Three healthy male albino Wistar rats weighing around (250±15g) were housed and acclimatized at the Laboratory Animal Research Unit at the Applied Science University. Each rat was individually housed in a cage and maintained under controlled conditions of temperature (20±3 °C), humidity (50±15%), and photoperiod cycles (12 light/12 h dark) with a conventional laboratory diet and unrestricted supply of drinking water. The Ethical approval was obtained from the Al-Ahliyya Amman University ethical committee (Decision no. AUP: AAU 1/2/2022-2023) which gained the approval of the Applied Science University.

Chromatographic separation and quantitative analysis of RP and VE was performed on a Thermo Finnigan Surveyor HPLC system (Germany). The column used was a Hypersil Thermo Electron Corporation, C18 250 × 4.6, mm, 5μm (Thermo Fisher Scientific/Germany). The mobile phase was composed of Acetonitrile: Methanol (70:30 v/v). The mobile phase was filtered through 0.45 µm nylon filter and degassed by an ultrasonic water bath (Model UCB 100, Spectralab). The flow rate was 1.0 mL/min and the column was maintained at ambient temperature. The injection volume of 10 µL was used, and detection wavelength was 220 nm.

Retinyl palmitate (RP) and vitamin E (VE) standard stock solutions were prepared by accurately weighing 10 mg of RP and 25 mg of VE and transferring them into 100 ml volumetric flasks. Then 70ml of mobile phase was added and sonicated for 15 minutes to solubilize RP and VE. The solutions were diluted with the mobile phase to give a final concentration of 0.1 mg/mL and 0.25 mg/mL for RP and VE respectively. Standard solutions were prepared by diluting stock solutions and injecting them into the HPLC system. The experiments were performed in triplicates and the average was calculated.

The solubility of RP and VE was measured in different types of oils to ensure the therapeutic concentration, and to choose the proper oily phase in the formulated emulsion. The solubility was tested in safflower oil, jojoba oil, coconut oil, olive oil, sesame oil, and liquid paraffin. Two mL of oil was added in a screw cap test tube and an excess of RP was added in each tube, shaken for few minutes on the vortex, then left to equilibrate in a water bath with a shaker at 25 rpm for 48 hours. Then, the samples were centrifuged at 3000 rpm and 0.5 ml of the supernatant was suitably diluted by the mobile phase and measured by the HPLC. The same procedure was repeated for VE. Measurements were taken in triplicate and the results were expressed as solubility in mg/g oil at ambient temperature.

Nanoemulsion (NE) was prepared using the suitable oily phase from the solubility study. Several formulations of nanoemulsion contain different ratios of oily phase, aqueous phase, and different ratios of the surfactant (Tween 20, Tween 60, Tween 80, Ceteareth 20, and cremophor RH40) and cosurfactant (ethanol and PEG 200) were prepared as shown in Table 1. The selection criteria of the best formula depended on the physical appearance (translucent), particle size (preferably less than 100 nm), and stability at room temperature.

The mixture of surfactant and co-surfactant (Smix)s was prepared by mixing them in a beaker at 300 rpm for 15–20 min. When the surfactant was very viscous like tween 80 or waxy like ceteareth 20, it was heated to 40 °C until suitable viscosity achieved. Then the cosurfactant was added while the surfactant cooled down. Twelve samples of (Smix)s were prepared as shown in Table 1.

The NE was prepared by high energy method (sonication). The ratio of the oily phase, aqueous phase, and the (Smix)s was decided based on preliminary experiments. All prepared (Smix)s were tried in the formulation of NE that contained fixed portions of oil and aqueous phases (15–30% of oil and 40–65% of water). Ten formulae were prepared as shown in Table 2. The oil phase, Smix, and the aqueous phase were accurately weighed, then Smix was mixed with the oily phase, and the aqueous phase was added gradually with continuous stirring using a vortex mixer for a few minutes. The probe sonicator was inserted in the primary emulsion and turned on at a frequency of 3.0 per second. When the mixture turned translucent, the emulsion was formed. The prepared emulsions were evaluated visually and then left aside for a few hours before choosing the proper formula. Five grams from each formulation were prepared.

Two selected formulae (FF1 & FF2) of the NE were prepared as shown in Table 3. The final formula contained Vitamin C (Vit C) as an antioxidant, methylparaben, propylparaben, and Germall plus as preservatives in addition to RP and VE. Twenty grams of each formula were prepared by dissolving RP, VE, and propylparaben (if added) in the oily phase. Then Smix was added to the oily phase and mixed thoroughly. Vit C, methylparaben, and Germall plus were dissolved in the aqueous phase. Then NE was prepared as described above.

The prepared NEs were characterized in terms of droplet size and charge, content of RP and VE, rheological behavior, and diffusion study.

Droplet size and charge were measured by zeta sizer (Malvern / UK). 100 µL of the sample was diluted with distilled deionized water to one mL and placed in the instrument. All formulae that gave translucent NE were subjected to this test. Average droplet size and polydispersity index (PDI) were recorded. Three measurements were made and the average ±SD was calculated. Zeta potential was also measured and recorded as average ±SD.

An assay of the prepared formula was made using the developed HPLC method. One gram of each formula was taken, and centrifuged at 3000 rpm to break the emulsion. 100 µL were taken from the oil phase, appropriately diluted with the mobile phase and measured for RP and VE concentration.

The viscosity and rheological behavior of the prepared serum were determined using a cone and plate viscometer (Anton Paar, Rheometer Germany GmbH, Model MCR 302). All measurements were carried out at a temperature of 25±1 °C, using spindle Cp 50. The formula was loaded between the concentric cylinders in a volume of 5–10 ml to ensure accurate results. To assure accuracy, the rheometer was calibrated and programmed via a computer controlled RheoCompass software (Anton Paar). The flow curve is constructed by plotting shear rate (1/s) versus shear stress (in Pascals) as well as the viscosity (in mPascas) versus temperature (in Celsius) curve. Ebanel/ Vitamin C serum (USA) was used as reference for the rheological properties. The same procedure was repeated for the reference serum.

The diffusion study was performed using the Franz Diffusion Cell model (SES GmbH-Analyse systeme/Fridhofstr 7-9D 55234 Bechenheim/Germany). The membrane used was dialysis tubing (Medical International Ltd/ M wt. 12–14000 daltons), with a volume of donner compartment = 12 mL, diffusion media was phosphate buffer (PB) (potassium dihydrogen phosphate buffer pH 7.0), and cross-sectional area of diffusion = 1.7 cm². To ensure sink conditions, 100 mg of each of RP and VE were added to 10 mL phosphate buffer pH 7.0 with 5% Tween 80, then placed in a shaking water bath for 24 hours at 26 °C. The next day the samples were centrifuged, and the aqueous phase was analyzed for both compounds. Samples were applied to the membrane and readings were taken at 0.5, 1, 2, 3, 4, 5, 6, 7, 8, and 24 hours and the volume was replaced by fresh media to keep the same sink conditions. The test was performed on the final formula (FF2) whereby 200 mg sample which theoretically contained 1 mg of each RP and VE was put in each of the 6 cells and average readings were calculated. Due to the sensitivity of RP to the light, the whole instrument was covered with foil and turning off all lights and shutting down the curtains of the Lab.

Three healthy male albino Wistar rats (250±15g) were used for assessment of skin irritation/corrosive potential and the reversibility of dermal effects of a topical preparation of RP and VE (FF2). The rats numbered as follows: 1 a negative control with no treatment, 2 a positive control given a blank formula without RP and VE, and 3 received the formula FF2. The presented test was conducted according to the OECD Guidelines for Testing of Chemicals, adopting Guideline 404 for Acute Dermal Irritation/Corrosion (OECD 2002). Before the experiment, fur was removed by closely clipping the dorsal area of the trunk of the rats using an electric clipper while having the rat restrained humanely, then the application of the formulation was performed as described by the guidelines. Table 4 shows the Draiz system guidelines of evaluation.

A preliminary stability study was conducted on formula FF2 to determine its physical stability and concentration of RP and VE at room temperature. FF2 was prepared and put in amber test tubes and wrapped with aluminum foil and left on the bench in the Lab where the temperature was around 24–26 °C during the day. Samples were taken at time zero and examined after 1, 2, 3 and 4 weeks, for physical appearance of the NE as well as estimation of RP and VE content. Particle size and zeta potential were also measured after 4 weeks.

All statistical analysis results were performed using Microsoft Excel. The results were presented as mean ± SD.

The particle size of O/W NE is a function of the oil phase to water phase ratio, type, and concentration of surfactant, type, and concentration of co-surfactant (if added), and the preparation method (Uchechi et al. 2014).

The results of measuring particle size are illustrated in Table 6. It was investigated that the major factors affecting the droplet size are composition of the emulsion (ratio of the oil phase to aqueous phase), the type and concentration of the emulsifying agent. Since the preparation was intended for skin care, and the active materials (VA and VE) are needed to be absorbed, that’s why small droplets were needed but not less than 20 nm. The evaluation of the best combination was made based on the smallest droplet size obtained and the smallest PDI, which reflect homogeneity of the NE. Bases on the solubility study, the volume of oil phase 25–30%, was enough to load the required dose of both VA and VE. Since all formulas were prepared using the safflower oil, the ratio of oil phase, type of surfactant and co-surfactant were the major factors affecting the particle size. Oil phase less than 25% was not enough to stabilize the emulsion and increasing the ratio of co-surfactant had no effect in decreasing the particle size. Cremophor RH40 without co-surfactant was also able to form NE. Based on the results of a particle size measurement, F8 which contained 25% safflower oil, 65% aqueous phase and 10% cremophor RH40-ethanol (1:1) was chosen to formulate the medicated NE.

Two medicated formulations were prepared from F8 named “FF1” which contained the paraben as a preservative and “FF2” that contained Germall plus as preservative in addition to the RP, VE and vitamin C were prepared as shown in Table 3. Germall plus is natural broad spectrum preservative that is compatible with most cosmetic products. Parabens have the advantage of mixing the hydrophobic ones with the oil phase and the hydrophilic with the aqueous phase. It was decided to give option according to the availability and cost of the final preparation for industrial purposes.

Particle size was measured in both formulas, FF1 gave a particle size of 50.3±5 nm with PDI of 0.32 and FF2 gave particle size of 52±3.6 nm with PDI of 0.34. These particle size values are not significant from F8 (before addition of APIs (p>0.05) .Small particle size would help in the good absorption and penetration of the APIs to the skin. Arianto & Cindy prepared sunscreen NE composed of sunflower oil as an oily phase and Tween 80 as a surfactant. The evaluation showed that the prepared formulation had an average particle size of 124.47 nm with yellowish color, a clear, transparent appearance (Arianto, and Cindy 2019). Uchechi et al. reported that nanoparticles ranging 50–500 nm are suitable for dermal and transdermal delivery of APIs. Very small of 1–20 nm might be cleared quickly and particles > 1 micron are retained on the surface of the skin (Uchechi et al. 2014). FF1 and FF2 have droplet size slightly larger than 50 nm which is suitable for topical absorption to the underlying skin layer. Zeta potential of FF1 and FF2 were +0.917 mV and +0.995 mV respectively. These were small values that might create stability problems. The small zeta potential is expected since the NE was prepared using a non-ionic surfactant. It could be increased by using a buffer with higher ionic strength as an aqueous phase.

Estimation of RP and VE in the final preparation showed that percent RP content was 97.3± 0.5% and 96.6±10% for FF1 and that of FF2 was 97.9±0.8 and 98.8±0.5 for RP and VE respectively.