M

RD

a

ARRAA

KMPNO5S

1

Srcsgocisn

gtc

nn

j

0d

Colloids and Surfaces B: Biointerfaces 86 (2011) 28–34

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

journa l homepage: www.e lsev ier .com/ locate /co lsur fb

odified PLA nano in situ gel: A potential ophthalmic drug delivery system

amesh C. Nagarwal, Rakesh Kumar, M. Dhanawat, J.K. Pandit ∗

epartment of Pharmaceutics, IT, Banaras Hindu University, Varanasi, UP 221005, India

r t i c l e i n f o

rticle history:eceived 4 February 2011eceived in revised form 5 March 2011ccepted 14 March 2011vailable online 23 March 2011

eywords:odified nano in situ system

oly lactic acidanoparticles

a b s t r a c t

A novel nano in situ gel forming system of 5-Fluorouracil (5-FU) was investigated for its potential usefor conjunctival/corneal squamous cell carcinoma (CCSC). The study was conducted in two steps, in thefirst step PLA nanoparticles were prepared and characterized; in the second step the drug loaded PLAnanoparticles were dispersed in sodium alginate solution yielding the modified nano in situ system,which were evaluated in rabbit eye. Size and morphology of prepared PLA particles were verified byusing dynamic light scattering (DLS), atomic force microscope (AFM) and scanning electron microscope(SEM). In vitro and in vivo study of free 5-FU, PLA nanoparticles and modified nano in situ system wereconducted in simulated tear fluid and in rabbit eye respectively. PLA nanoparticles were in size range of128–194 nm with spherical shape and smooth surface with narrow size distribution. No polymer drug

phthalmic delivery-fluorouracilodium alginate

interaction was found as confirmed by FTIR, NMR and DSC. XRD of PLA nanoparticles confirmed that 5-FUwas present in the crystalline state. In vitro experiments indicated a diffusion controlled release of 5-FUfrom both PLA nanoparticles and modified nano in situ system with high burst effect. Modified nanoin situ gel system (MNS) significantly increased the Cmax and AUC0-8 in aqueous humor as comparedto 5-FU solution and PLA nanoparticles. Higher 5-FU level in aqueous humor was possibly because ofincreased retention time of gel matrix-embedded drug loaded nanoparticles. Overall results showed the

halm

potential of MNS for opht

. Introduction

5-FU is a pyrimidine analogue and acts by interacting withphase cells (those actively synthesizing DNA) that has been

eported for the treatment of conjunctival/corneal squamous cellarcinoma (CCSC). It has limited side effects on the normal ocularurface epithelium [1]. Being an inexpensive drug and possessing aood stability in aqueous solution, 5-FU is a superior choice forcular chemotherapy. Due to low ocular bioavailability of topi-ally applied drugs, 1% (w/v) 5-FU solution is often used, whichs generally prepared extemporaneously. In case of topical solutionystemic side effects are more frequent due to absorption throughasolachrymal duct [1,2].

Drug loaded nanoparticles (DNPs) offer the advantage of tar-eting ocular tissues. It has potential to deliver the drug tohe anterior segment of eye with enhanced bioavailability. DNPsould also be employed to achieve multiple purposes, includ-

Abbreviations: NP, nanoparticles; DNPs, drug loaded nanoparticles; Modifiedano in situ gel system (MNS), drug loaded nanoparticles dispersed in sodium algi-ate solution.∗ Corresponding author. Tel.: +91 9935640101.

E-mail addresses: ramesh.nagarwal@gmail.com, dr.jkpandit@gmail.com,kpandit@bhu.ac.in (J.K. Pandit).

927-7765/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfb.2011.03.023

ic delivery in the therapy of CCSC.© 2011 Elsevier B.V. All rights reserved.

ing enhanced cellular uptake of poorly permeable drugs, reducedcellular and tissue clearance, and sustained drug delivery. DNPsof various sizes based on polymers and biomaterials such aspoly(lactide-co-glycolide; PLGA), poly(lactic acid; PLA), poly(�-caprolactone), albumin and chitosan have been developed asparticulate systems in drug delivery [3]. Microspheres and lipo-somal delivery systems of 5-FU have also been investigated toprolong ocular levels on topical administration. Giannavola etal. [4] have reported that acyclovir loaded nanoparticles showeda sustained acyclovir release and were well tolerated by therabbit’s eye that confirms the potential of PLA nanoparticu-late system for sustained ocular delivery with better ocularbioavailability.

In recent years, extensive investigations have been dedicated tothe development of newer systems of ocular drug delivery to attainmedications with prolonged retention time on the eye surface, min-imized dose frequency and improved transcorneal penetration ofnewly emerging drugs. Several attempts have been made to deliverophthalmic drugs to the eye by using polymeric vehicles such assodium alginate, chitosan, and gellan [5,6]. Drug delivery systems

based on the concept of in situ gel formation polymers that exhibitphase transition due to physiochemical changes in their environ-ment can be instilled as liquid drops into the cul-de-sac of the eyewhere they transform into a gel or semisolid phase. Thus, an in situgel forming delivery system has the ease of administration similar

R.C. Nagarwal et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 28–34 29

of DNP

tt

ebdsfisccdrtcXov

2

2

I(pnAD

2

2

oPawfpttbsa

2

wa

Fig. 1. Diagrammatic representation of fabrication

o an ophthalmic solution with extended retention time because ofhe gel formulation [7].

Apart from the above-mentioned system, a MNS can also benvisaged (Fig. 1). Application of nanoparticles and in situ gel haseen established in ocular delivery. Fabricated nanoparticles can beispersed in an in situ gelling vehicle: this combination of DNPs-initu gelling vehicle, at least theoretically, is likely to provide bene-ts of these components in modulating drug release at the ocularurface. Studies are required to examine the possibility of this con-eptual NPs-loaded gel matrix [3]. To examine the veracity of thisoncept we formulated a 5-FU loaded PLA nanoparticle (DNPs) andispersed in an in situ vehicle system capable of sustaining 5-FUelease that may be termed as a MNS. The particle size, its distribu-ion and surface morphology were characterized. Physiochemicalharacterization of nanoparticles was performed by FTIR, DSC andRD. Encapsulation efficiency was analyzed by using HPLC meth-ds. Finally, the potential of this MNS was tested for in vitro and inivo study in rabbit eye.

. Materials and methods

.1. Materials

5-Fluorouracil (5-FU) was generously gifted by Dabur Pharmandia Pvt. New Delhi, India. Poly(lactic acid) Mol.wt. 1,48,000PLA) was obtained from Huvis, Korea. Poly vinyl alcohol wasurchased from HiMedia Lab. Pvt. Ltd., Mumbai, India. Sodium algi-ate (medium viscosity) was obtained from Sigma–Aldrich, USA.ll other chemicals and solvents were of analytical grade (Merck,armstadt, Germany).

.2. Methods

.2.1. Preparation of MNSBriefly, first PLA-DNPs were prepared by dissolving 100–200 mg

f PLA in 20 ml of acetone and pouring it into 40 ml of aqueousVA (2.5%, w/v) containing a 5-FU (0.1%, w/v) at room temper-ture under magnetic stirring for 15 min. The milky suspensionas ultrasonicated (UP50H, Hielscher Ultrasonics Gmbh, Germany)

or 5 min at 60 magnitudes and 2 cycles for better colloidal dis-ersion. The organic solvent was removed on a Rotavapor (Buchiype). Afterwards, the DNPs were separated by cooling centrifuga-ion (Beckman, USA), at 18,000 rpm for 50 min. MNS was preparedy suspending the settled mass of PLA-DNPs in 1% sodium alginateolution containing 5-FU (0.1%, w/v) and stored at 4 ◦C for furthernalysis.

.2.2. Nanoparticles size and morphologyThe size (Z-average) and zeta potential (mV) of the nanoparticles

ere analyzed by dynamic light scattering (DLS), in triplicate usingZetasizer ZS (Malvern Instruments, UK). The nanosuspension of 5-

s dispersed in situ gel and DNPs loaded gel matrix.

FU was prepared with triple distilled water for size measurement.Scanning electron microscopy (SEM) was performed using a FEIQuanta 200 ESEM FEG operating between 5 and 20 kV. Liquid sam-ples were deposited on a thin aluminum plate (1 cm × 1 cm) anddried at room temperature and directly placed on the stub with-out staining and focused at different magnifications. Atomic ForceMicroscopy (NanoscopyE Digital Instruments, USA) was used forthe surface morphology and three dimensional organization and/orassociation of the nanoparticles. A drop of nanosuspension wasdeposited on a glass cover slip and was dried at room temperature.Contact mode was used with the tip mounted on a 100-�m-long,single-beam cantilever with resonant frequency in the range of240–255 kHz and corresponding spring constant of 11.5 N/m.

2.2.3. Physiochemical characterizationDNPs and blank NPs were separated by centrifugation from

the supernatant and freeze-dried. FTIR spectra were obtainedusing a KBr pellet in FTIR spectrophotometer (Shimadzu-8400S,Japan). Percentage transmittance (%T) was recorded in the spec-tral region of 500–4500 cm−1 using a resolution of 4 cm−1 and 40scans.

The 1H NMR spectra were recorded on JEOL AL300 (300 MHz)spectrometer. Chemical shift are reported in parts per million unitsrelative to Tetramethylsilane (TMS) used as internal standard. Cou-pling constants (J) are reported in Hertz (Hz). Dimethylesulfoxide(DMSO) and D2O were used for NPs, DNPs and 5-FU, respectively.

The XRD measurements were carried out using Bruker D8Advance X-ray diffractometer. The X-rays were produced using asealed tube and the wavelength of X-ray was 0.154 nm (Cu K-alpha).The X-rays were detected using a fast counting detector based onSilicon strip technology (Bruker LynxEye detector).

2.2.4. In vitro gelation studies and interaction study with mucinThe gelation studies of sodium alginate (SA) solution and PLA

nanoparticles dispersed in situ system were evaluated in simulatedtear fluid (STF), sodium chloride 0.67 g, sodium bicarbonate 0.20 g,calcium chloride dehydrate 0.008 g in distilled water q.s. to 100 mlusing in house fabricated Teflon gelation cell. The in vitro inter-action of SA and MNS with mucin (porcine stomach type II fromSigma Aldrich) was measured by viscosity change by Brookfieldviscometer using spindle 63 at 37 ◦C [5].

2.2.5. In vitro release studiesIn vitro release studies of DNPs were performed by dialysis tub-

ing membrane (Himedia Ltd., India) with a molecular weight cut-offof 12,000–14,000. The membrane opening was tied to the mouth of

a PVC test tube (1 cm diameter) and dipped in a 100 ml beaker con-taining STF (pH 7.4, 50 ml). The entire system was placed in beaker(250 ml) containing distilled water maintained at 37 ± 0.5 ◦C. Asmall magnetic bead (Sigma–Aldrich, USA) was placed in the beakerand was stirred at 100 rpm on a magnetic stirrer (Remi India Ltd.).

30 R.C. Nagarwal et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 28–34

Table 1Formulation composition and characterization of PLA nanoparticles and MNS.

Batch PLA (mg) 5-FU:PLA SA (%, w/v) Dynamic light scattering (DLS) particle size (nm) PDI %EE In vitro gelation

PLA 100 100 1:2 – 128 ± 3.5 0.015 19.89 ± 3.6 –PLA 150 150 1:3 – 145 ± 4.6 0.112 30.25 ± 4.2 –PLA 200 200 1:4 – 194 ± 3.5 0.136 48.35 ± 4.7 –MNS1 200 1:4 0.1 – – – +

S

At

2

ZtVodIotolt2tL(wwctibbtot

5pTTsa

2

atm

3

3

taf5oi

Fig. 2a and b shows the 2D and 3D AFM images of nanosus-pension of representative high molecular weight PLA with itshistogram (Fig. 2c). Spherical nature of the nanoparticles is evidentand interestingly the size of nanoparticle was similar to as observed

MNS2 200 1:4 0.5 –MNS3 200 1:4 1 –

low gelation (+), immediate gelation (++), Immediate stiff gelation (+++).

t predetermined times, 500 �l of the medium was removed andhe amount of 5-FU was analyzed by HPLC.

.2.6. In vivo studyIn vivo experiments were performed on groups of three New

ealand albino rabbits of either sex (Central Animal House, Insti-ute of Medical Sciences (IMS), Banaras Hindu University (BHU),aranasi, India) weighing from 2.1 to 2.4 Kg, free of any signsf inflammation or gross abnormality. All experiments were con-ucted with the permission of Central Animal Ethical Committee,

MS, BHU, Varanasi. The protocol of drug administration consistedf 4 instillations of 50 �l of 0.1% 5-FU solution, and PLA naopar-icles and MNS (5-FU equivalent to 1 mg ml−1) in the cul de sacf right eye in conscious rabbits. Normal saline was instilled ineft eye as control. After the last instillation, rabbits were main-ained in an upright position using restraining boxes. After 0.5, 1,, 4, 6, and 8 h 50 �l aqueous humor withdrawn after being anaes-hetized by 20 mg/kg of ketamine hydrochloride (Aneket®, Neonaboratories Ltd., Mumbai, India) and one drop of local anaesthetic0.5% proparacaine hydrochlodide). Aqueous humor was collectedith 26 gauge needle attached to a tuberculin syringe. Each eyeas examined after taking samples for any damage to iris, lens and

ornea using a slit lamp. Zinc sulphate (2%, w/v) solution was addedo the samples to precipitate the protein and separated by cool-ng centrifuge at 15,000 rpm for 15 min. Supernatant were filteredy 0.22 �m membrane filter and analyzed by HPLC. Tolerability oflank and drug loaded nanoparticles was tested by modified Draizeest using rabbit model and the congestion, swelling and dischargef the conjunctiva were graded on a scale from 0 to 3, 0 to 4, and 0o 3, respectively. Corneal integrity was checked by slit lamp [8].

High performance liquid chromatography was used to assayFU, on a Model CE4201, Cecil U.K. HPLC, a UV detector and a reversehase column (Hamilton, HxSil C18 5 �m 150 mm × 4.6 mm, USA).he mobile phase was 90% (v/v) water and 10% (v/v) of methanol.he flow rate was 1 ml/min. A 20 �l sample was injected by a microyringe (Hamilton, USA) and the eluent was monitored for thebsorbance of 5-FU by UV detector at 266 nm [9].

.2.7. Data analysisThe differences in average of data were compared by simple

nalysis of variance (one-way ANOVA) or independent sample-test (Origin 6.1 USA). The significance of the difference was deter-

ined at 95% confidence limit (˛ = 0.05).

. Results

.1. Size and morphology of nanoparticles

The fabricated 5-FU loaded PLA nanoparticles were obtained inhe size range of 128–194 nm (Table 1). Ratio of non-aqueous and

queous phase (20:40) was optimized for all the batches. At dif-erent amounts of PLA (100, 150 and 200 mg) and 50 mg (w/v) of-FU, nanoparticles of average size of 128, 145 and 194 nm werebtained respectively. The average particle size increased withncreasing PLA concentration. The particle size distribution also

– – ++– – +++

becomes wider at higher concentration of PLA. Polydispersity index(PDI) ranges from 0.015–0.136 confirming narrow distribution ofparticles. Encapsulation efficiency (%EE) increased with increase indrug and polymer ratio.

Fig. 2. (a, b) AFM 2D and 3D images of PLA nanoparticles and (c) histogram of PLAnanoparticles.

R.C. Nagarwal et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 28–34 31

Fi

iStea

3

fia5ts(nieabi

Nitsbcpgc

P

Fig. 4. SEM micrographs of (a) PLA 200 nanoparticles and (b) MNS3.

ig. 3. (a) AFM 2D, (b) 3D images of MNS3. The height scale has been given in 3Dmages to understand the depth of irregularities.

n DLS studies. Fig. 3a and b shows 2D and 3D AFM images of MNS3.EM of PLA 200 showed spherical and smooth surface of nanopar-icles (Fig. 4a). Fig. 4b clearly demonstrated the PLA nanoparticlembedded in sodium alginate solution. By SEM, image analysis theverage particle sizes was 137 nm for PLA 200.

.2. Physiochemical characterization of nanoparticles

FTIR spectra of PLA NPs, 5-FU and DNPs are presented ing. 5. The characteristic peaks of � (C–F) stretching frequencyt 1000 cm−1, � (N–H) at (3747 cm−1), and aromatic ring at00–600 cm−1 are prominent in 5-FU [10]. Similarly, the charac-eristic PLA peaks of � (C O) stretching at 1760 cm−1, � (C–H)tretching at 2900–3000 cm−1, � (C–C) stretching at 870 cm−1, �C–H) deformation at 1350–1500 cm−1 are also present in PLAanoparticles [11]. The presence of all the characteristic peaks of

ndividual PLA and drug in DNPs indicate the presence of drugmbedded in PLA nanoparticles. Interestingly, a small new peakt (1723 cm−1) has appeared in DNPs presumably due to hydrogenonded >N–H peak with >C O group of PLA nanoparticles–drug

nteractions.Existence of 5-FU in PLA nanoparticles is also confirmed by 1H

MR spectroscopy (Fig. 6). The peak at ı 7.489 ppm is a character-stic signal of the vinylic proton of 5-FU whose presence confirmshat 5-FU is embedded in PLA nanoparticles [12]. The peak has beenhifted to high field region (ı 7.722) indicating stronger hydrogenonded interaction of >N–H group with PLA nanoparticles (espe-ially with >C O group). Other NMR peaks assigned for methyl

roton of PLA remain unchanged in DNPs. Both FTIR and NMR sug-est strong interactions between drug (5-FU) and PLA nanoparticlesausing greater drug loading in DNPs

Fig. 7 shows the XRD spectra of 5-FU, PLA, and 5-FU-loadedLA nanoparticles. XRD patterns showed 5-FU crystalline peak at

Fig. 5. FTIR spectra of 5-FU (lower), PLA blank nanoparticle (middle) and drug loadedPLA nanoparticles (DNPs) (upper curve). Y-axis has been shifted for better clarity ofpresentation.

2� (28◦). Similarly, peak was also detected in 5-FU-loaded PLAnanoparticles that confirm the presence of drug in the nanopar-ticles as well as its crystalline nature.

3.3. Interaction of MNS with mucin

The interactions of SA solution and MNS with mucin (0.25%,w/v) were studied by measuring the change in viscosity at differ-ent rpm (Fig. 8). When SA solution was added to mucin, a slight

32 R.C. Nagarwal et al. / Colloids and Surfaces B: Biointerfaces 86 (2011) 28–34

Fig. 6. 1H NMR of pure 5-FU, blank PLA nanoparticles (NPs) and drug loaded PLAnanoparticles (DNPs).

cviN

3

irthwr

3

t(at52

Fig. 8. (A) Effect of different mediums on viscosity of MNS, (B) viscosity change dueto Mucin (M) interaction with SA and MNS3.

Fig. 9. In vitro release profile of 5-FU solution, PLA DNPs and MNS3.

Table 2Pharmacokinetic parameters of 5-FU in aqueous humor after topical application of5-FU solution, PLA nanoparticles and MNS3 in rabbit eye.

Pharmacokinetic parameters 5-FU PLA-DNPs MNS3

AUC0–8 (�g ml−1 h) 23.18 79.77** 150.87**

Cmax (�g ml−1) 6.14 16.41** 28.5**

Tmax (h) 1 2 2Kel (h−1) 0.245 0.24 0.160

Fig. 7. XRD spectra of 5-FU, PLA, PLA NPs and PLA DNPs.

hange in viscosity was observed. In case of MNS, initially a higheriscosity (at 10 rpm) was obtained that gradually decreased withncreased sheer rate revealing the sheer thinning behavior of MNS.o interaction was found in between mucin and MNS.

.4. In vitro release study

The in vitro release of PLA nanoparticles and MNS were stud-ed in simulated tear fluid, pH 7.4 (Fig. 9 Cumulative percentageelease vs. time). 5-FU from the PLA-DNPs was released in a sus-ained manner over a period of 8 h with low burst effect (19% in firstour). MNS showed high burst release during the first hour (47%)hile a sustained release was maintained further with additional

elease of drug from 5-FU loaded PLA nanoparticles for 8 h.

.5. In vivo study

The pharmacokinetic parameters like peak plasma concentra-ion (Cmax), time of peak concentration (Tmax), area under curve

AUC0–8), elimination rate constant (Kel), biological half-life (t1/2)nd mean residence time (MRT) of 5-FU after ocular adminis-ration into rabbit eye were compared for MNS, PLA-DNPs and-FU solution (Table 2 and Fig. 10). The observed Cmax value was8.5 �g ml−1 for modified system compared to 16.41 �g ml−1 for

t1/2 (h) 2.82 2.9 4.44MRT (h) 4.54 5.22 7.08

** Statistically significant (p<0.01) analyzed by one way ANOVA by Dunnett Mul-tiple Comparisons Test.

R.C. Nagarwal et al. / Colloids and Surface

FP

PAtTtMvP

4

toPetawc(wdopnwd

isiPtcntOa

trfi

tion time due to mucoadhesion) when compared with PLA-DNPs

ig. 10. 5-FU level in the aqueous humor after instillation of the free 5-FU solution,LA.

LA-DNPs and 6.14 �g ml−1 for 5-FU solution. With respect toUC0–8, MNS exhibited higher value (150.87 �g ml−1 h) compared

o that of PLA-DNPs (79.77 �g ml−1 h) and 5-FU (23.18 �g ml−1 h).he difference between the groups, with regard to these parame-ers, was significant at p ≤ 0.01. The Tmax has similar values for both

NS and PLA-DNPs (2 h) but higher than that of 5-FU (1 h). The MRTalues were clearly extended to 7.08 h for MNS in comparison toLA-DNPs (5.22 h) and 5-FU (4.54 h).

. Discussion

MNS was prepared by a two-step procedure. In the first stephe drug loaded PLA nanoparticles was prepared and in the sec-nd step, it was dispersed in sodium alginate solution. PreparedLA nanoparticles were in nano size range with lower PDI (<0.5)xhibiting a homogenous and stable nanosuspension. Larger par-icle sizes produced at higher concentrations of PLA, which ispparently due to agglomeration of particles. An enhanced %EEas observed with decreased drug:polymer ratio (1/2–1/4) that

ould be attributed to the increased viscosity of the solutionat higher polymer concentration), delaying the drug diffusionithin the polymer droplets [13]. It is interesting to note thatrug loaded nanoparticles were discrete while in the absencef any drug the particles were agglomerated. This suggests theresence of a high surface charge on 5-FU in the drug loadedanoparticles that keeps them separate; on the other hand, thereas clumping due to low surface charge in the absence ofrug.

In SEM images, individual particles were visible and size ranges very near to DLS data. AFM images confirmed smooth and roundhaped particles similar to SEM images. Solid phase character-zation by FTIR, XRD and DSC showed no interaction betweenLA and 5-FU. Extra peaks in drug-loaded nanoparticles similaro pure drug spectra showed the presence of drug in nanoparti-les without modification. XRD spectra supported the crystallineature of drug molecule in PLA DNPs. Similarly, the NMR spec-ra showed a separate drug peak in drug-loaded nanoparticles.verall results showed no chemical interaction between drugnd PLA.

The present study was focused on the potential of PLA nanopar-

icles in an in situ gel vehicle (sodium alginate). We have alreadyeported the potential of sodium alginate as in situ gel formulationor ocular application [3]. We attempted to explore the possibil-ty of enhancing the bioavailability as well as residence time of

s B: Biointerfaces 86 (2011) 28–34 33

formulation using a combination of PLA nanoparticles and in situgel system.

Optimization of in situ gel of sodium alginate (0.1–1%, w/v)was tested by visual inspection using gelation cell and 1% (w/v)of SA was found suitable for preparation of MNS as previouslyreported in literature for ophthalmic delivery for in situ vehicle[5,7]. Viscosity behavior of sodium alginate was analyzed in dif-ferent dissolution media. No significant difference was observedin case of distilled water and acetate buffer (pH 5.5) but whenSTF was added to sodium alginate solution, the viscosity wasincreased due to gel formation by ionic interaction; when highersheer rate was applied, the viscosity was decreased. This pseu-doplastic behavior of the formulation suggests its suitability forocular delivery. It could be further confirmed by the high shearrate of eye (4250–28,500 1/s) during blinking hence viscoelasticfluid with a viscosity that is high under low shear rate condi-tion and vice versa is often preferred [14]. The formulation shouldhave an optimum viscosity to allow sol to gel transition due toionic interactions. Also, the in situ formed gel should preserve itsintegrity without dissolving and eroding for prolonged period oftime. Similarly, the in vitro mucin interaction with SA solutionand MNS were studied. The viscosity was increased as com-pared to mucin alone. Results showed that the viscosity increasedsignificantly (p < 0.05) but it was in acceptable limit for ocularformulation.

In vitro release of the new system was tested and comparedwith PLA nanoparticles alone. A high burst release of drug wasobserved in case of MNS that can be explained by the fact that ini-tially the alginate polymer is completely hydrated when presentin the eye drop form. But, as it comes in contact with the sim-ulated lacrimal fluid and gelation occurs a prehydrated matrix isformed in which hydration and water penetration no longer limitdrug release, leading to a fast release [15]. Another possible reasonmay be the lag time of gelation of sodium alginate at ocular surface.Once the alginate has formed a gel with tear fluid containing Ca+2,further penetration of tear into the gel would be decreased. Dif-fusion of drug through gel and its surface erosion would dependlargely on the stiffness of the gel. This system showed bipha-sic release pattern. Initially the level of 5-FU was increased bythe in situ gel vehicle and then maintained by drug loaded PLAnanoparticles. Further, a gradual release shown in the next phaseis the consequence of the release of the drug fraction encapsu-lated in the core of the nanoparticles. PLA-DNPs showed slowerrelease with minimum burst effect and optimum loading capacity.Overall, the release of drug from nanoparticles was diffusion con-trolled as indicated by Higuchi model [16]. Drug release rate can becontrolled by controlling the size of the drug embedded in nanopar-ticles as well as in the in situ gelling system. In in vivo study, theMNS showed higher aqueous humor level of 5-FU as compared toPLA nanoparticles and free 5-FU solution. The AUC of this modi-fied system is almost 7 times higher than the pure drug solutionand 2 times than PLA-DNPs indicating very high ocular bioavail-ability. Furthermore, a higher half life and MRT of the modifiedsystem suggesting an enhanced retention time that could be justi-fied with the in situ gel forming property of sodium alginate and agood mucoadhesive nature of PLA. In our previous study we opti-mized PLA nanoparticles loaded with 5-FU and tested for ocularapplication which showed better tolerability in case of rabbit eye[5].

Hence, MNS fulfills the criteria for both high bioavailabilityand sustained dose maintenance for longer period (high reten-

alone. The overall results signify the potential topical applicationof MNS, which is in favour of sustaining drainage of drugs formthe conjunctiva sac of the eye, simultaneously without blinkingdifficulty.

3 urface

5

ehod

A

Nfi

R

[

[

4 R.C. Nagarwal et al. / Colloids and S

. Conclusion

The in vitro and in vivo results of MNS showed its potential for anffective ocular delivery targeting the anterior ocular segment andence could be further studied an efficient system in the therapyf CCSC. Besides, this novel approach could also be exploited forelivery of both water soluble and insoluble drugs.

cknowledgements

The authors are grateful to University Grants Commission (UGC),ew Delhi, and Banaras Hindu University, Varanasi-221005, India,

or Senior Research Fellowship to Ramesh C. Nagarwal for support-ng this research work.

eferences

[1] A. Al-Barrag, M. Al-shaer, N. Al-Matary, M. Al-hamdani, Clin. Ophthalmol. 4(2010) 801.

[[[

[[

s B: Biointerfaces 86 (2011) 28–34

[2] E. Midena, C.D. Angeli, M. Valenti, V. De Belvis, P. Boccato, Br. J. Ophthalmol. 84(2000) 268.

[3] R.C. Nagarwal, S. Kant, P.N. Singh, P. Maiti, J.K. Pandit, J. Control. Release 136(2009) 2.

[4] C. Giannavola, C. Bueolo, A. Maltese, M.A. Vandelli, G. Puglisi, V.H.L. Lee, M.Fresta, Pharm. Res. 20 (2003) 584.

[5] R.C. Nagarwal, P.N. Singh, S. Kant, P. Maiti, J.K. Pandit, J. Biomed. Nanotechnol.6 (2010) 648.

[6] R.C. Nagarwal, P.N. Singh, S. Kant, P. Maiti, J.K. Pandit, Chem. Pham. Bull. 59(2010) 272.

[7] J.K. Pandit, D. Bharathi, A. Srinatha, D.N. Ridhurkar, S. Singh, J. Pharm. Sci. 69(2007) 37.

[8] O. McDonald, J.S. Shedduck, Eye irritation, in: F.M. Marzulli, H.I. Maibach (Eds.),Advance in Modern Toxicology, vol. 4, Jone Wiley & Sons, New York, 1977.

[9] I.A. Alsarra, M. Alarifi, J. Chromatogr. B 804 (2004) 435.10] E. Bilensoy, Y. Cirpanli, M. Sen, A.L. Dogan, S. Calis, J. Incl. Phenom. Macrocycl.

Chem. (2006).11] H.C. Yang, M.H. Hon, Microchem. J. 92 (2009) 87.

12] Y.B. Luo, X.L. Wang, D.Y. Xu, Y.Z. Wang, Appl. Surf. Sci. 255 (2009) 6795.13] Y. Yeo, K. Park, Arch. Pharm. Res. 27 (2004) 1.14] C. Wu, H. Qi, W. Chen, C. Huang, C. Su, W. Li, S. Hou, YakugakuZasshi 127 (2007)

183.15] S. Cohen, E. Lobel, A. Trevgoda, Y. Peled, J. Control. Release 44 (1997) 201.16] C. Costa, C.J.M. Lobo, Eur. J. Pharm. Sci. 13 (2001) 123.