FORMULATION AND OPTIMIZATION OF CURCUMIN SOLID DISPERSION PELLETS FOR IMPROVED SOLUBILITY

Objective: The present study was aimed at designing of solid dispersion based pellets of curcumin (Cu) for improving its solubility. Methods: Solid dispersion (SD) of Cu was prepared by the melt method using Poloxamer 407 (Pol 407) at a different weight ratio of Cu-Pol 407 (1:2, 1:3, 1:5, 1:7, 1:10). The solid dispersion was characterised by FTIR, SEM, DSC, XRD and evaluated for saturation solubility in water, drug content and in vitro dissolution. The pellets of Cu solid dispersion were prepared by extrusion spheronization technique and optimization was performed by 3 2 full factorial design. The pellets were evaluated for size distribution, flow properties, hardness, disintegration and in vitro drug dissolution. Results: From the phase solubility analysis, Pol 407 was selected as a Solid dispersion carrier. The formation of Cu-SD by melt method using Pol 407, was confirmed from FTIR and DSC studies. XRD studies indicated a change of Cu from crystalline to amorphous form. There was a significant increase of Cu when formulated as SD compared to plain Cu. The optimization of extrusion spheronization process revealed the significant effect of Cu-Pol 407 ratio (p<0.0001) on in vitro dissolution of pellets. Higher Cu dissolution was obtained with Cu-SD pellets compared to plain Cu pellets. Conclusion: The present study demonstrated the potential of Cu-SD pellets in improving the solubility of poorly soluble Cu.


INTRODUCTION
Curcumin (Cu) is a phenolic phytoconstituent from the curcuminoid group and is obtained from curcuma Longa (family, zingiberaceae), traditionally known as turmeric. It is widely used in Indian and Chinese traditional medicine [1] and also has been extensively studied for its anti-inflammatory [2], anticancer [3], antioxidants [4], antibacterial and wound healing activity [5]. The efficacy of Cu in hepatoprotection [1], inflammatory bowel disease [4] and lifestyle associated diseases is also reported. The clinical potential of Cu is limited owing to its low aqueous solubility and poor permeation resulting in lower bioavailability and extensive metabolism. As a result, Cu is classified into the BCS Class IV compound and its therapeutic potential is limited due to its instability in physiological conditions [6]. Various formulation approaches have been extensively utilized to address the problems of solubility and permeability of Cu, out of which, improving solubility and dissolution of Cu has been a major area of interest. Solid dispersion (SD) of Cu using various water-soluble polymers, polyvinylpyrrolidone (PVP) [7], D α-tocopheryl polyethylene glycol 1000 succinate [8], chremophore and polyethylene glycol (PEG), Eudragit E100 [9], carboxymethyl cellulose acetate butyrate [10], hydroxypropyl methylcellulose (HPMC) [11], and hydroxypropyl methylcellulose acetate succinate [12] and polyethylene glycol-15hydroxy stearate (Solutol HS 15) [13] are reported.
However, the efficiency of Cu to form a stable complex with polymer is questionable and a large amount of polymers are needed to obtain a desirable effect. With the availability of self-emulsifying and surface-active carrier with low melting point, the major focus is now shifted to SD formulation using these carriers. One of the interesting strategy is the combination of nanotechnology and SD [14]. Nanomicellization is achieved with the help of amphiphilic carrier that can solubilize the drug in its hydrophobic core and also helps in improving the drug stability and solubility. Parika et al. [15] Reported self nanomicellizing SD of Cu using soluplus as the carrier. However, the use of a high amount of polymer in the selfnanomicellizing SD may pose a problem in handling and dispensing of this formulation. As pellets allow great design flexibility, ease of handling, can be easily packed in capsules, free-flowing and therefore commercially preferable, the pellet formulation of solid dispersion can address this problem.
In the present study, we have explored the potential of poloxamer 407 (Pol 407) as a carrier for the solid dispersion (SD) of the Cu. The solid dispersion of Cu was loaded on the pellet formulation and optimization was performed using design of experiment (DOE). The solid-state characterization of SD was done using FTIR, DSC, XRD AND SEM and the self micellizing properties were confirmed by particle size and zeta potential measurement. Pellet formulation was optimized on the basis of size distribution and in vitro dissolution.

Phase solubility analysis
A series of solutions (5, 10, 15, 20, 25, 30 mmol) of different carriers, HPβCD, Pol 188, Pol 407, PVP K30, and PEG 6000 were prepared in 0.1 N HCL. An excess amount of Cu was added to each carrier solution and the dispersion was kept to achieve equilibrium at 25 °C for 72 h with shaking at 100 rpm in an orbital shaker (Remi, CIS24b2, India). The dispersions were filtered through 0.45 µ membrane filter, suitably diluted with 0.1 N HCL and the drug concentration in solution was analyzed at 429 nm by UV spectrophotometry (Shimadzu, UV 1700, Japan). The experiment was performed in triplicate [16,17].

Preparation of solid dispersion by melt method
Solid dispersions of Cu were prepared with different weight ratio, 1:2, 1:3, 1:5, 1:7 and 1:10 of Cu-Pol 407. The carrier was melted at 55 °C, and then the Cu was dispersed to molten carrier with constant stirring followed by quick cooling. The obtained dried mass was then passed through sieve # 85 and stored in a refrigerator at 4 °C until further use [18,19].

Characterization of Cu-SD
Solid state characterization of Cu-SD and plain Cu were characterized by FTIR, DSC and XRD. The morphological characterization was done using SEM.

Scanning electron microscopy
The surface morphology of Cu-SD formulation was studied using field emission scanning electron microscopy (FESEM) (FEI, Nova Nano SEM 450, USA). The samples were coated with platinum layer of 20 nm to make them electrically conductive and the coated samples were then randomly scanned in the FESEM chamber at an acceleration voltage of 5.00 KV. The photomicrographs were obtained at different magnifications [16].

FT-IR study
IR absorption spectra of Cu, its physical mixture with excipients and SD formulation were recorded by potassium bromide dispersion technique in which dry samples and potassium bromide were placed in the sample holder and infrared spectrum was recorded using FTIR Spectrophotometer (Shimadzu, 8400S, Japan) over a range of 400-4000 cm -1 [20].

Differential scanning calorimetry study
In order to study the physical state of Cu and Pol 407, DSC (PerkinElmer, DSC 4000, USA) study was conducted for Cu, Pol 407 and Cu-SD. Accurately, 1 mg sample was weighed and placed in a sealed aluminum pan, and the sample was heated under nitrogen flow (20 ml/min) at a scanning rate of 10 °C per min from 30 to 350 °C. An empty aluminum pan was used as reference [10].

X-ray diffraction study
X-ray diffraction (Rigaku, Miniflex 600, Japan) was performed using Cu K 2α rays with a voltage of 40 kV and a current of 15 mA. Samples were scanned for 2θ from 20 ° to 80 °. Diffraction patterns for Cu, Pol 407 and SD of Cu-Pol 407 were obtained [21].

Particle size measurement
The particle size of Cu-SD was determined using dynamic light scattering (Horiba, SZ 100, Japan). Cu-SD was dissolved in 0.1 N HCL, and ultrasonicated for 10 min followed by measurement at fixed angle of 90 ° at 25 °C, carried out in triplicate [22].

Drug content study
Cu-SD equivalent to 5 mg of Cu was transferred to 50 ml volumetric flask, ultrasonicated to 15 min, and diluted to 50 ml with methanol. The solution was appropriately diluted by methanol and absorbance was noted at maximum wavelength, 419 nm. Drug content was determined using the calibration curve in methanol at the same wavelength. All the trials were done in triplicate [23,24].

Saturation solubility
The solubility of Cu and Cu-SD was determined in 0.1 N HCL. An excess amount of the Cu or Cu-SD was added into 5 ml of 0.1 N HCL up to saturation. Then the glass vials were placed into a water bath shaker for 72 h at 37 °C±0.5 and 100 rpm. The supernatant solutions were then passed through whatman filter paper and analyzed by UV spectrophotometry at 419 nm. All solubility measurements were performed in triplicate [25].

In vitro dissolution study
In vitro dissolution study was carried out using USP apparatus I (basket assembly). Accurately weighed sample equivalent to 20 mg of Cu-Pol 407 SD was filled into capsule and placed in a basket of dissolution vessel containing 900 ml of 0.1 N HCL as dissolution medium, maintained at 37±0.5 °C and 100 rpm. At each time interval, 5 ml of the sample was withdrawn and appropriately diluted. The equal volume of fresh dissolution medium was immediately replaced. The concentration of Cu in the sample was analyzed spectrophotometrically at 429 nm. The dissolution experiments were conducted in triplicate.

Preparation of Cu-SD pellets
The pellets of Cu and Cu-SD were prepared using extrusion and spheronization technique. The Cu or Cu-SD was mixed with all other excipients, except PVP K30 in a mortar. PVP K30 was dissolved in a mixture of IPA and water (2:1 ratio). The PVP solution was added to the powder mixture to form a damp mass. The wet mass was then passed through sieve number 18 (1000 µm) to obtain cylindrical extrudates. The extrudates were placed in a spheronizer (Shakti, SSP 120, India) fitted with the cross-hatched plate (2 mm) and spheronized for 5 min at varying rpm (1100, 1200, 1300 rpm). The resulting pellets were dried at 30 °C in a vacuum oven (Bio technique, BTI29, India) for 30 min [27]. The composition of pellet formulation is shown in table 1.

Experimental design
The optimization of the formulation was performed using 3 2 full factorial design. The design consisted of 2 variables at 3 levels. The first independent variable (X1) was Cu-Pol 407 ratio, (1:0, 1:3, 1:7) and the other variable (X2) was spheronization speed (1100, 1200, 1300 rpm). Total of nine formulations were prepared (table 2) and the effect on geometric mean diameter (Y1) of pellets, and drug release after 2 h (Y2) was evaluated. The data from 9 trial runs was analyzed using Design-Expert software (Stat-Ease, version 9.0, USA). The contour plots and 3-D surface response plots were generated in order to study the influence of independent variables on responses.

Particle size distribution
The particle size distribution of pellets was carried out by sieve analysis, using a set of USP standard sieves. Sieves set of #12, 16,18,22,24,30,36 and 44 were used along with a pellet load of 10 g. The sieve set was then mechanically shaken for 10 min. The net weight retained on each sieve was determined and these values were used for calculation of the particle size distribution [28].

Scanning electron microscopy
The surface morphology of Cu-SD pellet formulation was studied using field emission scanning electron microscopy (FESEM) (FEI, Nova Nano SEM 450, USA). The samples were coated with a platinum layer of 20 nm to make them electrically conductive and the coated samples were then randomly scanned in the FESEM chamber at an acceleration voltage of 5.00 KV. The photomicrographs were obtained at different magnifications

Micromeritic properties
Micromeritic properties like bulk and tapped density, Carr's index, compressibility index and angle of repose of pellet formulation were evaluated as per standard procedure mentioned in USP [28].

Hardness
The measurement of Hardness of pellets was done using a digital hardness tester (Veego, India).

Drug content
Drug content of pellets was estimated using UV visible spectrophotometric method. Pellets were crushed to powder using mortar and pestle. Finely crushed sample equivalent to 5 mg of Cu was transferred to 50 ml methanol, ultrasonicated to 15 min and after appropriate dilution, absorbance was noted at 419 nm. The experiment was performed in triplicate and drug content was calculated using a previously developed calibration curve in methanol [29].

In vitro dissolution study
In vitro dissolution study of pellets was performed using the same method as mentioned in the characterization of Cu-SD.

Stability study
The stability study of pellet formulation was performed at conditions as per ICH guidelines for Zone IV. Pellet sample was filled in capsules, which was further placed in HDPE container. The samples were maintained in the stability chamber with accelerated conditions (40 °C±2 °C, 75±5% RH), and intermediate condition (30 °C±2 °C, 65±5%). The sample were analyzed for appearance, drug content and in vitro drug release at 0 and 1 mo time points at both the conditions.

Phase solubility analysis
The solubility of Cu in 0.1 N HCL was observed as 0.0018 mmol. With increasing carrier concentration over the range of 5 to 30 mmol, the Cu solubility was found to be increased in a linear fashion in all the carriers as shown in fig. 1

Scanning electron microscopy
Scanning electron micrograph of Cu revealed its crystalline structure and Pol 407 appeared as perfectly spherical smooth-surfaced particles ( fig. 2). The SEM of Cu-SD showed agglomerated particles with a rough surface.

FT-IR study
The IR spectra of pure Cu ( fig. 3)

Differential scanning calorimetry study
The solid-state changes in Cu-SD were studied by comparing DSC of Cu, Pol 407, and Cu-Pol 407 SD (fig. 4). The crystallinity in Cu was confirmed by sharp endothermic peak at 178.66 °C. DSC of Pol 407 demonstrated endotherm at 54.9 °C. The endothermic peak was observed at 53.16 °C in Cu-SD. The peak for Cu was not observed. This confirmed molecular dispersion of Cu in Pol 407 carrier.

Fig. 4: Differential scanning calorimetry of (a) Cu (b) Pol 407 (c) Cu-SD X-ray diffraction study
The changes in the crystallinity of Cu in SD formulation were further confirmed by XRD.

Characterization of Cu-SD for self micellizing property
Pol 407 is a block copolymer with poly (ethylene oxide) (PEO) and poly (propylene oxide) (PPO) units. Above critical micelle concentration of 2.8x10 -6 M at 37 °C it forms micelles with PPO unit aggregated in the core and PEO units oriented towards the external aqueous phase. Micellar solubilization of lipophilic drugs could result in an increase in drug solubility. In order to confirm the formation of micelles by Pol 407, the particle size distribution was conducted after dispersion of SD in water (

Saturation solubility
The result of saturation solubility is depicted in fig. 6. The SD with ratio 1:3 demonstrated significant increase in solubility of Cu (158.71 mg/ml) as compared to only Cu (0.692 mg/ml) and SD with ratio 1:7 demonstrated solubility value of 333.72 mg/ml. Thus, the solubility of Cu was significantly increased with increasing ratio of Cu-Pol 407. This could be attributed to increase in formation of micelles with increase in Cu-Pol 407 ratio.  to raw Cu. When Cu-Pol 407 ratio was increased from 1:2 to 1:3 there was no significant increase in drug release. However, with increase in ratio to 1:7 and 1:10, the drug release was significantly increased. This is in agreement with the reports of Paradkar et al. [30] in which authors reported a solid dispersion system of Cu and PVP at different weight ratio. The drug release of solid dispersion was significantly higher than pure Cu and ratio 1:7 and 1:10 demonstrated a significant increase in drug release as compared to lower ratios, 1:1, 1:3 and 1:5.
Considering the results of saturation solubility, particle size and in vitro dissolution, the solid dispersion ratios of 1:3 and 1:7 were selected for the formulation of solid dispersion loaded pellets.

Characterization of Cu-SD pellets
Pellet formulations of Cu-SD were prepared by extrusion spheronization technique in order to overcome the problems of agglomeration and poor flowability of solid dispersion due to higher poloxamer concentration. Cu-SD was 30% w/w and other excipients were 70% w/w in the pellet formulation. Pellet formulation by extrusion spheronization takes place through various stages, attrition, plastic deformation and agglomeration resulting into shape change from cylindrical dumbbells, ellipsoid to finally spherical shape pellets. Preliminary trials were performed in order to optimize the solvents for the pelletization process. Water: IPA (isopropyl alcohol) in 1:4 ratio was used as solvent. Increase in water proportion resulted in sticky product whereas use of only IPA resulted in friable pellets. From the preliminary trials, the spheronization speed was set in the range of 1100 to 1300 rpm.
The feasibility of extrusion spheronization process for the formulation of pellets containing Cu in free form and in complex form with Pol 407 was studied using Design of Experiments (DOE). (table 4) depicts the effect of independent variables Cu-Pol 407 ratio (X1) and spheronization speed (X2) on evaluation parameters. The pellet formulation indicated good flow properties and compressibility as evident from Hausner ratio and compressibility index. As the spheronization speed was increased from 1100 to 1300 rpm, there was decrease in the values of angle of repose. For Cu pellets without SD (1:0 ratio), the dgeo (geometric mean diameter) of pellet increased with increase in spheronization speed ( fig. 8). However, with Cu-SD pellets, there was no significant change in dgeo with change in spheronization speed. During Cu pellet formation, higher speed of spheronization resulted in higher attrition producing large amount of fines. These fines, during dumbbell phase got deposited in the waist region resulting in formation of larger pellets. In Cu-SD pellet, presence of Pol 407 in high amount imparted excess cohesion and plasticity, thus decreasing the fine production. This resulted in narrow size distribution as compared to only Cu pellets. At higher speed, due to the deposition of excess fines, the pellet surface was rough resulting into increase in the value of angle of repose. The drug content for all the pellet formulation was in the range 81.69 to 93.69 %.  The high F value (1740.31) and p value less than 0.05 (p = 0.0001) imply the significance of the model. The predicted R 2 (0.9961) was in reasonable agreement with the adjusted R 2 (0.9991).  . 9). From the surface response plot, it was observed that when Cu-Pol 407 ratio was changed from 1:0 to 1:3, there was significant increase in the drug dissolution. This was due to presence of Pol 407 as solubility enhancer in the solid dispersion. Further increase in Cu-Pol 407 ratio to 1:7 increased the Cu dissolution, but the change was not very significant. The drug release from, SD pellets with Cu-Pol 407 ratio 1:3 was higher as compared to their SD formulation. This could be due to presence of hydrophilic carriers in pellet formulation. In addition, the uniform size of pellets provided a higher surface area for drug release than the agglomerated SD powdered formulations. There was no significant difference in the dissolution of pellets with different Cu-Pol 407 ratio SD. Change in spheronization speed did not significantly affect the dissolution of pellets.  . 9 A) indicated more than 50% Cu release for ratio of 1:3 and higher. Fig. 10 indicates dissolution profile of Cu and Cu-SD pellet. The result of micrometric properties indicated good flow and compressibility properties of pellet formulations (table 6). With increase in Cu-Pol 407 ratio in pellet formulation to 1:7, flow properties were adversely affected. Scanning electron micrograph of pellet formulation ( fig. 11) confirmed the spherical shape of pellets. The surface of pellets was found to be rough. This could be due to deposition of fines produced by attrition of pellets during spheronization. These results were in agreement with our observation during the optimization of the pelletization process.

Selection of optimized formulation of Cu-SD pellets
The criteria for the selection of optimized formulations were higher drug release and minimum geometric mean diameter. Based on the results of pellet size distribution and in vitro dissolution of pellets, F5 (Cu-Pol 407 ratio 1:3) was selected as an optimized formulation, since at this ratio, the higher release of Cu was obtained and further change in ratio did not significantly change dissolution profile. Fig. 12 indicates an overlay plot and table 7 indicates a comparison of predicted and experimental results. The percentage error in predicted and experimental values was only 0.14% and 0.48% indicating robustness of pellet formulation. This confirms the validation and pellet formulation process.

Stability study
The stability of optimized pellet formulation was studied for 1 mo and results are evident in table 8. From the data, no major change was observed in the appearance, drug content and drug release. This indicated the stability of the formulation at the test temperatures.

CONCLUSION
Cu-SD pellets were prepared by extrusion spheronization technique and the optimization of the formulation was performed by 3 2 full factorial design. Formulation of Cu-SD was confirmed from FTIR and DSC studies. The SD was found to increase the Cu solubility and dissolution significantly. The optimized Cu-SD pellets were spherical with narrow size distribution and indicated significant enhancement in Cu dissolution. The formulation was stable and the process was robust as confirmed from validation results. Conclusively, Cu-SD pellets presents a promising approach indicating an improvement in Cu dissolution.