Phenolic Compounds Contribution to Portuguese Propolis Anti-Melanoma Activity

Melanoma is the deadliest type of skin cancer, with about 61,000 deaths annually worldwide. Late diagnosis increases mortality rates due to melanoma’s capacity to metastasise rapidly and patients’ resistance to the available conventional therapies. Consequently, the interest in natural products as a strategy for drug discovery has been emerging. Propolis, a natural product produced by bees, has several biological properties, including anticancer effects. Propolis from Gerês is one of the most studied Portuguese propolis. Our group has previously demonstrated that an ethanol extract of Gerês propolis collected in 2018 (G18.EE) and its fractions (n-hexane, ethyl acetate, and n-butanol) decrease melanoma cell viability. Out of all the fractions, G18.EE-n-BuOH showed the highest potential as a melanoma pharmacological therapy. Thus, in this work, G18.EE-n-BuOH was fractioned into 17 subfractions whose effect was evaluated in A375 BRAF-mutated melanoma cells. The subfractions with the highest cytotoxic activity were analysed by UPLC-DAD-ESI/MSn in an attempt to understand which phenolic compounds could account for the anti-melanoma activity. The compounds identified are typical of the Gerês propolis, and some of them have already been linked with antitumor effectiveness. These results reaffirm that propolis compounds can be a source of new drugs and the isolation of compounds could allow its use in traditional medicine.


Introduction
Cancer is currently the second-leading cause of death worldwide, responsible for around 10 million deaths in 2020 [1]. Melanoma is the deadliest skin cancer type, causing about 61,000 deaths per year around the world [2,3]. If diagnosed at an early stage, it can be removed by surgery, improving the patient's survival rate. However, late diagnosis increases mortality rates due to its high capacity to metastasize, making treatment much more challenging [4,5]. One of the most recent options for melanoma treatment is targeted therapy, being the combination of BRAF (V-RAF murine sarcoma viral oncogene homolog B) and MEK (mitogen-activated protein kinase kinase) inhibitors used to enhance survival in BRAF-mutated patients [6][7][8]. The melanoma genome is highly mutated, with the most common mutation associated with the BRAF gene, a part of the mitogen-activated protein kinase (MAPK) signalling pathway [9][10][11]. BRAF mutation over-activates this pathway, leading to tumour aggressiveness and progression [12,13]. Approximately 40% to 60% of  1. Effect of G18.EE and G18.EE-n-BuOH on total cell biomass of A375 melanoma cell. Cells were treated with a range of concentrations (10 to 50 µg/mL) of G18.EE and G18.EE-n-BuOH for 72 h to determine the IC50 concentrations. Data were normalized for total biomass. Results are presented as mean ± SD (n ≥ 3).

Subfractionation of G18.EE-n-BuOH Fraction
The G18.EE was partitioned into four fractions (G18.EE-EtOAc, G18.EE-n-hexane, G18.EE-n-BuOH, and G18.EE-water) through a liquid-liquid partition. The G18.EE-n-BuOH was further fractionated in a silica gel column chromatography that separates the compounds based on their polarity, in order to obtain subfractions with less complex chemical composition. This process results in 17 subfractions ( Figure 2).  .EE-n-BuOH on total cell biomass of A375 melanoma cell. Cells were treated with a range of concentrations (10 to 50 µg/mL) of G18.EE and G18.EE-n-BuOH for 72 h to determine the IC 50 concentrations. Data were normalized for total biomass. Results are presented as mean ± SD (n ≥ 3).

Subfractionation of G18.EE-n-BuOH Fraction
The G18.EE was partitioned into four fractions (G18.EE-EtOAc, G18.EE-n-hexane, G18.EE-n-BuOH, and G18.EE-water) through a liquid-liquid partition. The G18.EE-n-BuOH was further fractionated in a silica gel column chromatography that separates the compounds based on their polarity, in order to obtain subfractions with less complex chemical composition. This process results in 17 subfractions ( Figure 2).

Subfractionation of G18.EE-n-BuOH Fraction
The G18.EE was partitioned into four fractions (G18.EE-EtOAc, G18.EE-n-hexane, G18.EE-n-BuOH, and G18.EE-water) through a liquid-liquid partition. The G18.EE-n-BuOH was further fractionated in a silica gel column chromatography that separates the compounds based on their polarity, in order to obtain subfractions with less complex chemical composition. This process results in 17 subfractions ( Figure 2).

Cytotoxic Effect of G18.EE-n-BuOH Subfractions
The cytotoxic capacity of all subfractions was tested in the A375 melanoma cell line at 25 µg/mL concentration, corresponding to an intermediate concentration between IC 50 obtained for both G18.EE and G18.EE-n-BuOH (Table 1). The subfractions exhibited different effects on A375 cell viability ( Figure 3) The most active subfractions were E, G, H, I, J, and K, while subfractions C, D, and F demonstrated behavior similar to the fraction they were obtained from (G18.EE-n-BuOH). The remaining subfractions had a less noticeable effect as they are less cytotoxic than the upstream fraction.

Cytotoxic Effect of G18.EE-n-BuOH Subfractions
The cytotoxic capacity of all subfractions was tested in the A375 melanoma cell line at 25 μg/mL concentration, corresponding to an intermediate concentration between IC50 obtained for both G18.EE and G18.EE-n-BuOH (Table 1). The subfractions exhibited different effects on A375 cell viability ( Figure 3) The most active subfractions were E, G, H, I, J, and K, while subfractions C, D, and F demonstrated behavior similar to the fraction they were obtained from (G18.EE-n-BuOH). The remaining subfractions had a less noticeable effect as they are less cytotoxic than the upstream fraction. The results are presented as mean ± SD (n ≥ 3). The statistical differences are in regard to the control, where "ns" means not significant and **** p < 0.0001.

Chemical Analysis of the Subfractions
The phenolic compounds of subfractions E, G, H, I, J, and K were analysed by UPLC-DAD-ESI/MS n as they demonstrate to be the most cytotoxic subfractions against the A375 melanoma cell line. Figure 4 shows the chromatographic profiles at 280 nm for G18.EE-n-BuOH and subfractions E, I, and K, and the phenolic composition is detailed in Table 2. Subfractions G, H, and J had chromatographic peaks not defined, probably due to the fact that phenolic compounds are present in very low amounts, not allowing the detection and identification of this type of compound in these subfractions. The results are presented as mean ± SD (n ≥ 3). The statistical differences are in regard to the control, where "ns" means not significant and **** p < 0.0001.

Chemical Analysis of the Subfractions
The phenolic compounds of subfractions E, G, H, I, J, and K were analysed by UPLC-DAD-ESI/MS n as they demonstrate to be the most cytotoxic subfractions against the A375 melanoma cell line. Figure 4 shows the chromatographic profiles at 280 nm for G18.EE-n-BuOH and subfractions E, I, and K, and the phenolic composition is detailed in Table 2. Subfractions G, H, and J had chromatographic peaks not defined, probably due to the fact that phenolic compounds are present in very low amounts, not allowing the detection and identification of this type of compound in these subfractions.   Table 2. Quercetin-dimethyl-ether ---+ "+" = compound detected; "-" = compound not detected.

Discussion
Propolis has proven antitumor activity against several types of cancer [42,[47][48][49][50][51][52] and, although scarce, some researchers demonstrated its anti-melanoma activity [43,53,54]. Some therapeutic options already exist for melanoma, but their effectiveness is still limited, mainly due to acquired resistance [55]. Thus, the investigation of propolis activity in this type of cancer is important due to the need for new treatments and sources of therapeutic agents [56].
The results mentioned above demonstrated that G18.EE and G18.EE-n-BuOH are capable of decreasing A375 cell viability (Table 1, Figure 1). Previous studies had already described the cytotoxic effect of propolis on melanoma. For example, a Chinese propolis ethanol extract had an anti-proliferative effect on A375 (IC 50 = 112 µg/mL) [57]. The cytotoxic effect of the water extract of propolis was also tested in this cell line, with an IC 50 value of 353 µg/mL [58]. These latest IC 50 values for propolis are much higher than the one obtained in this work for Gerês propolis against the A375 cell line (Table 1). An ethanol extract of propolis from Poland was demonstrated to decrease the viability of Me45 melanoma cells [59], as well as Dutch propolis isolated compounds (cinnamic acid derivatives, flavonoids and glycerol derivatives) in murine B16-BL6 melanoma [60]. The hydroalcoholic extracts (70%) of two types of Brazilian stingless bee propolis (Tubuna and Mandaçaia) have a cytotoxic effect in SK-MEL-28 (99 and 35 µg/mL reducing cell number by 50%, for Tubuna and Mandaçaia, respectively) [54]. The ethanol extract of Algerian propolis, as well as compounds isolated from green propolis (namely baccarin and p-coumaric acid), decreased murine melanoma tumour progression [61,62]. In general, Gerês propolis extract and its butanol fraction show lower IC 50 values (Table 1) than those described in the literature for propolis effects in melanoma, and therefore appear to be more active.
The anticancer effect of G18.EE and its fractions have previously been reported, proving, together with the results described in this work, the higher potential of Gerês propolis for cancer treatment. The cytotoxicity of Gerês propolis (G18.EE) and its four fractions: EtOAc, n-hexane, n-BuOH, and water were evaluated in renal cancer cell lines (A498, 786-O and Caki-2) and non-neoplastic renal cells (HK2). The EtOAc fraction was the most effective as it presented considerable cytotoxic activity (IC 50 from 0.16 to 30 µg/mL) and a high selectivity index [42]. The effect of this Gerês extract and its fractions was previously described in two BRAF-mutated melanoma cell lines. The highest cytotoxic activity was displayed by the n-BuOH fraction against both melanoma cell lines (IC 50 from 8.14 µg/mL to 11.22 µg/mL) [43].
As mentioned, propolis has a highly variable chemical composition, a major barrier to its use in the pharmaceutical industry and traditional medicine [33]. From this perspective, it is important to isolate and identify which compounds or mixtures of compounds allow G18.EE-n-BuOH to have this anti-melanoma activity. Since A375 melanoma cells seemed to be more sensitive to G18.EE-n-BuOH treatment (Table 1), G18.EE-n-BuOH was fractioned, having resulted in 17 subfractions (Figure 2). The cytotoxic effect of these subfractions was evaluated and some of the subfractions have a considerable impact on the A375 cell viability reduction (Figure 3). The most active fractions were E, G, H, I, J, and K, and in these most effective subfractions, compounds with anti-melanoma activity are expected to be present.
The pharmacological properties of propolis are often attributed to its phenolic compounds [35,63]. In light of this, the phenolic compounds of the most active subfractions were analysed by UPLC-DAD-ESI/MS n . Our research group has previously reported the chemical composition of G18.EE and its fractions [43]. In general, the compounds found in the subfractions (Table 2) fit well with what has been described for other Gerês propolis samples [40,42,43]. Moreover, some of the compounds identified here are present in other European propolis samples [64]. Indeed, the phenolic compounds identified in G18.EE-n-BuOH include phenolic acids such as ellagic acid, and flavonoids such as kaempferol and galangin. Pinobanksin, pinocembrin, chrysin, and acacetin were identified too, belonging to the class of flavanones and dihydroflavonols. Flavones such apigenin were also found.
Other propolis samples with proven cytotoxic effect on melanoma had their composition described. A Chinese propolis sample was also evaluated against the A375 melanoma cell line [57]. For this sample, the IC 50 concentration was 112 µg/mL [57], whereas for our most active subfractions, at a concentration of 25 µg/mL (about 4.5 times less concentrated), we obtained a cell viability of approximately 5%. We can observe that the Chinese extract and our subfractions have some compounds in common, such as apigenin, pinobanksin, kaempferol, chrysin, galangin, CAPE, pinobanksin 3-O-acetate, and pinocembrin [57]. However, we found that, in the subfractions, acacetin and kaempferide are not present in this Chinese propolis sample, and these active compounds may explain the higher cytotoxicity of the subfractions. Moreover, in this Chinese propolis, the most representative compounds are chrysin, pinobanksin 3-O-acetate, and pinocembrin [57], the last two being present in our subfraction E. As subfraction E is much more active than Chinese propolis, other compounds present in this subfraction probably contribute to its higher activity. Kubina et al. [59] demonstrated the anti-melanoma effect of a Polish propolis extract sample, describing a cytotoxicity of 12.2 ± 1.77% (which suggests cell viability of approximately 88%) in the Me45 melanoma cell line, for a concentration of 25 µg/mL of extract, seeming to be less active in comparison to our subfractions ( Figure 3). The difference between the activities could be justified by the use of the A375 melanoma cell line in our work, while the authors used the Me45 melanoma cell line. Therefore, to confirm the higher cytotoxic activity of our subfractions, it would be necessary to test them in the Me45 cell line. Even so, analyzing the composition of the extract, we noticed that the phenolic compounds present in higher amounts in this extract [59] have not been identified in our subfractions (caffeic acid, benzoic acid, and ferulic acid). However, the compounds present in a lower proportion in this extract, such as kaempferol, galangin, and apigenin, were also identified in our subfractions, which again reinforces the hypothesis that these are the compounds responsible for the high cytotoxic effect presented by the subfractions.
In addition, CAPE, a phenolic acid present in propolis of various geographical origins, is associated with significant chemoprotective and anticancer properties [33,[65][66][67], including against melanoma cells [68]. Acacetin identified in subfractions I and K exhibited anticancer effects in diverse types of cancer such as breast cancer, lung cancer, prostate cancer, and colon cancer [69]. The subfraction K appears to be the one that most affects cell viability, so it may be interesting to look at compounds that are only present in this subfraction-apigenin and kaempferol. Apigenin is a flavonoid with documented anticancer activity. This property is associated with multiple biological processes, including induction of cell cycle arrest, activation of cell apoptosis and autophagy, inhibition of cell migration and invasion, and initiation of an immunological response [70,71]. Moreover, kaempferol present in subfraction K proved to have anticancer properties associated with a mechanism that induces cancer cell apoptosis [72]. Regarding all these results, the compounds identified are potential targets for developing new drugs against cancer. The possible synergism between these compounds and their mechanism of action should be further investigated.

Propolis Sample and Preparation of Gerês Propolis Extract
The propolis sample (designated as G18) was obtained in 2018 from an apiary close to the Cávado River, between the villages of Paradela and Sirvozelo, in Montalegre, Gerês, Portugal (41 • 45 041.62 N; 7 • 58 003.34 W). The G18 propolis extract was prepared with absolute ethanol (Fisher Chemical) to obtain G18.EE, as previously described by Freitas et al. [40]. The ethanol extract (EE) was then dried by solvent evaporation (62% yield [41]) and stored at 4 • C, in the dark, until further use.

Subfractionation of G18.EE-n-BuOH Fraction
The G18.EE-n-BuOH was further fractionated in silica gel column chromatography. Briefly, silica gel 60 (0.035-0.070 mm; Acrós Organics, Morris Plains, NJ, USA) diluted in n-hexane (Fisher Chemical, Waltham, MA, USA) was placed on a glass column. G18.EEn-BuOH (156.6 mg) diluted in ethanol was deposited on top of the column and eluted successively with n-hexane, n-hexane/EtOAc mixtures of increasing polarity, and finally with mixtures of EtOAc/MeOH also with increasing polarity. The 54 subfractions obtained were analysed by thin-layer chromatography (TLC) (TLC silica gel F254, Merk, Darmstadt, Germany). The developed plates were examined under UV light (254 nm and 366 nm), followed by spraying with anisaldehyde-H 2 SO 4 (prepared by adding 0.5 mL of anisaldehyde (Acrós Organics) in 10 mL of fluvial acetic acid (PanReac, Chicago, IL, USA), 85 mL of MeOH, and 5 mL of concentrated H 2 SO 4 (Fisher Chemical) [73] adapted), and heated for 5 min at 100 • C. The fractions were pooled according to similarities in their TLC-chromatographic profile, finally obtaining 17 main subfractions, which were assigned a code from A to Q. After solvent remotion in a Rotavapor under low pressure at 40 • C, followed by nitrogen flow, subfractions were stored at 4 • C, in the dark, until further assays and dissolved in DMSO to prepare stock solutions for further experiments.

Chemical Analysis of the Subfractions: UPLC-DAD-ESI/MS n
The subfractions E, G, H, I, J, and K were analysed by UPLC-DAD-ESI/MS n as described [40]. The UPLC-DAD-ESI/MS n analysis was performed on an Ultimate 3000 (Dionex Co., Sunnyvale, CA, USA) apparatus equipped with an ultimate 3000 Diode Array Detector (Dionex Co.) and coupled to a mass spectrometer LTQ XL Linear Ion Trap 2D. The chromatographic system consisted of a quaternary pump, an autosampler, a photodiodearray detector, and an automatic thermostatic column compartment. Analysis was run on a Hypersil Gold (Thermo Scientific, Waltham, MA, USA) C18 column (100 mm length; 2.1 mm i.d.; 1.9 µm particle diameter, end-capped) and its temperature was maintained at 30 • C. The mobile phase was composed of (A) 0.1% of formic acid (v/v) and acetonitrile (B). The solvent gradient started with 20% of solvent (B), reaching 40% at 25 min, 60% at 35 min, and 90% at 50 min, followed by the return to the initial conditions. The flow rate was 0.1 mL/min, and UV-Vis spectral data for all peaks were accumulated in the range of 200 to 500 nm, while the chromatographic profiles were recorded at 280 and 320 nm. The mass spectrometer used was a Thermo LTQ XL (Thermo Scientific) ion trap MS equipped with an ESI source. Control and data acquisition were carried out with the Thermo Xcalibur Qual Browser data system (Thermo Scientific). Nitrogen above 99% purity was used, and the gas pressure was 520 kPa (75 psi). The instrument was operated in negative-ion mode with ESI needle voltage set at 5.00 kV and an ESI capillary temperature of 275 • C. The full scan covered the mass range from m/z 100 to 2000. CID-MS/MS and MS n experiments were simultaneously acquired for precursor ions using helium as the collision gas with a collision energy of 25-35 arbitrary units.

Cell Lines and Culture Conditions
The in vitro assays were performed using a human melanoma cell line: A375 (BRAF V600E mutated). This human cell line was established from cutaneous malignant melanoma and was obtained from the American Type Culture Collection (ATCC ® CRL-1619TM) and kindly provided by Dra. Marta Viana-Pereira (Life and Health Research Institute, University of Minho, Braga, Portugal). The A375 cell line was grown in Dulbecco's modified Eagle's medium (DMEM, PAN-Biotech, Aidenbach, Germany), supplemented with 10% FBS (PAN-Biotech) and incubated at 37 • C in a humidified environment with 5% CO 2 .

Cell Viability Assay
To determine the effect of the G18.EE and its fraction, G18.EE-n-BuOH, on A375 cell viability, the SRB (TOX-6, Sigma-Aldrich, St. Louis, MO, USA) assay was used as described previously [43]. Briefly, the cells were plated at a concentration of 25 × 10 4 cells/mL and left overnight in a complete medium to adhere. On the following day, they were subjected to serum starvation for 2 h. After the addition of the treatments (G18.EE and G18.EE-n-BuOH) at different concentrations (10 to 50 µg/mL), the plates were left for 72 h at 37 • C and 5% CO 2 . DMSO (Honeywell) (control) final concentration was 0.1% in all the wells. Cells were fixed to the plate with 100 µL of TCA (10%, Sigma-Aldrich) for 1 h at 4 • C. The next step was the addition of SRB (0.4% SRB in 0.1% acetic acid), which was left to react for 30 min at room temperature (RT). To solubilise the dye, Tris base (PanReac AppliChem ITW Reagents) was added and the plates were incubated for 10 min at RT. The absorbance was measured at 490 nm (Thermo Scientific Varioskan Flash) and IC 50 values were calculated using GraphPad Software Version 8.0.
To determine the effect of the G18.EE-n-BuOH subfractions on A375 cell viability, the SRB assay was also used. The methodology was the same as described above, with only one concentration (25 µg/mL, intermediate value between IC 50 obtained for both G18.EE and G18.EE-n-BuOH) tested for all of the 17 subfractions (A to Q). DMSO was used as a control.

Statistical Analysis
Each assay was performed in triplicate and repeated at least three times independently (n ≥ 3), and results were expressed as mean ± standard deviation (SD). For IC 50 determination, GraphPad Prism 8.0 software was used for logarithmic transformation after applying a non-linear sigmoidal dose-response regression. One-way analysis of variance (ANOVA) was used for the comparison between different conditions. The threshold used for statistical significance was p < 0.05.

Conclusions
Melanoma is the deadliest skin cancer type, partially due to the therapeutic resistance development. Propolis is a highly biologically active natural product produced by bees presenting antitumor properties. The ethanol extract of propolis from Gerês collected in 2018 (G18.EE) and its n-BuOH fraction (G18.EE-n-BuOH) were effective in decreasing cell biomass of the BRAF-mutated A375 melanoma cell line in a dose-dependent manner. In a previous paper for our research group, the G18.EE-n-BuOH fraction had already shown to be the most effective in reducing the viability of A375 melanoma cells, compared to the other fractions (EtOAc, n-hexane and water fractions) [43]. Based on that, G18.EE-n-BuOH was fractionated. The cytotoxic activity of the obtained subfractions was also evaluated in melanoma and the phenolic compounds present in the most active subfractions were identified. This research has made it possible to identify which phenolic compounds, among the several found in propolis, could contribute to the Gerês propolis anti-melanoma activity and, therefore, deserve more research regarding the development of new drugs against cancer. Propolis has an extremely variable chemical composition, hence the identification of the active compounds would facilitate its use in traditional medicine. Nevertheless, further studies are necessary. For instance, it would be useful to test the compounds identified in an isolated form, test different combinations of compounds, or test the compounds in combination with standard drugs, to understand the synergisms and mechanisms of action. Another question that arises is the selectivity index of the subfractions, making it necessary, in the future, to test them on normal melanocytes and/or other normal cells to evaluate their selectivity towards cancer cells.
Even so, this study confirms that Portuguese propolis is a very promising product for application in several areas, and Gerês propolis emerges as a potential adjuvant or even an alternative to conventional therapies. Funding: This work has been funded by national funds through the Foundation for Science and Technology (FCT)-projects UIDB/04050/2020, UDBI/04033/2020, UIDB/50026/2020, UIDP/50026/2020, UIDB/50006/2020, UIDP/50006/2020, and by the project NORTE-01-0145-FEDER-000055, supported by the Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF). S.P.C. is recipient of an FCT grant (2020.05779.BD). In addition, FCT and FEDER-COMPETE-QREN-EU have been funded the CQ/UM (UID/QUI/00686/2021). The NMR spectrometer Bruker Avance III 400 is part of the National NMR Network and was purchased within the framework of the National Program for Scientific Re-equipment, contract REDE/1517/RMN/2005 with funds from POCI 2010 (FEDER) and FCT.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data are contained within the article.

Acknowledgments:
The authors thank the beekeeper Amadeu Fortunas, from Casa do Couto, 5470-362 Montalegre, Portugal, for kindly donating the propolis sample used in this work; and to Marta Viana-Pereira, from the Life and Health Research Institute, University of Minho, Braga, Portugal, for generously providing the A375 melanoma cell line.