Journal of Controlled Release 56 (1998) 189–196

Application of vesicles to rat skin in vivo: a confocal laser

scanning microscopy study.
Marly E.M.J. van Kuijk-Meuwissen, Luc Mougin, Hans E. Junginger, Joke A. Bouwstra*
Division of Pharmaceutical Technology, Leiden /Amsterdam Center of Drug Research ( LACDR), Leiden University, P.O. Box 9502,
2300 RA Leiden, The Netherlands
Accepted 14 May 1998

Abstract

A major problem in (trans)dermal drug delivery is the low penetration rate of most substances through the barrier of the
skin, the stratum corneum. One of the methods to increase the penetration rate across the skin is encapsulation of a (model)
drug in lipid vesicles. In this study fluorescently labelled liposomes were applied on rat skin, in vivo. Bilayer labelled
gel-state and liquid-state liposomes (conventional or with flexible bilayers) were non-occlusively applied on the dorsal area
in the neck of the rat for 1, 3 or 6 h. Micelles were used as a control formulation. The penetration pathway and penetration
depth of the lipophilic fluorescent label into the skin was visualised by confocal laser scanning microscopy (CLSM). During
the first 3 h of application almost no differences in penetration depth were observed, when the label was applied in the
various formulations. After 6 h application, it was clear that the label applied in micelles and gel-state liposomes did not
penetrate as deep into the skin as the label applied in liquid-state vesicles. Among the liquid-state vesicles, the suspension
with the most flexible bilayers showed the highest fluorescence intensity in the viable epidermis and dermis, 6 h
post-application. Thus the vesicular form and the thermodynamic state of the bilayer and to a smaller extent the flexibility of
the bilayer influence the penetration depth of the label into the skin at longer application periods. These results are in good
agreement with CLSM results obtained from in vitro experiments with human skin.  1998 Elsevier Science B.V. All rights
reserved.

Keywords: Liposomes; Dermal delivery; Confocal laser scanning microscopy; In vivo

1. Introduction of differentiation. During the differentiation process

the cells migrate from the basal layer to the surface
The skin can be divided into two distinct layers; and cornify. The uppermost layer of the skin is the
the dermis and epidermis. The dermis consists of stratum corneum, which consists of flattened cor-
connective tissue and contains nerves, blood and neocytes embedded in lipid lamellar sheets.
lymph vessels, hair follicles, sebaceous and sweat The stratum corneum is known as the rate limiting
glands. The epidermis contains cells in several stages step for drug penetration through the skin. This
barrier is the major obstacle when attempting to
*Corresponding author. Tel.: 131 71 5274208; fax: 131 71 deliver drugs transdermally. One of the methods to
5274565 enhance drug penetration across the skin is to

0168-3659 / 98 / $ – see front matter  1998 Elsevier Science B.V. All rights reserved.
PII: S0168-3659( 98 )00087-X
190 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196

encapsulate the drug in vesicular carriers, like lipo- sources [17–19], reflectance imaging does not lean
somes. itself for detection of permeating compounds.
The first report on this subject was by Mezei and In the present study various types of fluorescently
Gulasekharam [1] in 1980, after which many other labelled liposomes were non-occlusively applied on
publications followed in this field (for a review see rat skin, in vivo. After various application periods
[2]). the rats were sacrificed and the treated skin was
Studies which are carried out for investigating the visualised perpendicular to the skin surface, using
skin-vesicle interactions are among others diffusion CLSM in combination with the cross-section device
experiments [3–5], visualisation studies by electron [20]. This mechanical cross-section method has the
microscopy [6–8] and by fluoromicrography [9–11]. advantage to visualise fluorescence in the stratum
The latter technique revealed that the fluorescent corneum, viable epidermis and dermis in one single
label, which was intercalated in the liposomal image. Hence the penetration pathway and penetra-
bilayer, remained in the stratum corneum [9,10] or tion depth of the bilayer marker into the skin can be
penetrated deeper in the epidermis predominantly examined simultaneously.
along the hair-shafts [11]. A disadvantage of the The penetration of the phospholipid bound label
fluoromicrograph technique is that the tissue needs to into the skin, applied in the various formulations,
be cryofixed, which may change skin lipid organisa- was compared to investigate the effect of (i) the
tion or may result in redistribution of the label [12]. thermodynamic state of the liposomes and (ii) the
The major advantage of confocal laser scanning flexibility of the bilayers of the several liquid-state
microscopy (CLSM) is that the tissue can be opti- liposomes and (iii) the vesicular shape.
cally sectioned and that the distribution of the
fluorescent probe in the sample can be visualised by
images parallel to the surface of the sample, without
2. Materials and methods
cryofixing or embedding the tissue. Like other
techniques in which a label is used, CLSM provides
only information about the localisation of the label 2.1. Materials
and not about the localisation of the entire liposome.
In recent studies on transdermal delivery by 1, 2 - Dilauroyl - sn - glycero - 3 - phosphatidylcholine
vesicles, a fluorescent probe was incorporated in the (DLPC) and 1,2-distearoyl-sn-glycero-3-phospha-
vesicular bilayer to track permeation [13,14]. Fol- tidylcholine (DSPC) were kind gifts from Natter-
lowing non-occlusive application of the vesicles, the mann Phospholipids (Cologne, Germany). Heptaox-
distribution of the probe in the skin was visualised yethylenelauryl-ether (C 12 EO 7 ) was purchased
by confocal optical sections made parallel to the from Servo (Delden, The Netherlands). N-(5-
plane of the skin surface. Zellmer et al. [13] incu- fluoresceinthiocarbamoyl) - 1,2 - dihexadecanoyl - sn-
bated human mammae skin in vitro for 18 h and glycero-3-phospho-ethanolamine (Fl-DHPE) and N-
could only detect the label on the surface of the skin. octadecyl-N9-(5-(fluoresceinyl))-thiourea (Fl-C 18 )
¨
Schatzlein and Cevc [14] applied flexible vesicles were both from Molecular Probes (Eugene,
(transfersomes TM ) on murine skin in vivo and found OR, USA). N-(2-hydroxyethyl)piperazine-N9-(2-
that the fluorescent label could be visualised by ethanesulfonic acid) ( HEPES ) and cholesterylhemisuc-
CLSM, until an apparent depth of 25 mm. The major cinate (CHEMS) were obtained from Sigma (Hilver-
disadvantage of this method is the progressive loss of sum, The Netherlands) and sodium chloride was
resolution and signal with depth in the tissue, due to purchased from Merck (Darmstadt, Germany).
scattering and absorption of both the laser excitation Ketamine (Ketalar, 10 mg / ml) was obtained from
and fluorescence emission light [15,16]. Parke Davis BV (Amsterdam, The Netherlands) and
While skin histology has been visualised in vivo rompun (5xylazine) (Rompun R ; 23.32 mg / ml 2-
by reflectance CLSM to a depth of 120–150 mm (2,6 - xylidino) - 5,6 - dihydro-4H - 1,3 - thiazine-hydro-
below the surface, using either white light or laser chloride) was supplied by Bayer (Leverkusen, Ger-
 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196 191

many) and atropine was from Brocacef (Maarsen, pared at approximately 658C, which is above the
The Netherlands). transition temperature of DSPC (568C). The final
lipid concentration of the liposomes was 50 mg / ml
2.2. Preparation of liposomes and the concentration of the fluorescent label was
2.75 mol %. The size and polydispersity were de-
The formulations used in this study are presented termined by dynamic light scattering (Malvern
in Fig. 1. Lipids, surfactant and the lipophilic 4700C, Malvern Instruments, Malvern, UK). The
fluorescent marker (Fl-DHPE) were dissolved in measurements were performed at 278C at an angle of
dichloromethane:methanol (3:1). After evaporation 908 between laser and detector.
of the solvent under vacuum (Speed Vac Concen- Mixing of C 12 EO 7 with Fl-DHPE resulted in the
trator, Savant, Dunee, Soest, NL), the lipid film was formation of vesicles, instead of micelles, and there-
hydrated with HEPES buffer at pH57.4 (20 mM HEPES, fore Fl-C 18 was used as a single chain fluorescent
135 mM NaCl). The sample was sonicated using a marker in the micelles. Before using the fluorescent
sonication bath (Transonic 460 / H, Elma, Singen, dye, its fluorescent intensity was compared to that of
Germany) at room temperature, except for the Fl-DHPE and appeared to be similar.
DSPC:CHEMS liposomes. This suspension was pre-
2.3. Animal experiments

The protocol for these experiments was approved

by the Leiden University ethical committee for
animal experimentation. Male Wistar rats ((WI)WU
BR) with SPF status were obtained from Broekman
Instituut BV, Someren, The Netherlands (via Charles
River Deutschland). The rats were housed for at least
one week in the animal facilities before the experi-
ments started. The animals received standard labora-
tory chow and water ad libitum, and were weighting
between 200 and 250 g. One day before the experi-
ment, the rats were individually placed in cages,
which were modified in such way that contact with
the application zone in the neck could be prevented.
The hair of the dorsal area of the neck was clipped
carefully under KRA anaesthesia (intraperitoneal
injection of ketamine 50 mg / kg, Rompun R 5 mg / kg
and atropine 0.05 mg / kg). Directly after clipping,
the fluorescently labelled liposomes or micelles were
non-occlusively applied on the skin (25 ml / 0.75
cm 2 ). The rat was lying on his abdomen on a
thermostated rug during the anaesthesia. Generally
the application zone was dry before the narcosis
ended, after approximately half an hour.
After the application period of 1, 3 or 6 h, the rats
Fig. 1. Summarized scoring data from CLSM pictures. CLSM (n52) were killed by a mixture of CO 2 and O 2 ,
images are from cross-sections of rat skin which was non-occlu- followed by cervical dislocation. The excess of
sively treated in vivo with fluorescently labelled liposomes or
micelles for 1, 3 or 6 h. The images were scored on penetration suspension was removed from the application zone
depth of the label and fluorescence intensities in stratum corneum, by dabbing three times with a water moisturised
viable epidermis and dermis. tissue. A piece of skin including the application zone
192 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196

was excised from the rat and fixed on a support for 3.2. Comparison between the various formulations
dermatomisation (Deca Dermatome, DePuy Health-
care, Leeds, UK). The dermatomed skin with a In Fig. 1 the results of the scoring are summarised
thickness of approximately 200 mm was immediately for each formulation and application period. The
examined by CLSM (see below). various liposome suspensions and micelles can be
compared with regard to the penetration depth of the
label and the intensity of the fluorescence in the
2.4. Confocal Laser Scanning Microscopy ( CLSM)
various skin layers. During the first 3 h the penetra-
tion of the label applied in the various suspensions
The liposome treated skin was cross-sectioned
was very similar, the label remained in the stratum
mechanically by means of a special designed cross-
corneum.
section device [20]. The cross-sectioned skin was
Fig. 2 depicts representative images of rat skin,
mounted in a sample holder in such a way that the
which was incubated with the different types of
freshly obtained cutting surface was positioned
liposomes or micelles for 6 h. Most images have an
against the coverglass. The mechanical cross-section
intense white band on the left-hand side corre-
of the skin was examined by CLSM, approximately
sponding to the skin surface. On the right-hand side
10 mm below the cutting surface to avoid interfer-
the area is less bright, which correlates to the dermal
ence from fluorescence of damaged cells.
side of the skin. When DLPC:C 12 EO 7 (60:40) was
The CLSM used was a BioRad MRC 600 confocal
applied to the skin, the fluorescent label can be
unit (BioRad, Hertfordshire, UK) equipped with a
detected in the stratum corneum (SC), viable epi-
Krypton–Argon laser and mounted on a Nikon
dermis (E) and dermis (D). In some images a striped
Optiphot (Nikon, Tokyo, Japan). For excitation of
structure within the white band can be observed.
the label the 488 nm laser line was used. The
This relates to the flattened stratum corneum cells
fluorescein–DHPE or the fluorescein-C 18 was de-
(corneocytes) which are embedded in lipid lamellae.
tected using the BioRad Blue High Sensitivity
The liposome size ranged from 70 to 250 nm, in this
(BHS) filterblock, which passes emitted light with a
size range no effect of size on the penetration of the
wavelength longer than 515 nm. Confocal images
fluorophore was observed in our studies (unpublished
were obtained using a Zeiss Plan Neofluor 253 / 0.8
results using in vitro human skin).
multi immersion objective, on its oil position. The
images were corrected for auto fluorescence of rat
skin, with the black level setting. All the pictures
3.2.1. Comparison between gel-state and liquid-
were averages of ten scans.
state liposomes
At skin temperature (328C), DSPC:CHEMS form
2.5. Scoring of the CLSM images gel-state liposomes, while DLPC:CHEMS liposomes
are in liquid-state [21]. The former were prepared
The fluorescence intensities of the CLSM pictures from phospholipids with a longer chain length and
were semi-quantitatively scored by three individuals consequently these liposomes have more rigid
and classified as follows: 2 no fluorescence, 6 weak bilayers than the liquid-state vesicles. These lipo-
fluorescence, 1 bright fluorescence. somes were used to study the effect of the thermo-
dynamic state of the bilayers on the penetration of
the label into the skin.
After 1 and 3 h of application the fluorescent label
3. Results and discussion
applied in the gel-state liposomes remained in the
stratum corneum, just like the label applied in
3.1. Size of the formulations DLPC:CHEMS (82:18) vesicles (Fig. 1). After 6 h
of application, a small amount of the dye applied in
The mean diameter of the various liposome sus- the gel-state liposomes was visualised in the viable
pensions ranged between 70 and 250 nm. epidermis, while the dye intercalated in the liquid-
 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196 193

Fig. 2. Representative CLSM images of a cross-section of rat skin treated with the various fluorescently labelled liposomes and micelles for
6 h; DLPC:C 12 EO 7 (60:40), DSPC:CHEMS (82:18), DLPC:CHEMS (82:18), DLPC:C 12 EO 7 (82:18) and C 12 EO 7 micelles. A high
magnification of skin (bottom-right) incubated with micelles for 6 h demonstrates that the fluorescent label is present around the
corneocytes. Stratum corneum (SC), viable epidermis (E) and dermis (D).

state suspension was visible in the viable epidermis 3.2.2. Comparison between the various liquid-state
and dermis (Fig. 2). The difference in penetration liposomes
depth of the label applied in gel-state versus liquid- A single chain surfactant, C 12 EO 7 , was incorpo-
state liposomes was also observed in the in vitro rated in liquid-state liposomes in order to create
CLSM study in which the same type of formulations deformable bilayers [14,23,24]. These
was applied on freshly excised human skin [22]. DLPC:C 12 EO 7 vesicles with a size .100 nm could
From these observations it can be concluded that be extruded through a 50 nm polycarbonate filter
the label applied in liquid-state vesicles penetrated using nitrogen pressure of 2 bar and they appeared to
deeper into the skin than when applied in gel-state be more deformable than conventional
vesicles. DLPC:CHEMS (82:18) liposomes [24]. The influ-
194 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196

ence of these flexible vesicles on transdermal label corneocytes can be observed. The fluorescent label is
penetration was investigated by using two suspen- present around these flattened cells and probably
sions with different phospholipid / surfactant molar associated with the lipid lamellae.
ratios; DLPC:C 12 EO 7 (60:40) and DLPC:C 12 EO 7 As shown in Fig. 1, the label applied in micelles
(82:18). had the same penetration profile as the label interca-
After the application periods of 1 or 3 h, the label lated in gel-state liposomes. Furthermore, although
in the DLPC:CHEMS (82:18) or the DLPC:C 12 EO 7 micelles had a smaller size than liposomes, the label
(60:40) liposomes could only be detected in the when applied in the micelles penetrated slower into
stratum corneum. However, when DLPC:C 12 EO 7 the skin than when it was applied in DLPC:C 12 EO 7
(82:18) vesicles were applied on the skin also a (82:18) vesicles. Additionally, the dye when applied
small amount of the dye could be observed in the encapsulated in the micelles penetrated less deep into
viable epidermis. But 6 h post-application, the dye the skin compared to the dye intercalated in
applied in the DLPC:C 12 EO 7 (82:18) liposome DLPC:C 12 EO 7 (60:40) liposomes, 6 h post-applica-
suspension did not penetrate further than the viable tion. These findings are in agreement with the
epidermis. In contrast, weak and bright fluorescence observations in the in vitro CLSM study using
were observed in the dermis, after administration of ¨
human skin [22] and studies performed by Schatzlein
DLPC:CHEMS (82:18) and DLPC:C 12 EO 7 (60:40) and Cevc [14], who reported that the label applied in
liposomes, respectively. micelles remained at the uppermost part of the
Thus the label applied in DLPC:C 12 EO 7 (60:40) stratum corneum. In recent in vitro studies, estradiol
liposomes, with the most flexible bilayers [24], loaded vesicles were compared to estradiol loaded
penetrated deeper and to a higher extent into rat skin micelles [25]. This study revealed that when es-
than when the label was applied in the other liquid- tradiol was incorporated in vesicles during applica-
state liposomes. This is in agreement with the in tion, the estradiol transport through the stratum
vitro CLSM results [22], although in the in vitro corneum was increased compared to estradiol applied
study the differences between the three liquid-state in micelles. This is in accordance with the findings in
suspensions were not as impressive as in this in vivo the present study. These studies suggest that the
study. vesicles have advantageous for increasing the pene-
¨
Schatzlein and Cevc reported similar CLSM re- tration depth of lipophilic substances into the skin.
sults, when they applied vesicles on murine skin in
vivo [14]. However, they described an important 3.3. Do liposomes penetrate into the skin as intact
discrepancy in penetration depth of the label, which vesicles?
was applied in transfersomes TM compared to applica-
tion in conventional liposomes. The disagreement in Since only the fluorescent marker can be visual-
the findings between the experiments carried out by ised by CLSM and not the entire liposome, CLSM
¨
Cevc and Schatzlein and the present study may be does not provide reliable information about the
due to (i) different composition of the vesicles or (ii) question whether or not liposomes can penetrate as
differences between rat and murine skin. intact vesicles. There are at least three mechanisms
by which the label can penetrate into the skin: (i) the
3.2.3. Comparison between micelles and vesicles label penetrates associated with the intact liposomal
The effect on label penetration of pure C 12 EO 7 , bilayer (penetration of intact vesicles), (ii) the label
forming micelles in the buffer solution, was studied penetrates associated with liposomal bilayer frag-
in order to establish whether C 12 EO 7 micelles have ments (which means that the liposomes disrupt) or
the same effect on label penetration into the skin as (iii) the label penetrates solitary. The phospholipids
the DLPC:C 12 EO 7 liposome suspensions or that the and / or surfactants penetrate molecularly dispersed
presence of vesicles is responsible for the increase in and might act as penetration enhancers. Recent
penetration of the label. studies [24] with electron microscopy using similar
Fig. 2 (bottom-right) shows a high magnification liposomes as have been used in the present study, did
of skin treated with micelles, in which the individual not reveal intact vesicles in the skin. From these
 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196 195

observations we can conclude that the label does not Influence of Particle Size of Liposomes on the Deposition of
Drug into Skin, Int. J. Pharm. 103 (1994) 277–282.
penetrate associated with intact vesicles in the skin.
[4] J. Du Plessis, K. Egbaria, N. Weiner, Influence of formula-
Whether or not the label is associated with bilayer tion factors on the deposition of liposomal components into
fragments or penetrates solitary cannot be conclude the different stata of the skin, J. Soc. Cosmet. Chem. 43
from these studies. However using electron micros- (1992) 93–100.
copy, it was clear that the DLPC:C 12 EO 7 liposomes [5] H.Y. Yu, H.M. Liao, Triamcinolone permeation from differ-
ent liposome formulations through rat skin in vitro, Int. J.
do interact with the bilayers in the stratum corneum Pharm. 127 (1996) 1–7.
and that lipids and / or surfactants penetrate into the [6] H.E.J. Hofland, J.A. Bouwstra, H.E. Bodde, ´ et al., Interac-
deeper layers of the stratum corneum [24]. tions between liposomes and human stratum corneum in
vitro: freeze fracture electron microscopical and small angle
X-ray scattering studies, Br. J. Dermatol. 132 (1995) 853–
866.
4. Conclusions [7] D.A. Van Hal, Non-ionic surfactant vesicles for dermal and
transdermal drug delivery. Ph.D. Thesis, Leiden University,
From the present study it can be concluded that Leiden, The Netherlands, 1994.
[8] M. Schaller, H.C. Korting, Interaction of liposomes with
vesicles are more favourable than a micellar solution,
human skin: The role of the stratum corneum, Adv. Drug
for an increased (model) drug penetration into the Deliv. Rev. 18 (1996) 303–309.
skin. Furthermore also the thermodynamic state of [9] J. Lasch, R. Laub, W. Wohlrab, How deep do intact lipo-
the liposomes play an important role in (trans)dermal somes penetrate into human skin?, J. Control. Rel. 18 (1991)
penetration of model compounds. A lipophilic sub- 55–58.
[10] ¨
K. Kriwet, C.C. Muller-Goymann, Diclofenac release from
stance applied in liquid-state liposomes penetrates
phospholipid drug systems and permeation through excised
deeper into the skin, than when the substance is human stratum corneum, Int. J. Pharm. 125 (1995) 231–242.
applied in gel-state vesicles. An increase in de- [11] D. Yarosh, C. Bucana, P. Cox, et al., Localization of
formability of the liposomal bilayer (e.g. liposomes containing a DNA repair enzyme in murine skin,
DLPC:C 12 EO 7 (60:40) liposomes) resulted in a J. Invest. Dermatol. 103 (1994) 461–468.
[12] D.M. Shotton, Confocal scanning optical microscopy and its
better penetration of the lipophilic probe compared to
applications for biological specimens, J. Cell Sci. 94 (1989)
less flexible liquid-state liposomes. Finally, it can be 175–206.
stated that the in vitro CLSM results with human [13] S. Zellmer, W. Pfeil, J. Lasch, Interaction of phosphatidyl-
skin are confirmed by the present reported in vivo choline liposomes with the human stratum corneum, Bio-
CLSM results with rat skin. However, CLSM does chim. Biophys. Acta 1237 (1995) 176–182.
[14] ¨
A. Schatzlein, G. Cevc, in: G. Cevc, F. Paltauf (Eds.),
not provide reliable information about the question
Proceedings of 6th International Colloquium on Phos-
whether or not liposomes can penetrate as intact pholipids; Phospholipids: Characterization, Metabolism and
vesicles. Novel Biological Applications, AOCS Press, Champaign,
Illinois, USA, 1995, pp. 189–207.
[15] A.J. Hoogstraate, C. Cullander, J.F. Nagelkerke et al, Diffu-
sion rates and transport pathways of fluorescein isothio-
Acknowledgements
cyanate (FITC)-labeled model compounds through buccal
epithelium, Pharm. Res. 11 (1994) 83–89.
The generous gift of the phospholipids by Natter- [16] M. Laurent, G. Johannin, N. Gilbert et al, Power and limits
mann Phospholipid GmbH was greatly appreciated. of laser scanning confocal microscopy, Biol. Cell 80 (1994)
229–240.
[17] ´ ˆ
P. Corcuff, C. Bertrand, J.L. Leveque, Morphometry of
human epidermis in vivo by real-time confocal microscopy,
References Arch. Dermatol. Res. 285 (1993) 475–481.
[18] ´
P. Corcuff, G. Gonnord, G.E. Pierard, et al., In vivo confocal
[1] M. Mezei, V. Gulasekharam, Liposomes – A selective drug microscopy of human skin: a new design for cosmetology
delivery system for the topical route of administration: I and dermatology, Scanning 18 (1996) 351–355.
Lotion dosage form, Life Sci. 26 (1980) 1473–1477. [19] M. Rajadhyaksha, M. Grossman, D. Esterowitz, et al., In
[2] H. Schreier, J.A. Bouwstra, Liposomes and niosomes as vivo confocal scanning laser microscopy of human skin:
topical drug carriers – Dermal and transdermal drug deliv- Melanin provides strong contrast, J. Invest. Dermatol. 104
ery, J. Control. Rel. 30 (1994) 1–15. (1995) 946–952.
[3] J. Du Plessis, C. Ramachandran, N. Weiner, et al., The [20] M.E.M.J. Meuwissen, J. Janssen, C. Cullander, et al., A
196 M.E.M. J. van Kuijk-Meuwissen et al. / Journal of Controlled Release 56 (1998) 189 – 196

cross-section device to improve visualisation of fluorescent Brain, V. J. James, K. A. Walters (Eds.), Prediction of
probe penetration into the skin by confocal laser scanning Percutaneous Penetration, 3rd International Conference, STS
microscopy, Pharm. Res. 15 (1998) 352–356. Publishing, Cardiff, 1993, pp. 226–236.
[21] B.L. Silver, The Physical Chemistry of Membranes, Pub- [24] M.E.M.J. van Kuijk-Meuwissen, Liposomes and phos-
lisher Creative Services Inc., 1985, pp. 33–56. pholipid / surfactant vesicles – in vitro and in vivo skin
[22] M.E.M.J. van Kuijk-Meuwissen, H.E. Junginger, J.A. Bouw- application. Ph.D. Thesis, Leiden University, Leiden, The
stra, Interactions between liposomes and human skin in vitro: Netherlands, 1998.
a confocal laser scanning microscopy study. Biochim. Bio- [25] D.A. van Hal, A. van Rensen, T. De Vringer, Diffusion of
phys. Acta, in press. estradiol from non-ionic surfactant vesicles through human
¨
[23] G. Cevc, A. Schatzlein, D. Gebauer, G. Blume, in: K.R. stratum corneum in vitro, STP Pharma Sci. 6 (1996) 72–78.