All chemicals and reagents utilized in the study were obtained from Sigma-Aldrich (St. Louis, MO, USA). Mitomycin C (MC) was purchased from Korea United Pharma. Rosemary oil (REO) was obtained from Sukar Nabat for natural oil (Jeddah, KSA). Breast cancer (MCF-7) and Human cervical cancer (HeLa) cells were donated by the Tissue Culture Unit at King Fahad Center for Medical Research (Jeddah, KSA). Cancer cell lines were cultured in a 25-cm² cell culture flask in Dulbecco’s modified Eagle’s medium containing antibiotics (0.1 mg/mL penicillin, 100 U/mL streptomycin and 10% v/v) and fetal bovine serum and incubated at 37 °C in a 95% humidified atmosphere with 5% CO₂. The confluent cells were trypsinized using 0.2 ml of trypsin, and they were seeded in 24- or 96- well plates, based on the assay. The antineoplastic and apoptotic effects of the studied formulation against the cancer cells were measured using MTT assay and DAPI assay, respectively.

Different weight fractions of oil phase (REO), hydrophilic phase (distilled water), and a mixture of surfactant and co-surfactant (tween 80 and span 20; 2:1 v:v) respectively, were mixed in Pyrex screw cap test tubes and heated up to 70 °C, with continuous mixing using a vortex mixer (VELP, Scientific, Italy) at 3000 rpm. This procedure was carried out to determine the best proportions of the constituents at which the one phase of REO/NE could be obtained. The resultant REO/NE was exposed to different temperatures (25 to 70 °C) to measure the opacity of the solution.

The REO/NE obtained was utilized as a carrier for MC. A stock solution of REO/NE+MC was produced by dissolving MC in REO/NE to yield solution of concentration 50 µg/ml of formula. Similarly, a 50 µg/ml stock solution of MC was prepared in normal saline (0.9% NaCl, w/v).

The measurements of particle size, charge and polydispersity index for REO/NE and REO/NE+MC were carried out 25 ± 0.2 °C using Zetasizer (Malvern, Nano-ZS Worcestershire, UK). Three independent measurements for each sample were analyzed to obtain the parameters. The data are presented as mean ± SD (x̄ ± SD).

Assay of in vitro anticancer activity was conducted to determine the potential antiproliferative and cytotoxic properties of the combination formula, REO/NE+MC on the cancer cells. The MCF-7 and Hela cells were seeded in triplicate in 96-well culture plates, each at a density of 10×10³cells/well and permitted to adhere for 24 h. Then, the cells were treated with different concentrations of REO/NE and MC for 24 h, to generate growth curves. The cells were incubated for 4 h with 0.5 ml of MTT solution (500 µg/ml) for 4 h 37 °C in the dark. Thereafter, the medium was discarded, and DMSO was added to the wells (0.1 ml/well) to dissolve the formazan crystals formed. The absorbance (Abs) of each well was recorded at 570 nm in a multiwell microplate reader (Bio Tek Instruments, Winooski, VT, USA). The Abs values were used to calculate percentage inhibition of cell growth, with the following equation:

$\text{ Growth inhibition }(\%)=100-\left(\frac{\text{ Abs }\left(\text{ treated cells }\right)-\text{ Abs }\left(\text{ blank }\right)}{\text{ Abs(untreated cells) }-\text{ Abs }\left(\text{ blank }\right)}\times 100\right)$

Then, the data were transformed to the fraction affected (Fa) ranging from 0 to 1, where Fa=0 and Fa=1, representing 100 and 0% viability, respectively.

The results of MTT were input into Compusyn software (Biosoft, Ferguson, MO, USA) and utilized to estimate the values of combination index (CI) using the Chou-Talalay Method for non-constant ratio of combination. The effect of combinations, REO/NE + MC, was primarily reflected by the combination index CI, where CI< 1 indicates synergism, CI> 1 indicates antagonism, and CI=1 demonstrates an additive effect.

The MCF-7 and Hela cancer cells were grown overnight in 24-well plates at a density of 50 × 10⁴ cells/well and treated for 24 h with three selected concentrations used in MTT assay of REO/NE, MC and the combined formula. Then, the cells were fixed with 4% paraformaldehyde solution in phosphate buffered saline (PBS, pH 7.4) for 30 min at room temperature, followed by rinsing with PBS for 5 min. The cells were permeabilized with Triton X-100 (0.1% in BPS) and stained using 4',6'-diamidino-2-phenylindole hydrochloride (DAPI) for 15 min in the dark. Finally, the images of morphological changes in the treated cells going through apoptosis were acquired using a fluorescence microscope (Leica CRT6000, Germany) with a blue filter at 437 µm. The mean nuclear area (in pixel²) for each group was quantified using ImageJ analysis software (Rasband, W.S., ImageJ, National Institute of Health, US).

The analysis of data was performed with MegaStat version 10.3 (Bulter University, Indianapolis, IN). Differences between the means of dependent and independent groups were determined using paired t-test and unpaired t-test, respectively. Significant differences were assumed at p values less than 0.05.

In this study, different fractions of the hydrophilic phase (distilled water), lipophilic phase (REO) and the mixture of surfactant and co-surfactant (tween 80 and span 20) at a ratio of 2:1, were used to determine the proportion of the constituents that resulted in one phase of the REO/NE product. Figure 1A–D show the three-, two- and bi-continuous phase emulsions produced at different proportions of the constituents. As displayed in Figure 1F, a clear one-phase REO/NE was obtained when 1.78% of REO was mixed with 92.8% of water in the presence of 5.41% of the mixture of tween 80 and span 20. Figure 2F shows the graph of turbidity of the resultant REO/NE at different heating temperatures. Turbidity was maintained from 25 to 60 °C, but it was markedly increased upon heating the system from 65 to 75 °C.

Figure 1. 

Pie charts showing three phases, two phases and one phase, based on the proportions of the constituents. A–D show three, two and bi-continuous phase emulsions. E: One phase of rosemary oil in water nano-emulsion (REO/NE). The green, blue and yellow slices represent the rosemary oil, distilled water and the mixture (2:1) of tween 80 to span 20, respectively. F: Turbidity – temperature curve of the resultant one-phase REO/NE showing the changes in the opacity of the formula at different temperatures. Data are presented as mean ±SD (n = 3).

The physical properties of the formulations are presented in Figure 2. The size distribution curves revealed that the mean diameters of synthesized REO/NE and REO/NE+MC were 59.22 ± 1.14 nm and 71.43 ± 1.32 nm, respectively, while both had similar polydispersity index of 0.02. In terms of zeta potential, the negative magnitude of REO/NE resulted in no remarkable change in the droplet charge after combined with MC.

The effect of various concentrations of REO/NE and MC on the growth of MCF-7 and HeLa cells, expressed as fractions of cells affected (Fa) for 24 h (based on MTT assay), are shown in Figure 3A, B. The data demonstrates that the fractions of growth inhibition of the HeLa cells were markedly increased when treated with either the selected concentrations of REO/NE or when subjected to < 1.98 µM MC > Fa values of MCF-7 cells (p < 0.05). Figure 3C shows that REO/NE and MC exhibited dose-dependent cytotoxicity, with IC₅₀ values of 1.83 ± 0.23% (v/v) and 49.65 ± 1.97 µM, respectively, on the MCF-7 cells, which were 1.30-fold and 1.66- fold, respectively, higher than those in HeLa cells. The half-maximal inhibitory concentration (IC₅₀) values of the individual drugs are shown in Figure 3C.

Figure 2. 

(A) Size and (B) zeta potential of REO/NE and REO/NE + MC, measured using DLS. The data are presented as mean ±SD (n = 3). The symbol NS indicates no significant difference; *** indicates significant differences between the formulations, (p < 0.001; paired sample t-test).

Furthermore, the MCF-7 and HeLa cells were treated with varying concentrations of the combination (REO/NE + MC) for 24 h, at levels that did not exceed the IC₅₀ values of MC alone. The results in Figure 3A show 50% inhibition of cell growth when MCF-7 cells were treated with 1% (v/v) REO/NE + 1.49 µM of MC. In contrast, treatment of MCF-7 cells with 1% REO/NE and 1.49 µM led to 15% and 18% inhibitions of cell growth, respectively. Regarding HeLa cells, 49 and 26% inhibitions of cell growth resulted when they were subjected separately to 1.33% (v/v) REO/NE and 1.98 µM MC, while 50% inhibition of cell growth occurred when the two were combined (Fig. 3B).

Figure 3. 

Inhibition of cell proliferation expressed as fraction affected (Fa) and assessed with MTT. Bar charts showing the Fa of (A) MCF-7 and (B) HeLa cells after 24 h of treatment with different drug concentrations. (C) IC₅₀ of individual drug treatments. Data shown are presented as mean ±SD (n=3). The symbols ^# and ^* indicate significant differences between MCF-7 and HeLa cells. ^$ indicates differences between MC and REO/NE + MC; ^#,*p <0.05, ^**P <0.01, ^{###, $$$} P <0.001.

Figure 4A shows the interactions between REO/NE and MC in MCF-7 and HeLa cells in dose-effect curves, which are represented by the m values obtained from their linearization with the median-effect plots (Fig. 4B) and summarized in Table 1. The result demonstrated that m values of the REO/NE dose-effect curves in both cell lines were similar (p > 0.05; m > 1), indicating steep curves, while the m value of MC was markedly lower in MCF-7 cells more than in HeLa cells (p < 0.05; m < 1), suggesting a shallower curve for MC in MCF-7 cells than in HeLa cells.

To determine whether the combination of REO/NE and MC produced higher cytotoxic effects on MCF-7 and HeLa cells than the predicted individual cytotoxicity, the Chou-Talalay method for non-constant ratio of REO/NE + MC combination was used to estimate the combination index (CI) through CompuSyn Software (Fig. 5). The results of Fa-CI plots revealed different patterns of interactions between REO/NE and MC in MCF-7 and HeLa cells. In particular, the CI values in MCF-7 cells increased in the direction of increasing the Fa values, with a value of 0.99 at Fa 0.54. However, a slight increase in the CI to 1.01 was observed when Fa increased to 0.7. At Fa 0.96, the CI value reached 0.6 and was similar to the CI at Fa 0.28. In contrast, for MCF-7 cells, CI values at Fa < 0.5 were above 1 and were slightly reduced to 1.1 at F 0.55. Thereafter, CI was decreased gradually to 0.43 at Fa 0.96.

To quantify the apoptotic effect of the individual and combined formulas of REO/NE and MC, the averages of nuclear areas visualized via DAPI staining in the microscopic images of MCF-7 (Fig. 5A) and HeLa cells (Fig. 6A) are summarized in Figures 5B, 6B, respectively. Cells were treated for 24 h with 3 different concentrations of the formulas, chosen according to the results of MTT assay. At a concentration of 0.47% (v/v) REO/NE and 0.7 µM MC, treatment of MCF-7 cells with REO/NE produced markedly higher reduction in the nuclear areas, when compared with MC treatment MC (p < 0.01). In contrast, the nuclear areas of HeLa cells treated with MC were smaller than the corresponding areas of cells treated with REO/NE (p < 0.05). In the case of combination therapy, a higher decrease in the nuclear area of MCF-7 was observed with 0.7 µM MC combined with 0.47% (v/v) REO/NE, when compared the nuclear areas of MCF-7 due to individual drug treatments. When HeLa cells were treated with a similar concentration of the combination, the nuclear area appeared smaller than that of REO/NE-treated cells but was larger than that in MC-treated cells (p < 0.01).

Figure 4. 

(A) Dose-Effect Curve, and (B) its linearization with Median-Effect Plots for effects of single and combination treatments with REO/NE and MC on MCF-7 and HeLa cells.

However, the treatment of MCF-7 with 1% REO/NE (v/v) and HeLa cells with 1.33% (v/v) REO/NE produced the least nuclear areas, when compared to the cells treated with MC at doses of 1.48 µM and 1.98 µM MC only, or with the combined formulations. As the concentration of REO/NE was increased to 2% (v/v) and that of MC to 4.94 µM, MCF-7 cells exhibited more marked reductions in nuclear areas than when subjected to MC (p < 0.01). Treatment of MCF-7 cells with REO/NE combined with MC led to the least nuclear areas, relative to individual treatment (p < 0.001). On the other hand, Hela cells treated with 2% (v/v) REO/NE or 4.94 µM MC revealed similar nuclear areas (p > 0.05). However, smaller nuclei areas were produced with the combined formula (p < 0.01).