Novel surface-engineered solid lipid nanoparticles of

0 downloads 0 Views 5MB Size Report
Jan 17, 2017 - tional review board approval or have followed the principles outlined in .... Lactoferrin bioconjugated solid lipid nanoparticles: a new drug delivery ... solid lipid nanoparticles (SLNs) for topical treatment of acne. Colloids Surf.

Research Article For reprint orders, please contact: [email protected]

Novel surface-engineered solid lipid nanoparticles of rosuvastatin calcium for low-density lipoprotein-receptor targeting: a Quality by Design-driven perspective Aim: The present studies describe Quality by Design-oriented development and characterization of surface-engineered solid lipid nanoparticles (SLNs) of rosuvastatin calcium for low density lipoprotein-receptor targeting. Materials & methods: SLNs were systematically prepared employing Compritol 888 and Tween-80. Surface modification of SLNs was accomplished with Phospholipon 90G and DSPE-mPEG-2000 as the ligands for specific targeting to the low density lipoprotein-receptors. SLNs were evaluated for size, potential, entrapment, drug release performance and gastric stability. Also, the formulations were evaluated for cellular cytotoxicity, uptake and permeability, pharmacokinetic, pharmacodynamic and biochemical studies. Results & conclusion: Overall, the studies ratified enhanced biopharmaceutical performance of the surface-engineered SLNs of rosuvastatin as a novel approach for the management of hyperlipidemia-like conditions. First draft submitted: 18 September 2016; Accepted for publication: 7 December 2016; Published online: 17 January 2017 Keywords:  cellular uptake • hyperlipidemia • nanomedicine • phospholipid • Quality by Design • targeting

Rosuvastatin calcium (referred as rosuvastatin) is a potent lipid lowering agent which acts on HMG-CoA-reductase enzyme for inhibiting the synthesis of cholesterol. It is commonly prescribed for the treatment and management of diverse conditions like dyslipidemia, familial hyperlipidemia, hypertriglyceridemia and atherosclerosis [1,2] . It is administered through oral route in clinically recommended doses, ranging between 10 and 40 mg, and is well tolerated in all age groups. However, rosuvastatin exhibits poor aqueous solubility, high hepatic first-pass metabolism and oral bioavailability of less than 20% [3,4] . This reduces the efficacy of rosuvastatin in lowering the elevated serum lipid levels in hyperlipidemia and atherosclerosis. Diverse formulation strategies like solid dispersions  [5] , inclusion complexes [6] , liquisolid compacts  [7] , nanocrystals [8] and self-nanoemulsifying systems [9,10] have been employed

10.2217/nnm-2016-0336 © 2017 Future Medicine Ltd

for oral delivery of rosuvastatin and for augmenting its biopharmaceutical performance. However, none of these reported strategies has been found to be highly satisfactory for addressing the aforesaid biopharmaceutical issues. Besides, a majority of such literature reports emphasizes only on improving the dissolution performance of rosuvastatin only. This calls for the development of novel nanostructured drug delivery systems not only for surmounting the biopharmaceutical challenges of rosuvastatin holistically, but also for augmenting the selective absorption of the drug thorough oral route for lowering the lipid levels for improving the treatment efficacy. Solid lipid nanoparticles (SLNs) are the well-known nanostructured colloidal carriers with particle size ranging between 1 and 1000 nm [11] . These are considered to be one of the effective alternatives over the poly-

Nanomedicine (Lond.) (Epub ahead of print)

Sarwar Beg1, Sanyog Jain2, Varun Kushwah2, Gurjit Kaur Bhatti3, Premjeet Singh Sandhu3, OP Katare1 & Bhupinder Singh*,1,3 1 University Institute of Pharmaceutical Sciences, UGC Centre of Advanced Studies, Panjab University, Chandigarh 160 014, India 2 Centre for Pharmaceutical Nanotechnology, Department of Pharmaceutics, National Institute of Pharmaceutical Education & Research, Mohali 160 062, Punjab, India 3 UGC-Centre of Excellence in Applications of Nanomaterials, Nanoparticles & Nanocomposites (Biomedical Sciences), Panjab University, Chandigarh 160 014, India *Author for correspondence: Tel.: +91 172 2534103 Fax: +91 172 2543101 [email protected] [email protected] gmail.com [email protected] pu.ac.in

part of

ISSN 1743-5889

Research Article  Beg, Jain, Kushwah et al. meric nanoparticles owing to several merits like low cost of excipients, ease of preparation, high drug loading potential for both lipophilic/hydrophilic drugs, controlled drug release profile, improved stability and enhanced biopharmaceutical performance [12] . Accordingly, several literature reports have been published on the application of SLNs for augmenting the oral bioavailability of drugs, plausibly through paracellular and transcellular pathways [13–18] . The nanosized colloidal structure of the SLNs, in this regard, potentiates oral drug absorption through these pathways. Since solid lipids particularly contain long-chain triglycerides, the SLNs also tend to undergo drug absorption via intestinal lymphatic pathways [19,20] . Moreover, the application of SLNs for site-specific targeting of the drugs to various receptors has also been demonstrated in various literature reports [21–29] . Of late, the approach of drug targeting to low-density lipoprotein (LDL) receptors has been considered to be highly effective in improving the efficacy of treatment of hyperlipidemia and atherosclerosis-like conditions [30] . LDL receptors are abundantly present on the intestinal epithelial cells, liver cells, macrophages and blood–brain barrier, subsequently regulate the absorption and metabolism of cholesterol. In this regard, drug targeting through LDL receptors is helpful in the management of different types of malignancies [31] . Moreover, it also plays a key role in the uptake and metabolism of sterols [32,33] . Thus, utilization of LDL receptors for the management of hyperlipidemia is quite novel, and hitherto unexplored. Since LDL receptors are cellsurface receptor that recognizes the apoprotein B100 and apoprotein E present on the outer phospholipid layer of LDL, very low-density lipoprotein and chylomicron remnants, utilization of these receptors in drug targeting for management of hyperlipidemia-like conditions is of great importance [31,34,35] . The present work, in this context, endeavors to explore the novel approach of delivering rosuvastatin using SLNs surface-engineered with phospholipids for drug targeting to the LDL receptors. Since none of the literature reports has investigated the utility of such a rational approach till date, the present work demonstrates the clinical implications for the management of hyperlipidemia and other related disorders. The systematic development of SLNs was carried out as per the Quality by Design-approach, which was extensively characterized through in vitro, in situ and in vivo studies. Materials & methods Materials

Rosuvastatin was obtained ex-gratis from M/s Mylan Laboratories Limited, Hyderabad, India. Compritol

10.2217/nnm-2016-0336

888 was received as gift sample from M/s Gattefosse, Cedex, France, while Phospholipon 90G (i.e., PL90G) was gifted by M/s Phospholipon GmbH, Ludwigshafen, Germany. PEGylated phospholipid (DSPEmPEG-2000) was received as gift sample from M/s Avanti Polar Lipid Inc., Alabama, USA. Soya lecithin was purchased from M/s Himedia Laboratories, Mumbai, India, and Tween-80 was purchased from M/s Fisher Scientific, Chandigarh, India. Deionized triple-distilled water (M/s Milipore, Mumbai, India) was used for all the experiments throughout study. All other chemicals, solvents and reagents used during the studies were of analytical reagent grade, and were used as obtained. Risk assessment studies

The risk assessment and factor screening studies were performed for identifying the high risk critical material attributes and/or critical process parameters influencing the critical quality attributes (CQAs) [36] . To perform risk assessment, Ishikawa fish-bone diagram was constructed employing Minitab 17 software (M/s Minitab Inc., PA, USA) to establish a cause–effect relationship among the plausible formulation attributes/process parameters and CQAs of the SLNs. Risk estimation matrix (REM) was constructed for identifying the medium to high risk factors with critical impact on the formulation CQAs. Low, medium and high risk ranking was used for discriminating the factors assigned to each of the factors based on the prior knowledge and experience [37,38] . Factor screening studies

Taguchi design was employed for performing factor screening studies to identify highly influential product and process variables on SLNs. The medium to high risk factors shortlisted from REM were subjected to factor screening. As a low resolution design (i.e., Resolution III), Taguchi design requires factors to be studied at two different levels, viz. low (-1) and high (+1), thus producing fewer experimental runs. Table 1 illustrates the experimental trials performed as per the Taguchi design, which were statistically analyzed using half-normal plots and Pareto charts for identifying the factor effects, while Table 2 shows the coded levels of the factors and their actual values. Selection of the lipid

Solubility of rosuvastatin was determined in various lipids, viz. Compritol 888, stearic acid, cetyl palmitate, glyceryl monostearate and soy lecithin. An excess amount of the drug was added to the vials containing 500 mg of the lipid and subjected to mechanical shaking for 24 h in a thermostatically controlled water

Nanomedicine (Lond.) (Epub ahead of print)

future science group

Novel surface-engineered solid lipid nanoparticles of rosuvastatin calcium 

Research Article

Table 1. Design matrix as per the 7-factor and 8-run Taguchi design, along with factor levels in coded and actual form for the solid lipid nanoparticle formulations. Trials

A: CP888

B: T80

C: MP

D: UT

E: UF

F: ST

G: SS

1

-1

-1

-1

-1

-1

-1

-1

2

1

-1

1

-1

1

-1

1

3

-1

1

1

1

1

-1

-1

4

1

-1

1

1

-1

1

-1

5

1

1

-1

-1

1

1

-1

6

-1

1

1

-1

-1

1

1

7

1

1

-1

1

-1

-1

1

8

-1

-1

-1

1

1

1

1

bath shaker maintained at 80 ± 2°C [37,39] . The clear supernatant fraction from each of the lipids was separated and dissolved in chloroform for measuring the drug content using previously developed and validated HPLC method of rosuvastatin [40] . Preparation of the SLNs

The SLNs were prepared by hot-microemulsification and solvent diffusion method using the lipids and their working ranges identified from the preliminary solubility study of the drug in it [29,41] . In brief, the solid lipid (i.e., 200 to 600 mg) was heated in up to 80°C and fixed quantity of drug (i.e., 20 mg) was added with gentle mixing for complete solubilization in lipid. Separately, a 5% w/v aqueous solution of Tween-80 was prepared in 10 ml of distilled water and heated at 80°C. The aqueous phase was then added to the organic phase under constant stirring to obtain a uniform dispersion, followed by ultrasonication in an ice bath for 3 min at a frequency of 9 mW. The solvent diffusion was carried out by addition of an excess amount of water (10 ml), followed by continuous stirring at 5000 r.p.m. for 30 min under ice bath conditions to obtain the SLNs. Systematic optimization of the SLNs using experimental design

Based on the risk assessment and factor screening studies, I-optimal design was employed for systematic optimization of the SLNs for the influential parameters viz. concentrations of lipid (i.e., Compritol 888), surfactant (i.e., Tween-80) and stirring speed. Table 3 summarizes an account of the 20 experimental runs for the SLNs with the factor combinations at three different level viz. low (-1), medium (0) and high (+1). The prepared formulations were evaluated for various CQAs like particle size, ζ potential, entrapment efficiency and time required for 80% drug release (T80%). The optimization data analysis was carried out by apt mathematical modeling of the experimental data with

future science group

quadratic polynomial model along with added interaction terms. The model fitness was evaluated by analyzing the statistical parameters like p-value, coefficient of correlation (r2) and predicted residual sum of squares. Response surface analysis was carried out for understanding the relationship among the studied factors on the formulation attributes. The optimized formulation was selected by numerical optimization and desirability function, followed by demarcation in the design space. Validation of the optimization methodology was conducted by preparing check-point formulations and comparing the predicted values of the CQAs with the observed ones using linear correlation plots and residual plots. Percent prediction error (or percent bias) was also calculated for ratifying the prognostic ability of the optimization methodology. Preparation of the PL-SLNs & PL-PEG-SLNs

For preparing PL-SLNs and PL-PEG-SLNs, an additional amount (i.e., 50 mg) of PL90G and DSPEmPEG-2000 was dissolved in ethanol and added to the aqueous phase for complete solubilization under magnetic stirring at 80°C. Further, the aqueous phase was added to the molten lipid phase containing drug solubilized in it. Characterization of the SLNs Particle size & ζ potential

The particle size distribution and ζ potential measurement of the SLNs, prepared as per the experimental design, was carried out by dynamic light scattering technique using Zetasizer ZS 90 (M/s Malvern Instruments, Worcestershire, UK). Transmission electron microscopy

An aliquot of 1 ml of the SLNs dispersion was diluted 100-fold with distilled water placed on copper grids, stained with 1% phosphotungstic acid solution and visualized under electron microscope (JEM-2100F, M/s Jeol, Tokyo, Japan).

www.futuremedicine.com

10.2217/nnm-2016-0336

Research Article  Beg, Jain, Kushwah et al.

Table 2. Coded levels of the factors and their actual values employed during Taguchi design. Factors

Type

 

 

Low (-1)

High (+1)

Conc. of Compritol 888 (CP888, mg)

Numerical

300

500

Conc. of Tween-80 (T80, %)

Numerical

150

200

Method of preparation (MP)

Categorical

Hot microemulsion

Cold microemulsion

Ultrasonication time (UT, min)

Numerical

1

3

Ultrasonication frequency (UF, mW)

Numerical

9

15

Stirring time (ST, min)

Numerical

30

60

Stirring speed (SS, r.p.m.)

Numerical

25

37

Encapsulation efficiency

Encapsulation efficiency of the SLNs was determined as per the reported method [42] . Briefly, an aliquot (2 ml) of the SLN dispersion was centrifuged at 10,000 r.p.m. (5590×g), the supernatant was discarded and pellet was harvested. The pellet containing SLNs was subjected for digestion in 0.1% w/v solution of TritonX100, followed by ultrasonication at 9 mW frequency for 15 min for complete removal of drug from the particles. The drug was extracted in methanol: chloroform mixture (96:4,%v/v) and the amount of drug entrapped was quantified using HPLC method [40] . The encapsulation efficiency was calculated using Eq. (1) as follows:

where, W1 represents the amount of drug encapsulated in the SLNs, and W2 represents the total amount of drug added to the formulation. In vitro gastrointestinal stability

The in vitro gastrointestinal stability of SLNs was carried out as per the method reported in literature [43] . The prepared formulation (aliquot 2 ml SLN dispersion) was added to 250 ml of simulated gastric fluid for 2 h and simulated intestinal fluid additional 6 h time period. At specified time interval of 2 h, aliquot (1 ml) of the samples were withdrawn for analyzing the p­article size, ζ potential and entrapment efficiency. In vitro drug release

The in vitro drug release studies for SLNs were carried out in dialysis bag method using 250 ml of simulated gastric fluid (pH 1.2) for 2 h and simulated intestinal fluid (pH 6.8) for 12 h at 100 r.p.m. and 37 ± 0.5°C. The dialysis membrane with molecular weight cut-off of 12 KDa (M/s Himedia Limited, Mumbai, India) was employed during the study. An accurately weighed amount of the SLNs (equivalent to 20 mg of the drug) was dispersed in 1 ml of the dissolution medium and

10.2217/nnm-2016-0336

Levels

placed in the dialysis bag. An aliquot (5 ml) of sample was withdrawn at periodic time intervals, followed by replenishment with an equal volume of fresh dissolution medium to maintain the sink conditions. The raw data obtained from dissolution studies were analyzed using an in-house software, ZOREL, with in-built provision for applying the correction factors for volume and drug losses during sampling [44,45] . Cell culture experiments

Caco-2 cells (ATCC, VA, USA) were grown in the tissue culture flasks (75 cm2), and were maintained at 37°C with 5% CO2 to simulate the physiological conditions (HeracellTM 150i, Thermo, USA). The growth medium comprised of Dulbecco’s Modified Eagle’s culture medium, 20% fetal bovine serum, 100 IU/ml of penicillin and 100 μg/ml of streptomycin. Proper care and maintenance of the cells were taken up to 21 days until their full growth. The growth medium was changed every alternate day with fresh medium for providing optimum nutrients to the cells. The cultured cells were trypsinized employing 0.25% trypsinEDTA solution (M/s Sigma Aldrich, MO, USA), after attaining 90% confluency. The harvested cells were then used for evaluating cellular cytotoxicity, passive cellular uptake and permeability studies. Moreover, the endocytotic uptake studies were also performed on Caco-2 cells in presence of uptake inhibitors for evaluating uptake potential of the SLNs through LDL receptors expressed on it. The presence of LDL receptors on Caco-2 cells and utility of them in exploring drug targeting through oral route has been well reported in literature [46,47] . Cellular cytotoxicity

The cytotoxicity studies were carried out using MTT assay on the previously grown Caco-2 cell monolayer, as per the procedure described in our previous reports  [37] . The cells were seeded in a 96-well plate at a density of 5 × 104 cells/well and incubated for the time period of 24 h with the pure drug suspension,

Nanomedicine (Lond.) (Epub ahead of print)

future science group

Novel surface-engineered solid lipid nanoparticles of rosuvastatin calcium 

SLNs and Triton X-100 (positive control), in the concentration of 0.1, 1, 10 and 20 μg/ml, respectively. Following incubation after 24 h, the medium containing the formulation was aspirated and cells were washed with phosphate buffer saline (PBS) pH 7.4. Subsequently, 150 μl of MTT solution (500 mg/ml in PBS) was added to each well plate and incubated at 37°C for another 3 h. The supernatant was removed and cells were aspirated with 200 μl of dimethyl sulfoxide to solubilize the formazan crystals. Optical density of the resulting solution was measured at 540 nm using an ELISA microplate reader. Based on the results of optical density, the percent cell viability was calculated using Eq. (2), as follows:

Research Article

a Millicell-ERS device (M/s Millipore Corporation, Darmstadt, Germany) and chopstick style electrodes. Cellular uptake & transport studies

All the uptake and transport studies were performed in Caco-2 cells, as per the procedure described in the literature  [48–50] . For the uptake studies, Caco-2 cells suspension containing 5 × 104 cells/well was seeded on each Transwell® permeable support (M/s Corning, MA, USA). The cell suspension was added with 0.1 ml of the growth medium at the apical side, while 0.5 ml of the medium was added on the basolateral side. Different uptake experiments were performed by adding the uptake and/or transport inhibitors in their specified concentration to the SLN formulation seeded on cells to evaluate the mechanistic pathways responsible for uptake and transportation through Caco-2 cells. Passive cellular uptake studies

In a separate set of experiments, safety of formulations was also assessed as a function of monolayer integrity by measuring the transepithelial electrical resistance (TEER) values up to 12 h postincubation with the test formulations. TEER was measured using

The passive cellular uptake was evaluated at varying concentrations of the prepared formulation seeded to the Caco-2 cells, followed by incubation for different time periods. For concentration-dependent uptake, cultured cells (5 × 104 cells/well) were incubated with

Table 3. Design matrix depicting the composition of solid lipid nanoparticle formulations prepared as per the I-optimal response surface design. Code

Build type

Space type

Factor X1 

Factor X 2 

Factor X 3 

1

Replicate

Plane

200

6

2400

2

Model

Vertex

200

3

1500

3

Model

Vertex

600

10

1500

4

Model

Plane

440

10

2400

5

Model

Edge

346

3

3000

6

Replicate

Plane

440

10

2400

7

Model

Edge

600

3

2025

8

Model

Vertex

200

10

3000

9

Lack of Fit

Interior

538

7

2228

10

Model

Plane

200

6

2400

11

Replicate

Interior

538

7

2228

12

Model

Plane

440

6

1500

13

Lack of Fit

Interior

366

7

2963

14

Replicate

Plane

440

6

1500

15

Lack of Fit

Interior

399

9

1650

16

Lack of Fit

Vertex

600

3

3000

17

Model

Edge

600

8

3000

18

Replicate

Edge

346

3

3000

19

Lack of Fit

Interior

240

7

1650

20

Model

Vertex

200

10

1500

X1: Concentration of CP 888 (mg); X2: Concentration of T80 (%); X3: Stirring speed (r.p.m.)

future science group

www.futuremedicine.com

10.2217/nnm-2016-0336

Research Article  Beg, Jain, Kushwah et al. varying concentrations (i.e., 5, 10, 15 and 20 μg/ml) of pure drug suspension and SLNs at 37 ± 1°C/5% CO2 for 2 h. For time-dependent uptake, cells treated with the same treatment formulations were incubated for specified time intervals (i.e., 1, 2, 3 and 4 h) in the culture medium and washed twice with PBS (pH 7.4). Further, the cells were lysed with 0.1% w/v Triton X-100 and washed with methanol to completely solubilize the drug internalized in the cells. The cell extract was centrifuged at 21,000 r.p.m. (18,498 × g) for 10 min and the supernatant was subjected to HPLC analysis for quantifying the amount of drug uptaken by the cells. Also, the visual observation of cellular uptake of the prepared formulations was carried out by preparing the SLNs loaded with Coumarin-6 dye (10 μg/ml), where the dye was added to the aqueous phase during preparation of SLNs. The confocal laser scanning microscopy (Olympus FV1000, M/s Olympus Corporation, Tokyo, Japan) was performed for evaluating fluorescence intensity as a function of the cellular uptake of the SLNs across the Caco-2 cells [37,43] . Endocytotic uptake studies

For endocytotic uptake evaluation, the cells were divided between control and treatment groups. The treatment groups were previously exposed to 5 μM concentration of filippin and 50 μM concentration of sucrose for 30 min to block the caveolae- and clathrinmediated endocytosis pathways [48,50] . After allowing incubation for the specified time period, the cells were washed with PBS and incubated with 5, 10 and 15 μg/ ml of prepared SLNs for 2 h. At the end of the experiment, cells were washed with ice-cold PBS and lysed by keeping overnight in 100 μl lysis buffer containing 0.1% w/v Tritin X-100 at 4°C, followed by HPLC analysis to evaluate the drug content as a measure of uptake potential of different formulations. Cellular permeability studies

The permeability experiment was carried out in Caco-2 cells as per the procedure described in our previous literature reports [37,51] . The drug loaded SLNs were subjected to permeability measurement from apical to basolateral (A→B) direction across the Caco-2 cell monolayer adhered on the surface of Transwell-96 permeable support (M/s Corning). The study was initiated by adding 250 μl of the test formulations to the transport medium in apical side and blank transport medium (2.6 ml) to the basolateral side. Aliquots of samples (200 μl each) were withdrawn from basolateral side during experiments at periodic time intervals, in other words, 0.5, 1, 2 and 3 h, followed by replacement with blank transport medium each time. The samples thus collected were analyzed using HPLC

10.2217/nnm-2016-0336

analysis. Cumulative amount of drug permeated was plotted as a function of time for calculating the apparent permeability (Papp), as per the Eq. (3) .

where, Papp is the apparent permeability coefficient expressed in 10 -6 cm/s, dq/dt (also called as flux) is the slope obtained from linear regression of cumulative amount of drug transported (in nM) as a function of time (seconds), C0 is initial concentration of the drug in donor compartment in μM/ml at the apical (for A→B) side and A is cross-sectional surface area of the filter (0.7 cm2 in 24 wells). Animal experiments

All animal experiments were carried out in accordance with the experimental protocols approved by the Institutional Animal Ethics Committee of the Panjab University, Chandigarh, India (Reference number: PU/ IAEC/S/14/110). Unisex Wistar rats (∼220–250 g) housed in polypropylene cages were kept under laboratory conditions at 25 ± 2°C and 55 ± 5% RH with free access to standard diet and water ad libitum. Proper care and maintenance of the animals were carried out as per the guidelines of Committee for Prevention, Control and Supervision of Experimental Animals, Govt. of India. In situ intestinal perfusion studies

The intestinal perfusion studies were carried out on PL-PEG-SLNs, PL-SLNs, Plain SLNs and pure drug suspension, as per the protocol described in our previous literature reports [37,52] . The animals were anesthetized using intraperitoneal injection of thiopental sodium in the dose of 50 mg/kg. The abdomen was opened with a midline incision and the entire small intestine was taken out carefully. The intestinal segment was perfused with Kreb-Ringer buffer maintained at 37 ± 1°C, until the perfusate was clear. The intestine was then perfused with the optimized SLN formulation and pure drug suspension at a rate between 0.2 and 0.3 ml/min. After the specified time intervals of 5, 15, 30 and 45 min, intestinal perfusates were collected and analyzed by HPLC for calculating intestinal permeability and absorption parameters like effective permeability (Peff ), wall permeability (Pwall), fraction absorbed (F) and absorption number (A n) of the drug from various prepared formulations [37,53] . In vivo pharmacokinetic studies

The animals were divided into four groups with six animals in each group and subjected to overnight fast-

Nanomedicine (Lond.) (Epub ahead of print)

future science group

Novel surface-engineered solid lipid nanoparticles of rosuvastatin calcium 

ing condition with free access to water. The pure drug suspension (dispersed in 0.25% w/v sodium carboxymethylcellulose), and 1 ml aliquot of the plain SLNs and modified SLNs (PL-SLNs and PL-PEG-SLNs), each containing rosuvastatin, were administered orally to the animals using a cannula. Blood samples (each around 0.2 ml) were periodically withdrawn from the retro-orbital plexus at specified time intervals in the heparinized micro-centrifuge tubes. Plasma was harvested by centrifugation at 10,000 r.p.m. (5590 × g) for 10 min and drug was extracted using acetonitrile as the extracting solvent. All the samples were filtered through 0.22 μm membrane filter and analyzed by HPLC [40] . Pharmacokinetic modeling and data analysis was performed using Win-Nonlin software version 5.0 (M/s Pharsight, CA, USA). One-compartment open body model (1-CBM) with zero lag-time using modified Wagner-Nelson method for peroral administration was chosen for computing various pharmacokinetic parameters.

Research Article

As per the protocol described in Section 2.9.2, the blood samples were collected from the hyperlipidemic animals in various groups at periodic time intervals postadministration of treatment formulations for estimation of biochemical parameters in erythrocytes and liver, followed by surface morphological characterization of erythrocytes. The detailed procedure for estimation of various biochemical parameters has been described in Supplementary data text Sections 1.1 and 1.2. Results Selection of the lipid

The equilibrium solubility data of rosuvastatin calcium observed in various lipids are depicted in Supplementary Figure 1. Equilibrium solubility of drug was found to follow the order of Compritol 888 > cetyl palmitate > stearic acid > glyceryl monostearate > Soy lecithin. As maximal solubility was observed in Compritol 888, it was selected as the lipidic constituent for further formulation development studies.

In vivo pharmacodynamic studies

Risk assessment studies

In vivo pharmacodynamic studies were carried out in hyperlipidemic Wistar rats treated with high fat diet and free access to water ad libitum up to 21 days [54] . After attaining the serum cholesterol level >250 mg/dl, the animals were divided into naive, control and treatment groups, with six animals in each group, respectively. The naive group contained animals, where neither disease was induced nor drug was administered. The control group, on the other hand, contained animals where only disease was induced but drug was not administered. The treatment groups included animals where disease was induced and animals were administered with various formulations viz. pure drug suspension, plain SLNs, PL-SLNs and PL-PEG-SLNs, each containing oral dose of rosuvastatin (i.e., 20 mg) equivalent to body weight of rats on 7, 14 and 21 days, respectively. At specified time periods, the animals were anesthetized using light ether anesthesia and blood samples (∼0.5 ml) were collected from the retroorbital plexus in heparin-coated glass vials. Serum was harvested by centrifugation at 10,000 r.p.m. (5579 × g) for 15 min and analyzed for the levels of total cholesterol (TC), LDL, triglyceride (TG) and high-density lipoprotein (HDL) in serum using ENZOPAK diagnostic kit (M/s Reckon Diagnostics, Gujarat, India).

Supplementary Figure 2 illustrates the fish-bone diagram drawn by highlighting various material attributes and process parameters having plausible influence on the CQAs of SLNs. The fish-bone diagram indicated plausible influence of the material attributes and process parameters on the CQAs of SLNs. Further, these factors were considered for constructing the REM. Supplementary Table 1 illustrates REM with factors ranked with low, medium and high risk levels on the basis of severity and occurrence of the risks on the CQAs. REM indicated that concentrations of lipid and surfactant were found to be associated with high risk, while concentration of edge activator, ultrasonication time, stirring speed, stirring time and method of preparation were associated with medium risk. The medium to high risk factors was further investigated through factor screening studies using Taguchi design for quantitative analysis of the factor effects on the CQAs.

Factor screening studies

Biochemical estimation studies

The factor screening was carried out with the help of Taguchi design, where obtained data subjected to mathematical modeling by selecting the linear polynomial model terms and two plausible two factor interaction(s), while ignoring the higher order interaction(s) terms owing to the confounding or aliasing of the combination of the factors.

Various oxidative stress markers like lipid peroxidase (LPO), superoxide dismutase (SOD), catalase (CAT), glutathione (GSH), acetylcholinesterase (AchE) and Na+ -K+ -ATPase were estimated in the erythrocytes and liver hepatocytes of the hyperlipidemic animals [55–57] .

where, Y = response variables, β 0 = intercept, β1 to β9 = coefficients of linear model terms. Among the various medium to high risk factors, the analy-

future science group

www.futuremedicine.com

10.2217/nnm-2016-0336

Research Article  Beg, Jain, Kushwah et al. sis of model through half-normal plots and Pareto charts revealed highly significant influence of factors B, C and G on the particle size, while factors B, C and D showed higher influenced on the entrapment efficiency, respectively. Overall, the factors B (i.e., Compritol 888), C (i.e., Tween-80), G (i.e., stirring speed) and D (i.e., method of preparation) on both the responses, as effects of these factors were found to be above t-value limit and Bonferroni limit (Supplementary Figure 3) . Preparation & optimization of the SLNs

The experimental data obtained as per the selected I-optimal design, fitted to the second-order quadratic polynomial model, is represented in Eq. (3) . Analysis of polynomial coefficients revealed the prevalence of significant interactions among the studied factors on the CQAs. Further evaluation of model fitting using parameters like correlation coefficient (R), ranging between 0.963 and 0.996, and insignificant lack of fit (p > 0.05), indicated aptness of the selected model and excellent goodness of fit of the experimental data (Supplementary Table 2) .

where, Y = response variables, β0 = intercept, β1 to β3 = coefficients of linear model terms; β4 to β6 = coefficients of linear interaction terms, β7 to β9 = coefficients of quadratic terms, β10 and β11 = coefficients of q­uadratic interaction terms. The 2D-response surface plots depicted in Supplementary Figures 4–7 reveals cause–effect relationship among the studied factors and corresponding response variables. Supplementary Figure 4(A–C) portrays the influence of studied factors on particle size. At lower levels of Tween-80, the particle size showed an inclining trend with increased concentration of Compritol 888 up to intermediate levels, followed by a distinct dip. On the contrary, increasing the levels of the Tween-80 and stirring speed revealed negative influence of both the factors on particle size. Smaller particle size was observed at lower levels of lipid, together with higher levels of surfactant and stirring speed. Supplementary Figure 5(A–C) depicts high influence of Compritol 888 on ζ potential with a curvilinear trend up to the intermediate levels of lipid, followed by a declining phase. Tween-80, however, shows a declining trend on the ζ potential at all the levels of lipid. Stirring speed was found to exert negligible influence on the values of ζ potential. Overall, high ζ potential values were observed at the intermediate levels of C­ompritol 888 and lower levels of Tween-80.

10.2217/nnm-2016-0336

Supplementary Figure 6(A–C) portrays a curvilinear trend for entrapment efficiency with increasing levels of Compritol 888 up to its intermediate levels, followed by a distinct declining trend. On the contrary, a linear increase in the values of entrapment efficiency was observed with increasing levels of Tween-80 and stirring speed. Maximal entrapment efficiency was observed at the intermediate levels of all the studied factors. Supplementary Figure 7(A–C) indicates the combined influence of all the studied factors on T80% . At low levels of Tween-80, the effect of Compritol 888 was found to be curvilinear and highly pronounced, from lower to intermediate levels on T80% , followed subsequently by a sharp declining phase. Likewise, the stirring speed also showed a curvilinear trend on the drug release profile and values of T80% . Higher value of T80% was observed at the intermediate levels of Compritol 888, and lower levels of Tween-80 and stirring speed.

Characterization of the prepared SLNs

illustrates the values of the CQAs (i.e., particle size, ζ potential, entrapment efficiency and T80%) for the SLN formulations, prepared as per the experimental design. Supplementary Table 3

Particle size

The SLNs exhibited particle size ranging between 32 and 141 nm, indicating nanostructured nature of the prepared formulations. Smaller particle size was observed for the formulations containing higher levels of lipid, lower levels of surfactant and higher stirring speed. ζ potential

The ζ potential values for SLNs were found to be ranging between -11.2 and -23.6 mV, indicating negative charge of the prepared formulations [58] . Magnitudinally high values of ζ potential were observed at high levels of lipid and surfactant, and vice versa. Entrapment efficiency

The prepared formulations showed high entrapment efficiency values ranging between 71 and 97%. High entrapment efficiency was observed at higher levels of lipid and surfactant, and intermediate levels of the s­tirring speed  [59] . In vitro drug release studies

The in vitro drug release profile of the formulations, prepared as per the experimental design, is depicted in Figure 1. Different drug release profiles were observed from various prepared formulations, with maximal drug release (i.e., >80%) observed within first 12 h time period. Faster drug release rate was noticeable from all those prepared formulations containing

Nanomedicine (Lond.) (Epub ahead of print)

future science group

Novel surface-engineered solid lipid nanoparticles of rosuvastatin calcium 

lower levels of lipid and higher amount of surfactant. However, the formulations containing intermediate to lower levels of lipid showed relatively slower drug release characteristics. Modeling of the drug release kinetics using Korsmeyer–Peppas model showed the values of ‘n’ to range between 0.278 and 0.541. Further, observations revealed that at higher levels of lipid, the values of ‘n’ were relatively higher in magnitude. Search for the optimum formulation & validation studies

Search for optimum formulation was carried out by ‘trading-off’ various CQAs to attain the desired objectives by minimization of the particle size (i.e., imperative for ease of permeation and absorption of the drugs), and maximization of ζ potential (i.e., imperative for formulation stability), entrapment efficiency and T80% (i.e., necessary for controlled drug release profile). Based on the aforesaid objectives, the selection criteria were embarked upon for locating the optimized formulation, with particle size 20 mV, encapsulation efficiency >85%, and T80% ranging between 6 and 8 h. The optimized formulation was identified by numerical optimization with the magnitude of desirability function close to unity. Figure 2 portrays the optimum SLNs formulation demarcated in the design space plot, construing a composition of Compritol 888 (412 mg), Tween-80 (6.25%) and stirring speed (2625 r.p.m.), while exhibiting the particle size of 63.5 nm, ζ potential of -25.5 mV, entrapment efficiency of 89.5% and T80% of 8.01 h. Figure 3 portrays the particle size distribution and transmission electron microscopy image of the optimized SLN formulation with spherical particles in the nanosize range. Evaluation of the

Research Article

surface-modified PL-SLNs and PL-PEG-SLNs showed particle size between 121.5 and 138.2 nm, and ζ potential between -28.4 and -32.3 mV. No significant difference, however, was observed in the parameters like entrapment efficiency and in vitro drug release characteristics. Validation of the experiential methodology, using the prepared check-point formulations revealed close similarity of the predicted responses with those of the observed ones, as revealed by high magnitude of ‘R’ of the correlation plots ranging between 0.948 and 0.972 (p  0.05) in the formulation characteristics was observed between the SLNs before and after incubation, thus confirming adequate stability of the ­optimized SLNs in the GI pH conditions.

Table 4

Cell culture experiments Cellular cytotoxicity Figure 4 depicts the percent cell viability data of Caco-2

cells treated with pure drug suspension, optimized SLN formulations (i.e., plain SLNs, PL-SLNs and PLPEG-SLNs) and Triton-X 100. The cytotoxicity studies showed cell viability >90% for pure drug suspension and the SLNs at all the studied concentrations with no significant (p > 0.05) difference between the number of cells seeded before and after incubation with the formulation. Further evaluation of cytotoxicity by mea-

Cumulative percent drug release

140 F16 F17 F1 F2 F4 F5 F7 F8 F9 F12 F13 F15 F19 F20 F20

120 100 80 60 40 20 0 0

6

12 Time (h)

18

24

Figure 1. In vitro drug release profiles of the solid lipid nanoparticles prepared as per the experimental design.

future science group

www.futuremedicine.com

10.2217/nnm-2016-0336

Research Article  Beg, Jain, Kushwah et al.

Design-Expert® Software Factor coding: Actual Overlay plot

1.0 T80%: 8.31299

Particle size Zeta potential Entrapment efficiency T80%

Particle size Zeta potential Entrapment T80%: Entrapment X1 Particle size X2

0.5

Actual factor C: Stirring speed = 0.0540541

B: Tween 80 (%)

X1 = A: Compritol 888 X2 = B: Tween 80

Zeta potential: 27.4934

0.0

Zeta potential: 22.8593

63.5117 25.5382 89.5804 8.0134 0.494602 -0.0375 Zeta potential: 22.8593

Zeta potential: 27.4934

T80%: 8.31299 Entrapment efficiency: 90.1628

-0.5

T80%: 6.2 -1.0 -1.0

-0.5

0.0

0.5

1.0

A: Compritol 888 (mg) Figure 2. Design space overlay plot depicting the optimized solid lipid nanoparticle formulation.

suring TEER values ratified insignificant (p > 0.05) change after incubation of Caco-2 cells, either with pure drug suspension or with SLN formulations (data not shown). On the contrary, the cells incubated with Triton-X 100 showed highly significant (p > PL-SLNs > plain SLNs. Enhancement in the cellular uptake was also confirmed with the help of confocal laser scanning microscopy imaging, where the optimized PLPEG-SLNs revealed much higher fluorescent intensity than that of the pure drug suspension (Figure 5C).

Cellular uptake & transport studies Passive cellular uptake

Endocytotic uptake

depicts a bar diagram of concentrationand time-dependent cellular uptake of the drug from various treatment formulations on Caco-2 cells. The results indicated significant (p  Plain SLNs vis-à-vis the pure drug (p 

Suggest Documents