Pharmaceutical nanotechnology

Optimization of long circulating mixed polymeric micelles containingvinpocetine using simple lattice mixture design, in vitro and in vivocharacterization

Rania Moataz El-Dahmya, Ibrahim Elsayed b,c,*, Ahmed Hassen Elshafeey b,d,Nabaweya Abdelaziz Abd El Gawad a, Omaima Naim El-Gazayerly b

a Pharmaceutics Department, Faculty of Pharmacy, October 6 University, EgyptbDepartment of Pharmaceutics and Industrial Pharmacy, Faculty of Pharmacy, Cairo University, Cairo, EgyptcCollege of Pharmacy, Gulf Medical University, Ajman, United Arab Emiratesd School of Pharmacy, University of Waterloo, ON, Canada

A R T I C L E I N F O

Article history:Received 10 September 2014Received in revised form 1 October 2014Accepted 3 October 2014Available online 5 October 2014

Pubchem classification:Vinpocetine (PubChem CID: 443955)Pluronics (PubChem CID: 24751)Mannitol (PubChem CID: 6251)Sodium lauryl sulphate (PubChem CID:3423265)

Keywords:PluronicVinpocetineMicellesLyophilizationSterilizationVivoMean residence time

A B S T R A C T

The aim of this study was to increase the in vivo mean residence time of vinpocetine after IV injectionutilizing long circulating mixed micellar systems. Mixed micelles were prepared using Pluronics L121,P123 and F127. The systems were characterized by testing their entrapment efficiency, particle size,polydispersity index, zeta potential, transmission electron microscopy and in vitro drug release. Simplelattice mixture design was planned for the optimization using Design-Expert1 software. The optimizedformula was lyophilized, sterilized and imaged by scanning electron microscope. Moreover, the in vivobehavior of the optimized formula was evaluated after IV injection in rabbits. The optimized formula,containing 68% w/w Pluronic L121 and 32% w/w Pluronic F127, had the highest desirability value (0.621).Entrapment efficiency, particle size, polydispersity index and zeta potential of the optimized formulawere 50.74 � 3.26%, 161.50 � 7.39 nm, 0.21 �0.03 and �22.42 � 1.72 mV, respectively. Lyophilization andsterilization did not affect the characteristics of the optimized formula. Upon in vivo investigation inrabbits, the optimized formula showed a significantly higher elimination half-life and mean residencetime than the market product. Finally, mixed micelles could be considered as a promising long circulatingnanocarrier for lipophilic drugs.

ã 2014 Elsevier B.V. All rights reserved.

1. Introduction

Vinpocetine, as a vincamine derivative, is used for the treatmentof cerebrovascular circulatory disorders as cerebral infarction andcerebral arteries cirrhosis (Hasa et al., 2011). It has limited oral usedue to its low water solubility (�5 mg/ml), slow dissolution rate,poor absorption and significant first-pass metabolism during firstpass that lead to very low bioavailability (�7%) (El-Laithy et al.,

2011). Additionally, it has very rapid elimination rate with shorthalf-life (2 h) that results in frequent drug dosing (three timesdaily) (Zhuang et al., 2010).

Vinpocetine was previously investigated after intravenousinjection for the control and treatment of both acute and chroniccerebral ischemia (Gulyás et al., 1998; Szakall et al., 1998).Intravenous administration could overcome some of vinpocetinedrawbacks as poor absorption and first pass, but its very low watersolubility and short elimination half-life still persist. Aqueoussolubility could be enhanced by decreasing the medium pH due tothe basic nature of vinpocetine (pKa = 7.1) (Hasa et al., 2011). Citricacid was previously utilized as an enhancer for the solubility anddissolution of vinpocetine (Ning et al., 2011). After vinpocetinesolubilization by pH modification, the only remaining difficultythat affects the reliability of intravenous administration is the

* Corresponding author at: Department of Pharmaceutics and IndustrialPharmacy, Faculty of Pharmacy, Cairo University, Kasr El-Aini, 11562, Cairo, Egypt.Tel.: +20 1149944306; fax: +20 233058256.

E-mail addresses: ibrahim.elsayed@pharma.cu.edu.eg, hema5th@yahoo.com(I. Elsayed).

http://dx.doi.org/10.1016/j.ijpharm.2014.10.0030378-5173/ã 2014 Elsevier B.V. All rights reserved.

International Journal of Pharmaceutics 477 (2014) 39–46

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsev ier .com/locate / i jpharm

rapid elimination rate. To overcome this problem, long circulatingnanocarriers could be utilized for formulation of intravenoussustained release vinpocetine.

Inclusion of hydrophobic drugs into polymeric micelles hasbeen found to be very attractive concept for solubilization andbioavailability enhancement (Zhao et al., 2012). Micelles contain-ing single polymer have been used in drug delivery for long time,but lately binary micelles replaced them. The disadvantages ofmono micellar systems include low drug loading, large particle sizeand low stability (Kulthe et al., 2011). On the other hand, mixedmicelles exhibit synergistic properties, such as increased micellestability, drug loading and entrapment efficiency (Gao et al., 2005,2008).

Pluronics are triblock copolymers of poly(ethylene oxide) (PEO)and poly(propylene oxide) (PPO) with the structure PEO–PPO–PEO.Previous studies demonstrated that pluronic copolymers can formspherical micelle by self-assembly (Parmar et al., 2011). Pluronicsshow low cytotoxicity and weak immunogenicity after topical andsystemic administration although they are not biodegradablepolymers. This may be due to having molecular weight less than15 kDa which could be easily filtered and cleared by the kidney(Batrakova et al., 2004; Huang et al., 2009).

Pluronic micelles have a core–shell structure with hydrophobiccore acting as solubilization depot for the non polar drugs andhydrophilic corona preventing aggregation, protein adsorption,and recognition by the reticulo-endothelial system (RES), whichleads to longer blood circulation time (Zhang et al., 2011). Also, thesmall size of polymeric micelles (<100 nm) could increase bloodcirculation time due to escaping from capturing by the mononu-clear phagocytic system in the liver and bypassing the filtration ofinter-endothelial cells in the spleen (Lu and Park, 2013). Presenceof pluronics with different hydrophilic–lipophilic balance (HLB)could help in achieving the optimum thermodynamic and kineticstabilities for the formed micelles. It was assumed that low HLBpluronics, would increase the thermodynamic stability of themicelles due to the tight hydrophobic interactions with hydropho-bic propylene oxide blocks. On the other hand, the high HLBpluronics would increase the kinetic stability of the micelles due tothe steric hindrance that minimize micelle aggregation (Abdelbaryand Tadros, 2013; Lee et al., 2011).

The aim of this work was to formulate mixed micelles using thelipophilic Pluronic L121 (HLB: 1–7), intermediate Pluronic P123(HLB: 7–12) and hydrophilic Pluronic F127 (HLB: 22) (Alexandridisand Lindman, 2000). Statistical design was utilized for optimiza-tion of the formulated micellar dispersion. The optimized formulawas lyophilized, sterilized and investigated for its in vivo behaviorafter intravenous injection in rabbits.

2. Materials and methods

2.1. Materials

Vinpocetine was kindly supplied by Nile Pharma, Cairo, Egypt.Pluronic L121 (PL121), Pluronic P123 (PP123), Pluronic F127(PF127), mannitol and sodium lauryl sulphate were purchasedfrom Sigma–Aldrich, St. Louis, USA. Potassium dihydrogenphosphate and disodium hydrogen phosphate were supplied bySISCO Research Laboratories Pvt. Ltd., Mumbai, India. All otherchemicals and solvents were of analytical grade and used withoutfurther purification.

2.2. Preparation of mixed micelles

Micelles were prepared using the thin film hydration technique(Wei et al., 2009). Briefly, vinpocetine (25 mg) and pluronic(s)(total weight: 1000 mg) were accurately weighed and dissolved in

10 mL chloroform:methanol mixture (in ratio, 2:1 v/v) in round–bottom flask (capacity, 250 mL). The organic solvent was evapo-rated under reduced pressure using rotary evaporator (Rotavapor,Heidolph VV 2000, Burladingen, Germany) revolving at 150 rpmfor 1 h at 60 �C. Hydration of the formed thin film was performedusing 20 mL distilled water. During hydration, the flask was rotatedat 150 rpm for 1 h at 60 �C under normal pressure. Finally, themicellar dispersion was sonicated using bath sonicator (Ultrasonicbath sonicator, Model SH 150-41, PCI Analytics Pvt. Ltd., Mumbai,India) for 2 min to reduce particle size (Oh et al., 2004).

2.3. Statistical design of the study

Simple lattice mixture experimental design was utilized toinvestigate the effects of the different variables on the character-istics of mixed micelles using Design-Expert1 software (Version 7,Stat-Ease Inc., MN, USA). The independent variables were thepercentages of PL121 (X1), PP123 (X2) and PF127 (X3). Monitoredresponses were entrapment efficiency (Y1: EE), particle size (Y2:PS), polydispersity index (Y3: PDI), as dependent variables. Table 1illustrates the composition of the prepared mixed micellesformulae. Ten formulae had been generated with 14 runs asformulae M1, M3, M8 and M10 were prepared twice. Formulaewere having the same total surfactant(s) to drug ratio withdifferent percentages of the three pluronics to investigate theeffect of hydrophilicity/lipophilicity on the micellar characteristics.Moreover, desirability values were calculated for the optimizationof variables to attain the superior formula.

2.4. Characterization of the mixed micelles formulae

2.4.1. Determination of entrapment efficiency (EE)Formulae were centrifugated at 14,000 rpm and temperature

4 �C for 1 h using cooling centrifuge (Beckman, Fullerton, Canada)to separate un-entrapped precipitated drug from the encapsulatednano-dispersed. Samples aliquots from supernatant were dilutedwith methanol to disrupt the micellar structure (Jin et al., 2011).Drug was spectrophotometrically analyzed in the diluted super-natant at lmax 268 nm. EE was calculated using the followingequation (Mu et al., 2010):

EE% ¼ weight of drug in micellesweight of added drug during preparation

� 100 (1)

2.4.2. Analysis of particle size, polydispersity index and zeta potentialThe particle size (PS) analysis of micelles was performed using

dynamic light scattering (Zetasizer Nano ZS-90, Malvern instru-ments, Worcestershire, UK). Before measurement, samples werediluted with distilled water, if required, until being translucent.Additionally, polydispersity index (PDI) was measured to assessthe particle size distribution. Finally, zeta potential (ZP) of thediluted formulae samples, having pH ranged between 5.5 and 6.5,was analyzed for evaluation of their physical stability. Measure-ments were done in triplicates for three independent samples ofeach formula and the average values �SD were calculated.

2.4.3. Morphological evaluation of the dispersed micelles bytransmission electron microscopy (TEM)

Morphology of the optimized formula was examined usingtransmission electron microscope (JEM-1230, Jeol, Tokyo, Japan).The negative staining technique was utilized as one drop of dilutedsample was placed on carbon coated copper grid and stained by 2%w/v phosphotungistic acid. The samples were investigated usingTEM at 100 kV, after drying at room temperature.

40 R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46

2.5. Lyophilization of the optimized micellar formula

The optimized formula was frozen at �20 �C in presence andabsence of 5% w/v mannitol as a cryoprotectant. Then, samples werelyophilized at �45 �C and pressure of 7 � 10�2mbar for 24 h(Novalyphe-NL 500; Savant Instruments Corp., USA). EE, PS, PDIand ZP were analyzed for the lyophilized samples after reconstitu-tion and compared to the original values recorded before lyophiliza-tion. Data was statistically analyzed applying one way ANOVA usingSPSS 19 software (IBM Corporation, Armonk, NY, USA). Post-hocmultiple comparisons were performed using Fisher’s least signifi-cant difference test and the results were considered significantlydifferent when p-values were less than 0.05.

2.6. In vitro vinpocetine release from the optimized formula

In vitro vinpocetine release from micellar dispersion, before andafter lyophilization was carried out in USP I dissolution apparatus(Pharm Test, Hainburg, Germany) using the membrane diffusiontechnique. Dialysis tubing membranes (Spectra Porã, molecularweight cut off 12–14 kDa, Sigma–Aldrich, St. Louis, USA) wereimmersed in the dissolution medium overnight prior to experi-ment. The lyophilized formula was reconstituted to its originalvolume using distilled water before conducting the dissolutiontesting. Samples from each formula (equivalent to 5 mg ofvinpocetine) were added to glass cylinders (6 cm length and2.5 cm internal diameter) tightly covered with the dialysismembrane tubing from one end. The loaded cylinders were fixedat the shafts of the USP dissolution tester apparatus (Abdelbary andTadros, 2013; Aburahma and Abdelbary, 2012). The used dissolu-tion medium was 500 mL phosphate buffer saline (PBS, pH 7.4)containing 0.5% sodium lauryl sulphate to achieve sink conditions(Nie et al., 2011). Rotation speed was set at 50 rpm and thetemperature was adjusted to 37 � 0.5 �C throughout the dissolu-tion study. Dissolution of vinpocetine market product (Vinporal1,Amriya Pharmaceuticals, Alexandria, Egypt) was carried out underthe same conditions for comparison. Samples (3 mL, each) werewithdrawn at time intervals of 0.25, 0.5, 1, 2, 3, 4, 5, 8, 24 and 48 hand spectrophotometrically analyzed at lmax 268 nm. The releasestudy was repeated three times and the average percentagesdissolved (�SD) were illustrated against time. Similarity factor (f2)was calculated for comparison of release profiles utilizing thefollowing equation:

f 2 ¼ 50 � log 1 þ 1n

� �Xnj¼1

jRj � Tjj224

35�0:5

� 100

8<:

9=; (2)

where n is the sampling number, T and R are the percentages ofdrug released from the test and reference micellar formulae,respectively, at each time interval j (Costa and Sousa Lobo, 2001).Moreover, half-lives of the drug release were calculated andstatistically compared to detect presence or absence of anysignificant difference between the calculated values.

2.7. Effect of sterilization by gamma radiation

Samples of the lyophilized formula were packed in dry iceinside a polyurethane container to avoid aggregation or meltingdue to the probable temperature elevation induced by F-radiation(F-irradiator, Gamma cell 1000; BEST theratronics, Ontario,Canada) (Yehia et al., 2012). Radiation dose was 2.5 Mrad as statedin the united states pharmacopoeia (Convention, 2006). Afterreconstitution of the sterilized formulae with sterile water forinjection, EE, PS, PDI and ZP were re-measured and statisticallycompared with the equivalent results before sterilization. More-over, testing of the in vitro drug release profile was re-conducted

and similarity factor (f2) was calculated to compare the drugrelease profiles before and after sterilization.

2.8. Morphological evaluation of the lyophilized matrices by scanningelectron microscopy (SEM)

The surface properties of the lyophilized optimized formula,before and after sterilization were evaluated using scanningelectron microscope (JXA-840; JEOL, Japan). Samples were goldcoated under vacuum using an ion sputter and then, examined.

2.9. In vivo quantification of vinpocetine concentrations in plasma

2.9.1. Study design and animalsSix healthy New Zealand male rabbits weighing 3–4 kg were

used throughout the study. The animals were randomly dividedinto two groups, each consisted of three: group I, administered theoptimized formula and group II, administered the marketedproduct (Vinporal1, Amriya Pharmaceuticals, Alexandria, Egypt),with volume equivalent to 3.2 mg vinpocetine. The equivalentrabbit dose was calculated according to the following equation,based on the body surface area (Reagan-Shaw et al., 2008):

Human equivalent dose ðmg=kgÞ¼ Animal dose ðmg=kgÞ � Animal Km

Human Km(3)

Briefly, rabbits were fasted overnight with free access to waterbefore experimentation. The protocol of the study was reviewedand approved by the institutional review board of GenuineResearch Center, Cairo, Egypt. Blood samples (approximately3 mL) were withdrawn from the ear vein into heparinized glasstubes at the following time intervals; 0.25, 0.5, 1, 2, 3, 4, 5, 8, 24 and48 h following the IV dosing. The withdrawn blood samples werecentrifuged at 4000 rpm for 15 min at 4 �C. Then, plasma wastransferred directly into 5 mL plastic tubes and stored at �70 �C tilldrug analysis. The study was repeated after washing period ofseven days to fulfill the crossover design.

2.9.2. Sample preparationA plasma sample of 500 ml was transferred to a 10 mL glass test

tube, then 50 mL of internal standard (IS; Torsemide) workingsolution (1000 ng/mL) was added. After vortex mixing for 10 s,4 mL aliquot of ether was added using Dispensette Organic (BrandGmbH, Postfach, Germany) (Lin et al., 2014). The sample wasvortex-mixed for 3 min using vortex. The organic layer wastransferred to clean glass tubes and evaporated to dryness usingcentrifugal vacuum concentrator Vacufuge 5301 (Eppendorf,Germany) at 40 �C. Dry residues were then dissolved in 200 mlof mobile phase and vortexed for 1 min for reconstitution, and20 ml was injected using the autosampler.

A sensitive, selective and accurate LC–MS/MS method wasdeveloped and validated before the study for determination ofvinpocetine. A shimadzu prominence (Shimadzu, Japan) series LCsystem equipped with degasser (DGU-20A3), solvent delivery unit(LC-20AB) with an auto-sampler (SIL-20 AC) was used to inject20 mL aliquots of the processed samples on a Luna C18 (phenom-enex, USA) (50 � 4.6) mm, 5 mm particle size. The guard columnwas phenomenex C18 (5 � 4.0) mm, 5 mm particle size. All analysiswas carried out at room temperature.

The isocratic mobile phase consisted of a mixture of methanol–2 mM ammonium acetate–formic acid (40:60:0.1, v/v/v); afterfiltration and degassing of the mobile phase, it was delivered at aflow rate of 0.3 mL/min into the mass spectrometer’s electrosprayionization chamber (Xia et al., 2010). MS/MS detection in negativeion mode using a MDS Sciex (Foster City, CA, USA) API-4000 mass

R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46 41

spectrometer was used for quantitation. The mass spectrometerwas equipped with a turbo ionspray interface at 450 �C. The ionspray voltage was set at 5500 V. The common parameters, viz.curtain gas, nebulizer gas, collision gas and auxillary gas were set at25 psi, 20 psi, 5 psi and 40 psi, respectively. The compoundparameters: collision energy, declustering potential, entrancepotential and collision exit potential were 100 V, 40 V, 8 V, 7 Vfor vinpocetine and 20 V, 30 V, 9 V, 12 V for Torsemide (IS),respectively. Multiple reaction monitoring (MRM) mode detectionof the ions was performed in the, monitoring the transition of them/z 351.17 precursor ion to the m/z 280.20 for vinpocetine and m/z348.99 precursor ion to the m/z 263.90 for IS. The analytical datawere processed by Analyst software version 1.6 (Applied Bio-systems Inc., Foster city, CA).

2.9.3. Pharmacokinetic and statistical analysisThe pharmacokinetic parameters of vinpocetine after IV

injection of the investigated formulae were analyzed by non-compartmental pharmacokinetic models using Kinetica1 softwareversion 5 (Thermo Fisher Scientific Inc., Waltham, MA). The peakplasma concentration (Cmax) was directly obtained from theconcentration–time data. In addition, the values of the area underthe plasma concentration–time curve from time zero to the lasttime point and to infinity (AUC0�48 and AUC0�1) were calculatedin accordance with the linear trapezoidal rule. Also, the terminalelimination rate constant (Ke), elimination half-life (t1/2) and meanresidence time (MRT) were calculated using the linear regressionof the terminal portion of the ln (concentration)–time curve.

3. Results and discussion

3.1. Characterization of the prepared micellar formulae

3.1.1. Determination of entrapment efficiency (EE)EE percentages for the prepared vinpocetine mixed micelles

ranged between 6.97% and 94.08% as shown in Table 1 and Fig. 1a.The calculated EE values were best fitted to polynomial analysiswith special cubic model. Adequate precision was calculated by theDesign-Expert software to demonstrate the signal to noise ratio toensure that the model could be applied to navigate the designspace, whereas a ratio greater than 4 is desirable (de Lima et al.,2011). On the other hand, predicted R2 was calculated as a measureof how good the model could predict a response value bycomparing the calculated value with the adjusted R2 (Annaduraiet al., 2008). Adequate precision was 17.85 with reasonabledifference between the predicted R2 (0.8619) and the adjusted R2

(0.9457). The calculated equation for the EE% analysis was:

EE% ¼ 92:55X1 þ 20:23X2 þ 14:27X3 � 140:02X1X2� 11:77X1X3 � 36:50X2X3 þ 864:35X1X2X3 (4)

Factorial statistical analysis revealed that changing surfactantspercentages had a significant effect on the EE% (p < 0.001). It wasobserved that increasing PL121 led to EE% increase. This might beattributed to the high lipophilicity of PL121 (HLB: 1–7) that providea favorable medium for the entrapment of a water insoluble druglike vinpocetine. Similar results were obtained by Xu et al. whoobserved the increase in folate loading after incorporation of PL121in the formulated mixed micelles (Xu et al., 2012).

3.1.2. Analysis of particle size, polydispersity index and zeta potentialThe mean PS of micellar formulae were in the nano range

between 10.70 and 315.39 nm as shown in Table 1. PS values weresubjected to polynomial analysis using quadratic model. Adequateprecision was 9.36 with reasonable difference between thepredicted R2 (0.7981) and the adjusted R2 (0.8243). The calculatedequation for the PS analysis was:

PS ¼ 218:83X1 þ 16:79X2 þ 30:56X3 þ 767:94X1X2þ 8:79X1X3 þ 32:74X2X3 (5)

Fig. 1b illustrates the effect of independent variables on themean PS. From the presented figure, it is clear that theincorporation of different pluronics percentages showed signifi-cant effects on the mean PS (p-value = 0.011). It could be noticedthat the formulae M1 (containing 100% w/w PF127), M2(containing PF127 and PP123, in ratio 1:1 w/w) and M3 (containing100% w/w PP123) were having the lowest particle size values. Thismight be attributed to the polar and semi-polar nature ofPF127 and PP123, respectively, which could enable them to formrelatively small micelles in water. Similar results were obtained inprevious studies utilizing PP123 alone or in a mixture withPF127 for the micellar solubilization of paclitaxel, whereas theparticle size was nearly 20 nm (Han et al., 2006; Wei et al., 2009).On the other hand, incorporation of PL121 in the remainingformulae (M4–M10) led to significant increase in the particle size.These results were in accordance with the previously noticedfindings by Abdelbary et al. who prepared mixed micellescontaining different concentrations of PL121 and PP123 for braintargeting of olanzapine (Abdelbary and Tadros, 2013).

Additionally, PDI values were also investigated and the obtainedresults were displayed in Table 1 and statistically illustrated inFig. 1c. It was found that all the prepared micellar formulae hadvalues ranged between 0.21 and 0.69 which could be considered asan acceptable mid range (Cho et al., 2014). Polynomial analysis

Table 1Experimental runs, independent variables, and measured responses of the simple lattice mixture design for vinpocetine micellar formulae.

Formulae PL121 (% w/w) PP123 (% w/w) PF127 (% w/w) EE (% w/w) PS (nm) PDI

M1 0.00 0.00 100.00 10.55 � 2.57 19.00 � 1.59 0.69 � 0.0513.91 � 2.01 22.34 � 1.92 0.54 � 0.03

M2 0.00 50.00 50.00 6.97 � 5.83 15.79 � 2.05 0.44 � 0.02M3 0.00 100.00 0.00 20.95 � 1.78 10.70 � 0.94 0.64 � 0.07

22.43 � 3.14 13.49 � 1.74 0.59 � 0.01M4 16.67 16.67 66.67 46.16 � 2.20 163.20 � 9.11 0.32 � 0.04M5 16.67 66.67 16.67 17.47 � 2.51 188.40 � 12.30 0.54 � 0.03M6 33.33 33.33 33.33 55.12 � 4.07 84.55 � 6.35 0.21 � 0.01M7 50.00 0.00 50.00 48.32 � 4.11 118.00 � 7.04 0.25 � 0.03M8 50.00 50.00 0.00 22.59 � 1.65 305.70 � 16.55 0.47 � 0.06

25.02 � 3.24 315.39 � 10.79 0.43 � 0.02M9 66.67 16.67 16.67 58.67 � 4.33 265.90 � 8.33 0.55 � 0.03M10 100.00 0.00 0.00 92.94 � 6.84 215.80 � 11.47 0.46 � 0.04

94.08 � 3.28 219.66 � 14.50 0.41 � 0.05

42 R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46

with quadratic model was utilized for the statistical evaluation ofthe PDI values. Adequate precision was 5.13 with reasonabledifference between the predicted R2 (0.2544) and the adjusted R2

(0.4286). The calculated equation for the PS analysis was:

PDI ¼ 0:46X1 þ 0:62X2 þ 0:60X3 � 0:24X1X2 � 1:05X1X3� 0:75X2X3 (6)

It could be concluded from the statistical analysis that changingpluronics percentages had no significant effect on the PDI valueswith p-value of 0.084.

Finally, the zeta potential of the prepared formulae wasinvestigated and negative charges were observed on their surfaces

with values ranging between �18.3 and �35.40 mV. As stated inliterature, the zeta potential is an indicator for the dispersionstability. High zeta potential could provide high repulsion forcesprotecting the dispersed nanoparticles from aggregation (Wanget al., 2013). From the obtained findings, it is evident that theobtained zeta potential range would present a reasonable degree ofstability for the prepared micellar formulae.

3.1.3. Morphological evaluation of the dispersed micelles bytransmission electron microscopy (TEM)

The morphological characteristics of the optimized micellarformula were investigated using TEM, as presented in Fig. 2. TheTEM micrographs showed vesicles with uniform spherical shape.

Fig. 2. Transmission electron micrographs of the optimized formula, at different magnifications.

Fig. 1. Response surface plot for the effects of incorporation of different pluronics percentages on the mean entrapment efficient (a), particle size (b), polydispersity index (c)and desirability (d) of vinpocetine micellar formulae.

R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46 43

These vesicles appeared to be well dispersed, non-aggregated withsmooth surface. The mean particle size demonstrated by the TEMmicrograph was in good agreement with that measured by theZetasizer. As previously mentioned, the observed PS was less than100 nm which could achieve long circulation through avoidance ofcapture and digestion by the RES (Cabral et al., 2011; Wei et al.,2009).

3.2. Selection of the optimized micellar formula

Optimization of all independent variables at the same time isnearly unachievable as the optimum condition obtained with oneresponse might have antagonizing effect on another one (Singhet al., 2012). The desirability function approach is widely used inresearches for the optimization of multiple response processes. Itwas calculated to choose the optimum composition with themaximal EE, minimal PS and PDI. The highest desirability value was0.621, as shown in Fig. 1d. This desirability value was for theoptimized formula containing 68% w/w PL121 and 32% w/w PF127.Such findings are in harmony with that obtained by Oh et al. whofound that PL121/PF127 mixtures (in ratio, 1:1 w/w) formed stabledispersions with small particle size. Furthermore, Oh et al. foundthat mixed PL121/PF127 micelles showed nearly 10-fold highersolubilization capacity compared to PF127 micelles (Oh et al.,2004). After preparation and investigation of the optimizedformula, EE, PS, PDI and ZP were 50.74 � 3.26%, 161.50 � 7.39 nm,

0.21 �0.03 and �22.42 �1.72 mV, respectively. Consequently, thisformula was selected for further investigations.

3.3. Characterization of the lyophilized optimized micellar formulabefore and after sterilization

Lyophilized samples of the optimized formula in presence andabsence of mannitol showed no significant differences in themeasured parameters (EE, PS, PDI and ZP), either to each other or tothe original formula before lyophilization (p � 0.05). So, it could beconcluded that the micellar dispersion do not need cryoprotectant.This might be due to the presence of PF127 which had longhydrophilic polyethylene oxide chain that might prevent stickingof the adjacent tubular aggregates formed by the lipophilicpluronic (Oh et al., 2004; Wei et al., 2009). These findings wereconsistent with the previous observations of Szleifer et al. whodemonstrated that lipid vesicles could be sterically stabilized byincorporation of PF127 (Szleifer et al., 1998). The mean EE of thereconstituted lyophilized formula, without mannitol, was found tobe 49.37 �4.10%. PS, PDI and ZP were 163.03 � 6.22 nm, 0.19 � 0.01and �23.55 � 2.16 mV. After sterilization, all the re-measuredparameters had no significant differences to the original formulabefore sterilization with p-value � 0.05.

3.4. In vitro vinpocetine release from the optimized formula

The release profiles of vinpocetine from the optimized micellarformula before, after lyophilization and after sterilization incomparison to the market product in PBS (pH 7.4) containingsodium lauryl sulphate are illustrated in Fig. 3. The optimizedformula showed significant sustainment of the drug release whencompared to the market product with f2 of 25. Moreover, the drugrelease half-life significantly increased from 2.38 h, in case of themarket product to 6.53 h, in case of the optimized formula(p < 0.001). Dissolution profile could be considered similar if f2value lies between 51 and 100 (Costa and Sousa Lobo, 2001). Fromthese data, it could be deduced that the prepared micellardispersion achieved the targeted sustained release profile forvinpocetine. On the other hand, drug release from the optimizedformula after lyophilization and sterilization were similar to theoriginal formula with f2 values of 55 and 52, with preserved releasehalf-life values of 6.41 and 6.76 h, respectively. These results couldindicate the capability of the used techniques during lyophilizationand sterilization in the preservation of the optimized formulacharacteristics.

Fig. 4. Scanning electron micrographs of the lyophilized optimized formula, before (a) and after (b) sterilization.

Fig. 3. Release profiles of vinpocetine from the optimized formulae before, afterlyophilization and after sterilization, in comparison with the market product in PBSat 37 �C.

44 R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46

3.5. Morphological evaluation of the lyophilized matrices by scanningelectron microscopy (SEM)

The scanning electron micrographs of the lyophilized optimizedformula are shown in Fig. 4, before (a) and after (b) sterilization.Both micrographs demonstrate the porous nature of the investi-gated matrices without any noticeable difference. These resultscould support the observed rapid reconstitution due to the rapidwater penetration through the lyophilized porous matrices.

3.6. In vivo quantification of vinpocetine concentrations in plasma

The LC–MS/MS assay has a good linearity from 0.5 ng/mL to500 ng/mL with acceptable within and between day reproducibili-ty. The lower limit of vinpocetine quantification in plasma was0.5 ng/mL. The intra-day accuracy of the method ranged from95.6% to 103.9% (data not shown), while the intra-day precisioncalculated as CV% ranged from 4.1% to 9.5%. The inter-day accuracyranged from 96.2% to 106.5%, while the inter-day precision rangedfrom 4.7% to 8.2%. The accuracy of freeze and thaw stability rangedfrom 89.0% to 93.4%, while its precision ranged from 4.4% to 7.9%.

The vinpocetine mean plasma concentration–time profilesfollowing single IV dose administration of the optimized formulaand the market product are shown in Fig. 5. Correspondingpharmacokinetic parameters are summarized in Table 2. It couldbe observed that the optimized formula had significantly lowermaximum plasma concentration (Cmax) and elimination rateconstant (Ke), when compared to the market product (p < 0.001).

On the other hand, significantly higher elimination half-life (t1/2)and mean residence time (MRT) have been reported in case of theoptimized formula (p < 0.001). These results could indicate theability of the optimized nanomicellar dispersion to achieveintravenous sustained profile when compared to the marketproduct containing the drug in solubilized form. These results werein accordance with that previously reported by Wei et al. whostudied the in vivo behavior of pluronics mixed micelles after IVinjection in rats (Wei et al., 2009).

4. Conclusion

Mixed micelles achieved significant in vitro and in vivo sustain-ment behavior and could be considered as a promising nanocarrierfor the intravenous delivery of the hydrophobic drug; vinpocetine,having rapid elimination rate with low elimination half-life.Additionally, it could be deduced that the presence of threesurfactants with different HLB in the same mixed micellar formuladid not offer any extra advantage over those containing twosurfactants only. This might be due to that the controlling factor isthe characteristics of the hydrophilic shell and the lipophilic core,nevertheless the number of surfactants incorporated.

References

Abdelbary, G.A., Tadros, M.I., 2013. Brain targeting of olanzapine via intranasaldelivery of core–shell difunctional block copolymer mixed nanomicellarcarriers: in vitro characterization, ex vivo estimation of nasal toxicity and invivo biodistribution studies. Int. J. Pharm. 452, 300–310.

Aburahma, M.H., Abdelbary, G.A., 2012. Novel diphenyl dimethyl bicarboxylateprovesicular powders with enhanced hepatocurative activity: preparation,optimization, in vitro/in vivo evaluation. Int. J. Pharm. 422, 139–150.

Alexandridis, P., Lindman, B., 2000. Amphiphilic Block Copolymers: Self-Assemblyand Applications. Elsevier Science, Amsterdam.

Annadurai, G., Ling, L.Y., Lee, J.F., 2008. Statistical optimization of mediumcomponents and growth conditions by response surface methodology toenhance phenol degradation by Pseudomonas putida. J. Hazard. Mater. 151,171–178.

Batrakova, E.V., Li, S., Li, Y., Alakhov, V.Y., Elmquist, W.F., Kabanov, A.V., 2004.Distribution kinetics of a micelle-forming block copolymer Pluronic P85. J.Control. Release: Off. J. Control. Release Soc. 100, 389–397.

Cabral, H., Matsumoto, Y., Mizuno, K., Chen, Q., Murakami, M., Kimura, M., Terada, Y.,Kano, M.R., Miyazono, K., Uesaka, M., Nishiyama, N., Kataoka, K., 2011.Accumulation of sub-100 nm polymeric micelles in poorly permeable tumoursdepends on size. Nat. Nanotechnol. 6, 815–823.

Cho, H.J., Park, J.W., Yoon, I.S., Kim, D.D., 2014. Surface-modified solid lipidnanoparticles for oral delivery of docetaxel: enhanced intestinal absorption andlymphatic uptake. Int. J. Nanomed. 9, 495–504.

Convention, U.S.P., 2006. USP 30. United States Pharmacopeial Convention.Costa, P., Sousa Lobo, J.M., 2001. Modeling and comparison of dissolution profiles.

Eur. J. Pharm. Sci.: Off. J. Eur. Fed. Pharm. Sci. 13, 123–133.de Lima, L.S., Araujo, M.D.M., Quináia, S.P., Migliorine, D.W., Garcia, J.R., 2011.

Adsorption modeling of Cr, Cd and Cu on activated carbon of different origins byusing fractional factorial design. Chem. Eng. J. 166, 881–889.

El-Laithy, H.M., Shoukry, O., Mahran, L.G., 2011. Novel sugar esters proniosomes fortransdermal delivery of vinpocetine: preclinical and clinical studies. Eur. J.Pharm. Biopharm.: Off. J. Arbeitsgemeinschaft fur Pharmazeutische Verfahren-stechnik e.V 77, 43–55.

Gao, Y., Li, L.B., Zhai, G., 2008. Preparation and characterization of Pluronic/TPGSmixed micelles for solubilization of camptothecin. Colloids Surf. B Biointerfaces64, 194–199.

Gao, Z.G., Fain, H.D., Rapoport, N., 2005. Controlled and targeted tumorchemotherapy by micellar-encapsulated drug and ultrasound. J. Control.Release: Off. J. Control. Release Soc. 102, 203–222.

Gulyás, B., Csiba, L., Kerényi, L., Galuska, L., Trón, L., 1998. The effect of a single-doseintravenous vinpocetine on chronic stroke patients. A PET study. In: Gulyás, B.,Müller-Gärtner, H. (Eds.), Positron Emission Tomography: A Critical Assessmentof Recent Trends. Springer, Netherlands, pp. 291–306.

Han, L.M., Guo, J., Zhang, L.J., Wang, Q.S., Fang, X.L., 2006. Pharmacokinetics andbiodistribution of polymeric micelles of paclitaxel with Pluronic P123. ActaPharmacol. Sin. 27, 747–753.

Hasa, D., Voinovich, D., Perissutti, B., Bonifacio, A., Grassi, M., Franceschinis, E.,Dall’Acqua, S., Speh, M., Plavec, J., Invernizzi, S., 2011. Multidisciplinary approachon characterizing a mechanochemically activated composite of vinpocetine andcrospovidone. J. Pharm. Sci. 100, 915–932.

Huang, S.J., Sun, S.L., Feng, T.H., Sung, K.H., Lui, W.L., Wang, L.F., 2009. Folate-mediated chondroitin sulfate-Pluronic 127 nanogels as a drug carrier. Eur. J.Pharm. Sci.: Off. J. Eur. Fed. Pharm. Sci. 38, 64–73.

Fig. 5. Mean plasma concentrations (ng/mL) of vinpocetine after IV injection of asingle bolus dose of the optimized formula compared to the market product, eachequivalent to 3.2 mg drug.

Table 2Vinpocetine pharmacokinetic parameters after IV injection of single bolus dose ofthe optimized formula and the market product, each equivalent to 3.2 mg drug.

Pharmacokinetics parameters Treatment (mean � SD)

Optimized formula Market product

Cmax (ng/ml)a 150.84 � 6.07 287.43 � 18.99AUC0�48 (ng h/ml)a 507.06 � 41.40 534.18 � 48.18AUC0�1 (ng h/ml)a 550.05 � 39.15 567.98 � 46.91Ke (l/h)a 0.05 � 0.01 0.07 � 0.00t1/2e (h)a 14.19 � 2.72 10.32 � 0.34MRT (h)a 9.94 � 1.96 5.48 � 0.50

a Data are the mean values (n = 6) � SD.

R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46 45

Jin, N., Zhao, Y.X., Deng, S.H., Sun, Q., 2011. Preparation and in vitro anticanceractivity of oxymatrine mixed micellar nanoparticles. Die Pharmazie 66,506–510.

Kulthe, S.S., Inamdar, N.N., Choudhari, Y.M., Shirolikar, S.M., Borde, L.C., Mourya, V.K., 2011. Mixed micelle formation with hydrophobic and hydrophilic Pluronicblock copolymers: implications for controlled and targeted drug delivery.Colloids Surf. B Biointerfaces 88, 691–696.

Lee, E.S., Oh, Y.T., Youn, Y.S., Nam, M., Park, B., Yun, J., Kim, J.H., Song, H.T., Oh, K.T.,2011. Binary mixing of micelles using Pluronics for a nano-sized drug deliverysystem. Colloids Surf. B Biointerfaces 82, 190–195.

Lin, C., Chen, F., Ye, T., Zhang, L., Zhang, W., Liu, D., Xiong, W., Yang, X., Pan, W., 2014. Anovel oral delivery system consisting in drug-in cyclodextrin-in nanostructuredlipid carriers for poorly water-soluble drug: vinpocetine. Int. J. Pharm. 465,90–96.

Lu, Y., Park, K., 2013. Polymeric micelles and alternative nanonized delivery vehiclesfor poorly soluble drugs. Int. J. Pharm. 453, 198–214.

Mu, C.F., Balakrishnan, P., Cui, F.D., Yin, Y.M., Lee, Y.B., Choi, H.G., Yong, C.S., Chung, S.J., Shim, C.K., Kim, D.D., 2010. The effects of mixed MPEG-PLA/Pluroniccopolymer micelles on the bioavailability and multidrug resistance of docetaxel.Biomaterials 31, 2371–2379.

Nie, S., Wu, J., Liu, H., Pan, W., Liu, Y., 2011. Influence of admixed citric acid andphysiological variables on the vinpocetine release from sodium alginatecompressed matrix tablets. Drug Dev. Ind. Pharm. 37, 954–962.

Ning, M., Zhou, Y., Chen, G., Mei, X., 2011. Preparation and in vitro/in vivo evaluationof vinpocetine elementary osmotic pump system. Adv. Pharmacol. Sci. 2011,1–7.

Oh, K.T., Bronich, T.K., Kabanov, A.V., 2004. Micellar formulations for drug deliverybased on mixtures of hydrophobic and hydrophilic Pluronic block copolymers. J.Control. Release: Off. J. Control. Release Soc. 94, 411–422.

Parmar, A., Singh, K., Bahadur, A., Marangoni, G., Bahadur, P., 2011. Interaction andsolubilization of some phenolic antioxidants in Pluronic(R) micelles. ColloidsSurf. Biointerfaces 86, 319–326.

Reagan-Shaw, S., Nihal, M., Ahmad, N., 2008. Dose translation from animal to humanstudies revisited. FASEB J.: Off. Publ. Fed. Am. Soc. Exp. Biol. 22, 659–661.

Singh, G., Pai, R.S., Devi, V.K., 2012. Optimization of pellets containing soliddispersion prepared by extrusion/spheronization using central compositedesign and desirability function. J. Young Pharm.: JYP 4, 146–156.

Szakall, S., Boros, I., Balkay, L., Emri, M., Fekete, I., Kerenyi, L., Lehel, S., Marian, T.,Molnar, T., Varga, J., Galuska, L., Tron, L., Bereczki, D., Csiba, L., Gulyas, B., 1998.Cerebral effects of a single dose of intravenous vinpocetine in chronic strokepatients: a PET study. J. Neuroimaging: Off. J. Am. Soc. Neuroimaging 8,197–204.

Szleifer, I., Gerasimov, O.V., Thompson, D.H., 1998. Spontaneous liposome formationinduced by grafted poly(ethylene oxide) layers: theoretical prediction andexperimental verification. Proc. Natl. Acad. Sci. U. S. A. 95, 1032–1037.

Wang, N., Hsu, C., Zhu, L., Tseng, S., Hsu, J.P., 2013. Influence of metal oxidenanoparticles concentration on their zeta potential. J. Colloid Interface Sci. 407,22–28.

Wei, Z., Hao, J., Yuan, S., Li, Y., Juan, W., Sha, X., Fang, X., 2009. Paclitaxel-loadedPluronic P123/F127 mixed polymeric micelles: formulation, optimization and invitro characterization. Int. J. Pharm. 376, 176–185.

Xia, H.M., Su, L.N., Guo, J.W., Liu, G.M., Pang, Z.Q., Jiang, X.G., Chen, J., 2010.Determination of vinpocetine and its primary metabolite apovincaminic acid inrat plasma by liquid chromatography–tandem mass spectrometry. J. Chroma-togr. B Anal. Technol. Biomed. Life Sci. 878, 1959–1966.

Xu, W., Cui, Y., Ling, P., Li, L.B., 2012. Preparation and evaluation of folate-modifiedcationic pluronic micelles for poorly soluble anticancer drug. Drug Deliv. 19,208–219.

Yehia, S.A., Elshafeey, A.H., Elsayed, I., 2012. Biodegradable donepezil lipospheres fordepot injection: optimization and in-vivo evaluation. J. Pharm. Pharmacol. 64,1425–1437.

Zhang, W., Shi, Y., Chen, Y., Ye, J., Sha, X., Fang, X., 2011. Multifunctional PluronicP123/F127 mixed polymeric micelles loaded with paclitaxel for the treatment ofmultidrug resistant tumors. Biomaterials 32, 2894–2906.

Zhao, L., Du, J., Duan, Y., Zang, Y., Zhang, H., Yang, C., Cao, F., Zhai, G., 2012. Curcuminloaded mixed micelles composed of Pluronic P123 and F68: preparationoptimization and in vitro characterization. Colloids Surf. B Biointerfaces 97,101–108.

Zhuang, C.Y., Li, N., Wang, M., Zhang, X.N., Pan, W.S., Peng, J.J., Pan, Y.S., Tang, X., 2010.Preparation and characterization of vinpocetine loaded nanostructured lipidcarriers (NLC) for improved oral bioavailability. Int. J. Pharm. 394, 179–185.

46 R.M. El-Dahmy et al. / International Journal of Pharmaceutics 477 (2014) 39–46