**Contents**



### **About the Editor**

**Carla Gentile** obtained her master's degree in Chemistry and Pharmaceutical Technology and her PhD in Medicinal Chemistry at the University of Palermo. During her PhD, she was a visiting researcher at the Respiratory and Molecular Biology Division of the School of Medicine, University of Southampton (UK) and at the INSERM, Villejuif (France), where she was also a post-doctoral fellow (French Hematology Society, SFH). After a post-doctoral fellowship (Italian Ministry of University and Research, MIUR), in 2008, she became an assistant professor at the University of Palermo. Currently, she is an associate professor in Biochemistry at the Department of Biological, Chemical, and Pharmaceutical Sciences and Technologies, University of Palermo. Her principal scientific interest lies in the evaluation of the biological properties of natural antioxidants, in particular dietary phytochemicals. She is the co-author of more than 70 ISI-indexed publications that have received a total of more than 2000 citations.

### *Editorial* **Biological Activities of Plant Food Components: Implications in Human Health**

**Carla Gentile**

Department of Biological, Chemical and Pharmaceutical Sciences and Technologies (STEBICEF), University of Palermo, Viale delle Scienze, 90128 Palermo, Italy; carla.gentile@unipa.it; Tel.: +39-091-23897423

Scientific data and epidemiological evidence collected over the last fifty years have shown that nutrition plays a decisive role in human health. While before it was considered essentially a non-disease condition, health is nowadays felt as a state of complete physical, mental, and social wellness, according to the definition of the World Health Organization. Eating properly is not only necessary to meet energy needs, avoiding syndromes associated with nutritional deficiency and excess, but it actively contributes to improve human health. The functional meaning of nutrition as surely protective and possibly therapeutic is today amply demonstrated by scientific evidence. Additionally, the increasing sensitivity to the healthy role of nutrition makes consumers more and more careful when choosing high quality foods.

The functional role of nutrition is due to specific small molecules with biological activity. These dietary compounds neither act as energy substrates or plastic materials for cells, nor as enzymatic cofactors. However, due to their peculiar bioactivity, they can benefit human health.

Plants are the most important source of bioactive molecules and still represent the main resource in the quest for new drugs. The interest in dietary phytochemicals is justified by the results from numerous epidemiological studies that demonstrate how diets rich in plant foods are able to prevent several human diseases, including cardiovascular, neoplastic, neurodegenerative, and metabolic pathologies. Of course, it is not possible to demonstrate that the positive effects resulting from the intake of a specific plant food are due to a particular component, for the presence of other phytochemicals and any synergistic effects cannot be neglected. However, experimental results showing biological activity for many pure dietary phytochemicals suggest that particular components may participate in the protective effects associated with the consumption of the plant food it can be sourced from.

The documented biological activity of phytochemicals is expressed through several protective effects, such as antioxidant, anti-inflammatory, antimicrobial, antitumor, immunomodulatory, neuroprotective, antihypertensive, antidiabetic actions. The commercial success of many plant-based supplements is based on these proven biological effects.

Although the protective effect associated with the consumption of plant foods is ascribable to a number of mechanisms, the biological activity of these phytocomponents has frequently been related to their ability to function as antioxidants. Many phytochemicals are in fact redox-active molecules and thus are able to influence the cellular redox balance. For this reason, besides protecting cells from oxidative stress phenomena, phytochemicals influence numerous redox-sensitive biological targets that regulate several important cellular functions. Additionally, the activity of phytochemicals can also involve interactions of these small molecules with various biological targets, including proteins, DNA and lipids, and consequently a possible alteration of their function [1].

This Special Issue addresses the biological activity of dietary phytochemicals, either purified [2] or in extracts from plant foods, and speculates on their potential effects on human health. The studied plant foods include edible parts (fruits, seeds, leaves, and flowers) of plants common in the Mediterranean area, such as pomegranate (*Punica granatum*) [3], oregano (*Origanum vulgare)* [4], and purslane (*Portulaca olearacea*) [5], but also of

**Citation:** Gentile, C. Biological Activities of Plant Food Components: Implications in Human Health. *Foods* **2021**, *10*, 456. https://doi.org/ 10.3390/foods10020456

Received: 29 January 2021 Accepted: 10 February 2021 Published: 19 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

several tropical plants, including annona (*Annona cherymola*) [6], herba mate (*Ilex paraguariensis*), guaranà (*Paullinia cupana*) [7], and ash gourd (*Benincasa hispida*) [8]. Bioactivity was studied in terms of radical scavenging and antioxidant [5–7,9], antimicrobial [10], and antiproliferative activity [4].

The collected results suggest that the intake of the aforementioned plant foods in the context of a balanced diet could be beneficial to health. In addition, considering the high concentration of bioactive molecules in some of the observed plant matrices, the presented results hint at the possibility of using some of the studied plant extracts in food functionalization or in the formulation of dietary supplements.

Due to their ability to exert several biological effects that are potentially useful for human health, over the years, dietary phytochemicals have drawn increasing interest in human nutrition research.

The papers collected in this Special Issue contribute to the growth of this research area. I would like to thank all the authors and the reviewers of the papers published in this Special Issue for their great contributions and effort. I am also grateful to the editorial board members and to the staff of the journal for their kind support during the preparation of this Special Issue.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


### *Article* **Pomological, Sensorial, Nutritional and Nutraceutical Profile of Seven Cultivars of Cherimoya (***Annona cherimola* **Mill)**

**Carla Gentile 1,\*, Giuseppe Mannino 2, Eristanna Palazzolo 3, Giuseppe Gianguzzi 3, Anna Perrone 1, Graziella Serio <sup>1</sup> and Vittorio Farina <sup>3</sup>**


**Abstract:** In this work, the food quality of four international (*Campas*, *Chaffey*, *Fino de Jete* and *White*) and three local (*Daniela*, *Torre1* and *Torre2*) cultivars of Cherimoya (*Annona cherimola* Mill) was investigated. With this aim, pomological traits, sensorial attributes, physiochemical parameters (pH, total soluble content and total acidity), nutritional composition (macro- and micro-nutrients) and nutraceutical values (bioactive compounds, radical scavenging and antioxidant properties) were evaluated. Among the seven observed cultivars, *Fino de Jete* was identified as the best, not only for its commercial attributes such as pomological traits and physiochemical values, but also for its nutritional composition. On the other hand, *Chaffey* and *Daniela* were the cultivars with the highest content of polyphenols, proanthocyanidins, and with the strongest antioxidant capacity. Concerning the two local ecotypes, *Torre1* and *Torre2*, they displayed a balanced nutritional profile that, if combined with their discrete nutraceutical, physicochemical and pomological values, may result in a reassessment of their commercial impact. In conclusion, our data provide interesting information about the pomological, nutritional, and nutraceutical properties of cherimoya fruits. Our results, in addition to promoting the commercial impact of local cultivars, may increase the use of individual cultivars in breeding programs.

**Keywords:** polyphenols; sensory analysis; mineral content; proanthocyanidins; carotenoids; antioxidant activity; FRAP; DPPH; ABTS; CAA

#### **1. Introduction**

The *Annonaceae* Juss. family covers more than 2000 species, of which 120 belong to the genus *Annona* L. [1]. The most famous species are *Annona cherimola* Mill (cherimoya), *Annona muricata* L. (soursop), *Annona squamosa* L. (conde fruit), *Annona reticulata* L. (custard apple), and the interspecific hybrid *Atemoya* (*A. cherimola* × *A. squamosa*). In particular, *Annona cherimola* is the most diffused specie in subtropical countries. It is an indigenous tree of Andean South America and it has naturalized in tropical highlands and subtropical areas of South America [2]. The marketable value of cherimoya is related to its big, heart-shaped and conical fruit [3], which may reach considerable weight and size. The edible flesh of these fruits is white, creamy, and with a custard-like consistency [4]. The aromatic flavor is a mix of papaya, banana, pineapple, and passion fruit [5]. Although cherimoya fruit is consumed as fresh fruit, it can be also processed making yogurt, ice creams and other desserts. It is not recommended to ripen the fruits on the tree, because they lose quality. Cherimoya fruits are generally harvested when not fully mature, and left for ripening

**Citation:** Gentile, C.; Mannino, G.; Palazzolo, E.; Gianguzzi, G.; Perrone, A.; Serio, G.; Farina, V. Pomological, Sensorial, Nutritional and Nutraceutical Profile of Seven Cultivars of Cherimoya (*Annona cherimola* Mill). *Foods* **2021**, *10*, 35. http://dx.doi.org/10.3390/ foods10010035

Received: 9 November 2020 Accepted: 21 December 2020 Published: 24 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/).

under controlled storage conditions. The time of harvesting is commonly determined by the changes of skin fruit color, which turn from green to yellowish-green in the proximity of physiological maturity [6]. In Europe, Spain is the most important producer of cherimoya fruits, and the most important cultivars are *Campas* and *Fino de Jete*, which are also the most widespread cultivars in the global market. In Italy, *A. cherimola* is well adapted to the pedoclimatic conditions of the tyrrhenian coastal areas of Sicily and Calabria, where there are good climate conditions for the production of other exotic and tropical fruits, such as loquat, mango, litchi, avocado, banana and papaya [7–10]. In particular, in Sicily, in addition to affirmed cultivars such as *Fino de Jete*, local ecotypes are also cultivated with very limited diffusion. Concerning the nutritional value, cherimoya flesh possesses a high content of sugars, while having low fat content and, in comparison with other tropical fruits, also good Ca and P content [11]. However, although the nutraceutical properties of other *Annona* species have been extensively studied, those related to cherimoya fruits are much less investigated. The majority of the available literature data suggests that this tropical fruit is an interesting source of bioactive compounds, including polyphenols (catechin, proanthocyanidins, hydroxytyrosol) [12,13], alkaloids (annocherines, norisocorydine, cheritamine, annonaine) [14,15], acetogenins (cherimolin-2 and almunequin) [16], terpenes (myrcene, pinene, linalool, caryophillene, terpenolene and germacrene) [17] and cyclopeptides (cherimola cyclopetide E and cherimola cyclopetide F) [18,19]. In addition, antioxidant [20], pro-apoptotic [3,16,17,21], anti-protozoal [22], and anti-diabetic [23] activities were also demonstrated for extracts obtained from different part of the fruit.

The aim of this study was the investigation of the pomological, physiochemical, sensorial, nutritional and nutraceutical attributes of seven cultivars of cherimoya fruits grown in Sicily. Our results provide comprehensive information on the quality of cherimoya fruits and can be useful for the improvement of the utilization of the specific genotype.

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

#### *2.1. Plant Material*

The fruits were obtained from trees grown in Vivai Torre (Milazzo, Sicily, Italy; 38◦19*-* N, 15◦24*-* E; 20 m a.s.l.). Four international affirmed CVs (*Campas*, *Chaffey*, *Fino de Jete* and *White*) and three local CVs (*Daniela*, *Torre1* and *Torre2*) were selected (Table 1). The fruits of each CV were picked from three 15-year-old trees, grafted on their own rootstock. The trees were planted in North-South direction, with an inter-tree spacing of 6 m and 6 m between rows. The yield per tree was measured by weighing and counting the total number of fruits per tree, and at each harvest time, the trunk circumference was measured at ~15 cm above the graft union. The yield efficiency and crop load were expressed in kilograms or number of fruits per trunk of a cross-sectional area (TCSA) or leaf area. A sample of 30 fruits per CV (10 fruits per 3 trees) was hand-picked when not fully mature, and the color changed from green to yellow. After harvest, the fruits were left to ripen under storage conditions (20 ◦C). Fifteen fruits were employed for pomological and physicochemical analysis, and 5 of them were employed for the sensorial evaluation. Finally, the others 15 fruits were immediately frozen in liquid nitrogen and then stored at −80 ◦C until the analysis of nutrients, bioactive compounds, and antioxidant activity was performed within 3 months. Immediately before the analysis, three fruits for each CV were thawed and peeled, and then the seeds were removed. The chopped pulp was homogenated. Three aliquots of each homogenate for each analysis were employed.


**Table 1.** The name of cultivar (CV), origin, yield, number of fruits per tree, yield efficiency percentage, and crop load of the observed seven *Annona cherimola* cultivars

Values are expressed as mean ± SD of data collected in two years. Among the same series, different lowercase letters indicate significantly different values at *p* ≤ 0.05, as measured by Tukey's tests.

#### *2.2. Pomological and Physiochemical Analysis*

Fruit weight (FW), longitudinal diameter (LD), transversal diameter (TD), seed weight (SW), peel weight (PeW), pulp weight (PW), flesh firmness (FF), total soluble solids content (TSSC), and titratable acidity (TA) were measured. FW, SW, PeW and PW (g) were determined using a digital scale (Gibertini EU-C 2002 RS, Novate Milanese, Italy); LD and TD (mm) using a digital calliper TR53307 (Turoni, Forlì, Italy); FF (kg/cm2) using a digital penetrometer TR5325 with a 8 mm diameter tip (Turoni, Forlì, Italy); TSSC (◦Brix) using a digital refractometer Atago Palette PR-32 (Atago Co., Ltd., Tokyo, Japan) and TA (g citric acid per L) using a CrisonS compact tritator (Crison Instruments, SA, Barcelona, Spain). Skin and flesh colors were calculated using a Konica Minolta Colorimeter based on the CIELAB system that measured the lightness (L\*) and the variation from red (+a\*) to green (−a\*), and from green (+b\*) to yellow (−b\*) in the fruits.

#### *2.3. Sensory Analyses*

A trained panel consisting of 10 judges (5 females and 5 males, between 22 and 45 years of age) performed the sensory profile analysis, as previously reported [7]. All panelists were trained and developed wide expertise in sensory evaluation of tropical fruits. The judges in preliminary sessions generated 24 sensory descriptors (Table 2), and they evaluated samples using a hedonic scale, assigning to each descriptor a score from 1 (absence of the sensation) to 9 (highest intensity). The order of each sample was randomized for each panelist, and water was provided for rinsing between the different samples.

**Table 2.** A list of the evaluated sensory descriptors and their definitions.



**Table 2.** *Cont.*

#### *2.4. Nutritional Parameters*

2.4.1. Content of Carbohydrates, Lipids, Proteins, Water and Ashes

Mealiness MEA A flour-like texture

The carbohydrate and protein content were evaluated as previously described, respectively using Anthrone's [24] and Kjedahl's [25] methods. Ash and water contents were determined through the procedure described in AOAC [26]. The content of lipids was calculated after lipid extraction with a gravimetric method, as previously described [27]. Data were expressed as g per 100 g of Pulp Weight (PW).

#### 2.4.2. Mineral Content

The contents of K, Na, Ca, Mg, Fe, Cu, Mn, and Zn were determined by atomic absorption spectroscopy following wet mineralization, and using the instrumental condition as previously described [28]. Briefly, the samples were digested, and approximately 100 mg of dried sample was weighed and incubated with 9 mL of 65% (*w/w*) HNO3, and 1 mL of 30% (*w/v*) H2O2 were added. The temperature was set at 200 ◦C for 20 min. Once cooled, the digested samples were diluted to a final volume of 50 mL with distilled H2O. All measurements were performed using an Agilent 4200 MP-AES fitted with a double-pass cyclonic spray chamber and OneNeb nebulizer. The calibration standards were prepared by diluting a 1000 mg/L multi-element standard solution (Sigma Aldrich and Scharlab S.L.) in 1% (*v/v*) HNO3. Finally, P was determined using a colorimetric method [29]. Data were expressed as mg per 100 g of Fresh Weight (PW).

#### 2.4.3. Vitamin Content

Retinol (Vit. A), Riboflavin (Vit. B2), Niacin (Vit. B3), and Ascorbic Acid (Vit. C) were extracted and determined according to previously reported methods. Briefly, Vit. A was extracted and quantified using a commercial kit (Vitamin A Food ELISA Kit, Crystal Chem, NL) and following the manufacturer's instructions. Vit. B1 and Vit. B2 were respectively extracted using 0.1 N HCl [30] or a solution of 1% (*v/v*) H2SO4 [31]. Quantification was performed via HPLC equipped with a fluorimetric detector [30,31]. Finally, Vit C was extracted with 10 mL of 1% (*v/v*) HPO3 for 45 min from dried extract, previously prepared [7]. After filtration, 1 mL was mixed with 9 mL of C12H7NCl2O2 and the absorbance was measured at 515 nm against a blank after 30 min. Vitamin C was quantified using a

calibration curve of authentic L-ascorbic acid (0.02–0.12 mg/100 g). Data were expressed as mg per 100 g of Fresh Weight (PW).

#### *2.5. Bioactive Compounds and Antioxidant Activities*

#### 2.5.1. Total Carotenoid Content (TCC)

TCC was determined in flesh homogenates via spectrophotometric analysis after extraction of carotenoids, as previously reported [8]. Data were expressed as β-carotene Equivalent per 100g of PW, using the molecular weight (536.87 g mol−1) and the molar extinction coefficient (2505 M−<sup>1</sup> cm<sup>−</sup>1) of β-carotene in Hexane.

#### 2.5.2. Preparation of Fruit Extracts

Three aliquots of each homogenate were extracted twice with 70% (*v/v*) EtOH using a 1:20 (*w/v*) ratio. After centrifugation (10 min at 10.000 *g*, 4 ◦C) and filtration through a Millex HV 0.45 μm filter (Millipore, Billerica, MA), the supernatants were recovered and combined together. Ethanolic extracts were used both for the determination of bioactive compounds and the antioxidant properties.

#### 2.5.3. Total Polyphenol Content (TPC)

The phenolic content of the flesh of the observed CVs of *A. cherimola* was determined in ethanolic extracts via the Folin-Ciocalteu method, with some minor changes as previously reported [32,33]. Results were expressed as mg Gallic Acid Equivalents (GAE) per 100 g of PW.

2.5.4. Determination of the Total ProAnthocyanidins Content (TPAC) and Investigation of the Polymerization Linkage via HPLC-MS/MS

The proanthocyanidins (PACs) were evaluated in the ethanolic extracts via BL-DMAC assay [34] with some minor changes, as previously reported [35]. The PAC concentration in the extracts was expressed as mg PAC-A equivalent per 100 g of PW.

In order to investigate PAC grade and type, polymerization-binding of catechins was investigated via High-Pressure Liquid Chromatography (HPLC, Agilent 1260, Technologies, Santa Clara, CA, USA) coupled with 6330 Series Ion Trap (Agilent Technologies, USA), as previously reported [9].

#### 2.5.5. Radical Scavenging and Metal Ion Reducing Activity

The antioxidant activity of the ethanolic extracts was measured evaluating both the radical scavenging activity via ABTS [36] and DPPH [37], and the reducing antioxidant power via FRAP [38] assays. Data were expressed as mmol Trolox Equivalent (TE) per 100 of PW as previously reported [35].

#### 2.5.6. Cellular Antioxidant Activity Assay (CAA)

The CAA assay was performed as previously described by Wolfe at al [39], with some minor changes [40]. For the experiments, we used HepG2 (human liver cancer cell line), obtained from American Type Culture Collection (ATCC) (Rockville, MD, USA). The antioxidant activity was expressed as CAA50 that is the amount of flesh in cell medium necessary to obtain the 50% of inhibition of oxidative stress, with respect to the positive control. CAA50 was calculated from concentration-response curves using linear regression analysis, and it was expressed as μg of PW per mL of cell medium.

#### *2.6. Statistical Analysis*

Each assay was repeated three times. All data were tested for differences between the CVs using one-way analysis of variance (ANOVA; general linear model) followed by Tukey's multiple range test for *p* ≤ 0.05, marking significant differences among the samples with different lowercase letters. Principal Component Analysis (PCA) and HeatMap Cluster Analysis were performed using covariant matrix of extraction and varimax rotation. All statistical analyses were performed using SPSS ver. 24. The nucleotide sequences were analysed via CLC software, and the cladogram of gene sequences was performed with ClustalX software by using the Neighbour Joining (NJ) method. Bootstrap values were calculated from 100 resampling of the alignment data.

#### **3. Results and Discussion**

#### *3.1. Pomological and Physiochemical Parameters*

The fruits of the observed *A. cherimola* CVs showed wide variability of the pomological (Table 3) and physiochemical (Table 4) parameters. All the observed CVs, except *Campas* and *White*, reached a considerable size, with a small incidence of SW and PeW on the total FW. For these CVs, the edible part ranged between 73% (*Daniela*) and 87% (*Fino de Jete*) of the total FW. Moreover, *Fino de Jete* and *Daniela* produced the largest and biggest fruits. On the other hand, *Campas* and *White* produced the smallest fruits, with an incidence of non-edible part of about 40%. In addition, significant differences in FF among the different CVs were not recorded. Concerning the edible part, generally, the largest fruits showed also the highest percentage of flesh (PW/FW). The highest yield per tree was obtained in *Daniela* and *Fino de* Jete, whereas *White*, *Campas* and *Torre 2* showed very low values (Table 1). Yield improvement was caused by the increase of the fruit size, rather than to the number of fruits; nevertheless, crop load was higher for *Fino de Jete* and *Chaffey*. Moreover, the highest yield efficiency was observed in *Chaffey* followed by *Fino de Jete.*

**Table 3.** The pomological traits of the seven observed *Annona cherimola* fruits. Data are expressed as mean ± SD. For each row, different lowercase letters mark significant (*p* < 0.05) differences among the samples, as measured by one-way ANOVA followed by Tuckey's test. The letter "a" denotes the highest content.


FW is the Fruit Weight; PW is the Pulp Weight; PeW is the Peel Weight; SW is the Seed Weight; SN is the Seed Number; LD is the Longitudinal diameter; TD is the Transversal diameter; F is the firmness.

> The L\*, a\* and b\* parameters of peel and pulp of Cherimoya fruits were minimally influenced by the genotype (Table 4). Low a\* and high b\* values were recorded in both peel and pulp for all the observed CVs, indicating a brown peel color and a yellow color of the pulp. It was previously suggested that during the maturation of *A. crassiflora* fruits, the decrease of a\* and the increasing of b\* may be related to chlorophyll degradation and carotenoid accumulation, typical of ripening processes [41].

> Also, TSSC varied low among the analyzed CVs, and the mean value recorded was 19.4 ± 1.82 ◦Brix, and the highest values were recorded in *Chaffey* and *Torre1* (Table 4). Our results were similar to those of Andrès-Augustin and colleagues who recorded comparable ranges for TSSC in fruits from commercial and local CVs of Cherimoya from Mexico [42]. Concerning TA, the observed CVs may be grouped in two different subgroups: the first one included *Fino de Jete*, *Torre1*, *Torre2* and *White*, showing a TA value more than 4.0 g malic acid per L; the second one included *Campas*, *Chaffey*, and *Daniela*, with a TA value less than 4.0 g of malic acid per L (Table 4). When the TSSC/TA ratio is considered, the fruits of the observed CVs may be divided between the sweetest (*Campas*, *Cheffey* and *Daniela*) and the bitterest (*Fino de Jete* and *White*). On the other hand, *Torre1* and *Torre2*, the two local CVs, displayed intermediate behaviors.

**Table 4.** Color and physiochemical parameters of the seven observed *Annona cherimola* fruits. For each parameter, different letters indicate significant (*p* < 0.05) differences among the cultivars as measured by one-way ANOVA followed by Tuckey's test. Letter "a" denotes the highest content.


TSSC is the Total Solid Soluble Content; TA is the Titratable Acidity; TSSC/TA is the ratio between the Total Solid Soluble Content; L\* is the lightness; a\* is the variation from red (+a\*) to green (−a\*); b\* is the variation from green (+b\*) to yellow (−b\*). Data are mean values ± SD.

#### *3.2. Sensorial Analysis*

The panel evaluation produced sensory profiles indicating that both commercial and local CVs have good organoleptic characteristics for fresh consumption, thanks to the good combination of some key attributes (Figure 1). In particular, the fruits recorded high values for appearance (APP), skin color (SC), flesh color (FC), consistency (C), juiciness (J), and melon, banana, and pear odor (MBO). The combination of these parameters with the low values for the medicine (MO) and grassy odor (GO) resulted in a good fruit appeal for the consumer.

Concerning the taste, sensorial analysis data showed that the observed fruits had high values of sweetness (S) and low values for astringent (AST), pungent (P), and acid (A) tastes. In particular, acidic taste is a sensorial attribute known to be one of the most important quality traits for the consumer, and its perception may be correlated with the TSSC/TA ratio. However, the lack of significant Pearson correlation between S and TSSC/TA ratio is not surprising since this correlation is typically stronger for more bitter fruits [43]. Sensorial analysis also suggested significant differences among the CVs concerning specific parameters such as juiciness (J), pulp (FC) and peel (PC) color and consistency (C). In particular, panel evaluations suggested that fruits with the lowest FF (*Fino de Jete* and *Torre1*) are those more appreciated because of their consistency.

**Figure 1.** Sensorial descriptors of the observed seven cultivars of *Annona cherimola* fruits. Values are represented as mean ± SD. For each series, the symbol "\*" indicate statistical (*p* < 0.05) differences among the different cultivars, as measured by Student's *t*-test.

#### *3.3. Nutritional Parameters*

The nutritional values per 100 g of the pulp of Cherimoya fruits are shown in Table 5. We recorded a mean moisture content equal to 79.33 ± 1.11 g per 100 g of PW. The mean value for the content of proteins, fats, and sugars was equal to 1.62 ± 0.14, 0.22 ± 0.04, and 14.09 ± 1.23 g per 100 g of PW, respectively. These values are in accordance with those reported from Morton [44]. Nerveless, strong differences in the macronutrients among the observed seven CVs were not measured, *Chaffey* was the CV with the highest sugar content, and *Fino de Jete* the CV with the lowest.


**Table 5.** The nutritional,

 mineral, and vitamin composition

 of the seven observed *Annona cherimola* fruits. Data are mean values ± SD, and they are expressed per 100

Additionally, *Fino de Jete* also showed the lowest fat content. Concerning fibers, they were about 3% of the PW, reaching more than 4% in *Campas*, *Daniela*, *Fino de Jete*, and *Torre1*. Our results, in agreement with other literature data, demonstrated that cherimoya is a tropical fruit with high nutritional value. Indeed, it had low fat content, while containing an amount of sugars and proteins generally higher than other common tropical fruits, including mango [7], kiwi [45], pineapple [46], and papaya [10].

Micronutrients, including minerals and vitamins, are involved in several biochemical processes, and their balanced intake is important to prevent deficiency diseases. Plant foods are important sources of these nutrients [47]. The mineral composition in 100 g of PW is reported in Table 5. K was the most abundant mineral in all the analyzed samples, ranging between 25% and 42% of the total mineral content. Moreover, our analysis showed how Cherimoya, as well as other tropical fruits, is a very reach source of Mg, Ca, and P. In particular, our results showed that the observed seven CVs of Cherimoya had an amount of these micronutrients from two- to four-fold higher than other common edible fruits, such peaches [48,49] and apple [50]. The amount of Na recorded had a mean value of 19.09 ± 2.02 mg per 100 g of PW; that is higher than that reported for others tropical fruits, such as banana, guava, mango, papaya and pineapple [51]. Concerning micro-minerals, our results suggest that cherimoya fruits are an extraordinary source of Zn, containing two-fold the amount normally present in red currant [51]. Moreover, we recorded high amounts of Mn and Cu, meanwhile the Fe contents were markedly lower. Globally, except for Ca and Zn content, our analysis showed a mineral content comparable to that obtained by Leterme [47]. Finally, concerning the general mineral composition among the different CVs, our results revealed significant differences only in K content, recording the highest content in *Fino de Jete* and lowest one in *Torre2*.

Concerning the quantified vitamins, our analysis revealed that the analyzed fruits are a good source of ascorbic acid, with a mean value equal to 37.66 ± 8.41 mg per 100 g of PW. On the other hand, we found a great variability among the observed seven CVs. Indeed, *Torre2* was the CV with the highest content of Vitamin C, recording an amount two-fold higher than *Chaffey*.

#### *3.4. Nutraceutical Parameters*

#### 3.4.1. Total Phenolic Content

Polyphenol compounds are the most abundant dietary phytochemicals [40], and several scientific reports demonstrate their positive influence on human health [9]. On the other hand, several biological actions are documented, including antioxidant [52], antinflammatory [53,54], antidiabetic [55], antiproliferative [3,8], antihypertensive [56], and antihyperlipidemic [57] effects. In this work, TPC in the flesh of the seven observed CVs was measured via Folin-Ciocalteu assay (Figure 2). Our analysis revealed that TPC varied between 28.50 ± 1.92 (*Fino de Jete*) and 174.90 ± 11.69 (*Chaffey*) mg GAE per 100 g of PW, recording an average value equal to 75.18 ± 57.94 mg GAE per 100 g PW. The mean value is higher than those reported for the flesh of other tropical fruits with high commercial impact, including kiwi, papaya, mango and avocado [7,10,45,58]. Furthermore, the mean value for TPC in our fruits was 10-fold higher than for cherimoya fruits from Portugal [59], but 3-fold lower than those obtained for the flesh of fruits from Ecuador [60]. The very large range suggests a significant variability among the analyzed genotypes. Although the contribution of reducing compounds different from polyphenols to TPC value cannot be excluded, the minimal differences in the content of protein and sugar (Tables 4 and 5) among the seven observed cherimoya fruits suggests that the observed range in TPC mainly depended on a different content of polyphenols. Among the observed CVs, the highest value was observed for *Chaffey*, followed by *Daniela*. On the other hand, the CVs with the highest commercial impact (*Campas*, *White* and *Fino de Jete*) recorded the lowest TPC. Finally, the TPC of the two local ecotypes, *Torre2* and especially *Torre1*, was considerable, and only lower than that recorded for *Chaffey* and *Daniela*.

**Figure 2.** The total polyphenolic content (TPC) in the flesh of the nine observed CVs of *Annona cherimola* fruits measured via Folin-Ciocolteau assay. The bars represent mean ± SD. Different lowercase letters on the top of bars indicate significant differences at *p* ≤ 0.05 as measured by one-way ANOVA followed by Tukey's multiple range test. The letter "a" denotes the highest value.

In our previous work, we also evaluated the phytochemical composition of leaves obtained from the same observed CVs of *A. cherimola* [3]. Comparing TPC values measured in the leaves with that recorded in this work for the flesh of the respective fruits, we found that for all the observed CVs, the leaves contained more polyphenols than the fruits. Moreover, for the leaves, a different ranking for TPC was observed. In particular, *Torre2*, *White*, and *Fino de Jete* were the CVs with the highest TPC in the leaves; meanwhile, *Daniela* and *Torre1* were those with the lowest [3].

#### 3.4.2. Content of Proanthocyanidins

Proanthocyanidins are polyphenols of high molecular weight with documented protective actions for human well-being [61]. In particular, their potential protective effect on the gastrointestinal tract is very interesting. Indeed, thanks both to their high digestive stability at gastrointestinal conditions, and to their reduced intestinal absorption [62–64], PACs may reach high concentrations in the intestinal lumen, producing significant biological effects at the local level. The potential benefit of PACs in chronic intestinal inflammation is supported by numerous studies [9,40,65–70], and epidemiological data show an inverse correlation between food intake rich in proanthocyanidins and the risk of developing colorectal cancer [9,71,72].

We evaluated tPACs in the flesh of the seven observed cherimoya fruits via BL-DMAC assay (Figure 3). We recorded a tPAC content ranging between 10.33 ± 4.51 (*Fino de Jete*) and 51.67 ± 4.04 (*Daniela*) mg PAC-A equivalent per 100 g of PW, with mean values of 28.54 ± 7.98 mg PAC-A equivalent per 100 g of PW. In particular, the highest tPACs was recorded for *Daniela*, followed by *Chaffey*, *Torre2* and *Torre1*; meanwhile, the lowest content was recorded for the *Fino de Jete* and *Campas*.

**Figure 3.** Total proanthocyanidin content (tPACs) in the flesh of the nine observed CVs of *Annona cherimola* fruits measured via BL-DMAC assay. The bars represent mean ± SD. Different lowercase letters on the top of the bars indicate significant differences at the *p* ≤ 0.05 level as measured by oneway ANOVA followed by Tukey's multiple range test. The letter "a" denotes the highest value. Inside each bar, the different colors indicate the percentage composition of PAC measured by HPLC-MS/MS, as reported in the Materials and Methods section.

Finally, the correlation coefficient (*p* = 0.76), measured by Pearson statistical analysis (Figure 4), suggested that tPCA strongly contributes to the previously measured TPC value. Our analysis is in accordance with García-Salas et al., who evaluated the tPAC and TPC contents in ethanolic extracts from fruits of two different CVs of Cherimoya (*Fino de Jete* and *Campas*) cultivated in Spain [12]. In particular, they showed that cherimoya pulp essentially contains PACs in addition to hydroxytyrosol and traces of luteolin [12].

Concerning the PACs profile, HPLC-MS/MS analysis revealed that in cherimoya fruits, B-type PACs represented about 90% of the total PACs. In particular, *Daniela* displayed the highest percentage, containing more than 94% of B-type PAC; meanwhile, *Chaffey* was the CV with the highest A-type PACs percentage, reaching more than 13%. Even if the greater bioactivity of PAC-A with respect to PAC-B type is well-known [3,9,32,73], the presence of the A-type PAC is very limited in food sources [74,75]. On the other hand, the literature data suggest a higher dosage of PAC-B can exert comparable bioactivity to the PAC-A type [73]. Regarding the polymerization grade of PACs contained in our fruit extracts, we found dimers and a small number of trimers. Our findings are in accordance with Garcia and colleagues, that reported almost exclusively low molecular weight PACs in cherimoya flesh [12]. Although the limited PAC bioavailability, experimental scientific data indicate that their intestinal absorption is inversely related to the polymerization degree [62]. Consequently, dimers and trimers of PAC may also poorly absorbed at the intestinal level [62,76]. Our results would therefore indicate that the cherimoya PAC fraction may be at least partially bioavailable.

**Figure 4.** A Pearson's correlation heat map displaying the correlation coefficient (ρ) based on physiochemical, nutritional, and nutraceutical data of the seven observed cultivars of *Annona cherimola* fruits. Different colors represent the negative (red) to positive (blue) correlation between two different parameters.

#### 3.4.3. Content of Carotenoids

Several scientific studies probe how carotenoid intake contributes to preventing human diseases related to oxidative stress [77]. On the one hand, animals are not able to synthesize carotenoids, and plant foods constitute the major carotenoid sources in the human diet [7,77]. The TCC of the seven observed CVs of *A. cherimola* fruits is reported in Figure 5. Our results showed that cherimoya fruits contain very low amounts of carotenoids. Indeed, the TCC ranged between 10.33 ± 4.51 (*Torre1*) and 51.67 ± 4.04 (*Torre2*) μg β-carotene per 100 g of PW, with an average value of 29.22 ± 12.47 μg β-carotene per 100 g of PW. This value is about 100-fold less than that recorded for mango and papaya fruits [7,10]. On the other hand, our results were in accordance with those listed in the USDA National Nutrient Database [78].

**Figure 5.** The Total Carotenoid Content (TCC) in the flesh of the nine observed CVs of *Annona cherimola* fruits. The bars represent mean±SD. Different lowercase letters on the top of bars indicate significant differences at *p* ≤ 0.05 as measured by one-way ANOVA followed by Tukey's multiple range test. The letter "a" denotes the highest value.

#### 3.4.4. Antioxidant Properties

Phytocomponents display various biological properties, and the determination of the potential bioactivity of plant extracts may contribute to their valorization for food fortification, but also be of use in the cosmetic and pharmaceutical fields [3,9]. Frequently, the bioactivity of phytochemicals is related to their antioxidative properties, not only preventing oxidative stress but also being useful for the modulation of important redoxdependent cellular functions [48,79].

In this work, the antioxidant properties of the ethanolic extracts of the flesh of the observed cherimoya fruits were evaluated in in solution assays and in a cell-based lipid peroxidation model. Regarding in solution assays, ABTS and DPPH were used to measure the radical scavenging activity; meanwhile, FRAP was used to evaluate the metal-reducing activity (Figure 6, Panel A). The mean values for the radical scavenging activity evaluated by DPPH and ABTS assay were 363.11 ± 153.59 and 228.75 ± 90.52 mmol TE per 100 g PW, respectively. A lower average value (1.36 ± 0.59 mmol TE per 100 g PW) was measured for the metal-reducing activity via FRAP assay. Peculiar characteristics of the reaction mixtures of the different assays and specific differences in the electronic transfer mechanism may explain the different antioxidant activities recorded [80]. Despite the variability evaluated in the antioxidant activity, the trend among the analyzed CVs was not influenced by the different assays, as suggested by the positive correlation (Figure 4) between the values obtained from the three assays (ρABTS/DPPH = 0.968; ]ρABTS/FRAP = 0.917; ρDPPH/FRAP = 0.949). On the other hand, the obtained results highlight a significant variability in the nutraceutical potential of the analyzed CVs. *Chaffey* and *Daniela* always showed the highest antioxidant activities; meanwhile, *White*, *Fino de Jete* and *Campas* displayed the lowest ones, both in terms of radical scavenging and reducing activity.

**Figure 6.** The antioxidant activities of ethanolic extracts of the flesh of the seven observed CVs of *Annona cherimola*. Panel (**A**) shows values measured by the radical scavenging (DPPH = violet; ABTS = light blue) and metal-reducing (FRAP = yellow) assays. Panel (**B**) shows THE CAA50 value. Within the same series, different lowercase letters indicate significant difference at the *p* ≤ 0.05 level as measured by one-way ANOVA followed by Tukey's multiple range test. The letter "a" denotes the value.

Additionally, the obtained values were positively correlated with TPC (ρTCP/DPPH = 0.983; ρTCP/ABTS = 0.940; ρTCP/FRAP = 0.928) (Figure 4), indicating that polyphenols contribute almost exclusively to the redox-active properties. Moreover, tPACs was also strongly correlated with FRAP (ρTPAC/FRAP = 0.844), but it found a lower correlation with ABTS and DPPH (ρTPAC/ABTS = 0.654; ρTPAC/DPPH = 0.664). The higher correlation found between FRAP and tPAC may be explained by the peculiar structural characteristics of proanthocyanidins, making it easier to bind metal ions thanks the presence of free metaoriented hydroxyl groups [3,7].

Although in solution assays are widely employed to preliminarily evaluate the antioxidant capacities, they cannot measure the antioxidant activity in a biological environment [40,80]. For this purpose, cellular-based lipid peroxidation models evaluate the potential ability of redox-active compounds to interact with biological membranes [39,81,82]. Therefore, they represent interesting alternatives to in vivo models, which instead may be expensive, unethical, and not easy to use [39]. In particular, the CAA assay is a very biologically relevant method because it also takes into account the uptake, metabolism, and location of antioxidant compounds within cells [39,81]. The antioxidant activity of the ethanolic extracts of the fruits from the seven observed CVs of *A. cherimola* expressed

as CAA50 is displayed in (Figure 6, Panel B). The average CAA50 value was 7.94 mg ± 1.81 of the PW per mL of cell medium was recorded. The obtained values are in the same range as those determined by Wolfe et al. under the same experimental conditions for hydrophilic extracts of other fresh fruits [81]. Among the observed CVs, a little variability in term of CAA50 was recorded. In particular, *Chaffey* displayed the highest activity, followed by *Daniela* and *Torre1*. Although in-solution assays are not always predictive of the antioxidant capacity in biological models, in our experimental conditions, the antioxidant activity of CAA is positively correlated with the redox-active properties evaluated by in solution assays (ρCAA/ABTS = −0.943; ρCAA/DPPH = −0.879; ρCAA/FRAP = −0.818) (Figure 4). Furthermore, the strong correlation between the CAA50 values and TPC (ρCAA/TPC = −0.923) and the lower correlation with tPACs (ρCAA/TPAC = −0.608) of the tested extracts suggested that the main contribution to CAA was not mainly given by the PACs, but by other polyphenol compounds. This result is not surprising, considering the low capacity of polyphenolic polymers to cross the cellular membrane [62].

#### *3.5. Cultivar Discrimination via Principal Component Analysis and HeatMap Cluster Analysis*

The Principal Component Analysis (PCA), calculated on the data matrices related to pomological, physiochemical, nutritional and nutraceutical values previously measured, allowed for the discrimination between the different fruits of the seven observed CVs of *A. cherimola* (Figure 7). In particular, PCA explained 27.32% and 51.73% of the total variance, respectively for PC1 and PC2. Positive factor scores discriminated *Fino de Jete* and *Daniela* from others CVs. In particular, *Fino de Jete* is the best CV for the highest values related to the most important commercial parameters, such as FW, LD, TD, F, and PW, but it also showed the best nutritional profile among the different CVs, having a good vitamin and mineral content. On the other hand, *Daniela* showed the best phytochemical profile and the highest antioxidant properties while having intermediate pomological, physicochemical, and nutritional values. *Campas* was completely separated from other CVs for positive PC2 and negative PC1. In particular, the separation is mainly due to the high mineral content, and to poor nutraceutical properties and very poor pomological traits. Finally, *Torre1*, *Torre2*, *White* and *Chaffey* were grouped for negative PC2. Especially, *Torre1* and *White* had also negative PC1 due to their similar pomological characteristics. On the other hand, positive PC1 and negative PC2 factor scores grouped *Torre2* and *Chaffey* for their particular PAC composition. However, *Chaffey* also had nutritional and nutraceutical properties better than *Torre2*.

The HeatMap coupled with Hierarchical Clustering Analysis confirmed the separation performed by PCA (Figure 8). In particular, *Fino de Jete* was completely separated from other CVs due to high values recorded both for some of the pomological traits, such as FW, LD, TD, PW, PeW and SN, and for the highest vitamin content. On the other hand, *White* were really far from *Fino de Jete* because the lowest values recorded for all the pomological traits of fruits, and also for the low vitamin and phytochemical content. For the same reason, *Campas* was very close to *White*. Clustering analysis revealed that the two local ecotypes, *Torre1* and *Torre2*, had not only similar pomological parameters, but they also displayed a comparable antioxidant activity both in solution and in cellular models. This proximity may also be explained by their similar PAC profile. Finally, *Daniela* and *Chaffey* take place in an intermediate position within the clustering due to their acceptable values recorded both for pomological traits, nutritional values, and nutraceutical properties.

**Figure 7.** Scatter plot of the principal components factor scores of the seven observed CVs of *Annona cherimola* fruits. Panel (**A**) shows the clear separation among the different CVs; meanwhile, Panel (**B**) reports the chemical portioning of the compounds.

**Figure 8.** Hierarchical clustering analysis and heatmap visualization of the physiochemical, nutritional, and nutraceutical data of the seven observed cultivars of *Annona cherimola* fruits. For each row, diverse colors indicate differences between the values measured for each parameter among the seven cultivars.

#### **4. Conclusions**

In the present work, we demonstrated the overall high quality of cherimoya fruits harvested from plants grown in Sicily, in terms of pomological, physiochemical, nutritional, and nutraceutical attributes. Our results showed great variability among the seven observed CVs and may contribute to better define the potential commercial impact of the different CVs. In particular, our analysis showed *Fino de Jete* as being a commercially appreciated CV for its pomological and physiochemical attributes; it also had high nutritional values. On the other hand, the local CV *Daniela*, together to good commercial attributes, also displayed good nutraceutical properties. Concerning *White*, a CV less requested on the market, in addition to having low pomological attributes, it had reduced nutritional and nutraceutical values. Finally, the two local ecotypes, *Torre1* and *Torre2*, had qualitative attributes comparable to those measured for the international CVs.

**Author Contributions:** Conceptualization, C.G.; Data curation, C.G., G.M., G.S., and V.F.; Formal analysis, G.M., E.P., G.G., A.P., G.S., and V.F.; Funding acquisition, C.G.; Investigation, G.M. and E.P.; Methodology, G.M.; Project administration, C.G.; Resources, C.G.; Software, G.M. and G.G.; Supervision, C.G., G.M. and V.F.; Validation, C.G., G.M. and E.P.; Visualization, G.M. and G.G.; Writing—original draft, C.G. and G.M.; Writing—review & editing, C.G. and G.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors would like to thank Vivai Torre s.r.l. for providing samples and for their kind hospitality. Moreover, we would like to express our gratitude to Valentina Sciacca and Stefano Puccio for their technical support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


#### *Article*

## **Antibiofilm and Enzyme Inhibitory Potentials of Two Annonaceous Food Spices, African Pepper (***Xylopia aethiopica***) and African Nutmeg (***Monodora myristica***)**

**Alfred Ngenge Tamfu 1,2,3,\*, Ozgur Ceylan 2, Selcuk Kucukaydin 4, Mehmet Ozturk 5, Mehmet Emin Duru <sup>5</sup> and Rodica Mihaela Dinica 3,\***


Received: 29 October 2020; Accepted: 27 November 2020; Published: 29 November 2020

**Abstract:** Food pathogens represent an important health threat, and it is relevant to study the effect of foodstuffs such as spices which can inhibit bacterial growth. This study reports the antimicrobial, antibiofilm, and enzyme (Acetylcholinesterase, Butyrylcholinesterase, urease, tyrosinase) inhibitory activities of two medicinal food spices belonging to the Annonaceae family, *Monodora myristica* and *Xylopia aethiopica*. GC-MS (gas chromatography mass spectrometry) analysis of silylated samples of Methanol-Dicloromethane (50:50) extracts of both plants led to the identification of nine compounds in *M. myristica* and seven compounds in *X. aethiopica*. *M. myristica* and *X. aethiopica* had the same minimum inhibitory concentration (MIC) values of 0.625 mg/mL and 2.5 mg/mL on *C. albicans* and *E. coli*, respectively. However, *M. myristica* had better activity than *X. aethiopica* on *Staphylococcus aureus*, while *Pseudomonas aeruginosa* was more susceptible to *X. aethiopica* than *M. myristica*. The lowest MIC value was 0.1325 mg/mL, exhibited by *M. myristica* on *S. aureus*. Both extracts showed good antibiofilm activity. On *S. aureus*, at the same concentration, *M. myristica* had better antibiofilm activity than *X. aethiopica*. On *E. coli* and *Candida albicans*, *X. aethiopica* had better antibiofilm activity than *M. myristica* at the same concentration. *X. aethiopica* showed better violacein inhibition in *Chromobacterium violaceum* CV12472, as its percentage inhibition of violacein varied from 80.5% ± 3.0% at MIC to 5.6 ± 0.2 at MIC/8, as compared to *M. myristica* with 75.1% ± 2.5% at MIC and 15.5% ± 1.1% at MIC/8. The anti-motility activity by swimming and swarming inhibition on *P. aeruginosa* PA01 was low at test concentrations and in both models, *M. myristica* showed higher motility inhibition than *X. aethiopica*. Although in enzyme inhibitory assays all extracts had low inhibitions compared to standards tested at the same concentrations, the results show that these plants can be used to manage food-borne infections.

**Keywords:** African food spices; GC-MS (gas chromatography mass spectrometry); antimicrobial; antibiofilm; violacein inhibition; swarming inhibition; swimming inhibition; anticholinesterase; antiurease; antityrosinase

#### **1. Introduction**

In every region of the world, selected indigenous plants are used as foods and spices, and investigating their chemical composition and bioactivities has become an interesting field of research. This is because it reveals both the nutritive value and medicinal potential of these food materials used as remedies especially for recalcitrant infectious diseases resulting from food contamination. Spices are defined by Corn et al. (1999) as ingredients usually from vegetables or different dried plant parts like barks, seeds, and leaves added in nutritionally small quantities to food in order to improve its color, taste, or flavor and may also play the role of preservatives that inhibit harmful bacterial growth [1,2]. Most spices are added to food recipes primarily for flavoring, seasoning, and imparting aroma to foods rather than for their nutritional benefits, and for this reason, their phytochemical compositions as well as bioactivities are still under-studied [3]. There is need for the search of locally available nutritional food excipients in Africa and other low income countries which are plagued with poor nutrition [4]. *Xylopia aethiopica* and *Monodora myristica* are two widely consumed food spices in Africa and both belong to the Annonaceae family of plants and have proven to be potent in managing microbial and fungal infections [5]. Annonaceous plants are commonly called custard apples and comprises over 2300 species which possess good biological activities [6].

*X. aethiopica*, also called African pepper, is a highly consumed food spice in Africa and it is used traditionally to manage rheumatism, bronchitis, headache, asthma, stomach-aches, neuralgia, dysenteric conditions, wounds and sores, constipation, epilepsy, fertility, and the ease of childbirth [7]. It has been shown to possess anticancer, antidiabetic, antimalarial, antioxidant, enzyme inhibitory, antimicrobial, and antibacterial properties and also protects against liver and kidney damage [3,7–14].

*Monodora myristica*, also called African nutmeg, bears many fruits. The seeds of this plant are mostly used in dry powdered form which is used as spice in desserts, stews, soups, and cakes. It is also marketed as whole seeds. These plants' seeds and powders are used in repelling insects, as a stimulant, and as remedy for sores, headache, and stomach disorders. In medicine, the bark is used in treatments of stomachaches, febrile pains, eye diseases, and hemorrhoids [15,16]. This plant possesses antioxidant, antisplasmodic, antiulcer, antimicrobial, cytotoxic, antinociceptive, anti-inflammatory, anticancer, and hepatoprotective activities, and it has been proven in male Wistar rats with induced hypercholesterolemic to modulate lipid peroxidation and bring down cholesterol levels [3,17–25].

These plants have been reported to possess antimicrobial activities, with more emphasis on their essential oil and little information about their extracts, meanwhile the plant seeds are consumed in whole. No studies have reported the ability of these plants to inhibit quorum-sensing-mediated effects in bacteria such as biofilm, violacein production, and swarming and swimming motilities. The aim of this work is to evaluate the ability of *M. myristica* and *X. aethiopica* extracts to inhibit biofilm formation, violacein production, and swimming and swarming motilities, as well as acetylcholinesterase, butyrylcholinesterase, urease, and tyrosinase enzymes.

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

#### *2.1. Plant Material and Extraction*

The seeds of *Xylopia aethiopica* and *Monodora myristica* were purchased from the Bamenda food market. Both spicy plants were identified by Mr. T. Fulbert, a botanist working at the National Herbarium of Cameroon and compared with existing voucher specimens 28725/SRF Cam. (*X. aethiopica*) and 49544/HNC (*M. myristica*). Of each of the seeds, 200 g were powdered and subjected to maceration extraction. Then, 2 L of dichloromethane/methanol mixture in the ratio 1:1 was used as the solvent for extraction process and the mixture was allowed to stand at room temperature for 48 h with intermittent stirring. After this, the supernatant was carefully decanted and filtered using a Whatman number 1 filter paper. This filtrate was evaporated on a Rotary evaporator to remove the solvent. This process was repeated three times for each sample to yield crude extracts of *X. aethiopica* (27 g) and *M. myristica* (41 g).

#### *2.2. GC-FID (Gas Chromatography Flame Ionization Detector) and GC-MS Analyses*

Prior to the GC-MS analyses, the samples were silylated using BSTFA [bis (trimethylsilyl)-trifluoroacetamide], according to the method described by Talla and coworkers [26]. GC-FID and GC-MS were performed as described elsewhere [27]. The GC-MS profiles of the extracts were achieved on a gas chromatograph (Hewlett-Packard 5890, Bunker Lake Blvd, Ramsey, MN, USA) with a JEOL MS-600H mass spectrometer (Tokyo, Japan) as detector. Prior to this, GC-FID was performed on a Shimadzu GC-17(Shimadzu Corp., Kyoto, Japan). We used helium (1 mL/min) as the carrier gas at a split ratio of 1:10 in a SPB-5VR capillary column of length 30 m and inner diameter 0.25 mm. The initial temperature of the oven was varied as follows: 60 ◦C for 3 min, increased at the rate of 5 ◦C/min to 180 ◦C, and finally at 7 ◦C/min to 300 ◦C final temperature. A ZB-5MSVR column of 30 m length and 0.25 mm inner diameter was used for GC-MS and the same temperature conditions as for GC-FID were applied. Next, 250 ◦C and 70 eV was applied for the ion source. Mass spectral fingerprints were used for identification on the NIST library and compared with some data reported.

#### *2.3. Determination of MIC (Minimal Inhibitory Concentration)*

The bacterial and fungal strains *Staphylococcus aureus* (ATCC 25923), *Escherichia coli* (ATCC 25922), *Candida albicans* (ATCC 10239), *Chromobacterium violaceum* (CV12472), and *Pseudomonas aeroginosa* (PA01) were used.

The broth dilution method, as described by the CLSI (Clinical and Laboratory Standards Institute, 2006), was applied to determine the MIC values [28]. The lowest concentration of extract at which no bacterial growth was visible was considered as the MIC. Mueller-Hinton Broth (MHB) was used as the medium and the bacterial concentration used had a density of 5 <sup>×</sup> <sup>10</sup><sup>5</sup> colony-forming units (CFU)/mL. Into 96-well microtiter plates, containing extracts at concentrations (10, 5, 2.5, 1.25, 0.625, 0.312 mg/mL), 100 μL of microbial cell solutions were inoculated and incubated for 24 h at 37 ◦C, after which the MICs were determined and recorded.

#### *2.4. Assay of Inhibition of Bacterial Biofilm Formation by Extracts*

The ability of the extracts to prevent biofilm formation by bacteria (*S. aureus, E. coli* and *C. albicans*) was evaluated at concentrations of 1, 1/2, 1/4, 1/8, and 1/16 (MIC) using the microplate antibiofilm method [29]. Tryptose-Soy Broth (TSB) containing 0.25% glucose were filled into wells with or without extracts, and 200 <sup>μ</sup>L of 1% overnight bacterial cell cultures (5 <sup>×</sup> 105 CFU/mL) were added and incubated at 37 ◦C for 48 h. The negative control wells contained only TSB and bacterial cells. The planktonic bacteria were washed after incubation, and the remaining bacteria were stained with 0.1% crystal violet solution and allowed to sit for 10 min. The dye was washed out carefully with distilled water, after which 200 μL of ethanol or 33% glacial acetic were filled into the wells of the microplates. Next, 125 μL of the resulting solution were transferred using a pipette into sterile tubes and the total volume was made up to 1 mL by adding distilled water. The optical density of each tube was read at 550 nm, and the equation below was used to calculate the percentage inhibition of the biofilm formation. Each experiment was done three times.

$$\text{Biofillm inhibition} \left( \% \right) = \frac{OD \left( \\$50 \text{ control} - OD \left( \\$550 \text{ sample} \right) \times 100 \right)}{OD \left( \\$50 \text{ control} \right)} \times 100$$

#### *2.5. Violacein Pigment Inhibition Assay*

This assay measures the qualitative quorum-sensing potential of the extracts using *Chromobacterium violaceum* ATCC 12472 strain [30]. *C. violaceum* CV12472 was grown overnight and 10 μL of it were put into microtiter sterile plates filled with 200 μL of LB broth. Sub-MIC concentrations of extracts were added and then incubated at 30 ◦C for 24 h. Control plates contained only LB broth and *C. violaceum* ATCC 12472. The decrease in the production of violacein pigment was measured by taking the absorbance at 585 nm. The violacein inhibition percentage of the extracts were calculated as follows:

$$\text{Vidacein pigment inhibition} \left( \% \right) = \frac{OD \ 585 \text{ control} - OD \ 585 \text{ sample}}{OD \ 585 \text{ control}} \times 100$$

#### *2.6. Swimming and Swarming Motility Inhibition on Pseudomonas Aeruginosa PA01*

The ability of the extracts to inhibit swarming motility in *P. aeruginosa* PA01 was performed as described elsewhere [30]. Briefly, swarming plates consisting of, 0.5% NaCl, 1% peptone, 0.5% agar, and 0.5% D-glucose together with extracts at three concentrations of 50, 75, and 100 μg/mL were prepared. *P. aeruginosa* PAO1 was grown overnight and 5 μL of it were point-inoculated at the center, and the swarm plates were wrapped with paraffin and inoculated in an upright position for 18 h. The plates not containing extracts were used as controls. The swarming movement was measured from the swarm diameter fronts.

The swimming plates consisted of 1.5% agar, 1% peptone, 0.5% NaCl, and 0.5% D-glucose together with the extracts at 50, 75, and 100 μg/mL concentrations. The same bacteria *P. aeruginosa* PAO1 was inoculated as in the swarming model. The inoculation was done for 18 h and control plates did not contain extracts. Swim zone diameter for samples and controls were used in calculating the percentage inhibition of swimming motility.

#### *2.7. Cholinesterase Inhibition Assay*

Ellman's Method was used to evaluate the acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) inhibition potential of extracts, and slight modifications were made [31]. Next, 96-well microplates were used and acetylthiocholine iodide (0.71 mM) was the substrate in AChE assay, while butyrylthiocholine chloride (0.2 mM) was the substrate for the BChE assay. In a 96-well plate containing 150 μL sodium phosphate buffer 100 mM (pH = 8), 10 μL of the sample were mixed with 20 μL of enzymes AChE or BChE. The mixture was incubated at 25 ◦C for 15 min, after which 10 μL of substrates and 10 μL of Ellman's Reagent (DTNB 0.5 mM) were added and the volume made up to 200 μL. The absorbance was measured at 412 nm for 10 min. The percentage inhibition of AChE or BChE was determined control using the formula:

$$(\text{E} - \text{S})\text{E} \times 100\tag{1}$$

where:

E: activity of enzyme with control. S: activity of enzyme with sample.

The experiments were repeated three times. Galantamine was used as the standard.

#### *2.8. Tyrosinase Inhibition Assay*

A spectrophotometrical method was used to evaluate the anti-tyrosinase activity of extracts in which tyrosinase enzyme from mushrooms was used following a method described elsewhere with a slight modification [32]. The substrate used in this assay was L-Dopa, while kojic acid was used as the standard inhibitor of tyrosinase. The percent inhibition of the enzyme (Inhibition %) by the extracts was calculated at each sample concentration (μg/mL) in a similar manner as in the AChE and BChE assay.

#### *2.9. Urease Inhibition Assay*

The indophenol method, in which the production of ammonia is measured, was used to determine the potential of extracts to inhibit urease [33]. A mixture of 25 μL of a Jack bean source urease enzyme, phosphate buffer 100 mM (pH 8.2), and 50 μL of urea 100 mM was prepared and after the adding of the samples (10 μL, 1 mM), it was incubated for 15 min at 30 ◦C. Subsequently, 45 μL of phenol reagent 1% (*w*/*v*) and 70 μL of 0.005% (*w*/*v*) alkali reagent were both added into each well and the mixture further

incubated for 50 min. The standard used was Thiourea. The absorbances were recorded at 630 nm and the % inhibitions were calculated.

#### *2.10. Statistical Analysis*

Each activity was done in triplicate. The results were recorded as the means ± standard error of the mean. Fisher's test was used to determine the significant differences between means; *p* < 0.05 were regarded as significant.

#### **3. Results**

#### *3.1. GC-MS Chemical Composition*

Nine compounds (Figure 1), including 1-monolinoleoylglycerol, 6,9,12-octadecatrienoic acid benzyl ester, 3-hydroxyspirost-8-en-11-one, ethyl-3,4,5-trimethoxybenzoate, palmitic (hexadecanoic) acid, ursodeoxycholic acid and sugars such as glycerol, glucose, and sucrose were identified in *M. myristica*. On the other hand, seven compounds (Figure 2), were identified in *X. aethiopica*, these were 3-carene, eucalyptol, 2-hydroxy-4-methylbenzoic acid, abietic acid, 3,21-dihydroxypregnan-4-one alongside two sugars, fructose, and glucose.

**Figure 1.** Chemical compounds identified in *M. myristica* seed extract by gas chromatography mass spectrometry (GC-MS).

**Figure 2.** Chemical compounds identified in *X. aethiopica* seed extract by GC-MS.

#### *3.2. Minimal Inhibitory Concentrations*

The minimal inhibitory concentrations (MICs) of both extracts are reported in Table 1. Both plants *X. aethiopica* and *M. myristica* had the same MIC values of 0.625 mg/mL and 2.5 mg/mL on *C. albicans* and *E. coli*, respectively. However, *M. myristica* had better activity than *X. aethiopica* on *S. aureus,* while *P. aeruginosa* was more susceptible to *X. aethiopica* than *M. myristica*. The lowest MIC value was 0.1325 mg/mL, exhibited by *M. myristica* on *S. aureus*.



#### *3.3. Percentage Biofilm Inhibition*

Prior to antibiofilm assay, the MIC values were determined and the biofilm inhibition potential of both extracts determined at MIC and sub-MIC concentrations and reported in Table 2. Both extracts had good antibiofilm activity on *S. aureus*. On *S. aureus*, the biofilm inhibition of *X. aethiopica* varied from 45.3% ± 1.5% at MIC to 5.2% ± 1.0% at MIC/8 while that of *M. myristica* varied from 52.6% ± 3.3% at MIC to 11.3% ± 1.5% at MIC/8. On *S. aureus*, at the same concentration, *M. myristica* had better antibiofilm activity than *X. aethiopica*. On *E. coli* and *C. albicans*, *X. aethiopica* had better antibiofilm than *M. myristica* at the same concentration (*p* < 0.01).


**Table 2.** Effects of different concentrations of *X. aethiopica* and *M. myristica* extracts on tested bacteria biofilm formations.

\*: Data are the mean of three replicates ± SD. Statistically different (*p* < 0.01).

#### *3.4. Violacein and Motility (Swimming and Swarming) Inhibition Percentages*

The percentage inhibition of violacein pigment synthesis by *C. violaceum* CV12472 was evaluated at MIC and sub-MIC concentrations and reported in Table 3. *X. aethiopica* inhibited violacein production better than *M. myristica*, as it percentage inhibition of violacein varied from 80.5% ± 3.0% at MIC to 5.6 ± 0.2 at MIC/8 as compared to *M. myristica* with 75.1% ± 2.5% at MIC and 15.5% ± 1.1% at MIC/4. At MIC/16, no inhibition was observed.

The swimming and swarming inhibition assay was carried on *P. aeruginosa* PA01 at 100 μg/mL, 75 μg/mL, and 50 μg/mL, and the results are given in Table 3. The anti-motility activity of these plant extracts were low at test concentrations. In both models, *M. myristica* showed higher motility inhibition than *X. aethiopica* (*p* < 0.05).


**Table 3.** Effects of *X. aethiopica* and *M. myristica* extracts at different concentrations on qualitative violacein inhibition and swarming/swimming motility.

\*: Data are the mean of three replicates ± SD. Statistically different (*p* < 0.05).

#### *3.5. Enzyme Inhibitory Activities*

The capacity of *X. aethiopica* and *M. myristica* to inhibit some selected enzymes acetylcholinesterase, butyrylcholinesterase, tyrosinase, and urease are evaluated and reported in Table 4. On AChE, the activities of *X. aethiopica* (7.80% ± 0.36%) and *M. myristica* (9.65% ± 0.11%) were almost the same but far lower than that of the standard galantamine (80.76% ± 0.52%). This was the same effect for BChE on which percentage inhibitions of *X. aethiopica* and *M. myristica* were 15.48 ± 0.57 and 13.64 ± 0.88 respectively against 74.43% ± 0.30% for galantamine.

*X. aethiopica* (30.45% ± 0.84%) had better inhibition on tyrosinase than *M. myristica* (23.44% ± 1.27%) while on urease, *M. myristica* (15.18% ± 0.86%) was more active than *X. aethiopica* (5.69% ± 0.33%).

**Table 4.** Anticholinesterase, antityrosinase, and antiurease activities (%inh.) of test samples at 100 μg/mL.


NT: Not tested. Statistically different (*p* < 0.01).

#### **4. Discussion**

The chemical composition of both plant extracts were effected using GC-MS. The GC-MS was done after the samples were silylated to enable the detection of a large number of compounds including polar ones. Silylation helps to reduce polarity of compounds contained in the extract while equally rendering them more volatile such that they become detectable on GC-MS. These enabled the identification of major compounds in both extracts. These two plants investigated in this study have evidently been proven to possess antimicrobial activity and have been used extensively to manage infectious diseases [34], but much of this has been attributed to its essential oils [5] with little attention on the extracts. In this study, the extracts have also demonstrated appreciable antimicrobial activity, and this is advantageous, since the plants are consumed principally in crude form. Antimicrobial activity was displayed by both plant extracts and these results corroborate with some findings in which these plant extracts showed antimicrobial activity against some pathogens [35]. However, microbial resistance arises due to quorum-sensing-mediated traits of pathogens which determines the severity of infections as well. The assays that involved quorum-sensing-mediated processes in bacteria are usually focused biofilm formation, violacein pigment production, and swimming and swarming motilities. Prior to these assays, MIC values are determined and they are performed at MIC and sub-MIC concentrations. Thus, working at sub-MIC concentrations, the hypothesis of bactericidal effect of extracts that occurs at high concentrations is eliminated, giving way for QS investigation.

In order to protect themselves during adverse conditions such as immunological defense systems of host, antibiotics, and starvation, many fungi and bacteria cells constitute self-organized and three-dimensional communities in which they will live. These constituted and protected communities are called biofilms and they contribute to the severity of chronic infections as well as the persistence of resistance to drugs and antibiotics [6,36]. Therefore, most antimicrobial agents treat symptoms of planktonic bacteria, and disease will resurface, due to the bacteria which were within biofilms and which will break out when the threat from the antibiotic is over. The potential of these plant extracts to inhibit biofilm formation by test bacteria is a very desirable effect. This can subsequently eliminate bacteria resistance or reduce the severity of microbial infections.

Violacein (violet color) production is one of the quorum-sensing-mediated traits of pathogenic bacteria *C. violaceum* that has received maximum attention and has the advantage that it is easily measurable and quantifiable. Violacein is a biomolecule that has antimicrobial activity against other microorganisms and also improves the *C. violaceum's* ability to cope with environmental stress [37,38]. *C. violaceum* has been highly applied in evaluating the antiinfective potential of many natural and synthetic products, and those with violacein inhibition find application potential alternatives to conventional bactericidal antibiotics.

The bacterial communities that are formed on surfaces constitute a serious health threat and potent contamination. The colonization of various surfaces by bacteria through bacterial motility can be reduced or limited by the use of natural substances and compounds which need to be identified and evaluated for their applications to impede bacterial colonizations [39]. Before forming biofilms, bacteria move by swimming and swarming and colonize surfaces, and this step is believed to be an initial step for the formation of the biofilm in flagellated *P. aeruginosa* PA01. This bacterium can use its flagella to swim towards surfaces followed by swarming and twitching motilities, which enables it to colonize surfaces and establish biofilm communities.

Cholinesterase inhibitors are a group of medicines that block the normal breakdown of acetylcholine. Deterioration of the central nervous system and Alzheimer's disease (AD) result from cholinergic deficiency and are associated with dementia. As a remedy to this situation, various cholinesterase inhibitors such as natural compounds and extracts and synthetic analogues and their hybrids are being employed [40]. Natural medicines are gaining ground because of their low toxicities and few side effects compared to synthetic ones. The enzyme-mediated browning reaction in harvested fruits is usually initiated by the copper-containing oxidase enzyme called tyrosinase. It is an enzyme with multiple functions and is responsible for the undesirable over-pigmentation of human skin in a process that is similar to that which takes place during the browning of fruits. The first two steps in mammalian melanogenesis are catalyzed by this enzyme, and hence the search for potent tyrosinase inhibitors, especially those that can be used safely in cosmetics and foods, is an attraction for researchers [41]. Many bacteria produce the nickel-dependent urease (ureolytic bacteria) enzyme, which is capable of hydrolyzing urea to produce ammonia with the emission of carbon dioxide as well. The emission of ammonia from agriculture is usually associated with these bacteria, and it also causes a number of infectious diseases [42]. Inhibiting urease is beneficial to human health and some notorious bacteria depend on urease-mediated processes for survival. Although in the enzyme inhibitory assays no extract had an inhibition close to that of the standards, these results are moderate because the extracts were tested at the same concentrations as the pure standard compounds.

Many of the compounds identified in these plants have previously been described in some medicinal plant extracts. 1-monolinoleoylglycerol, which is found in *M. myristica* has been described in the extract of *Datura stramonium* and *Salvadora persica* and the extracts of these plants have shown antimicrobial and anticancer activities [43,44]. Certain sugars were detected in both *M. myristica* and *X. aethiopica* plant extracts notably sucrose, glucose, glycerol, and fructose. These sugars were shown to possess antibacterial activities [45,46]. Ethyl-3,4,5-trimethoxybenzoate detected in *M. myristica* and other derivatives of 3,4,5-trimethoxybenzoic acid have been identified in medicinal plant *Mitracarpus scaber* and have demonstrated antimicrobial properties [47], and *Anarcadium occidentale* (cashew gum) also has been synthesized and evaluated for antioxidant and enzyme inhibition [48]. 6,9,12-octadecatrienoic acid phenylmethyl ester was found in the extract of *M. myristica*, and this compound was detected in *Croton bonplandianum* extract which possesses anti-inflammatory and enzyme inhibitory potentials [49]. 3-hydroxyspirost-8-ene-11-one, which was detected in *M. myristica*, has been previously described in *Artemisia annua*, and this plant extract studied showed antioxidant, antimicrobial, and anti-inflammatory activities [50]. Ursodeoxycholic acid was identified in *M. myristica,* and this compound has been synthesized from a natural source [51]. Hexadecanoic acid contained in *M. myristica* has been described in many plants and this compound has good medicinal properties including anti-inflammatory, antioxidant, and enzyme inhibitory activities including anticholinesterase

activity [52–54]. The compound 2-Hydroxy-4-methylbenzoic acid present in *X. aethiopica* has been detected previously in *Hemidesmus indicus* and mangrove fungus and has displayed anti-inflammatory, antipyretic, and antioxidant activities [55,56]. In the extract of *X. aethiopica*, abietic acid was identified, and this compound has been detected in *Isodon wightii*, which displayed antibacterial, antioxidant, and anticholinesterase activities [57]. A major component of eucalyptus called eucalyptol or 1,8-cineole identified in *X, aethiopica* extract has been detected in many species of eucalyptus, whose extracts possess antimicrobial and antiviral activities [58,59]. The volatile compound 3-carene was detected in the extract of *X. aethiopica* and has been described as an antimicrobial compound [60]. From these analyses of previous studies, it can be understood that the antimicrobial, antibiofilm, and enzyme inhibitory activities of these extracts of *M. myristica* and *X. aethiopica* can be attributed to these chemical compounds that they contain.

#### **5. Conclusions**

The investigation of bioactivities of food extracts and food-derived substances is attracting much attention because of their considerable and known safety. These food substances are exploited for their medicinal properties besides their nutritive values especially to combat food-borne infections resulting from various food pathogens. The emergence and spread of drug-resistant strains nowadays threatens the use of conventional antibiotics in order to treat bacterial infections and diseases. New strategies such as biofilm inhibition and quorum-sensing disruption are therefore necessary to overcome persistent infections especially those that involve resistant biofilm formation by the pathogens. Both food spices *X. aethiopica* and *M. myristica* have shown biofilm inhibitory potentials and also violacein inhibition and swimming/swarming motilities inhibitions. These traits are quorum-sensing-mediated processes which help to increase the resistance of bacteria and virulence of their infections. Both extracts displayed low to moderate anticholinesterase activities and tyrosinase and urease inhibitory potentials, which is interesting for these natural food spices. As an added value to the antimicrobial properties of these two natural food spices, they can be desirable as enzyme inhibitors which may present no or milder side effects than conventional synthetic drugs used for this purpose. The bioactivities of these extracts opens a space for the search of the bioactive molecules contained in them.

**Author Contributions:** Conceptualization, A.N.T. and O.C.; methodology, A.N.T., S.K., O.C., R.M.D., M.O. and M.E.D.; investigation, A.N.T., S.K. and O.C.; writing—original draft preparation, A.N.T. and O.C.; writing—review and editing, A.N.T., O.C. and R.M.D.; supervision, O.C., R.M.D., M.O. and M.E.D. Project administration, R.M.D.; funding acquisition, A.N.T. and R.M.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by 'Agence Universitaire de la Francophonie' (AUF) and the Romanian Government through Eugen Ionescu Postdoctoral Mobility Program 2020.

**Acknowledgments:** The authors are grateful to the University of 'Dunarea de Jos' of Galati, Romania and Mugla Sitki Kocman University, Turkey for collaborative material support and for providing a conducive platform for this research.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Oregano Phytocomplex Induces Programmed Cell Death in Melanoma Lines via Mitochondria and DNA Damage**

#### **Valentina Nanni 1,**†**, Gabriele Di Marco 1, Gianni Sacchetti 2, Antonella Canini <sup>1</sup> and Angelo Gismondi 1,\***


Received: 16 September 2020; Accepted: 15 October 2020; Published: 17 October 2020

**Abstract:** Plant secondary metabolites possess chemopreventive and antineoplastic properties, but the lack of information about their exact mechanism of action in mammalian cells hinders the translation of these compounds in suitable therapies. In light of this, firstly, *Origanum vulgare* L. hydroalcoholic extract was chemically characterized by spectrophotometric and chromatographic analyses; then, the molecular bases underlying its antitumor activity on B16-F10 and A375 melanoma cells were investigated. Oregano extract induced oxidative stress and inhibited melanogenesis and tumor cell proliferation, triggering programmed cell death pathways (both apoptosis and necroptosis) through mitochondria and DNA damage. By contrast, oregano extract was safe on healthy tissues, revealing no cytotoxicity and mutagenicity on C2C12 myoblasts, considered as non-tumor proliferating cell model system, and on *Salmonella* strains, by the Ames test. All these data provide scientific evidence about the potential application of this food plant as an anticancer agent in in vivo studies and clinical trials.

**Keywords:** oxidative stress; necroptosis; plant extract; secondary metabolite; γH2AX; copper

#### **1. Introduction**

More than one-third of all pharmaceutical molecules approved by the Food and Drug Administration and by the European Medicines Agency are natural compounds, or their derivatives, and about one-quarter of them specifically originate from plants [1–4]. In detail, over 60% of the anticancer drugs are phytochemicals, such as alkaloids and polyphenols, confirming that the plant kingdom is a valuable source of chemopreventive and chemotherapeutic agents [5–7].

Overall, scientific data have documented that this type of metabolites exerts an inhibitory effect on a broad range of mammalian tumor cell lines in in vitro and in vivo systems [8,9]. According to the literature, the main mammalian cellular and molecular mechanisms influenced by plant molecules are those that involve the following targets: nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB); protein tyrosine kinases (PTKs); target mitogen-activated protein kinases (MAPKs); cyclooxygenase (COX-2); cyclin-dependent kinases (Cdks); phosphoinositide 3-kinase (PI3K) interactors; activator protein 1 (AP1); and cytoskeleton components [10–22].

Among all biological properties, the free radical scavenging activity seems to be the most validated device employed by phytochemicals to inhibit tumor cell growth, counteracting reactive oxygen species (ROS) production and limiting protein, lipid, and DNA oxidation [23]. Nevertheless, a growing body of evidence brought this opinion into question, arguing that the antiradical activity of the natural compounds has a potential role in chemoprevention, but it cannot fully explain the relative antitumor effect [24,25]. In addition, it is important to underline that a wide number of plant secondary metabolites have also shown unexpected pro-oxidant consequences, including DNA damage and apoptosis, especially at high concentrations and in the presence of transition metal ions [26–29].

*Origanum vulgare* L., also known as oregano, is a Mediterranean plant species belonging to the Lamiaceae family which, nowadays, represents one of the most used culinary herbs. However, the application of oregano in several ethnobotanical practices, including folk medicine, dates to ancient times. Regarding its phytotherapic effect, various investigations have been performed, documenting that oregano essential oil possesses antimicrobial, antiviral, antifungal, antioxidant, anti-inflammatory, digestive, expectorant, neuroprotective, antispasmodic, and antidiabetic properties, simultaneously. Moreover, some literature works have also associated a strong anticancer activity to such type of plant extract. For these reasons, *O. vulgare* is capturing greatly the attention of the food, cosmetic, and pharmaceutical industries [30–35].

According to all this evidence, the present research aimed at investigating the molecular mechanism underlying the antineoplastic effect of *O. vulgare* L. ssp. *hirtum* phytocomplex against murine (B16–F10) and human (A375) melanoma cells.

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

#### *2.1. Plant Material*

*Origanum vulgare* L. plants were collected at the *Vatopedi Holy Monastery* on Mount Athos (Greece), in the summer of 2018. The plant material was transferred to the Botanical Garden of Rome "Tor Vergata", where its taxonomic identity was confirmed by Prof. Antonella Canini and Prof. Angelo Gismondi, based on morphological features. A part of the sample was deposited in the *Herbarium* of the Botanical Garden (voucher n. 127C), while the remaining portion was dried out (for 7 days at 37 ◦C) and used for the present research. In particular, the whole dried plants were powdered in liquid nitrogen, resuspended in 50% ethyl alcohol (200 mg/mL) and incubated, in agitation, for 24 h in the dark. After centrifugation for 20 min at 11,000 g, the supernatant was filtered (0.22 μm), completely desiccated at 30 ◦C by a vacuum drying system (Concentrator Plus, Eppendorf, Hamburg, Germany), and stored at −80 ◦C.

#### *2.2. Total Phenol and Flavonoid Content*

Hydroalcoholic oregano extract (HCOE) was solubilized in 50% ethyl alcohol at the final concentration of 200 mg/mL. The phenolic content in HCOE was measured according to the Folin–Ciocalteu modified method, as described in Impei et al. [36]. Results were reported as μg of gallic acid equivalents per gram of dried plant material (μg GAE/g DMW), applying a gallic acid calibration curve (0–30 mg/L). The amount of flavonoids in OE was assessed by the aluminium chloride colorimetric method [37]. Data were reported as μg of quercetin equivalents per gram of dried plant material (μg QE/g DMW), using a quercetin calibration curve (0–50 mg/L).

#### *2.3. High-Performance Liquid Chromatography-Diode Array Detector (HPLC-DAD) and Gas Chromatography-Mass Spectrometry (GC-MS) Analyses*

HCOE was characterized by an HPLC system (Shimadzu, Kyoto 604-8511, Japan) associated with an SPD-M20A diode array detector (DAD, Shimadzu, Kyoto 604-8511, Japan) and a Phenomenex Luna C18(2) (3 μm × 4.6 mm × 150 mm) column. A flow of 0.95 mL/min was applied, using formic acid 1% (buffer A) and methanol (buffer B). The following elution gradient was adopted: t0 min (A 85%, B 15%); t20 min (A 65%, B 35%); t55 min (A 10%, B 90%); t68 min (A 85%, B 15%); t70 min (A 85%, B 15%). UV–visible absorption spectra at 280 and 340 nm were monitored. Plant metabolites were identified and quantified comparing their retention time, absorbance spectrum, and chromatographic peak area

with those of relative pure standards (Sigma-Aldrich) at different concentrations. The amount of each detected molecule was reported as μg per 100 mg of dried plant material (μg/100 mg DMW).

GC-MS analysis was performed exactly as described in Nanni et al. [29]. In particular, to perform this investigation, HCOE was solubilized in 100% methanol at the final concentration of 200 mg/mL and then injected in the instrument.

#### *2.4. Cell Cultures and Plant Treatments*

Murine melanoma cells (B16–F10), human melanoma cells (A375), and murine myoblasts (C2C12) were cultured under standard conditions [38,39] in Dulbecco's Modified Eagle's Medium (DMEM). For cell experiments, HCOE was solubilized in sterilized PBS 1X, at the concentration of 1200 mg of dried plant material equivalent per mL, and added directly to the cell medium at specific doses. In particular, treatments were performed by exposing cells, for 4, 24 or 48 h, at 2, 4, 6, 8, or 10 mgs of dried plant material equivalent per mL of culture medium. Control cells (CNT) were treated only with PBS, at the highest volume of treatment, to check the influence of this non-toxic solvent on cells.

#### *2.5. Cell Proliferation, Selectivity Index, and Cell Cycle Analysis*

Cell viability was evaluated by the MTT kit (Sigma-Aldrich Merck, Darmstadt, Germany), as reported in the relative guidelines. Results were reported as percentage variation compared to the control (CNT), which was considered as a unit (100%). Plant extract cytotoxicity was measured using the Trypan Blue (1%, *w*/*v*) exclusion test and counting dead cells by a Neubauer-modified chamber. The selectivity index (SI) of HCOE on tumor and non-tumor cells was measured according to the following formula: SI = IC50 non-tumor cell line/IC50 cancer cell line (considering that IC50 represented the concentration at which 50% of cell proliferation was inhibited). Cell cycle analysis was performed by a FACSCalibur instrument (Beckton and Dickinson, Le Pont-de-Claix, France) associated to CellQuest software, counting 10,000 events per sample and using the protocol documented in Gismondi et al. [40]. Cytofluorimetric data were shown as a percentage of cells in G0/G1, S, G2/M, and sub-G1 phase. Other treatments were performed by necrostatin-1 (NEC-1; 20 μM, 48 h), Z-VAD-FMK (Z-VAD; 20 μM, 48 h), and Paclitaxel (TAX; 20 nM, 8 h) (Sigma-Aldrich).

#### *2.6. Mutagen and Mutagen-Protective Activity*

An Ames test was carried out on *Salmonella typhimurium* strains (TA97a, TA98, TA100 and TA1535) in order to evaluate the mutagen and mutagen-protective activity of HCOE. The assays were carried out as widely reported by Rossi et al. [41,42]. In detail, the mutagenic activity was determined by counting *Salmonella* colonies (Colony Counter 560; Suntex Instruments Company Ltd., New Taipei City, Taiwan) in plates treated with different concentrations of oregano extract in the presence and absence of S9 mix metabolic activation. The results were considered positive (potential mutagen) if the amount of revertant colonies was, at least, double that of the negative control. To determine the potential mutagen protection capacity of HCOE (concentration range: 0.01–0.1 mg/plate), bacteria were exposed to mutagenic agents, with or without S9 mix, and exposed or not to different concentrations of HCOE. Used mutagens were 2-nitrofluorene (2 μg/plate; Sigma-Aldrich) for TA97a, TA98, and TA1535; NaN3 (2 μg/plate; Sigma-Aldrich) for TA100 without S9; and 2-aminoanthracene (2 μg/plate; Sigma-Aldrich) for all *Salmonella* strains cultured with S9 mix. Data were expressed as CFU/plate. The inhibition rate (IR) of HCOE for mutagenic induction was measured according to the formula: IR (%) = (A−B) × 100/A (where A and B represent the number of revertants in positive controls or in plates with HCOE, without spontaneous colonies, respectively). Negative controls, represented by dimethylsulfoxide (DMSO) treated strains, were performed to evaluate the background of spontaneous revertants.

#### *2.7. Protein Analysis*

Cells were lysed in High Salt Buffer (2 mM CaCl2, 350 mM KCl, 50 mM Tris HCl pH 7.4, 1 mM MgCl2) containing 1% protease inhibitor cocktail and 1% NP40. Proteins whose concentration

was measured by the Bradford method [43] were separated by SDS-PAGE and transferred onto nitrocellulose membrane. Protein signals were detected by a chemiluminescent kit (Luminol Reagent; Santa Cruz Biotechnology, Dallas, TX, USA) and a VersaDoc Imaging System associated with Quantity One software (Bio-Rad). After normalization with GAPDH, the results were indicated as percentage change compared to the CNT, which was considered as a unit (100%). The antibodies (Santa Cruz Biotechnology) used for Western blotting analyses were as follows: mouse monoclonal GAPDH; mouse monoclonal microphthalmia-associated transcription factor (Mitf); mouse monoclonal p53; mouse monoclonal Parp-1; mouse monoclonal caspase-3 (Casp-3); rabbit polyclonal Bcl-2; rabbit polyclonal Bax; mouse monoclonal control outer mitochondrial membrane protein TOMM20; mouse monoclonal cytochrome c (Cycs); peroxidase-conjugated rabbit, and mouse secondary antibodies. Staurosporine (STS; 2 μM, 4 h) was used as an inducer for apoptosis (Sigma-Aldrich).

#### *2.8. Real-Time-PCR (RT-PCR) Assay*

Total RNA was extracted by a Pure Link RNA Mini Kit (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA). RNA concentration and purity were evaluated with a Nanodrop ND1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). For cDNA synthesis, 2.5 μg of RNA were incubated for 2 min at 65 ◦C with 0.4 mM of each dNTP (Euroclone, Milan, Italy). Then, 40 units of RNA inhibitor (Promega, Madison, WI, USA), 0.5 μg random hexamer primers (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA), 200 units of Moloney murine leukemia virus reverse transcriptase (Promega), 1× enzyme buffer, and 10 mM dithiothreitol were added to reach the final volume of 25 μL. The mix was incubated for 90 min at 37 ◦C. RT-PCR reactions were carried out in 20 μL of volume composed of 10 ng of cDNA, 5 μM of each primer, and 50% SYBR green (Kapa SYBR Fast qPCR kit; Kapa Biosystems, Roche, Wilmington, MA, USA, Country). cDNA amplification was carried out in an IQ5 thermocycler (Bio-Rad) with the following method: (i) initial denaturation at 95 ◦C, 4 min; (ii) 45 cycles of denaturation at 95 ◦C for 20 s (sec), primer annealing at 60 ◦C (for all genes) for 30 s, and extension at 72 ◦C for 30 s; and (iii) production of disassociation curve, from 50 to 90 ◦C (rate: 0.5 ◦C every 5 s), for the verification of the results. The 2−ΔΔCt formula was used to measure mRNA concentration for each gene: in detail, the threshold cycle (Ct) of the target gene monitored in the treated sample was normalized for the internal reference gene (β-actin, ACTB; ΔCt) and for the respective value of the control sample (ΔΔCt), which was considered as a unit (100%). Supplementary Materials Table S1 reports the list of primers used in this work: microphthalmia-associated transcription factor (MITF), tyrosinase-related protein 1 (TYRP1), tyrosinase (TYR), P21, P27, P53, cyclin-dependent kinase 1 (CDK1), cyclin B1 (CCNB1), and β-actin (ACTB) [44–48].

#### *2.9. Reactive Species Level and Mitochondrial Mass and Membrane Potential Measurement*

Intracellular reactive oxygen (ROS) and nitrogen (RNS) species, mitochondrial mass, and mitochondrial transmembrane potential were measured by 2 ,7 -dichlorodihydrofluorescein diacetate (DCFH-DA; green signal; 10 μM, 15 min), 4-amino-5-methylamino-2 ,7 -difluorofluorescein diacetate (DAF-FM DA; green signal; 2.5 μM, 30 min), MitoTracker Green (MTG; green signal; 250 nM, 30 min), and MitoTracker Red CMX ROS (MTR; red signal; 250 nM, 30 min) fluorescent assays (Sigma-Aldrich), respectively. The analyses were performed using the protocol described in Gismondi et al. [49] (FACSCalibur instrument; filters: FL-1<sup>+</sup> for green; FL-2<sup>+</sup> for red) and counting 10,000 cytofluorimentric events per sample. Negative controls were carried out treating cells with PBS 1X, whereas positive controls were produced incubating cells with hydrogen peroxide (H2O2; 5 mM; for DCFH-DA test), S-nitrosoglutathione (GSNO; 0.5 mM: for DAF-FM DA test), and carbonyl cyanide m-chlorophenyl hydrazone (CCCP; 10 μM; for mitochondrial tests) for 4 h before the exposure to the appropriate probe. All results were reported as a percentage variation of cell fluorescence compared to the CNT sample, which was considered as a unit (100%). Changes in the mitochondrial membrane potential were reported as MTR/MTG ratio, as suggested by Pendergrass et al. [50].

#### *2.10. Immunofluorescence Microscopy*

For γH2AX and 53BP1 foci detection, cells (grown on slides) were fixed in 4% paraformaldehyde for 15 min, permeabilized with 0.4% Triton X-100 in PBS for 10 min, blocked in PBS blocking solution (10% FBS, 0.1% Triton X-100) for 3 h, and incubated for 2 h with primary antibodies (mouse monoclonal γH2AX Ser-139; rabbit polyclonal 53BP1; Merck Millipore). Then, samples were exposed for 1 h to the respective secondary antibodies (goat anti-mouse IgG labeled with Alexa Fluor 488 and goat anti-rabbit IgG labeled with Alexa Fluor 594; Invitrogen, Eugene, OR, USA). Nuclei were stained with 0.1 mg/mL of DAPI for 1 min. Images were acquired by a Leica DMR microscope (Leica Microsystems, Wetzlar, Germany) equipped with a Leica DFC 350 FX digital camera, EBQ 100 isolated fluorescent lamp (Leistungselektronik Jena GmbH, Jena, Germany), UV/FITC/TRITC filters, and 40X and 100X objectives. All images were elaborated by Leica Qwin Pro image analysis software and captured at the same instrument settings and exposure times in order to ensure a correct comparison. For foci counting, 500 cells for each experimental condition were analyzed by ImageJ. Control treatments were performed by etoposide (ETO; 500 nM, 8 h) and triethylenetetramine (TETA; 50 μM, 48 h) (Sigma-Aldrich).

#### *2.11. Statistical Analysis*

Results were reported as mean value ± standard deviation (SD) of measurements obtained by independent experiments (*n* ≥ 3). Statistical significance was evaluated by one-way ANOVA test (Microsoft Excel software) vs. the respective control; a *p*-value < 0.05 was considered significant (\* < 0.05; \*\* < 0.01; \*\*\* < 0.001).

#### **3. Results**

#### *3.1. Chemical Characterization of the O. vulgare L. Phytocomplex*

For a preliminary typization of the plant sample, the concentration of simple phenols and flavonoids in the *O. vulgare* L. extract was measured by spectrophotometric analyses. The amount of total phenols in HCOE was equal to 107.50 ± 10.81 μg GAE/g DMW, while flavonoids were 230.79 ± 13.97 μg QE/g DMW. Then, in order to identify the main plant metabolites underlying the bioactivity of the oregano extract, two different chromatographic approaches were applied to characterize the biochemical profile of this natural matrix. As reported in Table 1, 13 compounds were identified and quantified in HCOE by the HPLC-DAD technique (Supplementary Materials Figure S1).

**Table 1.** High-performance liquid chromatography-diode array detector (HPLC-DAD) profiles of oregano extract. Plant molecules and their concentration, detected in hydroalcoholic oregano extract (HCOE) by HPLC-DAD analysis, are reported. Results were indicated as μg of metabolite per 100 mg of dried material (μg/100 mg DMW) and represent the mean ± SD of six independent experiments.


The most abundant molecule was chrysin (8.47 ± 0.06 μg/100 mg DMW), followed by quercetin-3-*o*-arabinoside (2.37 ± 0.04 μg/100 mg DMW) and rutin (2.15 ± 0.05 μg/100 mg DMW). Moreover, the phytocomplex extracted by *O. vulgare* L. samples was characterized by GC-MS analysis; in total, 45 secondary metabolites were detected and subjected to relative quantitation (Table 2).

**Table 2.** GC-MS profile of oregano extract. Plant metabolites and their relative abundance, detected in HCOE by GC-MS analysis, are reported. The relative abundance of each molecule was indicated as a percentage value with respect to the total mixture (100%). Values represented the mean of three independent experiments. The SD for each measurement was always <5% of the respective molecule peak area.



**Table 2.** *Cont.*

The most abundant molecules were carvacrol (34.82%), thymol (16.61%), and linolenic acid methyl ester (7.96%).

#### *3.2. O. vulgare L. Extract Reduces B16-F10 Cell Growth Not A*ff*ecting C2C12 Cell Viability*

The biological effect of *O. vulgare* L. hydroalcoholic extract on the proliferation of B16-F10 cells, a murine melanoma line characterized by high aggressiveness and drug resistance, was investigated, by MTT assay, after exposure for 24 and 48 h with different concentrations of plant phytocomplex (0.1–10 mg/mL). Simultaneously, to check the safety of HCOE on non-tumor cells, C2C12 myoblasts were exposed to similar treatments. HCOE did not affect C2C12 cell growth after 24 h of incubation (Figure 1A), whereas a slight decrease of myoblast viability was observed after 48 h of treatment, reaching 27% at the highest concentration of extract (Figure 1B). By contrast, the oregano sample significantly decreased B16–F10 proliferation: in particular, after 24 and 48 h of incubation, 10 mg/mL of HCOE caused a reduction of melanoma cell viability of 73.42% (Figure 1C) and 84.11% (Figure 1D), respectively. According to these results, IC50 values for C2C12 cells treated with HCOE, for 24 and 48 h, were estimated to be 55.44 and 14.28 mg/mL, respectively. For B16-F10, these values were 7.23 and 4.72 mg/mL, in that order. Consequently, for HCOE, the selectivity index was 7.66 after 24 h of treatment and 3.03 at 48 h.

The cytotoxicity of the oregano extract was evaluated by the Trypan Blue exclusion test. C2C12 and B16-F10 proliferation curves were generated, counting living cells after treatment with HCOE (2–10 mg/mL) or PBS (CNT) for 24 and 48 h. Simultaneously, dead cells were also counted in each sample. As expected, C2C12 cell growth was slightly affected by HCOE (Figure 1E). The strongest cytotoxic effect (21.25%) was observed after 48 h of incubation with the highest dose of HCOE. However, in all cases, the percentage of dead cells was always lower than 22% (Table 3). For B16-F10, cell proliferation significantly decreased in a dose-dependent manner (Figure 1F), and a remarkable percentage of dead cells was achieved at 10 mg/mL of HCOE (42% and 44.75% after 24 and 48 h of incubation, respectively) (Table 3).

To confirm the previous data, the B16–F10 cell cycle was analyzed after exposure to HCOE for 48 h. As shown in Figure 1G, low doses of plant extract induced a cell accumulation in the G1/G0 phase, whereas 10 mg/mL of HCOE determined an accumulation of cells in the G2/M phase equal to 45.24%.

**Figure 1.** C2C12 and B16-F10 cell growth. C2C12 (**A**,**B**) and B16-F10 (**C**,**D**) cell growth was measured after 24 h (**A**,**C**) and 48 h (**B**,**D**) of treatments with Phosphate Buffered Saline (PBS), as control, and different concentrations of *O. vulgare* plant extract (HCOE). Results expressed as percentage with respect to PBS represent the mean ± SD of four independent experiments (\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control). Proliferation curves of C2C12 (**E**) and B16-F10 (**F**) cells were generated counting, by a Neubauer modified chamber, the amount of alive cells after staining with Trypan Blue at 0, 24, and 48 h of treatments with HCOE. Results are indicated as the mean ± SD of four independent experiments. (\* *p* < 0.05 vs. control) (**G**) Cell cycle analysis of B16-F10 after treatment, for 48 h, with 2, 4, 6, 8, and 10 mg/mL HCOE is shown. For each sample, the percentage amount of cells in every cycle phase (G0/G1, S, and G2/M) was measured by cytofluorimetric analysis. Results are indicated as mean ± SD of four independent experiments (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control).


**Table 3.** Cytotoxicity analysis. The percentages of C2C12, B16-F10, and A375 dead cells identified by the Trypan Blue test after 24 and 48 h of HCOE treatment are reported. Data are reported as mean ± SD of four independent replicates. (\* *p* < 0.05 vs. control).

#### *3.3. Oregano Treatment Shows Antiproliferative Activity Also on A375 Human Melanoma Cells*

The antiproliferative properties of the oregano extract on B16–F10 murine melanoma cells encouraged us to continue our research, testing if the same plant preparation could also exhibit similar effects on the A375 human melanoma line. The data obtained by the MTT assay are reported in Figure 2A,B. In detail, HCOE treatments at selected doses (2, 4, 6, 8, 10 mg/mL) decreased A375 cell viability, respectively, by 20.73%, 26.42%, 39.02%, 38.21%, 58.40%, and 3.86% after 24 h of incubation, and by 4.84%, 17.95%, 39.43%, 54.77%, and 80.40% after 48 h compared to the corresponding controls. In addition, in this case, the selectivity index was estimated. IC50 values for A375 cells treated with HCOE, for 24 and 48 h, were 9.14 and 7.08 mg/mL, respectively. Consequently, the SI with respect to C2C12 cells was 6.07 after 24 h of treatment and 2.02 at 48 h.

The Trypan Blue exclusion test confirmed these results, evidencing reduced proliferation curves (Figure 2C) and significant cytotoxicity levels on human melanoma cells, especially after exposure to 10 mg/mL HCOE (47.14% and 55.13% of dead cells after 24 and 48 h of incubation, respectively) (Table 3).

Taking into account the great antiproliferative activity exerted by HCOE, the A375 cell cycle was analyzed after treatment with 10 mg/mL of oregano extract for 48 h. As reported in Figure 2D, a significant increase of cells in the G2/M phase (19.55%) was detected with respect to the control (CNT). In this context, the (TAX), a well-known plant drug able to induce G2/M phase arrest [51,52], was also used as positive control.

This evidence was consistent with the RT-PCR experiments performed to measure cyclin-dependent kinase 1 (CDK1), cyclin B1 (CCNB1), and P21 and P27 mRNA levels (Figure 2E). Indeed, the expression of CCNB1 and CDK1 genes, which are key factors in the transition from the G2 to the M phase [53], appeared reduced after 48 h of incubation with oregano extract. At the same time, P21 and P27 transcripts, which are CDK1/Cyclin B1 inhibitors [54–56], increased in the presence of HCOE. Similar results were obtained after exposure to TAX, although the P21 mRNA level remained unaltered compared to the control, as documented by the literature [57].

**Figure 2.** A375 cell growth and cycle. A375 cell viability was measured after 24 h (**A**) and 48 h (**B**) of treatment with 2, 4, 6, 8, and 10 mg/mL HCOE. Results, expressed as percentage with respect to the PBS control (CNT; considered as unit, 100%), represent the mean ± SD of four independent experiments (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control). (**C**) Proliferation curves of A375 cells were generated by counting, with a Neubauer modified chamber, the amount of alive cells after staining with Trypan Blue at 0, 24 and 48 h of treatment with HCOE. Results were indicated as mean ± SD of three independent experiments (\* *p* < 0.05 vs. control). (**D**) A375 cell cycle analysis after 48 h of treatment with paclitaxel (TAX) or 10 mg/mL HCOE is shown. For each sample, the percentage amount of cells in every cycle phase (G0/G1, S and G2/M) was measured by cytofluorimetric analysis. Results are expressed as mean ± SD of three independent experiments (\*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control). (**E**) P21, P27, CCNB1, and CDK1 mRNA levels, measured by RT-PCR, in A375 cells treated for 48 h with TAX or 10 mg/mL HCOE are reported. Gene expression, calculated as mRNA amount after normalization for ACTB mRNA (2-ΔΔCt), is reported as percentage with respect to the PBS control (CNT; considered as unit, 100%). Data represent the mean ± SD of three independent measurements (\* *p* < 0.05; \*\* *p* < 0.01 vs. control).

#### *3.4. Oregano Extract Has Neither Mutagenic Nor Mutagen-Protective E*ff*ects*

Before investigating, in depth, the antitumor effect exerted by HCOE on A375 cells, the potential mutagenic and mutagen-protective activities of the oregano extract were analyzed in order to confirm HCOE safety on non-tumor living model systems. For the assessment of both properties, an Ames test was carried out with *Salmonella typhimurium* strains TA97a, TA98, TA100, and TA1535 in the presence or absence of the metabolic activator S9 mix. As indicated in Supplementary Materials Table S2, HCOE did not show any toxicological evidence. Indeed, at all tested doses, the t/c values, namely the ratio between the number of colonies of *Salmonella* strains grown in the presence of oregano extract (t) and those on the control medium (c), were never higher or equal to 2 and never presented a dose–response trend [58].

As concerns the evaluation of the mutagen-protection effect mediated by HCOE, a properly modified Ames test [40] was carried out. The plant extract did not exhibit any protective activity against well-known mutagen compounds, as indicated by the inhibition rate (IR) percentages reported in Supplementary Materials Table S3. Indeed, despite the presence of IR-positive values, the HCOE dose–response effect was not observed with significant values.

#### *3.5. Oregano Phytocomplex Impairs MITF Pathway and Accumulates Reactive Species in A375 Cells*

The molecular mechanism underlying oregano antineoplastic activity was clarified by using A375 cells as a model system. In particular, according to previous data, 10 mg/mL of HCOE (thenceforth HCOE10) was selected as a treatment dose, showing strong antiproliferative results on tumor cells and minimal effects on non-tumor ones.

The expression of the main genes involved in the MITF pathway, which is a cell signal that plays a crucial role in melanoma progression [59], was analyzed by RT-PCR (Figure 3A). In A375 cells, HCOE10 treatment, for 48 h, drastically decreased MITF, TYR, and TYRP1 mRNA levels of 16.27%, 27.93%, and 27.74%, respectively, compared to the control. This result was also corroborated by Western blotting analysis of Mitf protein content (Figure 3B); densitometric quantitation evidenced that oregano extract determined a reduction of 38.3% of Mitf protein, with respect to the control (Figure 3C).

**Figure 3.** Analysis of mouse monoclonal microphthalmia-associated transcription factor (MITF) signaling and cell redox state. (**A**) *MITF*, tyrosinase (*TYR*)*,* and tyrosinase-related protein 1 (*TYRP1*) mRNA levels, measured by RT-PCR, in A375 cells treated for 48 h with 10 mg/mL HCOE are reported. Gene expression calculated as mRNA amount after normalization for β-actin (ACTB) mRNA (2-ΔΔCt) is reported as percentage with respect to the PBS control (considered as unit, 100%). Data represent the mean ± SD of four independent measurements (\*\* *p* < 0.01 vs. control). (**B**) Representative Western blotting membrane of Mitf and Gapdh protein levels is shown. (**C**) Quantitation of MITF protein in A375 cells treated for 48 h with HCOE 10 mg/mL is reported. The results obtained by the ratio between Mitf and Gapdh (used as internal loading control) are indicated as percentage values with respect to PBS control (considered as unit, 100%). Data indicate the mean ± SD of three independent experiments (\* *p* < 0.05 vs. control). (**D**) Intracellular reactive oxygen (ROS) and nitrogen (RNS) species levels were quantified in A375 cells, treated with 10 mg/mL of HCOE for 4, 24, and 48 h, by DCFH-DA and DAF-FM DA fluorescent assays, respectively. Radical species concentration is reported as percentage compared to PBS control (CNT). Hydrogen peroxide (H2O2) and S-Nitrosoglutathione (GSNO) treatments were performed as positive controls for ROS and RNS analysis, respectively. Results are expressed as mean ± SD of three independent measurements (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control).

As the inhibition of the MITF pathway has been associated to reactive species burst [29], the influence of the plant extract (that is HCOE10 treatment for 4, 24 and 48 h) on ROS and RNS levels was monitored in A375 by DCFH-DA and DAF-FM DA assays, respectively (Figure 4D). Oregano treatment for 4 h did not influence cell redox state, whereas a prolonged exposure caused a significant increase of reactive species: after 24 and 48 h, respectively, +92.82% and +32.51% for ROS and +37.42% and +22.98% for RNS, compared to control cells. Positive controls, using inducers of ROS (i.e., H2O2) and RNS (i.e., GSNO), were carried.

*Foods* **2020**, *9*, 1486

**Figure 4.** Cell death analysis and mitochondrial damage evaluation. A375 cell proliferation was analyzed after treatment, for 48 h, with 10 mg/mL HCOE, 10 mg/mL HCOE + 20 μM necrostatin-1 (HCOE10 + NEC-1), or 10 mg/mL HCOE + 20 μM Z-VAD-FMK (HCOE10+Z-VAD). (**A**) Cell death percentage was evaluated by flow cytometric assay, counting sub-G1 events. Results are expressed as mean ± SD of three independent experiments (\* *p* < 0.05; \*\* *p* < 0.01 vs. control). (**B**) Alive cells were counted by the Trypan Blue exclusion test at 48 h of treatment with HCOE10, HCOE10+NEC-1, and HCOE10+Z-VAD. Results expressed as percentage with respect to the PBS control (CNT; considered as unit, 100%) are expressed as the mean ± SD of three independent experiments (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control). (**C**) A375 cell viability was measured by MTT assay after 48 h of treatment with HCOE10, HCOE10+NEC-1, and HCOE10+Z-VAD. Results, expressed as percentage with respect

to the PBS control (CNT; considered as unit, 100%), represent the mean ± SD of three independent experiments (\*\* *p* < 0.01 vs. control). Mitochondrial mass (**D**) and membrane potential (**E**) measurements, after 4 h of treatment with carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (10 μM) (used as positive control) or 48 h of incubation with 10 mg/mL HCOE, were carried out by using MitoTracker Green (MTG) and MitoTracker Red CMX ROS (MTR), respectively. Results are expressed as percentage variation of fluorescence with respect to PBS control (CNT; considered as unit, 100%) (\* *p* < 0.05; \*\* *p* < 0.01 vs. control). (**F**) Changes in mitochondrial membrane potential are expressed as MTG/MTR ratio (\* *p* < 0.05 vs. control).

#### *3.6. Apoptosis*/*Necroptosis via Mitochondrial Pathway Is Induced in A375 Cells by HCOE*

The decrease of melanoma cell growth, together with the evidence of high toxicity, cell cycle arrest, and MITF pathway inhibition, suggested that HCOE could induce cell death. For this reason, A375 cell viability was evaluated after co-treatments with oregano extract and Z-VAD-FMK (an anti-apoptotic pan-caspase inhibitor; [60]) or necrostatin-1 (NEC-1) (an inhibitor of necrosis/necroptosis [61]). Flow cytometry analysis (Figure 4A) showed that both co-treatments partially suppressed cell death (−17.43% of cells in the sub-G1 phase for the HCOE10 + NEC-1 sample; −22.90% for the HCOE10 + Z-VAD sample), with respect to the HCOE10-treated sample. Trypan Blue exclusion test (Figure 4B) confirmed this result, evidencing an increase of alive cells of 9.31% and 24.82% after exposure to HCOE10 + NEC-1 and HCOE10 + Z-VAD, respectively, compared to the treatment with only HCOE10. On the contrary, by MTT assay (Figure 4C), only HCOE10 + Z-VAD treatment seemed to rescue A375 cell viability compared to the HCOE10 sample.

Taking into account that mitochondrial damage is one of the main consequences of the oxidative stress (previously documented by ROS and RNS monitoring) and that the MTT assay (reported above) is an indicator of mitochondria activity [62,63], mitochondrial mass and membrane potential were estimated, respectively, by MitoTracker Green (MTG) and MitoTracker Red CMX ROS (MTR) cytofluorimetric assays in A375 exposed to HCOE10 for 48 h. CCCP, a well-known mitochondrial uncoupler, was used as a positive control. As shown in Figure 4D–F, HCOE10 treatment slightly affected mitochondrial mass, while it induced a strong depolarization. MitoTracker Red fluorescence was also normalized with the MitoTracker Green signal (MTR/MTG); this ratio was reduced by 42.17% after oregano treatment compared to the control.

Mitochondrial membrane permeability loss suggested mitochondrial damage and apoptosis induction. To check this hypothesis, Western blotting analyses were carried out, monitoring specific markers of these phenomena. In particular, the protein levels of Bax, Bcl-2, cytochrome c (Cycs), and mitochondrial import receptor subunit TOMM20 were detected and quantified (Figure 5A–C). After 48 h of exposure to HCOE10, an approximately 2-fold increase of Bax/Bcl-2 ratio was observed. Moreover, as demonstrated in the case of staurosporin treatment (STS, used as positive control) [64], the plant extract caused the increase of Cycs (+135.2%) compared to TOMM20. The lack of changes in the TOMM20 level in oregano-treated cells confirmed the previous cytofluorimetric results for mitochondria mass..

Finally, additional immunoblots (and relative densitometric quantitations) were performed to study caspase-3 (Casp-3) and Parp-1 levels, as shown in Figure 5D–F. HCOE10 treatment, for 48 h, significantly decreased pro Casp-3 and full-length Parp-1 levels with respect to the negative control (CNT), inducing Parp-1 cleavage as also observed in the presence of STS. These effects were almost completely rescued by treating melanoma cells with HCOE10 and Z-VAD simultaneously.

**Figure 5.** Analysis of mitochondria, apoptosis, and DNA damage markers. (**A**) Representative Western blotting membranes of Bcl-2, Bax, TOMM20, Cycs, and Gapdh protein levels evaluated in A375 cells treated for 4 h with staurosporine (STS, 2 μM) (used as positive control) or for 48 h with 10 mg/mL HCOE are shown. Quantitation of Bcl-2, Bax, TOMM20, and Cycs proteins was performed and the ratio

between Bax/Bcl-2 (**B**) and Cycs/TOMM20 (**C**) are reported as percentage values, with respect to PBS control (CNT; considered as unit, 100%). (**D**) Representative Western blotting membranes of full length and cleaved form of Parp-1, pro caspase-3 (pro Casp-3), and Gapdh protein levels are shown. Quantitation of pro Casp-3 and Parp-1 proteins in A375 cells treated for 4 h with STS or for 48 h with 10 mg/mL HCOE and 10 mg/mL HCOE + 20 μM Z-VAD-FMK (HCOE10 + Z-VAD) are reported. (**E**) Pro Casp-3 levels evaluated by the ratio between pro Casp-3 and Gapdh (used as internal loading control) are expressed as percentage values with respect to PBS control (CNT; considered as unit, 100%). (**F**) The cleaved Parp-1/full-length Parp-1 ratio is expressed as percentage values with respect to PBS control (CNT; considered as unit, 100%). (**G**) P53 mRNA level were measured by RT-PCR in A375 cells treated for 48 h with 10 mg/mL of HCOE. Gene expression, calculated as transcript amount after normalization for ACTB mRNA (2−ΔΔCt), is reported as percentage with respect to the PBS control (CNT; considered as unit, 100%). (**H**) A representative Western blotting membrane of p53 and Gapdh protein levels is shown. (**I**) Quantitation of p53 protein in A375 cells treated for 48 h with 10 mg/mL HCOE is reported. Results obtained by the ratio between p53 and Gapdh (used as loading control) signals are indicated as percentage values with respect to PBS control (CNT; considered as unit, 100%). All data indicate the mean ± SD of three independent experiments (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. negative control).

#### *3.7. HCOE Triggers DNA Breakages Mediated by Metal Ions*

Since the pro-oxidant activity of several plant metabolites has been associated to DNA damage [65], the P53 gene expression level was monitored both in terms of transcript and protein amount, in A375 cells after exposure to HCOE10, for 48 h. As shown in Figure 5G–I, oregano treatment increased p53 mRNA (+102.1%) and protein (+95.95%) concentration, compared to the control.

To confirm the induction of DNA breakages by oregano extract, γH2AX and 53BP1 foci, two well-known markers of DNA damage [66,67], were detected by immunofluorescence (IF) analysis on A375 cells treated with HCOE10 for 48 h. In detail, an average of 14.86 ± 1.50 γH2AX foci and 14.10 ± 1.05 53BP1 foci per cell were measured with respect to 2.60 ± 1.56 γH2AX foci and 3.27 ± 1.14 53BP1 foci per cell found in the control sample (Figure 6A). Representative IF images per each sample were reported in Figure 6B; here, the treatment with etoposide (ETO), an inhibitor of topoisomerase II enzyme, represented the positive control. IF analysis showed that the major part (82% of cases) of γH2AX and 53BP1 foci co-localized (Figure 6B, panel l). In addition, considering that polyphenols, such as flavonoids, catalyze DNA breakages in the presence of metal ions (e.g., copper) [68], we evaluated the ability of oregano extract to trigger DNA damage in the presence of a copper chelator (TETA). The co-treatment HCOE10 + TETA determined a significant reduction of the level of DNA breakages (9.91 ± 2.91 γH2AX foci and 5.71 ± 1.87 53BP1 foci per cell) with respect to the pure treatment with HCOE10 (Figure 6A,B, panels q–t).

**Figure 6.** Immunofluorescence analysis. (**A**) The amount of γH2AX and 53BP1 foci detected in A375 cells treated with PBS (CNT), etoposide (ETO), 10 mg/mL HCOE (HCOE10), triethylenetetramine (TETA) or HCOE10+TETA were counted. Results are expressed as mean of foci per cell ± SD of three independent experiments (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001 vs. control). (**B**) Representative immunofluorescence images of A375 cells treated with PBS (CNT) (**a–d**), etoposide (ETO) (**e–h**), 10 mg/mL HCOE (HCOE10) (**i**–**l**), triethylenetetramine (TETA) (**m**–**p**), or HCOE10+TETA (**q**–**t**) are shown. γH2AX foci are in green, 53BP1 foci are in red, while nuclei were stained in blue with DAPI. Merged images are also reported. The white bars indicate 15 μm.

#### **4. Discussion**

Among tumors, skin cancer is the most common neoplasia worldwide. In particular, the more aggressive and deadliest form of this pathology is represented by melanoma [69]. Melanoma is a multi-factorial disease, depending on both environmental and endogenous factors. Indeed, about 90% of melanomas are caused by ultraviolet light exposure [70], while the remaining 10% has been associated to genetic defects [71]. Currently, such type of skin cancer is treated by surgical removal, which leads to a high survival rate except in the presence of metastases. In the latter case, a chemotherapeutic approach based on several drugs, such as Dacarbazine (an alkylating agent), Vemurafenib (BRAF kinase inhibitor), Ipilimumab (monoclonal antibody targeting for cytotoxic T-lymphocyte antigen-4), Pemrolizumab (monoclonal IgG4 antibody), and Nivolumab (monoclonal antibody targeting for Programmed Death-1 protein) [72], is the most efficient strategy to treat melanoma. However, the aggressiveness and the high rate of multi-drug resistance of this pathology highlight the need of new antineoplastic molecules.

Phytochemicals, namely secondary metabolites, produced by plants to protect themselves from environmental stresses and promote their reproduction [73], have been widely documented to exert a great variety of non-negligible bioactivities even on mammalian systems. Indeed, it has been documented that several plant compounds promote apoptosis and inhibit metastasis and angiogenesis [74–77]. For this reason, they have been taken under consideration thanks to their relevant medical and pharmaceutical properties. In this scenario, plant phytocomplexes can represent potential antiproliferative and anti-invasive cocktails for drug-resistant melanomas.

Based on the previous evidence, in the current research, the biological effect of a hydroalcoholic extract from plants of *Origanum vulgare* L. ssp. *hirtum* (HCOE) was investigated on highly metastatic and drug-resistant murine (B16-F10) and human (A375) melanoma cells. Indeed, although a potential antineoplastic effect has been associated to oregano extracts [30–35], the capacity of this herb to contrast the growth of the above-mentioned melanoma lines has never been elucidated in detail. Moreover, as one of the main goals of the cancer research is the discovery of new drugs with limited or without adverse side effects for healthy tissues, oregano extract was also tested on C2C12 myoblasts, which is a non-tumor cell model.

Since plant extract bioactivity cannot be attributed only to its more representative compounds, rather than to the synergic effect of plant molecules present both in high concentration and in trace [78–80], first of all, we investigated the biochemical profile of the oregano extract used in this study by chromatographic approaches (HPLC-DAD and GC-MS). A total of 58 metabolites were detected and recognized. Moreover, the amount of total phenols and flavonoids in HCOE was also measured, together with its in vitro antiradical power, to better characterize the plant extract.

The GC-MS chemoprofile obtained in the current research totally was in line with those documented in the literature. In this regard, *O. vulgare* ssp. *hirtum* essential oil can be classified in four chemotypes, according to its main constituents, especially thymol and carvacrol, and relative ratios. For instance, *O. vulgare* ssp. *hirtum* essential oil extracted from plants grown in Southern Italy and Northern Greece would seem rich in thymol, while that obtained from plant material propagated in Southern Greece was abundant in carvacrol [81–83]. Although purified from plants grown on Mount Athos (Northern Greece), our oregano extract showed a chemoprofile more similar to those of Southern Greece. To explain this phenomenon, it is important to keep in mind that several independent variables, such as plant growth stage and environment conditions, may strongly influence the phytocomplex. On the other hand, the oregano sample here studied revealed a content of phenolics double compared to that reported in the literature [84,85].

As concerns oregano biological activity, the plant phytocomplex determined a great reduction of B16-F10 cell growth, especially after 48 h of exposure with the highest doses (6, 8, 10 mg/mL), while it minimally influenced the myoblast division rate. B16-F10 and C2C12 proliferation curves confirmed the previous MTT outcomes. Moreover, HCOE induced a significant time- and dose-dependent toxicity on murine melanoma cells, whereas a low percentage of Trypan blue positive cells was detected in the case of C2C12. To validate these results, IC50 values and selectivity indexes (SI) for the plant treatments on both cell lines were calculated. According to the literature, a reliable SI value must be equal to or greater than 2 [86], and HCOE satisfied this requirement. Cell cycle analysis of B16-F10 demonstrated that oregano extract caused an increase of cells in G2/M phase.

This promising evidence encouraged us to check if oregano extract could exert antineoplastic effects also against a human melanoma cell line with the aim to lay the basis for future desirable application in translational medicine. As expected, significant antiproliferative activity, with a high selectivity index, together with a relevant cytotoxic effect of HCOE on A375 cells was confirmed.

Taking into account all previous data, the concentration of 10 mg/mL of oregano sample was selected for further experiments, showing the best antiproliferative effect.

Cytofluorimetric and RT-PCR analyses proved that the plant extract blocked cell division in the G2/M phase, acting on the expression of specific key genes implicated in the inception of the mitotic process (i.e., CDK1, CCNB1, P21, and P27) [53,55,56] such as paclitaxel, which is a well-known plant anticancer drug with antimitotic property.

Before proceeding with the other analyses, as the final objective of the current research was the valorization of an oregano hydroalcoholic extract for potential chemotherapeutic applications, the control of the safety for the plant preparation with reference to mutagenic properties was necessary. For this purpose, the Ames test, recommended by the European Food Safety Authority (EFSA) as a proper assay to assess food safety [87], was carried out. It provided the proof that HCOE did not have a non-mutagenic effect [88,89], as expected. Indeed, *O. vulgare* and its derivatives, which are accepted as food ingredients by the U.S. Food and Drug Administration, are listed among the GRAS (Generally Recognized As Safe) substances by the Code of Federal Regulations of the USA and are generally well tolerated by the human body, although gastrointestinal upset and skin allergic reactions have been associated to them [90]. However, the lack of data regarding oregano genotoxicity [91,92] highlights the need for further studies on this topic and valorizes the present preliminary results. The mutagen-protective activity of HCOE was also investigated in order to further analyze the plant extract under a healthy point of view. Unfortunately, at all tested doses, no protective property against known mutagenic compounds was documented.

In the second part of this work, the molecular mechanism underlying the bioactivity of HCOE on melanoma cells was investigated in depth. First of all, the efficiency of the MITF pathway was verified, studying the expression rate of MITF, TYR, and TYRP1 genes. The results revealed an antimelanogenic activity of the oregano extract on A375, as already suggested by the literature [93].

At low levels, reactive species play a key role in cell signaling, but their overproduction can lead to mitochondrial alterations (i.e., DNA mutations, respiratory chain damage, membrane permeability loss), oxidative stress, and an inhibition of specific genes, including those related to melanin synthesis [29,94]. Therefore, according to the previous results, intracellular ROS and RNS levels were measured in A375 cells, demonstrating that HCOE10 had a remarkable pro-oxidant effect, especially after 24 h of incubation. This evidence suggested that oregano antitumor activity could be based on a reactive species-mediated apoptotic process. Literature data about carvacrol bioactivity, one of the most abundant phenolic monoterpenoids of oregano (as also documented in the present research by GC-MS analysis, see Table 2), would support this hypothesis. Indeed, several published works showed that this compound induces apoptosis in different tumor cell lines by increasing the ROS amount and disrupting mitochondrial membrane potential [95–97].

All previous considerations suggested that HCOE10 could trigger cell death in A375. For this reason, to clarify which cell death pathway was induced by oregano extract, a pan-caspase inhibitor, Z-VAD-FMK, and an inhibitor of necroptosis, necrostatin-1 [98], were used in co-treatments with HCOE10 on the melanoma cells. The experiments demonstrated that the decrease of tumor cell proliferation was partially due to both apoptosis and necroptosis induction, as already suggested by Savini et al. [99] and Rubin et al. [100]. Indeed, these two phenomena are strongly correlated to each other because they share the same stimuli (such as TNF-α), ligands, and receptors [101,102]. The mechanism underlying the activation of apoptosis and/or necroptosis is still under investigation. Nevertheless, Annexin V/propidium iodide staining in the presence of necrostatin-1, as well as the analysis of TNF-α level, could be performed in the future to better clarify the role of these two pathways in oregano-induced cell death [103,104].

Since in MTT assay, only HCOE10 + Z-VAD double treatment rescued A375 viability, taking into account that mitochondria are both generators of and targets for reactive species (whose levels were previously observed to be increased in the presence of HCOE), a mitochondria impairment caused by oregano extract was hypothesized. To validate this theory, cytofluorimetric analyses, based on the use of MitoTracker Green and MitoTracker Red CMX-ROS probes, were performed. The data confirmed a loss of mitochondrial potential after HCOE10 treatment, while no significant change in mitochondrial mass was appreciated. By contrast, CCCP-treated cells (the positive control) showed an increase of mitochondrial mass, which was probably due to mitochondrial fragmentation [105,106] associated to an expected membrane depolarization.

As known in the literature, the permeabilization of the mitochondrial outer membrane is an event promoted by the pro-apoptotic protein Bax. Bcl-2, on the other hand, is an anti-apoptotic factor that prevents apoptosis by inhibiting Bax. Therefore, an elevated Bax/Bcl-2 ratio is a feature of apoptotis induction. As a consequence of Bax activation, damaged mitochondria release cytochrome c into the cytoplasm, leading to caspase-3 induction and Parp-1 cleavage [107,108]. In view of this, the amount of apoptosis-related proteins (Bax, Bcl-2, Cycs, Casp-3, and Parp-1) were evaluated in A375 after 48 h of incubation with HCOE10. Oregano extract was able to trigger apotosis in the human melanoma cells by increasing Bax and Cycs concentrations and decreasing Bcl-2 and pro Casp-3 amounts. Moreover, as expected, Parp-1 cleavage was evident in oregano-treated cells. HCOE10 + Z-VAD co-treatment reversed HCOE effect, showing protein levels similar to those detected in the respective negative control. All these results confirmed that the plant extract induced, in A375 cells, a caspase-dependent apoptosis, which was mediated by mitochondrial damage. However, considering that both ROS and RNS play an important role also in autophagy and that mitophagy is strictly involved in mitochondrial turnover [109,110], it would be interesting in the future to investigate if autophagy/mitophagy is also induced by oregano treatment.

Another important key element in the apoptotic process is the P53 gene, whose protein promotes BAX gene expression by direct activation of its promoter and BCL2 downregulation [111–113]. P53 is activated in response to a wide range of genotoxic insults. It is involved in several DNA-repair machineries, such as nucleotide excision repair, for the removal of helix-distorting lesions (typical of UV-damage) and base excision repair (BER) in case of base oxidative modifications [114,115]. As known in the literature, plant compounds, especially polyphenols, can cause oxidative DNA strand breakage, alone or in the presence of transition metal ions. Indeed, among all, the copper (that is the most abundant ion of the cell nucleus, together with zinc), after reduction from Cu(II) to Cu(I) form by the action of plant metabolites, is particularly prone to produce ROS (especially the hydroxyl radical) during its re-oxidation, bind chromatin (particularly guanines), and cause DNA breakages [116,117].

The existence of such types of event prompted us to analyze P53 mRNA and protein levels and evaluate DNA damages (detecting γH2AX and 53BP1 foci by IF) upon HCOE10 exposure. Oregano treatment induced in A375 p53 upregulation, at both the transcriptional and translational level, and DNA breakages. In detail, γH2AX is the phosphorylated form of the histone 2AX (H2AX). Its phosphorylation is an early consequence of double and single-strand breakages [118–121]; therefore, the detection of γH2AX is widely used as a marker of DNA damage. On the other hand, p53 binding protein 1 (53BP1) locates only to DNA double-strand breaks [122]. IF analyses highlighted a great number of γH2AX and 53BP1 foci, which often co-localized in HCOE10-treated cells with respect to untreated controls. This evidence, together with Ames test results, pointed out that oregano hydroalcoholic extract acts as a genotoxic but not mutagenic agent in A375 melanoma cells, causing DNA single and double-strand breaks [123]. Furthermore, taking into account that DNA damage often results from the binding of phytochemical to transition metal ions, such as copper [124–126], co-treatments with HCOE10 and a copper chelator (TETA) were performed. Surprisingly, the sequestration of copper by TETA protected A375 cells from HCOE10-induced DNA damage, confirming that this metal ion plays a fundamental role in oregano bioactivity and relative DNA cleavage reaction.

In cancer masses, the concentration of iron and zinc is lower, whereas the copper concentration is usually higher than in healthy tissues [127–132]. This feature can explain why plant compounds exert selective cytotoxic activity against tumor cells but not toward non-tumor ones [133–139], justifying our data about oregano cytotoxicity on B16–F10 and A375 but not on C2C12.

In conclusion, *O. vulgare* hydroalcoholic extract, due to its peculiar chemical profile and pro-oxidant effect, inhibits melanogenesis and melanoma cell proliferation. Through in-depth molecular analyses, the antineoplastic activity of the oregano extract was associated to its ability to trigger programmed cell death (apoptosis and necroptosis) in A375 human melanoma cells via mitochondria and DNA damage. As this molecular mechanism was correlated to the intracellular/nuclear concentration of copper ions, oregano phytochemicals appeared to be slightly toxic or non-toxic for non-tumor cells. All this evidence represents a robust starting point for further studied focused on the design of new anti-melanoma natural drugs. Indeed, according to the present data, *O. vulgare* phytocomplex, working in synergy, represent an excellent candidate as anticancer agent, being highly selective and effective against human melanoma cells.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2304-8158/9/10/1486/s1, Figure S1: HPLC analysis; Table S1: RT-qPCR primers; Table S2: Ames test for mutagenic analysis; Table S3: Ames test for mutagen-protective analysis.

**Author Contributions:** Conceptualization: V.N., A.G. and G.S.; Methodology: A.G.; Formal analysis: V.N., G.D.M. and G.S.; Writing—Original draft preparation: V.N. and A.G.; Writing—Review and editing: all authors; Resources: A.C., G.S. and A.G.; Supervision: Antonella Canini and A.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors want to thank Marco Sarandrea (SARANDREA MARCO & CO. srl) who performed the sampling of the plant material and monks of MOUNT ATHOS VATOPEDI HOLY MONASTERY (Iερα´ Moνη´ Bατoπεδι´oυ) in Greece who kindly provided us oregano plants. Moreover, we also thank Alessandro Grandini for technical assistance performing the Ames test, Miss Sophie Gart for the revision of the English, and Maria Rosa Ciriolo, Katia Aquilano, Carlo Rodolfo and Daniele Lettieri Barbato for their suggestions on mitochondrial analyses. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
