http://informahealthcare.com/ddiISSN: 0363-9045 (print), 1520-5762 (electronic)

Drug Dev Ind Pharm, Early Online: 1–10! 2013 Informa Healthcare USA, Inc. DOI: 10.3109/03639045.2013.850710

RESEARCH ARTICLE

Design, optimization and evaluation of poly-e-caprolactone (PCL) basedpolymeric nanoparticles for oral delivery of lopinavir

Punna Rao Ravi, Rahul Vats, Vikas Dalal, Nitin Gadekar, and Aditya N

Department of Pharmacy, BITS-Pilani Hyderabad Campus, Jawaharnagar, Ranga Reddy (Dist.), Andhra Pradesh, India

Abstract

Lopinavir (LPV)-loaded poly-"-caprolactone (PCL) nanoparticles (NPs) were prepared byemulsion solvent evaporation technique. Effects of various critical factors in preparation ofloaded NPs were investigated. Box–Behnken design (BBD) was employed to optimize particlesize and entrapment efficiency (EE) of loaded NPs. Optimized LPV NPs exhibited nanometericsize (195.3 nm) with high EE (93.9%). In vitro drug release study showed bi-phasic sustainedrelease behavior of LPV from NPs. Pharmacokinetic study results in male Wistar rats indicated anincrease in oral bioavailability of LPV by 4-folds after incorporation into PCL NPs. From tissuedistribution studies, significant accumulation of loaded NPs in tissues like liver and spleenindicated possible involvement of lymphatic route in absorption of NPs. Mechanistic studiesusing rat everted gut sac model revealed endocytosis as a principal mechanism of NPs uptake.In vitro rat microsomal metabolism studies demonstrated noticeable advantage of LPV NPs byaffording metabolic protection to LPV. These studies indicate usefulness of PCL NPs inenhancing oral bioavailability and improving pharmacokinetic profile of LPV.

Keywords

Box–Behnken design, design of experiments,endocytosis, everted gut sac, lopinavir,nanoparticles, pharmacokinetics, poly-"-caprolactone, Wistar rats

History

Received 9 May 2013Revised 13 September 2013Accepted 19 September 2013Published online 4 November 2013

Introduction

Human immunodeficiency virus (HIV) is a lentivirus from theRetroviridae family responsible for acquired immunodeficiencysyndrome (AIDS). At present there are two known types of HIV,HIV-1 and HIV-2 with HIV-1 being much more virulent,transmittable and prevalent and the cause of majority of HIVinfection in the world1. HIV mainly resides in the anatomical(CNS, lymphatic system, liver, spleen and lungs) and cellularreservoirs (i.e. CDþ T lymphocytes and monocytes/macrophages)of the human body2. Majority of the anti-retroviral drugs areunable to reach these inaccessible ‘‘viral reservoirs’’/HIV local-ization sites3.

Lopinavir (LPV) is a newer and more promising HIV proteaseinhibitor (PI). It is an essential part of Highly Active AntiRetroviral therapy (HAART) and a new standard of care for HIVinfected patients in anti-retroviral therapy4. Oral dose of LPV is400 mg twice daily. However, LPV shows poor oral bioavailabilityin humans due to low aqueous solubility (52 mg/ml), extensivefirst pass metabolism and P-glycoprotein (P-gp) efflux5. It hasbeen reported to undergo extensive pre-systemic metabolism viagut and hepatic cytochrome P450 (CYP3A)6. Therefore, it fails toachieve therapeutic concentration in blood and target viralreservoirs when given alone7. In order to improve the oralbioavailability of LPV, both pharmacokinetic8,9 and novelformulation approaches including peptide pro-drugs of LPV10,

melt-extruded LPV tablet formulation11, surface stabilized LPVnanocrystals12 and LPV-loaded nanoparticles (NPs)13,14 havebeen explored in recent past.

Nanotechnology-based drug carrier/delivery approaches arethe focal point of today’s therapeutic research. Drug nanocarrierslike NPs offer several advantages like improvement in oralpharmacokinetics of poorly bioavailable drugs by avoiding theirpre-systemic metabolism/P-gp mediated efflux and targeted drugdelivery to the reticulo-endothelial system (RES)15,16.

Poly (lactic-co-glycolic acid) (PLGA) NPs containing com-bination of anti-retroviral drugs including LPV has been reportedfor improving oral bioavailability and therapeutic efficacy ofthese drugs17. PLGA NPs showed a significant improvement inintracellular profile of LPV. However, the reported formulationwas aimed to include combination of anti-retroviral drugsresulting in low entrapment efficiency (EE) for LPV.

Poly-"-caprolactone (PCL) is semi-crystalline biodegradableand biocompatible polyester with low glass transition temperatureand melting point18. It has been widely investigated in recent pastfor drug delivery applications. It is non-toxic and non-muta-genic19. It is degraded slowly and rate of degradation is slowerthan any other biodegradable polyesters20 which makes thispolymer suitable for delivering drugs for chronic therapy. Beinghydrophobic and slow degrading polymer PCL deemed suitablecarrier to entrap lipophilic drug like LPV. Moreover, it isconsiderably economical than other polymers such as polyglyco-lide, polylactide and their copolymers21.

Primary objective of present work was to prepare andcharacterize LPV-loaded PCL NPs and to study in vivo perform-ance of optimized formulation. LPV NPs were rationallydesigned and optimized using Plackett–Burman Design (PBD)

Address for correspondence: Punna Rao Ravi, Department of Pharmacy,BITS-Pilani Hyderabad Campus, Jawaharnagar, Ranga Reddy (Dist.),Andhra Pradesh, India. Tel: þ91 40 66303539. Fax: þ91 40 66303998.E-mail: rpunnarao@hyderabad.bits-pilani.ac.in, rpunnarao@gmail.com

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

and Box–Behnken Design (BBD) and data were statisticallyanalyzed using Design Expert software (Full version 8.0.7.1, Stat-Ease Inc., Minneapolis, MN).

Materials and methods

Materials

Lopinavir was obtained as a gift sample from MatrixLaboratories, Hyderabad, India. Poly-"-caprolactone (Molecularweight 65 000 g/mol) was purchased from Polysciences Inc.,Warrington, FL. Rat intestinal microsomes (RIM), rat livermicrosomes (RLM) and NADPH were procured from BDGentest, Woburn, MA. HPLC grade acetonitrile, ammoniumacetate, methanol, methylene chloride (DCM) and sodium citratewere purchased from Merck Laboratories, Mumbai, India.Ingredients of Krebs-Henseleit bicarbonate buffer were purchasedindividually from Sigma–Aldrich, Mumbai, India. Methyl cellu-lose (molecular weight 14 000 Da, viscosity 15 cps) and Tween 80were purchased from S.D. Fine Chemicals Ltd, Mumbai, India.A Milli-Q water purification system (Millipore, Billerica, MA)was used for obtaining high quality HPLC grade water.

Methods

Preparation of drug-loaded NPs

Lopinavir-loaded polymeric NPs were prepared by previouslyreported oil-in-water emulsion-solvent evaporation technique22

with minor modifications. This method was adopted formanufacturing formulations listed in Box–Bhenken Design(BBD). The manufacturing conditions and component rangeswere selected after initial screening using Plackett–BurmanDesign (PBD). Briefly, LPV (10 mg) and specific amount ofpolymer (100–300 mg) were dissolved in 5 ml of DCM, whichconstituted organic phase. This organic phase was slowly addedinto the 50 ml aqueous phase containing Tween 80 (0.5–1.5% w/v)under magnetic stirring (800 rpm) to form a primary emulsion.Primary emulsion was further subjected to high speed homogen-ization (at 5000 rpm, Polytron PT 3100D, Kinematica, Lucerne,Switzerland) for specific time period (15–45 min). Resultantcolloidal preparation was centrifuged at 20 000� g for 45 min toobtain LPV-loaded NPs. Prepared NPs were washed three timeswith water to remove adherent free drug from the outer surface ofNPs. Washed NPs were re-suspended in water and subjected topre-freezing at �80 �C for 6 h. Further, freeze-drying was carriedfor 12 h at �110 �C in a lyophilizer (Coolsafe 110-4, Scanvac,Lynge, Denmark). Five percent mannitol was used as a cryopro-tectant. This lyophilized powder was stored in sealed glasscontainers at room temperature till further use.

Experimental design

Low resolution PBD, a factorial design matrix, was used toidentify critical formulation and manufacturing factors in prep-aration of LPV-loaded NPs. Total of eight factors were studied attwo levels to determine their effect on two responses, namely EEand particle size (nm) of loaded NPs. The variables studied were:Amount of polymer, type of surfactant (Tween 80 and poloxamer407), concentration of surfactant, type of external phase (DCMand chloroform), volume of external phase, speed of homogen-ization, time of homogenization, rate of external phase addition(rapid and slow).

Based on particle size and EE data obtained from PBD, weselected Tween 80 (as surfactant) and DCM (as external phasesolvent) for the preparation and optimization of LPV-loaded NPs.Three critical factors including concentration of surfactant,amount of polymer and time of homogenization were identified

for optimization process. From these initial experiments, limitsand range for identified critical factors were set for subsequentoptimization studies using BBD.

Box–Behnken design, a sub-type of response surface method-ology (RSM), was employed to develop quadratic models foroptimization process and to reduce number of experimental trials.A 17-run, 3-factor, 3-level BBD was constructed to evaluate maineffects, interaction effects and quadratic effects of identifiedinitial factors. Non-linear quadratic model generated by BBDdesign was in the following form (1):

Y ¼ b0 þ b1X1 þ b2X2 þ b3X3 þ b12X1X2

þ b13X1X3 þ b23X2X3 þ b11X21 þ b22X2

2 þ b33X23

Where, Y is measured response associated with each factorlevel combination; b0� b33 are regression coefficients of respect-ive factors and their interaction terms computed from observedexperimental values of Y and X1, X2, X3 are the coded levels ofindependent factors. The terms X1X2, X2X3, X3X1 and X2

i (i¼ 1, 2or 3) represent interaction and quadratic terms, respectively.Dependent and independent factors selected are shown in Table 1.Critical factors evaluated in this study were concentration ofsurfactant (X1), amount of polymer (X2) and time of homogen-ization (X3). Responses studied were particle size (Y1) and EE(Y2). Experiment design matrix generated by the software isshown in Table 2.

Table 2. Box–Behnken experimental design.

Critical factors Response variables

Run

Surfactantconcentration(X1, % w/v)

Polymeramount

(X2, mg)

Homogenationtime

(X3, min)

Particlesize

(Y1, nm)

Entrapmentefficiency

(Y2, %)

1 1.0 200 30 170.5 92.32 1.0 100 45 138.4 78.53 1.0 200 30 174.1 92.84 0.5 200 15 215.0 81.15 0.5 300 30 204.2 86.66 1.5 100 30 188.9 80.37 1.0 200 30 178.4 88.88 1.0 300 45 176.9 93.49 1.5 200 15 232.6 88.7

10 1.0 300 15 212.1 89.011 0.5 200 45 174.2 87.812 1.5 300 30 230.7 94.413 0.5 100 30 165.1 73.514 1.0 200 30 171.4 87.515 1.0 200 30 172.2 87.216 1.5 200 45 208.6 92.317 1.0 100 15 163.3 76.8

Table 1. Critical factors and their levels in the Box–Behnken Design(BBD).

Levels used

Critical factorsLow(�1)

Medium(0)

High(þ1)

IndependentX1¼ Surfactant concentration (% w/v) 0.5 1 1.5X2¼ Polymer amount (mg) 100 200 300X3¼Time of homogenization (min) 15 30 45

Dependent ConstraintsY1¼ Particle size (nm) MinimumY2¼Entrapment efficiency (%) Maximize

2 P. R. Ravi et al. Drug Dev Ind Pharm, Early Online: 1–10

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

Particle size and zeta potential analysis

Particle size and zeta potential of optimum NP dispersions weremeasured by Zetasizer (3000SH, Malvern Instruments Ltd.,Worcestershire, UK). All samples were diluted with doubledistilled water to reach a suitable concentration beforemeasurement.

Entrapment efficiency determination

Drug EE was determined by ultra filtration method23 usingAmicon Ultra centrifugal filters (Millipore, Billerica, MA)(mwco: 10 kDa). Briefly, microfilters containing 0.5 ml of NPssuspended in water were centrifuged at 6000� g for 30 min toseparate un-entrapped drug (free drug, Wfree) from total drug(Wtotal) added to the formulation. Un-entrapped drug (i.e. drugdiffused through the membrane) was quantified by RP-HPLCmethod24. EE was calculated by following equation (2):

EE %ð Þ ¼ Wtotal �Wfreeð Þ=Wtotal½ � � 100

Scanning electron microscopy analysis

Polymeric NPs were examined for morphology under scanningelectron microscope (JSM-6360LV Scanning Microscope; Jeol,Tokyo, Japan). Before analysis, 100ml of NP dispersion was driedovernight on aluminum stub under vacuum which was sputter-coated using a thin gold-palladium layer under an argonatmosphere using a gold sputter module in a high-vacuumevaporator (JFC-1100 fine coat ion sputter; Jeol, Tokyo, Japan).Coated samples were then scanned and photomicrographs weretaken at an acceleration voltage of 15 kV.

Differential scanning calorimetry

Differential scanning calorimetry (DSC) analysis was carried outusing DSC 60 (Shimadzu Corporation, Kyoto, Japan) instrument.Accurately weighted samples were taken in aluminum pan andcrimp sealed. Samples were equilibrated at 25 �C in DSCchamber. After sufficient equilibration time, samples weresubjected to heating run over temperature range of 25–200 �C ata heating rate of 5 �C/min.

In vitro release study

In vitro drug release study of optimized formulation wasperformed using dialysis bag method25. Dialysis bag (mwco 12–14 kDa, pore size 2.4 nm), containing 1.5 mg equivalent NPs, wasimmersed in 50 ml phosphate buffer saline (pH 7.4) containing0.1% w/v Tween 80 maintained at 37� 0.5 �C and stirred at50 rpm. At pre-determined time intervals, to maintain sinkcondition, drug release media was completely replaced. Ananalogous study was also preformed with an equal amount of freeLPV. Cumulative release of LPV in sample solution wasdetermined by RP-HPLC method.

Obtained data were fitted into zero order, first order, Higuchiand reciprocal-powered time mathematical models26 for evalu-ation of release kinetics. Regression coefficient (r2) and time for50% dissolution (t50%) were calculated for the best-fit model.

Stability studies

Optimized NPs were subjected to stability testing as perInternational Conference on Harmonization (ICH) Q1A (R2)guidelines (ICH, 2003)27. Briefly, optimized LPV-loaded NPswere stored in sealed glass vials at 25� 2 �C/60� 5% relativehumidity in stability chamber (Remi, Mumbai, India). Sampleswere analyzed over 3-month period for particle size, zeta potentialand EE. Statistical evaluation was done using GraphPad

Prism version 5.03 for Windows software (GraphPad Software,Inc., San Diego, CA).

Pharmacokinetic evaluation of optimized NPs in Wistar rats

Male Wistar rats, weighing 180–220 g, were used in this study.Experimental protocol was approved by the Institutional AnimalEthics Committee (Approval No.: IAEC-01/01-12). All animalswere housed under constant environmental conditions (22� 1 �Croom temperature; 55� 10% relative humidity; 12 h light/darkcycle) and were allowed access to food and water ad libitum.Animals were fasted overnight (12 h) before dosing and continuedon fasting until 4-h post-administration of formulation.Thereafter, rat chow diet was provided ad libitum. In all studies,freshly prepared drug formulations were administered.

Pharmacokinetic studies were conducted post-oral (20 mg/kg,10 ml/kg) and post-intravenous (IV, 4 mg/kg, 1 ml/kg) adminis-tration of free LPV suspension and LPV-loaded NPs. For bothoral and IV pharmacokinetic study, two groups with five animalsin each group were made. While control group received freesuspension (LPV suspended in 0.5% w/v methyl cellulose),treatment group received optimized LPV NPs formulation.Similarly, in IV pharmacokinetic study, control group receivedfree LPV solution (LPV dissolved in PEG:Ethanol:Water pre-mix) while the treatment group received LPV NPs formulation.

Blood samples (150 ml) were collected from the orbital sinusinto microfuge tubes containing anti-coagulant (3.8% w/v sodiumcitrate) at pre-dose, 0.17, 0.25, 0.5, 1, 2, 4, 6, 8 and 12 h post-dosefor oral studies and at pre-dose, 0.08, 0.17, 0.25, 0.5, 1, 2, 4, and6 h post-dose for IV studies. Collected blood samples were kepton ice until further processing. These samples were furtherharvested for plasma by centrifuging at 4 �C for 10 min at 650� gand then stored at �70 �C until further analysis. The samples wereanalyzed by a validated HPLC method for estimation of LPV inrat plasma matrix24.

Tissue distribution study

To study LPV bio-distribution, 24 male Wistar rats (180� 20 g)were divided into two groups of 12 rats each. They were giveneither free LPV (suspension prepared with 0.5% w/v methylcellulose) or LPV-loaded NPs at a dose of 20 mg/kg via oralgavage. Three rats per formulation were sacrificed at 0.5, 1, 2 and4 h post-dosing. Tissues of interest (spleen and liver) werecollected immediately after cervical dislocation and they wereblotted dry with tissue paper. Samples were frozen at �70 �C untilfurther analysis.

Prior to analysis, tissue samples were thawed to roomtemperature, chopped into small pieces and homogenized to afine paste in tissue homogenizer (Remi, Mumbai, India) alongwith water (20% w/v). LPV was extracted from tissue homogenateby adding protein precipitating agent, acetonitrile in the ratio of1:3 (v/v). Extracted samples were centrifuged (6000� g for15 min) and resultant clean supernatant (75ml) was injected intoHPLC to determine the concentration of drug in the samples.

In vitro metabolic stability study

In vitro metabolic stability studies were performed by incubatingfree LPV, free LPV with blank NPs and LPV-loaded NPs withRIM and RLM (1 mg/ml) at an effective concentration of 5mM.Reaction was initiated by addition of NADPH (2 mM) inphosphate buffer (100 mM, pH 7.4). Incubations were performedat 37 �C in a shaking water bath for 30 min. Reaction wasterminated by addition of cold acetonitrile. Samples werevortexed briefly and centrifuged at 6000� g for 15 min.Resultant clean supernatant (75ml) was injected. Percentage

DOI: 10.3109/03639045.2013.850710 Lopinavir-loaded poly-"-caprolactone nanoparticles 3

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

metabolism of LPV was determined in all three test conditions. Ina separate set of experiment, free LPV was incubated withreaction buffer without enzymes for 30 min to assure no non-enzymatic degradation occurred (data not shown).

Enzyme inhibition due to formulation excipients present inLPV NPs was investigated by incubating blank NPs (NPsprepared without drug) along with free LPV and difference inmetabolism (compared with free LPV alone) was taken as enzymeinhibition.

Uptake of LPV-loaded NPs into rat everted gut sac

Everted gut sac studies in rats were performed using estab-lished methods adapted from literature28. Briefly, rats werefasted overnight for 12 h but allowed water ad libitum beforethe experiment. After anesthesia (urethane, 1.25 g/kg), entireintestine was removed by cutting across upper end of theduodenum and lower end of the colon and stripping mesenterymanually. Intestine was washed carefully with normal saline(0.9% w/v NaCl) and different segments of small intestine wereidentified. A length of 8–10 cm from jejunum and ileum wasrapidly removed and gently everted over a glass rod. Evertedintestine was then slipped off the glass rod and placed in a flatdish containing Krebs–Henseleit bicarbonate (KHB) bufferoxygenated with O2/CO2 (95%/5%) at 37 �C. One end wasclamped and tied with a silk suture before filling it with KHBbuffer at 37 �C using 0.5 ml syringe. These intestinal sacs werethen slipped off the needle carefully and loose ligature atproximal end was tightened. Intestinal sacs were then placed inindividual incubation chambers containing either free LPV orLPV NPs (2.5 mg/ml) prepared in oxygenated KHB buffer atmaintained temperature of 37 �C.

To discern the uptake mechanism, everted gut sacs wereincubated at 4 �C and also in the presence of specificendocytosis inhibitors like chlorpromazine (CPZ) (10mg/ml)or nystatin (NYT) (25mg/ml) at 37 �C. In order to evaluateregional differences due to expected variation in expression ofendocytic uptake cells (M cells) on apparent permeability(Papp) of LPV, all experiments were conducted in both ratjejunum and ileum segments. After pre-set incubation time of60 min, intestinal sacs were carefully removed, blotted ontofilter paper and contents were collected. Intestinal sacs wererinsed thrice with KHB buffer and rinsings were pooled withoriginal content for analysis. Samples were analyzed with avalidated HPLC method.

Papp values, expressed in cm/s, were calculated in eachexperimental condition using the following equation (3):

Papp ¼ dQ=dt=A� Co

where dQ/dt is the rate of appearance of LPV in the everted gutsac (receiver compartment), Co is the initial concentration of LPVoutside everted gut sac (donor compartment) and A is total crosssectional area of tissue.

HPLC analysis

Previously reported24 simple and rapid RP-HPLC method wasused for the analysis of LPV in rat plasma under isocraticconditions. Briefly, Shimadzu LC-20 AD Series HPLC system(Shimadzu Corporation, Kyoto, Japan) consisting of ShimadzuLC-20 AD HPLC pump, Shimadzu series DGU-20A5 Degasserand a Shimadzu SIL HTC auto-sampler was used to inject 75 mlaliquots of the processed samples on an endcapped RP-C18column (Luna�, 250 mm� 4.6 mm, 5mm, Phenomenex, Torrance,CA) which was maintained at a temperature of 40 �C. Theisocratic mobile phase consisted of an aqueous phase (10 mMammonium acetate, pH 6.5) and acetonitrile (35:65 v/v). Buffer

was filtered through 0.22 mm Millipore membrane filter anddegassed ultrasonically for 5 min before use. LPV was monitoredat wavelength of 210 nm.

The above-reported method was partially validated for theanalysis of the drug in tissue samples like spleen and liver. Nointerference was observed by tissue matrices at the retention timeof LPV. Method was found to be reproducible with good recovery(490%) for the drug.

A single step protein precipitation technique was used toextract LPV from rat plasma matrix and tissue samples.The detector response was linear over the concentration rangeof 250–4000 ng/ml.

Statistical analysis

All in vitro studies were performed in triplicate and datafrom these experiments are expressed as mean�SD. Non-compartmental pharmacokinetic analysis was performed usingPhoenix� WinNonlin� (Pharsight Inc., Mountain View, CA)to determine various pharmacokinetic parameters. Unpaired t-test or Analysis of variance (ANOVA), followed by Dunnett’stest (Graphpad Prism, version 5.03) was used to assess anysignificance of difference between means. The significantlevel was set at 5%.

Results

Experimental design

Selected critical factors showed statistically significant effect onobserved responses for particle size and EE (Table 3). Quadraticequations establishing main effects and interaction factors weredetermined based on estimation of statistical parameters gener-ated by software. Statistical validation of quadratic equations wasconfirmed by ANOVA. Three dimensional response surfacegraphs for illustrating effects of selected critical factors onselected responses are shown in Figure 1.

Effects on particle size (Y1)

As shown in Table 2, particle size of formulations ranged between138.4 nm (run 2) and 232.6 nm (run 9), indicating sensitivitytoward critical factors selected for the study. Experiments carriedout at center points (run 1, 3, 7, 14 and 15; n¼ 5) of the designindicate reproducibility of experiment as coefficient of variation(CV) is 52%. Independent factors affecting particle size wereconcentration of surfactant (X1), amount of polymer (X2) andduration of homogenization (X3) (Table 3).

These effects can be explained by following quadraticequation (3):

Y1 ¼ 176:80þ 12:50 X1ð Þ þ 21:13 X2ð Þ � 15:63 X3ð Þþ 0:75 X1X2ð Þ � 4:25 X1X3ð Þ � 2:5 X2X3ð Þ þ 27:35 X2

1

� �

� 7:4 X22

� �þ 3:1 X2

3

� �

Regression coefficient (r2) of the above equation was 0.9563,indicating good correlation between observed response andselected critical factors. The residuals were distributed randomlyaround zero and there was no effect of experimental sequence ontrend of residuals.

Effects on entrapment efficiency (Y2)

As shown in Table 2, EE varied between 73.5% (run 13) and94.4% (run 12) which indicates that the response was sensitivetowards selected factors. Experiments performed at centerpoints of the design (run 1, 3, 7, 14 and 15; n¼ 5) confirmedthat experimental method was highly reproducible (CV53%).

4 P. R. Ravi et al. Drug Dev Ind Pharm, Early Online: 1–10

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

Figure 1. (a) Response surface plot showing effect of polymer amount and surfactant concentration on particle size. (b) Response surface plot showingeffect of time of homogenization and surfactant concentration on particle size. (c) Response surface plot showing effect of polymer amount andsurfactant concentration on entrapment efficiency. (d) Response surface plot showing effect of time of homogenization and surfactant concentration onentrapment efficiency.

Table 3. Statistical analysis results of particle size and entrapment efficiency (EE).

Particle size (Y1) EE (Y2)

Source Sum of squares DF F Value p Value Sum of squares DF F Value p Value

Model 10247.33 9 635.07 0.0001* 659.95 9 126.74 0.0001*X1 1250.00 1 697.21 0.0001* 91.12 1 157.50 0.0001*X2 3570.13 1 1991.30 0.0001* 378.12 1 653.55 0.0001*X3 1953.13 1 1089.39 0.0001* 32 1 55.31 0.0001*X1X2 2.25 1 1.25 0.2996# 0.25 1 0.43 0.5320#

X1X3 72.25 1 40.30 0.0004* 1 1 1.73 0.2300#

X2X3 25.00 1 13.94 0.0073* 1 1 1.73 0.2300#

X21 3149.57 1 1756.73 0.0001* 20.84 1 36.03 0.0005*

X22 230.57 1 128.60 0.0001* 114.95 1 198.67 0.0001*

X23 40.46 1 22.57 0.0021* 9.16 1 15.83 0.0053*

Residual 12.55 7 4.05 7Lack-of-fit 9.75 3 4.64 0.0861# 3.25 3 5.42 0.0681#

Pure error 2.80 4 0.8 4Total 10259.88 16 664 16

*Significant at �50.05.#Not significant at �50.05.

DOI: 10.3109/03639045.2013.850710 Lopinavir-loaded poly-"-caprolactone nanoparticles 5

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

Data presented in Table 3 show that independent factors affectingEE were concentration of surfactant (X1), amount of polymer (X2)and time of homogenization (X3).

Effect can be explained by following second-order polynomialquadratic equation (4):

Y2 ¼ 90:7þ 3:38 X1ð Þ þ 6:88 X2ð Þ þ 2:00 X3ð Þ þ 0:25 X1X2ð Þ� 0:5 X1X3ð Þ þ 0:5 X2X3ð Þ � 2:23 X2

1

� �� 5:23 X2

2

� �� 1:48 X2

3

� �

Regression value of above equation was found to be 0.922indicating suitability of selected design model. Analysis ofresiduals indicated that residuals were normally distributedaround zero.

Optimization and validation

Desirability function (0.98) was probed using Design-Expertsoftware to acquire optimized formulation. Selection of optimumformulation was based on pre-set criteria as shown in Table 1.Conditions for optimal formulation as predicted by the soft-ware were as follows: surfactant concentration¼ 1.36 mg/ml,polymer amount¼ 270.74 mg and duration of homogenization¼40.15 min. To prove validity of this statistical model, verificationruns (n¼ 6) with these conditions were carried out and Wilcoxon-Signed Rank test was used to identify any statistically significantdifference between actual and theoretical values. At a¼ 0.05,there was no statistically significant difference between actualand theoretical values for particle size (p50.0732) and EE(p50.8543) thus affirming validity of proposed model.Optimized formulation exhibited particle size of 195.3� 2.3 nmand EE of 93.95� 1.23%.

Physicochemical characterization of NPs

Scanning electron microscopy studies revealed that LPV-loadedPCL NPs were almost spherical in shape (Figure 2). Mean particlesize, poly dispersibility index (PDI) and zeta potential value ofoptimized LPV NPs (n¼ 6) were 195.3� 2.3 nm, 0.10� 0.01 and�19.74� 2.1 mV, respectively. Negative zeta potential wasattributed to the presence of lactone residues on polymericmatrix surface.

Figure 3 shows DSC thermograms for pure LPV, bulk PCL,physical mixture of LPV and PCL (1:1), blank NPs and LPV-loaded PCL NPs. DSC thermogram for pure LPV showed sharpmelting peak at 97.2 �C while bulk PCL showed melting peak at56.3 �C. In DSC thermograms of blank and PCL-loaded NPs, anadditional peak observed at 168.3 �C was of mannitol (used ascryoprotectant).

In vitro drug release profile of optimized NPs formulation andfree LPV is presented in Figure 4. Free LPV showed completedissolution within 5 h. LPV NPs showed a bi-phasic releasepattern, which was characterized by an initial rapid release(60%) in first 10 h followed by slow and continuous drugrelease up to 75 h. Drug release kinetics was studied by fittingdata into various mathematical models. From regression analysis,drug release from NPs was most appropriately described byreciprocal-powered time model (r2¼ 0.9877). In comparison,zero-order kinetics (r2¼ 0.2873), first-order kinetics (r2¼ 0.9399)and Higuchi kinetics (r2¼ 0.8987) showed relatively lower r2

values. Time taken for 50% drug release (t50%) from NPs wascalculated to be 7.34 h.

Stability studies

Stability estimation for optimized PCL NPs was done on basisof particle size, EE, zeta potential and PDI variations during3-month study period. Data are illustrated in Figure 5. Resultsshow that there was negligible change in assessed parameterswhen LPV NPs are stored at 25� 2 �C/60� 5% relativehumidity.

Figure 3. Overlaid DSC thermograms of LPV, PCL, PM, blank NPs andLPV NPs. LPV, lopinavir; PCL, poly-"-caprolactone; PM, physicalmixture (1:1); NPs, nanoparticles.

Figure 4. In vitro drug release profile of free LPV and LPV NPs insimulated blood pH condition (phosphate buffer saline (PBS), pH 7.4).The data are expressed as mean� SD. of six independent determinations(n¼ 6). LPV, lopinavir; NPs, nanoparticles.

Figure 2. Scanning electron micrograph of lopinavir-loaded solid lipidnanoparticles (�7500).

6 P. R. Ravi et al. Drug Dev Ind Pharm, Early Online: 1–10

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

Pharmacokinetic studies

Pharmacokinetic behavior of free LPV and LPV-loaded PCL NPsin male Wistar rats after oral administration is shown in Figure 6.Comparative pharmacokinetic parameters of free LPV and LPVNPs post-IV and oral administration are given in Table 4.Significant change in pharmacokinetics of LPV was observedfollowing IV administration of LPV NPs as compared to free LPV.Total plasma clearance (CL) of LPV from PCL NPs reduced by26% (p50.05), whereas, significant increase in mean retentiontime (MRT increased by 1.82-folds, p50.05), and area under thecurve (AUC increased by 1.91-folds, p50.01) was observed ascompared to free LPV formulation.

Following oral administration, polymeric NPs showed statis-tically significant improvement in pharmacokinetics of LPV asdetermined by AUC (increased by 4.4-folds, p50.001), Cmax

(increased by 3.0-folds, p50.001) and MRT (increased by 1.30-folds, p50.05).

Uptake of LPV-loaded NPs into rat everted gut sac

Data obtained from permeability studies are shown in Table 5.Experimental results indicate a significant increase in apparentpermeability (Papp) of LPV in LPV NPs as compared to free LPV.The Papp of loaded LPV was found to increase by 1.88-fold(p50.01) in jejunum and 2.37-fold (p50.01) in ileum region ofintestine. Low incubation temperature (4 �C) and presence ofspecific inhibitors significantly (p50.05) reduced the intestinaluptake of NPs. As shown in Table 5, intestinal uptake of NPsat 4 �C was only �39% (p50.01) of that at 37 �C in jejunumand �44% (p50.01) of that in ileum. The Papp of LPV NPswas reduced by 37% in jejunum and 33% in ileum region afterco-incubation with CPZ. Similarly, in presence of NYT Papp ofloaded NPs significantly decreased by 34% in jejunum and 40%in ileum region.

In vitro metabolic stability study of LPV

Results obtained from in vitro metabolic stability studies usingRIMs and RLMs are shown in Figure 7. Mean percentagemetabolism of LPV in loaded NPs (24% in RIMs; 22% in RLMs)was found to reduce significantly (p50.001) as compared to freeLPV (90% in RIMs; 85% in RLMs) after 30 min of incubationperiod in both microsomes. Statistically no significant change in

metabolism of free LPV was observed upon co-incubation withblank NPs.

Tissue-distribution study

In order to establish in vivo performance of LPV-loaded PCL NPsat targeted organs, tissue distribution study was conducted and

Figure 5. Stability characteristics of freeze-dried LPV-loaded PCL NPs interms of mean particle size, entrapment efficiency (EE), Zeta potential andpolydispersity index (PDI) stored at 25� 2 �C and 60� 5% RH. The data are expressed as mean� SD of six independent determinations (n¼ 6). LPV,lopinavir; PCL, poly-"-caprolactone; NPs, nanoparticles.

Figure 6. Mean plasma concentration versus time profile of free LPV andLPV NPs after oral administration to male Wistar rats (n¼ 5). The dataare expressed as mean� SD. LPV, lopinavir; NPs, nanoparticles.

Table 4. Pharmacokinetic parameteres of free LPV and LPV NPsfollowing oral and IV administration to Wistar rats (n¼ 5).

Route ParametersFree LPV(Group A)

LPV NPs(Group B)

IV(4 mg/kg)

Co (ng/ml) 2534.01� 134.12 1874.16� 120.14**MRT (h) 0.998� 0.04 1.813� 0.01**CL (ml/h/kg) 1543.59� 107.34 1134.28� 95.13*Vd (ml/kg) 1578.59� 116.54 1924.20� 109.45*AUC (ng/ml*h) 2592.41� 215.23 4951.00� 321.34**

Oral(20 mg/kg)

Cmax (ng/ml) 645.85� 67.7 1893.13� 20.16***Tmax (h) 0.85(0.75–1.0) 1.1 (0.75–2.0)MRT (h) 4.99� 0.50 6.51� 0.49*AUC (ng/ml*h) 1555.52� 53.34 6829.44� 98.59***BA (%) 12.43 54.60

The data are expressed as mean� SD. LPV, lopinavir.*p50.05, **p50.01, ***p50.001 as compared to Free LPV (Group A).

DOI: 10.3109/03639045.2013.850710 Lopinavir-loaded poly-"-caprolactone nanoparticles 7

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

compared with distribution pattern of free LPV. As shown inFigure 8, it is evident that LPV from loaded NPs was significantly(p50.01) accumulated in liver (AUC increased by 2.7-folds, Cmax

increased by 3.2-folds) and spleen (AUC increased by 2.10-folds,Cmax increased by 1.6-folds) tissues as compared to free LPVpost-oral administration.

Discussion

Experimental design

Results of experimental design provided considerable usefulinformation and reaffirmed utility of statistical design for conductof experiments. As shown in Figure 1(a), for a given concentrationof surfactant, an increase in polymer amount leads to significantincrease in particle size. This could be due to increase in viscosityof continuous phase with increase in amount of polymer whichresults in reduction of evaporation rate of organic phase therebyleading to formation of larger particles29. It is also evident that atconstant polymer amount, with increase in surfactant concentra-tion up to an optimal level, particle size decreases; beyond this, anincrease in particle size is seen. Initial reduction of particle sizeby surfactant was attributed to reduction of interfacial tensionbetween dispersed organic phase and dispersion media (aqueousphase)30. However, at higher surfactant concentrations,

hydrophobic interactions between surfactant molecules dominate,leading to aggregation and increase in particle size.

As shown in Figure 1(b), increase in the length of homogen-ization time resulted in reduced particle size. Longer duration ofhomogenization results in increased energy input that preventsagglomeration of NPs.

Effect of surfactant concentration and polymer amount on theEE at constant homogenization time is shown in Table 3 andillustrated in Figure 1(c). A steep curvature for EE when viewedfrom polymer axis indicates that, with increasing amount ofpolymer, EE increases. This is expected because with increasingamount of polymer, the lipophlic drug LPV gets better entrappedin hydrophobic polymeric matrix. Higher amount of polymer alsoprovides additional number of particles into which LPV getsentrapped.

Likewise, from Figure 1(c) at fixed homogenization time, it isevident that with increasing surfactant concentration, EE of LPVincreases. This is because more LPV molecules get entrappedwithin surfactant layer present at surface of NPs, leading to a highEE31. Further, with increasing polymer and surfactant concentra-tion, viscosity of the medium increases preventing rapid diffusionof LPV into the bulk of medium leading to higher EE32.

From polynomial equation for EE and Figure 1(d), it is evidentthat homogenization time has a positive effect on EE. Withincreasing homogenization time, energy input increases leading toreduction of particle size and increase in surface area whichincreases EE.

Physicochemical characterization of NPs

Optimized formulation exhibited particles in nano size range withhigh EE and low PDI value. Low PDI indicates that optimalconditions could be used to produce stable LPV NPs with arelatively narrow size distribution. In a previously reported work,

Figure 8. Tissue distribution study of free LPV and LPV NPs followingoral administration (20 mg/kg) to Wistar rats. Three animals weresacrificed at each time point (n¼ 3) to harvest (a) spleen and (b) livertissues. The data are expressed as mean� SD. LPV, lopinavir; NPs,nanoparticles.

Table 5. Effect of temperature and endocytic uptake inhibitors(chlorpromazine, 10 mg/ml and nystatin, 25 mg/ml) on intestinalapparent permeability of LPV NPs.

Papp (�10�5 cm/s)

Groups Jejunum Ileum

Free LPV at 37 �C 2.58� 0.31 2.94� 0.34Free LPV at 4 �C 2.63� 0.29 3.13� 0.32Free LPVþCPZ 2.72� 0.32 2.89� 0.33Free LPVþNYT 2.44� 0.22 2.93� 0.28LPV NPs at 37 �C 4.86� 0.35** 6.97� 0.45**LPV NPs at 4 �C 1.90� 0.24@@ 3.08� 0.29@@

LPV NPsþCPZ 3.06� 0.21@ 4.67� 0.31@

LPV NPsþNYT 3.22� 0.26@ 4.19� 0.27@

The data are expressed as mean� SD. LPV, lopinavir; CPZ, chlorpro-mazine; NYT, nystatin.

**p50.01 as compared with Free LPV at 37 �C.@@p50.01 as compared with LPV-loaded NPs at 37 �C.@p50.05 as compared with LPV-loaded NPs at 37 �C.

Figure 7. Metabolic stability of free LPV, free LPVþ blank NPs andLPV NPs after 30 min of incubation with RIM and RLM at 1 mg/mlprotein concentration. ***p50.001 versus free LPV. Data presented aremean of six independent determinations (n¼ 6). LPV, lopinavir; NPs,nanoparticles.

8 P. R. Ravi et al. Drug Dev Ind Pharm, Early Online: 1–10

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

LPV was loaded into PLGA NPs with maximum EE of 45% andparticle size of 334 nm. In stark contrast, our optimized formu-lation manufactured by similar method demonstrated high EE(95%) and lower particle size (195 nm) indicating usefulness oflogical design in optimization.

Drug release profile indicates bi-phasic release pattern. Initialrapid release may be due to presence of LPV molecules on surfaceof NPs entrapped in surfactant layer. In second phase, theobserved slow release is due to slow diffusion of entrapped drugfrom NPs matrix.

In Figure 3, absence of LPV peak in thermograms of loadedNP formulation indicates that majority of LPV is present inamorphous form within the polymeric matrix. No shift in peakposition for LPV or PCL was observed in physical mixtureindicating absence of incompatibility between PCL and LPV.

Ex vivo and in vivo studies

In vivo performance of optimized formulation was evaluatedin male Wistar rats. Series of comparative pharmacokineticstudies were conducted to understand mechanism involved inpharmacokinetic improvement of LPV NPs as compared tofree LPV.

From oral pharmacokinetic studies it is evident that LPV haspoor bioavailability; presumably due to both high first-passmetabolism and P-gp efflux. Significant improvement in plasmaexposure of LPV in NPs could be attributed to reducedmetabolism/P-gp efflux and/or increased intestinal permeabilityof LPV upon loading. A significant reduction in plasma clearanceand extended MRT following IV administration of LPV NPsindicates metabolic protection (in liver) by polymeric NPs to LPV.

In a previously reported study17, PLGA NPs for LPV wereprepared and administered by intra-peritoneal route. It was foundthat, compared to free LPV, serum concentration of LPV fromPLGA NPs increased by over 8-folds. Further, higher distributionof NPs to liver, kidney, spleen and brain was also reported. Ourdata are in reasonable agreement with the previously reporteddata. However, in our case, optimized LPV-loaded PCL wereadministered by oral route. Apparently, these results should beviewed with sagaciousness, as route of administration in bothcases are different.

Intestinal uptake of NPs was investigated in rat everted gut sacmodel to understand the mechanism of its absorption andcontribution of absorption in improving oral bioavailability ofLPV NPs. A significant increase in Papp of LPV in LPV NPs ascompared to free LPV suggested that NPs could efficiently crossintestinal barriers by active endocytosis (phagocytosis/pinocyt-osis) process via M-cells and also protect drug from P-gp effluxupon loading33. It is widely reported that at lower temperatures(4 �C), energy-dependent processes like endocytosis could beblocked34. Hence, we investigated uptake of LPV NPs at twodifferent temperatures, 37 �C (Control) and 4 �C. The dataindicate significant reduction in Papp value of LPV in LPV NPs.However, no significant effect of incubation temperature onintestinal permeability of free LPV across intestinal segments wasfound. This implied that uptake of NPs in everted gut sacs couldbe possibly a result of active uptake process.

Further, it has been reported that specific endocytic uptakeinhibitors like chlorpromazine (CPZ) and nystatin (NYT) couldreduce uptake of NPs by either inhibiting clathrin-coated pitassociated receptors at cell surface or abolishing caveolaefunction, respectively35. Hence, intestinal uptake study of LPVwas performed in the presence of CPZ and NYT. From the studyresults, it could be deduced that both clathrin- and caveolaemediated endocytosis mechanisms were involved in the uptake ofLPV NPs. However, no significant change in the Papp value of

free LPV in the presence of CPZ and NYT was found indicatingthe absence of active absorption process for free LPV. It alsooverrules interaction of CPZ and NYT with P-gp and othertransporter systems.

In vitro metabolic stability studies conducted with RIM andRLM show extensive metabolism of free LPV and alsodemonstrated that NPs could offer metabolic protection to drug.Metabolic protection offered by NPs to LPV (gut wall andhepatic) aids in achieving longer circulation time leading tohigher plasma exposure of LPV.

Data obtained from tissue distribution studies indicate signifi-cant accumulation of LPV in spleen and liver tissues from LPVNPs suggesting lymphatic uptake of NPs following oral admin-istration. Similar behavior has been reported in recent past fororally administered NPs36. It is reported that M-cells which coverlymphoid Peyer’s patches, take up NPs by combination ofendocytosis or transcytosis mechanism37. Size of particlesprecludes their absorption into blood capillaries and thereforethey are secreted into lymph.

It has been noticed that viral reservoirs present in lymphoidalorgans like spleen and liver are poorly accessed by conventionaltherapy. Minimum effective concentration of drug cannot bemaintained for the necessary time duration at the site of HIVlocalization38. Higher distribution of LPV-loaded NPs in spleenand liver tissues at all-time points assures higher LPV concen-tration in these reservoirs. Thus, better therapeutic outcome ofLPV from LPV NPs could be expected in comparison toconventional LPV therapy.

Conclusion

LPV was successfully fabricated in PCL NPs with high EE anddesirable particle size range. Critical factors for manufacturingthese LPV NPs were identified and optimized using DOE withgood correlation between actual and predicted values. OptimizedLPV NPs showed a significant increase in oral bioavailability ascompared to free LPV suspension. Metabolic protection offeredby polymeric NPs was demonstrated as one of the possiblereasons for improvement in oral bioavailability of loaded LPV. Itwas also demonstrated that NPs were taken up by endocytosisprocess. Higher plasma and tissue (liver and spleen) concentrationof LPV from PCL NPs was also attributed to lymphatic transportof NPs. In conclusion, formulating PCL NPs for LPV was aneffective approach in improving its oral bioavailability andreducing dos,e which may prove beneficial in the treatment ofHIV-infected patients.

Declaration of interest

The authors report no conflicts of interest. The authors alone areresponsible for the content and writing of the article.

References

1. Freed EO. HIV-1 replication. Somat Cell Mol Genet 2001;26:13–33.2. Schrager LK, D’Souza MP. Cellular and anatomical reservoirs of

HIV-1 in patients receiving potent antiretroviral combinationtherapy. JAMA 1998;280:67–71.

3. Pomerantz RJ. Reservoirs of human immunodeficiency virus type 1:the main obstacles to viral eradication. Clin Infect Dis 2002;34:91–7.

4. Farmer P, Leandre F, Mukherjee J, et al. Community-basedtreatment of advanced HIV disease: introducing DOT-HAART(directly observed therapy with highly active antiretroviral therapy).Bull World Health Organ 2001;79:1145–51.

5. Temesgen Z, Warnke D, Kasten MJ. Current status of antiretroviraltherapy. Expert Opin Pharmacother 2006;7:1541–54.

6. du Plooy M, Viljoen M, Rheeders M. Evidence for time dependentinteractions between ritonavir and lopinavir/ritonavir plasma levels

DOI: 10.3109/03639045.2013.850710 Lopinavir-loaded poly-"-caprolactone nanoparticles 9

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.

following P-glycoprotein inhibition in Sprague-Dawley rats. BiolPharm Bull 2011;34:66–70.

7. Pierson T, McArthur J, Siliciano RF. Reservoirs for HIV-1:mechanisms for viral persistence in the presence of antiviralimmune responses and antiretroviral therapy. Annu Rev Immunol2000;18:665–708.

8. Piacenti FJ. An update and review of antiretroviral therapy.Pharmacotherapy 2006;26:1111–33.

9. Ravi PR, Vats R, Thakur R, et al. Effect of grapefruit juice andritonavir on pharmacokinetics of lopinavir in Wistar rats. PhytotherRes 2012;26:1490–5.

10. Agarwal S, Boddu SH, Jain R, et al. Peptide prodrugs: improved oralabsorption of lopinavir, a HIV protease inhibitor. Int J Pharm 2008;359:7–14.

11. Else LJ, Douglas M, Dickinson L, et al. Improved oral bioavail-ability of lopinavir in melt-extruded tablet formulation reducesimpact of third trimester on lopinavir plasma concentrations.Antimicrob Agents Chemother 2012;56:816–24.

12. Jain S, Sharma JM, Jain AK, Mahajan RR. Surface-stabilizedlopinavir nanoparticles enhance oral bioavailability without co-administration of ritonavir. Nanomedicine (Lond) 2013;8:1639–55.

13. Aji Alex MR, Chacko AJ, Jose S, Souto EB. Lopinavir loaded solidlipid nanoparticles (SLN) for intestinal lymphatic targeting. Eur JPharm Sci 2011;42:11–18.

14. Shibata A, McMullen E, Pham A, et al. Polymeric nanoparticlescontaining combination antiretroviral drugs for HIV type 1treatment. AIDS Res Hum Retroviruses 2013;29:746–54.

15. Kingsley JD, Dou H, Morehead J, et al. Nanotechnology: a focus onnanoparticles as a drug delivery system. J Neuroimmun Pharmacol2006;1:340–50.

16. Sharma P, Garg S. Pure drug and polymer based nanotechnologiesfor the improved solubility, stability, bioavailability and targeting ofanti-HIV drugs. Adv Drug Deliv Rev 2010;62:491–502.

17. Destache CJ, Belgum T, Goede M, et al. Antiretroviral release frompoly(DL-lactide-co-glycolide) nanoparticles in mice. J AntimicrobChemother 2010;65:2183–7.

18. Prabu P, Dharmaraj AACN, Khil MS, et al. Preparation, character-ization, in-vitro drug release and cellular uptake of poly(caprolac-tone) grafted dextran copolymeric nanoparticles loaded withanticancer drug. J Biomed Mater Res A 2008;90:1128–36.

19. Shah LK, Amiji MM. Intracellular delivery of saquinavir inbiodegradable polymeric nanoparticles for HIV/AIDS. Pharm Res2006;23:2638–45.

20. Ulery BD, Nair LS, Laurencin CT. Biomedical applications ofbiodegradable polymers. J Polym Sci B Polym Phys 2011;49:832–64.

21. Woodruff MA, Hutmacher DW. The return of a forgotten polymer –polycaprolactone in the 21st century. Prog Polym Sci 2010;35:1217–56.

22. Byun Y, Hwang JB, Bang SH, et al. Formulation and characteriza-tion of a-tocopherol loaded poly-"-caprolactone (PCL) nanoparti-cles. LWT Food Sci Technol 2011;44:24–8.

23. Yue PF, Lu XY, Zhang ZZ, et al. The study on the entrapmentefficiency and in vitro release of Puerarin submicron emulsion.AAPS PharmSciTech 2009;10:376–83.

24. Vats R, Murthy AN, Ravi PR. Simple, rapid and validated LCdetermination of lopinavir in rat plasma and its application inpharmacokinetic studies. Sci Pharm 2011;79:849–63.

25. Shah M, Pathak K. Development and statistical optimization of solidlipid nanoparticles of simvastatin by using 23 Full-Factorial Design.AAPS PharmSciTech 2010;11:489–96.

26. Barzegar-Jalali M, Adibkia K, Valizadeh H, et al. Kinetic analysis ofdrug release from nanoparticles. J Pharm Pharm Sci 2008;11:167–77.

27. ICH Q 1 A (R2). Stability testing of new drug substances andproducts. (CPMP/ICH/2736/99). Available from: http://www.emea.eu.int/pdfs/human/ich/273699en.pdf.

28. Barthe L, Woodley J, Houin G. Gastrointestinal absorption ofdrugs: methods and studies. Fundam Clin Pharmacol 1999;13:154–68.

29. Lamprecht A, Ubrich N, Hombreiro Perez M, et al. Influences ofprocess parameters on nanoparticle preparation performed by adouble emulsion pressure homogenization technique. Int J Pharm2000;196:177–82.

30. Ma Y, Zheng Y, Zeng X, et al. Novel docetaxel-loaded nanoparticlesbased on PCL-Tween 80 copolymer for cancer treatment. Int JNanomed 2011;6:2679–88.

31. Mayank S, Krishna C, Mishra A, Pathak K. Oral solid compritol 888ATO nanosuspension of simvastatin: optimization and biodistribu-tion studies. Drug Dev Ind Pharm 2011;37:526–37.

32. Yang Y, Chung T, Bai X, Chan, W. Effect of preparation conditionson morphology and release profiles of biodegradable polymericmicrospheres containing protein fabricated by double-emulsionmethod. Chem Eng Sci 2000;55:2223–36.

33. Vyas TK, Shah L, Amiji MM. Nanoparticulate drug carriers fordelivery of HIV/AIDS therapy to viral reservoir sites. Expert OpinDrug Deliv 2006;3:613–28.

34. Zhang Z, Gao F, Bu H, et al. Solid lipid nanoparticles loadingcandesartan cilexetil enhance oral bioavailability: in vitro charac-teristics and absorption mechanism in rats. Nanomedicine 2012;8:740–7.

35. Roger E, Lagarce F, Garcion E, Benoit JP. Lipid nanocarriersimprove paclitaxel transport throughout human intestinal epithelialcells by using vesicle-mediated transcytosis. J Control Release 2009;140:174–81.

36. Albanese A, Tang PS, Chan WCW. The effect of nanoparticle size,shape, and surface chemistry on biological systems. Annu RevBiomed Eng 2012;14:1–16.

37. Porter C, Trevaskis N, Charman W. Lipids and lipid basedformulations: optimizing the oral delivery of lipophilic drugs. NatRev Drug Discov 2007;6:231–48.

38. Wong HL, Chattopadhyay N, Wu XY, Bendayan R. Nanotechnologyapplications for improved delivery of antiretroviral drugs to thebrain. Adv Drug Del Rev 2010;62:503–15.

10 P. R. Ravi et al. Drug Dev Ind Pharm, Early Online: 1–10

Dru

g D

evel

opm

ent a

nd I

ndus

tria

l Pha

rmac

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y St

Joh

ns U

nive

rsity

on

01/0

7/14

For

pers

onal

use

onl

y.