Multi-drug delivery to tumor cells via micellar nanocarriers

2.1. Chemicals

Paclitaxel and 17-AAG were purchased from LC Laboratories (Woburn, MA). 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (PEG-DSPE) was obtained from Avanti Polar Lipids (Alabaster, AL). PEG-PLA [The number average molecular weight (Mn) of PEG and PLA were 4.2K and 2.1K, respectively, the polydispersity index (PDI) = 1.05] was purchased from Advanced Polymer Materials (Montreal, Canada). D-α-tocopheryl polyethylene glycol 1000 succinate (TPGS) was a gift from Eastman Chemical Company (Kingsport, TN). All other chemicals were of analytical or reagent grade.

2.2. Preparation of paclitaxel/17-AAG-loaded micellar nanocarriers

The dual drug-loaded micelles were prepared by a solvent evaporation method. Briefly, in a 5-ml round-bottom flask, paclitaxel, 17-AAG and copolymers (up to 5 mg) were dissolved in 1 ml chloroform, which was removed under vacuum at the room temperature to form a homogenous thin drug-polymer film. Next, the film was hydrated with 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)-buffered saline (HBS, pH 7.4) and sonicated for 10 min at room temperature. The resulting mixture was then centrifuged at 12,000 g for 10 min to yield clear micelle dispersion. PEG-PLA micelles were prepared with 5.6 mM PEG-PLA, while PEG-DSPE/TPGS mixed micelles were prepared with 12.5 mM PEG-DSPE and 25.0 mM TPGS. Under these conditions, the resulting concentrations of both PEG-PLA and PEG-DSPE were identical at 35 mg/ml. Unless specified otherwise, the resulting concentrations of paclitaxel and 17-AAG in the micellar dispersion were 0.2 mM and 1.4 mM, respectively. The hydrodynamic diameter of the micelles was assessed by dynamic light scattering using a Nano Zetasizer (Malvern Instruments, UK).

2.3. Simultaneous quantification of paclitaxel and 17-AAG by HPLC analysis

The concentrations of paclitaxel and 17-AAG were simultaneously determined using a reverse phase HPLC methodology. The HPLC system (Waters, Milford, MA) has a Waters 2795 pump, a Phenomenex (Torrance, CA) C8 column (5 μm, i.d. 4.6 × 150 mm), and a Waters 996 photodiode array detector. As an internal standard, α-naphthoflavone was used at a concentration of 2 μM. The detection wavelengths for paclitaxel, 17-AAG and α-naphthoflavone were 227 nm, 333 nm and 281 nm, respectively. The isocratic mobile phase was comprised of 25% (v/v) sodium phosphate buffer (25 mM, pH 3.0) and 75% (v/v) methanol with 10 mM triethylamine, and the flow rate was 1.2 ml/min. Before being injected into the HPLC system, micelle samples were first diluted with the HPLC mobile phase. The linear range for the standard curves of paclitaxel and 17-AAG was between 0.5–10 μM.

2.4. Release kinetics of paclitaxel and 17-AAG from micelles

The release kinetics of paclitaxel and 17-AAG from micellar nanocarriers was studied using a previously reported dialysis method (Chandran et al, 2010). The dual drug-loaded micelles were prepared as described above, and diluted about 10 times with HBS or fetal bovine serum (Invitrogen, Carlsbad, CA), and loaded into a 3 ml Slide-A-Lyzer dialysis cassette (Thermo Scientific, Rockford, IL) with a 20 KD MWCO. To ensure the sink condition for the drug release, each cassette was placed in 500 ml phosphate-buffered saline (20 mM, pH 7.4) that was changed every 2 hours. At pre-determined time points, a sample (20–40 μl) was collected from each cassette, for which the concentrations of paclitaxel and 17-AAG were determined by HPLC. To derive the first-order release rate constant (k), the drug concentration in the dialysis cassette (C_t) as a function of release time (t) was fitted to the equation C_t/C₀ = e^−kt, wherein C₀ was the initial drug concentration (Chandran et al., 2010). The best-fit nonlinear regression was obtained by Sigma Plot (San Jose, CA), and the release half-life (t_{1/2, release}) was calculated by 0.693/k.

2.5. Storage stability of paclitaxel/17-AAG-loaded micelles

Freshly prepared paclitaxel/17-AAG-loaded PEG-DSPE/TPGS mixed micelles were stored at 4 °C or 37 °C. At pre-determined time points, the micelle samples were centrifuged at 12,000 g for 10 min. The supernatants were then analyzed by HPLC for changes in the drug content.

2.6. ¹H NMR spectroscopy

The dual drug-loaded micelles were prepared as described above, except that D₂O was used in HBS solution. As a control, identical amounts of drugs and copolymers were dissolved in DMSO-d₆. NMR data were collected on a 600 MHz NMR spectrometer (Agilent, Palo Alto, CA) at 20 °C with 2048 scans and a spectral width of 7.2 kHz. ¹H chemical shifts were referenced to DMSO at 2.50 ppm for the DMSO-d₆ sample and to internal standard sodium 3-(trimethylsilyl) propionate-2, 2, 3, 3-d₄ (TSP) at 0.0 ppm for the micelle sample. The residual HDO signal in the micelle sample was suppressed by a saturation pulse with a field strength of 40 Hz.

2.7. Cell proliferation assay

Human ovarian cancer SKOV-3 cells (ATCC, Manassas, VA) were cultured in DMEM (Invitrogen) and maintained at 37 °C with 5% CO₂ in a humidified incubator. Cells were seeded in 96-well plates and treated with paclitaxel (2.5–10 nM), 17-AAG (17.5 nM–70 nM), combinations of the two drugs at a 1:7 molar ratio as free drugs or in micelles. The dual drug-loaded micelles were prepared as described above. Stock solutions of free paclitaxel and 17-AAG were prepared in DMSO. At 72 h, cells were stained with 0.1% crystal violet, which were dissolved in 10% acetic acid and analyzed for their absorbance at 595 nm using a microplate reader (Bio-Tek, Winooski, VT). The relative cell number was calculated as the ratio of the absorbance of the treated well versus that of the untreated control (Roforth et al., 2008). The student’s t-test was used for the statistical analysis (Sigma Plot, San Jose, CA). A p value < 0.5 was considered statistically significant.

3.1. Preparation and size measurement of paclitaxel/17-AAG-loaded micellar nanocarriers

17-AAG has previously been shown to sensitize breast, ovarian and lung cancer cells to the cytotoxicity of paclitaxel (Solit et al., 2003; Sain et al., 2006; Sawai et al., 2008). In human ovarian cancer SKOV-3 cells, we observed enhanced cytotoxicity when paclitaxel and 17-AAG were combined at a 1:7 molar ratio (see data in cytotoxicity study below). We therefore selected this ratio for all our subsequent formulation and biological studies involving the dual drugs. To develop a safe formulation for combined delivery of paclitaxel/17-AAG without employing organic solvents, we hypothesized that both drugs could be simultaneously entrapped into micellar nanocarriers. We chose to study PEG-PLA micelles and PEG-DSPE/TPGS mixed micelles for the following reasons: (1) PEG-PLA micelles are being evaluated in the clinic for delivering paclitaxel (Kim et al., 2004; Saif et al. 2010; Kim et al., 2007); (2) we have recently shown that PEG-DSPE/TPGS mixed micelles may serve as a promising drug carrier for 17-AAG (Chandran et al., 2010). Here, we noticed that, the dual drug-loaded micelles of both types were formed with minimal precipitation of either polymers or drugs. Exhibiting unimodal and narrow distribution, the average hydrodynamic diameters of paclitaxel/17-AAG-loaded PEG-DSPE/TPGS mixed micelles and PEG-PLA micelles were 11 ± 1 nm and 32 ± 1 nm, respectively (Fig. 1).

3.2. Release of paclitaxel and 17-AAG from micellar nanocarriers

To serve as true drug carriers and promote drug extravastation into tumors via the EPR effect, it is crucial for micelles to be able to entrap the loaded drug molecules for an extended period of time during circulation. To this end, we examined the release profiles of paclitaxel and 17-AAG from PEG-DSPE/TPGS mixed micelles and PEG-PLA micelles. As controls, the release of free paclitaxel and 17-AAG was studied to verify that the diffusion of drug molecules across the dialysis membrane was not a rate-limiting step during the release process. As illustrated in Fig. 2A and Table 1, the incorporation of paclitaxel and 17-AAG into PEG-DSPE/TPGS mixed micelles resulted in significantly prolonged release half-lives (t_{1/2, release}) for both drugs (14.17 h and 6.85 h, respectively), compared to those from PEG-PLA micelles (6.36 h and 1.64 h, respectively). The release of both free drugs into the sink was much faster than those from either micellar formulation, which indicates that during the entire process of the drug release from the dialysis cassette, the liberation of drugs from micelles is the rate-limiting step, and the obtained release kinetics thus reflects the rate of drug release from micelles.

Next, we carried out the release study in the presence of 90% serum to mimic in vivo circulation environment. As illustrated in Fig. 2B and Table 1, the release of both paclitaxel and 17-AAG was again significantly slower from PEG-DSPE/TPGS mixed micelles (t_{1/2, release} = 7.09 h and 4.83 h, respectively) than from PEG-PLA micelles (t_{1/2, release} = 4.23 h and 1.92 h, respectively). The presence of serum made the release of drug molecules from both micelle types faster overall, suggesting that serum proteins may have weaken the hydrophobic interactions between drug molecules and micelle-forming polymers. The only exception was that, the presence of serum in the PEG-PLA micelle sample appeared to slightly deter the release of 17-AAG. This may be attributed to the strong binding of 17-AAG to serum proteins (Egorin et al., 2001), which became a rate-limiting step in this case. Meanwhile, the release rate of free paclitaxel and 17-AAG was both reduced, owing to the extensive binding of both drugs to serum proteins. Taken together, the above results indicate that PEG-DSPE/TPGS mixed micelles can function as better drug carriers than PEG-PLA micelles for combined delivery of paclitaxel and 17-AAG. And we therefore focused on paclitaxel/17-AAG-loaded PEG-DSPE/TPGS mixed micelles for the remainder of this work.

3.3. Incorporation of paclitaxel and 17-AAG into PEG-DSPE/TPGS mixed micelles

¹H NMR analysis was carried out to verify the simultaneous incorporation of paclitaxel and 17-AAG into PEG-DSPE/TPGS mixed micelles. When drug molecules are loaded into the core-shell micelles formed by the amphiphilic copolymers, the molecular motion within the micellar core is significantly restricted, leading to broadened line width of ¹H signals from protons of drug molecules and the hydrophobic segments of the copolymers located within the micelle core. As a control, we first obtained the ¹H spectrum of a mixture of paclitaxel, 17-AAG, PEG-DSPE and TPGS, that were dissolved in DMSO-d₆ (Fig. 3A). As expected, the detected ¹H signals from all molecules in the control sample had narrow line widths, because these protons experience rapid molecular motions in DMSO solution. In contrast, following the formation of micelles (Fig. 3B), the ¹H NMR peaks from both DSPE (0.87 ppm and 1.28 ppm) and the vitamin E portion of TPGS (1.90–2.00 ppm) were much broader compared to those in the DMSO sample. These results are indicative that the micellar core is formed by DSPE and the lipophilic portion of TPGS. On the other hand, the micelle formation had minor impact on the ¹H NMR line width at 3.71 ppm, which originated from oxyethylene (O-CH₂-CH₂) protons of PEG, compared to its counterpart in DMSO-d₆. This indicates that the PEG moiety resides in the corona region of the micelle, where the local motion of the protons is largely unrestricted by the micelle formation. Importantly, the fact that ¹H signals of paclitaxel and 17-AAG in the aqueous micelle sample were undetectable (Fig. 3B) clearly demonstrates that both drugs are buried within the hydrophobic core of the micelles.

3.4. Maximum loading of paclitaxel and 17-AAG into PEG-DSPE/TPGS mixed micelles

By modulating the input concentrations of both drugs, a combination of paclitaxel and 17-AAG could be loaded into PEG-DSPE/TPGS mixed micelles at a fixed ratio precisely. Aimed at a molar ratio of 1:7, paclitaxel and 17-AAG could be maximally incorporated into PEG-DSPE/TPGS mixed micelles at about 1.81 ± 0.04 mM and 12.57 ± 1.42 mM, respectively. At these concentrations, the aqueous solubility of each drug was increased by more than 5,000- and 700-fold. The loading efficiency, defined as the ratio of the incorporated versus the input drug concentrations, was about 95% for paclitaxel and 90% for 17-AAG.

3.5. Storage stability of paclitaxel/17-AAG-loaded PEG-DSPE/TPGS mixed micelles

We examined the storage stability of the dual drug-loaded micelles by monitoring the drug concentrations in the micelle dispersion. We found that, maximally loaded micelles remained stable at 4 °C for two weeks with both drug concentrations nearly unaltered (< 10% loss in the drug content). At 37 °C the same micelles were stable and maintaining above 95% of initial drug concentrations for 48 h. By 1 week, however, there were nearly 35% reductions in the concentrations of both drugs.

3.6. Cytotoxicity of paclitaxel/17-AAG-loaded PEG-DSPE/TPGS mixed micelles

In order to evaluate the cytotoxicity of paclitaxel/17-AAG-loaded PEG-DSPE/TPGS mixed micelles, cell proliferation assay was performed in SKOV-3 cells. For comparison purpose, we also examined the cytotoxicity of individual drugs and their combination in free forms. At a 1:7 molar ratio, combination of free paclitaxel (5–10 nM) and 17-AAG (35–70) exhibited significantly enhanced cytotoxicity in SKOV-3 cells, compared to those of individual drugs (p < 0.05, Fig. 4). When both drugs were simultaneously formulated into PEG-DSPE/TPGS mixed micelles, paclitaxel/17-AAG remained as cytotoxic to SKOV-3 cells as the combined free drugs. As a vehicle control, empty PEG-DSPE/TPGS mixed micelles had little inhibition (<10 %) on the proliferation of SKOV-3 cells at concentrations present in the above drug-loaded micelles. These results demonstrate that PEG-DSPE/TPGS mixed micelle formulation retains the antiproliferative activity of paclitaxel/17-AAG combination.