FORMULATION , CHARACTERIZATION , STABILITY , INVITRO EVALUATION AND OPTIMIZATION OF DIACEREIN NIOSOMES

Niosomes or non-ionic surfactants vesicles are microscopic lamellar structures formed on the admixture of a non-ionic surfactant, cholesterol and stearylamine with subsequent hydration in aqueous media. The delivery of drugs by “vesicular drug delivery system” such as niosomes provides several important advantages over conventional drug therapy. Diacerein is an Interleukin-1 inhibitor and it is highly effective in relieving the symptoms of osteoarthritis. Diacerein, in contrast to NSAIDs, are potent inhibitor of IL-1 beta induced nitric oxide production by chondrocytes and cartilage, without reducing prostaglandin E2 production. The main objective of this study was to design suitable niosome-encapsulated drug delivery for anti-inflammatory drugs like Diacerein and evaluate the vesicle size, entrapment efficiency, in vitro release and physical stability of the system. Non-ionic surfactants used were Tween (40&60), cholesterol and stearylamine in molar ratio 1:1:0.1. The niosomes were prepared by thin film hydration method. The higher entrapment efficiency was observed with niosome (F11) prepared from tween 60, cholesterol and 2.5 min sonication. The release pattern shown by these formulations were first order & Higuchi diffusion controlled mechanism. The physical stability study show that niosomal preparation stored at refrigerated temperature for 60 days show maximum drug retained compare to room temperature and elevated temperature conditions. Finding of all this investigation conclusively demonstrate prolongation of drug release at a constant and controlled rate after niosomal encapsulation of Diacerein.


INTRODUCTION
The basic goal of drug therapy is to achieve a steady state blood or tissue level that is therapeutically effective and nontoxic for an extended period.The design of proper dosage regimen is an important element in accomplishing this goal (Satturwar et al., 2002).Novel drug delivery systems aim to deliver the drug at a rate directed by the needs of the body during the period of treatment and channel the active entity to the site of action (Biju et al., 2006).Targeted drug delivery implies for selective and effective localization of pharmacologically active moiety at preidentified (preselected) targeted (s) in therapeutic concentration while restricting its access to nontarget normal cellular linings thus minimizing toxic effects and maximizing therapeutic index.Targeted drug delivery is an event where a drug carrier complex/conjugate delivers drug (s) exclusively to the preselected targeted cells in a specific manner (Vyas and Khar, 2004).To pursue optical drug action, functional molecules could be transported by a carrier to the site of action and released to perform their task (Shahiwala and Misra, 2002).
The targeting methods may be classified as chemical methods, co-valent bonding and physical methods.Chemical methods involve chemical modification of the parent compound to a derivative, which is activated only at the target site.Various physical methods make use of the carriers such as liposomes, niosomes, resealed erythrocytes, nano-particles, platelets, magnetic microspheres, and monoclonal antibodies.Recently niosomal drug delivery system (A particulate colloidal carrier system) is drawing attention due to its significant advantages over conventional drug delivery system.It is reported that niosomes are non-ionic surfactant vesicles inclosing an aqueous phase and a wide range of molecules could be encapsulated within aqueous spaces of lipid membrane vesicles.Niosomes or non-ionic surfactants vesicles are microscopic lamellar structures formed on the admixture of a non-ionic surfactant, cholesterol and stearylamine with subsequent hydration in aqueous media (Sheena et al., 1998).Diacerein directly inhibits IL-1 synthesis, release, down modulate IL-1 induced activities and have been shown to posses disease modifying effect in experimental models of osteoarthritis and in human subjects with finger joint and knee osteoarthritis, (Fidelix et al., 2006) The present study was aimed for formulating niosomes of diacerein (IL-1 inhibitor), optimizing the formulation, characterizing them and assessing in vitro performance of the system.

Materials and Equibement
Diacerein (NUTRA Specialities Private Limited) was obtained as a gift sample from ADCO, Egypt.Tweens (40 and 60) were procured from El-Nasr Pharmaceutical Chemicals Co (ADWIC), Cairo, Egypt.Cholesterol was procured MERCK, E.Merck, and Darmstadt.Stearylamine: purchased from Fluka, Sigma-Aldrich chemie Riedstr.2,Germany.Solvents and other reagents were of analytical grade.All the Ingredients were used without further purification.Phosphate Buffer saline (PBS) pH 7.4 was prepared as described in the Indian Pharmacopia1996.

Preparation of Diacerein Niosomes
Niosomes were prepared by using thin film hydration method.Drug, nonionic surfactants and cholesterol were taken in molar ratio 1:1.Different niosomal formulations were prepared by thin film hydration technique reported by Azmin et al, (1985).Accurately weighed quantities of surfactant (either tween 40 or 60) and cholesterol, were dissolved in 10ml of chloroform in a round bottom flask (Abdulhasan et al., 2010).The solvent mixture was evaporated in a rotary evaporator under reduced pressure at a temperature of 60 ± 5 0 C and the flask rotated until a smooth, dry film was obtained.The film was hydrated with 25 ml of PBS 7.4 containing Diacerein (0.5%) at 60 o C with gentle shaking on a water bath.The niosomal suspension was then transferred to a suitable glass container and sonicated using probe sonicator in an ice bath for heat dissipation.The sonicated dispersion was then allowed to stand for about 2 hours at room temperature to form niosomes.The formulation was stored in refrigerator (Sakthivel et al., 2012).

Photo microscopy
Vesicle dispersions were characterized by photo microscopy for vesicle formation and morphology.Samples of Niosomal formulations were examined under optical microscope by means of fitted camera and photographed at magnification of 40 to 100 X (Abdelbary and Elgendy, 2008).

Determination of vesicle size
This is performed for characterization of vesicle's size.Vesicle size of niosomes were determined by using Malvern Mastersizer (Abdelbary and Elgendy, 2008).

Determination of Diacerein entrapment efficiency
The prepared Diacerein niosomes were separated from unentrapped drug by centrifugation at 7000 rpm for 30 minutes.The isolated layers were washed twice with PBS 7.4 and recentrifuged again (El- Ridy et al., 2008;Hu et al., 1999 andSilver et al., 1985).The amount of Diacerein entrapped was estimated indirectly by measuring the unentrapped drug in the washing and subtracting it from the total initial amount of Diacerein used at the start of the niosomes preparation.

In vitro release of Diacerein from niosomes
All niosomal formulae were employed in this examination.Each preparation was separated, washed and the amount of Diacerein entrapped was determined (as mentioned above).
The amount of drug retained at zero time was considered as the total amount of drug.The pellet of each preparation was then suspended using phosphate buffer solution (PBS) 7.4 to exactly 500 ml.The Rotary shaker was used to carry out the experiment.The device was adjusted to a rate of 80 stroke/min and the temperature was adjusted to 37-40 o C. A 5 ml sample from each of the niosomal suspension was taken at different time interval.The samples were separated and filtered through 0.45 µm filter, the amount of Diacerein released was determined at each time interval and the amount of Diacerein retained was then calculated at each time interval for each formula.

Optimization of Diacerein niosomes:
Statistical Correlation between Independent Variables (Charge inducer percent X1, HLB value X2 and Sonication time X3) and dependent response of Diacerein niosomes (Particle size Y1, Entrapment efficiency Y2 and In vitro release after 8 hrs Y3) using Statistical package STATGRAPHICS plus.

Physical Stability of Diacerein Niosomes
Physical stability of the prepared diacerein niosomes were carried out to investigate the leaching of down from niosomes (in a liquid form) during storage.The samples of niosomal formulations were sealed in a glass vial and stored at refrigeration temperature (4o C), room temperature and elevated temperature (40oC) for a period of 2 months.Samples from each vial were withdrawn at definite time intervals, 15, 22, 30, 45 & 60 days, the residual amount of the drug in the vesicles was determined as described previously after separation from unentrapped drug (Singh et al., 2011).

Photo microscopy
The photomicrograph of Diacerein niosomes prepared by thin film hydration method is shown in figure (1).They reveal that the niosomes were spherical in shape and exist in disperse and aggregate collections.

Determination of vesicle size
The means particle diameters of niosomes, composed of tween 40 and 60 with cholesterol are shown in table (2).The results reveal that formula 9(tween 40) has the smallest particle diameter (7.33 um) while Formula 11(tween 60) has the largest particle diameter (23.66 um).

Determination of Entrapment efficiency:
The entrapment efficiencies of all niosomal formulations composed of tween 40 and 60 with cholesterol are reported in Table (3).The results reveal that formula 11(tween 60) has the highest entrapment efficiency (58.43 %) while Formula 15(tween 60&40) has the smallest entrapment efficiency (9.52 %).

Optimization: Factorial Characterization of Diacerein niosomes
The experimental runs and the observed responses for the Diacerein formulations are shown in table (5).The dependent variables studied were Y1 (particle size), Y2 (Entrapment efficiency) and Y3 (Release after 8 hrs) based on the experimental design.The range of the responses for Y1 was 23.66 um in F11 (maximum) and 7.33 um in F9 (minimum).While in Y2, the range of the responses was 58.43 % in F11 (maximum) and 9.52 % in F15 (minimum).The range of the responses for Y3 was 97.5 % in F1 (maximum) and 89.1 % in F7 (minimum).
The relationship between the dependent and independent variables was further elucidated using main effect plot.showed the effects of factors X1, X2 and X3 Formula 4 Formula 7 Formula 8 Formula 13 Formula 14 Formula 15 on the response Y1, Y2 and Y3.The results given in these figures were manipulated in details as following:

Table (5): Full factorial design layout
Effect of X1, X2 and X3 on Y1 (particle size) Figure ( 5) standardized Pareto chart and figures (6-7) showed the main effects, interaction effects and quadratic effects of charge inducer (X1), HLB values (X2) and sonication time (X3) on the particle size.From the figures it was obvious that (X2) had the main effects on the particle size.Also it was noted that increasing X1 from 0% to 10% resulted in decreasing particle size from 22.5 um to 19.5 um (negative effect); increasing X2 from 14.9 to 15.6 resulted in increasing particle size from 21.6 um to 22.78 um then decreasing to 14.7 um (negative effect); and increasing X3 from 0 min to 10 min resulted in increasing particle size from 18.4 um to 21.8 um then decreasing to 19.3 um (positive effect).Table (6) showed the ANOVA for the particle size.The statistical significance of each effect was tested by comparing the mean square against an estimate of the experimental error.In this case it was noted that the HLB value (X2) had p-value less than 0.05 indicating that it significantly different from zero at 95% confidence level.The R-squared statistic indicates that the model as fitted explains 80.75 % of the variability in the particle size.The adjusted Rsquared statistic, which is more suitable for comparing models with different number of independent variables, is 46.12 %.The standard error of the estimate shows standard deviation of the residuals to be 3.167.Since the DW value is greater than 1.4 (2.436), there is probably not any serious autocorrelation in the residuals.

Figure (7):
Main effect plot showing the interaction effect of X1, X2 and X3 on the particle size

Effect of X1, X2 and X3 on Y2 (entrapment efficiency)
Figure ( 8) standardized Pareto chart and figures (9-10) showed the main effects, interaction effects and quadratic effects of charge inducer (X1), HLB values (X2) and sonication time (X3) on the entrapment efficiency.From the figures it was obvious that (X3) 2 , X1, (X2) 2 , X2 and X2X3 respectively had the main effects on the entrapment efficiency.Also it was noted that increasing X1 from 0% to 10% resulted in decreasing entrapment efficiency from 28.2 to 2.2 (negative effect); increasing X2 from 14.9 to 15.6 decrease entrapment efficiency from 31.8 to 9.2 then increase to 15.8 (negative effect) and increasing X3 from 0 to 5 min resulted in decreasing entrapment efficiency from31.1 to 10.9 then increasing to 31 (no effect).Table (7) showed the ANOVA for the entrapment efficiency.The statistical significance of each effect was tested by comparing the mean square against an estimate of the experimental error.In this case it was noted that 5 effects (the charge inducer X1, HLB value (X2), (X2) 2 , sonication time (X2 X3) and (X3) 2 ) had p-value less than 0.05 indicating that it significantly different from zero at 95% confidence level.
The R-squared statistic indicates that the model as fitted explains 96.37 % of the variability in the entrapment efficiency.The adjusted R-squared statistic, which is more suitable for comparing models with different number of independent variables, is 89.85 %.The standard error of the estimate shows standard deviation of the residuals to be 5.75.The mean absolute error (MAE) of 2.83 is the average value of the residuals.The Durbin-Watson (DW) statistic tests the residuals to determine if there any significant correlation based on the order in which they occur in your data file.Since the DW value is less than 1.4 (1.2586), there is probably serious autocorrelation in the residuals.

Effect of X1, X2 and X3 on Y3 (release after 8 hours)
Figure ( 11) the standardized pareto chart and figures (12-13) showed the main effects, interaction effects and quadratic effects of charge inducer (X1), HLB value (X2) and sonication time (X3) on the release after eight hours.From the figures it was obvious that no factor had effects on release after eight hours.Also it was noted that increasing X1 from 0% to 10% resulted in decreasing release after eight hours from 96% to 93.9 (negative effect); increasing X2 from 14.9 to 15.6 decrease release after eight hours from 95.6% to 94.6% then increasing to 95.4%(negative effect); and increasing X3 from 0 to 5 min resulted in increasing release after eight hours from 91.8 % to 94.8% then decreasing to 93.2% (positive effect).
Table (8) showed the ANOVA for the release after eight hours.The statistical significance of each effect was tested by comparing the mean square against an estimate of the experimental error.In this case it was noted that none of the factors had p-value less than 0.05 indicating that it not significantly different from zero at 95% confidence level.The Rsquared statistic indicates that the model as fitted explains 54.89 % of the variability in the release after 8 hours.The adjusted R-squared statistics, which are more suitable for comparing models with different number of independent variables, is 0 %.The standard error of the estimate shows standard deviation of the residuals to be 2.827.The mean absolute error (MAE) of 1.428 is the average value of the residuals.The Durbin-Watson (DW) statistic tests the residuals to determine if there any significant correlation based on the order in which they occur in your data file.Since the DW value is greater than 1.4 (1.943), there is probably not any serious autocorrelation in the residuals.By applying the optimize response, the optimized formula containing Diacereinentrapped niosomes is obtained by using the independent variables as follow: Charge inducer (0 %), HLB (15.6) and sonication time (0 min).Table (9) showed the observed and the predicted values of the responses for the optimized formula of Diacerein niosome that suggested by Factorial design.

Release Kinetics
We determined the proper order of release of drug from different formulations by analyzing linear regression study.Zero, first and Higuchi diffusion controlled model equations were applied to all in vitro release results.From the results we can conclude that the drug was released from niosome by a zero, a first order and Higuchi diffusion controlled mechanism Table (10).

Physical Stability Study of Diacerein Niosomes
Physical stability study of the prepared niosomes was carried out to investigate the leaching of drug from niosomes during storage at refrigerator condition, room temperature and elevated temperature.The percentage of Diacerein retained after a period of 7, 15, 22, 30, 45 & 60 days in MLVs niosomes composed of tween 40 with cholesterol in molar ratio 1:1 are shown in table (11).Also the results indicate that maximum percentage drug retained was observed at refrigerated conditions than room temperature and elevated temperature, after 2 months study.This may be due to the higher fluidity of lipid bilayers at higher temperature resulting into higher drug leakage.

Figure ( 3 ):
Figure (3): In vitro release of Diacerein from tween 60 niosomes after 8 hrs residuals.The Durbin-Watson (DW) statistic tests the residuals to determine if there any significant correlation based on the order in which they occur in your data file.

Figure ( 8 ):
Figure (8): Standardized pareto chartshowing the quadratic effect and interaction effect of particle size X1, X2 and X3 on the entrapment Figure (11): Standardized pareto chart showing the quadratic effect and interaction effect of X1, X2 and X3 on the release after 8 hours.

Figure ( 12 ):
Figure (12): Main effect plot showing the effect of X1, X2 and X3 on the release after 8 hours.

Figure ( 13 ):
Figure (13):Main effect plot showing the interaction effect of X1, X2 and X3 on the release after 8 hours

In vitro release of Diacerein from niosomes
Results of an in vitro study on the release of diacerein niosomal vesicles prepared using Tween 40 and Tween 60 and mix of them are shown in Figs. 2, 3 and 4, respectively.The percentage of drug released after 8 h (Q8h) from the prepared niosomal vesicles are shown in Table (4).

Table (
The mean absolute error (MAE) of 1.516 is the average Interaction Plot for Particle size

Table ( 7
): Analysis of variance for Entrapement efficiency

Table ( 8
): Analysis of variance for Release after 8 hours

Table ( 9
): Observed and predicted values of the responses for the optimized Diacerein niosomes

Table ( 10 ) :
The Calculated Correlation Coefficients for The In-Vitro Release of Diacerein Niosomes prepared by Box-Behnken design Employing Different Kinetic Orders or Systems