Polymer assisted entrapment of netilmicin in PLGA nanoparticles for sustained antibacterial activity

http://informahealthcare.com/mncISSN: 0265-2048 (print), 1464-5246 (electronic)

J Microencapsul, Early Online: 1–14! 2014 Informa UK Ltd. DOI: 10.3109/02652048.2014.944951

ORIGINAL ARTICLE

Polymer assisted entrapment of netilmicin in PLGA nanoparticles forsustained antibacterial activity

Atul Kolate1,2, Girish Kore1, Pierre Lesimple2, Dipesh Baradia1, Sushilkumar Patil1, John W. Hanrahan2, andAmbikanandan Misra1

1Pharmacy Department, Faculty of Technology and Engineering, The Maharaja Sayajirao University of Baroda, Vadodara, Gujarat, India, and2Department of Physiology, McGill University, Montreal, Quebec, Canada

Abstract

This study was aimed to develop poly(DL-lactide-co-glycolide) (PLGA) nanoparticle of highlywater soluble antibiotic drug, netilmicin sulfate (NS) with improved entrapment efficiency (EE)and antibacterial activity. Dextran sulfate was introduced as helper polymer to formelectrostatic complex with NS. Nanoparticles were prepared by double emulsificationmethod and optimized using 25-1 fractional factorial design. EE was mainly influenced bydextran sulfate: NS charge ratio and PLGA concentration, whereas particle size (PS) was affectedby all factors examined. The optimized NS-loaded-NPs had EE and PS of 93.23 ± 2.7% and140.83 ± 2.4 nm respectively. NS-loaded-NPs effectively inhibited bacterial growth compared tofree NS. Sustained release protected its inactivation and reduced the decline in its killing activityover time even in presence of bronchial cells. A MIC value of 18 mg/mL was observed for NPs onP. aeruginosa. Therefore, NPs with sustained bactericidal efficiency against P. aeruginosa mayprovide therapeutic benefit in chronic pulmonary infection, like cystic fibrosis.

Keywords

Antibiotic, antimicrobial effect, nanoparticle,PLGA, Pseudomonas aeruginosa, sustainedrelease

History

Received 31 January 2014Revised 6 July 2014Accepted 8 July 2014Published online 19 September 2014

Introduction

Over the last decade, biodegradable polymeric nanoparticles(NPs) have gained increasing interest due to their many advan-tages in drug delivery. These include protection of the drugagainst degradation and deleterious biological interactions,improved bioavailability, solubility, retention time and intracel-lular penetration (Kumari et al., 2010). Thus NPs are a promisingapproach for the delivery of therapeutics such as hydrophilicand hydrophobic drugs, vaccines, proteins and biologicalmacromolecules that could be used for the treatment of diseasessuch as cancer, infection and inflammatory disorders(Soppimath et al., 2001; Kumari et al., 2010). NPs based onpoly(DL-lactide-co-glycolide) (PLGA) have been widely investi-gated due to their small particle size, favorable safety profile, andability to provide sustained release (Mundargi et al., 2008). Theyare approved for human use by the Food and Drug Administrationand several PLGA-based formulations are available in worldwidemarket (Okada, 1997). Lipophilic drugs can be encapsulatedeasily into hydrophobic PLGA matrix; however, entrapment oflow molecular weight hydrophilic drugs in PLGA matrix presentsa great challenge (Danhier et al., 2012).

Barichello et al. (1999) reported low entrapment efficiency(EE) of hydrophilic drugs in PLGA NPs when compared withlipophilic drugs. The major reason was the poor affinity between

the hydrophilic drug and PLGA, which promotes drug diffusioninto the external aqueous medium resulting in low EE (Barichelloet al., 1999). Thus, attempts have been made to enhance theentrapment of hydrophilic drugs into PLGA hydrophobic matrixby nanoprecipitation (Saxena et al., 2004), single emulsification(Tewes et al., 2007), W/O/W double emulsification (Lecarozet al., 2007; Tewes et al., 2007), and in the case of PLGA NPs,by use of reverse micelles (Gao et al., 2008). Nevertheless, theapplication of PLGA NPs for entrapping hydrophilic drugsremains limited.

Netilmicin sulfate (NS) and other aminoglycoside antibioticsare used for the treatment of P. aeruginosa infections in cysticfibrosis due to their concentration-dependent antibacterial activityand long-term post-antibiotic effect (Karlowsky et al., 1994;Prayle and Smyth, 2010). However their incorporation into aPLGA matrix is problematic because their amino groups andhydroxyl groups cause them to be polar, and their high watersolubility inhibits their incorporation into the hydrophobic core ofPLGA (Lecaroz et al., 2007). Moreover they penetrate into thesputum of CF patients poorly and their activity is reduced due tobinding with sputum components such as glycoproteins andcations (Honeybourne, 1994; Kuhn, 2001). Consequently highsystemic doses are needed to achieve effective concentrations ofthe drug at the infection site (Ratjen et al., 2009). High doses canresult in nephrotic and ototoxic effects, triggering permanent renalinsufficiency and auditory nerve damage, with deafness, dizzinessand unsteadiness, reducing the therapeutic window (Forge andSchacht, 2000).

Recently it has been demonstrated that nanoparticulate drugdelivery is a promising approach to deliver antibiotics (Sanderset al., 2000; Suk et al., 2009). NPs can provide controlledantibiotic release thereby maintaining a constant plasma

Correspondence: Dr. Ambikanandan Misra, 0265 Professor in Pharmacyand Dean, Faculty of Technology and Engineering, The MaharajaSayajirao University of Baroda, Kalabhavan, Vadodara 390001, Gujarat,India. E-mail: [email protected]

Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

concentration above the minimum inhibitory concentration (MIC)for a prolonged duration. This maximizes the therapeutic effectwhile minimizing antibiotic resistance, and also improves patientcompliance (Gao et al., 2011). However, attempts to use NPs forthe delivery of aminoglycosides have shown little drug entrap-ment in the PLGA hydrophobic core (Lecaroz et al., 2007;Abdelghany et al., 2012; Asadi, 2013). Thus more efficientformulation techniques are needed. To our knowledge there hasbeen no systematic study of the effects of processing andformulation parameters on the EE of NS in PLGA NPs.

The aim of the present study was to investigate the effectsof important process and formulation parameters on the entrap-ment of the highly water soluble ionic drug NS in PLGAhydrophobic matrix, and evaluate the antibacterial capacity ofthe resulting NPs against P. aeruginosa. This should provideinsight into how polymeric NPs can be used to improvebioavailability and provide sustained release of antibiotic drugs,and thus reduce the dosing frequency and side effects associatedwith the drug.

Materials and methods

Materials

Netilmicin sulfate was provided as a gift sample by Samarth LifeSciences Pvt. Ltd., Mumbai, India. Poly(DL-lactic-co-glycolicacid) (PLGA, 50:50) (Mw¼ 17 kDa) was received as a giftsample from Purac Biochem, Netherlands. Dextran sulfate sodium(DS, Mr¼ 5000) was purchased from Himedia, Mumbai, India.Pluronic� F68 (Poloxamer 188, Sigma-Aldrich, India), ethylacetate, and sucrose were purchased from Sigma-Aldrich, India.Amicon Ultra 15 filtration units were purchased from Millipore.

Development of drug-loaded nanoparticles

Nanoparticles (NPs) were prepared by a (w/o/w) double emulsionsolvent evaporation method as previously described with somemodifications (Bilati et al., 2005). Briefly, water in oil emulsionwas prepared by adding 0.15 to 0.35mL of an internal aqueousphase (W1) containing Netilmicin sulfate (NS) in double-distilledwater to 0.5mL ethyl acetate (organic phase, O) containing PLGApolymer (10–30mg/mL) by means of probe sonication (SartoriusLabsonic M., Gottingen, Germany) at 60 amplitude for 90 sec onice bath. Then primary (W1/O) emulsion was added drop-wise to2mL 1.5% w/v of Pluronic� F68 solution (external aqueous phase,W2), and further emulsified by sonication for another 30 sec bymean of probe sonication at 60 amplitude (on ice bath) to form aW1/O/W2 double emulsion. The resulting double emulsion wasfurther diluted with 5mL of 1.5% (w/v) Pluronic� F68 solution andstirred magnetically for 4h at room temperature to evaporate theethyl acetate. The remaining organic solvent and free drug wereremoved by washing the NPs 3 times with double-deionized waterusing an Amicon Ultra-15 centrifuge filter (10-kDa) (Millipore,Billerica, MA), then re-suspended in double-distilled water toobtain the desired final concentration. The NPs were usedimmediately or frozen at �70 �C and lyophilized using sucroseas a cryoprotectant to obtain solid product, which was further keptovernight in desiccators to remove remaining moisture (secondarydrying) and then stored at �20 �C for later use.

To examine the effect of helper hydrophilic polymerdextran sulfate (DS) on the EE of NS, NPs were prepared usingthe same procedure but with DS in the internal aqueous phaseof the emulsion (with different charge-to-drug ratios). Theblank PLGA NPs were also prepared using the same procedurewithout adding NS. Various factors affecting the entrapment ofthe drug in the polymer were evaluated by using the factorialdesign.

Experimental design

To elucidate the effect of different process conditions on theentrapment of NS in PLGA NPs and particle size of NPs,experiments were designed using the JMP 10.02 software (SAS,Cary, NC). A preliminary screening was carried out to determinewhich process parameters have the most significant effects on theNPs. Results of the preliminary study suggested that five factorscan influence the characteristics of the NPs. The chosenindependent process parameters included: the volume ratiobetween the inner water phase and the oil phase (X1), thePLGA concentration (mg/mL) (X2), the surfactant concentration(%) (X3), the DS-to-NS charge ratio (X4), and the sonication time(sec) of the primary emulsion (X5).

A 25-1 fractional factorial design (FFD) was then created inSAS-JMP 10.02 and used to determine the effect of these processparameters on NP characteristics. Each process parameter wasevaluated at the two levels; low and high levels assigned by (�1)and (+1), respectively as listed in Table 1. The levels were chosenin such a way that they provided a maximal design spaceand enabled feasible processing of the NPs. In addition, twocenter points were included to evaluate the curvature of thestatistical model. Each measurement was conducted in triplicate.The effects of the independent variables on two responsevariables: EE of the NS (Y1) and the particle size (Y2), aregiven in Table 2.

The applied 25-1 FFD is of resolution V, which indicatedthat the main effects are confounded with four-factor interactions,and two-factor interactions are confounded with three-factor interactions. The main effects and the two-factorinteractions are comprised in the statistical model stated inEquation (1):

Y ¼ �0 þX

�iXi þX

�ijXiXj þ " ð1Þ

Where, Y is the measured response, �0 is the arithmetic meanresponse, �ij is the interaction term �i, �j are the estimatedparameters, X is the coded value of the independent parameter ande is the residual error. The effects of the main factors and the two-factor interactions were analyzed by analysis of variance.Statistical analysis was considered significant when the p valueswere lower than 0.05. The coefficients in Equation (1) werecalculated using coded values of the independent parametersusing Equation (2):

xi ¼Xi � Xi;high þ Xi;low

� �=2

Xi;high � Xi;low

� �=2

ð2Þ

where Xi is the coded value of the independent variable Xi, andXi, high and Xi, low are the high and low levels of the independentvariable Xi. Statistical models were accepted when there was nolack of fit, no correlation in the residual plots and the residualswere normally distributed.

Table 1. Actual and coded values of the formulation parameters.

Levels

Variables Factors UnitLow

level (�1)Centrallevel (0)

HighLevel (+1)

X1 Vw1/VO ratio 0.3 0.5 0.7X2 CPLGA mg/ml 10 20 30X3 CPluoronic F68 % 0.5 1 1.5X4 DS:NS charge

ratioratio 0.5 1.5 2.5

X5 tsonication sec 30 60 90

2 A. Kolate et al. J Microencapsul, Early Online: 1–14

Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

Nanoparticle characterization

Particle size

NPs size (diameter, nm) and polydispersity index were deter-mined by dynamic light scattering using a Malvern ZetasizerNano ZS (Malvern Instruments, Malvern, UK) (Patel et al.,2012b). The NPs suspensions were diluted suitably with filtereddouble-distilled water to avoid multi-scattering phenomenaand placed in a disposable sizing cuvette. The polydispersityindex was also evaluated to investigate the narrowness of theparticle size distribution. All the analysis were carried out intriplicates.

Zeta potential

Zeta (z) potential measurements of samples were made with aMalvern Zetasizer Nano ZS (Malvern Instruments). Zeta potentialwas calculated by Smoluchowski’s equation from the electro-phoretic mobility. Measurements were taken after 10-fold dilutionof the NPS suspension with filtered, double-distilled water in adisposable zeta cuvette. All the analysis were carried out intriplicate.

Determination of entrapment efficiency

Actual loading of NS in NPs was evaluated by an indirect method(Ungaro et al., 2012). To determine the drug EE, NS-loaded NPswere centrifuge filtered in a 10-kDa Amicon Ultra-15 centrifugefilter for 4000g at 4 �C to remove free drug. Filtrate was collectedand the amount of unentrapped NS was determined in a UVspectrophotometer (UV 1800, Shimadzu, Kyoto, Japan) bymeasuring absorbance at 335 nm after derivatization with orthophthaldehayde (OPA) at 60 �C for 15min. A calibration plot wasdeveloped with NS concentration between 10–60 mg. The linearityof response was verified over the concentration range 10–60 mg/mL with a regression coefficient (R2) of 0.9996. The limit ofdetection (LOD; estimated as 3 times the background noise) was1.92mg/mL while the limit of quantitation (LOQ; estimated as10 times the background noise) was 5.82 mg/mL. The EE wascalculated from following equation. All the analysis were carriedout in triplicates.

Encalsulation efficiency %ð Þ

¼ Amount of drug loaded in NP

Initial amount of drug used in formulation� 100

Transmission electron microscopy

The Transmission electron microscopy (TEM) study was con-ducted to examine the morphology of NPs. The observations wereconducted by using a transmission electron microscope (HitachiH-7500, 120 kV). A suspension of NPs was coated on a coppergrid, air dried and stained with 2% (w/v) phospho-tungstic acid.Air-dried samples of NPs were then examined directly under thetransmission electronic microscope.

Differential scanning calorimetry

Differential scanning calorimetry (DSC) analysis was carried outto evaluate possible interactions between the NS and the excipientsused, and to evaluate the nature of NS within the NPs. DSC analysisof samples (NS, PLGA, blank NPs, and NS loaded NPs) wasperformed using a TA-60WS Thermal Analyzer (Shimadzu, Kyoto,Japan). The samples were sealed hermetically into an aluminumpan and heating curves were recorded at a scan rate of 10 �C/minfrom 20 to 350 �C in an inert atmosphere, which was maintained bypurging dry nitrogen at a flow rate of 40 mL/min.

FTIR studies

FTIR spectroscopy was used to characterize functional groups.A FTIR Spectrophotometer (BRUKER ALPHA FT-IRSpectrometer, Germany) was used to obtain the FTIR spectrogramof NS, PLGA, physical mixture, and NPs loaded with NS. Thesamples were mixed with potassium bromide (1:100 ratio) andcompressed into a pellet. Then pellet was scanned under the FTIRspectroscope.

In vitro drug release studies

Drug release from the NPs was studied in phosphate buffer pH 7.4.Briefly, 10mg of NPs were dispersed in 1mL phosphate buffer pH7.4 and incubated at 37 �C in a shaking incubator (Patel et al.,2012a). At specific time intervals, samples were centrifuged at13 000 rpm for 5min at 4 �C to isolate NPs; 0.2mL of medium werewithdrawn and replaced by the same amount of fresh medium. Thewithdrawn medium was analyzed for NS content by UV spectro-photometric analysis after derivatization with OPA as describedabove. Experiments were performed in triplicate.

Antibacterial susceptibility testing

The minimum inhibitory concentration (MIC) is defined as thelowest antibiotic concentration that inhibits a visible planktonic

Table 2. Effects of the independent variables on two response variables: entrapment efficiency and the particle size.

Run X1 X2 X3 X4 X5 % EE Particle size (nm) Polydispersivity index (PDI)

1 1 1 �1 1 �1 81.45 ± 1.5 140.79 ± 3.5 0.206 ± 0.0302 1 �1 1 �1 1 50.55 ± 2.3 127.44 ± 2.9 0.180 ± 0.0433 1 �1 �1 1 1 62.76 ± 1.7 134.55 ± 1.7 0.146 ± 0.0304 1 1 1 �1 �1 54.76 ± 1.1 137.36 ± 2.3 0.180 ± 0.0605 �1 �1 �1 �1 1 46.65 ± 1.4 121.43 ± 1.4 0.146 ± 0.0306 0 0 0 0 0 65.44 ± 2.5 133.23 ± 1.8 0.153 ± 0.0417 1 1 �1 �1 1 59.76 ± 3.1 137.94 ± 0.9 0.120 ± 0.0108 �1 1 �1 1 1 81.87 ± 1.0 138.33 ± 1.6 0.150 ± 0.0459 �1 1 1 1 �1 79.23 ± 1.3 134.43 ± 1.4 0.154 ± 0.045

10 1 �1 �1 �1 �1 52.76 ± 0.7 130.27 ± 1.3 0.160 ± 0.03411 1 �1 1 1 �1 65.34 ± 2.9 125.34 ± 1.6 0.140 ± 0.02012 1 1 1 1 1 93.23 ± 2.7 140.83 ± 2.4 0.130 ± 0.01813 �1 1 �1 �1 �1 66.87 ± 1.6 139.55 ± 2.9 0.123 ± 0.01114 �1 �1 �1 1 �1 43.98 ± 0.7 133.33 ± 3.5 0.133 ± 0.01215 0 0 0 0 0 66.85 ± 2.6 132.43 ± 1.6 0.176 ± 0.03716 �1 �1 1 1 1 73.88 ± 1.5 126.43 ± 2.9 0.170 ± 0.04517 �1 1 1 �1 1 63.45 ± 1.3 133.94 ± 0.8 0.166 ± 0.04618 �1 �1 1 �1 �1 41.54 ± 1.5 130.45 ± 2.2 0.160 ± 0.040

DOI: 10.3109/02652048.2014.944951 Netilmicin entrapped PLGA nanoparticles 3

Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

bacterial growth. The MIC of free NS and NS-loaded NPs weredetermined by the broth microdilution method in 96-wellmicroplates (Wikler, 2008). An optical density measurement at600 nm was used to examine the visible bacterial growth in whichOD 60050.1 indicated zero bacterial growth. Briefly,P. aeruginosa was grown in Trypticase soy broth (TSB) at37 �C to obtain an optical density (OD) of �1 at 600 nm(�1� 108 CFU/mL). The bacterial cell suspension was thenfurther diluted by 100-fold. Free NS and NS-loaded NPs wereserially diluted in TSB to make a final NS concentration rangingfrom 40 mg/mL to 0.2mg/mL (calculated from % drug loading) ina final volume of 100mL. Then, 100ml of diluted P. aeruginosainoculum were added to each well and the samples wereincubated at 37 �C for 24h. The MIC endpoints were determinedby reading the OD of the plate wells at 600 nm and confirmed byvisual inspection. The lowest concentration that yielded OD� 0.1was determined as the MIC.

Minimum bactericidal concentration (MBC) is the drugconcentration where no visible growth appears on agar plates.NS and NS-loaded NPs treated bacterial cultures showing growthor no growth in the MIC tests were used for this test. Bacterialcultures that were used for the MIC test were inoculated onto theagar and incubated at 37 �C for 24h. Microbial growth wasassessed by counting colonies. The minimal concentration of theNS and NS-loaded NPs that abolished colony growth wasdetermined as the MBC.

Control NPs were evaluated for possible antibacterial activityagainst P. aeruginosa. Bacteria were cultivated in a 96 well platefollowing the same procedures used above to determine the MIC/MBC. A bacterial suspension were exposed to the blank NPs (1 to5% w/v) and incubated overnight at 37 �C, then the number ofColony-Forming Units (CFU) was determined using the agar platemethod.

To mimic in-vivo conditions, human bronchial epithelial cells(CFBE41o-) were incubated with free NS or NS-loaded NPs, andantimicrobial activity was tested every 24h for 5 days. The cellswere seeded in 12 mm Transwell� (Corning, NY) inserts(250 000cells/well, 12wells, 0.4mm-pore Polyester membrane,Costar 3460). Culture medium was removed from apical surfaceon the 3rd day after plating and particles were added apically in500mL of EMEM (Eagle’s minimal essential medium) withoutserum. Totally 50 mL apical medium was then collected every 24hfor 5 days. Samples were stored at 4 �C. Standard inocula of P.aeruginosa were incubated overnight with shaking at 37 �C in2mL of TSB. Bacteria were diluted to 0.5� 106 cells/50 mL andadded to the 50mL of collected samples and incubated overnightat 37 �C. The bacteria-sample mixture was then diluted and platedonto agar plates and CFU were counted. To evaluate the effect ofNP on cell integrity TEER (transepithelial electrical resistance)was recorded with help of EVOM chopstick electrodes (WorldPrecision Instruments, New Haven, CT).

Cell Culture

CFBE41o- cells were kindly provided by Dr. Dieter Gruenert(California Pacific Medical Center Research Institute andDepartment of Laboratory Medicine, University of California,San Francisco, CA) and were cultured in Eagle’s minimumessential medium with Earls salt (EMEM) (Wisent, St-Bruno, QC,Canada) containing 10% fetal bovine serum (FBS), 1% penicillin/streptomycin, 2mM L-glutamine. Cells were maintained in ahumidified incubator with 5% CO2 at 37 �C.

Safety studies

Cell viability. The viability of CFBE41o- cells after exposureto NPs was evaluated using MTT (3-(4,5-dimethylthiazol-2-yl)-

2,5-diphenyltetrazolium bromide). Briefly, CFBE41o- cells wereseeded into 96 well plates (Falcon 3072) at density 4� 104 cells/well in EMEM containing 10% FBS and 1% antibiotic. Andincubated at 37 �C, 5% CO2 for 24h. After 24h, the cells wereexposed to different concentration of NPs (0.5mg/mL to 5mg/mL)in EMEM for 24h. Cells treated with EMEM and 0.1% sodiumdodecyl sulfate (SDS) were used as negative and positive controls,respectively. After 24h treatment, cells were rinsed thoroughlywith phosphate buffer saline (PBS), treated with 100mL MTTsolution, and incubated for 4h. After 4h of incubation, the mediumwith MTT was replaced with a solubilization solution (100 mLDMSO) to dissolve the formazan crystals formed after internal-ization of MTT by live cells. The resulting coloured solution wasanalyzed using a microplate reader (Synergy Mix, Biotek,Winooski, VT) at 570nm. Cell viability was expressed as thepercentage of absorbance of test samples relative to that of cellstreated with EMEM (n¼ 3).

Transepithelial electrical resistance analysis

Transepithelial electrical resistance (TEER) analysis was per-formed on CFBE41o- cells to determine the integrity of the cellmonolayer upon exposure to nanoparticle suspension. The cellswere seeded onto Transwell� clear permeable filter inserts asdescribed above. Exactly 500mL of medium was added to theapical side of the cell monolayer and 1500mL were added to thebasolateral side. TEER across the insert was measured using anohm meter using chopstick electrodes. TEER was measured fromdays 1 to 3 to ensure that the monolayer was confluent, and thiswas confirmed by light microscopy. The resistance of an insertlacking cells was subtracted from all measurements to correct forthe resistance of the Transwell. On day 3, the apical medium wasreplaced with fresh medium containing the NPs and TEER wasmeasured immediately and again after 1, 2, 3, 4 and 5 days.

Stability studies

As per the ICH guidelines for stability studies, the physical stabilityof the lyophilized NPs was assessed for 6 months (Lalani et al.,2013). The lyophilized NPs were stored in glass vials (USP type 1),closed with rubber closures and sealed with aluminum caps. Thevials were stored at 5 ± 3 �C and 25 ± 2 �C and 60 ± 5% relativehumidity for a period of 6 months. At predetermined time points,samples were evaluated for the particle size, zeta potential and drugcontent. For the test conditions 25 �C ± 2 �C and 60% ± 5% RH,sampling was done at 1, 2, 3 and 6 month intervals. When testing at5 �C ± 3 �C, samples were taken at the end of months 3 and 6. Thedrug content in the initial sample was considered as 100%. Data areexpressed as mean ± SD, n¼ 3. An in-vitro release study was alsoperformed at the end of the stability study. However, the releasedfraction was checked at the end of study because the samples weredry and had no risk of drug release. The NS release profile after thestability study was compared with that obtained before the stabilitystudy. The factor of difference (f1, Equation 3) and similarity (f2,Equation 4) were calculated by considering initial NPs sample asreference. The difference factor (f1) measures the percent errorbetween two curves over all time points (Costa and Sousa Lobo,2001).

f1 ¼Xn

t¼1

Rt � Ttj j" #

=Xn

t¼1

Rt

" #( )� 100 ð3Þ

f2 ¼ 50� log 1þ ð1=nÞXn

t¼1

ðRt � TtÞ2" #�0:5

�100

8<:

9=; ð4Þ


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

where, N¼ number of time points, Rt and Tt¼ dissolution ofreference and test products at time t.

The percentage error between two profiles is zero when the testand reference profiles are identical. In order to consider thesimilar release profiles, the f1 values should be closer to 0 andvalues f2 should be closer to 100 (between 50 and 100) (Costa andSousa Lobo, 2001).

Statistical analysis

The experiments were performed in triplicate, unless otherwisestated. All data were expressed as mean ± standard deviation. Thestatistical significance of the results was determined using aStudent’s t-test where p50.05 as minimum level of significance.

Results and discussion

Development of PLGA nanoparticles

Results of the preliminary study suggested that five factors greatlyinfluence the EE and particle size of NPs. These independentprocess parameters were varied according to the coded valuementioned in Table 1 to obtain the optimized formulation by useof the double emulsification method.

Optimization of the NS-loaded PLGA nanoparticleformulation

25-1 FFD was developed to analyze the effects of the fiveindependent variables on the response variables; i.e. the EE andparticle size (Table 2). These responses are discussed below.

Entrapment efficiency

While developing NPs formulations, entrapment of the hydro-philic drug into the hydrophobic polymer was the mostchallenging task. High solubility caused the antibiotic to diffuseeasily into aqueous phase during preparation of the NPs. The mainaim was therefore to maximize the EE (Y1) of the NS in the NPs.The developed formulations showed EE ranging from 41.55% to93.23% (Table 2). Based on the process parameters, a statisticalmodel was obtained with three significant main effects and onesignificant interaction describing 0.99 of the variation (Table 3).The most significant process parameters affecting the EE of NSinto NPs were the DS:NS charge ratio (X4, b 9.08), the PLGA

concentration (X2, b 8.94), and the Vw1/V0 ratio (X5, b 2.88).The positive sign of the X2, X4 and X5 independent variablesindicates they had positive correlations with the EE; i.e. anincrease in value of these parameters caused an increase in theEE. The level of significance of the charge ratio and PLGAconcentration suggests their direct interaction with drug, whichaffected EE of NS. NS is a water-soluble aminoglycosideantibiotic in which positively-charged groups are expected tolimit entrapment within the hydrophobic PLGA matrix. Indeed,EE varied with changes in the DS to NS charge ratio,demonstrating the important role of electrostatic interactionbetween DS and NS in the preparation of the NPs. Some recentstudies (Wong et al., 2006; Li et al., 2008; Lu et al., 2009) haveshown that complexation of ionic polymers and drugs results inincreased loading capacity of the drug in particles. The electro-static interactions have been validated between the water-solublecationic drugs and counter-ionic polymers (Li et al., 2006, 2008;Wang et al., 2010). The positive charges on the ionic NSmolecule are neutralized by the negative charges of DS. The DS isa clinically applicable polymer showing high charge density dueto high substitution of the sulfate group (Lu et al., 2009). TheNS-DS complexes formed under these conditions were insolublein the aqueous phase and their hydrophobicity enhancesthe incorporation of NS in NPs. Another important factor forthe entrapment of NS in NPs is the polymer concentration. It wasobserved that EE increased with increasing polymer concentration(Song et al., 2008a, 2008b, Lalani et al., 2013). Increasing thepolymer concentration resulted in increased viscosity of theorganic phase. High solution viscosity delays diffusion of the drugtowards the aqueous solution through the polymer drops. Also,high polymer concentration results in faster precipitation of thepolymer on the surface of the dispersed phase, increasing drugentrapment (Song et al., 2008a, 2008b).

The volume ratio of the inner water phase to the oil phase in theprimary emulsion also showed marked effects on the entrapment ofthe NS in NPs. We observed that a lower volume ratio enhancesentrapment. The efficiency was increased significantly when thevolume ratio was increased from 0.3 to 0.7. Increasing the innerwater phase volume should reduce the NS concentration gradientbetween the inner and outer water phase and, consequently limitthe diffusion of NS from the inner to the outer water phase andpromote its entrapment in the PLGA core. However increasing thevolume ratio of the inner water phase to oil phase needs to be done

Tables 3. Results of the fractional factorial design.

% EE (Y1) Particle size (nm) (Y2)

Term Estimate t Ratio Prob4jtj Estimate t Ratio Prob4jtj

Intercept 63.90 109.61 50.0001* 133.22 974.37 50.0001*X1 1.44 2.34 0.1443 1.03 7.17 0.0189*X2 8.94 14.47 0.0047* 4.62 31.86 0.0010*X3 1.61 2.62 0.1204 �1.24 �8.61 0.0132*X4 9.08 14.69 0.0046* 0.97 6.74 0.0213*X5 2.88 4.67 0.0429* �0.66 �4.58 0.0445*X1�X2 �1.72 �2.79 0.1082 0.29 2.03 0.1795X1�X3 �0.72 �1.17 0.3625 �0.32 �2.24 0.1548X1�X4 1.53 2.48 0.1317 0.08 0.58 0.6195X1�X5 �1.39 �2.25 0.1536 1.53 10.61 0.0088*X2�X3 �1.52 �2.47 0.1322 �0.0025 �0.06 0.9604X2�X4 2.28 3.69 0.0663 �0.27 �1.93 0.1939X2�X5 �0.88 �1.44 0.2872 0.52 3.64 0.0678X3�X4 3.58 5.80 0.0285* �1.24 �8.61 0.0132*X3�X5 2.14 3.46 0.0742 0.79 5.49 0.0316*X4�X5 2.32 3.77 0.0638 1.44 9.97 0.0099*R2 value 0.9964 0.9987F value 37.41 103.37Significance 0.0263 0.0096

*Indicates significant difference based on Prob5jtj.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

with care as it may also lead to formation of a less stable primaryemulsion due to coalescence of water droplets. This may increasethe particle size and decrease EE. The obtained results are inagreement with the previous reported studies (Jeffery et al., 1993;Bilati et al., 2005; Cun et al., 2010).

Sonication time plays important role in determining theparticle size of the primary emulsion droplets. A smallerdifference in droplet size between primary and secondaryemulsion results in the formation of a thin oil layer around thewater droplet and thus drug from inner aqueous phase can moreeasily pass through oil phase resulting in lower EE (Ito et al.,2008). Increasing sonication time results in larger size differencebetween primary and secondary emulsions, which enhancesentrapment of the drug.

The ratio between the inner water phase and the organic phasewas varied by changing the volume of the inner water phase, andthis also impacted the EE of NS in NPs. Surfactant concentrationdid not exert a significant direct effect on EE, however aninteraction is evident between the surfactant concentration (X3)and the DS:NS ratio (X4), indicating that a high DS: NS ratio isrequired to obtain high EE (Figure 1). The described two-factorinteraction between the surfactant concentration and the DS: NSratio shows that there is a positive effect of including surfactant inthe formulation at high DS: NS ratio. Surfactant plays a role instabilizing colloidal solutions (Blanco and Alonso, 1998). Theseresults indicate that the emulsion is unstable at high DS: NS ratioand thus require a high surfactant concentration for stability,which in turn enhances the EE.

Particle size

Particle size is responsible for the biopharmaceutical properties ofthe NPs formulation. Smaller particles have a larger surface area,

which can lead to a faster release of the encapsulated drug, thusparticle size affects the release characteristics of the NPs(Yoncheva et al., 2003). The size can also play key role inparticle endocytosis (Calvo et al., 1996). Thus, achieving theoptimal particle size is important for achieving desirablebiopharmaceutical characteristics. Thus, to have NPs withabove-mentioned biopharmaceutical properties, based on theliterature we defined 150 nm as the maximum acceptable diameterfor NPs.

The size of formed NPs varied between 121.43 nm and140.83 nm (Table 2). However the particle size depended onseveral process parameters, namely PLGA concentration (X2, b4.62), ratio of volume of inner aqueous phase to the organic phase(Vw1/Vo) (X1, b 1.03), and DS: NS charge ratio (X4, b 0.978).Apart from these, other parameters had little impact on theparticle size and were part of the significant interaction term,therefore all of them were included in the statistical model todescribe the particle size of NPs (Table 3).

The variation accounted for by the statistical model was0.9987. The positive sign of variables X1, X2 and X4 indicatedtheir positive correlations with particle size while the negativesign of variables X3, X5 indicated their negative correlation withthe particle size. A large interaction was observed between theratio of the inner water phase to the organic phase and sonicationtime (X1 & X5), concentration of the surfactant and sonicationtime (X3 & X5), drug charge ratio and sonication time (X2 &X5), and PLGA concentration and DS: NS charge ratio (X2 &X4). These provided a positive synergistic effect between therespective two factors. The interaction between the ratio ofthe PLGA concentration and sonication time (X2 & X5) mayindicate that a longer sonication time is require to obtain a stableemulsion with small particle size when PLGA added at highconcentration (Figure 1). The ratio of the inner water phase to the

Figure 1. Interaction plots showing significant two-way interaction terms for the dependent variables. Solid lines display factor at high level, whereasdashed lines are the low level of the factors. A: interaction term X3X4 – Puronic F 68 concentration/ DS:NS ratio. B: X1X5 – Volume ratio/sonicationtime. C: X3X4 – Pluronic F 68 concentration/ DS:NS ratio. D: X3X5 – Pluronic F 68 concentration/Sonication time. E: X4X5 – DS:NS ratio/sonicationtime. A: Encapsulation efficiency. B-E: particle size.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

organic phase was varied by changing the volume of the innerwater phase. Low ratio results in the formation of the unstableemulsion thus longer sonication time is require to form stableemulsion which in turn reduces particle size. Also it was observedthat high surfactant concentration is necessary to form a stableemulsion at high DS:NS charge ratio. Further synergistic effectwas observed at high surfactant concentration and long sonicationtime, which results in the lower particle size.

The observed effect of the significant factors on % EE and PSwere further supported by generating 3-D response surface plotsand contour plots using the JMP 10 software as shown inFigure 2(A–E) and Figure 3(A–E) respectively. Figure 2(A)depicts effect of DS: NS ratio & Pluronic F 68 concentration on %EE of NS in PLGA NPs. It can be observed that % EE increaseswith increase in DS: NS ratio, Pluronic F 68 concentrations.Similarly, Figure 3(B–E) shows that PS increases with increase inratio of volume of inner aqueous phase to the organic phase, DS:NS charge ratio and with decrease in surfactant concentration &sonication time. The contour plot for prefixed % EE value wasfound to be linear signifying % EE increases with increase inDS:NS ratio, Pluronic F 68 concentration. On the other hand, thecontour plot for prefixed PS values were found to be non-linearwith curved segment signifying non-linear relationship betweenvariables.

Based on the aforementioned experiments, the optimizedparameters were identified and are listed in Table 4. OptimizedNPs showed EE of 93.23 ± 2.7%, size 140.83 ± 2.4 nm, with apolydispersity index of 0.130 ± 0.018 (Figure 4A) and zetapotential �23.45 ± 3.06 mV. The TEM image of NPs revealsthat NPs are spherical in shape and uniform in size (Figure 4B).It was observed that the particle size of the NPs was increased to265.46 ± 5.01 nm after lyophilization without significant changein EE, or polydispersity index. The zeta potential of all theparticles was negative and in the range of �20.3 mV to �24.5 mV.High negative values of the zeta potential create an electrostaticrepulsion between particles, which helps to prevent aggregation

and thus stabilizes the NPs dispersion (Feng and Huang, 2001).Zeta potential values in the range �15 to �30 mV is common forstabilized NPs (Musumeci et al., 2006).

Differential scanning calorimetry

Differential scanning calorimetry (DSC) analysis is useful fordemonstrating possible interactions between different compoundsin a mixture. The DSC analysis was performed on NS, PLGA,blank NPs and NS-loaded NPs (Figure 5a). PLGA shows theendothermic peak at 44�C. DSC curves of the blank NPs show anendothermic peak at 50 �C owing to the PLGA. The DSC curve ofthe NS showed an endothermic peak at 187.74 �C and drugmelting was followed by thermal decomposition. However,endothermic peak of NS was absent in the DSC curve ofNS-loaded NPs, indicating that NS was present as an amorphousphase in the NPs.

FTIR Spectroscopy

FTIR spectra of all the samples are shown in (Figure 5b). The NSFTIR spectrum showed peaks of C–O and C–N stretching band at1010.91 cm�1 & 1125.94 cm�1, aliphatic C–H stretching band at3017.34 cm�1 & C–H bending band at 619.51 cm�1, �NHstretching and �OH stretching bands at 3100 to 3391.36 cm�1.N–H bending band observed at 1609.83, 1506, 1284.13 cm�1.In the PLGA FTIR spectrum, peaks at 3506.64 cm�1 for �OHstretching and 3000.18–2955.77 cm�1 for C-H stretching bandswere observed as the typical band of PLGA. The ester C¼Ostretching band of PLGA was observed at 1757.01 cm�1. All thetypical bands of PLGA and NS were also present in the FTIRspectrum of their physical mixture. However, the FTIR spectrumof NS loaded NPs had only the characteristic peaks of thepolymer. The absence of characteristic drug peaks in the NPssample indicates that nonencapsulated drug was not present,which is in agreement with previous studies involving PLGAparticulate systems (Gupta et al., 2010; Xiang-Gen et al., 2011).

Figure 2. Surface plot for PLGA NPs; (A) Effect of DS:NS ratio & Pluronic F 68 concentration on EE; (B) Effect of Sonication time & Volume ratio onPS; (C) Effect of DS:NS ratio & Pluronic F 68 concentration on PS; (D) Effect of sonication time & Pluronic F 68 concentration on PS; (E) Effect ofDS:NS ratio & sonication time on PS.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

In-vitro release study

The release profiles of NS from the NS loaded PLGA NPs inphosphate buffer (pH 7.4) at 37 �C is shown in Figure 4(C).Bi-phasic release behaviour was observed, characterized by initialburst release followed by sustained release of NS from NPs.Lower burst release was observed due to the uniform distributionof the NS in NPs rather than just on the surface of the NPs(Li et al., 2008). About 25% of drug content was released withinthe first 24 h, followed by sustained drug release. Though polarstructure of DS contained ether bonds and –OH groups, the strongelectrostatic interaction between NS and DS made the complexneutral and hydrophobic enough to be stored in the PLGA forprolonged time, thus sustaining the drug release from NPs. Inaddition, the lipophilic nature and solid matrix of NPs as well asimmobility of the polymer chains results into slower waterpenetration and drug diffusion. This can be explained byassuming that when the release medium penetrates into the tinypores of NPs and counter ions from the release medium competewith NS and release it from the NS–DS complex through ionexchange, the dissociated NS diffuses out through the pores.

As the ion exchange and drug release progresses, more flexibleand hydrophilic polymer chains would be produced. Eventuallythe dissociated DS molecules are dissolved and diffuse out,leaving the solid matrix more porous and accessible for aqueousmedium to diffuse in, which accounts for the later acceleration ofrelease (Li et al., 2008).

To evaluate the mechanism of drug release from NPs, therelease data were analyzed by using the Korsmeyer–Peppasequation:

Qt=Q1 ¼ k� tn

Where, Qt/Q1 is the fraction of released drug, t is the releasetime, k is a constant characteristic of the drug–polymer system,and n is release exponent that characterizes the mechanism ofdrug release (Costa and Sousa Lobo, 2001). When log Qt/Q1versus log t was plotted, it gave straight lines (R2 0.9946) and theexponents n, k were calculated from the slope and the intercept ofthe regression lines, respectively. The calculated diffusionalexponent (n) and kinetic constant (k) for the NPs formulationswas found to be 0.41 & 0.8908 respectively. As the n value is50.5, one can conclude that the release of NS from NPs was byFickian diffusion (Peppas, 1985).

Safety studies

Cell viability

Results of the MTT assay are reported in Figure 6(A). Thepercentage of cell viability was compared to control cells (100%)obtained at the same time under the same experimental condi-tions. Positive control SDS treatments showed cell viability ofonly 10.74 ± 1.92% after 24 h of incubation when compared withcontrol cells (100 ± 6.04%). However, test samples containingNS-loaded NPs and blank NPs formulations both demonstratedcell viability of �75% even after 24h at concentrations as high as5mg/mL compared to negative control. Results showed that NPsexhibit cell toxicity in a dose-dependent fashion. Further it was

Figure 3. Contour plot for PLGA NPs; (A) Effect of DS:NS ratio & Pluronic F 68 concentration on EE; (B) Effect of Sonication time & Volume ratioon PS; (C) Effect of DS:DS ratio & Pluronic F 68 concentration on PS; (D) Effect of sonication time & Pluronic F 68 concentration on PS; (E) Effect ofDS:NS ratio & sonication time on PS.

Table 4. Optimal parameters for NS-loaded PLGAnanoparticles.

Parameters Optimized level

Vw1 0.35 mlCPLGA 15 mgEthyl acetate 0.5 mlCPluoronic F68 (1.5%) 2 + 5 mlDS:NS charge ratio 2.5EEa 93.23 ± 2.7%,Particle sizea 140.83 ± 2.4 nmPolydispersivity indexa 0.130 ± 0.018Zeta Potentiala �23.45 ± 3.06 mV.

aValues represent mean ± SD (n¼ 3).


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

observed that presence of NS in NPs does not affect thecytotoxicity of NPs, suggesting it is safe to use in theseformulations.

TEER analysis

Transepithelial electrical resistance (TEER) measurements wereperformed on CFBE41o- cells to determine the effect ofNS-loaded NPs on cells under liquid/liquid culture conditions.The presence of an intact cell monolayer was confirmed by steadyTEER values (450–500 � cm2) after 3 days of culturing and themonolayer was visible by light microscopy. After exposure to theNPs, TEER was measured at a defined time interval for 5 days.TEER depends on tight junctions between cells, which can bedisrupted by noxious stimuli (Kapus and Szaszi, 2006) andtherefore provides a convenient measure of barrier function(Florea et al., 2003). Incubation of monolayers with NPsat a concentration of 2 mg/mL, which has been shown to ensureat least 80% cell viability, did not alter the resistance of theepithelial layer. This was confirmed by the progressive increasein TEER value over the period of time without any significantdifference (p¼ 0.832; p40.05) between test and controlcells (Figure 6B). Although Fiegel et al. (2003) reportedchange in the barrier properties of lung epithelial cells duringexposure to PLGA NPs, this was not observed in the presentwork. Taken together, these results suggest that the NPs do not

interfere with the integrity of the airway epithelial cells in vitroand may have little acute toxicity when used in vivo to treat lunginfections.

Antibacterial efficiency

Bacteria were exposed to different concentrations of blank NPsfor 24h and optical density was measured at 600 nm (OD 600).In addition, microbial growth was ascertained by growing bacteriaon agar plates after suitable dilution. For both control and blankNPs, bacterial growth and CFU counts were found to remainclosely similar (Figure 7A), which indicates that the blank NPs donot possess any intrinsic antibacterial activity. Moreover,increasing the concentration of the blank NPs did not affect thegrowth of bacteria. This was further evaluated by growingbacterial cells on agar plates. High CFU counts were observedon the agar plate after 24 h of incubation with blank NPs,confirming that the antibacterial activity of the NS-loadedNPs was exclusively due to the NS and not materials usedto prepare NPs. As the presence of the NPs did not have anyeffect on bacterial growth, bacterial susceptibility testingwas compared using only the NS solution collected from thein-vitro release study without the presence of NPs during theassay.

The antibacterial activity of NS-loaded NPs was determined asMIC on P. aeruginosa. A MIC value was found approximately

Figure 4. (A) The results of size measurement by DLS; (B) TEM image of NS-loaded NPs. (C) Cumulative percent drug released in PBS pH 7.4;(D) Drug release profile from NS from NPs before and after stability study.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

equal to18 mg/mL. In the same experimental conditions, the MICof control free NS was 2.4 mg/mL (Figure 7B), while the completeeradication of the bacteria was observed at 44mg/mL and 10mg/mL for NS-loaded NPs and native NS, respectively. The higherMIC value for NS-loaded NPs was attributed to the progressiverelease of the drug from the NPs; i.e. the amount of drug releasedby the particles after 24 h was 25%, which should result in aconcentration of available drug similar to the MIC observed forthe NS alone. Therefore, it could be estimated that the MIC and

MBC value for the native NS and NS released from NPs weresimilar.

Further effect of the entrapment of NS in NPs onthe antibacterial efficiency of NS was evaluated by comparingthe native NS and NS released form the NPs. Bacterial cellswere exposed to the same antibiotic concentration (2.4 mg/mL)and OD600 was measured after 6, 12 and 24 h. The results(Figure 7C) indicate no difference between the bacterialeradication rates of encapsulated NS and native NS after 6,

Figure 5. (a) Differential scanning calorimetry curves of (A) PLGA, (B) NS, (C) blank NPs, (D) NS-loaded NPs; (b) Overlay of FTIR spectra of(A) PLGA, (B) NS, (C) Physical mixture, (D) NS loaded NPs.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

12 and 24 h antibiotic exposures (p¼ 0.279; p40.05).Hence, the results confirm that NS can be efficientlyencapsulated into the polymeric NPs without affecting itsantibacterial activity.

To mimic the in-vivo condition, CFBE41o- cells wereincubated with free NS or NS-loaded NPs, and antimicrobialactivity was tested every 24h for 5 days. It was observed that NPswere more effective in eradicating the bacterial growth as

Figure 7. (A) CFU count from blank NPs susceptibility testing. (B) In vitro antibacterial activity of free NS and NS-encapsulated PLGA NPsagainst P. aeruginosa. (OD¼ absorbance at 600 nm of UV-spectrophotometer measurement); (C) Effect of the encapsulation on the antibacterialactivity of NS.

Figure 6. (A) Effects of NP formulations on viability of CFBE 41 o- cells after 24 h incubation. Test samples contained 0.5 mg/mL, 1 mg/mL, 2 mg/mLand 5 mg/mL of formulation. Data represent mean ± SD, n¼ 6; (B) TEER analysis of CFBE 41 o- cells layers exposed to NPs over time.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

compared to free NS. The cations present in the airway surfacefluid (Knowles et al., 1997) antagonized the activity of the NSwhen incubated with bronchial cells and decreased its antibac-terial efficiency (Kuhn, 2001). However improved efficacy wasobserved with NPs due to the protection of the NS by NPsduring incubation with the bronchial cells. NS was released slowlyfrom NPs over days and retained its antibacterial activity againstP. aeruginosa. Entrapment in NPs significantly reduced thedecline in the antibacterial activity of NS (Figure 8), thus

entrapment in NPs could provide also better activity againstbacteria in-vivo.

Stability studies

The stability studies were carried out in accordance with the ICHguidelines to evaluate the suitability of the storage condition forthe developed formulations since it is important to preserve thecritical parameters, i.e. mean particles size and size distributionuntil they are applied. Storing freeze-dried particles duringstability studies did not change particle size (Table 5). Therewas also no change (p40.05) in PDI, zeta potential or total drugcontent during the storage period. After storage for 6 months, thecontent of NS was 98.64 ± 1.33% at 5 ± 3 �C condition(p¼ 0.178), and 98.59 ± 2.35% at 25 ± 2 �C/60 ± 5% RH(p¼ 0.575). Also, there were no significant changes (p40.05)in NS release profiles before and after the stability studies(Figure 4D). The sample after the stability study exhibited f1 andf2 values of 2.03 and 94.61 at 5 ± 3 �C versus 3.03 and 84.01 at25 ± 2 �C/60 ± 5%, respectively (Table 6), demonstrating thatstorage of samples in stress conditions did not alter the release

Figure 8. Fate of NS antibacterial activity in-vivo.

Table 5. Stability studies data of the NS loaded NPs (n¼ 3).

Parameter evaluated

Storage condition Time PS (nm) PDI Zeta potential (mV) % Drug content

Initial 265.46 ± 5.01 0.155 ± 0.01 �23.05 ± 1.49 100.005 ± 3 �C 3 M 267.05 ± 3.43 0.281 ± 0.12 �22.49 ± 1.74 99.90 ± 1.87

6 M 268.38 ± 3.68 0.233 ± 0.07 �23.48 ± 2.94 98.64 ± 1.3325 ± 2 �C/60 ± 5% RH

1 M 271.38 ± 4.03 0.312 ± 0.10 �21.04 ± 0.24 99.37 ± 0.812 M 272.05 ± 2.13 0.320 ± 0.11 �22.19 ± 0.86 97.62 ± 2.283 M 270.38 ± 6.08 0.255 ± 0.08 �23.53 ± 2.78 97.28 ± 2.496 M 271.05 ± 3.19 0.313 ± 0.10 �23.78 ± 2.34 98.59 ± 2.35

Table 6. Release profiles comparison of the NS loaded NPs at the end ofstability study.

Compared batches f1 f2 Similarity p Value

NS-loaded NPs: NS loaded NPs(Stability, 5 ± 3 �C)

2.03 94.61 Yes 0.178

NS-loaded NPs: NS loaded NPs(Stability, 25 ± 2 �C/ 60 ± 5%)

3.03 84.01 Yes 0.575

f1: Difference factor; f2: Similarity factor; PCL NPs: Reference Samples,PCL NPs (Stability): Test sample.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

behaviour of the NS from NPs. It also indicates that lyophilisationof NPs enhance stability of the NPs.

Conclusion

PLGA has a limited ability to entrap the hydrophilic drug.However, complexation of NS with helper polymer DS enhancedthe entrapment of NS in the hydrophobic core of PLGA. TheNS-loaded PLGA NPs were spherical in shape, uniform in size,and had a net negative surface charge. The entrapment efficiencyand particle size of the NPs were influenced by variousformulation and processing parameters, i.e. DS:NS charge ratio,PLGA concentration, sonication time and Vw1/V0 ratio.Implementation of fractional factorial design to the developmentof PLGA NPs increased their NS entrapment efficiency up to 93%and gave a sustained release profile. Moreover, NS-encapsulatedPLGA NPs effectively inhibited the growth of bacteria over anextended time period due to the sustained release characteristicsof NPs, while the antibacterial activity of free NS declined. Thisindicates that NPs are able to prolong the bactericidal activity ofNS in-vivo. The sustained release from NPs should make itpossible to reduce the NS dose and administration frequency,thereby reducing the side effects associated with it.

Acknowledgements

Authors would also thanks to TIFAC CORE in NDDS for laboratoryfacilities for the work.

Declaration of interest

The authors report no conflict of interest. We are grateful to financialsupport from Indian council of Medical Research (ICMR), New Delhi,India and Department of Foreign Affairs and International Trade (DFAIT)Canadian Commonwealth Scholarship Program (CCSP) by the CanadianBureau of International Education (CBIE).

References

Abdelghany SM, Quinn DJ, Ingram RJ, Gilmore BF, Donnelly RF,Taggart CC, Scott CJ. Gentamicin-loaded nanoparticles showimproved antimicrobial effects towards Pseudomonas aeruginosainfection. Int J Nanomed, 2012;7:4053–63.

Asadi A. Streptomycin-loaded PLGA-alginate nanoparticles: Preparation,characterization, and assessment. Appl Nanosci, 2013;1–6.

Barichello JM, Morishita M, Takayama K, Nagai T. Encapsulation ofhydrophilic and lipophilic drugs in PLGA nanoparticles by thenanoprecipitation method. Drug Dev Ind Pharm, 1999;25:471–6.

Bilati U, Allemann E, Doelker E. Poly(D,L-lactide-co-glycolide) protein-loaded nanoparticles prepared by the double emulsion method –Processing and formulation issues for enhanced entrapment efficiency.J Microencapsul, 2005;22:205–14.

Blanco D, Alonso MJ. Protein encapsulation and release from poly(lac-tide-co-glycolide) microspheres: Effect of the protein and polymerproperties and of the co-encapsulation of surfactants. Eur J PharmBiopharm, 1998;45:285–94.

Calvo P, Alonso MJ, Vila-Jato JL, Robinson JR. Improved ocularbioavailability of indomethacin by novel ocular drug carriers. J PharmPharmacol, 1996;48:1147–52.

Costa P, Sousa Lobo JM. Modeling and comparison of dissolutionprofiles. Eur J Pharm Sci, 2001;13:123–33.

Cun D, Foged C, Yang M, Frokjaer S, Nielsen HM. Preparation andcharacterization of poly(DL-lactide-co-glycolide) nanoparticles forsiRNA delivery. Int J Pharm, 2010;390:70–5.

Danhier F, Ansorena E, Silva JM, Coco R, Le Breton A, Preat V. PLGA-based nanoparticles: An overview of biomedical applications. J ControlRelease, 2012;161:505–22.

Feng S, Huang G. Effects of emulsifiers on the controlled release ofpaclitaxel (Taxol) from nanospheres of biodegradable polymers.J Control Release, 2001;71:53–69.

Fiegel J, Ehrhardt C, Schaefer UF, Lehr CM, Hanes J. Large porousparticle impingement on lung epithelial cell monolayers–toward

improved particle characterization in the lung. Pharm Res, 2003;20:788–96.

Florea BI, Cassara ML, Junginger HE, Borchard G. Drug transport andmetabolism characteristics of the human airway epithelial cell lineCalu-3. J Control Release, 2003;87:131–8.

Forge A, Schacht J. Aminoglycoside antibiotics. Audiol Neurootol, 2000;5:3–22.

Gao H, Jones M-C, Chen J, Liang Y, Prud’homme RE, Leroux J-C.Hydrophilic nanoreservoirs embedded into polymeric micro/nanopar-ticles: An approach to compatibilize polar molecules with hydrophobicmatrixes. Chem Mater, 2008;20:4191–3.

Gao P, Nie X, Zou M, Shi Y, Cheng G. Recent advances in materials forextended-release antibiotic delivery system. J Antibiot (Tokyo), 2011;64:625–34.

Gupta H, Aqil M, Khar RK, Ali A, Bhatnagar A, Mittal G. Sparfloxacin-loaded PLGA nanoparticles for sustained ocular drug delivery.Nanomedicine, 2010;6:324–33.

Honeybourne D. Antibiotic penetration into lung tissues. Thorax, 1994;49:104–6.

ICH Harmonised Tripartite Guideline. Stability testing of new drugsubstances and products. Q1A (R2); February 2003.

Ito F, Fujimori H, Makino K. Factors affecting the loading efficiency ofwater-soluble drugs in PLGA microspheres. Colloids Surf BBiointerfaces, 2008;61:25–9.

Jeffery H, Davis SS, O’hagan DT. The preparation and characterization ofpoly(lactide-co-glycolide) microparticles. II. The entrapment of amodel protein using a (water-in-oil)-in-water emulsion solventevaporation technique. Pharm Res, 1993;10:362–8.

Kapus A, Szaszi K. Coupling between apical and paracellular transportprocesses. Biochem Cell Biol, 2006;84:870–80.

Karlowsky JA, Zhanel GG, Davidson RJ, Hoban DJ. Postantibiotic effectin Pseudomonas aeruginosa following single and multiple aminoglyco-side exposures in vitro. J Antimicrob Chemother, 1994;33:937–47.

Knowles MR, Robinson JM, Wood RE, Pue CA, Mentz WM, Wager GC,Gatzy JT, Boucher RC. Ion composition of airway surface liquid ofpatients with cystic fibrosis as compared with normal and disease-control subjects. J Clin Invest, 1997;100:2588–95.

Kuhn RJ. Formulation of aerosolized therapeutics. Chest, 2001;120:94S–8S.

Kumari A, Yadav SK, Yadav SC. Biodegradable polymeric nanoparticlesbased drug delivery systems. Colloids Surf B Biointerfaces, 2010;75:1–18.

Lalani J, Rathi M, Lalan M, Misra A. Protein functionalized tramadol-loaded PLGA nanoparticles: Preparation, optimization, stability andpharmacodynamic studies. Drug Dev Ind Pharm, 2013;39:854–64.

Lecaroz MC, Blanco-Prieto MJ, Campanero MA, Salman H, Gamazo C.Poly(D,L-lactide-coglycolide) particles containing gentamicin:Pharmacokinetics and pharmacodynamics in Brucella melitensis-infected mice. Antimicrob Agents Chemother, 2007;51:1185–90.

Li Y, Taulier N, Rauth AM, Wu XY. Screening of lipid carriers andcharacterization of drug-polymer-lipid interactions for the rationaldesign of polymer-lipid hybrid nanoparticles (PLN). Pharm Res, 2006;23:1877–87.

Li Y, Wong HL, Shuhendler AJ, Rauth AM, Wu XY. Molecularinteractions, internal structure and drug release kinetics of rationallydeveloped polymer-lipid hybrid nanoparticles. J Control Release, 2008;128:60–70.

Lu E, Franzblau S, Onyuksel H, Popescu C. Preparation of aminoglyco-side-loaded chitosan nanoparticles using dextran sulphate as acounterion. J Microencapsul, 2009;26:346–54.

Mundargi RC, Babu VR, Rangaswamy V, Patel P, Aminabhavi TM.Nano/micro technologies for delivering macromolecular therapeuticsusing poly(D,L-lactide-co-glycolide) and its derivatives. J ControlRelease, 2008;125:193–209.

Musumeci T, Ventura CA, Giannone I, Ruozi B, Montenegro L,Pignatello R, Puglisi G. PLA/PLGA nanoparticles for sustained releaseof docetaxel. Int J Pharm, 2006;325:172–9.

Okada H. One- and three-month release injectable microspheres of theLH-RH superagonist leuprorelin acetate. Adv Drug Deliv Rev, 1997;28:43–70.

Patel B, Gupta V, Ahsan F. PEG-PLGA based large porous particles forpulmonary delivery of a highly soluble drug, low molecular weightheparin. J Control Release, 2012a;162:310–20.

Patel D, Naik S, Misra A. Improved transnasal transport and brain uptakeof tizanidine HCl-loaded thiolated chitosan nanoparticles for allevi-ation of pain. J Pharm Sci, 2012b;101:690–706.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

Peppas NA. Analysis of Fickian and non-Fickian drug release frompolymers. Pharm Acta Helv, 1985;60:110–11.

Prayle A, Smyth AR. Aminoglycoside use in cystic fibrosis: Therapeuticstrategies and toxicity. Curr Opin Pulm Med, 2010;16:604–10.

Ratjen F, Brockhaus F, Angyalosi G. Aminoglycoside therapy againstPseudomonas aeruginosa in cystic fibrosis: A review. J Cyst Fibros,2009;8:361–9.

Sanders NN, De Smedt SC, Van Rompaey E, Simoens P,De Baets F, Demeester J. Cystic fibrosis sputum: A barrier to thetransport of nanospheres. Am J Respir Crit Care Med, 2000;162:1905–11.

Saxena V, Sadoqi M, Shao J. Indocyanine green-loaded biodegradablenanoparticles: Preparation, physicochemical characterization and invitro release. Int J Pharm, 2004;278:293–301.

Song X, Zhao Y, Hou S, Xu F, Zhao R, He J, Cai Z, Li Y, Chen Q.Dual agents loaded PLGA nanoparticles: Systematic study of particlesize and drug entrapment efficiency. Eur J Pharm Biopharm, 2008a;69:445–53.

Song X, Zhao Y, Wu W, Bi Y, Cai Z, Chen Q, Li Y, Hou S. PLGAnanoparticles simultaneously loaded with vincristine sulfate andverapamil hydrochloride: Systematic study of particle size and drugentrapment efficiency. Int J Pharm, 2008b;350:320–9.

Soppimath KS, Aminabhavi TM, Kulkarni AR, Rudzinski WE.Biodegradable polymeric nanoparticles as drug delivery devices.J Control Release, 2001;70:1–20.

Suk JS, Lai SK, Wang YY, Ensign LM, Zeitlin PL, Boyle MP, Hanes J.The penetration of fresh undiluted sputum expectorated by cysticfibrosis patients by non-adhesive polymer nanoparticles. Biomaterials,2009;30:2591–7.

Tewes F, Munnier E, Antoon B, Ngaboni Okassa L, Cohen-Jonathan S,Marchais H, Douziech-Eyrolles L, Souce M, Dubois P, Chourpa I.Comparative study of doxorubicin-loaded poly(lactide-co-glycolide)nanoparticles prepared by single and double emulsion methods. Eur JPharm Biopharm, 2007;66:488–92.

Ungaro F, D’angelo I, Coletta C, D’emmanuele Di Villa Bianca R,Sorrentino R, Perfetto B, Tufano MA, Miro A, La Rotonda MI, QuagliaF. Dry powders based on PLGA nanoparticles for pulmonary deliveryof antibiotics: Modulation of encapsulation efficiency, release rate andlung deposition pattern by hydrophilic polymers. J Control Release,2012;157:149–59.

Wang Y, Wang R, Lu X, Lu W, Zhang C, Liang W. Pegylatedphospholipids-based self-assembly with water-soluble drugs. PharmRes, 2010;27:361–70.

Wikler MA. Methods for dilution antimicrobial susceptibility tests forbacteria that grow aerobically; Approved Standard. 8th ed. Wayne, PA:CLSI; 2008.

Wong HL, Bendayan R, Rauth AM, Wu XY. Simultaneous delivery ofdoxorubicin and GG918 (Elacridar) by new polymer-lipid hybridnanoparticles (PLN) for enhanced treatment of multidrug-resistantbreast cancer. J Control Release, 2006;116:275–84.

Xiang-Gen W, Li-Na Y, Meng X, Hao-Ran J. Anti-infectious activity ofintravitreal injectable voriconazole microspheres on experimentalrabbit fungal endophthalmitis caused by Aspergillus fumigatus.J Pharm Sci, 2011;100:1745–59.

Yoncheva K, Vandervoort J, Ludwig A. Influence of process parametersof high-pressure emulsification method on the properties of pilocar-pine-loaded nanoparticles. J Microencapsul, 2003;20:449–58.


Jour

nal o

f M

icro

enca

psul

atio

n D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y B

rock

Uni

vers

ity o

n 10

/01/

14Fo

r pe

rson

al u

se o

nly.

Polymer assisted entrapment of netilmicin in PLGA nanoparticles for sustained antibacterial activity

Documents

Transcript of Polymer assisted entrapment of netilmicin in PLGA nanoparticles for sustained antibacterial activity