Sulforaphane-Loaded Ultradeformable Vesicles as A Potential Natural Nanomedicine for the Treatment of Skin Cancer Diseases.

Sulforaphane is a multi-action drug and its anticancer activity is the reason for the continuous growth of attention being paid to this drug. Sulforaphane shows an in vitro antiproliferative activity against melanoma and other skin cancer diseases. Unfortunately, this natural compound cannot be applied in free form on the skin due to its poor percutaneous permeation determined by its physico-chemical characteristics. The aim of this investigation was to evaluate ethosomes® and transfersomes® as ultradeformable vesicular carriers for the percutaneous delivery of sulforaphane to be used for the treatment of skin cancer diseases. The physico-chemical features of the ultradeformable vesicles were evaluated. Namely, ethosomes® and transfersomes® had mean sizes <400 nm and a polydispersity index close to 0. The stability studies demonstrated that the most suitable ultradeformable vesicles to be used as topical carriers of sulforaphane were ethosomes® made up of ethanol 40% (w/v) and phospholipon 90G 2% (w/v). In particular, in vitro studies of percutaneous permeation through human stratum corneum and epidermis membranes showed an increase of the percutaneous permeation of sulforaphane. The antiproliferative activity of sulforaphane-loaded ethosomes® was tested on SK-MEL 28 and improved anticancer activity was observed in comparison with the free drug.


Introduction
Sulforaphane (1-isothiocyanate-(4R)-(methylsulfinyl)-butane) is a natural dietary isothiocyanate. It is an enzymatic product obtained from the reaction between myrosinase and glucopharanin, a 4-methylsulfinylbutyl glucosinolate contained in cruciferous vegetables, e.g., broccoli, brussel sprouts, and cabbage. The attention given to this natural compound is constantly growing due to its multiple The aim of this work was to investigate ethosomes ® and transfersomes ® as potential carriers for the percutaneous delivery of sulforaphane, in order to propose an innovative clinical therapeutic treatment for skin cancer diseases. The ethosomes ® and transfersomes ® were characterized from a physico-chemical and technological point of view (i.e., mean size, size distribution, zeta potential, storage stability, entrapment efficacy, and drug release were evaluated), and ex vivo permeation through the human stratum corneum and epidermis membrane was also investigated.

Preparation of Ethosomes ®
Ethosomes ® were made up of PL90G, ethanol, and water in the percentages reported in Table 1. The ethosome ® colloidal suspensions were prepared as previously described [16]. Briefly, PL90G was poured into a hermetically-sealed Pyrex ® glass vial and solubilized with a suitable amount of ethanol under mixing at 700 rpm with a magnetic stirrer (Midi MR1 Digital IkamagR; IKA-WERKE GMBH and Co., Staufen, Germany). Double distilled water was slowly added at 25.0 ± 0.1 • C. The obtained ethosomes ® were then homogenized at 15,000 rpm for 1 min using an Ultra-Turrax T 25 equipped with an S25 N-8G homogenizing probe (IKA-WERKE) and then left at room temperature for 30 min under continuous stirring (Orbital Shaker KS 130 Control, IKA-WERKE). To achieve sulforaphane-loaded ethosomes ® , the drug (55 µg/mL) was dissolved in ethanol during the preparation procedure of the various formulations.

Preparation of Transfersomes ®
Transfersomes ® were made up of PL90G, SC, and water (Table 1) and were prepared using the thin layer evaporation method as previously reported [24]. Briefly, PL90G and SC were dissolved in a round-bottomed flask with absolute ethanol. The flask containing lipid solution was connected to a rotary evaporator (Rotavapor ® R-210, Büchi-Italia, Milan, Italy) at 60 • C and under a slow nitrogen flux, until all traces of solvent had evaporated and a thin-layer lipid film had formed. An ethanol/water (7:93 v/v) mixture (final volume equal to 6 mL) was added to the lipid film and a vesicular colloidal suspension was obtained by mixing at 700 rpm for 15 min using an orbital Shaker (KS 130 Control, IKA-WERKE). The carriers were left at 40 • C for 2 h in order to stabilize the obtained suspension. Transfersomes ® were then subjected to the extrusion procedure through polycarbonate membranes (400 and 200 nm), as reported in a previous investigation [25]. To achieve sulforaphane-loaded transfersomes ® , the drug (55 µg/mL) was dissolved in ethanol during the preparation procedure of the various formulations.

Physico-Chemical Characterization of Vesicle Formulations
The mean size, size distribution, and z-potential were evaluated by a Zetasizer Nano ZS (Malvern Instruments Ltd., Worchestershire, United Kingdom), following a 1:50 dilution of the samples. Zetasizer Nano is a dynamic light-scattering spectrophotometer (DLS) and a third-order cumulant fitting correlation function was used for sample analysis. This instrument was equipped with a 4.5 mW laser diode operating at 670 nm, which was the light source; the back-scattered photons were detected at 173 • . Before starting the analyses, the medium refractive index (1.330), medium viscosity (1.0 mPa × s), and dielectric constant (80.4) were set. The samples were placed in quartz cuvettes to be analyzed [26].
To evaluate the stability of the prepared vesicular carriers, Turbiscan Lab ® Expert, equipped with a Turbiscan Lab Cooler, was used. The photon transmitted (T) and backscattered (BS) through the samples, placed in a cylindrical glass tube, was detected [27]. The kinetic stability of the samples was evaluated from the data, obtained by TurbiSoft software (Formulaction, L'Union, France). Measurements were taken for 1 h at room temperature (24 ± 1 • C).

Entrapment Efficacy of Ethosomes ® and Transfersomes ®
The amount of entrapped sulforaphane was separated from the untrapped aliquot by using ultracentrifugation. The colloidal vesicles were poured into polycarbonate tubes and then centrifuged at 90,000× g for 1 h at 4 • C using an Avanti 30 Centrifuge (Beckman, Fullerton, CA, USA) equipped with a fixed angle rotor Beckman mod. F1202. The supernatant and the pellet were divided and separately analyzed using HPLC (A Jasco PU-1580 intelligent HPLC pump, Tokyo, Japan) (see Section 2.9).
To break the pellets, 4 mL of ethanol was used. Possible interference from vesicle components was avoided using empty ethosomes ® and transfersomes ® as blanks. To quantify the amount of sulforaphane entrapped in vesicular systems, the difference between the drug used during the preparation and the non-encapsulated drug was calculated. The following equation was used to determine the encapsulation efficiency (EE%) of sulforaphane in deformable vesicles: where De is the amount (mg) of entrapped sulforaphane and Da is the sulforaphane amount (mg) used to prepare ethosomal and transfersomal formulations. The reported results represent the average value of five different formulations ± standard deviation.

Sulforaphane Release Profiles
Dynamic skin permeation systems (Laboratory Glass Apparatus, Inc. 1200 Fourth Street Berkeley, CA 94710, USA) were used to evaluate the vesicles' ability to release the sulforaphane and to permeate through the skin. They were characterized by a surface area of 0.75 cm 2 and a nominal receiving volume of 4.75 mL. They were composed of two compartments, i.e., the donor, filled with vesicular colloidal suspensions (200 µL), and the receptor, filled with an ethanol/water mixture (20:80 v/v). For the release studies, a synthetic cellulose membrane (molecular cut-off weight of 10,000 Da) was interposed between the two compartments. Throughout investigations, sink conditions were maintained, the fluid receptor was constantly stirred with a small magnetic stirring bar, and the temperature was maintained at 32.0 ± 0.5 • C by means of a circulating water bath [28]. The duration of experiments was 24 h and at specific time intervals, 500 µL of the receptor phase was collected for HPLC (see Section 2.9) determination of the released sulforaphane. The amount of withdrawn receptor solution was replaced with the same amount of fresh solution. Experiments were carried out in triplicate and the results were the average of three different experiments ± standard deviations.

Percutaneous Permeation of Sulforaphane-Loaded Deformable Vesicles
Dynamic skin permeation systems were used to evaluate the in vitro percutaneous permeation of sulforaphane in free form and entrapped in deformable vesicles (ethosomes ® and transfersomes ® ) through human stratum corneum and viable epidermis (SCE) membranes. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Research Ethics Committee of the University "Magna Graecia" of Catanzaro (Italy). SCE membranes were prepared as described by Paolino [29], using fresh abdominal human skin obtained from the plastic reduction surgery of healthy adults (mean age 30 ± 4 years). Briefly, the subcutaneous fat was removed by a scalpel and skin samples were put in distilled water for two minutes at 60 ± 1 • C to obtain SCE membranes, which were peeled off from dermis. The obtained sheets (average thickness~40 µm) [30] were stored at 4 • C until being used. As described for release studies (Section 2.5), the receptor compartment was filled with an ethanol/water solution (20:80 v/v) and stirred at 600 rpm, while the donor compartment was filled with 200 µL of sample. In this case, the SCE membrane was interposed between the two compartments, after its hydration in isotonic sterile saline solution. At prefixed intervals, for all 24 h of the experiments, aliquots of the receptor phase were withdrawn and immediately analyzed by HPLC (see Section 2.9) to determine the amount of permeated sulforaphane. Additionally, in this case, the withdrawn volume was replaced with the same amount of fresh solution and the temperature was maintained at 32.0 ± 0.5 • C by means of a circulating water bath [28].
The skin samples obtained from skin permeation system studies were homogenized for 5 min in the presence of 1 mL of methanol and sonicated (Sonopolus GH70, Bandelin-Electonic, Berlin, Germany) at 50 cycles/s. A following centrifugation procedure was carried out on the tissue suspension for 10 min at 7000 rpm by Megafuge 1.0 Centrifuge (Heraeus Sepatech, Osterode/Harz, Germany). The obtained supernatant was finally analyzed by HPLC to determine the amount of drug trapped in the skin samples.
Experiments were carried out in triplicate and the results were the average of three different experiments ± standard deviations.

Cell Cultures
Plastic culture dishes (100 mm × 20 mm) were used for the incubation of Melanoma cell line SK-MEL-28 in a Guaire ® TS Autoflow Codue Water-Jacketed incubator at 37 • C (5% CO 2 ). EMEM, containing penicillin (100 UI/mL), streptomycin (100 µg/mL), amphotericin B (250 µg/mL), and FBS (10% v/v), was used as a medium for cell line growth and replaced with fresh EMEM every 48 h. Trypsin (2 mL) was used to remove the cells adhered to the plate as soon as 80% confluence occurred. Then, 4 mL of the culture medium was placed into a centrifuge tube and centrifuged (1000 rpm) at room temperature for 10 min with an Eppendorf Centrifuge 5810. Finally, before in vitro experiments, fresh EMEM medium was used to resuspend the pellet.

Evaluation of In Vitro Anticancer Activity
The cells were plated in 96-well culture dishes at a density of 10,000 cells/0.2 mL in triplicate, and were then maintained at 37 • C for 24 h before the start of the experiment. After this incubation time, fresh EMEM medium containing different concentrations of free and entrapped sulforaphane was added to the plates with cells in order to replace the culture medium, followed by reincubation for 24, 48, or 72 h. As a control, we used eight wells for each plate with untreated cells. After treatment of the cells with the different samples, 10 µL of MTT (5 mg/mL dissolved in PBS solution) was placed in each well; after 3 h of incubation, supernatants were removed and (200 µL) dimethyl sulfoxide/ethanol solution (1:1 v/v) was added to each well to solubilize the colored formazan crystals. The plates were gently shaken at 230 rpm (IKA ® KS 130 Control, IKA ® WERKE GMBH & Co, Staufen, Germany) for 20 min. The ELISA microplate reader (BIO RAD, xMark™ Microplate Absorbance Spectrophotometer) at λ abs 570 nm and λ em 670 nm was used to study the absorbance values of all the analyzed samples. The percentage of cell viability was calculated according to the following equation: where AbsT is the absorbance of treated cells and AbsC is the absorbance of control (untreated) cells.
Cell viability values were the average of three different experiments ± standard deviation.

Statistical Analysis
Statistical analysis of all experiments was performed by one-way ANOVA. A posteriori Bonferroni t-test was carried out to check the ANOVA test. A p value <0.05 was considered statistically significant. Values are reported as the average ± standard deviation.

Physico-Chemical and Technological Characterization of Ethosomes ® and Transfersomes ®
The design and development of a drug delivery system for an innovative purpose require a careful and in-depth investigation regarding the chemical-physical and technological features. Suitable features, such as the mean size, polydispersity index, and zeta-potential, are the basis of pre-formulation studies for the possible development of a new pharmaceutical preparation. In particular, a small vesicle size and negative superficial charge are generally required for topical drug delivery systems selected for therapeutic treatments of skin disorders [32]. For these reasons, deformable vesicles were submitted to light scattering analysis so as to choose the most suitable formulation to be tested in vitro. As shown in Table 2, ethosomes ® (formulations A-I) and transfersomes ® (formulation J) displayed a narrow particle size distribution (<300 nm). As previously demonstrated and reported [16,18], the mean size of ethosomes ® is influenced by the amount of ingredients, i.e., the mean size of ethosomes ® decreases when the ethanol amount increases and soybean phosphatidylcoline amount decreases. In Pharmaceutics 2020, 12, 6 7 of 13 our case, the dimensions of ethosomes ® remained in a rather narrow range, probably because the effect of ethanol was counterbalanced by the lecithin effect. All the investigated ethosomal and transfersomal formulations are characterized by negative zeta potential values that are a requirement for topical application and storage stability [16,33]. As shown in Table 2, four formulations (B, D, E, and J) are characterized by suitable zeta potential and polydispersity index values. In fact, a strongly negative zeta potential value and a low polydispersity index should bring no aggregation phenomena, and should thus guarantee a good colloidal stability. Moreover, Turbiscan Lab ® Expert analysis [27] was carried out to confirm the DLS results, by measuring the transmission and backscattering profiles as functions of time and sample height. As shown in Figure 1b According to the data obtained through dynamic light scattering and Turbiscan Lab ® Expert analysis, the formulations B, D, E, and J were chosen for sulforaphane delivery and subsequent characterization studies.
Drug entrapment within a vesicular carrier is an important parameter to be investigated to really evaluate the topical delivery potentiality of the systems. The entrapment efficiency of sulforaphane within chosen formulations was evaluated. In Table 3, entrapment efficiency values were reported. The chosen formulations were characterized by high drug encapsulation efficiency values, particularly formulations E and J (87.54% and 86.20%, respectively). formulations are overlapping Figure 1b,c (data not shown). This finding showed that no creaming or sedimentation occurred, thus confirming the stability of the abovementioned nanosystems. The formulations of ethosomes ® A, C, F, G, H, and I showed significant variations in their backscattering profiles, thus evidencing a certain destabilization of the colloidal suspension. In Figure 1a, the Δ transmission and Δ back scattering profiles of formulation A are shown as examples of instable profiles.  The presence of drugs in vesicular systems can influence and modify the physico-chemical characteristics and stability profile of them. In fact, as shown in Table 3, sulforaphane induced a significant increase in the mean size and polydispersity index values of formulations B and D, thus showing a destabilization phenomenon, as evidenced by Turbiscan Lab ® Expert analysis.
In particular, Figure 2 shows that the profiles of the stability kinetic value of sulforaphane-loaded formulation B and D did not fall within a narrow range of the TSI (Turbiscan Stability Index), thus further supporting the presence of instability phenomena. Instead, E and J samples showed a suitable tolerance with respect to the presence of the drug, by maintaining an unchanged mean size, polydispersity index, and stability profile. Probably, the presence of high ethanol concentrations increases the sulforaphane solubility in the polar phase of the vesicular colloidal formulations. The zeta potential values of the four drug-loaded deformable vesicle formulations were not affected (see Table 3).

Evaluation of Sulforaphane Release and the Percutaneous Permeation Profile
The previously described results allowed the formulations with the best physico-chemical features and the most suitable stability profiles with and without sulforaphane to be selected. The ability of E and J formulations to release the drug was also investigated by using dynamic skin permeation systems, by determining the amount of released sulforaphane as a function of time. As shown in Figure 3, formulation E released~67% of the entrapped drug during 24 h. The release profile of sulforaphane from formulation E was characterized by a biphasic trend. That is, the amount of sulforaphane entrapped in the outer surface bilayer was quickly released during the first five hours. After this initial rapid release, sulforaphane was gradually and steadily released over time, probably depending on the chemical characteristics of sulforaphane, which allowed a certain affinity with the vesicular structure. On the other hand, the sulforaphane release profile of transfersomes ® (Formulation J) was significantly lower (p value < 0.005) than the release from formulation E. In detail, this result was less than 10%. This poor release was probably due to the great affinity of sulforaphane to the chemical components Pharmaceutics 2020, 12, 6 9 of 13 (phospholipids and sodium cholate) of the transfersomes ® , as already demonstrated by Celia et al. [24] for linoleic acid-loaded transfersomes ® .
showing a destabilization phenomenon, as evidenced by Turbiscan Lab Expert analysis.
In particular, Figure 2 shows that the profiles of the stability kinetic value of sulforaphaneloaded formulation B and D did not fall within a narrow range of the TSI (Turbiscan Stability Index), thus further supporting the presence of instability phenomena. Instead, E and J samples showed a suitable tolerance with respect to the presence of the drug, by maintaining an unchanged mean size, polydispersity index, and stability profile. Probably, the presence of high ethanol concentrations increases the sulforaphane solubility in the polar phase of the vesicular colloidal formulations. The zeta potential values of the four drug-loaded deformable vesicle formulations were not affected (see Table 3).

Evaluation of Sulforaphane Release and the Percutaneous Permeation Profile
The previously described results allowed the formulations with the best physico-chemical features and the most suitable stability profiles with and without sulforaphane to be selected. The ability of E and J formulations to release the drug was also investigated by using dynamic skin permeation systems, by determining the amount of released sulforaphane as a function of time. As shown in Figure 3, formulation E released ~67% of the entrapped drug during 24 h. The release profile of sulforaphane from formulation E was characterized by a biphasic trend. That is, the amount of sulforaphane entrapped in the outer surface bilayer was quickly released during the first five hours. After this initial rapid release, sulforaphane was gradually and steadily released over time, probably depending on the chemical characteristics of sulforaphane, which allowed a certain affinity with the vesicular structure. On the other hand, the sulforaphane release profile of transfersomes ® (Formulation J) was significantly lower (p value < 0.005) than the release from formulation E. In detail, this result was less than 10%. This poor release was probably due to the great affinity of sulforaphane to the chemical components (phospholipids and sodium cholate) of the transfersomes ® , as already demonstrated by Celia et al. [24] for linoleic acid-loaded transfersomes ® . The two formulations (E and J) were further investigated to evaluate their percutaneous permeation through human SCE membranes, in comparison with a water-ethanol solution of sulforaphane. Figure 4 shows that ethosomes ® were able to improve the percutaneous permeation of sulforaphane. In particular, the amount of permeated drug was over 90% when delivered by ethosomes ® . In the case of sulforaphane-loaded transfersomes ® , the amount of drug detected in the receptor compartment of dynamic skin permeation systems was very low (p value <0.001) with respect to sulforaphane-loaded ethosomes ® , in agreement with the in vitro release studies, and similar to the amount permeated from the hydroalcoholic sulforaphane solution.
The homogenization of skin samples derived from percutaneous permeation studies and the subsequent HPLC analysis confirmed that sulforaphane-loaded transfersomes accumulated in the SCE membrane, with a concentration of sulforaphane equal to 34.14 μg/mL (72 ± 5.8% of the delivered drug). The two formulations (E and J) were further investigated to evaluate their percutaneous permeation through human SCE membranes, in comparison with a water-ethanol solution of sulforaphane. Figure 4 shows that ethosomes ® were able to improve the percutaneous permeation of sulforaphane. In particular, the amount of permeated drug was over 90% when delivered by ethosomes ® . In the case of sulforaphane-loaded transfersomes ® , the amount of drug detected in the receptor compartment of dynamic skin permeation systems was very low (p value <0.001) with respect to sulforaphane-loaded ethosomes ® , in agreement with the in vitro release studies, and similar to the amount permeated from the hydroalcoholic sulforaphane solution. Pharmaceutics 2019, 11, x FOR PEER REVIEW 10 of 14 These findings supported the hypothesis of the interaction and affinity of the drug with the components of the transfersomal formulation. Moreover, the transfersomes ® could establish an interaction with the outer structures of the skin, with consequent vesicle accumulation in the corneocytes. This strong interaction could allow systems to be released for more than 24 h. In fact, Celia et al. [24] previously evidenced, following a confocal microscopy study combined with dynamic skin permeation experiments, the interaction and hence the release of transfersomes ® from corneocytes after a 48 h incubation. In particular, the confocal microscopy experiments [24] showed that transfersomes ® were accumulated in the SCE membrane for a prolonged period of time and then released into the receptor compartment of dynamic skin permeation systems. Based on these findings, our investigation was further developed by choosing, for the following studies, only the ethosome ® formulation E.

In Vitro Anticancer Activity of Sulforaphane and Sulforaphane-Loaded Ethosomes ®
The anticancer properties of sulforaphane-loaded ethosomes ® were investigated on human SK-MEL 28 malignant melanoma cells; this cell line was chosen as a model of high proliferative cells. The cytotoxic effects of sulforaphane-loaded ethosomes ® were evaluated both as a function of the drug concentration (10, 20, and 50 μM) and the incubation time (24,48, and 72 h), in comparison with a simple drug solution.
The anticancer activity of sulforaphane has been well-described by many studies and the involved mechanisms of action seem to be multiple. Arcidiacono et al. [13], for example, demonstrated that the inhibition of A375 and 501MEL melanoma cells is correlated with reduced AKT phosphorylation induced by sulforaphane. Other studies have shown the proapoptotic effects of sulforaphane on human melanoma cells by p53 and p38 pathways. Moreover, studies have demonstrated that drugs cause G1/S and G2/M cell cycle arrest by altering the levels of cyclin A [13,34,35]. Therefore, the anticancer activity of sulforaphane is due to intracellular mechanisms. Our in vitro results ( Figure 5) showed that the ethosome ® formulation provided the best anticancer activity on SK-MEL 28 after 24 h and at all tested concentrations compared with the free drug. This trend was also confirmed after 48 and 72 h incubation. The ability of ethosomes ® to increase the anticancer activity of the drug is probably due to their fusion with the outer cell membranes, thus allowing cell permeation and drug release directly into the cytoplasm [36]. The homogenization of skin samples derived from percutaneous permeation studies and the subsequent HPLC analysis confirmed that sulforaphane-loaded transfersomes accumulated in the SCE membrane, with a concentration of sulforaphane equal to 34.14 µg/mL (72 ± 5.8% of the delivered drug).
These findings supported the hypothesis of the interaction and affinity of the drug with the components of the transfersomal formulation. Moreover, the transfersomes ® could establish an interaction with the outer structures of the skin, with consequent vesicle accumulation in the corneocytes. This strong interaction could allow systems to be released for more than 24 h. In fact, Celia et al. [24] previously evidenced, following a confocal microscopy study combined with dynamic skin permeation experiments, the interaction and hence the release of transfersomes ® from corneocytes after a 48 h incubation. In particular, the confocal microscopy experiments [24] showed that transfersomes ® were accumulated in the SCE membrane for a prolonged period of time and then released into the receptor compartment of dynamic skin permeation systems. Based on these findings, our investigation was further developed by choosing, for the following studies, only the ethosome ® formulation E.

In Vitro Anticancer Activity of Sulforaphane and Sulforaphane-Loaded Ethosomes ®
The anticancer properties of sulforaphane-loaded ethosomes ® were investigated on human SK-MEL 28 malignant melanoma cells; this cell line was chosen as a model of high proliferative cells. The cytotoxic effects of sulforaphane-loaded ethosomes ® were evaluated both as a function of the drug concentration (10, 20, and 50 µM) and the incubation time (24,48, and 72 h), in comparison with a simple drug solution.
The anticancer activity of sulforaphane has been well-described by many studies and the involved mechanisms of action seem to be multiple. Arcidiacono et al. [13], for example, demonstrated that the inhibition of A375 and 501MEL melanoma cells is correlated with reduced AKT phosphorylation induced by sulforaphane. Other studies have shown the proapoptotic effects of sulforaphane on human melanoma cells by p53 and p38 pathways. Moreover, studies have demonstrated that drugs cause G1/S and G2/M cell cycle arrest by altering the levels of cyclin A [13,34,35]. Therefore, the anticancer activity of sulforaphane is due to intracellular mechanisms. Our in vitro results ( Figure 5) showed that the ethosome ® formulation provided the best anticancer activity on SK-MEL 28 after 24 h and at all tested concentrations compared with the free drug. This trend was also confirmed after 48 and 72 h incubation. The ability of ethosomes ® to increase the anticancer activity of the drug is probably due to their fusion with the outer cell membranes, thus allowing cell permeation and drug release directly into the cytoplasm [36].

Conclusions
The obtained data highlighted the different ability of ethosomes ® and transfersomes ® to effectively deliver sulforaphane through the skin. The physico-chemical and technological characterization showed that stable sulforaphane-loaded ethosomes ® and transfersomes ® were obtained, but only one ethosomal preparation showed a suitable ability to contain and release the drug and to successfully permeate the skin. In particular, the different modes of action between ethosomes ® and transfersomes ® seemed to suggest differences in their efficiency as therapeutic agents for the treatment of melanoma. In fact, ethosomes ® elicited an increase of the in vitro percutaneous permeation of sulforaphane compared with transferosomal formulations and drug solutions. This effect is probably due to the presence of ethanol in the composition of ethosomes ® , which could promote interaction between carriers and lipids of the stratum corneum. These findings are very encouraging and suggest that ethosomes ® could be effective carriers for the topical administration of sulforaphane and a real opportunity for the development of an innovative suitable therapeutic strategy for skin cancer disease treatment, such as that for melanoma.  The results were normalized as a function of the blank formulation E cytotoxicity. Results are the mean of three different experiments ± standard deviation. The data obtained for SFN-Etho are statistically significant with respect to the same concentration of free SFN (* p < 0.05; ** p < 0.001).

Conclusions
The obtained data highlighted the different ability of ethosomes ® and transfersomes ® to effectively deliver sulforaphane through the skin. The physico-chemical and technological characterization showed that stable sulforaphane-loaded ethosomes ® and transfersomes ® were obtained, but only one ethosomal preparation showed a suitable ability to contain and release the drug and to successfully permeate the skin. In particular, the different modes of action between ethosomes ® and transfersomes ® seemed to suggest differences in their efficiency as therapeutic agents for the treatment of melanoma. In fact, ethosomes ® elicited an increase of the in vitro percutaneous permeation of sulforaphane compared with transferosomal formulations and drug solutions. This effect is probably due to the presence of ethanol in the composition of ethosomes ® , which could promote interaction between carriers and lipids of the stratum corneum. These findings are very encouraging and suggest that ethosomes ® could be effective carriers for the topical administration of sulforaphane and a real opportunity for the development of an innovative suitable therapeutic strategy for skin cancer disease treatment, such as that for melanoma.