European Journal of Pharmaceutics and Biopharmaceutics 86 (2014) 301–306

Contents lists available at ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics

journal homepage: www.elsevier .com/locate /e jpb

Research paper

Influence of massage and occlusion on the ex vivo skin penetrationof rigid liposomes and invasomes

0939-6411/$ - see front matter � 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.ejpb.2013.11.004

⇑ Corresponding author. Department of Dermatology, Venerology and Allergol-ogy, Charité – Universitätsmedizin Berlin Charitéplatz 1, 10117 Berlin, Germany.Tel.: +49 30 450 518 235; fax: +49 30 450 518 918.

E-mail address: juergen.lademann@charite.de (J. Lademann).

Sindy Trauer a,b, Heike Richter a, Judith Kuntsche c,d, Rolf Büttemeyer e, Manfred Liebsch b,Michael Linscheid f, Alfred Fahr c, Monika Schäfer-Korting g, Jürgen Lademann a,⇑, Alexa Patzelt a

a Center of Experimental and Applied Cutaneous Physiology, Department of Dermatology, Charité – Universitätsmedizin Berlin, Germanyb Zebet at Federal Institute of Risk Assessment (BfR), Berlin, Germanyc Department of Pharmaceutical Technology, Friedrich-Schiller-University Jena, Jena, Germanyd Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Odense, Denmarke Department of Surgery, Charité – Universitätsmedizin Berlin, Germanyf Department of Chemistry, Humboldt-Universität zu Berlin, Berlin, Germanyg Institute of Pharmacy (Pharmacology and Toxicology), Freie Universität Berlin, Germany

a r t i c l e i n f o

Article history:Received 5 June 2013Accepted in revised form 7 November 2013Available online 16 November 2013

Keywords:Massage techniqueFranz diffusion cellFollicular penetrationLiposomesInvasomesDrug delivery

a b s t r a c t

Liposomes are frequently described as drug delivery systems for dermal and transdermal applications.Recently, it has been shown that particulate substances penetrate effectively into hair follicles and thatthe follicular penetration depth can be increased by massaging the skin, which simulates the in vivomovement of hairs in the hair follicles. In the present study, massage was applied to skin mounted toFranz diffusion cells. By means of confocal laser scanning microscopy, the influence of massage and occlu-sion on the follicular penetration depths of rigid and flexible liposomes loaded with a hydrophilic andlipophilic dye was investigated. The application of massage increased follicular penetration significantly.Occlusion resulted in an increased follicular penetration depth only for rigid liposomes, whereas inva-somes did not penetrate more effectively if occlusion was applied. The results confirm that massage isan important tool for increasing follicular penetration in ex vivo studies using Franz diffusion cells. Occlu-sion may reduce the efficacy of follicular penetration depending on the specific liposomal preparation.Rigidity in particular appears to be a relevant parameter.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Liposomes have been adopted as drug delivery systems for sys-temic and topical use. Hydrophilic agents can be enclosed withinthe inner aqueous sphere, and lipophilic agents can be intercalatedinto the lipid bilayer [1–3]. Liposomes can effectively deliver drugsto and through the skin [4–6]. Whereas previous studies wereaimed elucidating the mechanisms of liposome penetration andsubsequent distribution of the active compounds being trans-ported [7–9], optimization of topical drug delivery by penetrationof intact liposomes, the adsorption effect and penetration throughthe transappendageal route are currently the subject of intense re-search [10]. Other studies investigated the relevance of liposomecomposition on penetration efficacy [11]. Classical liposomes donot penetrate deeply into the skin, but rather remain confined tothe upper layers of the stratum corneum [10], as was evidenced

by a study in which intact rigid liposomes were not detected inthe granular layers of the epidermis by confocal microscopy [12].A new class of highly deformable liposomes has recently beendeveloped, i.e., deformable liposomes which can increase dermaland transdermal delivery. In some cases, the efficiency describedwas compared with subcutaneous administration [13]. Elsayedet al. [10] reviewed the mode of action of deformable liposomesand proposed that both intact vesicular permeation into the stra-tum corneum and a penetration-enhancing effect contribute tothe enhanced topical delivery of drugs, whereby one of the twomechanisms may predominate, depending on the physico-chemi-cal properties of the drug.

A comparison between rigid and flexible liposomes can be as-sumed as penetration-enhancing properties of flexible liposomesdo not become relevant during follicular penetration, since pene-tration is stimulated mechanically by massage. A liposome sizeof approximately 130 nm (Table 1) was specifically chosen becauseit is in a range in which the structure of the cuticula of the movinghairs stimulates the transport of the liposomes into the hair folli-cles by a mechanism that is similar to a geared pump. The compar-ison serves to assess the use of massage and occlusion to allow

302 S. Trauer et al. / European Journal of Pharmaceutics and Biopharmaceutics 86 (2014) 301–306

conclusions to be drawn regarding the penetration behavior of ri-gid and flexible liposomes, respectively.

Thus far, penetration studies investigating the dermal or trans-dermal delivery of drugs via liposomes were mainly performedex vivo in Franz diffusion cell experiments [14–16]. It is widely rec-ognized that ex vivo penetration studies can demonstrate whethersuch delivery systems can enhance penetration into human skinwhen performed according to the standard guidelines [17]. Ex vivotests are required as in vivo testing in humans or animals is limiteddue to ethical reasons.

In the past, the penetration of liposomes through the stratumcorneum has been studied extensively, whereas the follicular path-way has rarely been regarded as relevant for liposomal transdermaldelivery [18,19]. Recently, this opinion has drastically changed. Thehair follicle infundibulum represents an invagination of the epider-mis that physiologically interrupts the skin barrier, thus providingan additional surface area that is available for penetration of topi-cally applied substances [20–22]. A clear advantage of follicular up-take is that there is no need for manipulation of the physiologicalfunction, such as by the use of penetration enhancers, once theagent gains access to the hair follicle infundibulum in sufficientamounts. Nanoparticles and microparticles in particular preferen-tially penetrate into the hair follicles [23]. It has been hypothesizedthat movement of hairs on the skin promotes the transportation ofparticles deeply into the hair follicles where the skin barrier is lessefficient and dispersion of the active agents into the viable tissuecan occur. In vivo, the movement of the hair is a physiological pro-cess, whereas it must be simulated ex vivo [24] such as by massageapplication, which leads to a significantly deeper penetration ofparticulate substances into the hair follicle [25]. However, the effectof massage on the penetration depth of rigid and flexible liposomesinto skin mounted to Franz cells, the standard method for ex vivotesting [17], is unknown to date, and was therefore investigatedin the present study. Additionally, the influence of environmentalhumidity (e.g., occlusion) on the penetration of rigid and flexibleliposomes was examined. While occlusion is known to enhancethe delivery of conventional liposomes [26], occlusion may dimin-ish the penetration efficacy of flexible liposomes [27,28].

2. Material and methods

2.1. Handling of the skin

Human full thickness breast or abdominal skin (1700 lm thickon average) was obtained during plastic surgery from 6 differentsubjects (female, aged 30–67 years). Approval for these experi-ments had been obtained from the Ethics Committee of the Charité– Universitätsmedizin Berlin. After surgery, the subcutaneous fatwas carefully removed and the remaining sample was wrappedin aluminum foil and stored at �20 �C (min 24 h, max 6 months).

2.2. Liposome formulations

Flexible and rigid liposomes sized approximately 130 nm (Ta-ble 1) were prepared and labeled with fluorescent dyes. 1,2-Diol-eoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamineB), ammonium salt (Rh-DOPE, Avanti Polar Lipids, Alabaster, USA)

Table 1Characterization of liposomes.

Liposomes Before dialysis After dialysis

Z – average(nm)

PDI Z – average(nm)

PDI

Flexible 139 ± 1.5 0.1 ± 0.005 146 ± 0.6 0.11 ± 0.005Rigid 115 ± 1.4 0.07 ± 0.009 117 ± 1.5 0.08 ± 0.003

is a lipophilic fluorescent dye located in the liposomal membrane.The excitation wavelength of rhodamine is 540 nm and the detec-tion wavelength of the red fluorescence is 570 nm. 5,6-Carboxy-fluorescein, disodium salt (CF, Sigma Aldrich, Taufkirchen,Germany) is a hydrophilic fluorescent dye, which is enclosed intothe aqueous core of the liposomes. The excitation wavelength ofcarboxyfluorescein is 490 nm and the detection wavelength is520 nm. Green fluorescence is detectable after degradation of theliposomes and subsequent release of the dye.

2.3. Preparation of rigid liposomes

For the preparation of the liposomes with a rigid bilayer,1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) and 1,2-dipalmitoyl-sn-glycero-3-phosphoglycerol, sodium salt (DPPG),were purchased from Genzyme Pharmaceuticals (Liestal,Switzerland).

90 mg DPPC, 10 mg DPPG and 0.5 mg Rh-DOPE were dissolvedin approximately 1.5 ml chloroform (Merck, Darmstadt, Germany).After evaporation of the chloroform at 50 �C under a vacuum (Rota-Vapor, Büchi Labortechnik, Konstanz, Germany), the obtained thinlipid film was hydrated with 1 ml 10 mM Tris buffer pH 7.4 (SigmaAldrich, Taufkirchen, Germany) containing 20 mM carboxyfluores-cein, and shaken for approximately 30 min at 50 �C. Afterequilibration at room temperature overnight under a nitrogenatmosphere, small unilamellar liposomes were prepared by extru-sion (LiposoFast basic, Avestin, Ottawa, Canada) with 21 repeti-tions through a polycarbonate membrane with a pore size of100 nm (Armatis, Mannheim, Germany) at 55 �C.

2.4. Preparation of flexible liposomes (invasomes)

For the preparation of the flexible liposomes, soybean lecithinin ethanol 75:20 w/w (NAT 8539) was purchased from Phospho-lipid GmbH (Cologne, Germany). 0.5 mg Rh-DOPE was dissolvedin 134 mg of NAT 8539 corresponding to 100 mg soybean lecithin.The mixture was vortexed at room temperature until a clear solu-tion was obtained. 20 ll of a mixture of limonene, citral and cineol(10:45:45 by volume, Sigma Aldrich, Taufkirchen, Germany) wasadded. Subsequently, 900 ll 10 mM Tris buffer pH 7.4 containing20 mM carboxyfluorescein was added under vortexing. The mix-ture was sonicated (sonication bath USR 54 h, Merck EuroLab NV,VWR, Germany) for approximately 15 min at room temperatureand small unilamellar liposomes were prepared by successiveextrusion (LiposoFast basic) at room temperature twice eachthrough the polycarbonate membranes with pore sizes of400 nm, 200 nm, 100 nm and 50 nm, respectively.

Non-entrapped carboxyfluorescein was removed by dialysis ofthe liposome formulations (MWCO 12–14 kDa, Visking, MedicalInternational, Great Britain) at room temperature against 10 mMTris buffer pH 7.4 (rigid liposomes) or against 10 mM Tris bufferpH 7.4 containing 3% ethanol (v/v) and a few droplets of the ter-pene mixture (flexible liposomes). Removal of the non-entrappedcarboxyfluorescein was completed after 30 h for both liposomedispersions with the dialysis fluid being replaced four times duringthis process. As a consequence, the dispersion was diluted untilapproximately one third of the dialyzed fluid remained. The lossof entrapped carboxyfluorescein can be excluded as the size ofthe liposomes increased as a result of the supplied fluid (Table 1).During the entire preparation process and dialysis, the formula-tions were stored at 4 �C under the exclusion of light.

2.5. Reference substance caffeine

Caffeine (Sigma Aldrich, Steinhagen, Germany) was used as acontrol substance. 2.5 g Caffeine was dissolved in 1000 ml

S. Trauer et al. / European Journal of Pharmaceutics and Biopharmaceutics 86 (2014) 301–306 303

Dulbecco’s phosphate-buffered saline (DPBS) from PANBiotecGmbH (Aidenbach, Germany). Subsequently, 10 ll/cm2 of the ref-erence solution was applied at a concentration of 2.5 mg/ml DPBS.

2.5.1. Franz diffusion cell experimentsThawed human full thickness skin was mounted to a static

Franz diffusion cell (FD-C) (surface area 1.76 cm2; Gauer–Glas,Püttlingen, Germany) [29] with the stratum corneum facing up-wards and the dermis positioned to contact the receptor mediumDPBS (Dulbecco’s Phosphate-Buffered Saline, PAA, Pasching,Austria), which was maintained at 32 �C ± 1 �C and continuouslystirred by a magnetic bar. According to the OECD guideline 428,caffeine was used as reference probe to control the quality of theexperimental series [17].

All test formulations (10 lg/cm2, finite dose) were applied for24 h, during which aliquots of the receptor solution were collected(0, 2, 4, 6 and 24 h after dosing; under replacement) for fluores-cence analysis. The lack of a fluorescence signal in the removedreceptor medium probe proved a valid test procedure. Parallel testswere performed, i.e., all test formulations and reference substanceswere treated equally. As caffeine penetrates into the receptor med-ium and the amount must be analyzed to validate the quality of thetests, it is indispensable that the receptor medium is also removedfrom the Franz diffusion cells with a test formulation and replacedby fresh medium. Even if the liposomes do not penetrate into thereceptor solution, this must be verified by fluorescence measure-ments. As expected, fluorescence was never detected in the recep-tor medium. It could be demonstrated that no fluorescent dyereached the receptor reservoir as the follicular pathway was usedfor penetration. Because vellus hairs do not extend into the subcu-taneous fat tissue, the samples did not exhibit any holes as it isusual for split skin, indicating that the full thickness skin was notdamaged. Also, the skin sample was not shifted during massageapplication. As the data established for caffeine are comparableto those already published, the experiment was correctlyperformed.

Each experiment was conducted with at least 6 FD-C (originat-ing from 6 donor subjects) and repeated three times so that thedata were acquired over the course of 4 days. The effects of mas-sage (study design 1) and occlusion (study design 2) were derivedfrom experiments on skin samples of the same donor skin, in orderto exclude donor-related skin properties. The structural integrity ofthe liposomes was not affected by sink conditioning as the lipo-somes penetrated exclusively via the follicular shunt and did notreach the receptor reservoir. The liposomes did not break whensubjected to massage or occlusion. This was shown in study design1, in which occlusion was applied and the effects of massage wereinvestigated. If the liposomes had been broken, the skin measure-ments would have shown strong carboxyfluorescein coloration.However, the CSLM images disclosed strong rhodamine fluores-cence, but only weak carboxyfluorescein coloration. In deeper der-mal layers, weaker rhodamine fluorescence and increasedcarboxyfluorescence intensity was detectable. Intact liposomesare represented by ample red and only minute green colorationsin the images. If the liposomes break, carboxyfluorescein solutionis released from the core, and the fragments migrate into deeperregions in the hair follicle.

2.5.2. Study design 1 – Application of massage using a deviceThe massage device used was an electrical toothbrush (Braun,

Oral B, Germany), in which the brush was replaced with a metallicball that was covered with a cut-off finger of a latex glove. Thevibration of the toothbrush led to an oscillation of the metallic balland thereby enabled massaging of the topically applied test formu-lations into the skin.

After mounting the human full thickness skin on the FD-C andhomogeneous application of the liposome formulations, massagewas applied for 3 min on the test samples, whereas no massagewas applied on the control samples. Tests were regarded valid ifa test probe removed from the receptor solution yielded no fluores-cent signal. The structural integrity of the liposomes was not af-fected by sink conditioning as the liposomes penetratedexclusively via the follicular shunt and did not reach the receptorreservoir. The donor compartments of test and control sampleswere then covered with parafilm and aluminum foil to ensureocclusion (and exclusion of light) during the 24 h exposure. Atthe end of the experiments, any donor formulation found on theapplication site was removed using cotton swabs, and 8 mm punchbiopsies were taken from the skin samples and shock frozen in li-quid nitrogen for preparation of 10 lm cyro-sections. The sectionswere then analyzed using confocal laser scanning microscopy(CLSM).

Anticipating that the application of massage led to significantlyincreased penetration depths of both liposome preparations, themassage device was also applied for the study of the influence ofocclusion on the follicular penetration depth of both liposomepreparations.

2.5.3. Study design 2 – Investigation of the influence of occlusionThe influence of occlusion on the formulations was investigated

in study design 2. Both the test formulations and the reference sub-stances were investigated with and without occlusion, respec-tively, at 70% environmental humidity, corresponding to OECDguidelines. Thus, the conditions for the occluded skin samples werenot significantly different from the non-occluded samples, suchthat the behavior of the liposome types permits conclusions as toan influence of occlusion only if the humidity of the ambient airdropped significantly. The results shown in Fig. 3 suggest that flex-ible liposomes are influenced much less by the environmentalhumidity than rigid liposomes.

Investigations were performed in FD-C, whereas the liposomeapplication and massage were performed as described for studydesign 1. During the experiments, part of the FD-C was occludedby covering the donor chamber with parafilm and aluminum foilto create a microclimate and to exclude light above the epidermis,while the remaining FD-C was non-occluded, i.e., the donor cham-ber remained open.

The humidity of the environment was 70% in the experimentswith non-occluded skin as recommended [17]. At the end of theexperiment, skin samples were prepared for CLSM as describedin the study design 1.

2.5.4. Confocal laser scanning microscopy (CLSM)All cryo-sections of the skin containing hair follicles were ana-

lyzed with a CLSM 410 (Zeiss, Jena, Germany) to determine the fol-licular penetration depths of the liposomes. The field of viewmagnification of the CLSM was 250 � 250 lm. As the liposomeswere double-labeled with two fluorescent dyes, the penetrationof the constituents of the outer (rhodamine) and the inner sphere(carboxyfluorescein) of the liposomes could be detected separatelyat 570 nm and 520 nm, respectively. The penetration depths of thefluorescent dyes were determined in lm. In order to exclude theinfluence of autofluorescence of the skin, a corresponding filterwas utilized to cut off the fluorescence below 510 nm, which cor-responds to the autofluorescence of the skin.

2.5.5. StatisticsStatistical analysis was performed with the software program

IBM SPSS Statistics 19�. Explorative data analysis and the non-parametric Wilcoxon test were used to compare the penetration

* *

* * * 1800

304 S. Trauer et al. / European Journal of Pharmaceutics and Biopharmaceutics 86 (2014) 301–306

depths of the fluorescence probes of test and control samples.p 6 0.05 was considered as significant.

-M +M -M +M

*

*

* *

* *

*

Rigid-Rh Rigid-CF Rigid-Rh Rigid-CF Flex-Rh Flex-CF Flex-Rh Flex-CF0

200

400

600

800

1000

1200

1400

1600

Pene

tratio

n de

pth

[µm

]

Fig. 1. Massage effect follicular penetration depths of rigid (Rigid) and flexible(Flex) liposomes with (+M) and without (�M) massage appliance. Rh-DOPE (Rh)was used as a label for the liposome bilayer and CF for the inner, aqueous liposomecore. (n = 6; �p < 0.05).

Fig. 2. Massage effect laser scanning microscope images of the penetration offlexible liposomes: (a) penetration of rhodamine without massage, (b) penetrationof carboxyfluorescein without massage, (c) penetration of rhodamine with massage,and (d) penetration of carboxyfluorescein with massage.

3. Results

It could be demonstrated that liposomes can penetrate effec-tively into the hair follicles via the follicular penetration pathway.Fig. 2 shows that the hair follicle comprises both fluorescence dyes.Although the hydrophilic carboxyfluorescein dye was also detect-able in the tissue adjacent to the follicle, it did not penetrate intothe receptor fluid. The carboxyfluorescein was not detectable inthe receptor fluid within 24 h as the hair follicle serves as a reser-voir. As a result, the fluorescent dye rhodamine, which binds to thelipids of the liposomes, was not detected outside the follicle.

3.1. Effect of massage on the follicular penetration of rigid and flexibleliposomes

After massage application, the follicular penetration depths ofthe fluorescent dyes were determined. Rhodamine covalentlybinds to phospholipids of the liposome membrane, and thus is amarker for the lipophilic agents forming the outer sphere of the lip-osomes, while carboxyfluorescein embedded into the inner sphereof the liposomes is a marker of hydrophilic agents dissolved in theaqueous phase. For all four tested variations (rigid and flexibleliposomes with and without massage), it can be concluded thatcarboxyfluorescein penetrated significantly deeper than the rhoda-mine-conjugate, the marker for the liposome bilayer (p 6 0.05;Fig. 1). Massage application favored penetration into the depth ofthe hair follicle infundibula of both fluorescent markers in thecases of rigid as well as flexible liposomes (p 6 0.05). Independentof the massage application, the flexible liposomes penetrated sig-nificantly deeper than the rigid liposomes (p 6 0.05).

The liposomes investigated in the present study penetrated to adepth of 93.2 ± 11.7 lm (rigid liposomes) and 137.3 ± 26.5 lm(flexible liposomes) when applied without massage, meaning thatneither the sebaceous gland nor the bulge region was reached. Theapplication of massage led to a penetration depth of 477.2 ± 61 lmand 698.8 ± 90.7 lm, respectively, which enabled the target sitesto be reached.

Fig. 1 shows the preparation of fresh liposomes. The influence ofmassage on the penetration depth was investigated using occludedsamples. Most liposomes remained undamaged, penetrating to adepth of 500 lm (rigid liposomes) and 700 lm (flexible lipo-somes), respectively, within 24 h. The released carboxyfluoresceinreached penetration depths of approximately 1200 lm, indepen-dent of the type of liposomes implemented. The liposomes appliedwithout massage showed penetration depths of less than 200 lmfor rhodamine and of less than 300 lm for carboxyfluorescein,independent of the liposome types. In Fig. 2, the penetration ofboth kinds of liposomes is depicted, demonstrating that the pene-tration of carboxyfluorescein (Fig. 2b and d) was more efficientthan the penetration of rhodamine (Fig. 2a and c). In the case ofmassage application (Fig. 2c and d), the fluorescence signal couldbe detected in deeper regions of the hair follicles.

3.2. Effect of occlusion on the follicular penetration depth of rigid andflexible liposomes

The microclimate generated by occlusion leads to a strongerhydration of the skin, which in turn promotes the penetration ofhydrophilic substances. Once released from the core of the lipo-somes, carboxyfluorescein, e.g., can automatically penetrate intodeeper skin layers because the cells are hydrated and the dye candisperse. Likewise, the effect of occlusion on the follicular penetra-

tion depth of rigid and flexible liposomes was investigated. The re-sults are summarized in Fig. 3. Again, for all four tested variations(rigid and flexible liposomes with and without occlusion), hydro-philic carboxyfluorescein penetrated significantly deeper than therhodamine-conjugate (p 6 0.05). For the rigid liposomes, occlusionhad a positive effect on the penetration depth (p 6 0.05), whereasfor the flexible liposomes, occlusion did not increase the penetra-tion depth of rhodamine (p > 0.05) and even appeared to haveinhibited the penetration of carboxyfluorescein. In the case of

-O +O -O +O

*

*

*

*

*

*

*

* *

*

Rigid-Rh Rigid-CF Rigid-Rh Rigid-CF Flex-Rh Flex-CF Flex-Rh Flex-CF0

200

400

600

800

1000

1200

1400

1600

1800

Pene

tratio

n de

pth

[µm

]

Fig. 3. Occlusion effect penetration depths of rigid (Rigid) and flexible (Flex)liposomes with (+O) and without (�O) occlusion after massage application. Theouter sphere of the liposomes was labeled with Rh-DOPE (Rh), the inner sphere waslabeled with CF. (n = 6; �p < 0.05).

S. Trauer et al. / European Journal of Pharmaceutics and Biopharmaceutics 86 (2014) 301–306 305

non-occlusion, the flexible liposomes penetrated significantly dee-per into the hair follicles (p 6 0.05). For the investigations on theinfluence of occlusion, the same batch of liposomes was used asfor the investigations on the influence of massage. The investiga-tions were not conducted before the massage experiments werecompleted and evaluated.

Fig. 3 illustrates the penetration of intact and broken liposomes.The broken liposomes migrate to deeper skin layers than intact lip-osomes, which explains why rhodamine is detectable in liposomefragments up to a depth of approximately 1200 lm. In non-oc-cluded samples, the rigid liposomes exhibit similar penetrationdepths as shown in Fig. 1 for the occluded samples. This is attrib-uted to the high environmental humidity, which under occlusionand through the development of a microclimate above the dermiscan exceed 70%. The results show that the rigid liposomes desic-cate more rapidly if a microclimate cannot develop above the skin.In this case, penetration ceases earlier and at lower depths. Once amicroclimate has formed above the epidermis, the rigid liposomesexhibit penetration depths comparable to those of the flexible lip-osomes. The flexible liposomes are less influenced by occlusion,i.e., they penetrate to the same depths, whether or not they areoccluded.

4. Discussion

The movement of hair in hair follicles has been described ascrucial to stimulating follicular penetration. Recent investigationshave shown that the pumping effect of the moving hairs is mosteffective when the particle size of the topically applied substanceis in the range of the cuticula thickness of the corresponding hair[24]. Larger and smaller particles did not penetrate as efficientlyinto the hair follicles [23]. On the contrary, the penetration of lip-osomes through the stratum corneum decreases as their diameterincreases [30]. While the movement of the hair occurs physiologi-cally in vivo and is additionally supported by the user when prod-ucts are topically applied by rubbing, this phenomenon must beinduced in ex vivo studies by the application of massage. The pres-ent investigation demonstrated that the application of massagehad a significant positive effect on the follicular penetration depthfor both liposome preparations. The penetration depth was in-creased by up to a factor of 5 when massage was applied. Lade-mann et al. observed a comparable increase in the follicular

penetration depth of 320 nm-sized particles into ex vivo skin sub-sequent to massage application [25].

Moreover, the results indicate liposomal degradation as thegreen fluorescence of carboxyfluorescein, which is the label ofthe inner aqueous liposome core, was detected independent ofrhodamine-related fluorescence of the phospholipid. It can be as-sumed that after the degradation at a significant depth withinthe hair follicle, both marker dyes penetrate independently. The re-sults show that liposomes can be utilized to deliver hydrophilicsubstances such as carboxyfluorescein deeply into the hair follicles.Within the hair follicle, several target sites of interest are located,such as the sebaceous gland, the bulge region, localization of thestem cells, or the infundibulum, which represents an interruptedbarrier with increased permeability in the lower part and is sur-rounded by a high density of immune cells and an extensive capil-lary network. This capillary network offers new therapeuticoptions concerning the therapy of sebaceous gland-associateddiseases such as acne vulgaris, but also concerning topicalvaccinations or regenerative medicine [23]. It has been shownpreviously that liposomal preparations can induce better resultsin the treatment of acne vulgaris than conventional treatments[19].

Recently, Vogt et al. [31] investigated the morphometry of vel-lus and terminal hair follicles. For vellus hair follicles, presentmainly in breast and abdominal skin, the average length of theinfundibulum, e.g., at the end of which the sebaceous gland is lo-cated, is 225 ± 34 lm. Additionally, the influence of environmentalhumidity on the penetration of rigid and flexible liposomes wasanalyzed. Percutaneous absorption is generally increased whenthe site of application is occluded [26]. However, in regard to theimplementation of flexible liposomes, occlusion has been de-scribed to be counterproductive [28,32]. The results of the presentstudy revealed that occlusion had a significant positive effect onthe follicular penetration depth for rigid liposomes, whereas forthe flexible liposomes, no significant difference could be detectedfor the lipids of the outer sphere (rhodamine). This is presumablyalso the case for other lipophilic agents embedded in the liposomalshell. However, occlusion led to a slightly but significantlydecreased penetration depth of the hydrophilic markercarboxyfluorescein.

This should also remain true for other hydrophilic agents dis-solved in the aqueous core. It has been suggested [28,32] that inthe case of non-occlusive application, the increased drug transportcan be caused by a more profound interaction between the liposo-mal constituents and the skin and/or the presence of a hydrationgradient in the skin [7]. According to Cevc and Blume [32], thewater gradient is an important driving force for drug diffusion.During the occlusive application, the water gradient is expectedto be absent [28]. The same explanation might be valid for follicu-lar penetration, but this must be investigated further. The results ofthe present study allow the conclusion that massage application isan important element when investigating follicular penetration ofliposomes and other particulates ex vivo. Massage application sim-ulates the physiological movement of hair, which is important foran effective follicular delivery, and additionally imitates the appli-cation behavior of the end user. Although valid for several otherdermatological topical preparations, occlusion did not increasethe follicular penetration of the flexible liposomes. Thus, a poten-tial beneficial effect of occlusion must to be evaluated for newformulations.

Acknowledgment

We would like to thank the Foundation ‘‘Skin Physiology’’ of theDonor Association for German Science and Humanities for financialsupport.

306 S. Trauer et al. / European Journal of Pharmaceutics and Biopharmaceutics 86 (2014) 301–306

References

[1] M.B. Pierre, I. Dos Santos Miranda Costa, Liposomal systems as drug deliveryvehicles for dermal and transdermal applications, Arch. Dermatol. Res. 303(2011) 607–621.

[2] S.F. Haag, B. Taskoparan, R. Bittl, C. Teutloff, R. Wenzel, A. Fahr, M. Chen, J.Lademann, M. Schafer-Korting, M.C. Meinke, Stabilization of reactivenitroxides using invasomes to allow prolonged electron paramagneticresonance measurements, Skin Pharmacol. Physiol. 24 (2011) 312–321.

[3] S. Scalia, M. Mezzena, D. Ramaccini, Encapsulation of the UV filters ethylhexylmethoxycinnamate and butyl methoxydibenzoylmethane in lipidmicroparticles: effect on in vivo human skin permeation, Skin Pharmacol.Physiol. 24 (2011) 182–189.

[4] N. Katahira, T. Murakami, S. Kugai, N. Yata, M. Takano, Enhancement of topicaldelivery of a lipophilic drug from charged multilamellar liposomes, J. DrugTarget. 6 (1999) 405–414.

[5] M.I. Nounou, L.K. El-Khordagui, N.A. Khalafallah, S.A. Khalil, Liposomalformulation for dermal and transdermal drug delivery: past, present andfuture, Recent Pat. Drug Deliv. Formul. 2 (2008) 9–18.

[6] M. Mezei, V. Gulasekharam, Liposomes – a selective drug delivery system forthe topical route of administration. Lotion dosage form, Life Sci. 26 (1980)1473–1477.

[7] J.A. Bouwstra, P.L. Honeywell-Nguyen, Skin structure and mode of action ofvesicles, Adv. Drug Deliv. Rev. 54 (Suppl. 1) (2002) S41–S55.

[8] M. Kirjavainen, A. Urtti, R. Valjakka-Koskela, J. Kiesvaara, J. Monkkonen,Liposome–skin interactions and their effects on the skin permeation of drugs,Eur. J. Pharm. Sci. 7 (1999) 279–286.

[9] N. Weiner, L. Lieb, S. Niemiec, C. Ramachandran, Z. Hu, K. Egbaria, Liposomes: anovel topical delivery system for pharmaceutical and cosmetic applications, J.Drug Target. 2 (1994) 405–410.

[10] M.M. Elsayed, O.Y. Abdallah, V.F. Naggar, N.M. Khalafallah, Lipid vesicles forskin delivery of drugs: reviewing three decades of research, Int. J. Pharm. 332(2007) 1–16.

[11] A. Gillet, F. Lecomte, P. Hubert, E. Ducat, B. Evrard, G. Piel, Skin penetrationbehaviour of liposomes as a function of their composition, Eur. J. Pharm.Biopharm. 79 (2011) 43–53.

[12] M. Kirjavainen, A. Urtti, I. Jaaskelainen, T.M. Suhonen, P. Paronen, R. Valjakka-Koskela, J. Kiesvaara, J. Monkkonen, Interaction of liposomes with human skinin vitro – the influence of lipid composition and structure, Biochim. Biophys.Acta 1304 (1996) 179–189.

[13] G. Cevc, A. Schatzlein, G. Blume, Transdermal drug carriers – basic properties,optimization and transfer efficiency in the case of epicutaneously appliedpeptides, J. Control. Release 36 (1995) 3–16.

[14] L. Ballam, C.M. Heard, Pre-treatment with aloe vera juice does not enhance thein vitro permeation of ketoprofen across skin, Skin Pharmacol. Physiol. 23(2010) 113–116.

[15] A. Otto, J.W. Wiechers, C.L. Kelly, J.C. Dederen, J. Hadgraft, J. du Plessis, Effect ofemulsifiers and their liquid crystalline structures in emulsions on dermal andtransdermal delivery of hydroquinone, salicylic acid and octadecenedioic acid,Skin Pharmacol. Physiol. 23 (2010) 273–282.

[16] S. Trauer, J. Lademann, F. Knorr, H. Richter, M. Liebsch, C. Rozycki, G. Balizs, R.Buttemeyer, M. Linscheid, A. Patzelt, Development of an in vitro modified skin

absorption test for the investigation of the follicular penetration pathway ofcaffeine, Skin Pharmacol. Physiol. 23 (2010) 320–327.

[17] R. Bussiahn, R. Brandenburg, T. Gerling, E. Kindel, H. Lange, N. Lembke, K.D.Weltmann, T. von Woedtke, T. Kocher, The hairline plasma: an intermittentnegative dc-corona discharge at atmospheric pressure for plasma medicalapplications, Appl. Phys. Lett. 96 (2010).

[18] G.M. El Maghraby, A.C. Williams, B.W. Barry, Skin hydration and possible shuntroute penetration in controlled estradiol delivery from ultradeformable andstandard liposomes, J. Pharm. Pharmacol. 53 (2001) 1311–1322.

[19] G.M. El Maghraby, A.C. Williams, B.W. Barry, Can drug-bearing liposomespenetrate intact skin?, J Pharm. Pharmacol. 58 (2006) 415–429.

[20] N. Otberg, H. Richter, H. Schaefer, U. Blume-Peytavi, W. Sterry, J. Lademann,Variations of hair follicle size and distribution in different body sites, J. Invest.Dermatol. 122 (2004) 14–19.

[21] B. Lange-Asschenfeldt, D. Marenbach, C. Lang, A. Patzelt, M. Ulrich, A.Maltusch, D. Terhorst, E. Stockfleth, W. Sterry, J. Lademann, Distribution ofbacteria in the epidermal layers and hair follicles of the human skin, SkinPharmacol. Physiol. 24 (2011) 305–311.

[22] N. Luther, M.E. Darvin, W. Sterry, J. Lademann, A. Patzelt, Ethnic differences inskin physiology, hair follicle morphology and follicular penetration, SkinPharmacol. Physiol. 25 (2012) 182–191.

[23] A. Patzelt, H. Richter, F. Knorr, U. Schafer, C.M. Lehr, L. Dahne, W. Sterry, J.Lademann, Selective follicular targeting by modification of the particle sizes, J.Control. Release 150 (2011) 45–48.

[24] J. Lademann, A. Patzelt, H. Richter, C. Antoniou, W. Sterry, F. Knorr,Determination of the cuticula thickness of human and porcine hairs andtheir potential influence on the penetration of nanoparticles into the hairfollicles, J. Biomed. Opt. 14 (2009) 021014.

[25] J. Lademann, H. Richter, A. Teichmann, N. Otberg, U. Blume-Peytavi, J. Luengo,B. Weiss, U.F. Schaefer, C.M. Lehr, R. Wepf, W. Sterry, Nanoparticles – anefficient carrier for drug delivery into the hair follicles, Eur. J. Pharm.Biopharm. 66 (2007) 159–164.

[26] P. Treffel, P. Muret, P. Muret-D’Aniello, S. Coumes-Marquet, P. Agache, Effect ofocclusion on in vitro percutaneous absorption of two compounds withdifferent physicochemical properties, Skin Pharmacol. 5 (1992) 108–113.

[27] Y. Shimoyama, M. Fujii, Y. Kanda, A. Mizoguchi, H. Oda, N. Koizumi, Y.Watanabe, Effects of application method on skin penetration ofcarboxyfluorescein incorporated in liposomes, Chem. Pharm. Bull. (Tokyo) 58(2010) 429–431.

[28] M.E.M.J. van Kuijk-Meuwissen, H.E. Junginger, J.A. Bouwstra, Interactionsbetween liposomes and human skin in vitro, a confocal laser scanningmicroscopy study, Bba-Biomembranes 1371 (1998) 31–39.

[29] T.J. Franz, Percutaneous absorption on the relevance of in vitro data, J. Invest.Dermatol. 64 (1975) 190–195.

[30] J. de Leeuw, H.C. de Vijlder, P. Bjerring, H.A. Neumann, Liposomes indermatology today, J. Eur. Acad. Dermatol. Venereol. 23 (2009) 505–516.

[31] A. Vogt, S. Hadam, M. Heiderhoff, H. Audring, J. Lademann, W. Sterry, U. Blume-Peytavi, Morphometry of human terminal and vellus hair follicles, Exp.Dermatol. 16 (2007) 946–950.

[32] G. Cevc, G. Blume, Lipid vesicles penetrate into intact skin owing to thetransdermal osmotic gradients and hydration force, Biochim. Biophys. Acta1104 (1992) 226–232.