All the chemicals and reagents were purchased from SigmaAldrich (St.Louis, Missouri, USA). Poly lactic-co-glycolic acid (PLGA) 50/50 (MW 153,000) was a generous gift from Purac Biomaterials (Gorinchem, Netherlands).

pH-Solubility profile: The aqueous solubility of FBZ in phosphate buffers with pH values ranging between 2 and 12 was determined. According to procedure described previously [15-17], excess FBZ was added to various vials containing 10 ml buffer, placed in a water-bath shaker at 37°C, and allowed to equilibrate for 24 hours. Following equilibration, the samples were centrifuged at 4,500 rpm for 5 minutes in a Thermo IEC Multi RF Centrifuge (Needham Heights, Massachusetts USA). The supernatants were collected, filtered through a 0.45 µm filter, and assayed for FBZ by HPLC. All experiments were performed in triplicates.

pKa-determination from FBZ-pH solubility profile: pKa values were determined from the pH-solubility profile based on the methods described previously [18,19] . From the equation provided:

EQUATION

where S_o is the solubility of the unionized form of FBZ in water, and S is the observed solubility of drug at specified pH values, a plot of the EQUATION versus pH was used to determine pKa from the x-axis intercept.

A 1-octanol/water partition coefficient was performed by placing 5 mg of FBZ in 5 mL phosphate buffer (0.1 M) with pH ranging from 2 to 10, that had been previously been saturated with 1-octanol. An equal volume of 1-octanol (5 mL), which served as the organic phase, was added to the aqueous phase. Tubes were rotated end-over-end for 5 hours and centrifuged for 10 minutes at 4,500 rpm. Finally the aqueous phase was carefully aspirated from the organic phase diluted, and amount of FBZ present was determined using UV/Vis spectrophotometry. The partition coefficient (Log P) was calculated using the equation:

EQUATION

Where V_aq is the volume of the aqueous phase, C_i and C_aq are the concentration of FBZ in the aqueous phase initially and after equilibration and V_o is the volume of 1-octanol.

Forced degradation studies were conducted as by Ohja et al. [20]. One milligram of FBZ was dissolved in 1M hydrochloric acid at room temperature and sampled over a period of 120 minutes and assayed for intact FBZ by HPLC.

HPLC analysis of FBZ: The mobile phase consisting of 60% acetonitrile and 40% of 50 mM phosphate buffer with 1mM EDTA was pumped at a flow rate of 1.0 mL/min through a Zorbax 300SBC18, 4.6×250 mm, 5 micron column (Agilent, Santa Clara CA) at room temperature. The sample injection volume was 20 µL and the FBZ was detected at 345 nm with SPD-M10AVP photodiode detector (Shimadzu, Columbia Maryland. USA). The data was collected and integrated using Shimadzu EZ Start 7.2.1

FBZ and methyl-β-cyclodextrin (mβCD) complexation: Solutions containing 1%, 2%, 5% and 10% mβCD were prepared in distilled water; FBZ was added to each solution, stirred for 48 hr and then centrifuged at 4,500 rpm for 10 minutes. The filtrate was obtained, by collecting the supernatant and subsequently filtering through a 0.45 µm filter [21,22]. The absorbances of the filtrates were obtained at various wavelengths using a SkanltMultiskan Spectrum spectrophotometer (Thermo Electron Corporation, Vantaa, Finland). The complexation constant (K1:1), according to the hypothesis of 1:1 stoichiometric ratio of complexes was calculated from phase solubility diagrams using the following equation:

EQUATION

Where, K_1:1 is the complexation constant, S_o is the intrinsic solubility, and the slope is calculated from a graph of the dissolved drug concentration versus mβCD concentration in the medium. The intrinsic solubility of FBZ in the absence of mβCD was determined directly in aqueous media.

The 1:2 complex (K_1:2) was calculated using an equation derived by Loftsson et al. below [23]:

EQUATION

Where S_t and CD is the total FBZ and mβCD concentration respectively, K_1:1 and K_1:2 represent the complexation constants for the 1:1 and 1:2 complexes respectively.

Chitosan gel containing FBZ/mβCD: Chitosan hydrogels were prepared by dissolving 400 or 800 mg chitosan in 20 ml of 1% glacial acetic acid solution. Twenty milliliters of 10% FBZ/mβCD complex was then added to each chitosan gel preparation and mixed thoroughly to form 1% and 2% chitosan gels containing FBZ/mβCD.

FBZ-PLGA nanoparticles preparation: Nanoparticles (FBZPLGA) were prepared using the nanoprecipitation technique previously described [24,25]. Poly (vinyl alcohol) (PVA) was employed in the formulation to increase drug uptake and release characteristics. Briefly, 1 mg of FBZ was added to 2 ml of 1% PLGA in acetone and stirred until dissolved and followed by the addition of 4% PVA solution in a drop-wise fashion in FBZ-PLGA under constant stirring at 1,500 rpm. The formed emulsion was stirred at room temperature for 2 hours to evaporate acetone. Larger particles and polymer clumps were removed by centrifuging the suspension at 5,000 rpm for 10 min. Blank PLGA nanoparticles (PLGAnps) without FBZ were also formulated similarly. The particle size and zeta potential of the all formulations (PLGAnps and FBZ-PLGAnps) were measured using BI-200SM laser light scattering system (Brookhaven Instruments, Holtsville, New York).

FBZ leakage from FBZ-PLGAnps: To assess the leakage of FBZ from PLGAnps, 3 ml nanoparticle suspension was transferred to a 50 kDa molecular weight cut off dialysis bag and dialyzed against 50 ml PBS. At predetermined intervals, 500 µl of dialysate was sampled and replaced with equal volume of fresh PBS buffer. The FBZ concentrations in the samples were determined by HPLC.

Madin Darby Canine Kidney (MDCK) cells were seeded on 96 well plates at 5 x 10³ cells per well and cultured in DMEM supplemented with 2 mM L-glutamine, 1.5 g/L sodium bicarbonate, 4.5 g/L glucose, 10 mM HEPES, 1 mM sodium pyruvate, 10% fetal bovine serum, and 1% penicillin/streptomycin. At 75% confluence, the MDCK cells were exposed to various formulations (PLGAnps, FBZ-PLGAnps, FBZmβCD and FBZ-1% chitosan gel) and cell viability was determined using Alamar blue assay at 6, 24 and 48 hr time points. At the end of each time point, the cells were washed with HBSS and further incubated for 5 h with 100 µl of Alamar blue plus cell medium. The absorbance of the produced color was measured using Skanlt Multiskan Spectrum spectrophotometer (Thermo Electron Corporation, Vantaa, Finland) at a wavelength of 550/580 nm.

Polarized MDCK cells were incubated with FBZ-PLGAnps conjugated-fluorescein isothiocyanate (FBZ-PLGAnps-FITC) at 37 °C with gentle shaking. After 1 hr, the FBZ-PLGAnps-FITC were removed, the cell surface was washed twice with PBS, fixed using 4% paraformaldehyde, mounted, and imaged with excitation wavelength of 488 nm and emission wavelength of 520 nm using Axiovert 100M microscope equipped with Zeiss LSM 510 laser confocal microscope (Carl Zeiss, Inc., Thornwood, NY).

No significant increase in FBZ solubility was observed from pH 2 to 10 as shown in Figure 1, however FBZ solubility increased significantly from pH 10 to pH 12 with a final solubility of 950 µg/ml. Incorporating pH-solubility data in equation 1, the pKa of FBZ was determined to be 10.95 (Figure 2) while the calculated log P using octanol/water (equation 2) was found to be 1.246.

Figure 3 shows log D versus pH profile of FBZ solubility at different pH points. From the profile, it is observed that the log D value increased slightly from 1.26 at pH 1 to 1.49 at pH 2 and thereafter remains unchanged to pH 10. Afterwards, log D decreased remarkably from 1.42 to -0.59 from pH range of 11 to 14. This suggests that log D against pH profile of drug FBZ is a characteristic of an acid drug where unionized group dominant at low pH. While at high pH, drug FBZ reaches ionized state and increases its hydrophilic nature. For the stability of FBZ in an acid environment, FBZ degradation was assessed in 1M HCL and the data obtained (Figure 4) revealed that over the period of 120 min, only 2.21% of FBZ was degraded. Overall, the stability profile of FBZ showed a degradation rate of 2.5×10^-5 min^-1.

Effects of mβCD on the aqueous solubility of FBZ was studied using the phase-solubility approach where FBZ was added to mβCD solution to determine the extent of FBZ solubility. From Figure 5 it was observed that solubility of FBZ increases as the concentration of cyclodextrin is increased showing a linear increase in solubility. This linear increase indicates a 1:1 complex formation with mβCD and the correlation coefficient of the linear regression of the phase solubility curve was than 0.99 indicating a good fit. The calculated complexation constant K1:1 of FBZ with mβCD was 0.00706 mM-1 using equation 3. In addition, the K1:2 complexation constant was calculated, based on equation 4, to be 8.98 x 10^-8 mM^-1 which was significantly lower than the K1:1 complex. The lower value of K1:2 compared with K1:1 indicates that the 1:1 complex was the predominant form.

In-vitro release of FBZ was conducted and the results show a linear relation between the FBZ released and amount of mβCD used to entrap FBZ. From Figure 6, we observed that the higher the amount of mβCD employed in loading FBZ, the greater the released amount of FBZ. This agrees with the findings in Figure 5 where increasing amount of mβCD improves the solubility of FBZ.

Particle size and zeta potential FBZ-PLGAnps were determined as 216.20 ± 6.20 nm and -5.47 ± 0.83 mV while that of PLGAnps were 147.68 ± 18.07 nm and -2.92 ± 0.64 mV (Table 1). FBZ-PLGAnps were about 1.46 fold larger than PLGA nanoparticles is mostly likely due to FBZ entrapment. We also observed that zeta potential of FBZ PLGAnps was 1.87 fold-high compared to that of PLGAnps. The negative charges exhibited by both nanoparticles are typical of PLGA polymers [24,26,27].

MDCK cells viability and internalization of 125 µg of FBZmβCD nanoparticles: To evaluate the effect of various FBZ entrapped nanoparticles on MDCK cells, we first determined the effect of increasing FZB concentration on the MDCK cells viability. After 48 hr of exposure, we observed that 0.01 µM of FBZ barely have any effect on MDCK cells while 60% of MDCK cells were still viable after incubating them with 10 µm FBZ (Figure 7A). Figure 7B shows the graphical representation of the viability of MDCK cells when exposed to varying concentrations of various nanoparticles. The MDCK cells did not show any significant decrease in viability when treated with PLGAnps, FBZ-PLGAnps and 125 µg of FBZ-mβCDnps for the exposed period of 6, 12 or 48 hr. However, 250 µg of FBZ-mβCDnps and 500 µg of FBZ-mβCDnps have a significant decrease on MDCK cells’ viability (p<0.05) when they were exposed to the cells for 48 hr. Of all the formulations, only 500 µg FBZ-1% chitosan gel was cytotoxic on MDCK cells compared with the control. On the cellular uptake of nanoparticles, 125 µg of FBZ-mβCDnps was poorly internalized by the polarized MDCK cell monolayer. Only chitosan coated FBZ loadedPLGAnps (FBZ-cPLGAnps), showed a significant accumulation in polarized MDCK cell monolayer as shown in Figure 8B, whereas, Figure 8A show the presence of nuclei in untreated MDCK cells.