Annona muricata Leaves as a Source of Bioactive Compounds: Extraction and Quantiﬁcation Using Ultrasound

: In this work, the ultrasound-assisted extraction (UAE) (operating conditions of sonication amplitude, pulse-cycle, and extraction time) was investigated to obtain an extract rich in biologically active compounds from Annona muricata leaves. In addition, the content of bioactive compounds from the extract by the optimal UAE conditions (UAE extract) was compared with extracts prepared by decoction and infusion. Moreover, Artemia salina toxicity was evaluated in all extracts. UAE extract (using optimal conditions: 80% amplitude, 0.7 s pulse-cycle, for 4.54 min) contained 178.48 mg/100 mL of soluble phenols, 20.18 mg/100 mL of total ﬂavonoids, 27.81 mg/100 mL of hydrolyzable polyphenols, 167.07 mg/100 mL of condensed tannins, 30.44 mg/100 mL of total alkaloids and 14.62 mg/100 mL of total acetogenins. The UAE extract exhibited a high antioxidant capacity and a higher content of bioactive compounds from ~6 to ~927-fold than decoction or infusion, depending on the type of compound. Twenty-four different phenolic compounds were identiﬁed in the samples, and UAE extract produced the highest concentration. All extracts were considered non-toxic using the A. salina test. The UAE extract from A. muricata leaves is a rich source of bioactive compounds and can be utilized to formulate therapeutic drugs or nutraceutical foods.


Introduction
Annona muricata L. is a native tree of America that has stirred great interest due to its various therapeutic effects. In traditional medicine, bark, fruit, seed, and leaves have been used to treat fever, pain, headache, insomnia, rheumatism, parasitic and bacterial infections, hypertension, inflammation, diabetes, cancer, respiratory, and skin diseases [1]. In native regions, infusions from dry leaves and decoctions from fresh leaves are used as sedatives for chronic diarrhea, dysentery, hypertension, kidney, and biliary ailments. In addition, the consumption of these beverages has increased in the past few years because they have been attributed to anticancer properties [2]. Table 1. Experimental values of total soluble phenols and yield after ultrasonic-assisted (Box-Behnken design) from extracts of Annona muricata leaves also showed predicted values and error rates after the response surface analysis.

Run Predictors
Final Temperature ( • C) X PC = Pulse-cycle. X SA = Sonication Amplitude. X ET = Extraction time. All entries represent the means ± standard deviation of three determinations and three replicates (n = 9). Different letters by column indicate significant statistical differences between treatments (α = 0.05). 1 The values were predicted using secondary polynomial equations, R 2 = 0.886.
TSP were extracted using 0.5 g of dried leaves powder and 20 mL of acetone: water (80:20 v/v). Ultrasound was applied with an ultrasonic generator (Hielscher, UP400S, Teltow, Germany) of high intensity (maximal nominal output power of the device, 400 W) at a frequency of 24 kHz and with an ultrasonic probe H7 Tip 7 (acoustic intensity 300 W/cm 2 and Hielscher, Teltow, Germany) that was immersed 2 cm into the extraction solution. The UAE conditions were applied according to the experimental design. A cold-water bath was employed to maintain the extraction temperature at 25 ± 2 • C. The samples were extracted twice, and the extracts were centrifuged (Hermle Z32HK, Wehingen, Germany) at 9380× g for 10 min at 4 • C. The supernatants were concentrated in a rotary evaporator (Yamato Scientific Co., Ltd., RE301, Tokyo, Japan) at 40 • C to a volume of 8 mL (acetone-free extract). TSP were measured in the concentrated extracts.

Total Soluble Phenols (TSP)
The TSP content was determined using the Montreau method [20] with slight modifications. The concentrated extracts (12 µL) were mixed with 12 µL of Folin-Ciocalteu reagent (Sigma-Aldrich, St. Louis, MI, USA), 116 µL of sodium carbonate solution (75 g/L), and 164 µL of distilled water; the mixture was stirred for 15 min in darkness. Absorbance was measured at 750 nm with a microplate reader (800TS, Biotek, Winooski, VT, USA). The results were calculated as milligrams of gallic acid equivalents per gram of dry extract or 100 mL extract (mg/g DW or mg/100 mL).

Yield of Total Soluble Phenols
The yield was calculated in percentage from the total weight of the sample (g) according to Equation (1) [21]. (1)

Response Surface Methodology Analysis (RSM) to Obtain the Optimal UAE Conditions
The TSP and yield data were analyzed for RSM to achieve the optimal extraction conditions. A second-order polynomial equation was used to calculate the predicted response (Equation (2)), where Y is the predicted response (TSP or Yield), Xi is the value of each factor (X A , X PC, and X ET ), β0 is a constant, and βi are the principal effect coefficients for each variable, and βij are the interaction effect coefficients. Model adequacy was assessed by analysis of variance (ANOVA) using the Statsoft ® statistical software v. 10 (Tulsa, OK, USA) to determine the effects of significant interactions in the model (p < 0.05) and by quantification of the coefficient of determination (R-squared and R-adjusted). The Fisher-LSD test assessed the differences between means (p < 0.05).

Model Reliability and Comparison of the Content of Bioactive Compounds from A. muricata Leaf Extract Using Optimal UAE Conditions with Extracts by Decoction and Infusion
The optimal conditions by UAE were performed experimentally to examine the model accuracy, and the TSP content was measured again. Moreover, the bioactive compound content was compared between UAE, decoction, and infusion extracts from A. muricata leave.
Thermal treatments were made as the consumers usually prepare the infusion and decoction from A. muricata leaves. The infusion was prepared by placing 2.5 g of dried crushed leaves wrapped into filter paper in 250 mL of hot water (98 • C) for 10 min. The decoction was performed with 5 g of fresh leaves placed in boiling water (98 • C) for 10 min [22]. The extracts were cooled down and analyzed. Response variables were: phenolic compounds, antioxidant capacity (AOX), total alkaloids, total acetogenins, and toxicity. Also, the effectiveness of the UAE concerning thermal treatments was calculated.

Phenolic Compounds
TSP were performed as was described in Section 2.4. The total flavonoid content was measured by a spectrophotometric assay [23]. The reaction consisted of 1 mL of extract, 1 mL of methanol, and 1 mL of 2%AlCl 3 solution. The mixture was maintained in the dark with continuous stirring for 10 min. The absorbance was measured at 415 nm in the microplate reader. Results were calculated as milligrams of quercetin equivalents per gram of dry extract or 100 mL extract (mg/g or mg/100 mL extract).
After soluble phenols, hydrolyzable polyphenols and condensed tannins were analyzed only in the dried extracts (200 mg). The samples were hydrolyzed with methanol/H 2 SO 4, 90:10 (v/v) at 85 • C for 20 h. After the samples were centrifuged (10 min, 25 • C, 5000× g) and the precipitates were resuspended twice with 10 mL of methanol and centrifuged. The supernatants were combined, and hydrolyzable polyphenols were measured by the Folin-Ciocalteu reagent [20,24]. To measure condensed tannins, the dried extracts were treated with 10 mL of a butanol/HCl/FeCl 3 (97.5:2.5, v/v) solution at 100 • C for 3 h. Afterward, they were centrifuged (10 min, 4 • C, 6000× g), and the residues were rewashed with the same solution. The supernatants were combined, and the absorbance was read at 555 nm in the microplate reader. Condensed tannins were calculated with a standard curve of proanthocyanidins from Mediterranean carob pods (Ceratonia siliqua L.) [25]. The results were expressed as mg/g or mg/100 mL.

Total Alkaloids
The extracts (38 mL) were added to 1.8 mL HCl (3%), and they were moderately shaken for 4 h at room temperature and kept in refrigeration (4 • C) overnight. The extract was centrifuged (21,105× g, 10 min at 4 • C), the supernatants were placed on ice, and the pH was adjusted to 10 with NaOH (15%). Then, CH 2 Cl 2 in a 1:2 ratio (extract:CH 2 Cl 2 , v/v) was added, and the organic layer was recovered. Next, CH 2 Cl 2 was added two times more, and the organic layers were combined. The organic phase was kept at 4 • C and concentrated in a rotary evaporator at room temperature to a volume of 8 mL. The samples (100 µL) were mixed with 1000 µL of Dragendorff reagent. The absorbance was measured at 530 nm in a spectrophotometer (Jenway 6705, Dunmow, UK). The results were expressed as milligrams of quinine equivalents per gram of dry extract or 100 mL of extract (mg/g or mg/100 mL) [26].

Total Acetogenins
Total acetogenins were determined as described by Aguilar-Hernández et al. [27] with slight modifications. The UAE, decoction, and infusion extracts were freeze-dried (Labconco FreeZone 2.5 Plus cascade benchtop 7670020, Kansas City, MO, USA). Freezedried samples (0.2 g) were mixed with 10 mL of chloroform and shaken moderately for 2 h. Then, the mixtures were centrifuged (9380× g, 10 min at 4 • C). The supernatant was saved, and the residues were resuspended with 10 mL chloroform, shaken for 1 h, and centrifuged again. The supernatants were combined and cleaned through a chromatographic column (25 cm long × 2.5 cm diameter) loaded with activated carbon (2.5 g) using chloroform (10 mL). Subsequently, petroleum ether (5 mL) was added to the samples, and the organic phase was discarded. The cleaned samples (250 µL) were mixed with Kedde's reagent (2 mL). The absorbance was measured at 505 nm in the spectrophotometer. The results were expressed in mg/g DW or mg/100 mL extract.

Effectiveness of Ultrasound to Extract Bioactive Compounds Compared with Decoction and Infusion
The UAE effectiveness [26] was calculated with Equation (3): where: BC-Content UAE = Bioactive compound content measured in UAE extract. BC-Content decoction or infusion = Bioactive compound content measured in infusion or decoction extract.

Toxicity with the Brine Shrimp Lethality Test
Toxicity was evaluated from UAE, decoction, and infusion extracts according to the Anaya-Esparza et al. [32] method. The test was performed using the A. salina in the early stage (3 weeks of growth, 2 cm in length). Ten adult organisms for each treatment were placed into a tube containing 10 mL of saline water (30 mg/mL, non-iodized sea salt). Each dried extract was resuspended in 1 mL of saline water. A. salina were exposed to 1500, 1000, 500, and 100 µg/mL for 24 h at 28 • C in darkness. A negative control (1 mL saline water) was used. Each concentration was tested in triplicate in three replicated experiments. The number of survivors was counted after 24 h of exposure, and dead A. salina was considered those that did not exhibit any internal or external movement during several seconds of observation.

Statistical Analysis
In the first stage, RSM was used. The experimental data were analyzed in a unifactorial design in the second stage. Data analyses were performed using analysis of variance (ANOVA) (p < 0.05) with the STATISTICA v.10 software (Stat soft, Tulsa, OK, USA). The LSD-Fisher test was employed (α = 0.05) to examine the significant difference between means. Table 1 shows significant statistical differences (p < 0.05) between treatments. Results showed a dependence on the experimental conditions. During UAE, the temperature was maintained at 25 ± 2 • C. The highest content of TSP and yield were 174.32 mg/100 mL (or 236.85 mg/g of dry extract) and 5.68% with a 0.7 s pulse-cycle and a sonication amplitude of 60% for 4 min. However, the conditions that produced the lowest TSP and yield values were 60% sonication amplitude, 1 s pulse-cycle, and 2 min. According to the suggested classification by Vasco et al. [33], phenolic compound levels can be divided into (i) low (5 mg/g DW), (ii) intermediate (5-25 mg/g DW), and (iii) high, when they are greater than 25 mg/g DW. A. muricata leaves contained an intermediate polyphenol level, with a phenolic content higher than the green tea infusion (66.77 mg/100 mL) [34].

Total Soluble Phenols and Yield from A. muricata Leaves by Ultrasound-Assisted Extraction
In this study, the TSP values from aqueous UAE extract were three times greater than the soluble phenols reported as gallic acid equivalents from aqueous extracts by stirring (68.37 mg/100 mL), maceration (26.2 mg/g of extract), and ultrasonic bath (42 mg/g of extract) from A. muricata leaves, [6,17,35]. It is attributed to the direct contact of the ultrasonic probe with the samples, which generate high-intensity ultrasonic waves. The energy released by the implosion of the microbubbles impacts the plant material dispersed in the solvent. The cell wall is thus broken, and larger pores are produced that facilitate mass transfer from plant material to the solvent, thus increasing the yield [18,36]. However, sometimes UAE causes adverse effects, although it is dependent on operating conditions [31].

Optimal UAE Conditions to Extract Total Soluble Phenols from Annona muricata Leaves
The response surface analysis on the TSP content and yield as a function of the independent variables (X SA , X PC , X TE ) was performed using multiple regression. The ANOVA proved that most of the parameters and their combinations were significant (p < 0.05) on TSP (Table S1). Also, ANOVA confirmed that experimental data had significant correlation coefficients (R 2 = 0.886 an R 2 = 0.873, respectively) and adjusted correlation coefficients (adjusted R 2 = 0.848 and 0.826) with the calculated models (quadratic polynomials). Moreover, b-coefficients of the fitted quadratic polynomial models for TSP and yield were significant (p < 0.05), except X 2 SA . The lack of fit test showed the adequacy of the model (p > 0.05), indicating a good approximation to the real system.
The regression model predicted the TSP content and yield with the quadratic polynomial equations (Table S2) 95% confidence level. The predicted values exhibited a close relationship to the experimental data (see Table 1). It is the first report where a Box-Behnken design and RSM are employed to attain optimal UAE conditions (sonication amplitude, pulse-cycle, and extraction time) of phenolic compounds from A. muricata leaves. However, some authors have evaluated the same independent variables in starfruit leaves [37] and Justicia spicigera leaves [38]. They found high correlation coefficients (R 2 = 0.884 and R 2 = 0.978, respectively), concluding that RSM is an efficient tool to optimize UAE conditions in extracting phenolic compounds from leaves.
The significant interactions of the UAE parameters on TSP content and yield are depicted in Figures 1 and 2. The three-dimensional (3D) response surface plots show elliptical contour shapes attributed to the interactions between the corresponding variables. At 40% sonication amplitude, the highest TSP ( Figure 1A) and yield ( Figure 2A) extraction were obtained with a 0.3 s pulse-cycle and 3.5 min extraction time. For 60% sonication amplitude, the highest TSP content ( Figure 1B) and yield ( Figure 2B) are observed with a 0.7 s pulse cycle with 4 min extraction time, while at 80% amplitude, the highest TSP content ( Figure 1C) and yield ( Figure 2C) are observed with 0.7 s pulse cycle and 4.5 min extraction time. Figures 1D and 2D (Pareto plots) validate the effect of the independent variables and their interaction on TSP and yield at a 95% confidence level. The main effects of linear or quadratic parameters were X 2 PC > X 2 ET > X SA × X 2 PC .  Pareto Chart for standardized effects estimates Total soluble phenols (mg/100 mL)  Pulse-cycle, extraction time, and sonication amplitude were the most critical factors involved in phenolics extraction and yield from starfruit leaves [37], Justicia spicigera leaves [38], and guava leaves [39]. However, 100% sonication amplitude, constant acoustic irradiation (1 s pulse-cycle), and long extraction times (>5 min) caused a high cavitation phenomenon and sonolysis that degraded thermolabile bioactive compounds [10,40].
In this study, the UAE application at 1 s pulse-cycle (constant ultrasonic waves), any extraction time, and sonication amplitude tends to extract low TSP content from A. muricata leaves. The results coincided with the study by Anaya-Esparza et al. [38]. These authors reported that the highest extraction time (12 min), 1 s pulse-cycle, and 100% sonication amplitude caused the lowest TSP content from Justicia spicigera leaves. In contrast, 0.4 s and 0.7 s pulse-cycle are pulsed acoustic irradiations that decrease the ultrasonic energy and the effect of sonolysis [10]. Therefore, the optimal UAE conditions (80% sonication amplitude, 0.7 s pulse-cycle, and 4.54 min) to extract the highest TSP content and yield from A. muricata leaves are shown in Table 2. The predicted optimal response of TSP was 180.52 mg/100 mL within the permitted confidence limit (95%) of 166.47 to 184.55 mg/100 mL of extract. With respect to yield, it was 5.86%, within the permissible confidence limit (95%) of 5.43 to 6.30%.  Pareto Chart for standardized effects estimates Yield (%)   Confidence limits of −95% lower limit, +95% confidence limit upper limit. The confidence interval is the difference between upper and lower limits.

Model Reliability and Comparison of the Content of Bioactive Compounds from A. muricata Leaf Extract Using Optimal UAE Conditions with Extracts by Decoction and Infusion
The optimal UAE conditions were conducted experimentally to verify the model reliability, resulting in an extract with 178.48 mg/100 mL (or 237.92 mg/g of dry extract) ( Table 3). In addition, flavonoids (20.18 mg/100 mL or 27.31 mg/g of dry extract), alkaloids (30.44 mg/100 mL or 51.21 mg/g of dry extract), and total acetogenins (14.62 mg/100 mL or 19.44 mg/g of dry extract) were mostly extracted (p < 0.05) from A. muricata leaves by the optimal UAE conditions compared with decoction and infusion ( Table 3). The hydrolyzable polyphenols have not been reported in A. muricata leaf extracts; however, they were quantified in the UAE extract (27.81 mg/100 mL or 38.80 mg/g of dry extract) as well as condensed tannins (167.07 mg/100 mL or 224.87 mg/g of dry extract). Furthermore, decoction and infusion showed low hydrolyzable polyphenols and condensed tannins content. It is inferred that these phenolic compounds (hydrolyzable and condensed tannins) are poorly extractable with water and conventional methods because they are embedded in the cell wall. However, heat modifies cellular structure, and some compounds are released due to the leaching effect [24,25]. However, ultrasound breaks the cell walls to remove non-extractable polyphenols and causes a UAE extract rich in phenolic compounds (extractable and non-extractable). Table 3. Phenolic compounds, total alkaloids, and total acetogenins obtained by optimal conditions of ultrasound-assisted extraction (UAE), decoction, and infusion from extracts of Annona muricata leaves and UAE effectiveness.  The reported flavonoid content from A. muricata leaves using three days of maceration was 19.14 mg/100 mL [41], although the extraction time was shortened with UAE in this experiment. In contrast, tannins (1.22 mg/g of dry extract) [42], alkaloids (39.37 mg/g of dry extract) [16], and acetogenins (0.318-15.05 mg/g of dry extract) [43,44] from A. muricata leaves are lower than found in this work using maceration, ultrasonic bath, decoction, and infusions; therefore, it can be inferred that the use of a direct ultrasonic probe to the matrix is much more effective to extract these bioactive compounds than conventional methods or ultrasonic bath.
In addition, the UAE was more effective in extracting bioactive compounds from 7-fold to~927-fold, dependent on each type of measured compound (Table 3). These findings agree with Anaya-Esparza et al. [38] when different methods for extraction (ultrasound and thermal decoction) of phenolic compounds from Justicia spicigera leaves were performed. They reported an 11-fold increase in total polyphenols when the sample was treated by UAE (55% amplitude, 0.7 s pulse-cycle, and 3 min) than thermal decoction. On the other hand, Aguilar-Hernández et al. [30] compared UAE (100% amplitude, 0.7 pulse-cycle and 15 min) and maceration for ACGs extraction from A. muricata seeds. They reported that~UAE extracted 10-fold more ACGs than maceration. Lee et al. [16] reported 2.3-fold more alkaloids from A. muricata leaves extract with UAE (340 W, 56 • C, 30 min) than soxhlet extraction (7 h, 80 • C).
The highest extraction of bioactive compounds by UAE adapted with an ultrasonic probe generates high-intensity waves with a high disrupting effect on the plant cell walls [14]. At the same time, thermal treatments (decoction or infusion) did not extract a high content of bioactive compounds. It can be attributed to elevated extraction temperatures, which degrade these compounds [45,46]. In addition, using whole fresh leaves or dried crushed leaves did not allow for good mass transfer compared with powdered leaves used in UAE [13]. Figure 3 shows that AOX was higher in the UAE extract than in decoction and infusion extracts. The order of AOX was as follows FRAP > DPPH > ABTS. Documented studies of A. muricata leaf extracts have exhibited AOX by FRAP > ABTS or DPPH assays using maceration and Soxhlet methods [6,47]; however, it is difficult to compare data because the reported units are different. The highest AOX by FRAP assay is attributed to flavonoids and tannins with chelating capacity. The hydroxyl groups (-OH) in the B ring of flavonoids (catechins, quercetins, and kaempferol derivates) and tannins (proanthocyanins) strongly support the hypothesis that these phenol families are responsible for the highest antioxidant activity through the hydrogen transfer mechanism [48], metal-chelating ability (Fe + ) and radical scavenging [49,50]. However, the reducing or chelating activities depend not only on the total polyphenol content but also on the specific chemical structure and the number and position of hydrogen donating hydroxyl groups on the aromatic cycles of the phenolic molecules [49,51]. On the other hand, the extract by decoction exhibited more TSP than the extract by infusion; thus, AOX was greater in decoction extract. This result was attributed to when preparing the infusion; the leaves were previously dried (40 • C) and then subjected to hot water (80 • C); therefore, the soluble phenolic compounds decreased most likely because they were more thermo-sensitive [51].

Phenolic Profile
In the extract by UAE, twenty-four phenolic compounds were identifi fied (Figure 4), while in the decoction and infusion extracts, twenty-three a were identified. The phenolic compounds found in the highest contents in t were epigallocatechin (14.80 mg/100 mL), gallocatechin (7.59 mg/100 mL), g mg/100 mL), -coumaric acid (6.79 mg/100 mL) and 4-hydroxyphenyl a mg/100 mL), while in the decoction and infusion extracts, the phenolic c tected in the highest proportion were gallocatechin with 7.59 mg/100 mL a mL respectively, followed by the 4-hydroxyphenyl acetic acid, homovanil maric acid and neochlorogenic acid (Table 4). Similarly, some phenolic co been reported from an infusion of A. muricata leaves, such as rutin (1.78 μ genic acid (1.40 μg/mg), catechin (1.11 μg/mg), and gallic acid (0.27 μg/m much lesser concentrations than those found in this work. The effect of method was notable; the UAE extract exhibited higher content of pheno

Phenolic Profile
In the extract by UAE, twenty-four phenolic compounds were identified and quantified (Figure 4), while in the decoction and infusion extracts, twenty-three and twenty-two were identified. The phenolic compounds found in the highest contents in the UAE extract were epigallocatechin (14.80 mg/100 mL), gallocatechin (7.59 mg/100 mL), gallic acid (6.80 mg/100 mL), ρ-coumaric acid (6.79 mg/100 mL) and 4-hydroxyphenyl acetic acid (6.28 mg/100 mL), while in the decoction and infusion extracts, the phenolic compounds detected in the highest proportion were gallocatechin with 7.59 mg/100 mL and 7.01 mg/100 mL respectively, followed by the 4-hydroxyphenyl acetic acid, homovanillic acid, ρ-coumaric acid and neochlorogenic acid (Table 4). Similarly, some phenolic compounds have been reported from an infusion of A. muricata leaves, such as rutin (1.78 µg/mg), chlorogenic acid (1.40 µg/mg), catechin (1.11 µg/mg), and gallic acid (0.27 µg/mg) [52], but in much lesser concentrations than those found in this work. The effect of the extraction method was notable; the UAE extract exhibited higher content of phenolic compounds than infusion and decoction extracts (p < 0.05). It can be explained because the thermal extraction methods could induce the degradation of phenolic acids [51]. However, this depends on the type of polyphenol moieties because there were some exceptions. The content of protocatechuic acid, chlorogenic acid, gallocatechin, and catechin was statistically similar between extracts (p > 0.05). UAE extract resulted rich in phenolic compounds, which coincided with antioxidant capacity.   In this experiment, the obtained UAE extract from A. muricata leaves is considered a source of bioactive compounds with potential health benefits. The phenolic compounds (extractable or non-extractable) are potent antioxidants. Also, it has been demonstrated that these same compounds exhibited antibacterial, anti-inflammatory, antiaging, anticarcinogenic, antidiabetic, hepatoprotective, and immunostimulatory activities [53,54].
Catechins (gallocatechin, epigallocatechin, epicatechin, catechins, and other isomers) have been reported to have antibacterial activity by binding to the cell membrane causing the membrane to burst by damaging the lipid layer and releasing its cytoplasmic contents, with subsequent cell death [55]. Other studies demonstrated that gallic acid and its derivatives have a potential anticancer activity due to the pro-oxidant action of the gallate compounds, whose mechanism of action is associated with oxidative stress, causing mitochondrial dysfunction and an increase in the level of intracellular Ca 2+ , which induces the death of tumor cells, since these cells lack the protection system of catalase [56].
On the other hand, reported isoquinoline alkaloids in A. muricata have exhibited anti-depressant, analgesic, and dopaminergic activities [7]. Acetogenins, correspondingly, are the specific compounds mainly studied in the Annonaceae family by their anticancer activity on different cancer cell lines such as breast, prostate, and liver, among others [2]. The UAE extract, rich in phenolic compounds, alkaloids, and acetogenins, is a potential source for alternative medicine. All values are means ± standard deviation of three determinations and three replicates (n = 9). Different letters in each file indicate significant statistical differences between treatments (α = 0.05). ND = not detected. 1

Toxicity with the Brine Shrimp Lethality Test
The UAE, decoction, and infusion extracts did not manifest any alteration of motility and behavior of A. salina larvae in any of the concentrations; therefore, the 100% A. salina survived. González-Esquinca et al. [57] demonstrated that the toxicity of the ethanolic dried extract (LC 50 = 831 µg/mL) was higher than the aqueous extract obtained by boiling 20 min (LC 50 > 1000 µg/mL) from A. muricata leaves against A. salina nauplii. The differences in the toxicity reports from A. muricata leaves extracts with this experiment can be by the age of larvae, solvents to obtain extracts, and extraction methods that can influence the type and quantity of bioactive compounds extracted.
According to the Clarkson s toxicity index, the plant extracts are considered toxic when LC 50 is <100 µg/mL, intermediate toxic when LC 50 is 100-500 µg/mL, low toxic when LC 50 is 500-1000 µg/mL and non-toxic when LC 50 is >1000 µg/mL) [58]; therefore, the evaluated extracts are considered non-toxic.

Conclusions
An aqueous extract rich in bioactive compounds (phenolic compounds, alkaloids, acetogenins, and antioxidant capacity) was obtained from A. muricata leaves using ultrasoundassisted extraction. In addition, UAE was more effective in extracting bioactive compounds from A. muricata leaves and increasing the antioxidant capacity of the extract than decoction and infusion extracts. UAE extract exhibited the presence of eighteen phenolic acids, five flavonoids, and one ellagic tannin. All extracts were assessed as non-toxic according to the brine shrimp lethality test. The present study highlights that ultrasound applied directly to the raw material with an ultrasonic probe is an efficient technology to obtain an extract from A. muricata leaves as a source of bioactive compounds. Thus, these extracts can be incorporated into capsules, food, and beverages to increase the content of phenolic compounds, antioxidant capacity, and other bioactive compounds that can contribute to consumer health.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/horticulturae8070560/s1, Table S1: Analysis of variance and regression coefficients of quadratic polynomial models with ultrasonic-assisted extraction on total soluble phenols and yield from extracts of Annona muricata leaves. Table S2: The predicted mathematical models for the extraction of soluble polyphenols (mg/100 mL) and yield from extracts of Annona muricata leaves after ultrasound-assisted extraction.