Presence of Electrostatically Adsorbed Polysaccharides Improves Spray Drying of Liposomes

E:FoodEngineering&PhysicalProperties

Presence of Electrostatically AdsorbedPolysaccharides Improves Spray Dryingof LiposomesAyse Karadag, Beraat Ozcelik, Martin Sramek, Monika Gibis, Reinhard Kohlus, and Jochen Weiss

Abstract: Spray drying of liposomes with conventional wall materials such as maltodextrins often yields nonfunctionalpowders, that is, liposomes break down during drying and rehydration. Electrostatically coating the surface of liposomeswith a charged polymer prior to spray drying may help solve this problem. Anionic lecithin liposomes (approximately 400nm) were coated with lower (approximately 500 kDa, LMW-C) or higher (approximately 900 kDa, HMW-C) molecularweight cationic chitosan using the layer-by-layer depositing method. Low (DE20, LMW-MD) or high molecular weight(DE2, HMW-MD) maltodextrin was added as wall material to facilitate spray drying. If surfaces of liposomes (1%)were completely covered with chitosan (0.4%), no bridging or depletion flocculation would occur, and mean particlediameters would be approximately 500 nm. If maltodextrins (20%) were added to uncoated liposomes, extensive liposomalbreakdown would occur making the system unsuitable for spray drying. No such aggregation or breakdown was observedwhen maltodextrin was added to chitosan-coated liposomes. Size changed little or even decreased slightly dependingon the molecular weight of maltodextrin added. Scanning electron microscopy images of powders containing chitosan-coated liposomes revealed that their morphologies depended on the type of maltodextrin added. Powders preparedwith LMW-MD contained mostly spherical particles while HMW-MD powders contained particles with concavitiesand dents. Upon redispersion, coated liposomes yielded back dispersions with particle size distributions similar to theoriginal ones, except for LMW-C coated samples that had been spray dried with HMW-MD which yielded aggregates(approximately 30 μm). Results show that coating of liposomes with an absorbing polymer allows them to be spray driedwith conventional maltodextrin wall materials.

Keywords: chitosan, layer-by-layer deposition, liposomes, maltodextrin, spray drying

Practical Application: Liposomes have attracted considerable attention in the food and agricultural, biomedical industriesfor the delivery of functional components. However, maintaining their stability in aqueous dispersion represents achallenge for their commercialization. Spray drying may promise a solution to that problem. However, prior to this studyspray drying of liposomes often led to the loss of structural integrity. Results of this study suggest that spray drying mightbe used to produce commercially feasible liposomal powders if proper combinations of adsorbing and nonadsorbingpolymers are used in the liquid precursor system.

IntroductionLiposomes are spherical bilayer vesicles formed from aqueous

dispersions of phospholipids with typical sources of phospho-lipids being soy or egg lecithins. Lecithins have long been usedas emulsifiers and texture modifiers in foods and are generallyrecognized as safe. Liposomes have attracted considerable atten-tion in the biomedical, food, and agricultural industries in recent

MS 20120759 Submitted 5/30/2012, Accepted 11/14/2012. Authors Karadagand Ozcelik are with Dept. of Food Engineering, Faculty of Chemical and MetallurgicalEngineering, Istanbul Technical Univ., 34469, Maslak, Istanbul, Turkey. AuthorsSramek and Kohlus are with Dept. of Food Processing, Inst. of Food Science andBiotechnology, Univ. of Hohenheim, Garbenstrasse 25, 70599 Stuttgart, Germany.Authors Gibis and Weiss are with Dept. of Food Physics and Meat Science, Inst. of FoodScience and Biotechnology, Univ. of Hohenheim, Garbenstrasse 25, 70599 Stuttgart,Germany. Direct inquiries to author Weiss (E-mail: [email protected]).

years because they are biocompatible, biodegradable, nontoxic,and have the ability to act as targeted release-on-demand carriersystems for both water and oil-soluble functional compounds suchas antimicrobials, flavors, antioxidants, and bioactive compounds(Keller 2001; Taylor and others 2005; Mozafari and others 2008;Malheiros and others 2010; Gibis and others 2012).

However, liposomes are generally unstable when suspended inaqueous systems for prolonged periods. They can undergo a physi-cal degradation, including vesicle fusion, aggregation, and leakageof entrapped material over time. Since such deterioration pro-cesses take place mostly in aqueous environments where lipo-somes are mobile, one possible approach may be to dry them.Freeze-drying is one of the most widely used processes to pro-duce liposomal powders and the structural changes brought aboutby freeze drying have been thoroughly investigated (Harrigan andothers 1990; Chen and others 2010; Ingvarsson and others 2011).Monosaccharides and more frequently disaccharides are utilized

C© 2013 Institute of Food Technologists R©E206 Journal of Food Science � Vol. 78, Nr. 2, 2013 doi: 10.1111/1750-3841.12023

Further reproduction without permission is prohibited

https://www.researchgate.net/publication/222773072_Liposomes_in_nutrition?el=1_x_8&enrichId=rgreq-8ab0ca08-298d-4f55-872d-5ffffe628c37&enrichSource=Y292ZXJQYWdlOzIzNDE1NjczMztBUzoxMzk5MDM4MDc1Mjg5NjBAMTQxMDM2NzA3NTQyOQ==

E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies

Spray drying of coated liposomes . . .

Table 1–Molecular characteristics of chitosans used in this study.

Chitosan type

Degree ofdeacylation

(%)∗

M. (averagemolecular

weight, kDa)Hugginsequation

M (averagecalculatedmolecular

weight, kDa)Kraemerequation

HMW-C 79 877 932LMW-C 81 532 589∗As per manufacturer specification.HMW-C, high molecular weight chitosan; LMW-C, low molecular weight chitosan.

as stabilizing excipients during drying (Koster and others 2000;Koster and others 2003). In comparison to freeze drying, spraydrying is less expensive, time and energy consuming. However,there are fewer studies available that report results of spray dryingof liposomes (Goldbach and others 1993a, 1993b; Lo and others2004; Wessman and others 2010). Some studies report contradict-ing results with respect to the integrities of the liposomal structuresduring spray drying. Goldbach and others (1993b) found only aslight decrease in the size of redispersed powders, when liposomaldispersions were spray dried in the presence of lactose. Conversely,Wessman and others (2010) produced spray dried liposomal pow-ders using lactose as a wall material and found that liposomeshad either completely collapsed or rearranged into multilamellarliposomes with membranes being composed of 2 or more bi-layers after rehydration of powders. They suggested that osmoticpressures induced by the increasingly concentrated wall materialswere responsible for the destabilization. Our own studies over thepast years have confirmed the difficulty of generating functionalliposomal powders using conventional wall materials.

Recently, we and others have used a method referred to as the“layer-by-layer” (LbL) depositing method to increase the stabilityof liposomes (Quemeneur and others 2007; Laye and others 2008;Chuah and others 2009; Mady and Darwish 2010; Quemeneurand others 2010; Gibis and others 2012). This method is based onthe deposition of oppositely charged biopolymers such as proteinsand carbohydrates onto the surfaces of liposomes. This approachhas been proven to provide better thermal as well as environmental(pH and ionic strength) stability (Ogawa and others 2003; Guzeyand McClements 2007; Shaw and others 2007; Laye and others2008; Klinkesorn and McClements 2009).

We hypothesize that the LbL approach may be able to allow forfunctional liposome powders to be prepared, that is, powders thatcontain liposomes that stay intact upon rehydration. We base thishypothesis on the fact that our previous studies had shown thatcoating of liposomal membranes improved their mechanical stabil-ity. The objective of this study was to test this hypothesis by coatingliposomes 1st with an adsorbing polymer (chitosan), adding 2nd aconventional nonadsorbing wall material (maltodextrin), and 3rdspray drying the combined system. To gain a better understandingof the role and influence of adsorbing compared with nonadsorb-ing polymers, chitosans and maltodextrins of different molecularweights were used.

Materials and Methods

MaterialsChitosans were donated by Primex (Siglufjordur, Iceland) and

had different molecular weights and degrees of deacetylation(Table 1). Lecithin (soybean phospholipids, 70% phosphatidyl-

choline) was provided by the American Lecithin Co. (Oxford,Conn., U.S.A.). Maltodextrins DE 2 and 21 were donated byRoquette-freres SA (Lestrem, France). Sodium acetate and aceticacid were purchased from Merck GmbH (Darmstadt, Germany)and Carl Roth Carl Roth GmbH & Co. KG (Karlsruhe, Germany)and used without further purification. 0.8 μm isopore mem-brane filters were obtained from Millipore GmbH (#ATTP14250,Billerica, Mass., U.S.A.). Double distilled water was used in thepreparation of all samples.

Solution preparationAn acetate buffer solution (pH = 3.5 ± 0.1; 0.1 M) was prepared

with 0.6 g/L sodium acetate and 5.75 mL/L acetic acid. A lecithinstock dispersion was prepared by dispersing 2% (w/v) lecithinpowder in acetate buffer. Chitosan stock solutions were preparedby dispersing 1.5% (w/v) into buffer solution, and the solutionwas stirred overnight to ensure complete hydration. Maltodextrinswere dissolved in acetate buffer to generate stock solutions withvarying concentrations.

Chitosan molecular weightThe shear viscosity of serially diluted chitosan solutions in acetic

acid was measured using a rotational rheometer (MCR 300, An-ton Paar, Stuttgart, Germany) equipped with a double cylindricalsystem (DG-26.7/TEZ 150 P-C). The intrinsic viscosities of chi-tosan in 0.2M CH3COOH/0.1 M CH3COONa solution werecalculated using the Huggins and Kraemer equation (Wang andothers 1991; Kasaai and others 2000). The reduced viscosity ηr wascalculated as

ηr =(

η

η0− 1

)c

(1)

where ηr is the viscosity of the chitosan solution at the polymerconcentration of c and η0is the viscosity of liquid phase. Theinherent viscosity was calculated as

ηi =ln

(nn0

)c

(2)

The intrinsic viscosity (η) of chitosan solution was obtainedfrom the intercept of an ηr and ηi compared with c plot when c wasapproaching zero. Finally, average molecular weights of chitosanwere calculated from the intrinsic viscosities using Mark–Houwinkequation (Wang and others 1991; Kasaai and others 2000; Pa andYu 2001; Baxter and others 2005).

[η] = KMaw (3)

where is intrinsic viscosity, K was a constant (1.424 ×10−3 mL g−1) and a was a constant (0.96) (Wang and others1991; Knaul and others 1998).

Preparation of uncoated and coated liposomesFine-disperse lecithin liposomes with average particle diameters

of approximately 400 nm were manufactured by homogenizinglecithin solutions with a high shear disperser (DI-25 Yellowline,IKA) for 10 min at 9500 rpm, and 3 min at 13500 rpm. Thereafterthe dispersion was filtered 5 times through 0.8 μm Isopore mem-brane filters (Millipore GmbH). Coated liposomes were produced

Vol. 78, Nr. 2, 2013 � Journal of Food Science E207



by electrostatic deposition of positively charged chitosan layer ontothe surface of negatively charged liposomes. To this purpose, li-posome suspensions (2% w/v) were added to chitosan solutions(0.002% to 1.5% [w/v] chitosan) at room temperature and stirredovernight.

Particle size distribution and mean sizeThe particle size distribution of samples was measured using

a laser light diffraction particle size analyzer (LA-950, Horiba,Japan). The instrument finds the particle size distribution that gives

the best fit between experimental measurements and predictionsmade using light scattering theory (Mie Theory). A refractiveindex for lecithin of 1.44 and 1.33 for the aqueous phase wasused to calculate particle size distributions. Laser diffraction resultswere displayed as volume based distributions, and the volume meandiameter (d . 4,3) was used to report average particle diameters. Thevolume mean diameter can be calculated as

−d4,3 =

∑ni .d 4

i∑ni d 3

i

(4)

10 2

10 3

10 4

105

10 6

0.0

Mea

n P

arti

cle

Dia

met

er (

nm)

10 2

10 3

10 4

105

10 6

0.0

Mea

n P

arti

cle

Dia

met

er (

nm)

0.2

Chit

0.2Chit

2 0.

tosan Concen

2 0.tosan Concen

4 0

ntration (w/v

Particle d

ζ -Potent

4 0ntration (w/v

Particle d

ζ -Potenti

0.6 0

v%)

diameter (nm)

tial (mV)

0.6 0v%)

diameter (nm)

ial (mV)

B

A

-40

-30

-20

-10

0

10

20

30

40

50

60

0.8

ζ -P

oten

tial

(m

V)

-40

-30

-20

-10

0

10

20

30

40

50

60

0.8

ζ -P

oten

tial

(m

V)

0.75%

0.5%

0.4%

0.25%

0.1%

0.05%

0.025%

0.01%

0.005%

0.0025

0.001%

Figure 1–Change in ζ -potential and mean particle diameter after addition of 0 to 0.75, w/v% chitosan to liposomes (1%, w/w) for (A) HMW-C (highmolecular weight chitosan) and (B) LMW-C (low molecular weight chitosan). Photographic images of dispersions (chitosan concentration increasingfrom 0.001% to 0.75% from left to right) were taken after 2 wk of storage at room temperature.

E208 Journal of Food Science � Vol. 78, Nr. 2, 2013

E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


where ni is the number of droplets of diameter di. All particle sizemeasurements were made on at least 2 freshly prepared sampleswith 3 readings made per sample.

ζ -potentialLiposomal dispersions were diluted to a particle concentration of

approximately 0.005% (w/v) with acetate buffer. Diluted disper-sions were then loaded into a cuvette of a particle electrophoresisinstrument (Nano ZS, Malvern Instruments, Malvern, UK), and

the ζ -potential was determined by measuring the direction andvelocity that the liposomes moved in the applied electric field. Theζ -potential measurements are reported as the average and standarddeviation of measurements made from 2 freshly prepared samples,with 3 readings made per sample.

Optical microscopySamples were gently shaken before analysis using a vortexer

to ensure homogeneity. One drop of sample containing liposomes

E

G

I

C

A

D

F

H

B

Figure 2–Microscopic images (at 100× magnification), A: liposomes (1%, w/w) with initial size of approximately 0.4 μm; B-G: after addition of highmolecular weight chitosans B: 0.01%, C: 0.05%, D: 0.1%, E: 0.25%, F: 0.5%, G: 0.75%; H and I: after addition of low molecular weight chitosans H:0.5% and I: 0.75%.




was placed on an objective slide and then covered with a cover slip.Light microscopy images were taken with an axial mounted CanonPowershot G10 digital camera (Canon, Tokyo, Japan) mounted onan Axio Scope optical microscope (A1, Carl Zeiss MicroimagingGmbH, Gottingen, Germany).

Figure 3–Microscopic images (at 100× magnification) of dispersions afteraddition of low molecular weight (LMW-MD) and high molecular weight(HMW-MD) maltodextrin to uncoated liposomes (0.5 w/v%). A: uncoatedliposomes; B: addition of LMW-MD (20 w/v%), C: addition of HMW-MD (20w/v%). The scale bar denotes a length of 10 μm.

Spray dryingUncoated and coated liposomal dispersions were mixed with

stock maltodextrin solutions to obtain mixtures containing 20%(w/v) MD, 0.5% (w/v) lecithin, and 0.175% (w/v) chitosan. Sam-ples were stirred overnight (12 to 15 h) prior to spray drying toensure homogeneity. The dispersions were dried at a feed rateof 2.5 cm3/min at an inlet temperature of 160 ◦C resulting in anoutlet temp of 90 ◦C, and 0.67 m3/min air flow using a laboratoryscale spray-drier equipped with a 1.5-mm nozzle atomizer oper-atorated at an atomizing air flow of 5 cm3/min (Mini spray-dryerB-290, BUCHI, Switzerland). Dried powders were collected andstored in airtight containers and placed in a desiccator (approxi-mately 3% RH) at room temperature. All analysis was completedwithin a week after the spray drying process.

Moisture contentA Karl Fischer system of Metrohm (Metrohm, Switzerland),

equipped with a Titrino KF 841 and 20 mL burette was usedto determine the residual moisture content of powders. A 2-component system containing Hydranal-Solvent and Hydranal-Titrant 5 (both Sigma-Aldrich Laborchemikalien, Germany) wasused for measurement.

Water activityThe water activity of samples was measured using an AW Sprint

TH500 Water Activity Meter (Novasina, Switzerland) at 25 ◦C.

Powder yieldFollowing the subtraction of moisture content in the spray dried

powders, the yield was calculated as the ratio of the total weight ofpowder obtained at the end of spray drying process and the massof initial solids (lecithin, chitosan, and maltodextrin) fed into thesystem:

Yield(%) =total weight of powder after spray drying × (1 − moisture content of powder)

total dry weight of lecithin, chitosan and maltodextrin added initially

× 100 (5)

Particle size distribution of reconstituted liposomesThe powder was reconstituted to 10 g solids/100 g reconstituted

liposomal dispersion by dissolving 0.5 g powder in 4.5 mL ofacetate buffer (pH = 3.5). The particle diameter was analyzedafter overnight reconstitution using the above-mentioned staticlight scattering method (Horiba LA-950, Japan).

Scanning electron microscopy (SEM)Spray-dried powders were mounted onto separate, adhesive-

coated aluminum pin stubs. Excess powder was removed by tap-ping the stubs sharply and then blowing dry air across. The stubswere sputter coated with a thin layer of gold in a Leica vacuumcoating unit at 40 mA for 100 s 3 times, at a working distance of50 mm by using an argon gas purge. The samples were examinedusing a NeoScope JCM-5000 SEM. The SEM was operated athigh vacuum with an accelerating voltage of 10 kV. Images weretaken at 1000× and 4000× magnifications.

Statistical analysisAll measurements were repeated at least 3 times using duplicate

samples. Means and standard deviations were calculated from these


E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


measurements using Excel (Microsoft, Redmond, Va., U.S.A.).One-way analysis of variance (ANOVA) was conducted using theStatistica 8 software (Stat Soft Inc., Tulsa, Okla., U.S.A.). Differ-ences were analyzed using Duncan’s Multiple Range Test compar-isons and P value of <0.05 were chosen to determine significantdifferences.

Results and Discussion

Influence of chitosan addition on the properties ofliposomes

The influence of chitosan molecular weight and concentra-tion on liposome charge and particle diameter was monitored to

Figure 4–Optical microscopic images (at 100× magnification) of uncoated liposomal dispersions containing various maltodextrin molecular weightsand concentrations. The region below the dotted line shows the concentrations where dispersions did not flocculate.




Table 2–Mean particle diameter change of uncoated liposomaldispersions after the addition of low molecular weight maltodex-trin (LMW-MD).

Liposome conc. (w/v%)LMW-MD conc.

(w/v%)Mean particlediameter (nm)

Initial diameter 427.4 ± 2.82 5 358.2 ± 1.6

10 333.2 ± 3.64 5 384.6 ± 2.0

10 350.0 ± 3.615 335.5 ± 1.2

6 10 354.0 ± 2.615 336.4 ± 2.4

determine the conditions where coated liposomes could be gen-erated (Figure 1). The surface charge of liposomes changed fromnegative to positive when chitosan was added to liposomal dis-persions. For both high and low molecular weight chitosan, thesurface charge of the initially anionic liposomes (0.5 w/v%) in-creased from –32 mV to approximately +56 mV with additionof chitosan (0 to 0.75 w/v%). Regardless of chitosan molecu-lar weight, the net surface charge of the liposomes became zeroafter the addition of approximately 0.008 w/v% chitosan suggest-ing that at this chitosan concentration charge neutralization oc-curred. When the chitosan concentration exceeded 0.05 w/v%,the ζ -potential became constant at approximately +50 mV, in-dicating that the vesicles’ surfaces had been fully covered withchitosan.

The mean diameter of particles depended on the chitosan con-centration added. The particle diameter of uncoated liposomeswas approximately 0.4 μm and increased to around 50 μm af-ter low amounts of chitosan (0.0025 to 0.01, [w/v%]) had been

0

2

4

6

8

10

12

14

16

18

20

0.01 0.1 1 10 100 1000

Fre

quen

cy (

%)

Diameter (µm)

Uncoated liposomes

LMW-MD addition

HMW-MD addition

Figure 5–Volume-based particle diameter distribution of uncoated lipo-somes (0.5 w/v%) before and after low molecular weight (LMW-MD) andhigh molecular weight maltodextrin (HMW-MD) addition (20% w/v%).

added. Formation of aggregated structures could be observed un-der the microscope, and these were after a few hours followedby phase separation and formation of a supernatant and a sedi-ment layer at the bottom of test tube (Figure 2). Large aggregateswere formed within a net negative to net positive surface chargetransition regime (ζ = –15 mV to +25 mV). Previous stud-ies have similarly indicated an increased likelihood for bridgingflocculation to occur when charges transition from net negativeto net positive or vice versa (Laye and others 2008). This hasbeen explained by the fact that insufficient amount of polymer ispresent causing a single polymer chain to attach simultaneously to2 or more liposomes. The formation of such polymer “bridges”yields very large aggregates that sediment rapidly. Above a crit-ical concentration windows, the mean diameter decreased againwith increasing chitosan concentration (>0.025 w/v%). The meandiameter was lowest (0.5 μm) at a chitosan concentration of ap-proximately 0.4 (w/v%), where the ζ -potential had reached aconstant value. There, microscopic images were void of any ag-gregated structures and no phase separation was visible in testtubes.

Further addition of chitosan to the system induced depletionflocculation, due to presence of free polyelectrolytes in the con-tinuous phase. The concentration gradient of polymers in theimmediate vicinity of liposomes and the bulk solution generatesan osmotic force which is strong enough to overcome the variousrepulsive forces thus favoring particle aggregation (D.J., 2005 #1)(McClements 2005). In our study, the formation of flocs was vis-ible in the microscopic images on the day of preparation but wasnot visible when inspecting the test tubes. Photographic images oftest tubes taken after 2 wk of storage however showed formationof a sedimentation layer in the bottom of the tubes (Figure 1 and2). Our results are in agreement with literature. Blijdenstein andothers (2004) suggested that flocs formed by bridging flocculationare mechanically strong and irreversible and grow to relativelylarge sizes, whereas flocs formed by depletion interactions aretypically smaller in size and weaker. Laye and others (2008) alsoreported that depletion flocculation occurred in the presence ofexcessive chitosan concentration, but this process was reversibleand aggregated liposomes dissociated upon dilution, stirring ormild sonication.

Depletion flocculation becomes more prevalent for polymerswith higher molecular weights and at increased polymer concen-trations. This is because the strength of the interaction increases asthe concentration of free polymer increases, and because the rangeof the interaction increases as the radius of gyration of the polymersincreases (McClements 2000). Mun and others (2006) proposedthat emulsion droplet aggregation through a depletion mechanismincreased with increasing chitosan molecular weight. Similarly inour study, high molecular weight chitosan promoted depletionflocculation at lower chitosan concentrations. Optical microscopyimages showed that at 0.5% (w/v) chitosan concentration, samplesprepared with high molecular weight chitosan had flocs in them.In contrast, samples prepared with low molecular weight chitosanflocculation had no flocs in them until a chitosan concentrationof approximately 0.75% (w/v) was exceeded (Figure 2).

Thus, stable chitosan-coated liposomes could be formed withina concentration range of Csat < C < Cdep, where Csat is the min-imum concentration of polymer required to cover the oppositelycharged particles and Cdep is the polymer concentration where de-pletion flocculation occurs. In this concentration range, surfacesof liposomes are completely saturated with chitosan, and there is


https://www.researchgate.net/publication/222593010_Comments_on_viscosity_enhancement_and_depletion_flocculation_by_polysaccharides?el=1_x_8&enrichId=rgreq-8ab0ca08-298d-4f55-872d-5ffffe628c37&enrichSource=Y292ZXJQYWdlOzIzNDE1NjczMztBUzoxMzk5MDM4MDc1Mjg5NjBAMTQxMDM2NzA3NTQyOQ==

https://www.researchgate.net/publication/227921550_Formation_of_Biopolymer-Coated_Liposomes_by_Electrostatic_Deposition_of_Chitosan?el=1_x_8&enrichId=rgreq-8ab0ca08-298d-4f55-872d-5ffffe628c37&enrichSource=Y292ZXJQYWdlOzIzNDE1NjczMztBUzoxMzk5MDM4MDc1Mjg5NjBAMTQxMDM2NzA3NTQyOQ==

https://www.researchgate.net/publication/227921550_Formation_of_Biopolymer-Coated_Liposomes_by_Electrostatic_Deposition_of_Chitosan?el=1_x_8&enrichId=rgreq-8ab0ca08-298d-4f55-872d-5ffffe628c37&enrichSource=Y292ZXJQYWdlOzIzNDE1NjczMztBUzoxMzk5MDM4MDc1Mjg5NjBAMTQxMDM2NzA3NTQyOQ==

E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


not yet sufficient free chitosan in the continuous phase to promotedepletion flocculation (Guzey and McClements 2006). The satu-ration concentration Csat can be estimated by an empirical modelvia the change of the ζ -potential:

�ζ (c )�ζsat

= ζ (c ) − ζsat

ζ0 − ζsat≈ exp

(− c

c ∗)

≈ exp(

− 3cc sat

)(6)

where ζ (c) is the potential of liposomes at a chitosan concentrationof c and ζ 0 and ζ sat are the potentials in the absence of chitosan andwhen liposomes are fully saturated with chitosan. c∗ is the chitosanconcentration where the change in ζ -potential is 1/e of the totalchange in ζ -potential at saturation: �ζ = �ζsat/e . The variablecsat can be estimated by determining the polymer concentration atwhich the ζ -potential has increased or decreased by 95%. csat =–c∗ln(0.05) or csat ≈ 3c∗ (Guzey and McClements 2007). Based on

A2

A1

A3

A4

A5

B

B5

B2

B3

B4

B1

Figure 6–Microscopic images (at 100× magnification) of structures formed after addition of high (HMW-MD) and low molecular weight (LMW-MD)maltodextrin (20 w/v%) to high molecular weight chitosan(0.175 w/v%) (A series) and low molecular weight chitosan (0.175 w/v%) (B series) coatedliposomes (0.5 w/v%) and the change in the structure upon redispersion of spray dried powders. 1: coated liposomes; 2: after addition of LMW-MD; 3:reconstituted spray dried samples prepared with LMW-MD; 4: after addition of HMW-MD; 5: reconstituted spray dried samples prepared with HMW-MD.


https://www.researchgate.net/publication/6579270_Formation_stability_and_properties_of_multilayer_emulsions_for_application_in_the_food_industry?el=1_x_8&enrichId=rgreq-8ab0ca08-298d-4f55-872d-5ffffe628c37&enrichSource=Y292ZXJQYWdlOzIzNDE1NjczMztBUzoxMzk5MDM4MDc1Mjg5NjBAMTQxMDM2NzA3NTQyOQ==



this information, we calculated a saturation concentration of 0.036(w/v%) for low molecular weight chitosan and 0.027 (w/v%) forhigh molecular weight chitosan. This concentration correspondedto the polymer concentration where particle diameters started tobecome smaller after having become aggregated.

The surface coverage at the point of saturation �sat can becalculated by considering the structure of liposomes. For phos-phatidylcholine vesicles with a single bilayer, a shell thickness ofapproximately 5 nm has been reported (Xu and others 2012). Thetotal surface area of liposomes is

A = 4πr 2n (7)

where n is the number of liposomes; and n can be calculated fromthe concentration of lecithin:

n = Clecithin,total

Clecithin,liposome= Clecithin,totalVsolution

43π (r 3 − (r − �r )3).ρlecithin

(8)

where Clecithin,liposome is the lecithin concentration needed to forman individual vesicle, r is the mean radius of the liposomes (200nm), ρ is the density of lecithin (1.015 g/cm3 at T = 25 ◦C),and �r is the thickness of the liposomal membrane (5 nm). Thesurface coverage � can then be calculated as

�sat = Csat

A(9)

For low molecular weight cationic chitosan (LMW-C) and highmolecular weight cationic chitosan (HMW-C), a mean surfacecoverage of 0.185 and 0.139 mg/m2 was calculated for our ap-proximately 400 nm liposomes. This value is similar to valuesreported previously where surface loads of biopolymers adsorbedto the surfaces of colloidal particles had been calculated (Helgasonand others 2009; Li and others 2010).

Effect of maltodextrin addition on the properties ofuncoated and coated liposomes

Uncoated liposomes. In our study, we used high or lowmolecular weight maltodextrin as wall material to facilitate spraydrying of uncoated or coated liposomes. The 2 maltodextrins arewall materials that are widely used in commercial spray dryingoperations. Lecithin and maltodextrin concentrations of 0.5 and20 (w/v%) were used, respectively. Regardless of its molecularweight, addition of maltodextrin to uncoated liposomes immedi-ately caused extensive flocculation and a complete breakdown ofthe system (Figure 3). Figure 4 shows a “phase diagram” illustrat-ing at which liposome-maltodextrin concentrations this break-down occurred. Excessive concentration of maltodextrin in thebulk solution appeared to cause depletion flocculation, albeit thistime due to the addition of maltodextrin. Depletion flocculationbegins to occur at a defined minimum polymer concentration,the so-called critical flocculation concentration, which dependson the volume fraction of liposomes and the molecular weight ofpolymer. The region where flocculation occurred was larger forhigh molecular weight maltodextrin (HMW-MD) than for lowmolecular weight maltodextrin.

When low molecular weight maltodextrin was added to li-posomes at large polymer concentrations, the particle diameterdecreased between 40 and 90 nm (Table 2). At higher concen-trations, flocculation occurred and liposomes aggregated whichwas visible under the microscope (Figure 4). In contrast, lipo-

Table 3–Mean particle diameter of HMW-C and LMW-C (0.175w/v%) coated liposomes (0.5 w/v%) in the absence of maltodex-trin, of coated liposomes mixed with LMW-MD and HMW-MD(20 w/v%) prior to spray drying, and of liposomes reconstitutedfrom powders of coated, with maltodextrin spray dried lipo-somes [liposome (0.5 w/v%); chitosan (0.175 w/v%); maltodextrin(20 w/v%)].

Initialmean

diameter(nm) MD type

Meandiameter of

predrieddispersions

(nm)

Meandiameter ofreconstituteddispersions

(nm)

Uncoated 427 ± 3HMW-C 483 ± 23 LMW-MD 381 ± 15 367 ± 22

HMW-MD 1040 ± 132 1255 ± 159LMW-C 532 ± 18 LMW-MD 409 ± 24 366 ± 16

HMW-MD 2054 ± 235 32302 ± 6652

HMW-C, high molecular weight chitosan; LMW-C, low molecular weight chitosan;HMW-MD, high molecular weight maltodextrin; LMW-MD, low molecular weightmaltodextrin.

Table 4–ζ -Potential of uncoated and HMW-C and LMW-C (0.175w/v%) coated liposomes (0.5 w/v%), of liposomes mixed withLMW-MD and HMW-MD (20 w/v%) prior to spray drying, andof liposomes reconstituted from powders of coated, with mal-todextrin spray dried liposomes.

Initialζ -potential

(mV) MD type

ζ -Potentialof predrieddispersions

(mV)

ζ -Potentialof recons-

tituteddispersions

Uncoated −31.5 ± 1.41 LMW-MD −28.7 ± 0.55 a

HMW-MD −26.0 ± 1.94b

HMW-C 56.0 ± 0.92 LMW-MD 55.0 ± 1.38 52.4 ± 1.10HMW-MD 50.5 ± 1.80 44.8 ± 1.04

LMW-C 55.8 ± 1.61 LMW-MD 54.7 ± 1.05 51.9 ± 1.66HMW-MD 51.0 ± 1.53 48.0 ± 2.04

HMW-C, high molecular weight chitosan; LMW-C, low molecular weight chitosan;HMW-MD, high molecular weight maltodextrin; LMW-MD, low molecular weightmaltodextrin.aSpray drying conducted only for chitosan coated liposomal dispersions.bPhase separation was observed after a few hours. The measurement was done after thesample had been mixed using a vortexer prior to analysis.

somes mixed with concentrations of high molecular weight mal-todextrin below the flocculation cutoff (2 w/v% lecithin, and 2.5w/v% HMW-MD) did not change in their size (415.0 ± 11.2nm) compared to uncoated ones (427.4 ± 2.8 nm). Liposomesare vesicles composed of an aqueous core surrounded by a phos-pholipid bilayer membrane. When these structures are immersedin a solution containing low molecular weight compounds withan appreciative potential to reduce water activity such as salts orsugars, water will migrate from the core to decrease the concentra-tion gradient between the inside and the outside of the liposomes.Such an osmotic driving force may explain the size reductionobserved.

Addition of maltodextrin (20% w/v) to uncoated liposomes ledto formation of different structures depending on maltodextrinmolecular weight. At this concentration, due to extensive floccu-lation, uncoated liposomes broke down in the presence of bothlow and high molecular weight maltodextrin. The results led usto the conclusion that uncoated liposomes combined with highconcentrations of maltodextrin are not suitable for spay drying(Figure 3 and 5).

The addition of high molecular weight maltodextrin slightlyreduced ζ -potential values (P < 0.05), whereas low molecularweight maltodextrin addition had no effect on the electrical chargeof uncoated liposomes (Table 4). Maltodextrin is a hydrophilic


E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


nonionic polysaccharide and as such no effect of its additionon electrical charge of liposomes is to be expected. However,Klinkesorn and coauthors (2004) also observed a slight decreasein ζ -potential of emulsions when maltodextrin concentrations in-creased; particularly so, if maltodextrins had a certain degree ofpolymerization (DE). They proposed that maltodextrin may havebecome attached to droplet surfaces due to hydrophobic inter-actions thereby reducing the negative charge of droplets. On theother hand, high molecular weight maltodextrin addition causeddepletion flocculation which may also have contributed to theobserved decrease in ζ -potential.

Table 5–Yield, moisture content, and water activity of chitosan-coated spray liposomal powders.

MD typeChitosan

type Yield (%)

Moisturecontent

(%)Water

activity (-)

LMW-MD LMW-C 64.7 ± 7.00 4.1 ± 0.62 0.09 ± 0.04HMW-C 61.7 ± 4.60 4.4 ± 0.35 0.09 ± 0.03

HMW-MD LMW-C 57.4 ± 6.33 3.8 ± 0.45 0.08 ± 0.03HMW-C 57.3 ± 2.41 3.7 ± 0.82 0.08 ± 0.03

HMW-C, high molecular weight chitosan; LMW-C, low molecular weight chitosan;HMW-MD, high molecular weight maltodextrin; LMW-MD, low molecular weightmaltodextrin.

0

2

4

6

8

10

12

14

16

0.01 0.1 1 10 100 1000

Fre

quen

cy (

%)

Diameter(µm)

LMW-C coated liposomes

HMW-MD addition

Reconstituted liposomes

0

2

4

6

8

10

12

14

16

0.01 0.1 1 10 100 1000

Fre

quen

cy (

%)

Diameter(µm)

HMW-C coated liposomes

HMW-MD addition


0

2

4

6

8

10

12

14

16

0.01 0.1 1 10 100 1000

Fre

quen

cy (

%)

Diameter(µm)

LMW-C coated liposomes

LMW-MD addition


0

2

4

6

8

10

12

14

16

0.01 0.1 1 10 100 1000

Fre

quen

cy (

%)

Diameter(µm)

HMW-C coated liposomes

LMW-MD addition


Figure 7–Volume based particle diameter distribution of low (LMW-C) and high molecular weight (HMW-C) chitosan (0.175 v/w%) coated liposomesbefore and after low (LMW-MD) and high molecular weight (HMW-MD) maltodextrin (20 w/v%) addition and after reconstitution of spray dried samples.




Coated liposomes. Liposomes coated with 0.35 (w/v%) highand low molecular weight chitosan could be used to preparespray dried powders, since at this concentration neither deple-tion nor bridging flocculation had been observed. In contrastto uncoated liposomes that had broken down after maltodex-trin addition (Figure 4), the structure of chitosan-coated lipo-somal dispersions did not change after maltodextrin addition(Figure 6). This suggests that the presence of a “protective coat”of an adsorbed biopolymer decreased susceptibility against subse-quent depletion flocculation caused by a 2nd nonadsorbing poly-mer. The adsorption of chitosan to the liposomal surfaces in-creases the thickness of the interfacial layer and alters its charge.In consequence, there may be an increased steric and electro-static repulsion between different liposomes which would reducethe extent of flocculation (Dickinson 2009). Many researchershave studied the increased stability of emulsions after adsorptionof a polysaccharide layer (Wollenweber and others 2000; Payetand Terentjev 2008). As mentioned above, an approximately 100nm decrease in diameter of liposomes (Figure 7) was observed

due to the previously mentioned osmotic effect after additionof low molecular weight maltodextrin to high and low molecu-lar weight chitosan coated liposomes. In contrast upon the addi-tion of high molecular weight maltodextrin, the dispersions stillshowed unimodal particle size distribution but increased in parti-cle size (Figure 7), for example, the diameter of high molecularweight chitosan coated liposomes increased from approximately0.5 μm to 1 μm while the diameter of LMW-C coated lipo-somes increased to 2 μm (Table 3). The differences in particlesize compared to their initial size before maltodextrin additionwas significant (P < 0.05). Since a high molecular weight chi-tosan layer on the particle surface can be expected to be thicker,liposomes coated with HMW-C appear to be less susceptible todepletion interaction upon addition of maltodextrin (Figure 6).Finally, similar to uncoated liposomes, addition of low molecu-lar weight maltodextrin to coated liposomes had no significanteffect on ζ -potential values, whereas high molecular weight mal-todextrin addition slightly decreased the ζ -potential (P < 0.05)(Table 4).

Figure 8–SEM images of coated liposomes (A: high molecular weight chitosan; B: low molecular weight chitosan) spray dried in the presence of lowmolecular weight maltodextrin. Pictures were taken at 1000× and 4000× magnifications.


E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


Moisture content and water activity of spray driedliposomal powders

The different liposomal preparations containing maltodextrinwere then spray dried. The moisture content and water activitiesof powders are listed in Table 5. Since the spray drying process pa-rameters were kept constant, moisture content and water activitydepended mainly on the feed solution properties. Indeed, therewas no significant difference among samples (P < 0.05) in terms ofmoisture content and water activity. Similar results had been ob-tained by Goula and Adamopoulos (2008) and Gharsallaoui andcoauthors (2012), who both stated that the final moisture con-tent of the spray-dried powders, at constant conditions, is mainlydetermined by the nature of the matrix material. Due to themore viscous nature of the low DE maltodextrin, and thereforecoarser droplets we expected a somewhat lower evaporation rate.In our study however, nosignificant differences were found be-tween powders containing different molecular weight maltodex-trins while the powder yield of all samples varied somewhat(58% to 65%).

Powder morphologyThe drying of liquid precursor solutions containing dispersed

polymers and discrete entities such as droplets or particles mayyield a variety of different shapes and structures depending onboth, the nature of the particulate matter and the dispersed poly-mer, and process parameters such as drying air and liquid precur-sor concentration. Many powder properties including particle size,flowability, and protection of core material from the environmenthave been reported to be directly related to their morphologies(Rosenberg and others 1985; Walton 2000). The SEM imagesof liposomal powders generated in this study demonstrate thatthe morphology of spray dried powders mainly depended on themaltodextrin type used (Figure 8 and 9). Powders prepared fromcoated liposomes with low molecular weight maltodextrin weremostly spherical with some small indentations and wrinkles ontheir surfaces. Their diameter ranged between 1 and 5 μm, andmost of the particles had average diameter of less than 5 μm. Thediameter range of coated liposomes spray dried with high molecu-lar weight maltodextrin was larger (approximately 2 to 10 μm) and

Figure 9–SEM images of coated liposomes (A: high molecular weight chitosan; B: low molecular weight chitosan) spray dried in the presence of highmolecular weight maltodextrin. Pictures were taken at 1000× and 4000× magnifications.




most of the particles had sizes of approximately 10 μm. Smallersized particles appeared broken and collapsed, and surfaces of allparticles had characteristic concavities with deep dents (Figure 9).Such dents and wrinkles on the surface of spray dried powdershave been reported in other studies also where carbohydrates ex-cipients had been used (Sheu and Rosenberg 1998; Tonon andothers 2011). It has been suggested that dent formation is relatedto skin formation around the particle. In the early stages of drying,

the particle surface is initially a liquid with high solvent content.As soon as the atomized droplets come into contact with the dryair, evaporation occurs and a significant solvent concentration gra-dient between the surface and the interior of the droplets beginsto develop. Depending on the evaporation speed and the rate withwhich the solvent may migrate from the interior to the surface,solids may precipitate from the solution at the surface of the par-ticles, leading to the formation of a crust or skin. Depending on

Figure 10–(A) Mechanistic image of different events occurred when chitosan added to liposomes and (B) when low (LMW-MD) and high maltodextrin(HMW-MD) added to uncoated liposomes and high (HMW-C) and low molecular weight chitosan (LMW-C) coated liposomes and when spray dried coatedliposome powders reconstituted.


E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


the mechanical properties of the developing skin, it may remainintact or may fracture (Walton 2000; Walzel and Furuta 2011).Moreover, if the drying process takes place at a high temperature,the crust development may be followed by an internal bubble for-mation. The bubbles expand and begin to rupture the crust, inturn causing the particle to collapse, shrivel, and re-inflate. Such athermal expansion can also smooth out dents to a varying extent.The effectiveness of dent smoothing is dependent on the dryingrate and the viscoelastic properties of the wall matrix. Neverthe-less, smoothing of dents can occur only prior to solidification ofthe matrix, when the wall matrix is still elastic enough to undergosuch structural changes (Sheu and Rosenberg 1998). As a result,relatively smooth and shriveled particles with uneven surface struc-tures are formed. According to the SEM pictures of the surfacestructure of powders prepared with low and high molecular weightmaltodextrin, the powders had less dents, and a smoother surfacewhen the molecular weight of the wall material was lower. Theless-dented structure of particles dried with low molecular weightmaltodextrin may be attributed to the higher glass transition tem-perature and difference in sugar composition. High DE maltodex-trins typically consist of a greater amount of low molecular weightsugars which may act as a plasticizer preventing irregular shrinkageduring drying. The average molecular weight of the low molecu-lar weight maltodextrin (DE21) can be estimated to approximately9000 kg/kMol while for the high molecular weight maltodextrin(DE2) a value of 155000 kg/kMol was given by Avaltroni and oth-ers (2004). The glass transition temperature of the maltodextrinsat outlet moisture content was determined based on a GordonTaylor approximation to be 176 ◦C and 86 ◦C for the high andlow molecular weight maltodextrin. In comparison to the inlet airtemperature of 160 ◦C and the outlet air temperature of 90 ◦Cthese values indicate the difference in viscosity of the skin materialat later stages of drying. The low molecular weight maltodextrinstays mobile through most of the drying, while the high molecularweight maltodextrin will not be able to shrink in the later phaseand the skin rather wrinkles. The inverse relationship betweensurface dents and DE values of maltodextrins in our study is thusin agreement with the findings of various other authors (Sheu andRosenberg 1998; Danviriyakul and others 2002; Walzel and Furuta2011).

Reconstitution of coated liposomesDue to the structural breakdown of uncoated liposomes upon

addition of either low or high molecular weight maltodextrin(20 w/v%) before spray drying (see above), we spray dried onlychitosan coated liposomes. The particle diameter of spray driedchitosan coated powders upon reconstitution was measured at dif-ferent reconstitution ratio (1:5, 1:10, and 1:20 weight of powderto weight of reconstituted dispersion) and at different times (1, 2,4, 8, and 16 h). High molecular weight maltodextrin spray driedpowders needed at least 16 h to be completely dissolved, in con-trast to low molecular weight maltodextrin powders which couldbe rehydrated in less than 1 h. Mean particle diameter of reconsti-tuted dispersions of low molecular weight maltodextrin powderswere slightly smaller than those of predried dispersions. The dif-ference (approximately 20 nm) was not significant for high molec-ular weight chitosan coated particles but was significant (approxi-mately 40 nm) for low molecular weight chitosan coated particles(P < 0.05) (Table 3). All reconstituted dispersions prepared withlow molecular weight maltodextrin (LMW-MD) had unimodalparticle size distributions (Figure 7).

A key property of dehydrated powders manufactured for con-sumer use is their ease of reconstitution (Hogekamp and Schubert2003; Klinkesorn and others 2006). We used a laser diffractiontechnique to determine the rate and efficiency of the powderdispersion. Approximately 0.85 g powder was added to a contin-uously stirred buffer solution contained within the stirring cham-ber of a laser light diffraction particle size distribution analyzer(LA-950, Horiba, Japan), using a low buffer volume in the mea-surement chamber (approximately 150 mL). The dispersibility ofpowders was then determined as the change in mean particlediameter as a function of time. The measurement was conductedonly for coated liposomes prepared with LMW-MD, since HMW-MD powders needed at least 16 h to dissolve. The particle sizereached a minimum value of approximately 350 nm for both highand low molecular weight chitosan coated liposomes after 6 minof stirring and remained constant thereafter. This rapid decreasein particle size indicated that the powder samples prepared withlow molecular weight maltodextrin dissolved quickly and yieldedhomogenous dispersions.

Particle size of reconstituted high molecular weight maltodex-trin powders increased compared to the size of predried disper-sions with the increase being significant for both high and lowmolecular weight coated liposomes (P < 0.05). The increase wasapproximately 250 nm for high molecular weight chitosan coatedliposomes whereas for low molecular weight chitosan coated ones,liposomes grew to more than 15 times their predried size (Table 3).Powders prepared with high molecular weight maltodextrin hada bimodal distribution of reconstituted low molecular weight chi-tosan coated liposomes (Figure 7), and optical microscopy showedformation of large aggregates (Figure 6). The stabilizing effect ofchitosan as a function of its molecular weight during spray dry-ing may be explained by “Hydration Forces Explanation” (HFE)theory. According to the HFE theory the presence of small so-lutes in membranes during dehydration prevents a close approachof the membrane bilayers, thereby reducing the probability of astructure breakdown that arises when the bilayers are in close prox-imity. The better protection provided by high molecular weightchitosan could be attributed to its molecular weight. Koster andcoauthors stated that as long as the solutes remained in inter-membrane regions, the protective effect of the solute was relatedto its molecular weight (Koster and others 2000). They reportedfor example that raffinose offered a substantially better protec-tion than sucrose or trehalose. Each dry raffinose molecule has amolar volume that is equal to approximately 30 times that of awater molecule compared to approximately 18 times for sucroseand trehalose. In another study, the authors suggested that largermaltodextrins with molecular weights of 5000 and 12000 Da didnot provide any stabilization since large polymers are partially orcompletely excluded from the interlamellar space and sequesteredin the bulk phase during dehydration (Koster and others 2003).The results with coated liposomes and maltodextrins of the givensize range (approximately 9000 Da) indicate the opposite.

Mechanistic insightsIn our study, the surface of anionic phosphatidylcholine lipo-

somes was coated with cationic chitosan using the LBL depositingmethod. Addition of chitosan to liposomes resulted in bridgingflocculation when the amount of chitosan was not sufficient, andin depletion flocculation when the amount of free chitosan in thecontinuous phase was too high (Figure 10). An optimum chitosanconcentration could be determined from optical microscopic im-ages and particle diameter measurements.




Maltodextrin was then mixed with coated and uncoated li-posomes to facilitate spray drying. There, liposome, chitosan, andmaltodextrin concentrations of 0.5, 0.175, and 20 (w/v%), respec-tively, were used. Regardless of maltodextrin molecular weight,its addition to uncoated liposomes causes extensive flocculationmaking the system unsuitable for spray drying (Figure 10). Incontrast, the structure of chitosan-coated liposomal dispersionsdid not change after maltodextrin addition. We suggest that theadsorbed chitosan layer increases steric and electrostatic repulsionbetween liposomes, which in turn prevents osmotic effects inducedby addition of nonadsorbing polymer that cause aggregation andbreakdown of liposomes.

When chitosan coated liposomes were spray dried with mal-todextrins and later reconstituted, all coated liposomes exceptthose that had been coated with low molecular weight chitosanand spray dried with high molecular weight maltodextrin yieldedback the original particle size distributions. Mechanistically, onecan therefore conclude that the adsorption of a protective layerof biopolymers on the surface of liposomes in combination withthe appropriate selection of a nonadsorbing polymer is requiredto generate a stable liquid precursor system that in turn resists thestresses of a spray drying process without compromising rehydra-tion to an unacceptable level.

ConclusionsLiposomes are increasingly of great interest to food manufac-

turers for the delivery of various functional components such asflavors, antioxidants, antimicrobials, and bioactives. Recent stud-ies have in particularly shown their ability to act as carriers forpolyphenolics, a class of bioactives that is notoriously difficult toinclude in traditional oil-in-water emulsions. However, the diffi-culty of maintaining the structural integrity of liposomes in aque-ous dispersion represents a challenge for their commercialization.Dehydration may increase their stability, resulting in an increasedshelf-life and decreased distribution costs. Spray drying is one ofthe most economical processes to convert liquid precursors intodry powders. Prior to this study, spray drying of liposomes oftenled to their structural integrity being lost. The results of our studyshow for the 1st time that spray drying may be used to generatecommercially feasible powders if combinations of adsorbing andnonadsorbing polymers are used in the formulation of the liquidprecursor system. Clearly more research is needed to elucidatewhether such powders also afford a higher chemical stability ofliposomes to, for example, hydrolysis and oxidation and whethersuch an approach would reduce the leakage of encapsulated ma-terial. Nevertheless, the results should be of substantial interestto the food and pharmaceutical industry interested in deliveringactive ingredients.

AcknowledgmentsWe gratefully thank Dipl.-Biol. Isabelle Schneider, Hohenheim

Univ. Inst. of Zoology for her help with the Scanning ElectronMicroscopy Analysis. Funding of the study was provided by theExperiment Station Hatch Funds of the Univ. of Hohenheim. Weparticularly thank the EuRopean Community Action Scheme forthe Mobility of University Students (ERASMUS) Program thatsupported the 1st author in the form of a scholarship.

ReferencesAvaltroni F, Bouquerand PE, Normand V. 2004. Maltodextrin molecular weight distribution

influence on the glass transition temperature and viscosity in aqueous solutions. CarbohydrPolym 58(3):323–34.

Baxter S, Zivanovic S, Weiss J. 2005. Molecular weight and degree of acetylation of high-intensityultrasonicated chitosan. Food Hydrocolloids 19(5):821–30.

Blijdenstein TBJ, Winden AJMv, Vliet Tv, Linden Evd, van Aken GA. 2004. Serum separa-tion and structure of depletion- and bridging-flocculated emulsions: a comparison. ColloidsSurfaces A Physicochem Eng Asp 245(1–3):41–8.

Chen C, Han D, Cai C, Tang X. 2010. An overview of liposome lyophilization and its futurepotential. J Control Release 142(3):299–311.

Chuah AM, Kuroiwa T, Ichikawa S, Kobayashi I, Nakajima M. 2009. Formation of biocom-patible nanoparticles via the self-assembly of chitosan and modified lecithin. J Food Sci74(1):N1–N8.

Danviriyakul S, McClements DJ, Decker E, Nawar WW, Chinachoti P. 2002. Physical stabilityof spray-dried milk fat emulsion as affected by emulsifiers and processing conditions. J FoodSci 67(6):2183–9.

Dickinson E. 2009. Hydrocolloids as emulsifiers and emulsion stabilizers. Food Hydrocolloids23(6):1473–82.

Gharsallaoui A, Saurel R, Chambin O, Voilley A. 2012. Pea (Pisum sativum, L.) protein isolatestabilized emulsions: a novel system for microencapsulation of lipophilic ingredients by spraydrying. Food Bioprocess Technol 5(6):2211–21.

Gibis M, Vogt E, Weiss J. 2012. Encapsulation of polyphenolic grape seed extract in polymer-coated liposomes. Food Funct 3:246–54.

Goldbach P, Brochart H, Stamm A. 1993a. Spray-drying of liposomes for a pulmonary admin-istration. I. Chemical stability of phospholipids. Drug Dev Ind Pharm 19(19):2611–22.

Goldbach P, Brochart H, Stamm A. 1993b. Spray-drying of liposomes for a pulmonaryadministration. II. Retention of encapsulated materials. Drug Dev Ind Pharm 19(19):2623–36.

Goula AM, Adamopoulos KG. 2008. Effect of maltodextrin addition during spray drying oftomato pulp in dehumidified air: II. Powder Properties. Drying Technol 26(6):726–37.

Guzey D, McClements DJ. 2006. Formation, stability and properties of multilayer emulsions forapplication in the food industry. Adv Colloid and Interface Sci 128–130(0):227–48.

Guzey D, McClements DJ. 2007. Impact of electrostatic interactions on formation and stabilityof emulsions containing oil droplets coated by β-Lactoglobulin−pectin complexes. J AgricFood Chem 55(2):475–85.

Harrigan PR, Madden TD, Cullis PR. 1990. Protection of liposomes during dehydration orfreezing. Chem Phys Lipids 52(2):139–49.

Helgason T, Gislason J, McClements DJ, Kristbergsson K, Weiss J. 2009. Influence of molecularcharacter of chitosan on the adsorption of chitosan to oil droplet interfaces in an in vitrodigestion model. Food Hydrocolloids 23(8):2243–53.

Hogekamp S, Schubert H. 2003. Rehydration of food powders. Food Sci Technol Int 9:223–35.Ingvarsson PT, Yang M, Nielsen HM, Rantanen J, Foged C. 2011. Stabilization of liposomes

during drying. Expert Opin Drug Deliv 8(3):375–88.Kasaai MR, Arul J, Charlet G. 2000. Intrinsic viscosity–molecular weight relationship for chi-

tosan. J Polym Sci B Polym Phys 38(19):2591–8.Keller BC. 2001. Liposomes in nutrition. Trends Food Sci Technol 12(1):25–31.Klinkesorn U, McClements DJ. 2009. Influence of chitosan on stability and lipase digestibility

of lecithin-stabilized tuna oil-in-water emulsions. Food Chem 114(4):1308–15.Klinkesorn U, Sophanodora P, Pavinee C DJM. 2004. Stability and rheology of corn oil-in-water

emulsions containing maltodextrin. Food Res Int 37 851–9.Klinkesorn U, Sophanodora P, Chinachoti P, Decker EA, McClements DJ. 2006. Characteri-

zation of spray dried tuna oil emulsified in two-layered interfacial membranes prepared usingelectrostatic layer-by-layer deposition. Food Res International 39:449–57.

Knaul JZ, Kasaai MR, Bui VT, Creber KAM. 1998. Characterization of deacetylated chitosanand chitosan molecular weight review. Can J Chem 76(11):1699–706.

Koster KL, Lei YP, Anderson M, Martin S, Bryant G. 2000. Effects of vitrified and nonvitrifiedsugars on phosphatidylcholine fluid-to-gel phase transitions. Biophys J 78(4):1932–46.

Koster KL, Maddocks KJ, Bryant G. 2003. Exclusion of maltodextrins from phosphatidyl-choline multilayers during dehydration: effects on membrane phase behaviour. Eur Biophys J32(2):96–105.

Laye C, McClements DJ, Weiss J. 2008. Formation of biopolymer-coated liposomes by electro-static deposition of chitosan. J Food Sci 73(5):N7–N15.

Li Y, Hu M, Xiao H, Du Y, Decker EA, McClements DJ. 2010. Controlling the functionalperformance of emulsion-based delivery systems using multi-component biopolymer coatings.Eur J Pharm Biopharm 76(1):38–47.

Lo Y-l, Tsai J-C, Kuo J-H. 2004. Liposomes and disaccharides as carriers in spray-dried powderformulations of superoxide dismutase. J Control Release 94(2–3):259–72.

Mady M, Darwish MM. 2010. Effect of chitosan coating on the characteristics of DPPC lipo-somes. J Adv Res 1(3):187–91.

Malheiros PdS, Daroit DJ, Brandelli A. 2010. Food applications of liposome-encapsulated an-timicrobial peptides. Trends Food Sci Technol 21(6):284–92.

McClements DJ. 2000. Comments on viscosity enhancement and depletion flocculation bypolysaccharides. Food Hydrocolloids 14(2):173–7.

McClements DJ. 2005. Food emulsions: principles, practices, and techniques. 2nd ed. BocaRaton, Fla.: CRC Press.

Mozafari MR, Khosravi-Darani K, Borazan GG, Cui J, Pardakhty A, Yurdugul S. 2008. En-capsulation of food ingredients using Nanoliposome Technology. Int J Food Prop 11(4):833–44.

Mun S, Decker EA, McClements DJ. 2006. Effect of molecular weight and degree of deacetyla-tion of chitosan on the formation of oil-in-water emulsions stabilized by surfactant–chitosanmembranes. J Colloid Interface Sci 296(2):581–90.

Ogawa S, Decker EA, McClements DJ. 2003. Production and characterization of O/W emul-sions containing cationic droplets stabilized by lecithin−chitosan membranes. J Agric FoodChem 51(9):2806–12.

Pa J-H, Yu TL. 2001. Light scattering study of chitosan in acetic acid aqueous solutions.Macromol Chem Phys 202(7):985–91.

Payet L, Terentjev EM. 2008. Emulsification and stabilization mechanisms of O/W emulsions inthe presence of chitosan. Langmuir 24(21):12247–52.

Quemeneur F, Rammal A, Rinaudo M, Pepin-Donat B. 2007. Large and giant vesicles “Dec-orated” with chitosan: Effects of pH, salt or glucose stress, and surface adhesion. Biomacro-molecules 8(8):2512–9.

Quemeneur F, Rinaudo M, Maret G, Pepin-Donat B. 2010. Decoration of lipid vesicles bypolyelectrolytes: mechanism and structure. Soft Matter 6(18):4471–81.


E:Fo

odEn

ginee

ring&

Phys

icalP

ropert

ies


Rosenberg M, Kopelman IJ, Talmon Y. 1985. A scanning electron microscopy study of mi-croencapsulation. J Food Sci 50(1):139–44.

Shaw LA, McClements DJ, Decker EA. 2007. Spray-dried multilayered emulsions as a deliverymethod for ω-3 fatty acids into food Systems. J Agric Food Chem 55(8):3112–9.

Sheu TY, Rosenberg M. 1998. Microstructure of microcapsules consisting of whey proteins andcarbohydrates. J Food Sci 63(3):491–4.

Taylor TM, Weiss J, Davidson PM, Bruce BD. 2005. Liposomal nanocapsules in food Scienceand agriculture. Crit Rev Food Sci Nutr 45(7–8):587–605.

Tonon RV, Freitas SS, Hubinger MD. 2011. Spray drying of acai (euterpe oleraceae mart.) juice:effect of inlet air temperature and type of carrier agent. J Food Process Preserv 35(5):691–700.

Walton DE. 2000. The morphology of spray-dried particles: a qualitative view. Drying Technol18(9):1943–86.

Walzel P, Furuta T. 2011. Morphology and properties of spray-dried particles. In: TsotsasE, Mujumdar AS, editors. Modern Drying Technology. Weinheim, Germany: Wiley-VCHVerlag GmbH & Co. KGaA. p 231–94.

Wang W, Bo S, Li S, Qin W. 1991. Determination of the Mark-Houwink equation for chitosanswith different degrees of deacetylation. Int J Biol Macromol 13(5):281–5.

Wessman P, Edwards K, Mahlin D. 2010. Structural effects caused by spray- and freeze-dryingof liposomes and bilayer disks. J Pharm Sci 99(4):2032–48.

Wollenweber C, Makievski AV, Miller R, Daniels R. 2000. Adsorption of hydroxypropylmethylcellulose at the liquid/liquid interface and the effect on emulsion stability. ColloidsSurfaces A Physicochem Eng Asp 172(1–3):91–101.

Xu X, Khan MA, Burgess DJ. 2012. Predicting hydrophilic drug encapsulation inside unilamellarliposomes. Int J Pharm 423(2):410–8.


Presence of Electrostatically Adsorbed Polysaccharides Improves Spray Drying of Liposomes

Documents