Ondansetron- Bilosomes- Transdermal- Sodium deoxycholate- Permeation.
Cancer treatment is commonly followed by nausea and emesis, which is called by Chemotherapy-induced nausea and vomiting (CINV) and may affect the quality of patient life, discontinuance or decreasing the dose of chemotherapy. Around 70 % of patients, subjected to chemotherapy, suffer from emesis within or after 24 hours of administration, and 30% of patients suffer from anticipatory emesis (1).
CINV can be managed by number of drugs including; Dopamine antagonists, steroids or 5-HT3 receptor antagonists (2). 5-HT3 receptor antagonists showed superiority over other drugs in the prohibition and curing of CINV (3).
Ondansetron (OND) is commonly used for management of emesis after operations, radiotherapy or even chemotherapy. It is available in number of dosage forms; intravenous or intramuscular injections, tablets, oral solution, orally disintegrating tablets and suppositories (4).
Ondansetron suffers from low bioavailability (about 60%) as a result of first-pass metabolism effect which results and rapid elimination (t1/2; 3-5 h). Moreover, frequent dosing and invasive dosage forms of ondansetron decrease patient compliance, hence, comes the importance of developing a more safe, effective and convenient OND delivery system for cancer patients who receive chemotherapy (4). Therefore, transdermal delivery may represent a good tool to increase convenience administration to the patient (5). Moreover, physicochemical properties of Ondansetron hydrochloride (Mw: 365.15; logP: 2.07; logD (pH5.5): 0.19) increase chances of the transdermal delivery of its molecules(6).
Treatment of the intercellular lipid barrier of stratum corneum (SC) by colloidal vesicles makes it more loose and permeable (7). Lipid components of these vesicles enhance penetration and drug permeation by fluidization effect of the SC layer (8).
Many vesicular carriers such as liposomes (9), niosomes (10), transferosomes (11) and ethosomes (12) have been tried for their transdermal delivery to avoid the problems associating administering drugs by oral dosage forms. Bilosomes vesicles have same composition of niosomes with the addition of bile salts and were first reported by Conacher et al. (13). Bilosomes were previously studied as gastro-intestinal resistant vesicular carriers for vaccines to promote their oral delivery (14–16). They were also studied for transdermal drug delivery of tenoxicam by Al-mahallawi et al. (17).
The aim of this study was to develop OND-loaded bilosomal systems in comparison to niosomes in an attempt to promote transdermal delivery of OND, thereby, increasing its bioavailability. A full 32.21 factorial design was used to determine the effect of different formulation variables and to calculate highest desirability values for the best achieved bilosomal systems in comparison to OND-loaded niosomes. In-vitro permeation through cellulose membrane was conducted and desirability of all the prepared systems was determined to select systems for further analysis. Ex-vivo, In-vivo histopathological and permeation studies were accomplished for further assessment.
Materials and methods
Ondansetron hydrochloride (OND) and Emerest® ampoules (Ondansetron hydrochloride 2mg/mL) were generously donated by Global Napi Pharmaceuticals (Cairo, Egypt). Ondansetron, olmesartan (internal standard), span® 60, span® 80, sodium deoxycholate (SDC) and cellulose acetate dialysis tubing (M.wt cut-off 12,000-14,000) were purchased from Sigma Aldrich (St. Louis, MO). Cholesterol (CHL) NF was purchased from Parchem Industries (New rochelle, NY). Sodium lauryl sulfate, sodium chloride, potassium chloride, potassium dihydrogen orthophosphate-1-hydrate, disodium hydrogen orthophosphate-1-hydrate, and absolute ethyl alcohol were purchased from El Nasr Chemicals (Abuzabal, Egypt).
Preparation of OND-loaded niosomes and bilosomes
The thin-film hydration method was applied for preparation of OND-loaded niosomes and bilosomes using a surfactant, CHL and SDC (18). The investigated variables included; surfactant type (Span® 60 or Span® 80), surfactant: CHL molar ratio (7:0, 7:1, and 7:3) and SDC concentration (0, 2.5, and 5%, w/v). To prepare OND-loaded niosomes, OND (20 mg), the surfactant (Span® 60 or Span® 80) and CHL (if present) were added to ethyl alcohol (25 mL) (19) in a round bottom flask and dissolved via utilizing ultrasonic bath sonicator (ElmaS30H; Wetzikon, Switzerland) for 10 minutes (40 ºC). Rotary evaporator (Büchi® Rotavapor®; Flawil, Switzerland) was used for evaporation of the organic solvent under reduced pressure (60 ºC, 150 rpm, 5 minutes), till formation of the desired film on the inner wall of the flask. Distilled water (10 mL) was then added for hydration of the produced film under normal pressure at 700 rpm for 45 to confirm complete hydration and formation of OND-loaded niosomes. With respect to OND-loaded bilosomes, the same technique was employed except for incorporating SDC in the hydrating aqueous phase. To obtain uniform particle size dispersions, the developed systems were sonicated for 10 minutes (25 ºC). The prepared niosomes and bilosomal systems were then placed at 4 ºC until use.
Characterization of OND-loaded niosomes and bilosomes
Evaluation of polydispersity index (PDI), particle size (PS), and zeta potential (ZP).
The average PS of each dispersion was measured by photon correlation spectroscopy (PCS) which depends on analyzing the changes in scattered light intensity due to the random motion of particles (20), each dispersion was diluted (10 times) with deionized water before being analyzed. Measurements were carried out, in triplicate, at 90º to the incident beam using a Zetasizer (Malvern; Worcester-shire, UK) at 25±0.5 ºC. The zeta potential of each dispersion was measured in triplicate, at the same temperature, based on electrophoretic light scattering technology via laser Doppler Anemometer attached to Zetasizer. Lower PDI values reveals uniform and better distribution for particle size.
Determination of OND entrapment efficiency percentage (EE %).
The percentage of OND entrapped in each prepared system was determined, in triplicate, by determining the free (non-entrapped) OND. One mL of the dispersion was centrifuged via a cooling centrifuge (Sigma Laborzentrifugen GmbH; Osterode am Harz, Germany) at 4 ºC and 15,000 rpm for 1 hour. The supernatant was separated, diluted, and the concentration of OND was calculated after measuring the UV absorbance at λ max 303 nm using a spectrophotometer (UV-1800 Shimadzu; Kyoto, Japan). OND EE% was calculated as following:
OND EE%=[(total amount of OND- amount of free OND)/(total amount of OND)]×100
Transmission electron microscopy (TEM)
The morphological characteristics of representative OND-loaded bilosomal system was observed using TEM (H-7500, Hitachi; Tokyo, Japan). The selected bilosomal system dispersion was diluted, then, one drop of the dispersion was placed on copper grid, stained with phosphotungstic acid 1%, w/v and then air-dried for 10 minutes at 25 ºC prior to TEM examination.
Solid state characterization of OND-loaded bilosomes
The following characterization studies were conducted on representative lyophilized OND-loaded bilosomal system using freeze dryer (Martin Christ Company; Osterode, Germany) under vaccum -0.016 mbar, at -57 ºC for 24 hours.
a) Differential scanning calorimetry (DSC). The thermal characteristics of pure OND hydrochloride, SDC, CHL, Span® 60, physical mixture of OND with other bilosomes’ ingredients and lyophilized OND-loaded bilosomal system were examined using DSC (Shimadzu; Kyoto, Japan). The device was calibrated with purified nitrogen (99.9%). Samples (4 mg) were weighed accurately and placed in standard aluminum pans. Heating temperature range was from 10 ºC to 400 ºC and scanning rate was 10 ºC/min.
b) Powder X-ray diffraction (PXRD). To affirm DSC analysis, the XRD analysis of the same samples were conducted (PertPro®, PANalytical; Arnhem, Netherlands) using Cu Ka radiation ( 50 Kv, 60 mA) in the angular region of 2Ө = 4º-70º (21).
In-vitro drug permeation study
Cellulose acetate dialysis tubing pieces were equilibrated overnight in Sorensens’ Phosphate buffer (pH 7.4) containing Sodium lauryl sulfate (0.4%, w/v) (22). The membrane was mounted on a diffusion cell (Hanson Research Corporation; chatsworth, CA). The membrane surface area available for diffusion was 1.767 cm2. OND-loaded bilosomal system dispersion (0.25 mL) was placed at the donor side, while the receptor side was filled with 7.2 mL of Sorensens’ phosphate buffer (pH 7.4) containing sodium lauryl sulfate (0.4%, w/v) to maintain sink conditions and stirred at 600 rpm (32 ± 0.5 ºC) (23). Samples from the receptor compartment (0.25 mL) were withdrawn at different time intervals till 8 hours, and immediately replaced with fresh buffer solution to keep up constant volume and sink conditions. The obtained samples were then analyzed spectrophotometrically. For comparative studies, the cumulative percentages of OND permeated after 0.5 hour (P0.5h) and 8 hours (P8h) were determined. The results were statistically analyzed at P= 0.05.
Studying the effect of the formulation variables via a 32.21 full factorial design
A full 32.21 factorial design was compiled to estimate and optimize the effect of formulation variables on OND-loaded bilosomal system’s characteristics using the minimum number of experiments (24). Three factors were selected as the independent variables. (i) The surfactant type (X1) was assessed at two levels (Span® 60 or Span® 80), (ii) surfactant: cholesterol molar ratio (X2) was assessed at three levels (7:0, 7:1, 7:3) and (iii) the bile salt concentration (X3) was evaluated at three levels (0, 2.5, 5%, w/v). The ZP (Y1), PS (Y2), PDI (Y3), EE% (Y4), P0.5h (Y5) and P8h (Y6) were specified as the dependent variables. Analysis of the experimental results was performed using Minitab software (ver.17, Minitab Inc.; London, UK).
Optimization of OND-loaded bilosomes
In order to determine best achieved systems, the desirability function was calculated for determination of the optimum levels of studied variables (25). The set criteria was achieving the minimal PS and PDI accomplished with the highest ZP (as an absolute value) and EE% and highest P0.5h and P8h.
Preparation of skin. Newly born rats (70 ± 20 g) were shaved at the abdominal surface 24 hours before the experiment day to allow healing of any possible inflammations that would happen (22). At the day of the experiment, rats were sacrificed prior to excising the abdominal skin (26). Skin surface was neatly cleaned and subcutaneous tissues adhering fats were removed without harming the epidermal surface. Skin was equilibrated in phosphate buffer saline for 2 hours prior to the experiment.
Ex-vivo permeation study. Permeation of OND, through excised rat skin, from the best achieved bilosomal systems was assessed in comparison to OND-loaded solution (Emerest® ampoules 2mg/mL) (control). The excised skin was mounted on a diffusion cell (dermis facing the receptor compartment and SC fronting the donor compartment) and experiment was completed as previously reported in the in-vitro study.
Data analysis. Plots were constructed using the cumulative amount of OND permeated per unit surface area of skin as a function of time (t, h). Slope of the steady state part of the plot was calculated which represents skin permeation rate at steady state; flux (Jss, µg/cm2/h). The permeability coefficient (Kp, cm/h) was calculated as previously reported (27).
Kp = Jss/C (1)
Where C is the drug concentration at the donor compartment.
The enhancement ratio (ER) was also calculated by dividing the flux values from bilosomal systems over that of the control.
In vivo studies
Experimental animals. Male Wistar rats (200 ± 20 g) were involved in histopathological and permeation studies. The design of the studies was approved by the ethical committee of Future University in Egypt (reference number: REC-FPSPI-2/15) and following the guidelines of Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC). The animals were derived from the animal house one week before starting the experiment in order to accommodate to the environment (temperature of 25 ± 1ºC, relative humidity of 55 ± 5% and alternate 12h light-dark cycles). The animals were put in rat cages and supplied with standard diet and water ad libitum. Twenty four animals were included in the in-vivo skin permeation study, while 6 animals were used in the histopathological study. A two-treatment, non-blind, randomized, parallel experimental design was followed in both studies.
Dose calculations. The calculations for estimating OND dose for rats were based on the body weight of rats according to surface area ratio (23). The surface area ratio would be 56 when the average body weight of rats and humans is 200 g and 70 kg respectively (28). Consequently the oral OND dose for rats would be 1.4 mg/kg while the transdermal dose would be 0.826 mg/kg (23,29).
In-vivo histopathological study. The studies were performed to evaluate the possible irritation potential and explore other tissue structural alterations (if present) in rat skin after application of the investigated treatments. The rats were randomly divided into 2 equal groups. Based on dose calculations, the animals in the first group were treated with 0.9% sodium chloride (treatment A; control), while those of the second group were subjected to the selected OND-loaded bilosomal system (B6) (treatment B; test) onto the skin surface over a constant area 1.767 cm2 and left for 8 hours. Following, the rats were sacrificed and the skin was excised for histopathological examination. Autopsy samples were taken from both groups, fixed in 10% formaline saline solution for 24 hours and dehydrated using serial dilutions of alcohol (methyl, ethyl and absolute ethyl). Specimens were cleared in xylene and immersed in paraffin in hot air oven (56 ºC, 24 h). Paraffin beeswax tissue blocks were sectioned, deparaffinized, stained with hematoxylin and eosin and finally examined under light electric microscope (30).
In-vivo skin permeation studies
Treatments were applied to constant shaved area of 1.767 cm2 according to the following pattern; OND solution (Emerest® ampoules 2 mg/mL) (Treatment A; control) and selected OND-loaded bilosomal system (B6) (Treatment B; test). Rats were mildly anaesthetized using ether. Blood samples were withdrawn from retro-orbital vein at time intervals; 0, 0.5, 1, 2, 4, 8, 24 hours. Samples were collected into heparin-treated tubes and centrifuged (4000 rpm, 10 min) for isolation of plasma (31). The plasma was stored at (-20 ºC) until analysis.
Sample processing. The thawed samples were mixed with ethylacetate (4 ml) and vortexed for 5 minutes. Organic layer was evaporated in vacuum concentrator (Eppendorf 5301; Hamburg, Germany) at 45 ºC. Residue was reconstituted in mobile phase (70% acetonitrile + 30% 0.01 m ammonium acetate) (0.5 mL) and analyzed.
Determination of OND concentration by LC–MS/MS. OND determination in human plasma was performed according to the method previously reported by Moreira et. al with slight modifications (32). Triple Quadrupole LC/MS/MS Mass Spectrometer (AB Sciex Instruments; Redwood City, CA) equipped with electrospray ionization (ESI) source operating at the positive ion mode (ES+) was used. Multiple reaction monitoring (MRM) mode was turned on for detection and mass analysis. The tuning parameters were adjusted for ondansetron and olmesartan. The used LC–MS/MS Mass system consisted of AGILENT ECLIPSE C18 column (50 mm*5 µm) (Agilent; Santa Clara, CA) operated at 25 ºC, with LC-20AD pump, and SIL-20A/HT autosampler (Shimadzu; Kyoto, Japan). Samples (7 µm) were injected. Mobile phase was eluted at a flow rate of 0.9400 mL/min. The M/Z ratios were 294/170 for Ondansetron and 447/207 for Olmesartan. The autosampler was maintained at 4ºC and the total run time was 2 min. A linear calibration curve (r2 = 0.999) was constructed between OND plasma concentrations (ng/mL) and ondansetron/olmesartan peak area ratios over the concentration range of 0.10–20 ng/mL. The lower limit of quantification was 0.10 ng/mL. Under these conditions, typical standard retention times were 0.46 and 0.54 min for ondansetron and olmesartan, respectively.
Pharmacokinetic and statistical analyses. The non-compartmental analysis was applied for calculation of the pharmacokinetic parameters of the two treatments using WinNonlin1 software (33). The area under the curve from zero to 24 h (AUC0–24, ng h/ml) and the area under the curve from zero to infinity (AUC0–∞, ng h/ml) were estimated by the log-linear trapezoidal rule with extrapolation of the terminal slope to infinity. The Maximum OND plasma concentration (Cmax, ng/ml), time elapsed to reach maximum OND plasma concentration (Tmax, h), mean residence time from zero to 24 h (MRT0-24) and relative bioavailability were determined. Results were expressed as mean values (±SD; n=6) except for Tmax which was expressed as median values. A one-way ANOVA test, at a P-value of 0.05, was applied to ensure the significance difference between the derived pharmacokinetic parameters of the two treatments except for Tmax values which was subjected to student’s T-test.
Results and discussion
Analysis of factorial design.
Factorial designs are employed to determine and analyze the factors affecting characteristics of a delivery system as previously reported by Araújo et al. (34). The factors selected and levels used were based on preliminary trials (data not shown) to determine the possible independent variables. Analysis of factorial design assured that the model can be used to navigate the design space as predicted R2 values were in agreement with the adjusted R2 (Table I) (35). The influence of the surfactant type (X1), surfactant: cholesterol molar ratio (X2), bile salt concentration (%, w/v) (X3) on the EE%, PS and ZP of vesicles is graphically illustrated as contour plots (Figure 1) and data are summarized in table II.
Effect of Formulation variables on PS and PDI values
Particle size influences vesicles’ penetration, as smaller vesicles penetrate deeper through the skin than larger ones(36). Particle size of the prepared bilosomes are presented as average diameter which represents the mean hydrodynamic diameter of the particles (37). The bilosomes were in the nano to microscale range as their PS fluctuated between 147.85 ± 4.35 nm (B3) to 3386.25±131.75 nm (B7). Based on the investigated design, the factors that have significantly contributed to the increase in the PS were surfactant type (X1) and SDC concentration (X3) (P= 0.000 and 0.001, respectively). Span® 60- based bilosomal systems showed smaller PS than those prepared with Span® 80 which came in agreement with what was previously reported by Taymouri et al. (38). Increasing SDC concentration resulted in decreasing the PS; this may be due to the effect of SDC on increasing membrane flexibility (39).
A zero polydispersity index values specifies wholly monodispersed particle population, while a PDI=1 suggests highly polydispersed vesicles (40). Nine bilosomal systems showed PDI values less than 0.5, revealing narrow size distribution and good homogeneity. Factors that influenced the PS also affected the PDI of vesicles’ dispersions (P< 0.0001). Vesicles prepared using Span® 60 showed smaller PDI values which may be due to their small PS compared to vesicles prepared using Span® 80 (41). It was found that there is an inverse proportion relationship between SDC concentration and PDI values.
Effect of formulation variables on ZP.
Zeta potential value gives an indication to the overall charges acquired by vesicles and helps to judge the stability of colloidal dispersions. The system is considered stable when ZP value is around ±30 mV due to electric repulsion between particles (42). Negative charges acquired by the investigated bilosomal systems are due to the presence SDC in the vesicular constructs (18). The obtained ZP values from investigated vesicles ranged from -34.85 ± 0.059 mV (N3) to -4.5 ±1.56 mV (N4). Most of the systems possessed sufficient charges that would inhibit the aggregation of the vesicles. Both surfactant type (X1) and SDC (X3) concentration significantly influenced ZP (P = 0.009 and 0.001 respectively). Absolute ZP values increased with the increase in bile salt concentration as they acquire negative charge. Vesicles prepared using Span® 60 showed higher absolute ZP values may be due to its greater ability to encapsulate OND compared to Span® 80. Ondansetron, as an acidic drug, ionizes and gains negative charge at neutral or alkaline pH, so, its presence in the vesicles in high concentration leads to the increase in the charge density of the vesicles (17).
Effect of formulation variables on EE%.
Entrapment efficiency gives a prospective for the ability of investigated vesicles to entrap significant amount of OND. The percent of OND entrapped within the vesicles ranged from 4% ± 0.24 (N5) to 47.17% ±6.01 (N1). High EE% values were observed with; Span® 60-based bilosomal systems prepared at Span®: CHL molar ratios of 7:0, 7:1 and SDC concentration of 2.5%, w/v. The surfactant type (X1), was found to have a significant effect on EE% (P < 0.0001). Span® 60 showed higher EE% than Span® 80, these results were found to be in agreement with the reports of Al-Mahallawi et al. (17) and Aburahma et al. (21) who investigated the influence of Span® type on drug entrapment in bilosomes and niosomes vesicles, respectively. The unsaturated bond in the alkyl chain of Span® 80 might have decreased EE% by increasing vesicles permeability (43). In addition, the lower transition temperature of span® 80 (Tc= -12 ºC) than that of Span® 60 (53 ºC) may be a reason for low EE% values of span® 80-based bilosomes (44). Surfactant: cholesterol molar ratio (X2) was not found to significantly influence the EE% (P< 0.401) (45). Regarding SDC concentration (X3), it was found to significantly influence the EE% (p< 0.001). Increasing SDC concentration from 0% to 2.5%, w/v was found to increase EE%. Further increase to 5%, w/v was not found to significantly increase EE% (46). Bile salts were reported by Chen et al.(46) and Sun et al. (47) to enhance drug solubility and consequently higher EE% at SDC concentration of 2.5%, w/v. Interestingly, further increase in SDC concentration had a negative impact on EE%. This may be due to increasing drug solubility in the dispersion medium due to the formation of mixed micelles (48,49). High concentration of bile salts can cause fluidization of the vesicles’ lipid bilayers and leakage of the entrapped drug as a consequence (21,50).
Transmission electron microscopy (TEM).
TEM analysis helps in the explanation of the morphological characteristics of dispersed systems (51). As observed from TEM images of the representative bilosomal system (B6), vesicles were with a predominant spherical shape as seen in figure 2A. The particle size was in good agreement with the results obtained with particle size analysis (Figure 2B).
Solid state characterization of OND-loaded bilosomes
DSC thermograms and X-ray diffractograms of pure OND, SDC, CHL, Span® 60, physical mixture and the lyophilized bilosomal system (B6) are shown in figure 3.
Differential scanning calorimetry (DSC) (Figure 3A). DSC was used to determine whether the drug has crystalline or amorphous characters and to detect any possible interactions with other components of the bilosomal system. All components confirming that it still retained its crystalline nature (17). The DSC scan of Span® 60, CHL exhibited single endothermic peaks at 57 ºC and 148.5 ºC corresponding to their melting points, respectively. OND showed a sharp endotherm at 186.27 ºC, corresponding to its melting point and a comparatively larger endotherm at 290.25 ºC corresponding to the degradation process (degradation from ondansetron pure drug) (52). Several endothermic peaks were observed in SDC thermogram at 82.86 ºC, 125.50 ºC, 147.51ºC, and 197.33 ºC, 233.94 ºC, 368.05 ºC; these peaks were similarly reported in reference (17), the broad endotherm that started at 125.86 ºC probably due to the loss of water molecules and was followed by an exothermic recrystallization peak at 147.51°C (17,53). The DSC of the physical mixture displayed the characteristic peaks of Span® 60, CHL, and SDC, while OND characteristic peak of each component with decreased intensity probably due to dilution (54). The DSC thermogram of the bilosomal system showed disappearance of peaks of all components of the system, suggesting possible conversion to the amorphous state and entrapped within bilosomes (55,56). Nasr et al. (57) reported that the absence of drug’s peak might indicate interaction with the surfactant bilayers of the vesicles. Disappearance of the SDC peak suggesting its fluidization in the surfactant lipid bilayer and formation of the bilosomes (58). Peaks of both Span® 60 and CHL disappeared which suggests the interaction between surfactant intercalated in the lipid bilayer, CHL and SDC (59). To confirm this suggestion, PXRD analysis was conducted.
b) Powder X-ray Diffraction analysis (PXRD) (Figure 3B). OND exhibited diffraction peak at 2Ө value of 24.20º. Cholesterol showed intense peaks at 2θ values of 15.66º, 16.85º and 18.65˚ (60). Span® 60 showed intense peak at 2θ value of 21.27º (61,62), SDC showed low intensity peak at 2θ value of 15.27º (63). The disappearance of the characteristic sharp peaks of all components in the bilosomal system and their permanence in that of the physical mixture confirm the DSC results and agrees with what was reported by Pattnaik et al. (52). Bilosomes XRD analysis shows the presence of lamellar lattice in case of bilosomes (64). This also indicates that components may have undergone solid state transition from crystalline to amorphous form or crystallinity was reduced in case of optimized bilosomes (65).
Effect of formulation variables on in-vitro drug permeation.
The influence of the surfactant type (X1), surfactant: cholesterol molar ratio (X2) and bile salt molar concentration (X3) on P0.5h and P8h was studied. Transdermal OND permeation at 0.5 h after application of the dispersion, gives a prospective whether system will give burst OND release. Both surfactant: cholesterol ratio (X2) and SDC concentration (X3) significantly influenced P0.5h (P= 0.009 and 0.002, respectively). Surfactant: cholesterol molar ratio of 7:1 showed higher permeation values than other systems. Bilosomal systems prepared at a SDC concentration 2.5%, w/v showed higher permeation values than other systems. These results comes in agreement with what was reported by Kravchenko et al. (66) that the inclusion of the cholesterol increased transdermal permeation of phenazepam. Needless to say SDC acts as penetration enhancer (67). Decreasing OND permeation upon increasing SDC concentration more than 2.5%, w/v may be due to increasing OND affinity to vehicle and decreasing its thermodynamic activity, which resulted in slower release and poor transfer from vehicle to skin (68).
Transdermal permeation values at 8 hours give an indication about the ability of the system to allow a sustained release profile. P8h was significantly affected by surfactant type (X1) and surfactant: cholesterol molar ratio (X2) (P= 0.001 and 0.000, respectively). Span® 60-based bilosomal systems showed higher P8h values than those prepared with Span® 80. In a parallel line, Span® 80 niosomes were reported to show slower permeation profiles than Span® 60 niosomes via rat skin (69). Surfactant: CHL molar ratio of 7:1 showed the highest permeation values among other used ratios (66). This came in agreement with previously observed P0.5h values.
Selection of the optimized systems
Desirability values revealed the superiority of Span® 60-based bilosomes and that the optimum SDC concentration and Surfactant: CHL molar ratio are 2.5% w/v and 7:1, respectively. Based on the calculated desirability values displayed in table II, six bilosomal systems, namely; (B1, B3, B4, B5, B6, and B10) exhibiting desirability values > 0.77 were selected for further investigations.
Ex-vivo skin permeation studies
Ex-vivo permeation gives prospective for the in-vivo performance of a transdermal drug delivery system. Figure 4 displays the cumulative amount of OND/cm2 permeated via selected bilosomal systems as a function of time in comparison with control. The permeation profiles and estimated flux values demonstrated superiority of bilosomes over OND solution. Highest drug permeation profile was observed with B6, while lowest permeation values were observed with the control solution.
Flux of all bilosomal systems were in the range from 18.39 ± 1.97 (B3) to 89.17 ± 6.66 µg/cm2/h (B6) (Table III). These values were significantly higher (p< 0.001) than the control formula (1.45±0.05 µg/cm2/h). Permeability coefficient values of the investigated systems ranged from 0.04±0.003 (B3) to 0.18±0.013 (B6) which was higher than control (0.003±0.000). These results indicate that bilosomes considerably improved the OND transdermal permeation. It was suggested that bile salts included in the construction of bilosomes causes fluidization effect of both the vesicular lipid bilayers and SC lipids (70). Bile salts included in the vesicles may act as an edge activator by destabilizing lipid bilayers, increasing membrane flexibility, hence, increasing drug permeation through the skin. On the other hand, because of being a penetration enhancer (71), bile salts alter the nature of skin horny layer and enhances transcellular permeation, also they have the ability to sequester calcium ions that maintain the integrity of the tight junction, causing them to open the paracellular route that becomes more leak (72). References also reported that water content in the vesicles increase penetration by hydration of SC also by causing widening of the channels in keratin layer and distortion of lipid bilayer which may be one of the causes of high bilosomes permeation (23). All these factors collectively help improve transdermal delivery of drugs using bilosomes.
In-vivo histopathological studies. Light microscopical examination of stained skin sections belonging to group (I) showed normal skin architecture with well-defined epidermis, dermis, subcutaneous tissue and muscles (Figure 5; A and B). As depicted in Figure 5 C and D (group II). In a parallel line, the application of bilosomes to rat skin led to very minor wrinkles which could be attributed to the action of the penetration enhancer, SDC (17). There were no severe signs of skin irritation or inflammation (e.g., edema or erythema). All layers of the skin were histologically intact. These results were in line with the work done by Al-mahallawi et al.,who showed that Span® 60, Span® 80, Cholesterol and SDC did not result in histopathological alterations after application of tenoxicam-loaded bilosomal systems (17). Minor changes observed from histopathological study in the top most keratinized layer (SC)
In-vivo transdermal permeation studies. In-vivo permeation profiles of the best achieved OND-loaded transdermal bilosomal system (B6) in comparison to oral OND solution are displayed in figure 6. The pharmacokinetic parameters are summarized in table IV. Following oral administration of OND solution, OND significantly (P = 0.001) showed higher Cmax value (153.773 ± 34.57 ng/ml) at Tmax 0.5 h. In fact, OND plasma concentration declined rapidly within the following hours. On contrary, after administration of the transdermal B6 bilosomal system, OND showed Cmax value of 85.972 ± 144.72 ng/ml at 2 hours and OND plasma concentrations declined slowly within the following hours. Based on the calculated AUC0-∞ values, B6 showed 164.2% increase in bioavailability compared to oral OND solution. The delayed median Tmax from 0.5 to 2 hours and the prolongation in the MRT from 2.91±0.744 h to 5.25±1.10 h could prove the sustained release characteristics of the investigated bilosomal system. These results clearly demonstrate that the transdermal permeation properties of OND and its delivery across the skin can be improved using bilosomes this may be due to the solubilization effect reported on bile salts which lead to enhancing vesicles permeation through SC. These results come in agreement with what was reported by Moghimipour et al. (66) that bile salts and specially SDC act as a very good transdermal penetration enhancer (73). Cholesterol as well resulted in increasing OND in-vivo transdermal permeation upon increasing its concentration at the carrier bilosomal constructs.
In this work, bilosomes were investigated as a possible carrier for transdermal delivery of OND in comparison to niosomes. A complete 32.21 factorial design was set and the desirability values recommended the use of SDC at concentration of 2.5%, w/v and surfactant: CHL molar ratio of 7:1 and Span® 60. Bilosomal systems displayed spherical morphology, reasonable drug EE% along with higher OND permeation capability compared to OND solution. The existence of OND within the structure of bilosomes was demonstrated by DSC and PXRD studies. The ex-vivo skin permeation, histopathological and in-vivo skin permeation studies suggested that bilosomes promoted drug permeation through rat skin and represented a good safe model for OND transdermal delivery.