**1. Introduction**

The *Myrtus* genus, belonging to the Myrtaceae family, comprises about 50 species that are native of the Mediterranean basin. Among those, *M. communis* is an aromatic evergreen perennial sub-shrub (high: 1–3 m) with white flowers (blossoming time: June to July) and dark blue ripe berries [1]. It is native to Southern Europe, North Africa and West Asia and widespread in the Mediterranean region. Its fruits are consumed either raw or processed in diverse products such as canned fruits, yogurts, beverages, jams and jellies. In addition, there has been a growing interest in the use of berry extracts as ingredients in functional foods and dietary [2]. Volatile oils, tannins, anthocyanins, fatty acids, sugars, and organic acids such as citric and malic acids are important components of these fruits [3]. In general, myrtle berries are accepted as being rich in phenolic compounds, which in turn are associated to the fruits claimed health e ffects, including the prevention of degenerative diseases, such as cancer and cardiovascular diseases [4]. In fact, phenolic compounds are accepted as potent antioxidants due to their double bonds and hydroxyl groups, being capable of preventing the oxidation of free radicals that may damage physiological molecules cells, such as lipid proteins and DNA [5]. Many studies have shown a positive relationship between the phenolics content and the antioxidant capacity of fruits and

vegetables [4–6]. Moreover, the regular consumption of fruits and vegetables is believed to prevent oxidative stress events and oxidative-stress related diseases [7–9].

Due to countless beneficial characteristics of phenolic compounds in human health, research has been intensified, aiming to find fruits, vegetables, plants, agricultural and agroindustrial residues as sources of these bioactive components. Obtaining such compounds often requires many long and costly steps, such as extraction, isolation and identification [5], and often result in thermal degradation of various bioactive constituents [6,7]. In this context, the development of new extraction methods is one of the major challenges in technological innovation towards the direction of "Green chemistry" [8,9]. Among them, ultrasound-assisted extraction (UAE) and microwave-assisted extraction (MAE) are particularly attractive because of their simplicity, low cost of equipment, e fficiency in extracting analytes from di fferent matrices and the requirement of low energy, reduced quantity of solvent and/or time consumption [10], compared to conventional extraction methods, which have several disadvantages, such as the use of volatile and hazardous solvents, the long extraction time and more recovery energy [11]. The enhancement of the extraction process by ultrasounds is attributed to the disruption of the cell walls, reduction of the particle size and the increased mass transfer of the cell content to the solvent, caused by the collapse of the bubbles produced by acoustic cavitation [8]. The processing parameters optimization and interpretation of experiments compared to others has been previously done through response surface methodology (RSM) [12]. This latter has been shown to be a powerful tool in optimizing experimental conditions (factors) to maximize the response. With the experimental results of a response surface design, a mathematical polynomial model, describing the relation between a response (dependent variable) and the considered factors (independent variable), is built. The mathematical model, usually a second-order polynomial model, can be visualized graphically by drawing 2D contour plots or 3D response surface plots [13]. The model allows determining the optimum value of the independent variables ( *Xi*), as well as those of the dependent ones ( *Y*).

Previous studies focusing on phenolic compounds and/or the antioxidant abilities of myrtle pericarp have been performed with extracts obtained by conventional methods [14–16], while, to our knowledge, there is no available information on the optimization of ultrasonic procedure for the extraction of phenolic compounds from this matrix, using a safer solvent such as ethanol, which is an organic solvent used in the food and pharmaceutical industries [17]. Therefore, the present study aimed at the optimization of UAE process parameters using RSM, including ethanol concentration, extraction time, irradiation amplitude and liquid-to-solid ratio, to maximize the content of the extracted phenolics. Levels of phenolic compounds and the antioxidant activity of pericarp *M. communis* extract obtained under the optimum setting parameters (UAE-OPT extract) were compared with those of extracts obtained by microwave-assisted extraction (MAE) and conventional solvent extraction (CSE) methods, using previously established conditions [18]. Then, the individual phenolic compounds present in the optimized extract obtained by UAE were identified by UHPLC-DAD-ESI-MSn.

#### **2. Materials and Methods**

## *2.1. Plant Material*

The fruits of *M. communis* were harvested from spontaneous plants in Adakar, Bejaia, located in the northeast of Algeria. The collected samples were identified by the Vegetable Ecological Laboratory of the Algiers University, Algeria and a voucher specimen was deposited at the Herbarium of Natural History Museum of Aix-en-Provence, France, under the voucher number D-PH-2013-37-12. Berries were washed and then dried in a static oven at 40 ◦C for one week. Pericarps were separated manually from seeds and further grounded in an electrical grinder (A11Basic, IKA, Retsch, Germany), which was then sieved to obtain a fine powder (<250 μm).

#### *2.2. Extraction of Phenolic Compounds*

### 2.2.1. Ultrasound Extraction

UAE was performed in an ultrasonic apparatus (Vibra cell, VCX 75115 PB, SERIAL No. 2012010971 MODEL CV 334, SONICS, Newtown, Connecticut, USA) with a working frequency fixed at 20 kHz. For extraction, 1 g of the pericarp powder was placed in a 250 mL amber glass bottle containing ethanol. The suspension was exposed to acoustic waves under distinct setting parameters (solvent concentration, irradiation time, ultrasound amplitude and solvent-to-solid ratio). The temperature was maintained constant by circulating external cold water and checking the temperature using a T-type thermocouple [5]. Indeed, ultrasound is considered a non-thermal technology, since it increases only the local temperature without affecting the surrounding environment [19]. After extraction, the solution was filtered through a sintered glass filter of porosity 2.

To determine the effect of ethanol concentration, irradiation time, ultrasound amplitude and solvent-to-solid ratio on the extraction yield of phenolic compounds from myrtle pericarp, RSM was applied with a Box–Behnken Design (BBD) [5]. This design resulted in the testing of four factors in a single block of 30 sets of test conditions (Table 1). The constant values for irradiation time, liquid-to-solid ratio and ethanol concentration in the UAE trials were 10 min, 50 mL/g and 50% (*v*/*v*), respectively.

$$Y = \beta\_0 + \sum\_{i=0}^{k} \beta\_i X\_i + \sum\_{i=1}^{k} \beta\_{ii} X^2 + \sum\_{i>1}^{k} \beta\_{ij} X\_i X\_j + E \tag{1}$$

where *Xi*, *Xj*, ... , *Xk* are the independent variables affecting the responses *Y* (the yield of total phenolic compounds); β*<sup>i</sup>*, β*ii* and β*ij* are the regression coefficients for linear, quadratic and interaction terms, respectively; and k is the number of variables.

The factor levels were coded as −1 (low), 0 (central point or middle) and 1 (high), respectively, according to Equation (2):

$$X\_i = (X\_i - X\_0) / \Delta X \dots \tag{2}$$

where *Xi* is the coded value of the variable *Xi*; *X0* is the value of *X* at the center point; and Δ*X* is the step change.

## 2.2.2. Microwave-Assisted Extraction

Phenolic extracts were obtained using a domestic microwave oven (Samsung MW813ST, Kuala Lumpur, Malaysia) adapted by adding of a condenser [18]. The apparatus operated at a frequency of 2450 MHz and a maximum output power of 1000 W with a 100 W increment. The size of the heating cavity was 37.5 cm (L) × 22.5 cm (W) × 38.6 cm (D). The applied extraction conditions corresponded to those previously optimized [18]. A volume of 32 mL of 42% ethanol concentration was added to 1 g of pericarp *Myrtus* powder in a flat-bottomed flask. The mixture was irradiated at 500 W for 62 s. The resultant extract was then filtered through a sintered glass filter of porosity 2 and was stored at 4 ◦C until further analysis.

#### 2.2.3. Conventional Solvent Extraction

Conventional solvent extract followed the procedure established by Dahmoune et al. [5]. One gram of myrtle powder was placed in a conical flask, and 50 mL of 50% (*v*/*v*) ethanol were added. After stirring for 2 h, the mixture was filtered through a sintered glass filter of porosity 2 and the extract was stored at 4 ◦C until further use.

## *2.3. Analytical Determinations*

#### 2.3.1. Total Phenolic and Flavonoid Contents

The total phenolic content (TPC) of the UAE, MAE and CSE extracts was assessed according to the method of George et al. [20] and expressed as mg of gallic acid equivalent (GAE) per gram of myrtle pericarp powder on dry weight (DW) basis (mg GAE g–1 DW). The total flavonoid content was estimated by the aluminum trichloride method according to Quettier-Deleu et al. [21] and the results were expressed as mg of quercetin equivalent per g of myrtle pericarp powder, on a DW basis.

#### 2.3.2. Total Monomeric Anthocyanins and Condensed Tannin Contents

Total monomeric anthocyanin content was determined by the pH-di fferential method [22], and the results were expressed as mg cyanidin-3- *O*-glucoside equivalents per g of myrtle pericarp powder on a DW basis. The condensed tannin content was determined by the HCl–vanillin method as described by Aidi Wannes et al. [23] and the results were expressed as mg catechin equivalents per g of myrtle pericarp powder on DW basis.

## 2.3.3. Antioxidant Activity

The antioxidant activity of all samples was tested by using two different tests, namely 1,1-diphenyl-2-picrylhydrazyl radical (DPPH -) scavenging activity and reducing power methods [24]. DPPH - solution (60 μM) was prepared in absolute methanol and reaction was performed by the adding of 3 mL of this solution to 1 mL of the extracts, during 20 min at 37 ◦C in the dark. Thereafter, the absorbance was measured at 515 nm. The inhibition rate of the extracts was calculated according Equation (3).

$$\% \text{Scavenging} = \frac{(A\_{\text{control}} - A\_{\text{extract}})}{A\_{\text{control}}} \times 100\tag{3}$$

where *Acontrol* is the absorbance of DPPH - and distilled water *Asample* is the absorbance of DPPH - and sample extract. α-tocopherol and BHA (250 μg/mL) were used as positive controls.

For reducing power assay, 1 mL of desired dilution was mixed with 2.5 mL of sodium phosphate bu ffer (0.2 M, pH 6.6) and 2.5 mL of 1% ( *m*/*v*) potassium ferricyanide K3[Fe(CN)6)], followed by incubation in a water bath at 50 ◦C for 20 min and the addition of 2.5 mL of 10% ( *m*/*v*) trichloroacetic acid. At last, an aliquot of the resulting solution (1 mL) was added to 5 mL of distilled water and 1 mL of 0.1% ( *m*/*v*) of FeCl3·6H2O. Note that this method estimates the ability to reduce Fe3<sup>+</sup> to Fe2+. Antioxidant compounds present in the samples form a colored complex with potassium ferricyanide, trichloroacetic acid and ferric chloride, which is measured at 700 nm.

#### *2.4. Identification of Phenolic Compounds by UHPLC-DAD-ESI-MSn*

The phenolic compounds of the UAE-OPT extract were characterized by UHPLC-DAD-ESI-MSn (DAD: diode array detector; ESI: electrospray ionization) analysis on an Ultimate 3000 (Dionex Co., USA) apparatus equipped with an ultimate 3000 Diode Array Detector (Dionex Co., San Jose, CA, USA) and coupled to a mass spectrometer. Analysis was run on a Hypersil Gold (Thermo Scientific, San Jose, CA, USA) C18 column (100 mm length; 2.1 mm i.d.; 1.9 μm particle diameter, end-capped) and its temperature was maintained at 30 ◦C. The mobile phase was composed of (A) 0.1% of formic acid (*v*/*v*) and acetonitrile (B). The solvent gradient started with 5% of Solvent B, reaching 40% at 14 min and 100% at 16 min, followed by the return to the initial conditions. The flow rate was 0.1 mL min–<sup>1</sup> and UV–Vis spectral data for all peaks were accumulated in the range 200–700 nm while the chromatographic profiles were recorded at 280, 340 and 530 nm.

The mass spectrometer consisted of a Thermo LTQ XL (Thermo Scientific, San Jose, CA, USA) ion trap MS (mass spectrometer) apparatus equipped with an ESI source, operating in negative and positive modes, under the pre-established conditions [25].
