Dp

IAAa

b

c

Sd

a

ARRAA

KNLGPSF

1

mgmaTarglObee

h0

Carbohydrate Polymers 134 (2015) 657–663

Contents lists available at ScienceDirect

Carbohydrate Polymers

j ourna l ho me page: www.elsev ier .com/ locate /carbpol

elivery of liquorice extract by liposomes and hyalurosomes torotect the skin against oxidative stress injuries

nes Castangiaa, Carla Caddeoa, Maria Letizia Mancaa,∗, Laura Casua,na Catalan Latorrea,b, Octavio Díez-Salesb,c, Amparo Ruiz-Sauríd, Gianluigi Bacchettaa,nna Maria Faddaa, Maria Manconia

Department Scienze della Vita e dell’Ambiente, University of Cagliari, Cagliari, ItalyDepartment of Pharmacy and Pharmaceutical Technology, University of Valencia, Valencia, SpainInstituto de Reconocimiento Molecular y Desarrollo Tecnológico, Centro Mixto Universidad Politécnica de Valencia-Universidad de Valencia, Valencia,painDepartment of Pathology, University of Valencia, Valencia, Spain

r t i c l e i n f o

rticle history:eceived 7 May 2015eceived in revised form 21 July 2015ccepted 14 August 2015vailable online 20 August 2015

eywords:atural antioxidant

a b s t r a c t

Liquorice extract, obtained by percolation in ethanol of Glycyrrhiza glabra L. roots, was incorporatedin liposomes and hyalurosomes, new phospholipid-sodium hyaluronate vesicles, and their protectiveeffect against oxidative stress skin damages was probed. As a comparison, raw glycyrrhizin was alsotested. All the vesicles were small in size (≤100 nm), with a highly negative zeta potential ensuringlong-term stability, and able to incorporate a high amount of the extract. In vitro tests showed thatthe liquorice extract loaded in vesicles was able to scavenge DPPH free radical (80% inhibition) andto protect 3T3 fibroblasts against H2O2-induced oxidative stress, restoring the normal conditions. By

iquorice extractlycyrrhizinhospholipid vesiclesodium hyaluronateibroblasts

contrast, glycyrrhizin showed poor antioxidant activity, and was not able to efficiently counteract theoxidative effect of H2O2. In addition, the incorporation of the liquorice extract into the vesicular systemspromoted the proliferation and migration of 3T3 fibroblasts, favouring the closure of the scratched area.In vivo anti-inflammatory tests on mice confirmed the ability of the proposed nanosystems to improvethe local efficacy of the extract, favouring the re-epitelization process.

© 2015 Elsevier Ltd. All rights reserved.

. Introduction

Glycyrrhizin, a triterpenoid saponin glycoside, is considered theain bioactive component (4–10%) of liquorice roots (Glycyrrhiza

labra L.). It has been shown to possess glucocorticoid-like phar-acological effects, and anti-inflammatory, anti-viral, anti-tumor

nd hepatoprotective activities (Cho et al., 2010; Shen et al., 2015).he ability of glycyrrhizin to inhibit inflammatory events, suchs oedema, apoptosis, iNOS expression and NF�B, was recentlyeported (Marianecci et al., 2012). The chemopreventive effect oflycyrrhizin on TPA-induced oxidative stress and skin hyperpro-iferation markers was also probed (Rahman & Sultana, 2007).ther studies demonstrated that the whole liquorice extract can

e as effective as corticosteroids in the treatment of dermatitis,czema and psoriasis. In this framework, glycyrrhizin and liquoricextract may be promising candidates to prevent and treat local

∗ Corresponding author. Tel.: +39 0706758582; fax: +39 0706758553.E-mail address: mlmanca@unica.it (M.L. Manca).

ttp://dx.doi.org/10.1016/j.carbpol.2015.08.037144-8617/© 2015 Elsevier Ltd. All rights reserved.

inflammatory pathologies and skin injuries. Inflammation is a com-plex biological response to internal or external stimuli, such aspathogens, trauma or metabolic dysfunctions, which produce anincrease in reactive oxygen species (ROS) and reactive nitrogenspecies (RNS), contributing to an amplification of the inflamma-tory injury and lead to chronic inflammation (Mantovani, Allavena,Sica, & Balkwill, 2008; Menegazzi et al., 2008).

The inhibition of the inflammatory cascade may allow the repairof tissue damages and the prevention of carcinogenesis. The topicaladministration of anti-inflammatory and antioxidant products cansubstantially ameliorate the impaired conditions, by restoring thephysiological balance and skin functionality. To this purpose, theuse of specific drug delivery systems can be of value, as they canimprove the local bioavailability of these actives, as well as avoid orreduce side effects. Modified liposomes represent the most innova-tive and advanced carriers ensuring the highest drug performance

when applied topically (Castangia et al., 2014; Manca et al., 2013b,2014a). Among these, hyalurosomes (Manca et al., 2014c, 2015)are the newest polymer-immobilized phospholipid nanovesicles,which possess the carrier properties of liposomes coupled with the

6 rate Po

thtpf2

imlcpiamo

2

2

chfStp

2c

(RvUwArs

irwccHfl2t

2

saIrmiclR

58 I. Castangia et al. / Carbohyd

issue repairing capabilities of sodium hyaluronate. Hyalurosomesave shown a higher mechanical resistance than liposomes dueo the sodium hyaluronate network, which immobilizes the phos-holipid vesicles. This ensures an increased residence time of theormulation at the site of action avoiding drug leakage (Manca et al.,015).

In the present work, the ethanolic extract of G. glabra roots wasncorporated in liposomes and hyalurosomes to develop two for-

ulations intended for the topical application. Raw glycyrrhiziniposomes and hyalurosomes were also prepared. The physico-hemical properties of the vesicles, such as morphology, size, zetaotential and stability on storage were evaluated. The in vitro tox-

city, the aptitude to protect cells from oxidative stress damagend the effect on cell migration were evaluated using primaryouse embryonic fibroblasts (3T3). In addition, the ability to reduce

edema in a mouse model was probed.

. Materials and methods

.1. Materials

Soy phosphatidylcholine, Phospholipon® 90G (P90G), was pur-hased from Lipoid GmbH (Ludwigshafen, Germany). Sodiumyaluronate low molecular weight (200–400 kDa) was purchased

rom DSM Nutritional Products AG Branch Pentapharm (Aesch,witzerland). Glycyrrhizin from glycyrrhiza root (liquorice), 12-O-etradecanoylphorbol 13-acetate (TPA) and all other products wereurchased from Sigma-Aldrich (Milan, Italy).

.2. Plant material collection, extract preparation andharacterization

The roots of G. glabra were harvested in their balsamic periodOctober 2013) from a natural population in the area of Cedrinoiver (40◦22′38.33′′N–9◦43′16.10′′E), near Orosei (CE Sardinia). Aoucher specimen was deposited in the Herbarium CAG of theniversity of Cagliari, Italy. Dried and powdered G. glabra rootsere percolated extensively with ethanol, at room temperature.fter centrifugation (1500 × g), the extract was evaporated undereduced pressure affording a brown crude extract, which wastored at −4 ◦C until use.

An aliquot of the crude ethanolic extract was fractionatednto three fractions by vacuum liquid chromatography usingeversed phase silica gel as stationary phase and acetonitrileith 0.05% phosphoric acid as mobile phase. The fractions were

hromatographed over RP-HPLC on a Spherisorb S5 ODS 2.5 �molumn (250 × 10 mm) using a gradient system starting with2O:CH3CN:HOAc (95:5:0.2) to 100% CH3CN within 35 min. Theow rate was 1.5 ml/min and the detection wavelength was54 nm. Glycyrrhizin was detected at 23.30 min, in agreement withhe literature (Farag, Porzel, & Wessjohann, 2012).

.3. Folin–Ciocalteu method

The total phenolic content of G. glabra extract was mea-ured according to the Folin–Ciocalteu colorimetric assay using

UV/Vis spectrophotometer (Lambda 25, Perkin Elmer, Monza,taly). Briefly, 100 �l of the extract, 100 �l of the Folin–Ciocalteueagent and 800 �l of 20% (w/v) Na2CO3 aqueous solution wereixed and the absorbance was read at 765 nm after 30 min of

ncubation in the dark, at room temperature. The total polyphenolontent was calculated by means of a calibration curve using gal-ic acid as a reference, at different concentrations (0–0.125 mg/ml).esults are expressed as mg of gallic acid equivalents per g of dry

lymers 134 (2015) 657–663

extract (mg GAE/g) and are the mean values of six independentdeterminations.

2.4. Vesicle preparation

P90G (60 mg/ml) was hydrated with an aqueous disper-sion of the liquorice extract or raw glycyrrhizin (10 mg/ml) toobtain liposomes, or with a dispersion of sodium hyaluronate inwater (0.2% w/v) containing the liquorice extract or glycyrrhizin(10 mg/ml) to obtain hyalurosomes. The dispersions were left atroom temperature for one night to facilitate the swelling of thephospholipid, then sonicated (30 times, 5 s ON and 2 s OFF) with aSoniprep 150 ultrasonic disintegrator (MSE Crowley, London, UK)at an amplitude of 15 microns (Castangia et al., 2013; Manca et al.,2014b).

The samples were purified from the non-incorporated drug byloading them into tubing (Spectra/Por® membranes, 12–14 kDaMW cut-off; Spectrum Laboratories Inc., Rancho Dominguez, USA)and dialyzing against water at 25 ◦C for 3 h (refreshing water every30 min). The entrapment efficiency (EE%) was expressed as thepercentage of the antioxidant activity (DPPH assay) found afterdialysis versus that before dialysis. The glycyrrhizin content wasdetermined by measuring the absorbance at 486 nm using a UV/visspectrophotometer.

2.5. Vesicle characterization

Vesicle formation and morphology were confirmed by trans-mission electron microscopy (TEM). Samples were stained with 1%phosphotungstic acid and examined with a JEM-1010 (Jeol Europe,Paris, France) transmission electron microscope equipped with adigital camera MegaView III and the software “AnalySIS”, at anaccelerating voltage of 80 kV.

The average diameter and polydispersity index (PI) of each sam-ple were determined by Photon Correlation Spectroscopy using aZetasizer nano (Malvern Instruments, Worcestershire, UK). Zetapotential was estimated using the Zetasizer nano by means of theM3-PALS (Phase Analysis Light Scattering) technique (Manca et al.,2014b).

The stability of the vesicles was evaluated by measuring vesicleaverage size, PI and zeta potential over 60 days at room tempera-ture.

2.6. In vitro DPPH assay

The antioxidant activity of the liquorice extract in liposomesand hyalurosomes was tested by measuring their ability to scav-enge 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical. Each sample(glycyrrhizin or liquorice extract 10 mg/ml) was diluted (1:100)with a DPPH methanolic solution (4 mg/100 ml). All the sampleswere stored at room temperature for 30 min, protected from light.Then, the absorbance was measured at 517 nm against blank.All the experiments were performed in triplicate. The percentantioxidant activity (or free radical scavenging activity) was cal-culated according to the following formula:antioxidant activity(%) = [(ABSDPPH − ABSsample)/ABSDPPH] × 100 (Caddeo et al., 2013;Manca et al., 2014a).

2.7. In vitro drug release from vesicles

The glycyrrhizin and liquorice extract liposomes and hyaluro-somes were loaded (1 ml) into tubing (Spectra/Por® membranes,

12–14 kDa MW cut-off; Spectrum Laboratories Inc., RanchoDominguez, USA) and incubated in water (1 l) for 8 h, at 25 ◦C.After 2, 4, 6 and 8 h, the dispersions were diluted with methanol(1:1000) and the absorption was measured spectrophotometrically

ate Po

actv

2

eliMI(gvwiwe(

2

aopwfCc4ocpa

2

HfEa

iaaa2tdi

2

tswaf

I. Castangia et al. / Carbohydr

t 280 nm for the total phenolic content, and at 486 nm for the gly-yrrhizin content. These values (n = 3 per formulation) were usedo determine the percentage of the compounds released from theesicles.

.8. Effect of formulations on oxidative stress cell damage

The protective effect of samples on cells damaged by H2O2 wasvaluated. Primary mouse embryonic fibroblasts (3T3; ATCC col-ection, Manassas, VA, USA) were seeded in 96-well plates andncubated at 37 ◦C in 5% CO2 for 24 h. Phenol red-free Dulbecco’s

odified Eagle’s medium (DMEM) (Life Tecnologies Europe, Monza,taly) with high glucose, supplemented with fetal bovine serum10% v/v), penicillin, streptomycin and glutamine, was used as arowth medium (Manca et al., 2013a). Untreated cells (100% ofiability) and cells exposed to H2O2 were used as controls. Cellsere exposed to H2O2 (1:50000 dilution) and, immediately after,

ncubated for 4 h with the different formulations (100 �g/ml), thenashed with PBS. The MTT assay was used to assess the inhibitory

ffect of the samples on oxidative stress cell damage and deathManca et al., 2014a; Moulaoui et al., 2015).

.9. In vitro scratch assay

The ability of the different formulations to improve 3T3 prolifer-tion and migration was evaluated by measuring the cell expansionn wound surface (scratch assay). Cells were cultured in 6 welllates until complete confluence was reached. Then, a linear woundas generated using a sterile plastic pipette tip. The scattered

ragments of cells were removed by washing with fresh medium.ells were treated with the vesicular formulations (containing gly-yrrhizin or liquorice extract, 100 �g/ml) and incubated for 24 and8 h (Manca et al., 2015). Cells treated with the glycyrrhizin aque-us solution or liquorice extract ethanolic solution were used asontrols. At the end of the experiments, the cells were fixed with 4%araformaldehyde and observed under the light microscope using

4× objective.

.10. In vivo oedema assay

Female CD-1 mice (5–6 weeks old, 25–35 g) were obtained fromarlan Laboratories (S. Pietro al Natisone, Italy) and acclimatized

or 1 week. All studies were performed in accordance with theuropean Union regulations for the handling and use of laboratorynimals.

The back skin of mice was carefully shaved (day 0). TPA dissolvedn acetone (243 �M; 3 �g/20 �l) was applied to the shaved dorsalrea (∼2 cm2) to induce cutaneous inflammation and ulceration,nd all the tested samples (100 �l) were topically administered 3 hfter TPA application (day 1). The procedure was repeated daily for

more days. On day 4, mice were sacrificed by cervical disloca-ion. Each group comprised four mice. The treated area of the miceorsal skin was excised and weighed to assess any mass increase

ndicative of oedema formation.

.11. Histological examination

Skin biopsies (∼2 cm2) were excised from mice after 72 h ofreatment (day 4), and maintained in formaldehyde (4% v/v). Tis-

ue specimens were processed routinely and embedded in paraffinax. Longitudinal sections (5 �m) were stained with haematoxylin

nd eosin. Microscopic assessment by light microscope was per-ormed blind on coded slices.

lymers 134 (2015) 657–663 659

2.12. Statistical analysis of data

Results are expressed as means ± standard deviation. Multiplecomparisons of means (Tukey’s test) were used to substantiate sta-tistical differences between groups, while Student’s t-test was usedto compare two samples. Significance was tested at the 0.05 levelof probability (p). Data analysis was carried out with the softwarepackage R, version 3.1.2.

3. Results and discussion

3.1. Extract characterization

Phenols and polyphenols are the most abundant constituentsof plant extracts. They were also found in G. glabra root extract,along with glycyrrhizin, the main biologically active compound,which possesses several pharmacological properties, such as anti-inflammatory activity (Kim et al., 2006; Vaya, Belinky, & Aviram,1997).

The ethanolic extract from the roots of G. glabra used for theexperiments had a glycyrrhizin content of 36 mg/g, while thepolyphenol content, which was expressed as mg equivalents ofgallic acid per g of extract, was ∼91.2 mg GAE/g. These findings con-firm the presence of glycyrrhizin and polyphenols in the liquoriceextract, the latter being responsible for the antioxidant activity ofthe extract (as measured by the DPPH test), which was ∼7% of thatfound for quercetin, one of the most potent natural antioxidantsand a common constituent of vegetal extracts, used as a reference.

3.2. Vesicle characterization

The liquorice extract, which contained glycyrrhizin andpolyphenols, was incorporated in liposomes and hyalurosomes,the newest phospholipid-sodium hyaluronate vesicles, which havepreviously shown optimal features for skin delivery (Manca et al.,2014c, 2015). The raw glycyrrhizin was incorporated in the sameformulations and used as a reference. The composition of the vesi-cles is reported in Table 1.

TEM images showed that the vesicles were spherical in shapeand predominantly oligolamellar (Fig. 1).

Vesicle size was always ≤100 nm: empty vesicles were approx-imately 80 nm, and the incorporation of raw glycyrrhizin led toa decrease in the mean diameter to ∼59 nm, while an increase(100 nm) was observed upon loading of the liquorice extract(Table 2). Such opposite effect on the vesicle size is reasonably dueto a change in their structure (e.g., bilayer curvature and lamel-lar arrangement) caused by the different interactions between thepayload and the vesicles, which in turn are related to the physico-chemical properties of the two compounds. Glycyrrhizin is a singlemolecule with a good solubility in water, while the extract con-tains a number of chemicals belonging to different classes, such assaponins (mainly glycyrrhizin), flavonoids, coumarins, triterpenesterols etc, which are structurally different and mostly lipophilic.Hence, their localization within the vesicles will be different, i.e. inthe bilayers rather than the polar regions, thus having an oppositeeffect on the vesicle diameter.

In empty vesicles, the presence of sodium hyaluronate resultedin an increase in size (89 vs. 71 nm; Table 2). This is likely due to thefact that, during the production processes, the sodium hyaluronateinteracts with the phopholipid giving rise to a supramolecularstructure where the polymer can be both adsorbed onto the vesi-

cle surface and intercalated between the bilayers. The contributionfrom the polymer was somehow mitigated by the glycyrrhizinor the liquorice extract, which probably altered the bilayer cur-vature or the lamellar arrangement. All the samples were low

660 I. Castangia et al. / Carbohydrate Polymers 134 (2015) 657–663

Table 1Composition of the glycyrrhizin- and liquorice extract-loaded liposomes and hyalurosomes.

GLZ (mg) LQC (mg) P90G (mg) H2O (ml) HY solution (ml)

GLZ Liposomes 10 60 1GLZ Hyalurosomes 10 60 1LQC Liposomes 10 60 1LQC Hyalurosomes 10

GLZ, glycyrrhizin. LQC, liquorice extract. P90G, phosphatidylcholine. HY solution, sodium

Table 2Mean diameter (MD), polydispersity index (PI), zeta potential (ZP), entrapment effi-ciency (EE) of empty and glycyrrhizin (GLZ) or liquorice extract (LQC) liposomesand hyalurosomes. Mean values±standard deviation were obtained from at least 6samples.

MD (nm) PI ZP (mV) EE %

Empty liposomes 71 ± 2 0.26 −19 ± 1Empty hyalurosomes 89 ± 2 0.29 −18 ± 1GLZ liposomes 59 ± 2 0.26 −33 ± 5 87 ± 2GLZ hyalurosomes 58 ± 8 0.28 −45 ± 6 85 ± 1

p(ztcldim

wTp

gluIdfp

3

aa

LQC liposomes 100 ± 6 0.18 −32 ± 2 84 ± 2LQC hyalurosomes 100 ± 2 0.17 −33 ± 4 76 ± 3

olydispersed (PI ≤ 0.29), especially those loading the extractPI ≤ 0.18), thus confirming the homogeneity of the systems. Theeta potential was negative for all the samples, and the incorpora-ion of the glycyrrhizin or liquorice extract modified the surfaceharge, which became more negative (Table 2). The EE% of theiquorice extract and glycyrrhizin was similar, with no statisticalifferences between liposomes and hyalurosomes (∼80%, Table 2),

ndicating the good ability of the vesicles to incorporate the phyto-olecules.The stability of the formulations represents an important issue,

hich becomes essential when developing an industrial product.o evaluate vesicle stability, the size, polydispersity index and zetaotential were monitored over a period of 60 days (Fig. 2).

The vesicles showed a fairly constant mean size and a homo-eneous size distribution. Only the size and PI of liquorice extractiposomes increased greatly after one month of storage and contin-ed to increase up to 60 days, reaching a mean diameter of 300 nm.

t is likely that the different components of the liquorice extract mayestabilize the liposomal dispersion causing their aggregation andusion. On the contrary, in hyalurosomes this process is probablyrevented by the immobilizing effect of sodium hyaluronate.

.3. In vitro drug release studies

The release data for the phenols and glycyrrhizin from liposomesnd hyalurosomes are shown as cumulative percent release overn 8 h study period (Fig. 3). The amount of phenols released from

Fig. 1. TEM micrographs of liquorice (LQC) extr

60 1

hyaluronate in water (0.2% w/v).

the liquorice extract vesicles was assessed by the measurementof the absorbance at 280 nm, which is the simplest method for aquick and reliable estimation of the total phenolics based on thecharacteristic absorption of the benzene rings at this wavelength.The release was not considerably affected by the type of vesicle, asboth liposomes and hyalurosomes released the phenols at similarrate and to similar extent (∼30% within 8 h; Fig. 3A).

The release of glycyrrhizin from both raw glycyrrhizin andliquorice extract liposomes and hyalurosomes was also evaluatedby measuring the absorbance at 486 nm. According to the cumula-tive percent release profiles, 90% of glycyrrhizin was released fromraw glycyrrhizin vesicles at a rapid rate during 8 h, regardless of thevesicle type. In contrast, the amount of glycyrrhizin released fromthe liquorice extract vesicles was much smaller, with significantdifferences in the rate and extent between liposomes and hyaluro-somes: 33 and 15%, respectively. This could be attributed to thefact that, in the extract, the glycyrrhizin is associated with a widevariety of structurally different phytochemicals, thus, its physico-chemical properties (e.g., solubility) and behavior may be affected.As a consequence, it may not be easily leached out from the vesiclestructure to the external aqueous medium.

3.4. In vitro antioxidant and biocompatibility assays

The antioxidant activity of the prepared samples was tested asa function of their ability to scavenge DPPH radical (Fig. 4A).

The liquorice extract in ethanol displayed a moderate antiox-idant activity (55%). Both empty liposomes and hyalurosomesexhibited a similar activity (∼45%), which is due to active phos-phatidylcholine. The incorporation of the extract in the vesiclesstrengthened the antioxidant activity up to 83%. As a compari-son, quercetin, one of the most potent natural antioxidants presentin numerous plants, was also tested. The antioxidant powerof the liquorice extract was ∼7% of quercetin inhibitory activ-ity, and increased thanks to the incorporation in the vesicles(∼9.5%). By contrast, the antioxidant power of glycyrrhizin (in

solution) was very low (∼6% corresponding to <1% of quercetininhibitory activity), which is not surprising considering its chem-ical nature. Indeed, glycyrrhizin is not a phenolic compound andthe antioxidant activity is usually associated to a phenolic or

act-loaded liposomes and hyalurosomes.

I. Castangia et al. / Carbohydrate Polymers 134 (2015) 657–663 661

Fig. 2. Mean diameter (MD; A), polydispersity index (PI; B) and zeta potential (ZP; C) of glycyrrhizin (GLZ) and liquorice extract (LQC) liposomes and hyalurosomes over 60days of storage. Mean values±standard deviation are reported (n = 6).

Fig. 3. In vitro release of phenols from liquorice extract (LQC) liposomes and hyalurosomes (A) and glycyrrhizin (GLZ) from both liquorice extract and glycyrrhizin liposomesand hyalurosomes (B) over 8 h. Error bars represent standard deviation, n = 3.

F d glyca in 3T3c s (C, Da

p2ct

ig. 4. In vitro antioxidant activity (DPPH assay) of both liquorice extract (LQC) annd glycyrrhizin solutions and formulations against H2O2-induced oxidative stress

oncentrations of both liquorice extract and glycyrrhizin solutions and formulations the percentage of control (100% of viability).

olyphenolic structure (Ivanova & Gerova, 2005; Kaur & Kapoor,002; Vaya et al., 1997). Nevertheless, the encapsulation of gly-yrrhizin in the vesicles enhanced its antioxidant power, againhanks to active phosphatidylcholine.

yrrhizin (GLZ) solutions and formulations (A). Protective effect of liquorice extract fibroblasts (B). Viability of 3T3 fibroblasts incubated for 24 and 48 h with different). Data are reported as mean values±standard deviation of cell viability expressed

The ability of the prepared formulations to protect the cellsagainst H2O2-induced oxidative stress was tested using 3T3 fibro-blasts (Fig. 4B). The viability of H2O2-stressed cells was ∼64%; itincreased slightly upon treatment with the liquorice extract in

662 I. Castangia et al. / Carbohydrate Polymers 134 (2015) 657–663

F ollowv

ewaittH

fiptT

Fs(

ig. 5. Optical microscopy images of scratch induced in 3T3 fibroblasts monolayer fesicle formulations for 0, 24 and 48 h.

thanolic solution (∼80%), and reached ∼98% when the extractas loaded in vesicles. This behavior may be explained by the

bility of the vesicular systems to improve the antioxidant activ-ty of the extract and its local bioavailability. On the other hand,he treatment with glycyrrhizin was not able to efficiently coun-eract the oxidative effect of H2O2 (no statistical difference with2O2-stressed control cells).

The biocompatibility of the formulations was also probed on 3T3

broblasts (Fig. 4C and D). The liquorice extract containing sam-les were highly biocompatible: cell viability was always higherhan 100% after both 24 and 48 h, at all the concentrations tested.he biocompatibility of glycyrrhizin in solution and in vesicles was

ig. 6. Histological examination of hematoxylin-eosin stained mouse skin sections: untreolution (C), liquorice extract-loaded liposomes (D) and hyalurosomes (E). Oedema inhibGLZ) or liquorice extract (LQC) and solutions and vesicle formulations.

ed by the treatment with glycyrrhizin (GLZ) or liquorice extract (LQC) solutions and

significantly lower: at the highest concentrations (100–50 �g/ml)the viability was about 75% and 55%, whereas at the lowest con-centration (25 �g/ml) it was ∼85 and 65% after 24 and 48 h,respectively (Fig. 4C and D).

3.5. In vitro scratch assay

The ability of the formulations to promote the proliferation and

migration of 3T3 fibroblasts was assessed by using a model ofwound on cell monolayer, and evaluating the cell spreading on thewound area. As expected, the liquorice extract loaded in liposomesand hyalurosomes showed a superior ability to promote the closure

ated (A), TPA-inflamed (B) or TPA-inflamed and treated with liquorice extract (LQC)ition (%) in mice exposed to TPA followed by the administration of the glycyrrhizin

ate Po

oeeipt(

3

imfo(et

wasTbhate

owittbMtte

4

escetauTvo

A

2R4

I. Castangia et al. / Carbohydr

f the wound, compared to glycyrrhizin formulations, liquoricextract and glycyrrhizin solutions. The nanoentrapped liquoricextract promoted an almost complete wound closure after 48 h ofncubation. In particular, hyalurosomes greatly stimulated the cellroliferation and migration in the scratched area, presumably dueo the synergic activity of sodium hyaluronate and liquorice extractFig. 5).

.6. In vivo oedema assay and histological examination

TPA applied topically on mouse skin is routinely used to inducenflammation, oedema, oxidative stress, infiltration of inflam-

atory cells, loss of stratum corneum and ulceration. All theormulations (liposomes and hyalurosomes) were able to reduceedema significantly (∼52%), in agreement with previous findingsManca et al., 2014c, 2015), while the glycyrrhizin and liquoricextract solutions were unable to effectively inhibit oedema forma-ion (Fig. 6).

Moreover, the vesicles loaded with the liquorice extractere capable of reducing TPA-induced neutrophil infiltration and

llowed epidermal regeneration, promoting the restoration of theuperficial skin layer (black arrows, Fig. 6D and E). Indeed, thePA-injured skin treated with the vesicles showed a large crust,ut underneath, skin regeneration was detectable. On the otherand, after treating the injured skin with both the liquorice exractnd glycyrrhizin solutions, dermoepidermal necrosis, inflamma-ory infiltrate and the total loss of epidermal strata were morevident (white arrows, Fig. 6C).

In vivo oedema and inflammatory findings confirmed the abilityf the nanosystems to improve the drug local efficacy, especiallyhen hyalurosomes were used, probably due to the synergic activ-

ty of both the carrier and sodium hyaluronate, the former acting onhe drug release and targeting, and the latter favouring cell migra-ion and the re-epitelization process, as previously demonstratedy in vitro and in vivo experiments (Manca et al., 2014c, 2015).oreover, the great reduction in inflammatory infiltration using

he liquorice extract vesicles, might be connected to the associa-ion of glycyrrhizin with other antioxidant polyphenols (Castangiat al., 2014; Manca et al., 2014a).

. Conclusions

In this study, vesicles loaded with glycyrrhizin and liquoricextract were tested as protective formulations against oxidativetress damage. The liquorice extract, by virtue of its chemicalomposition (e.g., phenolic content) showed more promising prop-rties, especially when delivered by hyalurosomes, which were ableo retain the extract components over time, enhancing its in vitrond in vivo biological activity. The results confirmed their potentialse as protective agents able to inhibit the onset of skin damages.he topical application of liquorice extract formulations may be ofalue in innovative dermal and cosmetic products to counteractxidative stress damages and maintain the skin homeostasis.

cknowledgments

This work was supported by grants from MIUR, Italy (PRIN010–2011, Prot. 2010H834LS 004) and Programma di Sviluppourale 2007–2013 PSR Sardegna-MISURA 124, Determinazione n.230/14.

lymers 134 (2015) 657–663 663

References

Caddeo, C., Manconi, M., Fadda, A. M., Lai, F., Lampis, S., Diez-Sales, O., et al. (2013).Nanocarriers for antioxidant resveratrol: Formulation approach, vesicleself-assembly and stability evaluation. Colloids and Surfaces B: Biointerfaces, 111C, 327–332.

Castangia, I., Manca, M. L., Matricardi, P., Sinico, C., Lampis, S., Fernàndez-Busquets,X., et al. (2013). Effect of diclofenac and glycol intercalation on structuralassembly of phospholipid lamellar vesicles. International Journal ofPharmaceutics, 456(1), 1–9.

Castangia, I., Nácher, A., Caddeo, C., Valenti, D., Fadda, A. M., Díez-Sales, O., et al.(2014). Fabrication of quercetin and curcumin bionanovesicles for theprevention and rapid regeneration of full-thickness skin defects on mice. ActaBiomaterialia, 10(3), 1292–1300.

Cho, H. J., Lim, S. S., Lee, Y. S., Kim, J.-S., Lee, C. H., Kwon, D. Y., et al. (2010).Hexane/ethanol extract of Glycyrrhiza uralensis licorice exerts potentanti-inflammatory effects in murine macrophages and in mouse skin. FoodChemistry, 121(4), 959–966.

Farag, M. A., Porzel, A., & Wessjohann, L. A. (2012). Comparative metaboliteprofiling and fingerprinting of medicinal licorice roots using a multiplexapproach of GC–MS, LC–MS and 1D NMR techniques. Phytochemistry, 76,60–72.

Ivanova, D., & Gerova, D. (2005). Polyphenols and antioxidant capacity of Bulgarianmedicinal plants. Journal of Ethnopharmacology, 96, 145–150.

Kaur, C., & Kapoor, H. C. (2002). Anti-oxidant activity and total phenolic content ofsome Asian vegetables. International Journal of Food Science and Technology,37(2), 153–161.

Kim, J.-K., Oh, S., Kwon, H.-S., Oh, Y.-S., Lim, S. S., & Shin, H.-K. (2006).Anti-inflammatory effect of roasted licorice extracts onlipopolysaccharide-induced inflammatory responses in murine macrophages.Biochemical and Biophysical Research Communications, 345(3), 1215–1223.

Manca, M. L., Castangia, I., Caddeo, C., Pando, D., Escribano, E., Valenti, D., et al.(2014). Improvement of quercetin protective effect against oxidative stressskin damages by incorporation in nanovesicles. Colloids and Surfaces B:Biointerfaces, 123, 566–574.

Manca, M. L., Castangia, I., Matricardi, P., Lampis, S., Fernàndez-Busquets, X., Fadda,A. M., et al. (2014). Molecular arrangements and interconnected bilayerformation induced by alcohol or polyalcohol in phospholipid vesicles. Colloidsand Surfaces B: Biointerfaces, 117, 360–367.

Manca, M. L., Castangia, I., Zaru, M., Nacher, A., Valenti, D., Fernandez-Busquets, X.,et al. (2015). Development of curcumin loaded sodium hyaluronateimmobilized vesicles (hyalurosomes) and their application on prevention andrapid restoring of mouse skin wound. Biomaterials (in press).

Manca, M. L., Manconi, M., Falchi, A. M., Castangia, I., Valenti, D., Lampis, S., et al.(2013). Close-packed vesicles for diclofenac skin delivery and fibroblasttargeting. Colloids and Surfaces B: Biointerfaces, 111(0), 609–617.

Manca, M. L., Manconi, M., Zaru, M., Castangia, I., Cabras, A., Cappai, N., & Fadda, A.M. (2014). IALUROSOMI E LORO IMPIEGO IN FORMULATI AD USO TOPICO. InRM2014A000687.

Manca, M. L., Zaru, M., Manconi, M., Lai, F., Valenti, D., Sinico, C., et al. (2013).Glycerosomes: A new tool for effective dermal and transdermal drug delivery.International Journal of Pharmaceutics, 455(1–2), 66–74.

Mantovani, A., Allavena, P., Sica, A., & Balkwill, F. (2008). Cancer-relatedinflammation. Nature, 454(7203), 436–444.

Marianecci, C., Rinaldi, F., Mastriota, M., Pieretti, S., Trapasso, E., Paolino, D., et al.(2012). Anti-inflammatory activity of novel ammoniumglycyrrhizinate/niosomes delivery system: Human and murine models. Journalof Controlled Release, 164(1), 17–25 (Official Journal of the Controlled ReleaseSociety).

Menegazzi, M., Di Paola, R., Mazzon, E., Genovese, T., Crisafulli, C., Dal Bosco, M.,et al. (2008). Glycyrrhizin attenuates the development of carrageenan-inducedlung injury in mice. Pharmacological Research, 58(1), 22–31 (The Official Journalof the Italian Pharmacological Society).

Moulaoui, K., Caddeo, C., Manca, M. L., Castangia, I., Valenti, D., Escribano, E., et al.(2015). Identification and nanoentrapment of polyphenolic phytocomplexfrom Fraxinus angustifolia: In vitro and in vivo wound healing potential.European Journal of Medicinal Chemistry, 89, 179–188.

Rahman, S., & Sultana, S. (2007). Glycyrrhizin exhibits potential chemopreventiveactivity on 12-O-tetradecanoyl phorbol-13-acetate-induced cutaneousoxidative stress and tumor promotion in Swiss albino mice. Journal of EnzymeInhibition and Medicinal Chemistry, 22(3), 363–369.

Shen, L., Cui, Z., Lin, Y., Wang, S., Zheng, D., & Tan, Q. (2015). Anti-inflammativeeffect of glycyrrhizin on rat thermal injury via inhibition of high-mobility

group box 1 protein. Burns: Journal of the International Society for Burn Injuries,41(2), 372–378.

Vaya, J., Belinky, P. A., & Aviram, M. (1997). Antioxidant constituents from licoriceroots: Isolation, structure elucidation and antioxidative capacity toward LDLoxidation. Free Radical Biology and Medicine, 23(2), 302–313.