ± 0.02
	Isovanillin	0.34 ± 0.12
	Sinapaldehyde	0.32 ± 0.03
	Syringaldehyde	0.17 ± 0.04
∑ Phenolic aldehydes		6.25 ± 1.51
Alkaloid	Caffeine	0.13 ± 0.02

∑—summatory, d.b.—dry basis, * extraction conduced at 650 W, 50% (v/v) ethanol, and 20 min.

).

The major compound was malic acid (825.71 µg g⁻¹). This organic acid has antioxidant, anti-inflammatory, anti-aging, antidepressant, antihypertensive, and hypoglycemic properties [46,47,48,49]. It is listed as a food additive by regulatory agencies, owing to its technological functions, such as acidulant, acidity regulator, and sequestrant. It is applied in food formulations mainly to adjust the pH, intensify flavor, and/or increase product stability, contributing to the sensory quality and preservation of food (INS 296) [50].

Regarding phenolic acids, protocatechuic and resorcylic acids were identified at high levels. These compounds exhibit antioxidant, anti-inflammatory, anticarcinogenic, and antimicrobial properties [51,52,53,54]. The extract also had a high content of the flavonoid rutin, which is widely recognized for its multiple pharmacological activities, including antioxidant, anticarcinogenic, neuroprotective, cardioprotective, antidepressant, antihypertensive, antimicrobial, and antifungal actions [55,56,57,58].

A lower proportion of flavonoids in relation to phenolic acids was also observed by Santos et al. [22] in acerola by-product extract. The authors found that flavonoids represented 22.29% of the TPC, with rutin as one of the main representatives of the class. In the current study, the proportion was similar, with flavonoids accounting for 18.74% of the TPC.

Marques et al. [59] reported the presence of several phenolic compounds in acerola pomace extract, such as gallic acid, syringic acid, p-coumaric acid, catechin, epicatechin, and quercetin. Complementarily, Gualberto et al. [33] identified the flavonoids kaempferol, naringenin, and rutin, as well as traces of chrysin. Syringic acid was also detected by Marques et al. [60], whereas Poletto et al. [4] identified caffeic acid, salicylic acid, and ferulic acid in acerola by-product. Here, fumaric acid, resorcylic acid, morin, coniferyl aldehyde, hydroxybenzaldehyde, isovanillin, sinapaldehyde, syringaldehyde, and caffeine were identified for the first time in acerola by-product extract.

3.5. Antibacterial Activity

The MIC and MBC values of the acerola by-product extract were determined against various bacteria and compared to those of sodium nitrite, a commercial preservative. The results are presented in

Table 5. Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of acerola by-product extract obtained in the optimized conditions and positive control (sodium nitrite).

Bacteria	Parameter	Extract ** (mg mL−1)	Sodium Nitrite (mg mL−1)
E. coli	MIC *	11.32 ± 0.01 a	12.50 ± 0.00 b
	MBC *	22.65 ± 0.02 a	25.00 ± 0.01 b
L. monocytogenes	MIC *	2.89 ± 0.00 a	25.00 ± 0.01 b
	MBC *	11.58 ± 0.02 a	>100.00 ± 0.00 b
S. sonnei	MIC *	2.89 ± 0.00 a	12.50 ± 0.00 b
	MBC *	2.89 ± 0.00 a	>100.00 ± 0.00 b
S. aureus	MIC *	5.79 ± 0.00 a	25.00 ± 0.02 b
	MBC *	5.79 ± 0.01 a	100.00 ± 0.03 b
P. aeruginosa	MIC *	5.79 ± 0.00 a	12.50 ± 0.00 b
	MBC *	23.15 ± 0.01 a	100.00 ± 0.01 b
S. enterica subs. enterica Typhi	MIC *	2.89 ± 0.00 a	100.00 ± 0.02 b
	MBC *	17.37± 0.18 a	100.00 ± 0.01 b

* Mean values with different letters within the same column are significantly different according to the t-Student test at p ≤ 0.05. ** Extraction conduced at 650 W, 50% (v/v) ethanol, and 20 min.

. The extract exhibited MIC values of 2.89 to 11.32 mg mL⁻¹ and MBC of 2.89 to 23.15 mg mL⁻¹, showing greater antibacterial activity than sodium nitrite. The importance of assessing the extract’s activity against these selected strains is highlighted by their profound influence on global food safety. As major contributors to foodborne illness and spoilage, these species annually account for a substantial burden of disease and mortality, a public health issue well-documented in recent studies [61]. The standard exhibited MIC values of 12.50 to 100 mg mL⁻¹ and MBC values of 12.50 to >100 mg mL⁻¹. These results demonstrate the antimicrobial efficacy of the acerola by-product extract, particularly against S. sonnei and L. monocytogenes. Similar results were reported by Comichio et al. [62], who observed an MIC of 15.6 mg mL⁻¹ against E. coli for freeze-dried acerola concentrate. Stafussa et al. [63] reported an MIC of 6.25 mg mL⁻¹ against S. aureus for acerola pulp extract. Additionally, Paz et al. [64] found that acerola pulp extract exhibited an MBC of 12.5 mg mL⁻¹ against L. monocytogenes.

The results for antibacterial activity can be correlated with extract composition. González-Fandos and Herrera [65] reported that the application of 2% malic acid on poultry legs was effective in reducing Listeria monocytogenes populations, resulting in a reduction of 1.66 log CFU g⁻¹. Ajiboye et al. [66] demonstrated that protocatechuic acid exerts bacteriostatic and bactericidal action against strains of E. coli (MIC 0.55 mg mL⁻¹), P. aeruginosa (MIC 0.30 mg mL⁻¹), and S. aureus (MIC 0.45 mg mL⁻¹); such effects were attributed to the oxidative stress induced in bacterial cells. Alvarado-Martinez et al. [67] observed an inhibitory effect of protocatechuic acid on S. enterica Typhi (MIC 2.00 mg mL⁻¹). Pimentel et al. [68] reported the inhibitory effects of rutin on S. sonnei, and Orhan et al. [69] observed strong antibacterial action against S. aureus (MIC 0.45 mg mL⁻¹) and P. aeruginosa (MIC 16 mg mL⁻¹).

Thus, the observed antimicrobial activity can be attributed to the presence of various bioactive compounds with antimicrobial properties in the extract. Such compounds belong to different chemical classes, and the combined presence of phenolic compounds, flavonoids, and other natural bioactive agents may potentiate antibacterial effects, possibly through synergistic and complementary modes of action [70]. The most common mechanisms of action include alterations in the bacterial cell membrane, such as increased permeability, rupture, and changes in membrane potential. Other mechanisms include intracellular action, such as inhibition of essential enzymes, changes in pH, and alterations in DNA and protein synthesis, among others [71,72].

These results are particularly relevant given that the microorganisms evaluated are among the primary agents responsible for foodborne disease outbreaks. This association underscores the importance of investigating natural extracts with antimicrobial potential, as the presence of these pathogens in food poses a public health risk [73,74,75].

4. Conclusions

The results showed that ultrasonic power, ethanol concentration, and extraction time influence the antioxidant properties of the resulting acerola by-product extracts. The highest yields were achieved under the following conditions: ultrasonic power of 650 W, ethanol concentration of 50% (v/v), and extraction time of 20 min, resulting in extracts with a TPC of 3.36 g GAE 100 g⁻¹, DPPH value of 4.97 mM TE 100 g⁻¹, and FRAP value of 11.35 mM TE 100 g⁻¹. The mathematical model exhibited a good fit to the experimental data, enabling the prediction of the maximum TPC and antioxidant capacity. The extract obtained under optimized conditions exhibited malic acid, protocatechuic acid, resorcylic acid, and rutin as the major components. The extract also demonstrated bactericidal and bacteriostatic effects against various microorganisms, with MIC values ranging from 2.89 to 11.32 µg mL⁻¹ and MBC values ranging from 2.89 to 23.15 µg mL⁻¹. These results demonstrate the potential of acerola by-product extract as an alternative to synthetic additives in the food industry, which can contribute to the sustainability of the sector, support the circular economy, and add value to the acerola production chain. Further works must investigate the impacts of the extract addition on the sensory properties of food products and the stability of the antioxidants and antibacterial activity during their shelf lives.