3.1. Yield of Arbutus unedo Leaves Extracts
Many researchers are currently focusing on replacing harmful solvents with environmentally friendly substitutes. The use of green and sustainable solvents such as ethanol and water, combined with the use of low environmental impact technologies, is a promising holistic approach to the development of “green” extraction processes. Ethanol is recognized as a green solvent and offers greater industrial safety. On the other hand, water has a low environmental impact and is inexpensive in terms of production, transport, and disposal. On the other hand, the possibility of modifying the physico-chemical properties of water by varying the conditions (temperature, pressure, etc.) has increased the interest in its use as an extraction solvent. In this study, ethanol and water were used in order to evaluate and optimize the extracting ability of phenolic compounds from the leaves of
Arbutus unedo, as well as their influence on the biological activities of this specie, by varying the extraction process. The extraction of phenolic compounds by “green” non-toxic solvents was carried out by two conventional (or classic) methods, maceration and reflux extraction, and an unconventional (or innovative) namely extraction by sonication (
Table 2).
Yield results are significantly different and vary markedly depending on the solvent and extraction technique (
Table 2). Generally,
Arbutus unedo seems rich in polar compounds. In fact, the best performance was observed using the reflux extraction technique and water as solvent (38%), followed by extraction with 50% ethanol by maceration and sonication with yields of 35% and 30%, respectively. Cold water extraction and sonication have also shown high yields (28% and 29%). Absolute ethanol, especially by maceration, proved to be the least efficient with the lowest yields (5%). The variation in yields agrees with data from the literature, which indicates that the yield depends on the extraction solvent’s nature and polarity [
29]. Our results suggest that the yield increases with increasing solvent polarity.
Regardless of the extraction method, we observed that the yields using 50% ethanol were much higher than those using pure ethanol, which indicated that the extracting power of ethanol is improved by adding water following the increase in its polarity. Moreover, our results are consistent with several works in the literature indicating the efficiency of water in extracting soluble compounds from
Arbutus unedo leaves. In effect, Oliveira et al. [
30] reported a significant yield difference as a function of the polarity of the solvent used, varying from 2.8% using petroleum ether to 32.4% using boiling water. These authors showed an average yield of 15% using ethanol. Similarly, Malheiro et al. [
18] worked on 19
Arbutus ecotypes Portuguese and noticed hot water extraction yields ranging from 27% to 61%. However, Orak et al. [
31] noticed that ethanol and water showed yields close to 40% and 39%.
3.3. Phytochemicals Identification
Although the total contents of bioactive molecules are valuable data regarding the phytochemical profile of a plant, they do not allow the separate quantification of the main bioactive compounds. For this reason, a detailed assessment of the phenolic compounds in the seven different extracts was performed by HPLC-DAD analysis, and the results are displayed in
Table 3. The main compounds in the extracts of
Arbutus unedo leaves were flavonoids and phenolic acids. The obtained data highlighted the richness of extracts on hyperoside (quercetin 3-
O-galactoside) with amounts ranging from 30 to 56 mg/g DR. The highest content was observed in reflux water (56 mg/g DR) and ultrasound ethanol (55 mg/g DR). As previously reported in the literature, hyperoside was the most predominant phenolic compound in Croatian
Arbutus unedo leaves [
19].
Moreover, samples contained interesting amounts of arbutin. The highest quantity was obtained in water maceration extract (3 mg/g DR) followed by ethanol maceration and ethanol 50% ultrasound by about 2 mg/g DR. This hydroquinone-β-D-glucopyranoside is common in the leaves of several plant species, particularly in the Ericaceae. [
36]. Jurica et al. [
33] compared the effectiveness of different solvents (methanol, methanol 50%, ethyl acetate, and dichloromethane) and techniques for arbutin extraction from
A. unedo leaves and reported that ultrasound-assisted extraction with methanol was the most suitable extraction procedure for the recovery for arbutin. The arbutin content in
A. unedo leaves has been reported to vary according to the origin due to the effect of climate and soil characteristics, ranging from 0.6 mg/g to 12.4 mg/g [
36,
37].
Other flavonoids detected in our study were rutin, catechin, and epigallocatechin (
Table 3). Among the phenolic acids, ellagic acid was the most abundant; the highest quantities were recovered in water and ethanol ultrasound extracts (1.5 mg/g DR). Our results agree with several studies which have established that the main phenolic compounds present in
A. unedo leaves are catechin, epicatechin, catechin gallate, quercetin, gallic acid, ellagic acid and
p-hydroxybenzoic acid [
34,
38,
39,
40].
3.4. Effect of Solvent and Extraction Method on Antioxidant Activities
To understand the mechanism of action of the antioxidants present in the extracts of
Arbutus unedo leaves, four different methods were used, and the results are depicted in
Table 4.
The results showed that the nature of the solvent and the extraction technique influence the total antioxidant activity of
A. unedo (
Table 4). Water was found to be the most powerful for extracting antioxidant compounds. In particular, water reflux extraction exhibited a very high value of 194 mg GAE/g DW. Water cold and ultrasound extractions also showed interesting activities (86 mg GAE/g DW and 63 mg GAE/g DW, respectively), while ethanol and ethanol 50% maceration presented the lowest activity, around 32 and 52 mg EAG/g DW. We noted that the ultrasound technique improves the activity in ethanol extracts (TAA = 65 and 59 mg GAE/g DW in ethanol and ethanolic 50%, respectively).
Moreover, the results showed that the significant total antioxidant activity of water reflux extract was positively correlated to its richness in flavonoids and tannins. Indeed, this extract showed the highest levels of these compounds, which suggests that flavonoids and total tannins participated actively in the antioxidant activity of Arbutus unedo extracts.
The scavenging activity measured by the DPPH assay showed that the different extracts from
A. unedo exhibited high capacity with low IC
50 values that do not exceed 17 μg/mL (
Table 4). The ethanolic extract (50%) obtained by sonication and water reflux extract showed the highest trapping power of the DPPH (6 and 7 μg/mL, respectively). It should be noted that apart from the cold aqueous extract, the antiradical capacities obtained are higher than the synthetic antioxidant standard BHT (IC
50 value is equal to 11.5 μg/mL).
Besides, the results relating to the antiradical activity against the ABTS radical are in line with those of antiradical activity against the DPPH radical (
Table 4) and indicate that water reflux extract exhibits the best activity displaying a low IC
50 (58 μg/mL). However, this capacity becomes lower when the solvent used is ethanol. Ethanol 50% (IC
50 = 68 μg/mL), ethanolic extracts presented moderate antiradical activity since the IC
50 varies from 94 to 114 μg/mL. Our results agree with other reports, which underlined a significant antiradical potential of
Arbutus extracts. Moderate scavenging activity has been reported for the Portuguese and Turkish
A. unedo, with IC
50 values ranging from 73 to 487.2 μg/mL [
30,
31,
41]. Indeed, the high antiradical potential of the various extracts, especially that of the water extract by reflux, could be due to their richness in flavonoids, in particular, hyperoxide present with levels varying from 28 to 37 mg/g DW (
Table 3). The antioxidant activity of hyperoxide has been described as being related to the hydroxyl groups of the A and B rings and the glycosides bound to the C ring [
42]. Liu et al. [
43] found that hyperoxide could effectively protect PC12 cells from ROS-induced cytotoxicity, including hydrogen peroxide and tert-butyl hydroperoxide, without being damaging. Flavonoids are known to be powerful scavenger agents and effective hydrogen donors, acting as primary antioxidants and stabilizing radicals. In addition,
A. unedo leaf extracts are rich in
p-coumaric acid. This last one is one of the most active free radical scavenging hydroxycinnamic acids [
44]. Finally, the antiradical activity of the
Arbutus could also be due to the presence of arbutin. Indeed, this flavonoid has been reported as a powerful radical scavenger [
45].
According to the results illustrated in
Table 4, we noticed that the reducing powers of
A. unedo leave extracts were significantly different and varied markedly depending on the solvent and extraction technique. Hence, the ethanolic extracts, as well as that of the water reflux method, presented the lowest values of the EC
50 (varying from 60 and 83 μg/mL), therefore, the strongest reducing activities of the iron. Indeed, we noted that the pure ethanol extract of the ultrasound is distinguished by the best-reducing power of iron (EC
50 = 60 μg/mL). As for the previous tests, the water reflux extract showed a very high reducing power with an EC
50 of 83 μg/mL. On the contrary, lower activity was observed in the aqueous extract obtained by maceration, of which the EC
50 value is equal to 112 μg/mL. Our results reveal that the Tunisian
Arbutus exhibits an important reducing power. The lower reducing activity was noted by Malheiro et al. [
18] in aqueous extract leaves of 19 Portuguese
A. unedo genotypes. These authors explained that the antioxidant power of the
Arbutus would be due to its richness in reducing phenolic compounds.
Concerning the chelating power of iron which was measured by the inhibition of the formation of ferrozine-Fe
2 complex, our study revealed a low chelating capacity of ferrous iron; the aqueous extracts showed moderate activity displaying EC
50 values ranging from 33 to 40 mg/mL (
Table 4).
Our results revealed a strong antioxidant potential of the Tunisian A. unedo. In particular, the aqueous extraction under reflux was distinguished by a very high power to trap free radicals as well as a high total and reduced antioxidant activities.
3.5. Effect of Solvent and Extraction Method on Anti-Tyrosinase Activity
A. unedo leaves extracts were analyzed for tyrosinase inhibition activity. Tyrosinase is a copper enzyme that plays a main role by catalyzing the first two steps in melanogenesis. Firstly, it converts L-tyrosine into L-DOPA by hydroxylation, and secondly, this compound is converted into o-dopaquinone by oxidation which polymerizes spontaneously to form melanin, which is the key molecule for skin color [
46]. Melanin overproduction can produce hyperpigmentation disorders, such as lentigo, melasma, and hyperpigmentation. As a consequence, tyrosinase inhibitors are promising potential skin-whitening agents [
5]. There are well-known tyrosinase inhibitors such as hydroquinone; nevertheless, their adverse effects are a serious concern, which leads to the search for natural compounds that have tyrosinase inhibitory effects [
47].
The inhibition of tyrosinase by
A. unedo extracts was assessed on its two catalysis functions, i.e., monophenolase (the inhibition of L-tyrosine hydroxylation to L-DOPA) and diphenolase (L-DOPA oxidation to dopaquinone) activities. Interestingly and considering monophenolase inhibition, all the extracts were able to inhibit tyrosinase activity efficiently, although to a different extent (
Table 5). Ethanol and ethanol 50% maceration extracts exhibited the strongest activities (IC
50 = 90 μg/mL), followed by ethanol 50% ultrasound (IC
50 = 150 μg/mL) and decoction (IC
50 = 200 μg/mL), whereas ultrasound ethanol was the less potent inhibitor for monophenolase activity.
Some extracts displayed moderate diphenolase activity. Ethanol 50% obtained by ultrasound and maceration as well as ethanol maceration showed the highest activities with IC
50 values of 400, 450, and 500 μg/mL, respectively. The other extracts were rather ineffective (IC
50 > 1 mg/ mL). These results confirm previous observations on moderate diphenolase
A. unedo activity. Recently, Deniz et al. [
48], have investigated the enzyme inhibitory activity of 92 herbal ethanol extracts and showed that
A. unedo ethanol 80% leaves extract by maceration showed 32% diphenolase inhibition at 666 μg/mL.
These data pointed out the high potency of
A. unedo monophenolase activity and revealed strong lightening capacity. This remarkable activity is likely to result from high arbutin content in the different extracts (
Table 3), which act as the major tyrosinase-modulating compound. In our study, this phenolic exhibited high monophenolase activity (IC
50 = 100 μg/mL,
Table 5) while it was inactive in diphenolase activity. Recently, the effect of α-arbutin on the monophenolase and diphenolase activities of tyrosinase was analyzed and reported that this compound inhibits monophenolase activity and activates diphenolase activity [
49]. In addition, it has been shown that arbutin is also safe and can potentially prevent melanin formation without cytotoxicity [
50]. Data on the antioxidant capacity of arbutin are emerging, and these antioxidant properties are proposed to contribute to the skin-lightening effect of the molecule.
There are several other compounds present in extracts from
A. unedo plants with tyrosinase-modulating properties. Accordingly, Huang et al. [
51] explained that ellagic acid inhibited tyrosinase activity in a reversible manner and was a mixed tyrosinase inhibitor. Moreover, catechins and
p-coumaric acid were reported to inhibit tyrosinase by acting as alternative substrates [
52].
3.6. Effect of Solvent and Extraction Method on Cytotoxic and Anti-Inflammatory Activities
Due to the significant side effect profiles of drugs, the use of natural compounds for preventing or reducing inflammation has recently received a lot of attention. The cytotoxicity of
A. unedo extracts was studied using the resazurin assay. RAW 264.7 cells were treated with different concentrations of extracts (25–300 μg/mL), as shown in
Figure 1. Independently of technique, both water and ethanol of
A. unedo extracts exhibited no cytotoxic effect for the tested concentrations up to 300 μg/mL since cell viability exceeded 80%. Interestingly,
A. unedo extracts, especially water reflux and ethanol extract by maceration at 50 μg/mL, caused an increase in cell viability by 11% compared to the control. Based on this, concentrations ranging from 25 to 150 μg/mL were selected for further studies.
Nitric oxide (NO) is a strong mediator in numerous cellular processes, such as the regulation of neurotransmission, vasodilatation, inhibition of platelet adherence, host defense mechanisms, and inflammation [
53]. LPS can activate macrophage cells to initiate proinflammatory mediators, including TNF-α, IL-6, and NO. Therefore, the utilization of NO inhibitors constitutes a substantial therapeutic advance in the treatment of inflammatory diseases. Noncytotoxic
A. unedo extracts concentrations were examined to explore their potential to inhibit NO production in LPS-treated RAW 264.7 macrophages. Extracts showed anti-inflammatory activity in a concentration-dependent manner (
Figure 2). The most active extracts were ethanol maceration and water extract obtained by ultrasound with percentage NO production inhibition of 37 and 35%, respectively, at a high dose of 150 μg/mL. Some studies highlighted the potential activity of
A. unedo extracts as anti-inflammatory candidates. Tenuta et al. [
16] showed that
A. unedo ethanolic and hydroalcoholic macerations were able to reduce nitrite production in HFF1 cells stimulated with interleukin-2β.
Mariotto et al. [
54] reported that treatment with an aqueous extract of
A. unedo decreased acute lung inflammation in an animal model. Among
A. unedo phenolics, arbutin suppressed LPS-induced production of NO and expression of inducible NO synthase (iNOS) and cyclooxygenase-2 (COX-2) in LPS stimulated BV2 microglial cells in a dose-dependent manner and without causing cellular toxicity [
55].
3.7. Characterization of Emulsion Containing A. unedo Extract
After evaluating the phytochemical composition and the activities of the seven A. unedo extracts, it was found that the reflux water has the best performance with satisfactory compositional characteristics that allowed its use in the elaboration of a cosmetic emulsion based on a high phenolic compounds content and elevated antioxidant and anti-tyrosinase activities. Thus, a formulation that contained 1% of the selected A. unedo extract was developed.
Initially, the formulation loading phenolic-enriched A. unedo extract was light yellow and homogeneous, while the base formulation was white homogeneous. This color difference is caused by the color of the extract influencing the color of the end product.
On the other hand, total polyphenols retention evaluation showed a satisfactory level of phenolics retention (60.32%) in the nanoemulsion.
Stability tests are crucial because of their predictive nature. The centrifugation test was carried out to obtain information on possible instability processes. Formulations were centrifuged 24 h after preparation at 3000 rpm for 30 min. The emulsion loading phenolic-enriched A. unedo extract showed no change in its initial characteristics and stability. The emulsion was described as stable, as no cremation, flocculation, or phase separation was observed during the available time. Similar behavior was observed in the thermal stress test in which the cream did not show changes during the whole process.
Regarding pH analysis, the emulsion loading phenolic-enriched
A. unedo extract initially presented a pH value of about 6.9 (
Table 6). The value for the emulsion subjected to thermal stress for eight days was 6.5. Both values are compatible with skin [
56]. Therefore, these emulsions were considered suitable for use.
Zeta potential and particle size measurement are commonly employed methods to assess the stability of emulsions. Initially, the emulsion possesses 197 nm. After thermal stress for eight days, the sample showed a slight increase in the mean droplet size to 243 nm, as shown in
Table 6. However, the radius of the droplets was still within the range of 20–500 nm, which corresponds to the size of a nanoemulsion [
57]. The stability of the emulsion is strongly related to the droplet size distribution. Generally, the smaller the droplet size, the more stable the W/O nanoemulsions and the longer their shelf life. The small droplet size of nanoemulsions stabilizes them against gravitational separation and flocculation. Large droplet size can promote Ostwald ripening, which increases droplet size leading to coalescence and creaming [
58]. The zeta potential value obtained for the nanoemulsion on day zero was −56 ± 2 mV. Alternatively, the value presented for the nanoemulsion subjected to thermal stress for eight days was −64 ± 1 mV (
Table 6). Both values reflect the stability of the emulsion even when subjected to changes in temperature. The zeta potential shows the force of repulsion between adjacent equally charged droplets and consequently proves the important role of the droplets’ surface layer in the stabilization process. High values of zeta potential (>ǀ30ǀ mV) indicate resistance to particle aggregation and, therefore, greater stability [
59].