Paclitaxel and Taxol^® were purchased from Beijing Union Pharmaceutical Factory (Beijing, China). RPMI-1640 medium, fetal bovine serum (FBS), 0.25% (w/v) trypsin solution and antibiotic agents were purchased from Invitrogen Life Technologies (Carlsbad, CA, USA). Phosphate-buffered saline (PBS, pH, 7.4) was purchased from Hyclone (Logan, UT, USA). Cell Counting Kit-8 (CCK-8) was obtained from Dojindo Laboratories (Tokyo, Japan). Nile red, fluorescein and Hoechst 33258 were procured from Sigma-Aldrich (St. Louis, MO, USA). DIR was obtained from AAT Bioquest, Inc. (Sunnyvale, CA, USA). All solvents such as acetone and acetonitrile were of analytical or high performance liquid chromatography (HPLC) grade and were used according to the manufacturer's instructions.

PTX-CH Emul was prepared using a solvent evaporation method and high-pressure homogenization, respectively, as previously reported (22). The PTX Emul was prepared using the same method except that the process of dissolving PTX in oil was different. Briefly, PTX was dissolved in acetone. Then soybean oil-medium chain triglyceride (1:1, w/w) and vitamin E acetate were added to this organic phase. Acetone was removed by evaporation and the oily paste was used as the oil phase of emulsion. Fluorescein-labeled emulsion (FL emulsion) was prepared using the same procedure, except that fluorescein or the fluorescein-cholesterol complex was dissolved in oil in advance.

For the in vivo tumor-targeting imaging studies, DiR-encapsulated PTX-CH Emul and DiR-encapsulated PTX Emul were prepared using the same method as that of PTX-CH Emul and PTX Emul, respectively.

The emulsion particle size was measured by a PSS NICOMP Particle Sizing System (PSS, Port Richey, FL, USA) after appropriate dilution with double-distilled water. The paclitaxel content of the emulsion was measured using an HPLC system with a UV detector (Agilent Technologies Inc., Cotati, CA, USA). The emulsion encapsulation efficiency was determined by measuring free paclitaxel in the aqueous phase.

Confocal laser scanning microscopy (CLSM) (FV1000; Olympus Corp., Tokyo, Japan) was employed to visualize the microenvironment of the major components in the emulsion, lipid based depot and lipid nanoparticles (6,24). The fluorescent emulsions were labeled with two fluorescent probes, Nile red and fluorescein-cholesterol complex. Nile red- and fluorescein-labeled emulsions were prepared at a load ratio of 0.005% (w/v), diluted with appropriate quantities of distilled water, and a droplet of the emulsion dispersion was then sealed between two coverslips for CLSM measurement.

MDA-MB-231 cells were purchased from the Department of Cell Resources, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences & Peking Union Medical College (Beijing, China). The cell line was authenticated by the suppliers and cells were passaged in the laboratory. Cells were maintained in a 37°C/5% CO₂ humidified chamber in RPMI-1640 media supplemented with 10% FBS, 100 u/ml penicillin and 100 µg/ml streptomycin.

Female BALB/c nude mice with a body weight of 21 to 25 g and ages ranging from 6 to 8 weeks were obtained from Beijing HFK Bioscience Co., Ltd. (Beijing, China). All animal protocols were approved by the Institutional Animal Care and use Committee. Female BALB/c nude mice were inoculated subcutaneously in the right armpits with 0.2 ml (2×10⁶) MDA-MB-231 cell suspension. Cells were propagated in vivo as solid tumors. The tumors were implanted subcutaneously as fragments and allowed to grow to a median size of 100–200 mm³ before treatment was initiated.

The cytotoxicity of Blank Emul, PTX-CH Emul, PTX Emul and Taxol in the MDA-MB-231 cells was measured using CCK-8 (25). Cells were seeded in 96-well plates at a density of 3,000 cells/well. After 24 h, the cells were incubated with PTX-CH Emul, PTX Emul or Taxol at different PTX concentrations. The sequence of Blank Emul concentrations was exactly the same as PTX-CH Emul. After 72 h of incubation, the number of viable cells was determined using CCK-8 according to the manufacturer's instructions. Untreated control cells were considered to be 100% viable. The concentration required for 50% or 70% growth inhibition (IC₅₀ or IC₇₀ values, respectively) was determined with SPSS19.0 software (SPSS Inc., Chicago, IL, USA) as the mean ± standard deviation (SD).

MDA-MB-231 cells were seeded into 6-well chamber slides and allowed to attach overnight. The medium was replaced with culture medium containing fluorescein-DMSO solution (FL-DMSO), FL emulsion or emulsion loaded with fluorescein-cholesterol complex (FL-CH emulsion). After incubation for 4 h at 37°C, cells were washed three times with cold PBS, fixed with 4% paraformaldehyde (PFA), and the nuclei were counterstained using Hoechst 33258. The slides were observed using an Olympus FV1000 CLSM (26).

Ex vitro 3D breast tumor spheroids of MDA-MB-231 cells were developed using a liquid overlay system (27). An agarose solution (1.5%, w/v) was prepared in serum-free RPMI-1640 media by heating at 80°C for 30 min, followed by sterilization in an autoclave. Each well in the 96-well plates was coated with a thin layer (50 µl) of the sterilized agarose solution. For optimal spheroid formation, 2,000 cells were added to each well and medium containing 2.5% Matrigel (BD Biosciences, San Jose, CA, USA) was added. The plate was centrifuged for 10 min at 1000 × g and cultured in a standard cell incubator at 37°C with 5% CO₂. Cell culture medium was changed every 2 days. After 7 days, separate groups of spheroids were incubated with culture medium, PTX-CH Emul, PTX Emul or Taxol, each at the PTX concentration of 0.5 µg/ml. growth inhibition was monitored by measuring the size of the spheroids using an inverted phase microscope at 0, 1, 3, 5 and 7 days of treatment. The major (d_max) and minor (d_min) diameter of each spheroid was determined, and spheroid volume was calculated using the following formula: V = (π × d_max × d_min)/6. The spheroid volume ratio was calculated using the following formula: R (%) = (V_dayi/V_day0) × 100, where V_dayi is the spheroid volume at the nth day (day 1, 3, 5 or 7) of treatment, and V_day0 is the spheroid volume prior to treatment.

DiR, a near-infrared fluorescent probe, was encapsulated into PTX Emul and PTX-CH Emul. To investigate the tumor-targeting efficiency of PTX-CH Emul in the MDA-MB-231 tumor-bearing nude mice, PBS, DiR-encapsulated PTX Emul or DiR-encapsulated PTX-CH Emul was injected into the mice via the tail vein at a dose of 200 µg of DiR/kg body weight. The in vivo imaging was performed at different time points (2, 4, 8, 12 and 24 h) post-injection using the In vivo IVIS spectrum imaging system (Perkin-Elmer, Wellesley, MA, USA). At 24 h post-injection, the mice were sacrificed and the tumors and other major organs (heart, liver, spleen, lung and kidney) were harvested. The ex vivo imaging of the organs was also carried out.

Healthy female BALB/c nude mice were used to determine the MTD of PTX-CH Emul and Taxol after intravenous administration. Eight groups of 5–8 female BALB/c nude mice received intravenous injections of Taxol (20, 30 or 45 mg/kg), PTX-CH Emul (30, 45, 67.5 or 101.25 mg/kg) or saline (as a control) via the tail vein on days 0, 4 and 8. The effects of the treatments were investigated by close observation of survival rates. The MTD was defined as the dose that caused neither death resulting from toxic effects nor significant changes in general clinical signs within 1 week after intravenous administration of the aforementioned formulations.

The antitumor efficacy and toxicity profiles of the different PTX formulations were evaluated in a subcutaneous MDA-MB-231 xenograft model in female BALB/c nude mice. The treatments commenced when the tumors in the nude mice reached a volume of 100–200 mm³, and this day was designated as day 0. On day 0, the mice were randomly divided into 3 groups of 8–9 mice each. The mice were intravenously administered PBS, Taxol (20 mg/kg) or PTX-CH Emul (45 mg/kg), respectively. The treatments were conducted every 4 days for a total of 3 doses. Tumor size was measured with a digital caliper every 3 days. Tumor volume was calculated by the formula of (LxW²)/2, where L is the longest tumor dimension (mm) and W is the shortest tumor dimension (mm). The time required for a tumor to double in volume was calculated based on the initial tumor volume at the beginning of the treatment. Tumor recurrence was defined as the first observation of increased tumor size following tumor regression. The toxicity after treatment was monitored by observing behavior and measuring body weight every 3 days.

All data subjected to statistical analysis were obtained from at least 3 parallel experiments, and the results are expressed as mean ± standard deviation (SD). Tumor doubling time and time to tumor recurrence were analyzed by Kaplan-Meier analysis, according to GraphPad Prism version 5.01 for Windows (GraphPad^®, Inc., Software, San Diego, CA, USA). Statistical analysis was performed by Student's t-test for two groups, and one way ANOVA for multiple groups. A value of P<0.05 was considered statistically significant.

The characteristics of PTX-CH Emul were previously investigated (22). PTX-CH Emul, which is composed of a paclitaxel-cholesterol complex surrounded by a phospholipid monolayer, resembles the lipid structure of native LDL (Fig. 1A). The resulting emulsion had an encapsulation efficiency of 97.3% and a mean particle size of 150 nm (Fig. 1B). Moreover, the formulation survived autoclaving at 115°C for 30 min and remained stable for at least 12 months at 6°C. PTX-CH Emul was also substantially better tolerated than the corresponding dosages of Taxol in guinea pigs, as no evident hypersensitivity reaction was observed. These results demonstrated that PTX-CH Emul has excellent potential for industrial-scale production and clinical application.

CLSM is a useful tool in the investigation of the microstructure of emulsions (26,27). In our study, the distribution of the fluorescein-cholesterol complex and oil in the emulsion was determined by CLSM. Oil was labeled with a lipid probe, Nile red, a hydrophobic fluorescent marker with special fluorescent properties. Nile red emitted red light and was excited to visualize the distribution of the oil core in the emulsion. Fluorescein emitted green light was excited to visualize the distribution of the fluorescein-cholesterol complex in the emulsion. Yellow light was observed when Nile red and fluorescein were both excited and localized in the same region. The micrographs revealed a uniform distribution of Nile red and fluorescein in the emulsion. The green and red regions completely overlapped, indicating that the fluorescein-cholesterol complex and oil phase were evenly distributed throughout the emulsion, without phase separation (Fig. 1C–F).

The cytotoxicity of Blank Emul, PTX-CH Emul, PTX Emul and Taxol was evaluated using MDA-MB-231 cells. PTX decreased cell viability in a concentration-dependent manner in the concentration range of 0.001–0.05 µg/ml (Fig. 2A). The cell-survival rate in the group treated with PTX-CH Emul was significantly lower than that in the groups treated with PTX Emul and Taxol after 72 h, demonstrating that PTX-CH Emul had better anticancer efficacy than PTX Emul and Taxol.

An important parameter for quantitatively evaluating the in vitro cytotoxicity of an anticancer drug is the IC₅₀ or IC₇₀, which represents the drug concentration needed to kill 50% or 70% of the cancer cells at a designated time (28,29). Fig. 2B and C shows the IC₅₀ and IC₇₀ values of PTX-CH Emul, PTX Emul and Taxol in the MDA-MB-231 cells after a 72-h incubation, respectively. Among the different PTX formulations, PTX-CH Emul had the lowest IC₅₀ value (2.13±0.28 ng/ml), followed by PTX Emul (3.02±0.19 ng/ml, P<0.05) and Taxol (4.22±0.52 ng/ml, P<0.001). If IC₇₀ value was used, the differences were more pronounced since the IC₇₀ value of PTX-CH Emul (4.69±0.36 ng/ml) was significantly lower than that of all the other formulations, only ~1.5- and 2.4-fold lower than that of PTX Emul (7.14±0.63 ng/ml, P<0.01) and Taxol (11.07±0.93 ng/ml, P<0.001) respectively. These results demonstrated that PTX-CH Emul exhibited superior cytotoxicity as compared to PTX Emul and Taxol. Additionally, there was no cellular cytotoxicity caused by the Blank Emul at the investigated concentrations, eliminating the possibility that the lipid emulsions themselves were responsible for the cytotoxicity.

The cellular uptake profiles of FL-DMSO, FL emulsion and FL-CH emulsion in the MDA-MB-231 cells were assessed qualitatively using CLSM. As shown in Fig. 3, FL-CH emulsion exhibited significantly greater uptake in comparison with FL-DMSO and FL emulsion. This result demonstrated that emulsion loaded with a fluorescein-cholesterol complex significantly facilitated the uptake of fluorescein by the MDA-MB-231 cells.

In vitro 3D tumor spheroids provide an important link between monolayer cell cultures and animal models for evaluating drug delivery efficiency (30,31). 3D tumor spheroids generated by the liquid overlay technique contain aggregates of cells in close contact and an organized extracellular matrix consisting of fibronectin, laminin and collagen, which is similar to the extracellular matrix of tumors in vivo (32). Therefore, 3D breast tumor spheroids were used to evaluate the inhibitory effect of PTX-CH Emul on cell growth. 3D tumor spheroid growth was evaluated following the treatment of cells with culture medium containing PTX-CH Emul, PTX Emul or Taxol. In regular cell culture medium, 3D tumor spheroids of MDA-MB-231 cells grew rapidly and became more compact with the increase in culture time. However, in the presence of PTX Emul or Taxol, the tumor spheroids stopped growing or even became smaller. Distinctly, treatment of cells with PTX-CH Emul resulted in shrinkage of the spheroids, with the concomitant loss of 3D structures due to the dissociation of some cells from the spheroids, indicating that the PTX-CH Emul effectively inhibited tumor cell proliferation (Fig. 4A). After 7 days of treatment, the tumor spheroid volume ratios were 190.2±21.2% for the control group, 69.5±11.8% for the Taxol group, 54.8±6.2% for the PTX Emul group and 19.9±4.0% for the PTX-CH Emul group (Fig. 4B). All things considered, PTX-CH Emul exhibited much stronger inhibitory effects on tumor spheroids in comparison with Taxol and PTX Emul.

The tumor-targeting efficiency of PTX-CH Emul was evaluated in nude mice bearing MDA-MB-231 tumors. After the DiR-labeled emulsions were injected through the tail vein, the time-dependent biodistribution of different emulsions was monitored using non-invasive NIR fluorescence imaging in live animals. As shown in Fig. 5A, at all time points post-injection, the tumor fluorescence in DiR-PTX-CH Emul-treated mice was higher than that in the DiR-PTX Emul-treated mice at equivalent doses of DiR. Twenty-four hours after injection, 3D reconstruction of DiR-PTX-CH Emul-treated mice was performed. As shown in Fig. 5C, the fluorescence was primarily found at tumor sites, indicating that the DiR-PTX-CH Emul may serve as an effective tumor-targeting vehicle by which to deliver chemotherapeutic drugs to tumor tissues. Additionally, the ex vivo fluorescent image from the excised tumors also clearly showed higher fluorescence from the DiR-PTX-CH Emul-treated group over that from the DiR-PTX Emul-treated one (Fig. 5B), which was further evidence of the higher tumor-targeting efficiency of PTX-CH Emul. All these findings indicated that PTX-CH Emul enabled specific tumor-targeting efficiency, which could contribute to the selective increase in the therapeutic efficiency.

The MTD of Taxol in healthy BALB/c nude mice is 20 mg/kg, consistent with the previously reported results (33). PTX-CH Emul was significantly less toxic than Taxol, and the MTD of PTX-CH Emul was 2.25-fold higher than that of Taxol (45 mg/kg vs. 20 mg/kg, respectively). A previous study reported that the MTD of Abraxane^® is 2.24-fold higher than that of Taxol (30 mg/kg vs. 13.4 mg/kg, respectively) (34), which is essentially the same as that reported here. Based on these results, MTD values of 45 mg/kg and 20 mg/kg were chosen for PTX-CH Emul and Taxol, respectively, for the in vivo antitumor efficiency study.

Both Taxol and PTX-CH Emul significantly inhibited tumor growth in comparison with the control group (Fig. 6A). At the MTDs of PTX-CH Emul and Taxol (45 and 20 mg/kg, respectively), the proportion of tumor-free survivors in the PTX-CH Emul group (7 of 9) was higher than that in the Taxol group (0 of 8) (Table I). Equally as important, PTX-CH Emul showed superior antitumor activity to Taxol as measured by median time to tumor recurrence [>79 days vs. 32 days (P=0.0001)] (Fig. 6B and Table I), tumor doubling time [>79 days vs. 7.5 days (P<0.0001)] (Fig. 6C and Table I), tumor volume (P<0.01) (Fig. 6A) and tumor weight (P<0.001) (Fig. 6D). These results demonstrated that PTX-CH Emul possessed higher antitumor efficacy as compared with Taxol at the MTDs.

Toxicity was assessed by direct observation of animal behavior and body weight. Mice treated with 20 mg/kg Taxol showed reduced overall activity, which was likely a sign of a hypersensitivity reaction to the diluent. In contrast, the mice treated with PTX-CH Emul tolerated the regimens well. As shown in Fig. 6E, there was no significant difference in body weight loss between mice treated with PTX-CH Emul and Taxol at the MTDs. Mice had reduced body weight initially after treatment with the aforementioned formulations but gradually began to grow normally after they adapted to the challenges.