Journal of Molecular Liquids 218 (2016) 219–232

Contents lists available at ScienceDirect

Journal of Molecular Liquids

j ourna l homepage: www.e lsev ie r .com/ locate /mol l iq

Self-nanoemulsifying drug delivery system (SNEDDS) with enhancedsolubilization of nystatin for treatment of oral candidiasis: Design,optimization, in vitro and in vivo evaluation

Ahmed Alaa Kassem a,⁎, Amira Mohamed Mohsen a, Reham Samir Ahmed b, Tamer Mohamed Essam b

a Pharmaceutical Technology Department, National Research Centre, El-Buhouth Street, Dokki, Cairo 12622, Egyptb Microbiology and Immunology Department, Faculty of Pharmacy, Cairo University, Kasr El-Aini Street, Cairo 11562, Egypt

⁎ Corresponding author.E-mail address: aa.kassem@hotmail.com (A.A. Kassem

http://dx.doi.org/10.1016/j.molliq.2016.02.0810167-7322/© 2016 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 25 August 2015Accepted 16 February 2016Available online xxxx

The aim of the present study is to develop and optimize self-nanoemulsifying drug delivery systems (SNEDDSs)to improve the per-oral bioavailability of poorly soluble polyene antifungal drug, nystatin (NYS), and to evaluateits in vitro and in vivo performance. Solubility of NYS was estimated in various vehicles to select proper compo-nents combinations. Oleic acid (oil), Tween® 20 (Tw20) and Tween®40 (Tw40) (surfactants) aswell as dimeth-yl sulfoxide (DMSO) and propylene glycol (PG) (co-surfactants) were employed to construct pseudo-ternaryphase diagrams. Thermodynamic stability, dispersibility and robustness to dilution tests were performed to op-timize formulations fromphase diagram. Five optimized formulations composed of oleic acid, Tw20 andDMSOorPG at Smix ratios (1:1, 2:1 or 3:1)were selected. Theywere spherical in shape ofmeandroplet size b100 nmwithnegatively charged zeta potential b−15mV. The in vitro release profile of NYS-SNEDDSswas found significant incomparison to the plain NYS suspension. In vitro and in vivo evaluations against Candida albicans depictedpromoted antifungal efficacy of selected NYS-SNEDDS formulations compared to marketed and plain NYSsuspensions. The results indicate that NYS loaded SNEDDS, with enhanced solubilization and nanosizing, haspotential to improve the absorption of drug and increase its oral antifungal efficacy.

© 2016 Elsevier B.V. All rights reserved.

Keywords:NystatinSelf-nanoemulsifying drug delivery system(SNEDDS)Poorly water soluble drugPseudo-ternary phase diagramImmunosuppressed mouse-modelOral candidiasis

1. Introduction

Gastrointestinal candidiasis (GIC) is one of themost commonmuco-sal infections which may become pathogenic in patients with humanimmunodeficiency virus (HIV) or acquired immune deficiencysyndrome (AIDS) [1,2]. Oropharyngeal candidiasis (OPC) is a commonhuman fungal infection characterized by an overgrowth of Candidaspecies in the superficial epithelium of the oral mucosa [3,4]. The vastmajority of these infections are caused by Candida albicans [5]. Immuno-compromised patients, including those with HIV infection or cancer,are at an enhanced risk of OPC [4,6]. In addition, OPC can be triggeredin healthy patients by transient risk factors such as antibiotic, corticoste-roid treatment or dental prosthesis [7,8]. The infection can be com-plicated by esophageal candidiasis and, in the worst cases, fungalsepticemia [3]. OPC is not a lethal disease but must be treated toavoid chronicity, other tissue invasions or systemic infection [9].Although it is infrequent, disseminated candidiasis has a mortalityrate of 47% [6,10].

Nystatin (NYS) and fluconazole are the most widely employed anti-fungal agents in the treatment of OPC [5]. NYS is a polyene antifungalantibiotic, one of the oldest antifungal drugs, produced by Streptomyces

).

noursei strains [8,11], commonly used for the prophylaxis and treatmentof candidiasis. It acts by interferingwith the fungal cellmembrane of theantibiotic-sensitive organism by binding to sterols, chiefly ergosterol,and the formation of barrel-like membrane spanning channels [12].NYS possesses a broad antifungal spectrum; it has been reported to beeffective against azole-resistant strains of Candida and, in some cases,amphotericin B-resistant strains of C. albicans [13]. NYS is a yellow orslightly brownish hygroscopic substance [1,14], practically insoluble inwater and alcohol, slightly soluble in methanol, and freely soluble in di-methyl formamide and dimethyl sulfoxide [14]. NYS is poorly absorbedfrom the gastrointestinal tract, and detectable blood concentrations arenot obtained after usual doses [15]. Following oral administration, NYSis excreted almost entirely in feces as an unchanged drug [15]. It is notabsorbed through the skin or mucous membranes when applied topi-cally [15,16]. The chemical structure of NYS reveals formulation chal-lenges, because it is characterized by the presence of a large lactonering containing several double bonds conferring an amphiphilic andamphoteric nature [17,18], which in turn contributes to its low solubil-ity in aqueous media and poor bioavailability. Special care needs to betaken for the delivery of this drug as it cannot be simply introducedinto an aqueous solution because it forms aggregates that lead to formu-lation challenges [19]. NYS aggregates formed in aqueous mediaare non-selective and able to disrupt the integrity of both fungal andmammalian cell membranes, inducing toxicity and host cell death

220 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

[20]. Therefore, it is proposed that NYS delivery in an un-aggregatedform may improve its therapeutic index [19]. For this reason, thecontrolled-release delivery of this drug is a complex task. To overcomethis problem, several approaches have been reported [11]. NYS hasbeen formulated in micellar gels [18], mucoadhesive devices for topicaluse [21,22], liposomes [23], niosomes [24], nanoemulsions [19,25], lipidintravenous emulsions [11], intralipids [26], nanosuspensions [5],microparticles [27], pellets [1], solid lipid nanoparticles [28] and nano-structured lipid carriers [29].

The use of lipid formulations such as conventional microemulsionand self micro/nano emulsifying drug delivery systems (SMEDDSs/SNEDDSs) have generated much academic and industrial interest aspotential formulations for improving the oral bioavailability of drugs[30] and representing a unique solution to delivery of poorly solublecompounds [31]. The advantages of SNEDDSs over the conventionalemulsions or other lipid carriers are the significantly reduced energy re-quired for their preparation, their physical stability upon storage [32]and easier tomanufacture in a large scale [33]. The SNEDDSs can reducethe limitation of slow and incomplete dissolution of poorly water solu-ble drugs and facilitate the formation of solubilized phases from whichabsorption might take place [34]. Hydrophobic drugs can be dissolvedin these systems, enabling them to be administered as a unit dosageform for per-oral administration. This leads to in situ solubilization ofdrug that can subsequently be absorbed by lymphatic pathways, bypassing the hepatic first-pass effect [35]. The rationale to use SNEDDSsfor the delivery of poorly soluble drugs is that, they are presented inthe form of pre-concentrated solution. Hence, the dissolution steprequired for solid crystalline compounds shall be avoided [33,36]. Inaddition, the formation of a variety of colloidal species on dispersionand subsequent digestion of SNEDDSs facilitates drug absorption [37].SNEDDSs have been described in the literatures as homogenous (trans-parent) complex systems consisting of oils, surfactants, co-solvents andwater,which are thermodynamically stable [34]. SNEDDSs provide ultralow interfacial tensions and large o/w interfacial areas. Therefore,SNEDDSs have the advantages in possessinghigher solubilization capac-ity than simple micellar solutions, leading to the incorporation of poorwater-soluble pharmaceutical inside the oil phase [38]. They rapidlyform oil in water (o/w) nanoemulsion when exposed to aqueousmedia upon gentle agitation or digestive motility of GI tract [35,39].When a SNEDDS is introduced into the body, it is rapidly dispersed toform droplets of approximately nano-size range (b200 nm) [40].Furthermore, these particulate delivery systems show a prolongedresidence time on mucosal membranes [41] and they could reachgreater mucosal surface areas, resulting in a comparatively higherdrug uptake [40,42]. Because of the small droplet size, nanoemulsionskeep the transparency after aqueous dilution [34].

On the basis of these considerations, the major aim of the presentstudy is the development, optimization, characterization and evaluationof NYS-SNEDDS,with objectives of enhanced solubilization of thepoorlysoluble drug, NYS, for the treatment of oral fungal infections.

2. Materials and methods

2.1. Materials

NYS was kindly donated by GlaxoSmithKline Co., Cairo, Egypt. Oleicacid (extra-pure, 99%) was purchased from Alpha Chemika, India.Tween® 20 (Tw20) (extra-pure, 99%)was obtained from VWR Interna-tional, France. Tween® 40 (Tw40) (extra-pure, 99%) and Tween® 60(Tw60) (extra-pure, 99%) were bought from Sisco Research Laborato-ries (SRL) PVT Ltd., India. Tween® 80 (Tw80) (extra-pure, 99%) wasprocured from Riedel-de Haën, Italy. Dimethyl sulfoxide (DMSO)(HPLC grade, 99.5%)was purchased fromTedia, USA. Dimethyl formam-ide (DMF) (99.5%)was obtained fromS.D. Fine-Chem Ltd., India. Propyl-ene glycol (PG) (99.5%)was bought fromBDHLaboratory, UK.Methanol

(HPLC grade, 99.7%) was procured from Fisher Scientific, UK. All otherchemicals were of analytical grade.

2.2. Methods

2.2.1. Solubility studyThe solubility of NYS in various oils (Isopropyl myristate, Isopropyl

palmitate, Oleic acid, Methyl laurate and Miglyol), surfactants (Tw20,Tw40, Tw60, Tw80, Diacetin, Captex 200 and Span® 80) and co-surfactants (PG, DMSO, Polyethylene glycol (PEG) 400 and Ethanol)was determined. An excess amount of NYS (≈50 mg) was added to2 g of each component in screw-capped glass vials. The obtained mix-tures were mixed continuously for 2 min using vortex mixer (Julabo,Paramix II, Germany) to facilitate proper mixing of NYS with the vehi-cles. The mixtures were shaken (100 rpm) for 72 h at 25 ± 0.5 °C in athermostatically controlled shaking water bath (Memmert, SV 1422,Germany) followed by equilibrium for 24 h [31,43]. The equilibratedsamples were removed and centrifuged at 5000 rpm for 30 min. Thesupernatant solution was taken and filtered through a Milliporemembrane filter (0.45 μm) and then suitably diluted with methanol.The concentration of NYS was determined spectrophotometricallyusing UV–Visible recording spectrophotometer (Shimadzu, UV-2401PC, Japan) at 304 nmusingmethanol as a blank [14,24]. The experimentwas repeated in triplicates.

2.2.2. Preliminary screening of surfactants for their emulsification abilityDifferent surfactants for the per-oral usewere screened for emulsifi-

cation ability according to the method described by Date andNagarsenker [44]. Briefly, 300 mg of each selected surfactant (Tw20,Tw40, Tw60 or Tw80) was added to 300 mg of the chosen oily phase(oleic acid). The mixtures were gently heated at 50 °C for homogeniza-tion of the components. Fifty mg of each mixture was then diluted withdouble distilled water to 50 ml in a stoppered volumetric flask to yieldfine emulsion. Ease of emulsification was judged by the number offlask inversions required to yield homogenous emulsion. The resultingemulsions were allowed to stand for 2 h and their % transmittancewas evaluated spectrophotometrically at 638 nm using double distilledwater as a blank. Emulsionswere furthermore observed visually for anyturbidity or phase separation [45,46].

2.2.3. Construction of pseudo-ternary phase diagramsSelf-nanoemulsifying systems form fine o/w emulsions when intro-

duced into aqueous media with gentle agitation. Surfactant and co-surfactant get preferentially adsorbed at the interface, reduce thetension, and provide mechanical barrier to prevent the globules fromcoalescence. The decrease in free energy required for emulsion forma-tion consequently improves the thermodynamic stability [47]. On thebasis of the solubility studies (Section 2.2.1) and preliminary screeningof surfactants (Section 2.2.2), oleic acid was selected as the oil phase,Tw20 and Tw40 as the surfactants and DMSO and PG as the co-surfactants. Double distilled water was used as the aqueous phase forconstruction of phase diagrams. Pseudo-ternary phase diagrams ofmixed surfactant and co-surfactant (Smix), oil, and water but withoutdrug incorporation were plotted, and each of them represents a sideof the triangle. For any mixture, the total of surfactant, co-surfactantand oil concentrations always added to 100% [46]. Ternary mixtureswith varying compositions of surfactant, co-surfactant, and oilwere prepared resulting in a total amount of 1 g. Surfactant and co-surfactant were mixed in three ratios, namely; 1:1, 2:1 and 3:1 (Smix,w/w). For each phase diagram, oil and specific Smix ratio was mixedthoroughly in nine different weight ratios from 1:9 to 9:1 in differentglass vials. The nine different combinations of oil and Smix; 1:9, 2:8,3:7, 4:6, 5:5, 6:4, 7:3, 8:2, and 9:1, were made so that maximum ratioswere covered for the study to delineate the boundaries of phase precise-ly formed in the phase diagrams. Pseudo-ternary phase diagrams weredeveloped using the aqueous titration method. Slow titration with

221A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

aqueous phase was done to each weight ratio of oil and Smix. The totalwater consumedwas noted in terms ofw/w during titration of oil–Smixratio and observations were made for phase clarity [47]. The formationof the nanoemulsion was visually observed as clear/transparent andeasily flowable/dispersible with low viscosity o/w nanoemulsion andmarked on the pseudo-ternary phase diagram [38,48]. The point atwhich the mixture appeared turbid, viscous, or separated in phaseswas considered as the endpoint of titration. The amount of each compo-nent (oil, Smix andwater) at this point was recorded and presented in apseudo-ternary phase diagram [34]. The phase diagram was plottedusing ternary plot software (SigmaPlot forwindows, version 11.0, SystatSoftware, Inc., Ca, USA). All the studies were repeated thrice, withsimilar observations being made between repeats.

2.2.4. Preparation of NYS-SNEDDSOnce the self-emulsifying region was identified, the desired compo-

nent ratios of SNEDDS were selected (Table 1) for drug incorporationand further optimization. Ten mg of NYS was dissolved in the Smix, attheir determined ratios, and mixed continuously for 2 min using thevortex mixer. The oil phase was then finally incorporated into themixture containing drug and mixed using the vortex mixer to obtain ahomogenousmixture. The preparedNYS-SNEDDSswere kept in a tight-ly closed glass bottles at 25 °C and from these, the stable formulationswere subjected to further studies.

2.2.5. Optimization of formulations

2.2.5.1. Thermodynamic stability studies. The objective of these tests wasto evaluate the stability and phase integrity of NYS-SNEDDS underdifferent conditions of temperature variation and centrifugal force[34]. Three different tests were carried out in this study [49]:

2.2.5.1.1. Centrifugation study. Formulations were centrifuged at5000 rpm for 30 min and were then checked visually for instabilitysuch as phase separation, creaming, cracking or drug precipitation.The formulations that did not show any signs of instability were chosenfor heating–cooling cycle.

2.2.5.1.2. Heating and cooling cycle. Heating cooling cycle so per-formed involved three cycles between 4 °C and 45 °C with storage ateach temperature for not less than 48 h. The formulations that passedat these temperatures, without undergoing any creaming, cracking, co-alescence, phase separation or phase inversion, were chosen for freezethaw stress test.

2.2.5.1.3. Freeze thaw cycle (accelerated aging). Freeze thaw cycleinvolved three freeze thaw cycles at temperatures between −20 °Cand +25 °C with storage at each temperature for not less than 48 h.

Table 1Composition, thermodynamic stability and dispersibility studies of prepared NYS-SNEDDS form

Formulation Oil (%) Surfactant (%) Co-surfactant (%) Smix Water

Oleicacid

Tw20 Tw40 DMSO PG Ratio % %

F1 9 54 – 27 – 2:1 81 10F2 9 60.75 – 20.25 – 3:1 81 10F3 9 40.5 – – 40.5 1:1 81 10F4 9 54 – – 27 2:1 81 10F5 9 60.75 – – 20.25 3:1 81 10F6 18 36 – – 36 1:1 72 10F7 18 48 – – 24 2:1 72 10F8 18 54 – – 18 3:1 72 10F9 27 47.25 – – 15.75 3:1 63 10F10 9 – 40.5 – 40.5 1:1 81 10F11 9 – 54 – 27 2:1 81 10F12 9 – 60.75 – 20.25 3:1 81 10F13 18 – 36 – 36 1:1 72 10F14 18 – 48 – 24 2:1 72 10F15 18 – 54 – 18 3:1 72 10

The formulations were then visually observed for phase separation.Only formulations that were stable to phase separation were selectedfor dispersibility study.2.2.5.2. Dispersibility study. The dispersibility studies were carried out toobserve the self-emulsification efficiency and self-emulsification time[47]. Self emulsification time is the time required by the pre-concentrate to form a homogeneous mixture upon dilution, when dis-appearance of SNEDDS is observed visually [33]. Briefly, 0.5 ml of eachformulation was added drop wise to 250 ml of double distilled waterin a glass beaker with gentle agitation by placing it on a hotplate mag-netic stirrer at 100 rpm with temperature adjusted to 37 °C ± 0.5 °C[47]. The time required for the disappearance of the SNEDDS was re-corded [50]. The efficiency of self-emulsification of SNEDDSwas visuallyassessed using the five grading system [30,51].

Grade A: refers to rapidly forming (within 1 min) nanoemulsion,having a clear or bluish appearance.Grade B: is rapidly forming (within 2 min), slightly less clearnanoemulsion, having a bluish white appearance.Grade C: is fine milky emulsion that was formed within 2 min.Grade D: is dull, grayish white emulsion having slightly oil appear-ance that is slow to emulsify (longer than 2 min).Grade E: is a formulation, exhibiting either poor or minimal emulsi-fication with large oil globules present on the surface (longer than3 min).

Formulations those passed this test in grade A or B were selected forfurther optimization.2.2.5.3. Robustness to dilution. Robustness of selected NYS-SNEDDSformulations to dilution was studied by diluting it 50, 100 and 1000times with various diluents i.e. double distilled water, 0.1 N HCl andphosphate buffer (pH 6.8). The diluted samples were stored for 24 hand observed for any signs of physical changes i.e. phase separation ordrug precipitation [33,45].

2.2.6. Characterization of formulationsBased on optimization results, five NYS-SNEDDS formulations were

selected to perform characterization and further investigations. NYS-SNEDDS formulationswere characterized for their cloudpoint, electricalconductivity, optical clarity, droplet size, polydispersity index, zetapotential and morphology.

2.2.6.1. Cloud point (TCloud) determination. The effect of temperatureon the phase behavior of NYS-SNEDDSwas evaluated throughmeasure-ment of the cloud point value (TCloud) [45,49]. Each optimized

ulations.

Oil/Smix Thermodynamic stability studies Dispersibility study

Ratio Centrifugationstudy

Heating andcooling cycle

Freeze–thawcycle

Grade Inference

1:9 Passed Passed Passed A Passed1:9 Passed Passed Passed A Passed1:9 Passed Passed Passed A Passed1:9 Passed Passed Passed A Passed1:9 Passed Passed Passed A Passed2:8 Passed Passed Passed D Failed2:8 Passed Passed Passed E Failed2:8 Passed Passed Passed E Failed3:7 Failed Failed Failed – –1:9 Passed Passed Passed C Failed1:9 Passed Passed Passed E Failed1:9 Passed Passed Passed E Failed2:8 Passed Passed Passed E Failed2:8 Passed Passed Passed E Failed2:8 Passed Passed Passed E Failed

222 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

formulationwas dilutedwith double distilledwater in the ratio of 1:100and placed in a water bath with gradual increase in temperature (5 °Cincrements). Each samplewas equilibrated at the specified temperaturefor at least 2min. At the end of eachmeasurement the samples were vi-sually assessed for optical transparency, sings of phase separation ordrug precipitation [52].

2.2.6.2. Electrical conductivitymeasurement. The electrical conductivity ofselectedNYS-SNEDDS, before and after 50-fold dilutionwith double dis-tilled water, was determined at 25 °C using advanced electrochemistrymulti-parameter meter (Thermo Fisher Scientific Inc., Orion VSTAR 92,USA).

2.2.6.3. Spectroscopic characterization of optical clarity. Nanoemulsion isdefined as “a system of water, oil, and surfactant (or amphiphile)which is a single optically isotropic and thermodynamically stable solu-tion” [31]. Therefore, to meet the isotropic parameter, optical clarity in-vestigation is important. The optical clarity of the aqueous dispersions ofthe selected NYS-SNEDDS formulations was measured spectrophoto-metrically upon dilution [43,53]. Briefly, 1ml of each SNEDDSwas dilut-ed with 50 ml double distilled water. The absorbance of each solutionwas measured at 638 nm using double distilled water as a blank.

2.2.6.4. Droplet size analysis. The droplet size of the selected formulationswas determined by dynamic light scattering (DLS) using Zeta-Sizer(Malvern, Nano Series ZS90, Malvern Instruments, Ltd., UK). A He–Nelaser beam at 632 nm wavelength was used and light scattering wasmonitored at 25 °C at a fixed angle of 90°. The analytical sensitivityrange of the analyzer is 1 nm to 6 μm. The selected formulations werediluted at 1:50 (v/v) employing double distilled water and mixed for1 min using a magnetic stirrer to assure the homogeneity. After com-plete dispersion, aliquots were transferred to a cuvette and loadedinto the apparatus for measurement. The polydispersity index (PDI)reflects the uniformity of particle diameter and it can be used to depictthe size distribution of nanoemulsion. All measurements were done intriplicate from three independent samples.

2.2.6.5. Zeta potential measurement. The nanoemulsion stability is direct-ly related to the magnitude of the surface charge [33,54]. The zetapotential of the selected formulations was determined by laser diffrac-tion analysis using Zeta-Sizer (Malvern, Nano Series ZS90, Malvern In-struments, Ltd., UK). The samples were diluted at a ratio of 1:50 (v/v)with double distilledwater andmixed for 1min using amagnetic stirrerto guarantee the homogeneity. After complete dispersion, aliquotsweretransferred to a folded capillary cell and loaded into the apparatus forinvestigation. All studies were repeated in triplicate at 25 °C.

2.2.6.6. Transmission electron microscopy (TEM). The morphology of theselected NYS-SNEDDS was observed by TEM (JEOL Co., JEM-2100,Tokyo, Japan), under a high tension electricity of 160 kV. After sampledilution with double distilled water (1:50), a sample drop was placedon a carbon coated copper grid and air-dried for 10min at room temper-ature. The excess was instantly drawn off with a filter paper. Sampleswere subsequently negatively stained with 1% (w/v) phosphotungsticsolution for 60 s and excess of solutionwas also removed, before loadingin the microscope. Subsequently, the shape and surface characteristicswere evaluated at appropriate magnifications.

2.2.7. In vitro drug release studyThe release profile study of selected NYS-SNEDDS formulations was

performed using the dialysis bag method [45,55] with some modifica-tions. In all release experiments the dialysis bag (Dialysis tubing cellu-lose membrane, Sigma Co., USA; Molecular weight cutoff 12,000–14,000) was used. One g of NYS-SNEDDS containing 2 mg NYS wereinstilled into the dialysis bag, firmly sealed with dialysis clamp andplaced in a 100 ml screw-capped glass container filled with methanol/

DMF/water (55:15:30, v/v/v) [8,11,19,25,27] as the release medium.DMF had been utilized by others authors in dialysis membranes [56].This release medium allows maintaining sink conditions in the wholeexperiment. In this way, the released NYS would be soluble in the re-lease medium in contrast to other previous investigations [57,58], inwhichwater was utilized as release medium,wherein NYS is practicallyinsoluble. The experiment was conducted in a thermo-stated shakingwater bath (Memmert, SV 1422, Germany) equipped at 37 °C ± 0.5 °Cand 100 rpm. Samples were drawn at predetermined time intervals(0.5, 1, 2, 3, 4, 5, 6, 7, 8, 24 and 48 h) and replenishedwith the same vol-ume of fresh release medium. The samples were filtered through0.45 μm pore-sized nylon syringe filter and NYS concentrations weredetermined spectrophotometrically at 304 nm using the regressionequation of a standard curve developed in the same medium. The re-lease profile of NYS from SNEDDS formulations was compared to plainNYS suspended in phosphate buffer, containing the same quantity ofdrug. The cumulative release percentages were calculated as the ratioof the amount of drug released to the initial amount of drug in thedialysis bag. All measurements were performed in triplicate fromthree independent samples.

2.2.8. Stability studyLong-term stability was assessed by keeping the optimized NYS-

SNEDDS formulations into sealed amber glass vials at ambient roomtemperature (20–25 °C) and refrigeration temperature (4–8 °C) for6 months. The physical stability of the optimized formulations wasevaluated by monitoring the time-dependent change in the physicalcharacteristics (e.g., drug precipitation, phase separation and clarity)in addition to the drug content [31,38]. Samples were withdrawn atspecified time intervals of 15, 30, 60, 90, 120 and 180 days and thedrug content analysis was carried out spectrophotometrically at304 nm after appropriate dilutions using methanol. All measurementswere repeated for three independent samples.

2.2.9. In vitro antifungal activity

2.2.9.1. Organism. C. albicans ATCC 60193 was used for the induction oforal candidiasis in mice. C. albicans was cultured, from a stored stock at−80 °C, on Sabouraud dextrose (SD) agar plate and incubated at 37 °Cfor 24 h. Then, yeast cells were harvested into 6ml SD broth and left over-night in a shaker incubator at 37 °C. The overnight broth was centrifugedand the pellet was diluted using sterile saline to optical density (OD) =1.6 which corresponds to a cell count of 3 × 108 colony forming units(CFU)/ml. The cell countwas verified by viable count of serial 10-folds di-lutions spotted on SD agar plate incubated at 37 °C for 48 h.

2.2.9.2. Determination of minimum inhibitory concentration (MIC). TheMIC is defined as the lowest concentration of an antimicrobial agentthat inhibits the growth of a microorganism [19,25]. The MICs of the 5optimized formulations (F1–F5) compared to marketed NYS suspen-sion, containing the same quantity of drug, as well as plain NYSsuspended in phosphate buffer, were assessed against C. albicans using2-fold serial dilution method [59]. A disk impregnated with 10 μl ofeach dilution was applied on the surface of SD agar media inoculatedby an overnight culture of C. albicans. The plates were incubated at25 °C for 24 h. The lowest concentration that gave considerable inhibito-ry zone was recorded.

2.2.9.3. Time-kill test. The time needed for NYS-SNEDDS formulations,to achieve their lethal effect against C. albicans, compared to marketedNYS suspension, containing the same quantity of drug, as well as plainNYS suspended in phosphate buffer, was calculated according to Clancyet al. [60] with somemodifications. Concentrations of 256 and 512 μg/μlof each NYS-SNEDDS, suspensions of marketed NYS and plain NYSwereinoculated by 5% v/v of an overnight culture of C. albicans adjusted to anOD of 1.6 and incubated at room temperature in glass test tubes. A

Table 2Solubility of NYS in various vehicles at 25 °C.

Vehicle Solubility of NYS (mg/g) (n = 3)

OilsOleic acid 2.111 ± 0.042Isopropyl myristate 0.063 ± 0.003Isopropyl palmitate 0.053 ± 0.002Methyl laurate 0.075 ± 0.005Miglyol 0.194 ± 0.009

SurfactantsTween 20 1.792 ± 0.053Tween 40 1.987 ± 0.039Tween 60 1.425 ± 0.028Tween 80 1.369 ± 0.019Dicetin 0.797 ± 0.014Captex 200 0.135 ± 0.008Span 80 1.227 ± 0.022

Co-surfactantsPolyethylene glycol 400 3.625 ± 0.067Propylene glycol 4.697 ± 0.096Ethanol 0.966 ± 0.025Dimethyl sulfoxide N50 mg

223A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

volume of 100 μl of each test tube was collected after contact time of 0,5, 10, 15 and 20 min. The antifungal effect of the formulations and NYSsuspensions were neutralized by dilution into Dey–Engley neutralizingmedia. A volume of 10 μl of the previousmixturewas spotted on SD agarplates and incubated at upright position at 25 °C for 24 h. The least con-tact time that showed a total absence of any sign of fungal growth foreach formula was recorded.

2.2.10. In vivo antifungal activity

2.2.10.1. Animals.All the experiments involving animalswere conductedin accordance with the institutional ethical and regulatory guidelinesand were approved prior by the Institutional Animal Ethics Committee(Medical Research Ethics Committee (MREC) of the National ResearchCentre, Cairo, Egypt). Fifty male albino Swiss mice weighing 20–25 gwere employed in the present study. The mice were kept in polyethyl-ene cages, fed with pelleted food and tap water ad libitum, and exposedto alternating 12 h of light and darkness. The care of the rats was in ac-cordance with the institutional guidelines.

2.2.10.2. Experimental oral candidiasis in mice. Based on the in vitrorelease study, long-term stability study and in vitro antifungal activityexperiments, two NYS-SNEDDS formulations (F1 and F2) were chosento investigate their in vivo antifungal activity. The study of experimentaloral candidiasis in mice was performed according to the methoddescribed in previous reports [61,62] with some modifications. Allmice were immunosuppressed with three subcutaneous injections ofprednisolone (Depo-Medrol®, EIPICO Pharmaceutical Co., Egypt,under license of Pfizer Co., USA) at a dose of 100 mg/kg of body weightat days−1,+1 and+3 of C. albicans infection. Tetracycline hydrochlo-ride (0.1%) (Tetracid®, CID Pharmaceutical Co., Egypt) was given in thedrinking water beginning 1 day before the infection and maintainedduring the whole experiment. Each mouse was orally infected (day0) with 0.1 ml saline suspension containing 3 × 108 viable cells ofC. albicans. Oral infection was performed by means of a cotton swabrolled twice over all parts of the mouth. The animals were then afterrandomly divided into five groups (n = 10 per group): Group I wastreated with marketed NYS suspension, group II with plain NYSsuspended in phosphate buffer, group III with F1, group IV with F2and group V with phosphate buffer, as a control. All treatment groups(Groups I–IV) received a NYS dose of 80,000 IU/kg body weight by avolume of 0.1 ml, and group V received phosphate buffer (0.1 ml) [5].Just before the start of treatment (day 1), sample swabs were takenfrommicemouths to confirm the presence of C. albicans and to quantifythe number of CFU in the oral cavity. Treatments were administeredintra-orally twice daily for 7 consecutive days, beginning 1 day afteroral inoculation of C. albicans. The treatments were topically adminis-tered into the oral cavity using a 1.0 ml syringe equipped with a gavageneedle. The mice were immobilized during dosing in a supine positionwithout anesthesia for 1 min, allowing the animals to gargle the treat-ments and then swallow it [5].

Microbiological evaluation of progression of the infection was car-ried out as follows: the whole oral cavity including the buccal mucosa,tongue, soft palate and other oral mucosal surfaces was swabbedusing a cotton swab at days 1, 3, 5 and 7 post infection, and analyzedfor the viable count of the yeast cells. The end of the cotton swab wasthen cut off and placed in a tube containing 1 ml sterile saline. Aftermixing on a vortex mixer, to release Candida cells from the swab intothe saline, serial 10-fold dilutions of the cell suspension were spottedon SD agar plates and incubated at 37 °C for 48 h. The CFU of Candidacolonies were counted. All results were expressed as log10 CFU/ml.Mice were monitored once daily to evaluate their survival during the7 days treatment period.

2.2.10.3. Histopathological findings. At the end of the experiment (24 hafter the last administration of treatments), 3 mice from each group

were sacrificed by cervical dislocation. The tongues were removed byincision at the base, rinsed gently with normal saline and processedseparately for microscopic examination. Briefly, the tongues werefixed in 20% formalin solution followed by embedding in paraffin. Six-μm-thick sections were taken from paraffin blocks and stained withhematoxylin and eosin (HE) for histopathological examinations.

2.2.11. Statistical analysisDatawere expressed asmean± standard deviation (S.D.). Statistical

analysis was performed using (SPSS® statistics for windows, version17.0). The significance of differences between the mean values of thein vivo evaluation was determined by one-way analysis of variance(ANOVA) followed by Fisher LSD's post-hoc test, and a P-value b0.05was considered statistically significant.

3. Results and discussion

Poor water solubility and consequent restricted absorption is amajor limitation with many drugs despite their good efficacy. SNEDDSprovides an opportunity for the improvement of the in vitro andin vivo performance of poorly water soluble drugs and thus serves asan ideal carrier for the delivery of drugs belonging to BiopharmaceuticsClassification System (BCS) classes II and IV [54]. The current study wasperformed to define the role of self-nanoemulsifying formulations toenhance the solubility and bioavailability of the antifungal drug, nysta-tin. SNEDDS represents a possible alternative to the more traditionaloral suspensions for lipophilic compounds.

3.1. Solubility study

Solubility of drug in excipients plays an important role in determin-ing stability of formulation, asmany formulations undergo precipitationbefore undergoing in situ solubilization [31]. In Addition, for a successfulformulation of NYS loaded SNEDDS, the entire dose of NYS should besoluble in SNEDDS ingredients. The formulations of self-emulsifyingproperty should be carried out following combination of oils, surfac-tants, and co-surfactants. This mixture should be clear, isotropic,monophasic liquid at room temperature [63] and should have good sol-ubilizing capacity to incorporate dose of drug in minimum volume ofthe mixture [64]. The solubility of NYS in various oils, surfactants andco-surfactants is presented in (Table 2). Among various vehiclesscreened, oleic acid was selected as the oil phase showing the highestsolubilization capacity (2.111 ± 0.042 mg/g). Tw20 (1.792 ±0.053 mg/g), Tw40 (1.987 ± 0.039 mg/g), Tw60 (1.425 ± 0.028 mg/g) and Tw80 (1.369± 0.019mg/g) were used as surfactants possessing

224 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

the highest solubilities. Additionally, DMSO (N50 mg/g) and PG(4.697 ±mg/g) were chosen as the co-surfactants due to their efficientand moderate solubilizing effect, respectively. Oleic acid was reportedfor its enhanced intestinal absorption of drugs [64,65]. Tweens arenon-ionic surfactants providing formulation benefits in a number ofpharmaceutical applications having long standing food and pharmaco-peia approval [66]. DMSO and PG are safe co-surfactants previouslyreported as SNEDDS excipients for oral administration [30].

3.2. Preliminary screening of surfactants for their emulsification ability

Emulsification process and efficiency are controlled bymultiple var-iables including lipid–surfactant affinity, hydrophilic–lipophilic balance(HLB) value of surfactant and viscoelasticiy of the emulsion base [67].HLB value of surfactant affects the spontaneity of emulsification andemulsion droplet size. Surfactants with HLB N 10 are suitable for forma-tion of o/w nanoemulsions [68] and hence all screened surfactants haveHLB N 10 (Table 3). Non-ionic surfactants are often considered for phar-maceutical applications and nanoemulsion formulations since these areless toxic [45], less affected by pH and ionic strength changes [32] andtypically have lower critical micelle concentration compared to theirionic counterparts [67,69]. They are usually accepted for oral ingestion[70]. The selected surfactantswere compared for their emulsification ef-ficiencies using oleic acid as the oily phases. It has been reported thatwell formulated SNEDDS is dispersed within seconds under gentle stir-ring conditions [70]. Transmittance values aswell as number of flask in-versions of differentmixtures are demonstrated in (Table 3). The resultsinferred that highest % transmittance, i.e.highest emulsification efficien-cy, is acquired by Tw20 (HLB 16.7) followed by Tw40 (HLB 15.6) thenTw80 (HLB 15.0) and Tw60 (HLB 14.9) in the same order. The observeddifference in their emulsifying ability could be attributed to the HLBvalues of the examined surfactants, where Tw20 possessed the highestvalue and Tw60 the lowest, in addition to the difference in theirstructure and chain length [44,46]. Higher HLB surfactants lead to theformation of more stable nanoemulsion upon exposure to water [68].Tw20 and Tw40 showed highest solubility as well as emulsification ef-ficiency (yielded clear nanoemulsions requiring shorter time for nano-emulsification). Thus, both were selected for further investigations.

3.3. Construction of pseudo-ternary phase diagrams

Ternary phase diagrams were constructed in the absence of NYS toidentify the self-nanoemulsifying region and to select a suitable concen-tration of oil, surfactant and co-surfactant for the development of liquidSNEDDS formulations. These phase diagram plays important role instudying phase behavior of the formed nanoemulsions [31,33]. A simpleternary phase diagram comprises oil, water, and Smix, where each cor-ner in the phase diagram represents 100% of that particular component.Based on the data obtained from the solubility study and preliminaryscreening of surfactants, oleic acid was used as the oil phase, Tw20and Tw40 as surfactants and PG and DMSO as co-surfactants for con-structing different phase diagrams. Fig. 1 shows the constructed phasediagrams, where the translucent and low viscosity nanoemulsion areais presented as shaded area in the phase diagrams. It was clearlyshown that the use of DMSO as co-surfactant at Smix ratio (1:1) withTw20 could not form clear isotropic nanoemulsion region while atSmix ratio (2:1) and (3:1) could form nanoemulsion but with not

Table 3Emulsification ability of selected surfactants using oleic acid as the oily phase.

Trade name Chemical name

Tween® 20 Polyoxyethylene sorbitan monolaurateTween® 40 Polyoxyethylene sorbitan monopalmitateTween® 60 Polyoxyethylene sorbitan monostearateTween® 80 Polyoxyethylene sorbitan monooleate

more than 20% oil. Employing Tw40 could not form any nanoemulsionregion with DMSO as co-surfactant. The use of PG as co-surfactantgave clear isotropic regions employing not more than 30% of oil forall constructed phase diagrams except for Tw20/PG (3:1) systemwhich could emulsify up to 40% of oil. Tw20 showed larger self-emulsification region compared to Tw40, due its higher HLB andhence more hydrophilic nature. Higher HLB value is required forforming a good o/w emulsion [68,71] as higher hydrophilicity propertyfavors faster emulsification of the oil–surfactantmixture in contact withwater [47,72]. Self-nanoemulsification is spontaneous and the resultingdispersion is thermodynamically stable [38,73]. Free energy ofnanoemulsion formation depends on the extent to which the surfactantlowers the surface tension of the oil–water interface and the change indispersion entropy [69], and the present results demonstrated thatincreasing surfactant proportion (Smix 1:1, 2:1 and 3:1) led to a morefavorable formation of nanoemulsion. The surfactant forms a layeraround oil globule in such a way that polar head lies toward aqueousand non-polar tail pull out oil and thereby reduces surface tensionbetween oil phase and aqueous phase [73,74]. Further, increasing theconcentration of surfactant increased the spontaneity of the self-emulsification process [47]. The increase in co-surfactant decreasesthe region of emulsion formation, as co-surfactants have very little ef-fect on reducing the interfacial tension directly rather they help the sur-factants to reduce the interfacial tension [68]. Results obtained indicatedthat apart from HLB value and type of surfactant, other factors such asstructure and relative length of hydrophobic chains of co-surfactanthad influence on nano-emulsification and therefore nanoemulsionarea [31]. Moreover, since the drug is co-surfactant more soluble (espe-cially DMSO), co-surfactant concentration is an important factor tosolubilize required amount of drug dose.

3.4. Optimization of formulations

3.4.1. Thermodynamic stability studiesThe main difference between emulsions and nanoemulsions is

kinetic stability, reflecting the thermodynamic stability of the two sys-tems. SNEDDS undergoes in situ solubilization to form nanoemulsionsystem, and it should have stability such that it does not undergo pre-cipitation, creaming or cracking. However, in many cases, prolongedstorage might cause the drug to precipitate from the nanoemulsion;seed crystals start to appear and might grow to large crystalline mate-rials that will precipitate out at the bottom of the vessel [31]. Thereforeto check the stability, formulation was exposed to centrifugation study,heating and cooling cycle and freeze thawing cycle in order to eliminatethemetastable ones. Fourteen formulations passed the test (Table 1) i.e.there was no sign of phase separation, turbidity or drug precipitationobserved. One formulation (F9), containing oleic acid in a concentrationmore than 25%, were unstable and showed phase separation as well asturbidity, which may be due to the coagulation of the internal phasewhich led to phase separation [43,53]. The stable formulations werefurther tested for dispersibility.

3.4.2. Dispersibility studyAs SNEDDSs are released in the lumen of the gastrointestinal tract, it

disperses to form a fine nanoemulsion with the aid of GI fluid. Thus, it isimportant that formed nanoemulsion does not undergo precipitationfollowing phase separation with infinite dilution in the GI fluids. It is

HLB value No. of inversions % transmittance

16.7 3 58.47915.6 4 42.85514.9 7 37.67015.0 5 41.115

Fig. 1. Pseudo-ternary phase diagrams of oleic acid, Tw20, Tw40, DMSO and PG at Smix ratios 1:1, 2:1 and 3:1 indicating the clear o/w nanoemulsion region as shaded regions.

225A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

observed more prominently with drugs having poor aqueous solubilityor nanoemulsion which undergoes phase transition [31]. The SNEDDSshould disperse completely and quickly when subjected to dilutionunder mild agitation of GIT due to peristaltic activity [31,33]. It hasbeen reported that self-emulsification mechanism involves the erosionof a fine cloud of small droplets from the monolayer around emulsiondroplets, rather than progressive reduction in droplet size [31,75]. Theease of emulsification was suggested to be related to the ease of waterpenetration into the colloidal or gel phases formed on the surface ofthe droplet [76].

Accordingly, dispersibility study in double distilled water wasconducted. In the present study, double distilled water was used as adiluent for self-nanoemulsification test because it is well known thatthere is no significant difference in the SNEDDSprepared using pharma-ceutically acceptable surfactants, dispersed in either water or GI fluids[77,78]. All SNEDDSs that passed this test either in grade A or B wereselected for further investigation, because grade A or B formulationswill remain as SNEDDS when dispersed in GI fluids [48] (Table 1). Itwas clearly revealed that only SNEDDSs prepared using Tw20 as thesurfactant at ratio oil/Smix (9:1) i.e. F1–F5, could pass the test. Thismight be explained by the higher HLB value (16.7) possessed by Tw20and consequently higher hydrophilicity compared to Tw40 (15.6).Higher HLB value is required for forming a good o/w emulsion [68,71]since higher hydrophilicity property favors faster emulsification of theoil–surfactant mixture in contact with water [47,72], in addition to re-duced free energy of the system [79]. It was also observed that the

increase in oil concentration (F6–F8 and F13–15), required more timefor emulsification. The reason that can be put forward is the increasein interfacial tension between larger volume of oil and aqueous phasewith a net decrease in surfactant system, which makes the emulsifica-tion time longer [80]. Pouton found that the formulations containingmore hydrophilic components was superior in self-nanoemulsifyingability and provided smaller droplets than those containing more lipo-philic components [81]. Eventually, five formulations (F1–F5), whichpassed dispersibility study, were selected for further optimizations asthese formulations were certain to form nanoemulsion upon dilutionin the aqueous environment.

3.4.3. Robustness to dilutionUniform emulsion formation from SNEDDS is very important at dif-

ferent dilutions because drugs may precipitate at higher dilution in vivowhich affects the drug absorption significantly [32,33]. OptimizedSNEDDS formulations were exposed to different folds of dilution in dif-ferent media to mimic the in vivo conditions where the formulationwould encounter gradual dilution. Hence, each formulation was sub-jected to 50, 100 and 1000 times dilution in double distilled water,0.1 N HCl, and phosphate buffer (pH 6.8). Even after 24 h, all investigat-ed formulations showed no signs of precipitation, cloudiness or separa-tion which ensured the stability of the reconstituted emulsion. Thisreveals that all mediawere robust to dilution. These findingswill ensurethe prospect of uniform drug release profile in vivo.

226 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

3.5. Characterization of formulations

3.5.1. Cloud point (TCloud) determinationTemperature dependent phase behavior is one of the major prob-

lems associated with nanoemulsions, especially when dealing withnon-ionic surfactants [49,52]. An ideal formulation should remain as aone-phase transparent system at its storage temperature and at thetemperature of its intended use. The cloud point (TCloud) is the temper-ature above which the formulation clarity turns into cloudiness [45]. Athigher temperatures, phase separation can occur due to the decrease inthe solubility of the surfactant in water, the dehydration of the headgroup and phase separation [52]. Since both drug solubilization andformulation stability will decline with this phase separation, thecloud point of the formulation should be over 37 °C [45]. The cloudpoint value is affected by factors such as drug hydrophobicity, kind,combination, mixing ratio and amount of each of the oils, surfactantsand co-surfactants used [50,82]. The cloud points for the five selectedNYS-SNEDDS formulations were much higher (90 °C) as reported in(Table 4), indicating that a stable and transparent systemwill be formedat physiological temperature in vivo.

3.5.2. Electrical conductivity measurementTable 4 shows the electrical conductivity results of the selected for-

mulations, where the values ranged between (0.54–2.25 μS/cm) and(20.17–36.02 μS/cm) before and after 50-fold dilution using double dis-tilledwater, respectively. Itwas clearly depicted that before and after di-lution, the higher the co-surfactant concentration the higher theconductivity value (Smix 3:1, 2:1 and 1:1). Such increase in the electri-cal conductivity may be attributed to greater interaction of water mole-cules with the polar co-solvents (DMSO and PG) that can highlyincorporate into water [53]. In addition, increasing the amount ofwater upon dilution led to noticeable increase in the conductivityvalues. This finding agrees with a previous report [83]. Furthermore,conductivity values N1 μS/cm have been reported to be indicative ofbicontinuous or solution type systems which possess ultra low interfa-cial tension [83]. These dynamic structures include water and oilpseudo-domains which rapidly exchange [84]. In the presence ofthese structures, the conductivity is comparable to that of electrolyte so-lutions and decreases with decreasing water content [83].

3.5.3. Spectroscopic characterization of optical clarityLower spectroscopic absorbance should be obtained with optically

clear solutions because cloudier solutions will scatter more of the inci-dent radiation, resulting in higher absorbance [79]. To assess the opticalclarity quantitatively, UV–Visible spectrophotometer was used to mea-sure the amount of light of a givenwavelength (638 nm) transmitted bythe solution. Absorbance values of optimized formulations F1–F5, upondilution with double distilled water (1 × 50) varied between 0.034–0.189 (Table 4). As expected, the compositions with the lower absor-bance values showed the smallest droplet size values (Table 4) becauseaqueous dispersions with small absorbance values are optically clear,and oil droplets are thought to be in a state of finer dispersion [46,53].It was observed that the spectroscopic clarity (i.e. droplet size) was in-versely proportional to the concentration of surfactant in the Smix(1:1, 2:1 and 3:1). Such a decrease in droplet size may be the result of

Table 4Characterization parameters of optimized NYS-SNEDDS formulations.

Formulation Cloud point (°C) Conductivity (μS/cm) Spectroscopic

No dilution Dilution (1:50)

F1 90 2.25 36.02 0.172F2 90 1.2 32.42 0.068F3 90 0.91 31.59 0.189F4 90 0.73 28.68 0.154F5 90 0.54 20.17 0.079

more surfactants being available to stabilize the oil–water interface[79]. Furthermore, the decrease in the droplet size behavior reflectsthe formation of a better closed packed film of the surfactant at theoil–water interface, thereby stabilizing the oil droplets [68].

3.5.4. Droplet size analysis and polydispersity indexDroplet size is one of the most important characteristics of

nanoemulsion for stability evaluation and a critical step in the pathwayof enhancing drug bioavailability [33,38]. Smaller droplet size leads tolarger interfacial surface area for drug absorption and hence may leadto more rapid absorption and improved bioavailability [33]. Hence, thedroplet size of the nanoemulsion may govern the effective drug release[31,63,79]. The droplet sizes of optimized formulations are given inTable 4. The average droplet size of the five investigated formulationsat 50 times dilution ranged from 26.45–85.94 nm, which indicatedthat emulsion droplets are in nanometric range (b100 nm). The dropletsize decreased as the concentration of surfactant in the Smix increasedi.e. F2 b F1 and F5 b F4 b F3. This could be attributed to an increased sur-factant proportion relative to co-surfactant which is probably explainedby stabilization of the oil droplets as a result of the localization of thesurfactant molecules at the oil–water interface [31,74]. The surfactantmay cause the interfacial film to condense and stabilize, resulting insmaller droplet diameters, whereas the addition of the co-surfactantmay cause the film to expand [43,53]; thus, the relative proportion ofsurfactant to co-surfactant has varied effects on the droplet size [85].These observations are in consistent with earlier author reports [43,49,53]. Polydispersity index (PDI) of all formulations was less than 0.5,indicating uniform globule size distribution [33,48].

3.5.5. Zeta potential measurementEmulsion droplet polarity is also a very important factor in charac-

terizing emulsification efficiency [31,86]. Zeta potential is the potentialdifference between the surface of tightly bound layer (shear plane andelectroneutral region of the solution). The significance of zeta potentialis that its value can be related to the stability of colloidal dispersions.The zeta potential indicates the degree of repulsion between adjacent,similarly charged particles in dispersion. For molecules and particlesthat are small enough, a high zeta potential will confer stability i.e. thesolution or dispersion will resist aggregation [31,33]. When the poten-tial is low, attraction exceeds repulsion and the dispersion will breakand flocculate. So, colloids with high zeta potential (negative or posi-tive) are electrically stabilized. In the present study, the zeta potentialof the nanoemulsionwas determined after 50-fold dilution of the select-ed NYS-SNEDDSwith double distilled water (Table 4). The results showthat the optimized formulations were negatively charged, with valuesranging from −15.3 to −23.9 mV, indicating stable system and wellseparated emulsion globules [49]. The electrostatic repulsion forcesbetween the negatively charged droplets get an avoidance of the coales-cence of the nanoemulsion [63]. Non-ionic surfactants produce a nega-tively charged interface at neutral pH because of differential adsorptionof the hydroxyl ion (OH‾) and hydrated oxonium ion (H3O+) [87]. Thismight explain the negative zeta-potential values of all the SNNEDSformulations.

absorbance Droplet size (nm) ± S.D. PDI Zeta potential (mV) ± S.D.

83.89 ± 9.02 0.487 −19.5 ± 4.2026.45 ± 2.75 0.435 −22.3 ± 4.3285.94 ± 10.11 0.495 −15.3 ± 4.0874.20 ± 6.72 0.488 −19.8 ± 4.0248.56 ± 423 0.441 −23.9 ± 5.15

227A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

3.5.6. Transmission electron microscopy (TEM)TEM was used to explore the structure and morphology of formed

nanoemulsions of two selected NYS-SNEDDS formulations (F1 andF2), after dilution using double distilled water. The TEM images aredepicted in Fig. 2a and b. The images clearly indicate the spherical na-ture of nanoemulsions, with no signs of coalescence. The nanoemulsiondroplets emerged as dark spheres, with smooth surface, and the sur-roundings were found to be bright. Furthermore, the droplets appeareddistinct and non-aggregated with no signs of drug precipitation infer-ring the physical stability of formed nanoemulsions [31,63,79].

3.6. In vitro drug release study

When SNEDDSencounter aqueousmedia, the drug consequently ex-ists in the system in different forms including a free molecular form, ormixed in micelles or entrapped in the nanoemulsion droplets. Underthese circumstances, it is necessary to separate free drug moleculesfrom those entrapped in the nanoemulsion droplets or micelles to

Fig. 2. TEM micrographs of (A) F1 and (B) F2 NYS-SNEDDS stained with 1% (w/v)phosphotungstic acid.

assess the real release pattern [38,45,50]. Thereby, conventional releasetesting is not adequate for this system. Recently, dialysis bag methodwas recently adopted for SNEDDS in vitro release study [69,88]. In thedesigning of colloidal drug-carrier formulations, the study of the rateat which the drug is released from the vehicle is particularly important,because it can be used as quality control data to predict in vivo behavior,or to study the structure and releasemechanismof the system [11]. Pre-vious studies [8,11,19,25,27] demonstrated the ability of the mixturemethanol/DMF/water (55:15:30, v/v/v) to maintain sink conditions inNYS release experiments and excellent compatibility with cellulosemembranes. Fig. 3 shows NYS release profiles from their respectiveSNEDDS formulations at 37 °C for 48 h. The results show that NYS-SNEDDS prepared using DMSO as co-solvent (F1–F2) possessed higherrelease compared to those prepared employing PG as the co-surfactant (F3–F5). In addition, considering F1 and F2, it was observedthat the higher the co-surfactant (DMSO) concentration in the formula-tion (Smix 2:1 and 1:1), the higher the release rate. This might be ex-plained by the higher solubilization of NYS in DMSO compared to PG,which contributes in higher availability of NYS in the release media.On the contrary, in case of F3–F5, it was found that the higher the co-surfactant (PG) concentration in the formulation (Smix 3:1, 2:1 and1:1) the lower the release rate. This could be clarified by the moderatesolubilization efficiency of PG compared to DMSO, where another factormay governs the drug release; the droplet size [63]. As the droplets sizeof the nanoemulsion decreases, a larger interfacial surface area is pro-vided [34]. Therefore, the drug release from small droplets will be fasterandmore frequent than from the large oil droplets [89]. Fig. 3 also dem-onstrates comparison of drug release from SNEDDS formulations toplain NYS suspended in phosphate buffer, containing the same quantityof drug. Release patterns reveal that NYS release was significantlyhigher from SNEDDS formulations compared to plain drug suspension.Drug release from suspension did not exceed 47% after 48 h, whereas,the drug was 98% released from F1 after 48 h. These results contendthe role of SNEDDS formulation to improve NYS solubilization andin vitro release. The faster drug release from SNEDDS is attributed tothe spontaneous formation of nanoemulsion due to low surface free en-ergy at oil–water interface, which causes immediate solubilization ofdrug in the release medium [47]. During emulsification with water,the nanoemulsion components (oil, surfactant and co-surfactant) effec-tively swell and decreases the globule size, which leads to increase insurface area and decrease of surface free energy, thus eventually

Fig. 3. In vitro release profiles of F1–F5 NYS-SNEDDS compared to plain NYS suspension inmethanol/DMF/water (55:15:30 v/v/v) at 37 °C for 48 h. Data presented as mean ± S.D.(n = 3).

Table 5NYS retained in optimized SNEDDS formulations stored at different temperatures for 6 months.

Formulation Temperature (°C) NYS retained (% ± S.D.) (n = 3)

15 days 30 days 60 days 90 days 120 days 180 days

F1 Aa 63.70 ± 0.07 56.16 ± 5.69 41.55 ± 4.03 38.34 ± 2.09 36.15 ± 0.72 32.74 ± 1.37Bb 81.82 ± 1.37 79.43 ± 3.17 71.33 ± 0.22 63.49 ± 1.37 57.08 ± 1.08 54.89 ± 0.14

F2 A 65.74 ± 0.76 60.69 ± 1.98 47.48 ± 0.76 42.53 ± 1.37 39.90 ± 4.94 37.76 ± 1.29B 85.02 ± 3.11 81.69 ± 3.72 74.97 ± 0.15 65.41 ± 1.98 62.51 ± 1.52 60.74 ± 0.08

F3 A 28.54 ± 2.04 24.34 ± 2.37 21.76 ± 0.76 20.26 ± 0.17 19.84 ± 0.09 18.76 ± 1.10B 73.92 ± 2.63 56.06 ± 1.61 42.75 ± 3.31 33.57 ± 2.88 30.22 ± 0.17 29.08 ± 1.27

F4 A 42.25 ± 0.62 30.92 ± 1.16 23.74 ± 0.27 22.86 ± 0.45 22.36 ± 0.09 22.04 ± 0.18B 77.39 ± 2.23 60.77 ± 5.25 48.11 ± 2.85 38.60 ± 0.80 36.21 ± 0.09 31.80 ± 3.12

F5 A 52.00 ± 2.48 40.01 ± 0.43 32.08 ± 0.17 28.63 ± 0.43 26.82 ± 0.43 24.76 ± 3.34B 79.43 ± 2.98 68.84 ± 2.67 57.18 ± 5.12 48.00 ± 0.23 44.33 ± 0.23 38.61 ± 5.73

a A = Ambient room temperature (20–25 °C).b B = Refrigeration temperature (4–8 °C).

228 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

increases the drug release rate [47]. Thus, it can be concluded that theformulated SNEDDS improved the solubilization and release rate of NYS.

Fig. 4.Microbiological study of therapeutic efficacy of selected NYS-SNEDDS formulationsand NYS suspensions against oral candidiasis inmice. Data presented asmean± S.D. (n=10). Same letters in the same time interval are statistically non-significant different anddifferent letters in the same time interval are statistically significant different at P b 0.05.

3.7. Stability study

All investigated NYS-SNEDDSs were physically stable with no obvi-ous change in visual appearance i.e. no precipitation, phase separationor color change when being stored at ambient room temperature (20–25 °C) and refrigeration temperature (4–8 °C) for 6 months. The initialconcentrations (time 0) of NYS were referenced as 100% and all subse-quent concentrations were expressed as percentages of the initial con-centration. The stability of NYS was assessed by evaluating the percentchange in NYS concentrations from time 0. A decrease of N10% of theinitial concentration was considered to represent a significant loss ofdrug [90,91]. At ambient room temperature (with light protection) theconcentrations of NYS decreased below 70% of the initial value after15 days for all investigated formulations. On the other hand, at refriger-ation temperature, the concentrations of NYS decreased below 90% ofthe initial value after 15 days (Table 5). Thesefindings are in accordancewith previous reports [92–94], where NYS was found to be more stableat refrigeration temperature (up to 15 days) in contrast to ambientroom temperature (up to 4 days). It was clearly shown, that the formu-lations prepared using DMSO (F1 and F2) as the co-surfactant, revealedbetter stability profiles compared to their counterparts (F3–F5)prepared employing PG as the co-surfactant. This might be explainedby the higher solubilization power of DMSO toward NYS in contrast toPG. Also, it was obviously depicted that increasing amounts of Tw20 inthe Smix ratio (1:1, 2:1 and 3:1) has led to better NYS stability. Asreported in the droplet size analysis (Section 3.5.4), the nanomulsiondroplet size was influenced by surfactant concentration [67]. Increasein surfactant concentration produced smaller emulsion droplet size.Increased availability of the surfactant to adsorb around oil–water inter-face may decrease emulsion droplet size owing to decreased interfacialtension of the system [39]. Moreover, decrease in droplet size indicatesclose packed arrangement of surfactant at the oil–water interface andimproved stabilizing effect on oil droplets by forming a strongmechan-ical barrier to coalescence, which consequently led to higher entrappedNYS stability [68].

NYS is practically insoluble in water and is easily degraded by light,heat, oxygen, moisture, humidity and extreme pH values [93,94]. NYSshould be stored in airtight containers at a temperature of between 2and 8 °C and protected from light [92], and at 25 °C maximum afteropening for 7 days [16]. Some ready-to-use commercial NYS suspen-sions (100,000 IU/ml) containing preservatives have expiration datesof many weeks [15,94]. On the contrary, extemporaneously preparedoral suspensions of NYS powder should be used immediately aftermixing and should not be stored when this product contains no preser-vatives [15,94].

The explanation of the observed rate of degradation would requiredetailed kinetic studies, which were beyond the scope of this investiga-tion. The reaction mechanisms involved may be very complicated, asthere are six double bonds in the NYS molecule [93]. In conclusion, theprevious findings postulate the necessity for a detailed stability study,to obtain themaximumbenefit of drug delivery. The addition of preser-vatives and stabilizing agents to NYS-SNEDDSs is noticeably needed, inlight of the obtained data.

3.8. In vitro antifungal efficacy

3.8.1. Determination of minimum inhibitory concentration (MIC)The 2-fold serial dilution method was used to assess the in vitro

antifungal activity of NYS-SNEDDS and NYS suspensions againstC. albicans. SNEDDS excipients showed no inhibitory effect on theyeast grown, proving that the excipients did not interfere the test. Byassessing the MIC of the 5 selected NYS-SNEDDSs, only 2 formulations,namely F1 and F2, showed a comparable MIC (256 μg/μl) to that of themarketed NYS suspension. All other formulations (F3–F5), in additionto plain NYS suspension, showed higher MIC values. F1 and F2 containDMSO as co-surfactant in contrast to F3–F5 containing PG as the co-

229A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

surfactant. As stated previously, DMSO possesses higher drug solubilityand hence F1 and F2 revealed better release and stability profiles whichmight allow the system to provide better interaction with the fungi cellmembrane, and thus decrease theMIC value and increase the potency ofthe drug. The enhancement of the in vitro antifungal activity of NYS,when formulated in a lipid based formulation, compared with freeNYS, was previously reported [11,19,25,26], confirming the necessityto formulate NYS in lipid-based system to improve its efficacy.

3.8.2. Time-kill testMarketedNYS suspension showed the shortest time-kill values (less

than 10min at 256 μg/μl and less than 5min at 512 μg/μl) followedby F1and F2 (less than 20min at 256 μg/μl and less than 15min at 512 μg/μl).All other investigated formulations as well as plain NYS suspensionsrevealed more than 20 min time-kill values. Marketed NYS suspensionexhibited a much quicker effect, probably due to the direct contactwith the yeast on the culture medium [8].

Fig. 5.Microscopic observations of the tongues of mice with oral candidiasis on (day 8) after indegeneration of the filiform papillae and the keratinized spine (asterisk), hemorrhagic area (armice tongue showing large hemorrhagic areas (arrows) in the lamina propria. (C) Sectionpapillae (arrow). (D) Section of mice tongue treated with marketed NYS suspension showinfibers (asterisk). (E) Section of mice tongue treated with F1 showing slight disturbance of thtongue treated with F2 showing the filiform papillae (arrow); lamina propria (arrowhead) and

3.9. In vivo antifungal efficacy

The present study assesses the antifungal activity of selected NYS-SNEDDS formulations administered per-orally against oral candidiasiswith an immunosuppressed-mouse model, compared to marketedNYS suspension, containing the same quantity of drug, as well as plainNYS suspended in phosphate buffer. In thismodel, the severity of Candi-da infection was estimated both by measuring the number of viableCandida cells in the oral cavity and bymanifestation of histopathologicalfindings on the tongue.

3.9.1. Microbiological evaluation of the fungal burdenAt day 1 after oral inoculation of C. albicans (just before beginning of

treatment) sample swabswere taken frommicemouths and cultured toconfirm the presence of C. albicans and to quantify the number of CFUin the oral cavity. The results demonstrated in Fig. 4 confirmed theyeast presence with a non-significant difference (P N 0.05) between alltreatment groups. At day 3; F1, F2 and marketed NYS suspension

fection stained with HE with scale bar 20 μm. (A) Section of control mice tongue showingrow) and inflammatory infiltration (arrowhead) in lamina propria. (B) Section of controlof mice tongue treated with plain NYS suspension showing dislocation of the filiformg the filiform papillae (arrow); lamina propria (arrowhead); and transverse musculare filiform papillae and overlapping of the keratinized spine (arrow). (F) Section of micetransverse muscular fibers (asterisk).

230 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

significantly (P b 0.05) reduced the oral burden of C. albicans comparedto the plain NYS suspension aswell as the control group (Fig. 4). Indeed,F1 and F2 showed a significant reduction (P b 0.05) in oral C. albicansburden, in contrast to marketed NYS suspension, at day 5 of treatment.Afterwards, both treatments (F1 and F2)were statistically superior to allother treatment groups. Of particular interest, the NYS-SNEDDS formu-lations were significantly more efficacious than the marketed NYSsuspension in reducing the oral burden of C. albicans on treatmentdays, 5 and 7 (Fig. 4). Mice were assessed for survival during the treat-ment period, where no mortality was recorded.

Both F1 and F2 showed the highest release rate, long-term stabilityand in vitro antifungal efficacy compared to F3–F5. As discussed previ-ously, the presence of DMSO as co-surfactant in F1 and F2, led to highersolubilization and hence better physico-chemical properties wasreported. In addition, formed nanoemulsions possessed small dropletsize (b100 nm), which is one of the most important characteristics ofemulsion for stability evaluation and a critical step in the pathway of en-hancing drug bioavailability [33,38]. Smaller droplet size leads to largerinterfacial surface area for drug absorption and hence may lead to morerapid absorption and improve bioavailability. Furthermore, these partic-ulate delivery systems are reported to prolong residence time onmuco-sal membranes [41] and they could reach greatermucosal surface areas,resulting in a comparatively higher drug uptake [40,42].

3.9.2. Histopathological findingsUntreated (infected) control mice as well as mice treated with plain

NYS suspension consistently showed significant histopathologic chang-es of the stratified squamous epithelium of the entire dorsal area of thetongue such as abundant inflammatory cell infiltrates, hemorrhagicareas, degeneration of the filiform papillae, intercellular edema, and de-tachment of the cell layers, leading to disorganization of tissue structure(Fig. 5a–c). These histolopathologic changes may be directly related tothe lesions induced by Candida colonization in the dorsal epithelium[95]. In contrast, in mice treated with either marketed NYS suspensionor the two selected NYS-SNEDDS formulations, the histolopathologicchanges in the epithelium were sharply reduced, and, in some areas,changes are entirely absent (Fig. 5d, e). Itwas not possible to distinguishmarketed NYS suspension from the NYS-SNEDDS formulations due tothe qualitative nature of histopathology investigation.

4. Conclusions

In the present study, self-nanoemulsifying drug delivery systems ofnystatin were prepared, and in vitro and in vivo evaluated. Followingoptimization, five nystatin SNEDDS formulations, composed of oleicacid as the oily phase, Tween 20 as surfactant, dimethyl sulfoxide orpropylene glycol as co-surfactants, were selected. The formulationswere robust to dilution in different media using different dilutionfolds, exhibiting no signs of precipitation or separation. The preparedSNEDDSs showed cloud point of 90 °C and their electrical conductivityincreased as the water content increased. The globule size was foundto be in the nanometric range (b100 nm) and the zeta potential rangedfrom −15.3 to −23.9 mV indicating good stability. The morphologyinvestigation revealed spherical shaped globules using transmissionelectron microscopy. In vitro release profiles in methanol/DMF/water(55:15:30, v/v/v) depicted a gradual release pattern with 100% releaseafter 48 h. Nystatin loaded SNEDDS formulations showed a significantincrease in release rate compared to the plain drug suspension underthe same conditions. Long-term stability revealed that the formulationswere physically stable with no obvious change in visual appearance buta remarkable decay of nystatin after 6 months storage at ambient roomand refrigeration temperatures was reported. In vitro antifungal efficacydepicted the superiority of F1 and F2 NYS-SNEDDSs containing oleicacid, Tw20/DMSO (2:1) and (3:1), respectively. In vivo antifungal effica-cy in immunosuppressedmicewith oral candidiasis showed supremacyof F1 and F2 compared to marketed NYS suspension after a 7-day

treatment period as well as improved histopathological findings. Inconclusion, solubilization and nano-scaling by means of SNEDDSprovided a promising approach for increasing the efficacy of topicallyadministered NYS in the treatment of OPC. This novel application ofnano-emulsification could lead to the development of improved formu-lations of other antimicrobial agents.

Declaration of interest

The authors report no declarations of interest.

Acknowledgments

The authors extend their appreciation to the Project's Sector at theNational Research Centre, Egypt for funding this work through the re-search group project fund no. 10070107.

The authors would like to thank Teaching Assistant Salma Hussein(Microbiology and Immunology Department, Faculty of Pharmacy,Cairo University) for her great effort and collaboration in conductingthe in vivo microbiological evaluation study.

References

[1] S. Pal, S. Nagy, T. Bozo, B. Kocsis, A. Devay, Technological and biopharmaceutical op-timization of nystatin release from a multiparticulate based bioadhesive drug deliv-ery system, Eur. J. Pharm. Sci. 49 (2013) 258–264.

[2] L.L. Patton, A.J. Bonito, D.A. Shugars, A systematic review of the effectiveness of an-tifungal drugs for the prevention and treatment of oropharyngeal candidiasis inHIV-positive patients, Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 92(2001) 170–179.

[3] A. Akpan, R. Morgan, Oral candidiasis, Postgrad. Med. J. 78 (2002) 455–459.[4] H. Gonzalez Gravina, E. Gonzalez de Moran, O. Zambrano, M. Lozano Chourio, S.

Rodriguez de Valero, S. Robertis, L. Mesa, Oral candidiasis in children and adoles-cents with cancer. Identification of Candida spp, Med. Oral Patol. Oral Cir. Bucal 12(2007) E419–E423.

[5] A.Melkoumov, M. Goupil, F. Louhichi, M. Raymond, L. de Repentigny, G. Leclair, Nys-tatin nanosizing enhances in vitro and in vivo antifungal activity against Candidaalbicans, J. Antimicrob. Chemother. 68 (2013) 2099–2105.

[6] B. Spellberg, Novel insights into disseminated candidiasis: pathogenesis researchand clinical experience converge, PLoS Pathog. 4 (2008), e38.

[7] R. Lopez-Martinez, Candidosis, a new challenge, Clin. Dermatol. 28 (2010) 178–184.[8] M.J. Martin, A.C. Calpena, F. Fernandez, M. Mallandrich, P. Galvez, B. Clares, Develop-

ment of alginate microspheres as nystatin carriers for oral mucosa drug delivery,Carbohydr. Polym. 117 (2015) 140–149.

[9] J.A. Ship, A. Vissink, S.J. Challacombe, Use of prophylactic antifungals in the immuno-compromised host, Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 103 (2007)(Suppl., S6 e1–14).

[10] P.G. Pappas, Invasive candidiasis, Infect. Dis. Clin. N. Am. 20 (2006) 485–506.[11] D. Marin-Quintero, F. Fernandez-Campos, A.C. Calpena-Campmany, M.J. Montes-

Lopez, B. Clares-Naveros, A. Del Pozo-Carrascosa, Formulation design and optimiza-tion for the improvement of nystatin-loaded lipid intravenous emulsion, J. Pharm.Sci. 102 (2013) 4015–4023.

[12] A. Coutinho, M. Prieto, Cooperative partition model of nystatin interaction withphospholipid vesicles, Biophys. J. 84 (2003) 3061–3078.

[13] A.N. Ellepola, L.P. Samaranayake, The in vitro post-antifungal effect of nystatin onCandida species of oral origin, J. Oral Pathol. Med. 28 (1999) 112–116.

[14] Nystatin, In: British Pharmacopoeia 2007, pp. 1497–1498.[15] G.K. Mc Evoy, Nystatin, In: Drug Information, American Society of Health-System

Pharmacists, American Hospital Formulary Service (AHFS) 2004, pp. 528–529.[16] S.C. Sweetman, Nystatin, In: Martindale, The Complete Drug Reference, Pharmaceu-

tical Press, London, Chicago 2009, pp. 543–544.[17] I.P. Kaur, S. Kakkar, Topical delivery of antifungal agents, Expert Opin. Drug Deliv. 7

(2010) 1303–1327.[18] S.R. Croy, G.S. Kwon, The effects of pluronic block copolymers on the aggregation

state of nystatin, J. Control. Release 95 (2004) 161–171.[19] F. Fernandez-Campos, B. Clares Naveros, O. Lopez Serrano, C. Alonso Merino, A.C.

Calpena Campmany, Evaluation of novel nystatin nanoemulsion for skin candidosisinfections, Mycoses 56 (2012) 70–81.

[20] J. Bolard, P. Legrand, F. Heitz, B. Cybulska, One-sided action of amphotericin B oncholesterol-containing membranes is determined by its self-association in the me-dium, Biochemistry 30 (1991) 5707–5715.

[21] J.M. Llabot, R.H. Manzo, D.A. Allemandi, Double-layered mucoadhesive tablets con-taining nystatin, AAPS PharmSciTech 3 (2002), E22.

[22] K. Sakeer, H. Al-Zein, I. Hassan, S. Desai, A. Nokhodchi, Enhancement of dissolutionof nystatin from buccoadhesive tablets containing various surfactants and a soliddispersion formulation, Arch. Pharm. Res. 33 (2010) 1771–1779.

[23] A.W. Ng, K.M. Wasan, G. Lopez-Berestein, Development of liposomal polyene antibi-otics: an historical perspective, J. Pharm. Pharm. Sci. 6 (2003) 67–83.

231A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

[24] M.S. El-Ridy, A. Abdelbary, T. Essam, R.M. El-Salam, A.A. Kassem, Niosomes as a po-tential drug delivery system for increasing the efficacy and safety of nystatin, DrugDev. Ind. Pharm. 37 (2011) 1491–1508.

[25] F. Fernandez-Campos, A.C. Calpena Campmany, G. Rodriguez Delgado, O. LopezSerrano, B. Clares Naveros, Development and characterization of a novel nystatin-loaded nanoemulsion for the buccal treatment of candidosis: ultrastructural effectsand release studies, J. Pharm. Sci. 101 (2012) 3739–3752.

[26] R. Semis, I. Polacheck, E. Segal, Nystatin-intralipid preparation: characterization andin vitro activity against yeasts and molds, Mycopathologia 169 (2010) 333–341.

[27] M.J. Martin-Villena, F. Fernandez-Campos, A.C. Calpena-Campmany, N. Bozal-deFebrer, M.A. Ruiz-Martinez, B. Clares-Naveros, Novel microparticulate systems forthe vaginal delivery of nystatin: development and characterization, Carbohydr.Polym. 94 (2013) 1–11.

[28] R. Khalil, M. Kassem, A. Abdelbary, M. El-Ridy, M. AbouSamra, Preparation and char-acterization of nystatin-loaded solid lipid nanoparticles for topical delivery, Int. J.Pharm. Sci. Res. 4 (2013) 2292–2300.

[29] R.M. Khalil, A. Abdelbary, M.A. Kassem, M.S. El-Ridy, M.M. Abou Samra, G.E.A. Awad,S.S. Mansy, Formulation and characterization of nystatin-loaded nanostructuredlipid carriers for topical delivery against cutaneous candidiasis, Br. J. Pharm. Res. 4(2014) 490–512.

[30] D. Sakloetsakun, S. Dunnhaupt, J. Barthelmes, G. Perera, A. Bernkop-Schnurch, Com-bining two technologies: multifunctional polymers and self-nanoemulsifying drugdelivery system (SNEDDS) for oral insulin administration, Int. J. Biol. Macromol. 61(2013) 363–372.

[31] N. Parmar, N. Singla, S. Amin, K. Kohli, Study of cosurfactant effect onnanoemulsifying area and development of lercanidipine loaded (SNEDDS) selfnanoemulsifying drug delivery system, Colloids Surf. B: Biointerfaces 86 (2011)327–338.

[32] P.P. Constantinides, Lipidmicroemulsions for improving drug dissolution and oral ab-sorption: physical and biopharmaceutical aspects, Pharm. Res. 12 (1995) 1561–1572.

[33] K. Balakumar, C.V. Raghavan, N.T. Selvan, R.H. Prasad, S. Abdu, Self nanoemulsifyingdrug delivery system (SNEDDS) of rosuvastatin calcium: design, formulation, bio-availability and pharmacokinetic evaluation, Colloids Surf. B: Biointerfaces 112(2013) 337–343.

[34] N. Heshmati, X. Cheng, G. Eisenbrand, G. Fricker, Enhancement of oral bioavailabilityof E804 by self-nanoemulsifying drug delivery system (SNEDDS) in rats, J. Pharm.Sci. 102 (2013) 3792–3799.

[35] K. Kohli, S. Chopra, D. Dhar, S. Arora, R.K. Khar, Self-emulsifying drug delivery systems:an approach to enhance oral bioavailability, Drug Discov. Today 15 (2010) 958–965.

[36] N. Thomas, R. Holm, M. Garmer, J.J. Karlsson, A. Mullertz, T. Rades, Supersaturatedself-nanoemulsifying drug delivery systems (Super-SNEDDS) enhance the bioavail-ability of the poorly water-soluble drug simvastatin in dogs, AAPS J. 15 (2013)219–227.

[37] S. Chakraborty, D. Shukla, B. Mishra, S. Singh, Lipid—an emerging platform for oraldelivery of drugs with poor bioavailability, Eur. J. Pharm. Biopharm. 73 (2009) 1–15.

[38] J. Xi, Q. Chang, C.K. Chan, Z.Y. Meng, G.N. Wang, J.B. Sun, Y.T. Wang, H.H. Tong, Y.Zheng, Formulation development and bioavailability evaluation of a self-nanoemulsified drug delivery system of oleanolic acid, AAPS PharmSciTech 10(2009) 172–182.

[39] L. Wang, J. Dong, J. Chen, J. Eastoe, X. Li, Design and optimization of a new self-nanoemulsifying drug delivery system, J. Colloid Interface Sci. 330 (2009) 443–448.

[40] R.N. Gursoy, S. Benita, Self-emulsifying drug delivery systems (SEDDS) for improvedoral delivery of lipophilic drugs, Biomed. Pharmacother. 58 (2004) 173–182.

[41] P.P. Constantinides, J.P. Scalart, C. Lancaster, J. Marcello, G. Marks, H. Ellens, P.L.Smith, Formulation and intestinal absorption enhancement evaluation of water-in-oil microemulsions incorporating medium-chain glycerides, Pharm. Res. 11(1994) 1385–1390.

[42] A. Bernkop-Schnurch, Nanocarrier systems for oral drug delivery: do we really needthem? Eur. J. Pharm. Sci. 49 (2013) 272–277.

[43] U.A. Fahmy, O.A. Ahmed, K.M. Hosny, Development and evaluation of avanafil self-nanoemulsifying drug delivery systemwith rapid onset of action and enhanced bio-availability, AAPS PharmSciTech 16 (2015) 53–58.

[44] A.A. Date, M.S. Nagarsenker, Design and evaluation of self-nanoemulsifying drug de-livery systems (SNEDDS) for cefpodoxime proxetil, Int. J. Pharm. 329 (2007) 166–172.

[45] Y.S. Elnaggar, M.A. El-Massik, O.Y. Abdallah, Self-nanoemulsifying drug delivery sys-temsof tamoxifen citrate: designandoptimization, Int. J. Pharm. 380 (2009) 133–141.

[46] E.B. Basalious, N. Shawky, S.M. Badr-Eldin, SNEDDS containing bioenhancers for im-provement of dissolution and oral absorption of lacidipine. I: Development and op-timization, Int. J. Pharm. 391 (2010) 203–211.

[47] S. Beg, S. Swain, H.P. Singh, N. Patra Ch, M.E. Rao, Development, optimization, andcharacterization of solid self-nanoemulsifying drug delivery systems of valsartanusing porous carriers, AAPS PharmSciTech 13 (2012) 1416–1427.

[48] F. Shakeel, N. Haqa, F.K. Alanazi, I.A. Alsarra, Polymeric solid self-nanoemulsifyingdrug delivery system of glibenclamide using coffee husk as a low cost biosorbent,Powder Technol. 256 (2014) 352–360.

[49] A.G. Agrawal, A. Kumar, P.S. Gide, Formulation of solid self-nanoemulsifying drugdelivery systems using N-methyl pyrrolidone as cosolvent, Drug Dev. Ind. Pharm.41 (2015) 594–604.

[50] P. Zhang, Y. Liu, N. Feng, J. Xu, Preparation and evaluation of self-microemulsifyingdrug delivery system of oridonin, Int. J. Pharm. 355 (2008) 269–276.

[51] M. El-Badry, N. Haq, G. Fetih, F. Shakeel, Solubility and dissolution enhancement oftadalafil using self-nanoemulsifying drug delivery system, J. Oleo Sci. 63 (2014)567–576.

[52] S. Zargar-Shoshtari, J. Wen, R.G. Alany, Formulation and physicochemical character-ization of imwitor 308 based self microemulsifying drug delivery systems, Chem.Pharm. Bull. 58 (2010) 1332–1338.

[53] K.M. Hosny, Z.M. Banjar, The formulation of a nasal nanoemulsion zaleplon in situgel for the treatment of insomnia, Expert Opin. Drug Deliv. 10 (2013) 1033–1041.

[54] A.K. Singh, A. Chaurasiya, A. Awasthi, G. Mishra, D. Asati, R.K. Khar, R. Mukherjee,Oral bioavailability enhancement of exemestane from self-microemulsifying drugdelivery system (SMEDDS), AAPS PharmSciTech 10 (2009) 906–916.

[55] R. Yang, X. Huang, J. Dou, G. Zhai, L. Su, Self-microemulsifying drug delivery systemfor improved oral bioavailability of oleanolic acid: design and evaluation, Int. J.Nanomedicine 8 (2013) 2917–2926.

[56] C.S. Chaw, K.W. Chooi, X.M. Liu, C.W. Tan, L. Wang, Y.Y. Yang, Thermally responsivecore-shell nanoparticles self-assembled from cholesteryl end-capped and graftedpolyacrylamides: drug incorporation and in vitro release, Biomaterials 25 (2004)4297–4308.

[57] J.M. Llabot, R.H. Manzo, D.A. Allemandi, Novel mucoadhesive extended release tab-lets for treatment of oral candidosis: “in vivo” evaluation of the biopharmaceuticalperformance, J. Pharm. Sci. 98 (2009) 1871–1876.

[58] K. Sakeer, H. Al-Zein, I. Hassan, G.P. Martin, A. Nokhodchi, Use of xanthan and its bi-nary blends with synthetic polymers to design controlled release formulations ofbuccoadhesive nystatin tablets, Pharm. Dev. Technol. 15 (2010) 360–368.

[59] J.M. Andrews, Determination of minimum inhibitory concentrations, J. Antimicrob.Chemother. 48 (Suppl. 1) (2001) 5–16.

[60] C.J. Clancy, H. Huang, S. Cheng, H. Derendorf, M.H. Nguyen, Characterizing the effectsof caspofungin on Candida albicans, Candida parapsilosis, and Candida glabrata iso-lates by simultaneous time-kill and postantifungal-effect experiments, Antimicrob.Agents Chemother. 50 (2006) 2569–2572.

[61] N. Takakura, H. Wakabayashi, H. Ishibashi, S. Teraguchi, Y. Tamura, H. Yamaguchi, S.Abe, Oral lactoferrin treatment of experimental oral candidiasis in mice, Antimicrob.Agents Chemother. 47 (2003) 2619–2623.

[62] K. Ninomiya, N. Maruyama, S. Inoue, H. Ishibashi, T. Takizawa, H. Oshima, S. Abe, Theessential oil ofMelaleuca alternifolia (tea tree oil) and its main component, terpinen-4-ol protect mice from experimental oral candidiasis, Biol. Pharm. Bull. 35 (2012)861–865.

[63] M.M. Badran, E.I. Taha, M.M. Tayel, S.A. Al-Suwayeh, Ultra-fine self nanoemulsifyingdrug delivery system for transdermal delivery of meloxicam: dependency on thetype of surfactants, J. Mol. Liq. 190 (2014) 16–22.

[64] X. Qi, L. Wang, J. Zhu, Z. Hu, J. Zhang, Self-double-emulsifying drug delivery system(SDEDDS): a newway for oral delivery of drugs with high solubility and low perme-ability, Int. J. Pharm. 409 (2011) 245–251.

[65] Y. Onuki, M. Morishita, K. Takayama, Formulation optimization of water-in-oil–water multiple emulsion for intestinal insulin delivery, J. Control. Release 97(2004) 91–99.

[66] M. Verbrugghe, E. Cocquyt, P. Saveyn, P. Sabatino, D. Sinnaeve, J.C. Martins, P. Vander Meeren, Quantification of hydrophilic ethoxylates in polysorbate surfactantsusing diffusion H1 NMR spectroscopy, J. Pharm. Biomed. Anal. 51 (2010) 583–589.

[67] P.R. Nepal, H.K. Han, H.K. Choi, Preparation and in vitro–in vivo evaluation ofWitepsol H35 based self-nanoemulsifying drug delivery systems (SNEDDS) of coen-zyme Q(10), Eur. J. Pharm. Sci. 39 (2010) 224–232.

[68] T.R. Kommuru, B. Gurley, M.A. Khan, I.K. Reddy, Self-emulsifying drug delivery sys-tems (SEDDS) of coenzyme Q10: formulation development and bioavailability as-sessment, Int. J. Pharm. 212 (2001) 233–246.

[69] S. Shafiq, F. Shakeel, S. Talegaonkar, F.J. Ahmad, R.K. Khar, M. Ali, Development andbioavailability assessment of ramipril nanoemulsion formulation, Eur. J. Pharm.Biopharm. 66 (2007) 227–243.

[70] C.W. Pouton, C.J. Porter, Formulation of lipid-based delivery systems for oral adminis-tration: materials, methods and strategies, Adv. Drug Deliv. Rev. 60 (2008) 625–637.

[71] Y. Zhao, C. Wang, A.H. Chow, K. Ren, T. Gong, Z. Zhang, Y. Zheng, Self-nanoemulsifying drug delivery system (SNEDDS) for oral delivery of zedoary essen-tial oil: formulation and bioavailability studies, Int. J. Pharm. 383 (2010) 170–177.

[72] M. Rizwan, M. Aqil, A. Azeem, S. Talegaonkar, Y. Sultana, A. Ali, Enhanced transder-mal delivery of carvedilol using nanoemulsion as a vehicle, J. Exp. Nanosci. 5 (2010)390–411.

[73] M.J. Lawrence, G.D. Rees, Microemulsion-based media as novel drug delivery sys-tems, Adv. Drug Deliv. Rev. 45 (2000) 89–121.

[74] A.R. Dixit, S.J. Rajput, S.G. Patel, Preparation and bioavailability assessment ofSMEDDS containing valsartan, AAPS PharmSciTech 11 (2010) 314–321.

[75] C.W. Pouton, Formulation of self emulsifying drug delivery systems, Adv. Drug Deliv.Rev. 25 (1997) 47–58.

[76] M.J. Rang, C.A. Miller, Spontaneous emulsification of oils containing hydrocarbon,nonionic surfactant, and oleyl alcohol, J. Colloid Interface Sci. 209 (1999) 179–192.

[77] S. Shafiq, F. Shakeel, S. Talegaonkar, F.J. Ahmad, R.K. Khar, M. Ali, Design and devel-opment of oral oil in water ramipril nanoemulsion formulation: in vitro and in vivoassessment, J. Biomed. Nanotechnol. 3 (2007) 28–44.

[78] F. Shakeel, N. Haq, M. Elbadry, F.K. Alanazi, I.A. Alsarra, Ultra fine super self-nanoemulsifying drug delivery system (SNEDDS) enhanced solubility and dissolu-tion of indomethacin, J. Mol. Liq. 180 (2013) 89–94.

[79] S.K. Singh, P.R. Verma, B. Razdan, Glibenclamide-loaded self-nanoemulsifying drugdelivery system: development and characterization, Drug Dev. Ind. Pharm. 36(2010) 933–945.

[80] S. Kanuganti, R. Jukanti, P.R. Veerareddy, S. Bandari, Paliperidone-loaded self-emulsifying drug delivery systems (SEDDS) for improved oral delivery, J. Dispers.Sci. Technol. 33 (2012) 506–515.

[81] C.W. Pouton, Lipid formulations for oral administration of drugs: non-emulsifying,self-emulsifying and ‘self-microemulsifying’ drug delivery systems, Eur. J. Pharm.Sci. 11 (Suppl. 2) (2000) S93–S98.

[82] K. Itoh, Y. Tozuka, T. Oguchi, K. Yamamoto, Improvement of physicochemical prop-erties of N-4472 part I formulation design by using self-microemulsifying system,Int. J. Pharm. 238 (2002) 153–160.

232 A.A. Kassem et al. / Journal of Molecular Liquids 218 (2016) 219–232

[83] R.M. Hathout, T.J. Woodman, S. Mansour, N.D. Mortada, A.S. Geneidi, R.H. Guy,Microemulsion formulations for the transdermal delivery of testosterone, Eur. J.Pharm. Sci. 40 (2010) 188–196.

[84] R.C. Baker, A.T. Florence, R.H. Ottewill, T.F. Tadros, Investigations into the formationand characterization of microemulsions. II. Light scattering conductivity and viscos-ity studies of microemulsions, J. Colloid Interface Sci. 100 (1984) 332–349.

[85] A.K. Singh, A. Chaurasiya, M. Singh, S.C. Upadhyay, R. Mukherjee, R.K. Khar,Exemestane loaded self-microemulsifying drug delivery system (SMEDDS): devel-opment and optimization, AAPS PharmSciTech 9 (2008) 628–634.

[86] N.H. Shah, M.T. Carvagal, C.I. Patel, M.H. Infeld, A.W. Malick, Self-emulsifyingdrug delivery systems (SEDDS) with polyglycolyzed glycerides for improvingin vitro dissolution and oral absorption of lipophilic drugs, Int. J. Pharm. 106(1994) 15–23.

[87] K.O. Choi, N.P. Aditya, S. Ko, Effect of aqueous pH and electrolyte concentration onstructure, stability and flow behavior of non-ionic surfactant based solid lipid nano-particles, Food Chem. 147 (2014) 239–244.

[88] W. Wu, Y. Wang, L. Que, Enhanced bioavailability of silymarin by self-microemulsifying drug delivery system, Eur. J. Pharm. Biopharm. 63 (2006)288–294.

[89] N. Thomas, A. Mullertz, A. Graf, T. Rades, Influence of lipid composition and drugload on the in vitro performance of self-nanoemulsifying drug delivery systems, J.Pharm. Sci. 101 (2012) 1721–1731.

[90] R.M. Lopez, A. Ayestaran, L. Pou, J.B. Montoro, M. Hernandez, I. Caragol, Stability ofamphotericin B in an extemporaneously prepared i.v. fat emulsion, Am. J. HealthSyst. Pharm. 53 (1996) 2724–2727.

[91] P.J. Dentinger, C.F. Swenson, N.H. Anaizi, Stability of amphotericin B in an extemporane-ously compounded oral suspension, Am. J. Health Syst. Pharm. 58 (2001) 1021–1024.

[92] J. Groeschke, I. Solassol, F. Bressolle, F. Pinguet, Stability of amphotericin B and nys-tatin in antifungal mouthrinses containing sodium hydrogen carbonate, J. Pharm.Biomed. Anal. 42 (2006) 362–366.

[93] H. van Doorne, E.H. Bosch, Stability and in vitro activity of nystatin and its y-cyclodextrin complex against Candida albicans, Int. J. Pharm. 73 (1991) 43–49.

[94] N. Vermerie, C. Malbrunot, M. Azar, P. Arnaud, Stability of nystatin in mouthrinses;effect of pH temperature, concentration and colloidal silver addition, studied usingan in vitro antifungal activity, Pharm. World Sci. 19 (1997) 197–201.

[95] A.Martinez, J. Regadera, E. Jimenez, I. Santos, D. Gargallo-Viola, Antifungal efficacy ofGM237354, a sordarin derivative, in experimental oral candidiasis in immunosup-pressed rats, Antimicrob. Agents Chemother. 45 (2001) 1008–1013.