Lecithin-linker formulations for self-emulsifying delivery of nutraceuticals

1

2

3Q1

4

5

6

Q2

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

International Journal of Pharmaceutics xxx (2014) xxx–xxx

G Model

IJP 14058 1–11

Lecithin-linker formulations for self-emulsifying delivery ofnutraceuticals

Jacquelene Chu a, Yu-Ling Cheng a, A.Venketeshwer Rao b, Mehdi Nouraei a,Silvia Zarate-Muñoz a, Edgar J. Acosta a,*aUniversity of Toronto, Department of Chemical Engineering and Applied Chemistry, 200 College Street, Room 131, Toronto, Ontario M5S 3E5, CanadabUniversity of Toronto, Department of Nutritional Sciences, 150 College Street, Room 315, Toronto, Ontario M5S 3E2, Canada

A R T I C L E I N F O

Article history:Received 18 February 2013Received in revised form 24 March 2014Accepted 2 May 2014Available online xxx

Keywords:MicroemulsionOral deliverySEDSSMEDSPhase behaviorLinkersLecithin

A B S T R A C T

Lecithin-linker microemulsions are formulations produced with soybean lecithin in combination with ahighly lipophilic (lipophilic linker) and highly hydrophilic (hydrophilic linkers) surfactant-like additives.In this work, lecithin-linker systems were formulated to produce self-emulsifying delivery systems forb-carotene and b-sitosterol. The concentration of the lipophilic linker, sorbitan monooleate, wasadjusted to minimize the formation of liquid crystals. The concentration of hydrophilic linkers,decaglyceryl caprylate/caprate and PEG-6-caprylic/capric glycerides, was gradually increased (scanned)until single phase clear microemulsions were obtained. For these scans, the oil (ethyl caprate) to waterratio was set to 1. The single phase, clear microemulsions were diluted with fed-state simulated intestinalfluid (FeSSIF) and produced stable emulsions, with drop sizes close to 200 nm. Using pseudo-ternaryphase diagrams to evaluate the process of dilution of microemulsion preconcentrates (mixtures of oil,lecithin and linkers with little or no water) with FeSSIF, it was determined that self-emulsifying systemsare obtained when the early stages of the dilution produce single phase microemulsions. If liquid crystalsor multiple phase systems are obtained during those early stages, then the emulsification yields unstableemulsions with large drop sizes. An in vitro permeability study conducted using a Flow-Thru Dialyzerrevealed that stable emulsions with drop sizes of 150–300 nm produce large and irreversible permeationof b-carotene to sheep intestine. On the other hand, unstable emulsions produced without the linkercombination separated in the dialyzer chamber.

ã 2014 Published by Elsevier B.V.

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsev ier .com/locate / i jpharm

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

1. Introduction

Microemulsions are one type of nano-sized oral deliveryvehicles capable of enhancing the bioavailability of poorly solubledrugs (Lawrence and Rees, 2012). These systems enhance thebioavailability of hydrophobic drugs by reducing drop size,increasing the solubilization of the active ingredient, and increas-ing the residence time of the drug in the intestine (Acosta et al.,2011a; McClements, 2012). Microemulsions have oil and/or waternano-domains that exist in thermodynamic equilibrium due to theadsorption of surfactants at the oil/water interface. Microemulsioncan be composed of oil nano-domains solubilized in micelles(water continuous); or water nano-domains solubilized in reversemicelles (oil continuous); or they can be composed of coexisting oiland water nano-domains in bicontinuous systems (Kesisoglouet al., 2007; Rosen, 2004; Huang et al., 2010; McClements, 2012).

38

39

40

41* Corresponding author. Tel.: +1 416 946 0742; fax: +1 416 978 8605.E-mail address: [email protected] (E.J. Acosta).

http://dx.doi.org/10.1016/j.ijpharm.2014.05.0010378-5173/ã 2014 Published by Elsevier B.V.

Please cite this article in press as: Chu, J., et al., Lecithin-linker formulati(2014), http://dx.doi.org/10.1016/j.ijpharm.2014.05.001

Winsor used the terms Type I, II and III to identify oil-in-water,water-in-oil, and bicontinuous “saturated” microemulsions, re-spectively, that coexisted with an excess phase (oil for Type I, waterfor Type II, and oil and water for Type III) (Salager et al., 2005). Thetransition from oil to water continuous microemulsions can beaccomplished by changing the formulation conditions, for exam-ple, increasing the hydrophilicity of the surfactant or surfactantmixture (Salager et al., 2005). One characteristic property ofsaturated microemulsions near or at the phase inversion point isthat the interfacial tension with the excess phase is ultralow(10�2–10�5mN/m) (Salager et al., 2005). For single phase micro-emulsions (no excess phases), the transition from oil to watercontinuity can also be accomplished by increasing the water/oilratio. For these unsaturated microemulsions, the phase inversion iscaptured by changes in electrical conductivity, self-diffusioncoefficients (NMR techniques), and micelle/reserve micelle mor-phology (neutron or X-ray scattering) (Garti et al., 2004; Datemaet al., 1992; Fisher et al., 2013).

Self-emulsifying delivery systems (SEDS) represent a form ofdelivery systems that are related to microemulsion systems.

ons for self-emulsifying delivery of nutraceuticals, Int J Pharmaceut

mailto:[email protected]

http://dx.doi.org/10.1016/j.ijpharm.2014.05.001



http://www.sciencedirect.com/science/journal/03785173

www.elsevier.com/locate/ijpharm

42 SE43 ta44 w45 di46 2047 of48 2049

50 fo51 seQ3

52 Am53 ha54 of55 et56 ph57 is

58 its59 ca60 pr61 ge62 in63 sh64 oi65

66 th67 pr68 or69 2070 lo71 al72 hy73 co74 na75 po76 m77 an78 lip79 th80 ba81 lik82 et

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117Q4

118

119

120

2 J. Chu et al. / International Journal of Pharmaceutics xxx (2014) xxx–xxx

G Model

IJP 14058 1–11

DS are ideally isotropic, water-free mixtures of oils, surfac-nts, or co-solvents/surfactants that emulsify to form fine oil inater emulsions or microemulsions (SMEDS) upon aqueouslution and gentle agitation (Pouton, 1997; Gursoy and Benita,04). These systems have been used to enhance the absorption

lipophilic drugs (Narang et al., 2007; Pouton, 2000; Acosta,09).In addition to sufficient drug absorption, oral delivery

rmulations should be safe for consumption. Despite the limitedlection of food-grade ingredients (Acosta, 2009; Garti and Yuli-ar, 2008; Calderon et al., 2010), biocompatible microemulsionsve been developed using triglycerides, terpenes, and ethyl esters

fatty acids as the oil phase (Gupta et al., 2006; Von Corswant al., 1997; Gupta and Moulik, 2008; Torchilln, 2008). The use ofospholipids (lecithin mainly) as the main surface active materialalso desirable due to lecithin’s GRAS status (21 CFR 184.14), and

long alkyl tail (C16–C18) that provides greater oil solubilizationpacity (Acosta et al., 2005). Unfortunately, lecithin tends tooduce liquid crystals that promote the formation of undesirablels and emulsions. This limitation has been overcome with thetroduction of linker molecules in lecithin microemulsions. Fig. 1ows a schematic of the surfactant and linker assembly at thel/aqueous interface (Acosta et al., 2005).Linker molecules can be described as unbalanced surfactantsat segregate or adsorb at the oil/water interface, however, theirimary interaction is with either the oil phase (lipophilic linkers)

the aqueous phase (hydrophilic linkers) (Acosta et al., 2003,04, 2005, 2011b; Sabatini et al., 2003). A lipophilic linker has ang (C10+) hydrophobe (tail) and a weak hydrophile such as ancohol (R��OH) group or one or two ethylene oxide groups. Adrophilic linker has a short hydrophobe (typically C6–C9) but anventional surfactant hydrophile such as carboxylates, sulfo-tes, polyethylene glycol (more than 3 ethylene oxide groups),ly-glucoside, poly-glycerols, etc. (Sabatini et al., 2003). Althoughost of the literature does not differentiate between surfactantsd linkers, it has been shown that combination of hydrophilic andophilic linkers alone cannot produce microemulsions becauseey do not associate with each other. The linkers associate with alanced surfactant (lecithin in this case), producing a “zipper”-e self-assembly of hydrophilic and lipophilic linkers (Acosta

al., 2004).

Fig. 1. Schematic of the linker e

Please cite this article in press as: Chu, J., et al., Lecithin-linker formula(2014), http://dx.doi.org/10.1016/j.ijpharm.2014.05.001

In this work, we hypothesize that linker-based lecithin micro-emulsions can be formulated as oral drug delivery vehicles forhydrophobic drugs using food and pharmaceutical grade ingre-dients, using the lecithin-linker formulations of Yuan et al. (2008,2010) as starting point. The transdermal formula of Yuan consistedof lecithin, isopropyl myristate as the oil, sorbitan monooleate aslipophilic linker, and caprylic acid and sodium caprylate ashydrophilic linkers. In order to avoid pH sensitivity, the caprylicacid and its salt were substituted with decaglycerylcaprylate/caprate (food additive) and PEG-6-caprylic/capric glyc-erides (pharmaceutical grade). A similar mixture of non-ionichydrophilic linkers was used by Xuan et al. (2012). Isopropylmyristate (IPM) was replaced with ethyl caprate, a food-grade oil.

To evaluate the linker-based SEDS as oral delivery vehicles,b-sitosterol and b-carotene, both lipophilic nutraceuticals, wereselected as model drugs. Their chemical structures are shown inFig. 2. b-Sitosterol is a plant sterol that has low water solubility,and that slows or inhibits the incorporation of dietary and biliarycholesterol (Rozner et al., 2007; Rozner and Garti, 2006).b-Carotene is a carotenoid (yellow-orange color) insoluble inwater that has antioxidant properties and is affirmed as GRAS foruse as a nutrient (21 CFR 184.1245) by the FDA. b-Carotene wasalso used as a lipophilic marker in the in vitro permeability tests.Pseudo-ternary phase diagrams of the lecithin-linker mixture, fedstate simulated intestinal fluid (FeSSIF), and ethyl caprate with andwithout b-sitosterol were developed to guide the formulation ofself-emulsifying delivery systems (SEDS). An in vitro method wasused to assess active ingredient permeability by outfitting a Flow-Thru Dialyzer (Xuan et al., 2012; Yan and Gemeinhart, 2005), withexcised lamb intestine.

2. Materials and methods

2.1. Materials

2.1.1. ChemicalsThe following chemicals were purchased from Sigma–Aldrich

Canada (Oakville, ON, Canada): Sorbitan monooleate (span 80,+99%), taurocholic acid sodium salt hydrate (+95%), ethyl caprate(+99%), Dulbecco’s phosphate buffer saline (PBS), albumin frombovine serum (BSA), b-carotene (95%), sodium chloride (+99.5%,

ffect at the oil/water interface.

tions for self-emulsifying delivery of nutraceuticals, Int J Pharmaceut


121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

OH

CH3

CH3

CH3CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3 CH3

CH3

Fig. 2. Chemical structure of b-sitosterol (left) and b-carotene (right).

J. Chu et al. / International Journal of Pharmaceutics xxx (2014) xxx–xxx 3

G Model

IJP 14058 1–11

Fluka brand), and b-sitosterol (assay �60%, Fluka brand). Sodiumhydroxide pellets were purchased from Caledon Laboratories Ltd.(Georgetown, ON, Canada). Glacial acetic acid (+99.7%) waspurchased from EMD Chemicals Inc. (Philadelphia, PA, USA).Samples of decaglyceryl caprylate/caprate (drewpol 10-1-CC) andPEG-6-caprylic/capric glycerides (softigen 767) were kindlydonated by Stepan Company (Northfield, IL, USA) and Sasol NorthAmerica (Houston, TX, USA), respectively. Laboratory gradesoybean lecithin (+99%) was purchased from Fisher Scientific(Fairlawn, NJ, USA). Frozen sheep intestine was purchased from alocal market. The compositions listed are on a mass basis (wt.%)unless stated otherwise.

2.1.2. TissuePrior to use, the frozen sheep intestine was thawed by soaking it

in water at 35–40 �C for 1 h. This source of small intestine produceshomogenous and reproducible samples to measure drug perme-ation (Kumar et al., 2008). A three-inch section of the smallintestine (jejunum) was excised and the exterior impermeablemembrane was removed. The excised section was rinsed 4 timeswith distilled water and then soaked in PBS for 30 min to render itready for use in the in vitro study.

2.2. Microemulsion formulation

Phase behavior studies were required to determine optimalsurfactant and linker ratios for microemulsion formation. Thephase behavior was investigated using equivalent masses of oil andaqueous phases at room temperature (�25 �C) and pressure(1 atm). The oil and aqueous phases were then combined andvortexed for 2 min at 3200 rpm. The formulations were set aside toequilibrate for two weeks. Fed state simulated intestinal fluid(FeSSIF) (Dressman and Reppas, 2000) was used as the aqueousmedium. The composition of FeSSIF is shown in Table 1. Thepresence of surface active species (taurocholate and lecithin) inFeSSIF contributes to the emulsification of lipids and lipid-baseddelivery systems (Pouton, 1997, 2000). On the other hand, thepresence of sodium chloride helps in FeSSIF mimics the ionicstrength of the intestinal fluid. The effect of electrolyte concentra-tion in the phase behavior of lecithin-linker has been reportedelsewhere (Acosta et al., 2005). In lecithin-linker microemulsionsprepared without ionic surfactants, the effect of electrolyteconcentration is relatively minor. The mixture of acetic acid andsodium hydroxide produces a buffer with pH 5. For lecithin-linker

Table 1Composition of the fed state simulated intestinal fluid (FeSSIF) based on theformulation of Dressman and Reppas (2000).

Fed state simulated intestinal fluid (FeSSIF)

Sodium taurocholate 15 mMLecithin 3.75 mMAcetic acid 8.65 gNaCl 11.9 gNaOH qs pH 5Deionized water qs 1 L


microemulsion systems, it has been shown that within the range ofpHs of 3.5–6.5 there is little change in the phase behavior of thesystem. Highly alkaline systems (e.g., pH 10.5 or greater) tend toproduce hydrophilic systems with different phase behavior (Acostaet al., 2005).

Based on preliminary scans, when using 4% lecithin, a minimumof 7% sorbitan monooleate, and 3% decaglyceryl caprylate/capratewas required to prevent the formation of lecithin gels. Ahydrophilic linker scan was conducted maintaining constant theconcentrations of 4% lecithin, 7% sorbitan monooleate and 3%decaglyceryl caprylate/caprate in the system, and graduallyincreasing the PEG-6-caprylic/capric glycerides content from 0%to 20% to obtain transparent isotropic regions. The formulationsproduced with this scan are referred to as S1 systems, and theircomposition is summarized in Table 2.

2.3. Physicochemical characterization

2.3.1. Emulsion drop sizeThe average drop diameters were determined using dynamic

light scattering (DLS) measurements from a BI90Plus particle sizeanalyzer instrument (l = 635 nm; Brookhaven Instruments, LongIsland, NY, USA). The photomultiplier detector is positioned at ascattering angle of 90�. The cuvettes were set aside for 2 min toequilibrate prior to the measurements. The data were obtained byaveraging five measurements per vial set at 1 min per measure-ment. The FeSSIF used to dilute the SEDS formulations has micellesof 2–8 nm (Fatouros et al., 2009), but due to the low concentrationand small size of those micelles, the intensity of the scattered lightat 90� was low, particularly when compared to the scatteringobtained with diluted SEDS. The viscosities of the samples weremeasured in triplicate using a CV-2200 falling ball viscometer(Gilmont Instruments, IL, USA). The measurements were con-ducted at 25 �C.

2.3.2. ConductivityA VWR conductivity meter (model 21800-012) was used, which

was equipped to measure in the range of 0.001 mS–14.67 mS.Conductivity was measured at 25 �C.

2.3.3. Interfacial tensionA model 500 spinning drop interfacial tensiometer from

Temco, Inc. (Tulsa, OK, USA) was used to measure at 25 �C, theinterfacial tension between the FeSSIF-diluted microemulsion

Table 2Composition of S1 phase scan.

Phase Component S1 wt.%(%)

Aqueous Lecithin 4Decaglyceryl caprylate/caprate 3PEG-6-caprylic/capric glycerides 0–20FeSSIF Balance to 100

Oil Sorbitan monooleate 7Ethyl caprate 43



202 an203 de204 et

205 2.206

207 in208 tu209 I

210 in211 ca212 se213 re214 dr215 20216 lig217 Pa218 4.

t

219 2.

220

221 ph222 m223 ph224 (d225 te226 an227 id228 ph229 bi230 di231 Ca232

233 sc234 ce235 id236 40237 di238 1:239 Fe240 an241 be242 T1243 T1244 pr245

246 an247 liq248 eq249 re250 iso251 ph252 te

253 2.

254

255 di256 vo257 m258 re

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

Fig. 3. Flow-Thru DialyzerTM device to measure permeation of b-carotene.


G Model

IJP 14058 1–11

d the excess oil phase released after dilution. The principles andtails of the instrument have been described elsewhere (Kiran

al., 2009).

3.4. TurbidityTurbidity can be examined by the changes in transmitted light

tensity. It is represented mathematically in Eq. (1) where t isrbidity having units of inverse length, L is the path length of light,is transmitted light intensity, and I0 is the transmitted lighttensity of the “clear” sample (Acosta et al., 2003), which in thisse is distilled water. A red line was drawn on white paper andrved as a reference point. The changes in the light intensity of thed line through the sample served as a turbidity indicator. Sinceug residence time in the small intestine is about 3 h (Fadda et al.,09), images were captured over 4 h to examine stability. Theht intensities were assessed via image analysis tools CorelintShop Pro X3 (RGB tool, blue channel) and Scion Image Beta03 (Plot Profile tool).

¼ 1LIn

I0I

� �(1)

4. Pseudo-ternary phase diagram

A pseudo-ternary phase diagram was generated to examinease behavior and identify the range of isotropic regions withicroemulsion structures. The diagram was created along aqueousase dilution lines where mixtures of surfactant, linkers and oilesignated as preconcentrates) were diluted with FeSSIF at roommperature. Oil (ethyl caprate) dilution lines of surfactant, linkersd fed state were also used to further assist in phaseentification. The fed state medium was used as the aqueousase to generate the diagram rather than water to better simulateological conditions and assess the in vivo phase behavior uponlution (Dressman et al., 1998; Staggers et al., 1990; Wang andrey, 1996).The selected surfactant and linker mixture (from the phase

ans) was comprised of lecithin, sorbitan monooleate, decagly-ryl monocaprylate/caprate, and PEG-6-caprylic/capric glycer-es, in the ratio of 4:7:3:13. Preconcentrates containing 10, 20, 30,, 50, 60, 70, and 80% surfactant and linker mixture in oil wereluted with FeSSIF at FeSSIF:preconcentrate ratios of 0.2:1, 0.5:1,1, 2:1, 5:1, 10:1, 100:1. Similarly, 10–80% surfactant and linker inSSIF were diluted with oil to characterize the preconcentratesd phases near the surfactant–FeSSIF axis. Shorter notations will

used to denote specific preconcentrate dilutions, for instance0 represents the preconcentrate containing 10% surfactant in oil,0-1 represents the first dilution (16.7% aqueous phase) of the T10econcentrate.Each dilution was vortexed thoroughly for 1 min at 3200 rpmd kept in a separate vial. Cross polarizers were used to identifyuid crystal regions. The formulations were set aside touilibrate for 2 weeks before analysis. Formulations thatmained transparent were categorized as monophasic andtropic in the phase diagram. In addition, the effect ofytosterols was investigated by incorporating 3.35% of b-sitos-rol in oil (ethyl caprate).

5. Self-emulsification studies

The self-emulsification studies consisted of mimicking in vivolution with simulated intestinal media. The procedure involvesrtexing the fed state simulated intestinal fluid (FeSSIF)-dilutedicroemulsion preconcentrate for 2 min, and evaluating thesulting drop size and its turbidity over time. This method assists


in assessing whether the microemulsion preconcentrate willproduce a SMEDS, SEDS, an insoluble phase or a conventionalemulsion upon dilution. The criteria distinguishing these systemswill be based on the drop sizes discussed earlier, where SEDSproduce drop sizes between 100 nm and 1 mm, and conventionalemulsions are classified as systems with drop size exceeding 1 mm.

2.6. Permeation studies

The Flow-Thru DialyzerTM (Harvard Apparatus, St. Lawrent, QC,Canada) setup shown in Fig. 3 shows the receiver chambercontaining 4% bovine serum albumin in PBS (simulating humanplasma concentration). The SEDS (T50) dosed with b-carotene(0.22% in oil) was diluted with FeSSIF and circulated through thedonor chamber. The donor and receiver chambers were separatedby the excised intestine (6 mm diameter cross section). A FeSSIFdilution factor of 500 was used to approximate the dilution ratio inthe small intestine (Hauss, 2007). Both reservoirs were mixed atthe lowest setting. An Ocean Optics (Dunedin, FL, USA) UV/visSpectrophotometer (HR2000) and software (OOIBase32) trackedchanges in absorbance of the active. After the 2 h permeationstudy, the donor chamber was rinsed with 6 mL of FeSSIF and thiseffluent was discharged. The loop of the donor side was then closedto recirculate 15 mL of FeSSIF for 1 h to simulate the desorptionprocess. Additional simulations consisted of diluted oil (0.22%b-carotene in ethyl caprate diluted with FeSSIF) as the donorsolution to serve as a baseline comparison. The dialyzer fluidscirculated at a rate of 3 mL/min, which was consistent with themeasured postprandial flow rate in the jejunum in the first hour ofa study (Fadda et al., 2009; Kerlin et al., 1982). The absorbance datawas obtained by subtracting the absorbance at 450 nm (signaldetection wavelength) from the baseline. The integration time,average, boxcar, and strobe frequency parameters were set at 8, 10,2, and 10, respectively. The volumes of simulated plasma andsimulated intestinal fluid after consumption of SEDS were 30 mLand 10 mL. These volume ratios were designed based on the ratiosfound in literature (Sherwood, 2004).

To validate the use of the excised sheep intestine, thepermeability of lidocaine through the tissue was evaluated andcompared to literature values. The permeability of lidocaine wasmeasured using a protocol previously developed to measure thepermeability of lidocaine through skin (Yuan et al., 2008). To thisend, a section of the intestine was prepared following theprocedure of Section 2.1.2 and placed in a MatTek permeationdevice (MPD) (MatTek Corporation, Ashland, MA, USA). Theintestine section was placed into the MPD with the apical sidefacing up. After assembling the device with the membrane, thepermeation device was placed in a 6-well plate containing 5 mL ofphosphate buffer solution (PBS, the receiver solution). The donorcompartment (0.5 mL) of the MPD was filled with a solution of2000 mg/L lidocaine in PBS. At predetermined times (0.25, 0.5, 1, 2,3, 4, and 6 h), the MPD was moved to another well containing 5 mLof “fresh” PBS. The concentration of the permeated lidocaine in



310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

Fig. 4. PEG-6-caprylic/capric Q7glycerides (% w/w) scan for the S1 formulations of Table 2. (For interpretation of the references to colour in this figure legend, the reader isreferred to the web version of this article.)


G Model

IJP 14058 1–11

each well was measured using a Dionex ICS-3000 liquidchromatography system equipped with an AD25 UV–vis detector(set to l = 230 nm) and a reverse phase column (Genesis, C18,4 mm, 150 mm � 4.6 mm). A mixture of acetonitrile and 0.05 MNaH2PO4�H2O (pH 2.0) (30:70, v/v) was used as the mobile phase atflow rate of 1.0 mL/min. The retention time of lidocaine under thedescribed conditions was approximately 2.7 min, and the calibra-tion curve for the area under the peak vs. concentration was linear(R2 = 0.995). Additional details for this method can be foundelsewhere (Yuan et al., 2008; Xuan et al., 2012). The permeability oflidocaine was calculated by simply dividing the permeated flux oflidocaine (permeation area 0.256 cm2) by the concentration in thedonor phase, which assumed sink conditions (the highestconcentration measured in the receiver was still less than50 ppm). The experiment was run in triplicate and the calculatedlidocaine permeability was 7.5 � 0.8E-5 cm/s. This permeability isconsistent with the range of 7–8E-5 cm/s obtained by Berggrenet al. (2004) for intestines excised from male Sprague-Dawley rats.The permeability obtained in this work is slightly higher than thevalue of 6.2E-5 cm/s obtained for lidocaine with Caco-2 cellcultures (Camenisch et al., 1998).

3. Results

3.1. Phase scans

The phase behavior obtained with the S1 scan is shown in Fig. 4.Similar hydrophilic linker scans undertaken with lecithin-linkermicroemulsions have shown a conventional Type II (water-in-oil)– Type III/IV (bicontinuous) – Type I (oil-in-water) microemulsionwith increasing hydrophilic linker concentration (Acosta et al.,2005; Yuan et al., 2008). The Type II microemulsions in these scansproduce a clear or translucent upper yellow phase (due to thepresence of lecithin), capable of scattering light (signaling thepresence of water-swollen reverse micelles), that contain solubi-lized water (more than half the volume of the liquid phase isoccupied by the microemulsion) coexisting with an excess loweraqueous phase (Acosta et al., 2005; Yuan et al., 2008). This generalbehavior is consistent with the behavior observed for the systemscontaining from 0% to 4% PEG-6-caprylic/capric glycerides(hydrophilic linker). A substantial change in the phase behavioris observed at 6% PEG-6-caprylic/capric glycerides where a singlephase microemulsion with no excess phases is produced, akin tothe description of a Type IV microemulsion. However, the systemscontaining 6, 8, 10, and 12% PEG-6-caprylic/capric glycerides

Table 3Viscosity of selected S1 formulations.

% PEG-6-caprylic/capric glycerides w/w(%)

Mean viscosity (cP) + SD

3 16.4 + 0.514 15.5 + 0.515 18.8 + 0.520 28.6 + 1.0

SD – Standard deviation.


showed birefringence when examined with polarized light,suggesting the presence of anisotropic structures. Higher PEG-6-caprylic/capric glycerides concentrations, in the range of 13–20%resulted in isotropic systems with no excess phases. At thesehigher PEG-6-caprylic/capric glycerides concentrations the sys-tems have relatively low viscosities, summarized in Table 3.Although the structure of the systems obtained with 6–20% PEG-6-caprylic/capric is not clear, the purpose of the phase scan was tonarrow the phase transition Type II–III/IV–I, and certainly thesystem at 6% PEG-6-caprylic/capric glyceride represents a transi-tion point. Further studies are required to determine themorphology of the 6–20% PEG-6-caprylic/capric systems usingscattering techniques (SANS, SAXS) and self-diffusion coefficientsobtained via NMR techniques (Garti et al., 2004; Datema et al.,1992; Fisher et al., 2013; Fanun, 2010).

Fig. 5 shows the S1 drop sizes (accurate measurement given thewater continuity) after a fed state dilution ratio of 500. The dilutionof 0–6% PEG-6-caprylic/capric glycerides resulted in emulsion dropsizes between 570 nm and 830 nm. In comparison, the higheramount of PEG-6-caprylic/capric glycerides in the range of 8–20%produced nanoemulsions with more uniform sizes (smallerdeviation) of 105–175 nm. 13% PEG-6-caprylic/capric glyceridesconcentration was the lowest linker content that resulted in atransparent phase with small drop size upon dilution. Thus, theratio of 4% lecithin/7% sorbitan monooleate/3% decaglycerylmonocaprylate/caprate/13% PEG-6-caprylic/capric glycerides wasselected to generate the pseudo-ternary phase diagram.

3.2. Pseudo-ternary phase diagram

According to the phase scan (S1) studies, a “surfactant + linkermixture” with the mass ratio of 4:7:3:13 of lecithin:sorbitanmonooleate:decaglyceryl monocaprylate/caprate:PEG-6-caprylic/

Fig. 5. Drop sizes of S1 formulations after FeSSIF dilution factor of 500.



383 ca384 ta385 ex386 th387 6.388 no389 di390 su391

392 af393 Th394 fo395 Fe396 co

397

398

399

400

401

402

403

404

405

406

407

408

409

410

Fig. 6. Pseudo-ternary phase diagrams for systems prepared with the surfactant–linker (SL) mixture lecithin:sorbitan monooleate:decaglyceryl monocaprylate/caprate:PEG-6-caprylic/capric glycerides, using a ratio of 4:7:3:13. The oil phase contains ethyl caprate (EC) in (a) and (b), and EC + 3.35% b-sitosterol (b-sitosterol) in (c). The aqueousphase is FeSSIF (F). (a) presents the phase diagram after 4 h. (b) and (c) present the phase diagrams after 2 weeks. (For interpretation of the references to colour in this figurelegend, the reader is referred to the web version of this article.)


G Model

IJP 14058 1–11

pric glycerides produced clear, non-birefringent systems con-ining nearly 50% water and oil, and bluish emulsions upontreme dilution with FeSSIF. Pseudo-ternary phase diagrams foris surfactant mixture, oil, and FeSSIF system are depicted in Fig.

By visual inspection, any changes in the mixtures occurredtably within the first week. The blue lines illustrate aqueouslution lines for preconcentrates containing 30 and 50% ofrfactant + linker mixture (T30 and T50, respectively).Fig. 6a shows the pseudo-ternary phase diagram for the system

ter 4 h. A transparent, single phase region was identified as 1Ø.e single phase regions were observed upon the initial dilutionsr systems T30, T40, and T50. T30 could incorporate up to 17%SSIF. The T40 and T50 dilution lines within the monophasic areauld incorporate about 33% and 50% aqueous phase respectively.

Fig. 7. Pictures of the T40 and T50 dilution line containing


The remainder of the phase diagram designated by mØ representsa turbid area with multiphase systems.

Fig. 6b displays the phase behavior after two weeks where thesingle phase limits and multiphase regions are identified.Preconcentrates containing low surfactant concentrations resultedin microemulsions with excess oil (mE + oil) after FeSSIF dilutions.A microemulsion region with no excess phases (mE region) wasobtained with T30, T40, and T50 preconcentrates at low tointermediate dilutions. Dilutions of preconcentrates containinghigher surfactant concentration (T60+) led to two phases wherethe lower and upper phase comprised of liquid crystal andmicroemulsion, respectively (LC + mE). Going further along the T70dilution line resulted in liquid crystals with excess oil (LC + oil).Multiple phases were present near the surfactant–oil and

(from left to right) 17, 33, 50, 67, 83, 91, and 99% FeSSIF.



411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

Fig. 8. FeSSIF dilution (500 times) of surfactant in oil preconcentrates T10–T80.


G Model

IJP 14058 1–11

surfactant–FeSSIF axis. The S + O + P region was comprised ofsurfactant, oil, and precipitate while the S + F + P was comprised ofsurfactant, fed state, and precipitate.

The changes in phase behavior when incorporating a nutraceu-tical can be observed in Fig. 6c, where 3.35% b-sitosterol wasincorporated into the oil phase. At this b-sitosterol (below itssolubility in the oil phase), no hydrate formation was observedupon surfactant addition. In comparison to Fig. 6b, the addition ofb-sitosterol decreased the isotropic microemulsion areas andenlarged the microemulsion + liquid crystal (LC + mE) region. Phaseseparation and b-sitosterol precipitate was observed for systemswith high FeSSIF content. Self-diffusion NMR data suggest thatsterols solubilize in the core and the palisade of micelles (Fisheret al., 2013). Similar self-diffusion data could help to understandthe role of b-sitosterol in lecithin-linker systems.

From the phase diagrams, we were able to identify thatT40–T50 preconcentrates were the optimal systems as theyformed stable microemulsion structures along the dilution line.The T50 system was selected for in vitro permeability evaluation asit was able to dissolve more oil along the dilution line incomparison to T40, as illustrated by the T50 vials in Fig. 7.

3.3. Formulation of preconcentrates

Drop size, conductivity, interfacial tension, and turbiditymeasurements were obtained to characterize and examine theperformance of surfactant–oil mixtures upon dilution. T10–T80systems were diluted by a ratio of 500:1 with FeSSIF as shown inFig. 8. The preconcentrates T30–T80 have drop sizes under 500 nm,while T10 and T20 sizes are near a micron as reported in Fig. 9a. T40and T50 preconcentrate produce the smallest sizes of 151 nm and170 nm. These systems are, therefore, SEDS. This was expectedsince the pseudo-ternary phase diagrams showed that systemsformed microemulsions along most of the dilution line. Fig. 9bshows the dilution of the T50 preconcentrate at a ratio of

Fig. 9. (a) Mean drop sizes of preconcentrate containing 10–80% surfactant, and diluteddilution factor.


100–1000. The trend shows drop sizes decreased with increasingdilution. However, the sizes remained between 120 nm and245 nm.

The conductivity of the T10–T80 dilution lines were measuredfrom the phases containing the surfactants (not excess phase) andplotted in Fig. 10. For the T10 and T20 systems, even a smalladdition of FeSSIF produces a significant increase in conductivity,reflecting the highly hydrophilic nature of those formulations. ForT30, T40 and T50 systems, the addition of FeSSIF produced agradual transition from oil continuity (low conductivity) to watercontinuity (high conductivity). The dilution of T60, T70, and T80systems, on the other hand, led to formulations that were primarilyoil-continuous and only inverted to water continuous at highFeSSIF content, revealing the more hydrophobic nature of thosesystems. It is important to clarify that although electricalconductivity is commonly used to test phase continuity inmicroemulsions (Fisher et al., 2013; Fanun, 2010), it does notprove that the system is a microemulsion. One unique property ofmicroemulsions is that they produce ultralow interfacial tensionsclose/at the phase inversion point (Salager et al., 2005; Acosta et al.,2005). Given the important of this property, the interfacial tensionof the T50 dilution was evaluated for the systems containing morethan 50% FeSSIF that, according to Fig. 7, have an excess oil phase.For the T50 systems containing 60%, 70% and 80% FeSSIF, theinterfacial tension against the excess oil phase were 3.6E-2, 5.1E-2,and 9.1E-2 mN/m, respectively. These interfacial tensions arewithin the expected range of ultralow tensions for lecithin-linker(Acosta et al., 2005).

To interpret the results of Fig. 10, it is important to rememberthat below the T30 line in Fig. 6b one obtains water-continuousmicroemulsions in the presence of excess oil. Also, above the T50line one obtains an oil-continuous microemulsion (top phase) inequilibrium with a lower phase of liquid crystals. These findingsare consistent with the emulsion phase inversion observed fromthe changes in conductivity of Fig. 10.

One necessary feature of SMEDS and SEDS is that they should bestable over the course of the digestion process (Pouton, 1997;Gursoy and Benita, 2004). The stability of the diluted preconcen-trates and diluted oil (ethyl caprate) were assessed via turbiditymeasurements.

Fig. 11 shows the change in turbidity after 4 h of the dilutedmicroemulsion preconcentrate. The percent change was deter-mined by taking measurements at the bottom of each vial where

500 times with FeSSIF. (b) Mean drop size of T50 preconcentrate as a function of



487 th488 sig489 w490 ch491 vi492 T4493 al494 T8495 an496 oi497 Ov498 th499 fo500 re

501 3.

502

503 50504 de505 id506 la507 ill

508

509

510

511

512

513

514

515

516Q5

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

Fig. 10. Conductivity plots of T10–T80 dilution lines. Each point along the dilutionline reflects the same FeSSIF dilution increment used in the pseudo-ternary phasediagram. The connected lines are for ease of visualization. Note that the maximummeter reading is 14.67 mS. The last segment of the conductivity line is only includedas a guide.

Figtra

Fig. 12. Concentration of b-carotene in the donor compartment (0.11% b-carotenein SEDS).


G Model

IJP 14058 1–11

e greatest separation would occur. The diluted oil has shown anificant reduction in turbidity (44%) indicating separationithin 4 h. The T10–T30 systems were not stable and showedanges in turbidity in the range of 12–24%. T10 however, had moresible phase separation in comparison to T20 and T30. Systems0 and T50 were the most stable emulsions. The systems T60–T80so experience some 5–10% change in turbidity. The fact that the0 system experiences a smaller change in turbidity than the T60d T70 systems could be associated with the higher surfactant tol ratio in T80, which should facilitate the emulsification of the oil.erall, the T50 system, containing a mixture of 50% oil and 50% ofe surfactant–linker mixture, shows the most promise as SEDSrmulation given the small drop size and stability of the emulsionsulting from the dilution with FeSSIF.

4. Dialyzer study

To evaluate the hypothesis that the T50 formulation (50% oil,% of the 4:7:3:13 mixture of lecithin:sorbitan monooleate:caglyceryl monocaprylate/caprate:PEG-6-caprylic/capric glycer-es) is a candidate SEDS formulation, its permeation to excisedmb intestine was evaluated using the Flow-Thru DialyzerTM

ustrated in Fig. 3. To this end, 0.02 g of the T50 formulation,

537

538

539

. 11. Change in turbidity (4 h after emulsification) of microemulsion preconcen-te containing 10 to 80% surfactant (T10–T80) diluted 500 times with FeSSIF.


containing 0.11% b-carotene, were diluted 500 times with FeSSIF(to a final mass of 10 g) and recirculated through the donorchamber of the Flow-Thru DialyzerTM for up to 4 h. During thattime, the removal of b-carotene was tracked by visible spectros-copy. Fig. 12 presents the concentration of b-carotene in the donorchamber as a function of time in three different permeation trials.In all these trials, the concentration of b-carotene plateaus after72 min. The desorption trial, also included in Fig. 12, wasundertaken right after the completion of one of the permeationtrials, and consisted in recirculating through the donor chamber a15 mL of FeSSIF solution to determine if any T50 drops that mighthave been weekly trapped by the mucin layer of the lamb intestinecan be easily removed from the intestine. However, only a smallabsorbance signal was measured throughout this desorptionexperiment, suggesting that the permeation was largely irrevers-ible. The maximum permeation of the SEDS was 56.0 + 0.8%, asshown in Fig. 13. To normalize this permeation, it is necessary toconsider how much SEDS permeated per unit intestinal surfacearea. This maximum permeation was 39.6 mg/cm2, based on thepermeation of the initial dose (0.02 g SEDS) divided by theintestinal surface area. Considering that the area of the smallintestine (not considering the villi area) is close to 9800 cm2 (Grassiet al., 2007), one can expect a large capacity for SEDS permeation.After completing the 2 h permeation experiment we could notdetect the presence of b-carotene in the receiver solution,suggesting that the b-carotene was still retained in the intestineand that little, if any, fully permeated through the intestine. This isnot surprising because the transport of b-carotene from theintestine and into plasma requires chylomicrons, which are notpresent in the in vitro model used (Harrison, 2012).

Additional permeation trials were also conducted by using onlyFeSSIF-diluted oil (no linkers) as the donor solution. The

Fig.13. Percentage of SEDS permeated for a dilution containing 0.22% b-carotene inoil. Results from three separate trials.



540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

Fig.14. (a) The donor side chamber interior showing oil phase separation after an oil dilution (no linker) permeation test. (b) Pictures of FeSSIF dilution (factor 500) of T50 andethyl caprate (both dosed with 0.22% b-carotene in ethyl caprate) taken 2 h after mixing. (For interpretation of the references to colour in this figure legend, the reader isreferred to the web version of this article.)


G Model

IJP 14058 1–11

absorbance signal for b-carotene in these experiments wasdifficult to follow because the baseline absorbance spectrachanged during the experiment as a result of the coalescence ofoil drops. As seen in Fig. 14a, the lipophilic marker (yellow-orangecolor) accumulated within the dialyzer, indicating that the dilutedoil separated in the donor chamber. No oil film or drops wereobserved on the intestine layer, perhaps due to the hydrophilicityof the intestine membrane. Fig. 14b shows a comparison of T50 andthe oil (ethyl caprate) diluted 500 times with FeSSIF after 2 h.

The T50 dilution line was also evaluated at 37 �C and at 25 �C,with b-carotene (0.22% in ethyl caprate) to determine theinfluence of these parameters. Fig. 15 shows that there is noapparent difference in phase behavior obtained with theseconditions, as evidenced by the onset of 2-phase systems whenthe dilution ratio increased from 1 of 2. The temperatureinsensitivity within the 25–37 �C range is desirable in drugdelivery vehicles, and suggests the potential effectiveness of theformulation in vivo.

4. Discussion

The original objective of this work was to assess the possibilityof using lecithin-linker microemulsions to formulate self-emulsi-fying oral delivery systems. The relative complexity of the lecithin-linker formulation, compared to other formulations, lies in theselection of the appropriate hydrophilic linkers and in determiningthe ratios of surfactants and linkers in the mixture (Acosta et al.,2011b). The phase scans obtained during the first part of the workfollowed a pattern similar to that observed in other lecithin-linkerformulations described in the past (Acosta et al., 2005; Yuan et al.,2008). With increasing hydrophilic linker, the formulationexperienced a phase transition from an upper phase micro-emulsion in equilibrium with an excess aqueous phase to a systemthat appears bicontinuous in nature (Acosta et al., 2005). Although

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

Fig. 15. T50 dilution line showing the phase behavior after 2 h at (a) 25 �C, (b) 37 �Cand with (c) b-carotene at 25 �C. Dilution ratio of FeSSIF:preconcentrate are asfollows: 0.2:1, 0.5:1, 1:1, 2:1, 5:1, 10:1, 100:1.


the detailed structure around the phase transition point requiresfurther investigation, this simple phase scan helped narrowing theformulation space to explore SEDS formulations. The ternarydiagram mapped the behavior of the optimal lecithin-linkercombination that produced SEDS with the lowest and more stabledrop sizes. Most lecithin-linker systems display relatively largesingle phase regions but tend to produce excess oil or water nearthe oil and water vertices, likely due to the partition of lipophilicand hydrophilic linkers, respectively, in those phases (Yuan et al.,2008). Liquid crystals have also been observed in lecithin-linkerformulations that have low oil content (Xuan et al., 2012).Surprisingly, even a relatively small amount of b-sitosterolsubstantially modified the features of the ternary phase diagram.The enlargement of the liquid crystal region is consistent withother findings in the literature (Spernath et al., 2003) and is likelyassociated with the formation b-sitosterol hydrates in thepresence of water (Bonsdorff-Nikander et al., 2005).

The dilution studies helped connect the phase behavior of themicroemulsion with the behavior of the self-emulsified phasesproduced with FeSSIF. This type of connection between theequilibrium phase behavior and the self-emulsification process islacking in the area of SMEDS and SEDS formulation (Lawrence andRees, 2012), but it has been amply investigated in the formation ofnanoemulsions (McClements, 2012; Rao and McClements, 2012;Salager et al., 1983; Solans et al., 2005). According to that literature,the key for producing stable nanoemulsions is to produce balancedmicroemulsions with bicontinuous character in order to produceultralow interfacial tensions that facilitate the emulsification. Tostabilize the resulting nanoemulsion, it is necessary to end up in aregion far from the phase inversion point. The results of thedilution studies are consistent with this strategy as the systemsobtained with T30–T50 produce intermediate electrical conduc-tivity, ultralow interfacial tension (characteristic of bicontinuoussystems), and emulsions with the lowest drop size and maximumstability at extreme dilutions.

The fact that the dilution with FeSSIF results in emulsions andnot microemulsions could result in an inefficient delivery system.It has been proposed that one of the best methods to guaranteeimproved drug or nutraceutical permeation is via the use of fully-dilutable (U-type) microemulsions (Garti et al., 2004). However,there have been reports that stable emulsions with sizes near100 nm can also improve this permeation (Khoo et al., 1998;Odeberg et al., 2003). The in vitro permeation tests were carried outfor the optimal formulation in the system with oil only (no lecithin-linker combination) to evaluate the added value of introducing thelecithin-linker formula. The results show that introducing thelecithin-linker formula does produce stable nanoemulsions thatcan be irreversible retained by the intestine, even when the dropsize is slightly higher than 100 nm. As indicated by other studies, aslong as the particles or emulsions are below 500 nm, then there is achance for increasing the bioavailability of the drug or



623 nu624 dr625 w626 6-627 20628

629 th630 it

631 SE632 of633 et634 na635 (s636 re637 au638 sm639 co640 di641 be642 an643 +5

644 5.

645

646 so647 de648 id649 ca650 lic651 co652 m653

654 st655 us656 re657 em658 pr659 de660 id661 co662 m663 dr664 fo665 ca666 be667 st668 re669 su670 th671 ev672 th673 cr674 pr675 ob676 m677 re678 ov679 th680 th

681 UnQ6

682

683

684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766


G Model

IJP 14058 1–11

traceutical (Acosta, 2009; Sugano, 2009). The affinity of theops for the surface of the excised intestine might be associatedith the affinity of the polyethylene glycol chains, present in PEG-caprylic/capric glycerides, for biological membranes (Liu et al.,09; Wang et al., 2008; Efremova et al., 2002).There is limited information on the role of the mucin layer ofe intestine on the permeation of self-emulisified drops. However,has been shown that the permeation of ibuprofen formulated inDS is reduced in mucin-disrupted intestines but the permeation

dissolved ibuprofen is not affected by this disruption (Araya al., 2006). Norris and Sinko (1997) studied the absorption ofnoparticles on mucin layers mounted on Ussing chambersimilar device to the Flow-Thru DialyzerTM) and obtained particletention kinetics curves that were very similar to Fig. 12. Thesethors also showed that the difference in permeation is relativelyall when comparing 10 nm and 200 nm particles, which isnsistent with the some of the earlier observations that thefference between the effectiveness of SEDS and SMEDS might not

significant as long as the drop sizes are below 500 nm. Norrisd Sinko argued that, based on the Stokes–Einstein diffusivity,00 nm drops are too big to diffuse into the mucin layer.

Conclusion

In this work, alcohol-free SEDS have been developed usingrbitan monooleate as a lipophilic linker and a combination ofcaglyceryl caprylate/caprate and PEG-6-caprylic/capric glycer-es as hydrophilic linkers. Increasing the concentration of PEG-6-prylic/capric glycerides in the systems increased the hydrophi-ity of the formulations and produced systems with composition,nductivity, and interfacial tensions similar to bicontinuousicroemulsions upon FeSSIF dilutions.Phase behavior scans (S1) and measurements of drop size and

ability of emulsions produced upon dilution with FeSSIF wereed to select the optimal surfactant and linker combination thatsulted in the largest solubilization of the oil while facilitating theulsification process upon dilution. Surfactant–linker mixturesepared with the ratio 4:7:3:13 of lecithin:sorbitan monooleate:caglyceryl monocaprylate/caprate:PEG-6-caprylic/capric glycer-es produced relatively large areas of systems with composition,nductivity, and interfacial tensions similar to bicontinuousicroemulsions that eventually resulted in stable emulsions withop size close to 200 nm. Using this optimal ratio, preconcentratermulations containing the surfactant + linker mixture and ethylprate as the oil phase were evaluated. Only systems containingtween 30 and 50% surfactant + linkers (T30–T50) producedable emulsions suitable for designing SEDS. Comparing thesesults with the ternary phase diagrams produced for the optimalrfactant/linker ratio, one observes that systems that contain lessan 30% surfactant tend to form water-continuous emulsionsen at small dilution ratios. On the other hand, systems with morean 50% surfactant + linkers tend to form systems with liquidystals. Based on these observations and self-emulsificiationinciples, one concludes that potential SEDS formulations aretained with systems that undergo a phase of single phaseicroemulsion in the early stages of the dilution with FeSSIF. Thesults also suggests that as long as the resulting emulsion is stableer the permeation time, and has drop sizes smaller than 500 nm,ere is a chance of improving the permeation of the emulsion toe intestines.

cited reference

Acosta (2012) and Yuan and Acosta (2009).


References

Acosta, E., 2012. Association colloids as delivery systems. In: Huang, Q. (Ed.),Nanotechnology in the Food, Beverage and Nutraceutical Industries. WoodheadPublishing Co., Cambridge, England, pp. 263–298.

Acosta, E., Le, M., Harwell, J.H., Sabatini, D.A., 2003. Coalescence and solubilizationkinetics in linker-modified microemulsions and related systems. Langmuir 19,566–574.

Acosta, E.J., Harwell, J.H., Sabatini, D.A., 2004. Self-assembly in linker-modifiedmicroemulsions. J. Colloid Interface Sci. 274, 652–664.

Acosta, E.J., Nguyen, T., Witthayapanyanon, A., Harwell, J.H., Sabatini, D.A., 2005.Linker-based bio-compatible microemulsions. Environ. Sci. Technol. 39, 1275–1282.

Acosta, E., 2009. Bioavailability of nanoparticles in nutrient and nutraceuticaldelivery. Curr. Opin. Colloid Interface Sci. 14, 3–15.

Acosta, E., Diosady, L., Venketeshwer, A.R., Rousseau, D., Yip Ch, 2011a. Self-emulsifying delivery systems and lipid transport. In: Ahmad, M. (Ed.), Lipids inNanotechnology. AOCS Press, pp. 135–170.

Acosta, E., Chung, O., Xuan, X.Y., 2011b. Lecithin-linker microemulsions intransdermal delivery. J. Drug Deliv. Sci. Technol. 21, 77–87.

Araya, H., Tomita, M., Hayashi, M., 2006. The novel formulation design of self-emulsifying drug delivery systems (SEDDS) type O/W microemulsion III: thepermeation mechanism of a poorly water soluble drug entrapped O/Wmicroemulsion in rat isolated intestinal membrane by the Ussing chambermethod. Drug metab. Pharmacokinetics 21, 45–53.

Berggren, S., Hoogstraate, J., Fagerholm, U., Lennernäs, H., 2004. Characterization ofjejunal absorption and apical efflux of ropivacaine, lidocaine and bupivacaine inthe rat using in situ and in vitro absorption models. Eur. J. Pharm. Sci. 21, 553–560.

Bonsdorff-Nikander, A., Lievonen, S., Christiansen, L., Karjalainen, M., Rantanen, J.,Yliruusi, J., 2005. Physical changes of b-sitosterol crystals in oily suspensionsduring heating. AAPS Pharm. Sci. Tech. 6, E412–E420.

Calderon, M., Quadir, M.A., Sharma, S.K., Haag, R., 2010. Dendritic polyglycerols forbiomedical applications. Adv. Mater. 22, 190–218.

Camenisch, G., Alsenz, J., Van De Waterbeemd, H., Folkers, G., 1998. Estimation ofpermeability by passive diffusion through Caco-2 cell monolayers using thedrugs’ lipophilicity and molecular weight. Eur. J. Pharm. Sci. 6, 313–319.

Datema, K., Bolt-Westerhoff, J., Jaspers, A., Daane, J., Rupert, L., 1992. Demonstrationof bicontinuous structures in microemulsions using automatic-mode NMR(self-)diffusion measurements. Magn. Reson. Chem. 30, 760–767.

Dressman, J., Reppas, C., 2000. In vitro–in vivo correlations for lipophilic, poorlywater-soluble drugs. Eur. J. Pharm. Sci. 11, 73–80.

Dressman, J., Amidon, G., Reppas, C., Shah, V., 1998. Dissolution testing as aprognostic tool for oral drug absorption: immediate release dosage forms.Pharm. Res. 15, 11–22.

Efremova, N., Huang, Y., Peppas, N., Lecknamd, D., 2002. Direct measurement ofinteractions between tethered poly(ethylene glycol) chains and adsorbedmucin layers. Langmuir 18, 836–845.

Fadda, H., McConnell, E., Short, M., Basit, A., 2009. Meal induced acceleration oftablet transit through the human small intestine. Pharm. Res. 26, 35–356.

Fanun, M., 2010. Properties of microemulsions with mixed nonionic surfactants andcitrus oil. Colloids Surf. A 369, 246–252.

Fatouros, D., Walrand, I., Bergenstahl, B., Mullertz, A., 2009. Physicochemicalcharacterization of simulated intestinal fed-state fluids containing lyso-phosphatidylcholine and cholesterol. Dissolut. Technol. 16, 47–50.

Fisher, S., Wachtel, E.J., Aserin, A., Garti, N., 2013. Solubilization of simvastatin andphytosterols in a dilutable microemulsion system. Colloids Surf. B 107, 35–42.

Garti, N., Yuli-Amar, I., 2008. Micro- and nano-emulsions for delivery of functionalfood ingredients. In: Garti, N. (Ed.), Delivery and Controlled Release of Bioactivesin Foods and Nutraceuticals. Woodland Publishing Co., Cambridge, England, pp.149–166.

Garti, N., Amar-Yuli, I., Spernath, A., Hoffman, R., 2004. Transitions and loci ofsolubilization of nutraceuticals in U-type nonionic microemulsions studied byself-diffusion NMR. Phys. Chem. Chem. Phys. 6, 2968–2976.

Grassi, M., Grassi, G., Lapasin, R., Colombo, I., 2007. Understanding Drug Release andAbsorption Mechanisms: A Physical and Mathematical Approach. CRC Press,Florida.

Gupta, S., Moulik, S., 2008. Biocompatible microemulsions and their prospectiveuses in drug delivery. J. Pharm. Sci. 97, 22–45.

Gupta, S., Sanyal, S., Datta, S., Moulik, S., 2006. Preparation of prospective plant oilderived micro-emulsion vehicles for drug delivery. Indian J. Biochem. Biophys.43, 254–257.

Gursoy, R., Benita, S., 2004. Self-emulsifying drug delivery systems (SEDS) forimproved oral delivery of lipophilic drugs. Biomed. Pharmacother. 58, 173–182.

Harrison, E., 2012. Mechanisms involved in the intestinal absorption of dietaryvitamin A and provitamin A carotenoids. Biochim. Biophys. Acta Mol. Cell Biol.Lipids 70–77.

Hauss, D., 2007. Oral Lipid-based Formulations Enhancing the Bioavailability ofPoorly Water-Soluble Drugs. Informa Healthcare, New York.

Huang, Q., Yu, H., Ru, Q., 2010. Bioavailability and delivery of nutraceuticals usingnanotechnology. J. Food Sci. 75, R50–R57.

Kerlin, P., Zinsmeister, A., Phillips, S., 1982. Relationship of motility to flow ofcontents in the human small-intestine. Gastroenterology 82, 701–706.

Kesisoglou, F., Panmai, S., Wu, Y., 2007. Application of nanoparticles in oral deliveryof immediate release formulations. Curr. Nanosci. 3, 183–190.



767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807

808

809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848


G Model

IJP 14058 1–11

Khoo, S., Humberstone, A., Porter, C., Edwards, G., Charman, W., 1998. Formulationdesign and bioavailability assessment of lipidic self-emulsifying formulations ofhalofantrine. Int. J. Pharm. 167, 155–164.

Kiran, S., Acosta, E., Moran, K., 2009. Evaluating the hydrophilic–lipophilic nature ofasphaltenic oils and naphthenic amphiphiles using microemulsion models. J.Colloid Interface Sci. 336, 304–313.

Kumar, M., Ahuja, M., Sharma, S., 2008. Hepatoprotective study of curcumin–soyalecithin complex. Scientia Pharmaceutica 76, 761–774.

Lawrence, M., Rees, G., 2012. Microemulsion-based media as novel drug deliverysystems. Adv. Drug Deliv. Rev. 64, 175–193.

Liu, G., Fu, L., Zhang, G., 2009. Role of hydrophobic interactions in the adsorption ofpoly(ethylene glycol) chains on phospholipid membranes investigated with aquartz crystal microbalance. J. Phys. Chem. B 113, 3365–3369.

McClements, D., 2012. Nanoemulsions versus microemulsions: terminology,differences, and similarities. Soft Matter 8, 1719–1729.

Narang, A., Delmarre, D., Gao, D., 2007. Stable drug encapsulation in micelles andmicroemulsions. Int. J. Pharm. 345, 9–25.

Norris, D., Sinko, P., 1997. Effect of size, surface charge, and hydrophobicity on thetranslocation of polystyrene microspheres through gastrointestinal mucin. J.Appl. Polym. Sci. 63, 1481–1492.

Odeberg, J., Kaufmann, P., Kroon, K., Höglund, P., 2003. Lipid drug delivery andrational formulation design for lipophilic drugs with low oral bioavailability,applied to cyclosporine. Eur. J. Pharm. Sci. 20, 375–382.

Pouton, C., 1997. Formulation of self-emulsifying drug delivery systems. Adv. DrugDeliv. Rev. 25, 47–58.

Pouton, C., 2000. Lipid formulations for oral administration of drugs: non-emulsifying, self-emulsifying, and self-microemulsifying drug delivery sys-tems. Eur. J. Pharm. Sci. 11, S93–S98.

Rao, J., McClements, D., 2012. Lemon oil solubilization in mixed surfactantsolutions: rationalizing microemulsion & nanoemulsion formation. FoodHydrocolloids 26, 268–276.

Rosen, M., 2004. Surfactants and Interfacial Phenomena, 3rd ed. John Wiley & Sons,Hoboken, New Jersey.

Rozner, S., Garti, N., 2006. The activity and absorption relationship of cholesteroland phytosterols. Colloids Surf. A 282–283, 435–456.

Rozner, S., Aserin, A., Wachtel, E., Garti, N., 2007. Competitive solubilization ofcholesterol and phytosterols in nonionic microemulsions. J. Colloid InterfaceSci. 314, 718–726.

Sabatini, D., Acosta, E., Harwell, J., 2003. Linker molecules in surfactant mixtures.Curr. Opin. Colloid Interface Sci. 8, 316–326.

Salager, J., Minana-Perez, M., Perez-Sanchez, M., Ramirez-Gouveia, M., Rojas, C.,1983. Surfactant-oil-water systems near the affinity inversion. Part III: the twokinds of emulsion inversion. J. Dispersion Sci. Technol. 4, 313–329.


Salager, J., Antón, R., Sabatini, D., Harwell, J., Acosta, E., Tolosa, L., 2005. Enhancingsolubilization in microemulsions – state of the art and current trends. J.Surfactants Deterg. 8, 3–21.

Sherwood, L., 2004. Human Physiology: from Cells to Systems, 5th ed. Brooks Cole.Solans, C., Izquierdo, P., Nolla, J., Azemar, N., Garcia-Celma, M., 2005. Nano-

emulsions. Curr. Opin. Colloid Interface Sci. 10, 102–110.Spernath, A., Yaghmur, A., Aserin, A., Hoffman, R., Garti, N., 2003. Self-diffusion

nuclear magnetic resonance, microstructure transitions, and solubilizationcapacity of phytosterols and cholesterol in Winsor IV food-grade micro-emulsions. J. Agric. Food Chem. 51, 2359–2364.

Staggers, J., Hernell, O., Stafford, R., Carey, M., 1990. Physical–chemical behavior ofdietary and biliary lipids during intestinal digestion and absorption. 1. Phasebehavior and aggregation states of model lipid systems patterned after aqueousduodenal contents of healthy adult human beings. Biochemistry 2028–2040.

Sugano, K., 2009. Introduction to computational oral absorption simulation. ExpertOpin. Drug Metab. Toxicol. 5, 259–293.

Torchilln, V., 2008. Nanoparticulates as Drug Carriers. Imperial College Press,London, England.

Von Corswant, C., Engstrom, S., Soderman, O., 1997. Microemulsions based onsoybean phosphatidylcholine and triglycerides. Phase behavior and micro-structure. Langmuir 13, 5061–5070.

Wang, D., Carey, M.C., 1996. Complete mapping of crystallization pathways duringcholesterol precipitation from model bile: influence of physical–chemicalvariables of pathophysiologic relevance and identification of a stable liquidcrystalline state in cold, dilute and hydrophilic bile salt-containing systems. J.Lipid Res. 37, 606–630.

Wang, Y., Lai, S., Suk, J.S., Pace, A., Cone, R., Hanes, J., 2008. Addressing the PEGmucoadhesivity paradox to engineer nanoparticles that “slip” through thehuman mucus barrier. Ang. Chem. 47, 9726–9729.

Xuan, X.-Y., Cheng, Y.-L., Acosta, E., 2012. Lecithin-linker microemulsion gelatin gelsfor extended drug delivery. Pharmaceutics 4, 104–129.

Yan, X., Gemeinhart, R.A., 2005. Cisplatin delivery from poly(acrylic acid-co-methylmethacrylate) microparticles. J. Control. Release 106, 198–208.

Yuan, J.S., Acosta, E., 2009. Extended release of lidocaine from linker-based lecithinmicroemulsions. Int. J. Pharm. 368, 63–71.

Yuan, J., Ansari, M., Samaan, M., Acosta, E., 2008. Linker-based lecithin micro-emulsions for transdermal delivery of lidocaine. Int. J. Pharm. 349, 130–143.

Yuan, J., Yip, A., Nguyen, N., Chu, J., Wen, X.-Y., Acosta, E., 2010. Effect of surfactantconcentration on transdermal lidocaine delivery with linker microemulsions.Int. J. Pharm. 392, 274–284.



Documents

Lecithin-linker formulations for self-emulsifying delivery of nutraceuticals