Synthesis and properties of isocannabinoid and cholesterol derivatized rhamnosurfactants: Application to liposomal targeting of keratinocytes and skin

Article abstract of DOI:10.1016/j.ejmech.2005.04.009

The usefulness of vesicles to cargo material depends on the design of new ligands able to incorporate easily inside the bilayer and also to direct the vesicles to the targeted site. Therefore, the synthesis of two new rhamnose-bearing surfactants is described. The hydrophobic part consists of cholesterol (in compound 3) and citrylidene phloroglucinol (in compound 6). The ability of these two rhamnolipids to incorporate into a DPPC membrane and to form aggregates is investigated, respectively, by differential scanning calorimetry and by surface tension measurements. Those two new surfactants were incorporated in fluorescent liposomes to study their interactions with keratinocytes and skin sections. Intraliposomal delivery to keratinocytes was observed in both cases, even if the kinetics of delivery were different according to the rhamnosurfactant used. Skin sections were stained by both liposomal formulations, and different interactions between the liposomes and skin cells according to the surfactant used were noted.

Full text of DOI:10.1016/j.ejmech.2005.04.009

European Journal of Medicinal Chemistry 40 (2005) 1022–1029
www.elsevier.com/locate/ejmech
Original article
Synthesis and properties of isocannabinoid and cholesterol derivatized
rhamnosurfactants: application to liposomal targeting
of keratinocytes and skin
Véronique Barragan-Montero a,*, Jean-Yves Winum a, Jean-Pierre Molès b, Emmanuelle Juan a,
Caroline Clavel a, Jean-Louis Montero a
a Laboratoire de Chimie Biomoléculaire, UMR 5032, Université Montpellier II, ENSCM, 8, rue de l’Ecole Normale, 34296 Montpellier cedex 5, France
b Laboratoire de Dermatologie Moléculaire, EA no. 3754, Institut Universitaire de Recherche Clinique, 641, avenue du Doyen G. Giraud,
34093 Montpellier cedex 5, France
Received 9 February 2005; received in revised form 18 April 2005; accepted 22 April 2005
Available online 13 June 2005
Abstract
The usefulness of vesicles to cargo material depends on the design of new ligands able to incorporate easily inside the bilayer and also to
direct the vesicles to the targeted site. Therefore, the synthesis of two new rhamnose-bearing surfactants is described. The hydrophobic part
consists of cholesterol (in compound 3) and citrylidene phloroglucinol (in compound 6). The ability of these two rhamnolipids to incorporate
into a DPPC membrane and to form aggregates is investigated, respectively, by differential scanning calorimetry and by surface tension
measurements. Those two new surfactants were incorporated in fluorescent liposomes to study their interactions with keratinocytes and skin
sections. Intraliposomal delivery to keratinocytes was observed in both cases, even if the kinetics of delivery were different according to the
rhamnosurfactant used. Skin sections were stained by both liposomal formulations, and different interactions between the liposomes and skin
cells according to the surfactant used were noted.
© 2005 Elsevier SAS. All rights reserved.
Keywords: Rhamnosurfactant; Cannabinoid; Cholesterol; Liposomes; Skin
1. Introduction
mulation in specific cells or tissues involves the presence of
surface-associated targeting molecules known as ligands [4].
The ligands are natural or synthetic surfactants that are able
(a) to recognize specific cells [5]; (b) to incorporate easily
inside the liposome bilayer [6]. For example, a-L-rhamnose
cell surface receptors are present on the surface of kerati-
nocytes [7]. Therefore a-L-rhamnose bearing surfactants can
be used to target liposomes toward these cells. To our knowl-
edge, no publication dealing with such synthetic rhamnosur-
factants [8] except alkylrhamnosides [9] can be found in the
open literature. In contrast, biological produced rhamnolip-
ids have been largely reported [10].
The ability of cholesterol to bind with phospholipids in a
membrane [11] prompted us to synthesize a cholesteryl–
rhamnose surfactant (rhamnosurfactant 3). In order to enlarge
the investigation on interactions between the liposome bilayer
and the ligand, we also focused our attention on the use of a
new hydrophobic moiety and a citrylidene phloroglucinol (in
The past 30 years have seen new vesicles prepared with
synthetic lipids for drug delivery purposes [1,2]. These closed
vesicles can entrap water-soluble substances in the inner aque-
ous phase or lipophilic substances in the bilayer. Liposomal
drugs [3] have proven their clinical utility, but tremendous
efforts have still to be made to improve their efficiency, par-
ticularly in cell targeting. A liposome designed to serve as a
vehicle for bioactive compounds has to be stable under physi-
ological conditions, non-damaging for the environment and,
in order to be totally efficient, able to target specific cells or
tissues. Targeting can enhance drug uptake by the selected
cells and reduce drug concentration into the other cells. The
development of new liposomes which can increase drug accu-
* Corresponding author. Tel./fax: +33 4 67 14 43 43.
E-mail address: vbarragan@univ-montp2.fr (V. Barragan-Montero).
0223-5234/$ - see front matter © 2005 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2005.04.009

V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
1023
Fig. 1. Rhamnosurfactants 3 and 6.
rhamnosurfactant 6) was prepared as isocannabinoid deriva-
tives are also believed to interact with lipids in membranes
[12] (Fig. 1).
refluxing with triethyleneglycol in dioxane [14].A Fisher reac-
tion [15] with a-L-rhamnose led to the a anomer of rhamno-
surfactant 3 in 30% yield Scheme 1.
How these variations in the structure would influence the
surfactant properties had to be investigated. Therefore, the
amphiphilic properties of surfactants 3 and 6 were studied
using two different methods: (a) tensiometric determination
of the critical micelle concentration (CMC) and (b) thermo-
tropic phase transitions of 1,2-dipalmitoyl-rac-glycero-3-
phosphocholine (DPPC) membranes containing 3 and 6 by
differential scanning calorimetry (DSC). Those two new sur-
factants were incorporated in liposomes to study their inter-
action with keratinocytes and skin sections. Liposomal deliv-
ery to keratinocytes and skin section was done using three
types of palmitoyl-oleoylphosphatidylcholine/palmitoyl-
oleoylphosphatidylglycerol (POPC/POPG) liposomes: a)
liposomes loaded with rhamnosurfactant 3, (b) liposomes
loaded with rhamnosurfactant 6, and (c) for comparison pur-
poses, liposomes loaded with a cholesteryl succinate [7]
instead of one of the rhamnose ligand.
Under acidic conditions, citrylidene phloroglucinol leads
to a new isocannabinoid due to the opening of the pyrane
ring [16,17]. For this reason, the glycosylation cannot be done
in the presence of citrylidene phloroglucinol and another strat-
egy had to be followed for the synthesis of compound 6
(Scheme 2). The glycosylation was therefore carried out start-
ing from a-L-rhamnose and triethyleneglycol under acidic
conditions [14] to give compound 4 in 52% yield. In the sec-
ond step, selective bromination of the primary alcohol with
triphenylphosphine and carbon tetrabromide [18] gave the
expected compound 5 in 55% yield. This compound was then
condensed with citrylidene phloroglucinol [19] under basic
conditions [20] to afford surfactant 6 in 65% yield (a/b ano-
mer 90/10) Scheme 2.
The different compounds were fully characterized using
1H, 13C NMR and mass spectroscopy.
3. Results and discussion
The capacity of each surfactant to form aggregates was
studied by CMC determinations. The surface tension method
2. Chemistry
The synthesis of rhamnosurfactant 3 was easily accom-
plished as outlined in Scheme 1. Starting from the choles-
teryl toluenesulfonate 1 [13], compound 2 was prepared by
Scheme 2. Synthesis of rhamnosurfactant 6. Reagents and conditions: (i)
triethyleneglycol, p-toluenesulfonic acid D; (ii) PPh3, CBr4; (iii) Cs2CO3,
DMF, D.
Scheme 1
.
Synthesis of rhamnosurfactant 3. Reagents and conditions: (i)
triethyleneglycol, dioxane, D; (ii) p-toluenesulfonic acid, THF, D.

1024
V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
Fig. 2. Plot of surface tension vs. the log of the concentration of 3 and 6.
[21] was used to measure the CMC from the plots of the sur-
face tension (r) against the log of the surfactant concentra-
tion, log(C). Clear break points were observed at 0.1 mM for
rhamnosurfactant 3 and 0.025 mM for rhamnosurfactant 6.
Fig. 2 gives the data graphically.
intensively washed and then incubated to the cells. The tracer,
namely IP, was observed after 10 min of incubation in all the
conditions tested. However, liposomes loaded with rhamno-
surfactants 3 and 6 showed a strong nuclear staining to 10%
or 27% of the cells, respectively, when liposomes without
The structures of the two lipids are similar in many ways.
Although they both have a terpene-derived cyclic structure,
the hydrophobic moieties differ appreciably. Micellization is
driven, among others, by hydrophobic interactions between
the surfactants. Changing the hydrophobic chain from cho-
lesterol to citrylidene phloroglucinol decreases the CMC
about fourfold, which can reflect a more hydrophobic behav-
ior of citrylidene phloroglucinol with respect to the choles-
terol moiety. This observation is reasonable as the tremen-
dous affinity of cannabinoids for membranes was the main
reason for the difficulty to detect their receptors [22].
The use of liposomes depends on their chemical compo-
sition [23]. The nature of lipids has considerable influence on
the properties of the liposomes. For this reason, it is neces-
sary to characterize the thermal behavior of liposomes pre-
pared with our synthetic lipids. The influence of rhamnosur-
factants 3 and 6 on the thermotropic behavior of a membrane
was assayed by DSC with DPPC as the model membrane.
The calorimetric scans were made on DPPC multilamellar
vesicles in the presence of several concentrations of our syn-
thetic lipids. The hydrated DPPC lipids undergo a low
enthalpy pretransition prior to a chain-melting transition [24]
which were detected at 33 and 41 °C, respectively. The adsorp-
tion of surfactants 3 and 6 within the DPPC bilayer had a
pronounced effect upon the thermotropic properties of the
DPPC membrane (Fig. 3a,b).
Fig. 3a. DSC thermogram for aqueous dispersions of 3 with DPPC.
In both cases, the endothermic adsorption shifts regularly
to lower temperature as the amount of synthetic lipids was
increased. However, the cholesteryl surfactant 3 broadens and
shifts more significantly the phase transition of DPPC in com-
parison to the citrylidene phloroglucinol surfactant. The
decrease in phase transition temperature indicates that the gly-
colipids we have synthesized can indeed bind to the phospho-
lipids bilayer.
3.1. Intraliposomal delivery to cultured human
keratinocytes (HK)
To follow the intraliposomal delivery to keratinocytes, lipo-
somes were formed in presence of propidium iodide (IP),
Fig. 3b. DSC thermogram for aqueous dispersions of 6 with DPPC.

V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
1025
Fig. 4. Liposome delivery to cultured HK. Liposomes (cholesterylsuccinate
(A, B), compound 3 (C, D), compound 6 (E, F)) containing propidium iodide
(IP) were loaded onto cultured keratinocytes. After 40 min of incubation,
cultures were observed by Hoffman contrast (A, C, E) and by epifluores-
cence (B, D, F). Note that IP staining was weak and mainly nucleolar when
cells are incubated with cholesterylsuccinate-liposome whereas a strong
nuclear staining was observed fort both rhamnosurfactant 3- and rhamnosur-
factant 6-liposomes. Arrows point to negative cells, arrowheads to positive
cells. Bar represents 50 µm.
Fig. 5
.
Liposome fixation to skin sections. Liposomes were tagged with a
fluorescent lipid (green) and then applied to the skin section counterstained
with Hoescht 33258 (blue). The nuclei staining (blue) allowed us to visua-
lize the different compartment of the skin, namely the epidermis (e) and the
dermis (d). No staining was observed in the mock control (A) as in the skin
section incubated with chol-liposome (B). Rhamnosurfactant 3-liposome
strongly stained all the layers of the epidermis (C) whereas rhamnosurfac-
tant 6-liposome staining was restricted to the anuclei layers of the epidermis
i.e. the cornified layers (sc, D). This staining was inhibited when the section
was incubated with L-rhamnose prior rhamnosurfactant 6-liposome incuba-
tion (E), but not when both incubations were preformed simultaneously (F).
Bar represents 50 µm.
ligand showed only a weak nucleolar staining (Fig. 4). After
40 min of incubation the same profile was observed except
for liposomes loaded with rhamnosurfactant 3 which stained
cells up to 23%.
These results showed that both ligands increased the intrali-
posomal delivery to the cells suggesting that a physical inter-
action could occur between the L-rhamnose and the cells. Two
kinetics were observed according the compound tested, that
also suggested either that the lipophilic part influenced the
transfer or that different receptors existed at the surface of
the cells.
was different, through out the epidermis for rhamnosurfac-
tant 3 and the uppermost layers of the epidermis for rhamno-
surfactant 6. No clear staining has been noticed in the der-
mis.
To further test this interaction, free L-rhamnose was used
in competition with the rhamnosurfactant 6-liposome. When
free L-rhamnose was incubated prior to the liposome, a com-
plete inhibition of the liposome–skin cell interaction was
observed. However, no difference was shown when
L-rhamnose and rhamnosurfactant 6-liposome were incu-
bated simultaneously. The same result was obtained with the
rhamnosurfactant 3-liposome (data not shown).
These observations could fit with what was observed pre-
viously with cultured keratinocytes. Indeed, two different
kinetics according to the compound used were obtained for
the intraliposomal delivery and two different staining were
detected on human skin sections. However, rhamno-
3.2. Liposomal adhesion to human skin cells
In order to characterize the interactions between the lipo-
somes and human skin cells, liposomal solutions were formed
in presence of Oregon Green® 488 1,2-dihexadecanoyl-sn-
glycero-3-phosphoethanolamine (DHPE-fluoresceine, Inter-
chim, France). This tag allowed us to trace the liposomes fixed
on the skin sections.
Cholesterylsuccinate-liposomes did not fix skin cells as
shown in Fig. 5 whereas rhamnosurfactant 3- and rhamnosur-
factant 6-liposomes stained the skin section. Their staining

1026
V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
surfactant-6 liposome stained the uppermost layer of the epi-
dermis that is known to produce high background interac-
tions and also that is composed of terminally differentiated
cells that were not present in cultures. Hence, even if results
concordate between the two set of experiments, one might be
cautious with their interpretations. Nonetheless, in the open
literature two rhamnose-receptors have been characterized in
steelhead trout. They are also characterized by different affin-
ity to L-rhamnose and different activity in hemagglutination
assay. The search for L-rhamnose receptors in human deserves
full attention.
din (12.5 ml) was stirred for 12 h at room temperature. Fol-
lowing addition of water (25 ml), the mixture was extracted
with diethyl ether (3 × 25 ml). The organic phases were dried
over anhydrous sodium sulfate, and concentrated in vacuo to
give the expected compound as a white powder in 90% yield.
The compound was recrystallized in petroleum ether. Rf
= 0.82 (silica, methylene chloride); m.p. 130–132 °C (Litt.
132 °C); 1H NMR (CDCl3, 200 MHz) d 7.82 (d, 2H,
J = 8.3 Hz); 7.36 (d, 2H, J = 8.6 Hz); 5.32 (d, 1H, J = 5.3 Hz);
4.29–4.40 (m, 1H); 1.62 (s, 3H); 0.99 (s, 3H); 0.92 (d, 3H,
J = 7.6 Hz); 0.88 (d, 6H, J = 7 Hz); 0.68 (s, 3H).
5.1.3. 8-(Cholest-5-en-3b-yloxy)-3,6-dioxaoctan-1-ol (2)
4. Conclusion
To a solution of cholesteryl tosylate (1 g, 1.85 mmol) in
dry dioxane (40 ml) was added tri(ethyleneglycol) (10 ml,
74 mmol). The mixture was refluxed for 5 h under a nitrogen
atmosphere. Then the reaction mixture was poured into water
and extracted with ethyl acetate. The organic layer was washed
with 5% aqueous solution of NaHCO3, water, dried over anhy-
drous Na2SO4 and concentrated under reduced pressure to
give an yellow oil which was chromatographed on silica gel
(petroleum ether/ethyl acetate 8:2). The expected compound
was obtained as a white oily solid in 78% yield. Rf = 0.17
(silica, petroleum ether/ethyl acetate 6:4); 1H NMR (CDCl3,
200 MHz) d 5.34 (d, 1H, J = 5.16 Hz); 3.59–3.73 (m, 12H);
3.17–3.21 (m, 1H); 2.79 (m, 1H); 1.02 (s, 3H); 0.91 (d, 3H,
J = 6.45 Hz); 0.87 (d, 6H, J = 6.55 Hz); 0.68 (s, 3H). 13C
NMR (100 MHz, CDCl3) d 141.3; 122.0; 80.0; 73–67.6; 62.2;
57.2; 56.6; 50.0; 42.7; 40.2; 39.9; 39.4; 37.6; 37.3; 36.6; 36.2;
32.4; 32.3; 28.7; 28.6; 28.4; 24.7; 24.2; 23.2; 23.0; 21.5; 19.8;
19.1; 12.3 . FAB-MS: m/z: 541 [M + Na+].
The synthesis of two rhamnose-bearing surfactants has
been described. The ease of their synthesis, their ability to
incorporate liposome bilayers, and the lack of toxicity [25]
of their natural constituents make these sugar surfactants
appropriate tools for the studies of liposomes–cell interac-
tions and further development to human. Thus, liposome bear-
ing ligands reported there-in have been successfully used to
deliver material to HK. Moreover, experiments on skin sec-
tions have shown specific interaction with epidermis. Con-
sidering these results, such ligands are interesting candidates
for the development of new tagged liposomes which can find
applications in medicine as well as in cosmetic fields.
5. Experimental section
5.1. Materials and methods
5.1.1. Chemistry
5.1.4. 8-(Cholest-5-en-3b-yloxy)-3,6-dioxaoctanyl ␣-L-
rhamnopyranoside (3)
Melting points were determined on a Büchi apparatus and
are uncorrected. TLC analysis were performed on aluminum
sheets coated with Merck silica gel 60 F254. Compounds were
visualized by spraying the TLC plates with dilute 10% aque-
ous sulfuric acid solution or anisaldehyde solution, followed
by charring at 150 °C. 1H and 13C NMR spectra were recorded
on Brüker AC-200 and AC-400 spectrometers, respectively.
Chemical shifts (d) are reported in parts per million from inter-
nal standard tetramethylsilane. Coupling constants (J) are
measured in Hertz. Multiplicity is reported as follows: s (sin-
glet), d (doublet), t (triplet), m (multiplet) and combination
of these signals. Optical rotations were recorded on a Perkin
Elmer 241 polarimeter. Fast atom bombardment mass spec-
tra (FAB-MS) were recorded in positive mode on a JEOL
DX 300 spectrometer using NOBA (nitro-benzylic alcohol)
as matrix. Column chromatographies were performed on silica
gel 60.All solvents used for reactions were anhydrous and all
reactions were carried out under inert atmosphere. Com-
pounds are named according to IUPAC rules.
To a solution of L-(+)-rhamnose (60.6 mg, 0.33 mmol) and
8-(cholest-5-en-3b-yloxy)-3,6-dioxaoctan-1-ol (85 mg,
0.166 mmol) in dry tetrahydrofuran (4 ml) was added
p-toluenesulfonic acid (72.4 mg, 0.38 mmol). The reaction
mixture was stirred at 60 °C for 2 days. After addition of
anhydrous sodium carbonate, the mixture was filtered, and
the filtrate concentrated in vacuo. The residue was chromato-
graphed on silica gel (methylene chloride with gradient of
methanol 1–5%) to give in 30% yield the expected com-
pound as a white oily solid. Rf = 0.47 (silica, methylene
chloride/methanol 95:5); [␣]25 = –33 (c = 1; CHCl3); H
1
D
NMR (CDCl3, 200 MHz) d 5.35 (d, 1H, J = 5.13 Hz); 4.84 (d,
1H, J = 1.32 Hz); 3.97 (br, 2H); 3.73–3.84 (m, 2H); 3.65–
3.67 (m, 12H); 3.46–3.52 (m, 2H); 3.15–3.23 (m, 1H); 2.66
(br.s, 1H); 1.32 (d, 3H, J = 6.17 Hz); 1 (s, 3H); 0.93 (d, 3H,
J = 6.44 Hz); 0.88 (d, 6H, J = 6.54 Hz), 0.69 (s, 3H). 13C
NMR (CDCl3, 100 MHz) d 141.2; 122.1; 100.3; 80.0; 73.5;
72.1; 71.2; 71.2; 71.0; 70.7; 68.7; 67.6; 66.9; 57.2; 56.6; 50.6;
42.7; 40.2; 39.9; 39.4; 37.6; 37.3; 36.6; 36.2; 32.4; 32.3; 28.7;
28.7; 28.4; 24.7; 24.3; 23.2; 23.0; 21.5; 19.8; 19.1; 18.0; 12.3.
FAB-MS: m/z: 665 [M + H+], 687 [M + Na+], 703 [M + K+].
5.1.2. Cholest-5-en-3b-ol-p-toluenesulfonate (1)
A solution of cholest-5-en-3b-ol (5 g, 12.93 mmol) and
p-toluenesulfonylchloride (2.71 g, 14.22 mmol) in dry pyri-

V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
1027
5.1.5. 1,3,6-Trioxanonanyl L-rhamnopyranoside (4)
2.69 Hz); 1.83 (dd, 1H, J = 13.22, 1.70 Hz); 1.74 (ddd, 1H,
J = 13.54, 5.69, 5.05 Hz); 1.54 (s, 3H); 1.45 (dt, 1H, J = 13.54,
13.28, 5.25 Hz); 1.39 (s, 3H); 1.24 (dt, 1H, J = 13.28, 5.25,
5.05 Hz); 1.05 (s, 3H); 0.69 (dddd, 1H, J = 13.54, 13.28,
11.38, 5.69 Hz). 13C NMR (CDCl3, 100 MHz) d 157.8; 157.5;
156.0; 109.8; 98.9; 97.3; 85.0; 75.5; 47.1; 37.7; 35.7; 30.0;
29.4; 28.1; 24.1; 22.4. FAB-MS: m/z: 261 [M + H+].
To a solution of L-(+)-rhamnose (5 g, 27.4 mmol) and tri-
(ethyleneglycol) (11 ml, 82.34 mmol) in dry tetrahydro-
furane (100 ml) was added p-toluenesulfonic acid (6.78 g,
35.68 mmol). The reaction mixture was stirred for 2 days at
60 °C.After addition of anhydrous sodium carbonate, the solu-
tion was filtered and concentrated in vacuo to give a yellow
oil which was chromatographed on silica gel (methylene chlo-
ride with methanol gradient 1–10%). The expected com-
pound was obtained as a colorless oil in 51% yield (a/b ano-
mer 95/5). Rf = 0.55 (silica, methylene chloride/methanol 8:2);
[␣]25D = –38 (c = 1; CHCl3); 1H NMR (CD3OD, 200 MHz) a
anomer: d 4.74 (d, 1H, J = 1.65 Hz); 3.77–3.85 (m, 2H); 3.58–
3.71 (m, 12H); 3.32–3.43 (m, 2H); 1.28 (d, 3H, J = 6.2 Hz). b
Anomer: d 4.62 (s, 1H); 3.77–3.85 (m, 2H); 3.58–3.71 (m,
12H); 3.32–3.43 (m, 2H); 1.28 (d, 3H, J = 6.2 Hz). 13C NMR
(100 MHz, CD3OD) d 100.8; 73.0; 72.7; 72.7; 71.2; 70.7;
70.4; 70.4; 68.8; 66.7; 61.2; 17.0. FAB-MS: m/z: 297 [M +
H+], 319 [M + Na+], 335 [M + K+].
5.1.8. 1-Citrylidenephloroglucinyl-3,6-dioxaoctanyl
L-rhamnopyranoside (6)
A solution of citrylidene phloroglucinol (75 mg,
0.289 mmol) and cesium carbonate (282 mg, 0.867 mmol) in
dry DMF (3 ml) was stirred at 70 °C for 1 h. 1-Bromo-3,6-
dioxaoctanyl-L-rhamnopyranoside (217.4 mg, 0.577 mmol)
was then added and the mixture was stirred at 50 °C for 2 h.
The mixture was diluted with methylene chloride and washed
three times with brine. The organic phase was dried over anhy-
drous sodium sulfate, filtered and concentrated in vacuo. The
residue was chromatographed on silica gel (methylene chlo-
ride with methanol gradient 1–5%) to give the expected com-
pound in 65% yield (a/b anomer 90/10) as a white oily solid.
5.1.6. 1-Bromo-3,6-dioxaoctanyl L-rhamnopyranoside (5)
To a solution of 1,3,6-trioxanonanyl-L-rhamnopyranoside
(2.4 g, 7.66 mmol) and tetrabromomethane (5.1 g,
15.32 mmol) in dry tetrahydrofurane (100 ml) was added
triphenylphosphine (4.03 g, 15.32 mmol). The mixture was
stirred at room temperature for 12 h. Then the mixture was
filtered and the filtrate was concentrated in vacuo. The resi-
due was chromatographed on silica gel (methylene chloride
with gradient methanol 1–5%) to give in 55% yield the
expected compound as a colorless oil (a/b anomer 90/10). Rf
= 0.42 (silica, methylene chloride/methanol 9:1); [␣]25D = –42
Rf = 0.39 (silica, methylene chloride/methanol 95:5); [␣]25
D
= –31 (c = 1, CHCl3); 1H NMR (CDCl3, 200 MHz): d 6.04 (s,
2H); 4.74 (d, 1H, J = 1.7 Hz); 4.62 (s, 1H); 3.85–3.77 (m,
2H); 3.71–3.58 (m, 12H); 3.43–3.32 (m, 2H); 2.84 (ddd, 1H,
J = 3.2, 2.7, 1.70 Hz); 2.23 (ddd, 1H, J = 13.3, 3.2, 1.6 Hz);
2.03 (ddd, 1H, J = 11.4, 2.7, 5.1 Hz); 1.83 (dd, 1H, J = 13.2,
1.70 Hz); 1.74 (ddd, 1H, J = 13.5, 5.7, 3.1 Hz); 1.54 (s, 3H);
1.45 (dt, 1H, J = 13.5, 13.3, 5.3 Hz); 1.39 (s, 3H); 1.28 (d,
3H, J = 6.2 Hz); 1.24 (dt, 1H, J = 13.3, 5.3, 5.1 Hz); 1.05 (s,
3H); 0.639 (dddd, 1H, J = 13.5, 13.3, 11.4, 5.7 Hz). 13C NMR
(CDCl3, 100 MHz) d 159.2; 157.9; 157.5; 110.1; 100.3; 98.2;
98.2; 96.1; 84.6; 75.3; 73.4; 72.1; 71.3; 71.2; 71; 70.6; 70.1;
68.5; 68.0; 66.9; 47.1; 37.7; 35.7; 30.1; 29.4; 28.2; 24.2; 22.4;
18.0. FAB-MS: m/z: 539 [M + H+], 561 [M + Na+].
1
(c = 1; CHCl3); H NMR (CDCl3, 200 MHz) a anomer: d
4.83 (s, 1H); 3.99–4.13 (m, 2H); 3.68–3.87 (m, 13H); 3.51
(m, 4H); 1.33 (d, 3H, J = 6.13 Hz). b Anomer: d 4.55 (s, 1H);
3.99–4.13 (m, 2H); 3.68–3.87 (m, 13H); 3.51 (m, 4H); 1.33
(d, 3H, J = 6.13 Hz). 13C NMR (CDCl3, 100 MHz) d 100.4;
73.1; 72.9; 72.0; 71.6; 71.2; 70.9; 70.6; 68.6; 66.9; 30.8; 18.0.
FAB-MS: m/z: 359 [M79Br + H+], 361 [M81Br + H+], 381
[M79Br + Na+], 383 [M81Br + Na+], 739 [2M79Br + Na+],
743 [2M81Br + Na+].
5.2. Critical micelle concentration (CMC)
5.2.1. Materials
CMCs were determined at 24 °C with a Du Noüy tensi-
ometer. The apparatus was calibrated before measurements.
5.2.2. Method
5.1.7. Citrylidene phloroglucinol
Deionized water and dioxane (1%) were added to the
appropriate amount of surfactant to a final volume of 20 ml
and a concentration of 10–3 M. Several solutions of decreas-
ing concentrations were prepared by diluting the primary one
until a 10–6 M solution was obtained. Surface tensions (mN/m)
measurements were determined for each solution. The CMC
values were given by the intersection of the curve slopes. The
reproducibility of the experiments was checked by, at least,
three consecutive measurements for each concentration.
A solution of phloroglucin (6 g, 37 mmol) and freshly dis-
tilled citral (7 ml, 40.7 mmol) in dry pyridin (3.6 ml,
44.4 mmol) was refluxed for 6 h. After cooling, chloroform
was added to the mixture and the solution was washed with
10% aqueous solution of HCl and with water. The organic
layers were dried over anhydrous sodium sulfate and concen-
trated in vacuo. The residue was chromatographed on silica
gel (petroleum ether/diethyl ether 9:1) to give a white solid
(40% yield) which was crystallized from a mixture petro-
leum ether/methylene chloride. m.p = 151–152 °C. (Ref. [19],
150–152 °C); Rf = 0.44 (silica, petroleum ether/diethyl ether
5.3. Differential scanning calorimetry (DSC)
1
6:4); H NMR (CDCl3, 200 MHz) d 6.04 (s, 2H); 4.77 (s,
5.3.1. Materials
1H); 2.84 (ddd, 1H, J = 3.16, 2.69, 1.70 Hz); 2.23 (ddd, 1H,
J = 13.25, 3.16, 1.55 Hz); 2.03 (ddd, 1H, J = 11.38, 5.05,
DPPC was purchased from Sigma-Aldrich. The DSC
experiments were performed on a T.A. Instrument Q 100.

1028
V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
5.3.2. Method
derivative of cholesterol or compound 3 or compound 6
(15%), 1-palmitoyl-2-oleoyl-3-phosphoglycerol (POPG, 5%,
Sigma) and Oregon Green® 488 1,2-dihexadecanoyl-sn-
glycero-3-phosphoethanolamine (Oregon Green® 488 DHPE,
5%, Molecular Probes), were prepared by the hydration
method with Dulbecco’s phosphate buffer saline (DPBS,
Eurobio) as described above.
Appropriate amounts of DPPC and surfactant were dis-
solved in a small volume of chloroform. The solution was
then evaporated under vacuum with a bath temperature around
50 °C (temperature above the phase transition temperature of
DPPC) and the last traces of solvent were removed with high
vacuum for 2 h. Deionized water was added to the dried
samples to a final concentration of lipids around 0.5 mg/ml.
The mixtures were heated with a water bath to 50 °C for
30 min then vortexed for 1 min. The operation was repeated
three times.An equivalent volume of each sample was placed
into small aluminium pans which were then sealed. Prior to
scanning, samples were held at 50 °C for 5 min to ensure
complete equilibrium. During scanning, a reference pan con-
taining an identical volume of water was used. The heating
and cooling rates were 10 °C/min in all the experiments. The
measurements were performed in the temperature interval
from 10 to 60 °C. The reproducibility of the DSC experi-
ments was checked by three consecutive scans for each
sample.
Skin biopsies obtained from plastic surgery after collect-
ing informed patient consent, were embedded into OCT-
compound (Labonord, Templemars, France), snap-freezed in
liquid nitrogen and store at –80 °C until used. Six µm section
were obtained using the microcryotome (Slee, Mainz, Ger-
many) and incubated with a saturating solution (0.1% gelatin
in PBS (Sigma) for 2 hr in a humidified chamber. Ten micro-
liters of a liposomal solution (average density 50,000 per ml)
were then loaded onto the section and incubated for 2 hr in a
dark humidified chamber, washed twice in 0.1% gelatin in
PBS for 15 min and finally washed once in 0.1% gelatin in
PBS containing Hoescht 33258 to stain cell nuclei and once
with PBS. Observations were performed using the micro-
scope described above. For the L-rhamnose competition, sec-
tions were either preincubated in L-rhamnose (5 × 10–5 mol/l
final concentration) and then with the liposomal solution or
incubated simultaneously with both solutions. Sections were
then treated as described above.
5.4. Liposome preparation, cell culture and liposome
delivery
Large vesicles were prepared by hydration of a film com-
prised of 80% of 1-palmitoyl-2-oleoyl-sn-glycero-3-phos-
phocholine (POPC, Sigma-Aldrich), succinate derivative of
cholesterol or compound 3 or compound 6 (15%), and 5% of
1-palmitoyl-2-oleoyl-3-phosphoglycerol (POPG, Sigma-
Aldrich)) (5 × 10–7 mol total lipid) with a 0.5 ml phosphate
buffer saline (PBS) solution (1 µM) of propidium iodide, fol-
lowed by 19 extrusions through a 100 nm polycarbonate fil-
ter (LiposoFast, Avestin). The lipid mixture allowed, at room
temperature, to prepare the liposomes above the phase tran-
sition temperature.
After centrifugation the supernatant was discarded and the
pellet resuspended in PBS. This operation was repeated three
times, the final volume of the liposome suspension being 30 µl.
HK were isolated from foreskin following the Rheinwald
and Green [26] method. They were cultured in serum free
medium (SFM) supplemented with epidermal growth factor
(EGF, 5 ng/ml) and bovine pituitary extract (BPE, 50 µg/ml,
Invitrogen Europe) at 37 °C, 5% CO2 in a humidified atmo-
sphere.
Two to three days before the liposome delivery assay, cells
were split on a coverslip at 1,000 cells per cm2. Then, 20 µl
of liposome solutions (average density 50,000 per ml) were
loaded onto the cells and incubated for 10 and 40 min. At
each time point, cells were observed using an inverted micro-
scope (Nikon Eclipse TE300) equipped with Hoffmann con-
trast, epifluorescence and a digital camera (Nikon DXM1200)
for picture recording.
6. Acknowledgments
We thank the Ministère de la Recherche et de la Technolo-
gie for a Doctoral fellowship (E.J.), the Université de Mont-
pellier II for a faculty research grant (J.-Y.W.), the Ligue Con-
tre le Cancer (Comité Départemental des Pyrénées-Orientales)
for research grant (V.B.) and the University of Montpellier I
for financial support (UPRES EA no. 3754).
References
[1] D.D. Lasic, F. Martin, in: Eds (Ed.), Stealth Liposomes, CRC Press,
Boca Raton, 1995.
[2] R. Langer, Nature 392 (1998) 5–10.
[3] J.D. Baldeschwieler, P.G. Schmidt, Chemtech. 27 (1997) 34–42.
[4] C.M. Lehr, J. Control. Release 65 (2000) 19–29.
[5] P. Sears, C.H. Wong, Chem. Commun. 11 (1998) 1161–1170.
[6] V. Barragan, F.M. Menger, K.L. Caran, C. Vidil, A. Morere,
J.L. Montero, Chem. Commun. 1 (2001) 85–86.
[7] D. Cerdan, C. Grillon, M. Monsigny, G. Redziniak, C. Kieda, Biol.
Cell. 73 (1991) 35–42.
[8] V. Barragan-Montero, J.L. Montero, J.Y. Winum, E. Juan, J.P. Moles,
FR 2854896.
[9] (a) J.P. Houlmont, K. Vercruysse, E. Perez, I. Rico-Lattes, P. Bordat,
Comunicaciones presentadas a la Jornadas del Comite Espanol de la
Detergencia 32, 2002, pp. 131–142. (b) J.P. Houlmont, K. Vercruysse,
E. Perez, I. Rico-Lattes, P. Bordat, M. Treilhou, Int. J. Cosmet. Sci. 23
(2001) 363–368.
5.5. Liposome preparation, skin section and liposome
adhesion
[10] S. Lang, D. Wullbrandt, Appl. Microbiol. Biotechnol. 51 (1999)
22–32.
[11] D. Lafont, P. Boullanger, S. Chierici, M. Gelhausen, B. Roux, J. New
Chem. 20 (1996) 1093–1101.
Large vesicles comprised of 1-palmitoyl-2-oleoyl-sn-
glycero-3-phosphocholine (POPC, 75%, Sigma), succinate

V. Barragan-Montero et al. / European Journal of Medicinal Chemistry 40 (2005) 1022–1029
1029
[12] D.P.Yang, T. Mavromoustakos, K. Beshah, A. Makriyannis, Biochim.
Biophys. Acta 1103 (1992) 25–36.
[20] E.E. Dueno, F. Chu, S.I. Kim, K.W. Jung, Tetrahedron Lett. 40 (1999)
1843–1846.
[13] E.S. Wallis, E. Fernholz, F.T. Gephart, J. Am. Chem. Soc. 59 (1937)
137–140.
[14] M.J. Davis, Chem. Soc. (1962) 178–191.
[15] H.W.C. Raaijmahers, B. Zwanenburg, J.F. Chittenden, Recl. Trav.
Chim. Pays Bas 113 (1994) 79–84.
[16] L. Crombie, R. Ponsdorf, Tetrahedron Lett. 43 (1968) 4557–4560.
[17] J.L. Montero, F. Winternitz, Tetrahedron Lett. 29 (1973) 1243–1252.
[18] J. Hooz, S.S. Gilani, Can. J. Chem. 46 (1968) 86–87.
[19] L. Crombie, R.J. Ponsdorf, Chem. Soc. Chem. Commun. 7 (1968)
368.
[21] P.J. Lecomte du Nouy, Gen. Physiol. 1 (1918) 521–524.
[22] W.A. Devane, F.A. Dysarz III, M.R. Johnson, L.S. Melvin,
A.C. Howlett, Mol. Pharmacol. 34 (1988) 605–661.
[23] D.D. Lasic, in: Liposomes: From Physics to Applications, Elsevier,
Amsterdam, 1993.
[24] M.J. Janiak, D.M. Small, G.G. Shipley, Biochemistry 15 (1976)
4575–4580.
[25] W.L. Dewey, Pharmacol. Rev. 38 (2003) 151–178.
[26] J.G. Rheinwald, H. Green Cell 6 (1975) 331–343.