Soft Matter c4sm00981a
PAPER
detail after you have returned it.
systems can occur so the whole proof needs to be read. Please
rical data; figures and graphics; and references. If you have not
name(s) with an asterisk. Please e-mail a list of corrections or the
the PDF file or send a revised manuscript. Corrections at this stage
must be sent at the same time.
ne breaking, table widths and graphic placement, are routinely
ng your corrections; no late corrections will be made.
rs of receipt by e-mail to: softmatter@rsc.org
1
Characteristics and behaviour of liposomes when
incubated with natural bile salt extract: implications
for their use as oral drug delivery systems
Laura G. Hermida, Manuel Sabe´s-Xaman´ı
and Ramon Barnadas-Rodr´ıguez*
The use of liposomes3 for oral administration of drugs and for
food applications is based on their ability to preserve
entrapped substances and to increase their bioavailability.
1
5
10
15
20
25
30
35
40
45
50
1
5
10
15
20
25
30
35
40
45
50

Please check this proof carefully. Our staff will not read it in
Translation errors between word-processor files and typesetting
pay particular attention to: tabulated material; equations; nume
already indicated the corresponding author(s) please mark their
PDF with electronic notes attached - do not change the text within
should be minor and not involve extensive changes. All corrections
Please bear in mind that minor layout improvements, e.g. in li
applied to the final version.
We will publish articles on the web as soon as possible after receivi
Please return your final corrections, where possible within 48 hou

ART C4SM00981A_GRABS

Queries for the attention of the authors
Journal: Soft Matter
Paper: c4sm00981a
Title: Characteristics and behaviour of liposomes when incubated with natural bile salt extract: implications for
their use as oral drug delivery systems
Editor's queries are marked like this... 1 , and for your convenience line numbers are inserted like this... 5
Please ensure that all queries are answered when returning your proof corrections so that publication of your
article is not delayed.
Query
Reference Query Remarks
1
For your information: You can cite this article before you receive
notication of the page numbers by using the following format:
(authors), Soft Matter, (year), DOI: 10.1039/c4sm00981a.
2
Please carefully check the spelling of all author names. This is
important for the correct indexing and future citation of your
article. No late corrections can be made.
3 Please check that the inserted GA text is suitable.
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

ART C4SM00981A_GRABS

1
t
Sa2
d
)
na
structure
1 Introduction
Liposomes as (so) drug delivery
parenteral and topical administrations
uncertainty the oral route potentia
their structure, liposomes can protec
from the environment of the gastrointestina
vivo eectiveness depends on, amon
that they undergo when they interac
in the intestines. In this way, th
phospholipids can undergo severa
total disruption of the vesicles. Th
liposomes and other colloids as dru
to the digestion process, as it has
of some of them in the particle
macological activity.8,9
At the same time, the evaluatio
predict their bioavailability an
greatly depends on the existence
Caco-2 cultures.10–13 In these cases
1
5
10
15
20
25
30
35
40
45
50
1
5
10
15
20
25
30
35
40
45
Soft Matter
PAPER

aCentre of Research and Development in Chemi
Technology (INTI), Av. Gral. Paz e/ Constituye
Aires, Argentina
bCentre d'Estudis en Biof´ısica, Unitat de Biof´ı
Molecular. Faculty of Medicine, Univer
Cerdanyola, Catalonia, Spain. E-mai
+34 935811907; Tel: +34 935868476
This journal is ? The Royal Society of Che

systems are commonly used in
,1,2 but aer a period of
l has now emerged.3–7 Due to
t the entrapped substances
l tract, and their in
g other factors, the changes
t with the bile salts present
e macromolecular assembly of
l changes that can lead to the
e role of bile salts in the use of
g carriers is not only limited
been shown that the inclusion
structure enhances their phar-
n of oral formulations to
d mucoadhesion properties
of in vitro models, such as the
, and prior to the incubation
with cell cultures, formulations are usually exposed to in vitro
digestion that mimics the gastrointestinal tract. The intestinal
step of the previous process usually requires sample incubation
with a bile salt extract (BSE) from a biological source.14–16
Consequently, a detailed knowledge of the processes
involved in the interactions of liposomes and BSE would
contribute to the improvement of the functionality of the oral
formulations that contain these vesicles. In fact, the eect of
bile salts on phospholipid bilayers and monolayers is a partic-
ular case of the well-known solubilisation process17–19 caused by
surfactants which involves three dierent stages: (a) during the
vesicular stage the liposome bilayer becomes progressively
enriched in the surfactant, (b) liposomes are gradually
destroyed as the surfactant concentration increases and mixed
micelles are formed and, (c) only (mixed) micelles exist in the
bulk. Solubilisation curves can be obtained by monitoring
absorbance/turbidity changes in dierent liposome suspen-
sions upon increasing the surfactant concentration. Some
characteristic points that indicate the phase boundaries
(usually the onset and full solubilisation points) can be
observed from these curves. These points are then used to

Characteristics an
incubated with natura
implications for
systems
Laura G. Hermida,a Manuel
The use of liposomes for oral administra
preserve entrapped substances an
aect the liposome structure durin
salt systems used only one bile salt
with a natural bile salt extract (BSE
gel-state and liquid-ordered bilayers
behaviour was found. Fluid bilayer
complete disruption of vesicles (
to the formation of large mixed
(1000 nm) and, surprisingly, retaine
consequence, each type of liposom
specic interaction with bile salts.
Cite this: DOI: 10.1039/c4sm00981a
Received 6th May 2014
Accepted 30th June 2014
DOI: 10.1039/c4sm00981a
www.rsc.org/softmatter

stry, National Institute of Industrial
ntes y Albarellos San Mart´ın, Buenos
sica, Departament de Bioq´ımica i Biologia
sitat Auto`noma de Barcelona, 08193
l: ramon.barnadas@uab.cat; Fax:
mistry 2014

d behaviour of liposomes when
l bile salt extract:
heir use as oral drug delivery
be´s-Xaman´ıb and Ramon Barnadas-Rodr´ıguez*b
tion of drugs and for food applications is based on their ability to
to increase their bioavailability. Bile salts are one of the agents that
g intestinal digestion and the main reported studies on liposome/bile
. The aim of this work is to characterise the interaction of liposomes
at physiological pH and temperature. Three types of liposomes (uid,
) were studied. Phase diagrams were obtained and a very dierent
s were completely permeable to an entrapped dye with partial or
l size 10 nm). Gel-state bilayers released their content but BSE led
s (2000 nm). Liquid-ordered bilayers formed mixed vesicles
d a high percentage of their aqueous content (about 50%). As a
e oers singular features to be used in oral applications due to their

obtain the phase diagram of the system. It is described in the
literature that, for a given surfactant, liposome solubilisation
depends to a great extent on the size, lamellarity and compo-
sition of the vesicles, as well as temperature and ionic strength
of the medium.20–22 Most of the liposome solubilisation studies
performed with bile salts have been made using pure molecules
Soft Matter, 2014, xx, 1–9 | 1
50

Soft Matter Paper
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

such as sodium cholate, sodium deoxycholate, or some articial
mixtures.23–25 Even though the eect of natural BSE on lipo-
somes implies a closer approximation to the in vivo intestinal
digestion, and BSE is commonly used in in vitro liposome
digestion models, to our knowledge there are no detailed
studies focused on the eects of natural BSE on these vesicles.
Moreover, no detailed studies providing phase diagrams for
mixtures of bile salt extracts with liposomes are available. If
liposomes are currently intended to be used for oral drug
delivery, knowledge of their behaviour in the presence of a
complex natural mixture of bile salts could provide important
information for the optimisation of the system.
Our work endeavours to ll these gaps in the scientic
literature. On the one hand, it analyses and provides a
hypothesis, not previously described, about the formation of
mixed structures with BSE, especially in the cases of saturated
and cholesterol containing liposomes. The phase diagrams of
the studied systems are also obtained.
We report and analyse the interactions of a natural BSE with
three types of liposomes that exhibit very dierent phase
membrane properties at human body temperature (37 C): (a)
soy phosphatidylcholine (SPC), which has a negative transition
temperature and, therefore, forms uid bilayers; (b) gel-state
membranes composed of hydrogenated SPC (HSPC), with a
transition temperature higher than that of the human body, and;
(c) liquid-ordered bilayers of HSPC and cholesterol 3 : 2 mol
mol1 (HSPC/CHOL), which exhibit the characteristics of both
uid and gel phases, and have no transition temperature.26,27
Consequently, a very dierent behaviour is expected when they
interact with a natural BSE. These lipids are puried by large
scale methods, are commercially available, and are usually used
by food and pharmaceutical industries. The BSE commercial
extract used has a bile salt composition that is very similar to
that of human bile as previously determined.28,29 The present
paper specically describes the physical evolution of these
liposomes at 37 C and pH 6.5 (with the absence of intestinal
enzymes) by measuring the turbidity changes of the sample and
by dynamic light scattering. From these measurements, phase
diagrams were obtained and the temperature was eventually
raised to achieve full solubilisation. The capacity of the lipo-
somes to maintain the entrapped aqueous medium was also
evaluated by measuring the leakage of pyranine, a non-bilayer
permeable uorescent probe. We found that each type of
vesicle exhibits a very dierent behaviour not only with regard to
the susceptibility to be fully solubilised by BSE, but also in the
vesicular stage. The results set suggests the formation of three
dierent molecular aggregates which, depending on the initially
entrapped substances, may determine the eectiveness of each
type of liposome in nutritional formulations.
2 Experimental section
2.1 Chemicals and reagents
SPC (minimum 95%) and HSPC (minimum 95%) were obtained
from Degussa (Germany). BSE (hyocholic acid 5.3%, cholic acid
18.5%, deoxycholic acid 2.5%, glyco and taurocholic acid
37.5%, glyco and taurodeoxycholic acid 21.6%) and CHOL

2 | Soft Matter, 2014, xx, 1–9

($99%) were purchased from Sigma-Aldrich (USA). Pyranine
was from Kodak (USA) and DPX (a pyranine quencher) was from
Molecular Probes (The Netherlands). All other reagents were of
analytical grade.
2.2 Liposome preparation
SPC and HSPC were added into 10 mM TRIS, 145 mM sodium
chloride, pH 6.5 buer solutions (since pH changes during
intestinal digestion from, approximately, 5.7 to 7.4, this mean
value was selected and used throughout the work) and stirred 60
minutes at 40 C or 55 C respectively. The resulting multi-
lamellar vesicles (33 mM) were further homogenized30 using a
Microuidizer 110S at the previously indicated temperatures.
HSPC/CHOL liposomes were prepared at a lipid molar ratio of
3 : 2. Lipids were dissolved in chloroform : methanol (2 : 1 v/v)
and solvent was eliminated by rotary-evaporation. The dry
lm obtained was hydrated with the buer solution (nal lipid
concentration 41 mM) and vortexed at 55–60 C before
homogenization at the same temperature. When required,
2 mM pyranine (a water soluble uorescent dye) was added to
the buer.
2.3 Vesicle size determination
The particle size was measured by dynamic light scattering
(Ultrane Particle Analyzer UPA150, USA). Cell temperature was
controlled by an external bath, and the change of water viscosity
with temperature was considered in the soware presets.
Analyses were performed without sample dilution in order not
to alter the phase equilibrium. Results are expressed as the
mean diameter of the volume distribution and SD (n $ 2).
2.4 Solubilisation assays
Solubilisation curves of liposomes at several concentrations
were obtained by monitoring sample absorbance (600 nm) on a
double beam spectrophotometer Varian CARY 3Bio. The wave-
length was chosen in order to minimize the interference of BSE
absorption. Appropriate dilutions of the BSE in TRIS buer (pH
6.5) were used as reference solutions. The required volumes of
concentrated BSE aliquots were added to the continuously
stirred samples. Results are expressed as mean SD (n $ 2).
The phase diagrams where obtained by calculating the charac-
teristic points of the solubilisation curves. At a given concen-
tration of lipid (each one of the curves), the corresponding BSE
concentrations were calculated from the break points of the
curve, from it the rst derivative (0 value) and, in the case of
total liposome solubilisation, the BSE concentration that
caused an absorbance value equal or smaller than 0.03 was also
considered. Results of the phase diagrams are expressed as the
mean SD (n $ 2).
2.5 Fluorescence assays
Fluorescence assays were performed to study the eect of BSE on
the aqueous content of liposomes. Vesicles were obtained in
buer containing pyranine 2 mM and subsequently puried
by size exclusion chromatography (Sephadex G-25) to remove the

This journal is ? The Royal Society of Chemistry 2014

non-entrapped dye. Aer adjusting the lipid concentration,
sample aliquots were incubated during 1 h at 37 C with
increasing concentrations of BSE. The percentage of pyranine
retention was calculated from the ratio of the corrected uo-
rescence measured before and aer the addition of 150 ml of DPX
200 mM to 3 ml of sample. As BSE exhibits intrinsic uores-
cence, curves of BSE uorescence in the absence and presence of
DPX were acquired for data correction. Fluorescence was
measured with a SLM Aminco 8100 Spectrouorometer using
417 nm and 511 nm as excitation and emission wavelengths
respectively. Results are expressed as mean SD (n $ 2).
2.6 Dierential scanning calorimetry (DSC)
The main phase transition temperature of HSPC and HSPC/
CHOL liposomes was determined using a Microcal MC-2 DSC
microcalorimeter (USA). Measurements were performed at a
heating rate of 90 C h1, from 25 to 80 C, and using TRIS
buer as a blank.
3 Results
3.1 SPC liposomes
are formed (2 mM to approximately 6 mM of BSE) and; (III) only
micelles are present in the sample (BSE concentration > 6 mM).
It can be observed that stage I (vesicular domain) exhibited
particular absorbance changes. There was a clear initial
decrease of the absorbance (stage Ia) and a subsequent increase
(stage Ib). The uorescent changes shown in Fig. 1 indicate that,
in one hour, the dye was released at BSE concentrations that did
not cause bilayer disruption, that is, during stage I. This was
especially true during stage Ib, as the retention of the dye
drastically diminished (inset Fig. 1), and pyranine was almost
completely released before the onset of formation of mixed
micelles. Fig. 2 shows the absorbance variations of dierent
concentrations of SPC liposomes aer the addition of
increasing amounts of BSE. From the similarity of the proles,
in all cases, it can be assumed that equivalent morphological
changes and processes took place. Several characteristic points
can be obtained from the curves at dierent SPC concentra-
tions: CIab, the BSE concentration that caused the change from
stage Ia to stage Ib (minimum absorbance value of stage I); Csat,
the BSE concentration that caused the saturation of the bilayer
(limit between stages I and II); and Csol, the BSE concentration
that caused full solubilisation of liposomes (limit between
stages II and III). The inset of Fig. 2 was obtained by plotting the
total BSE concentration vs. the total lipid concentration at the
Paper Soft Matter
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

Fig. 1 Absorbance (circles) and uorescence (squares) changes of
suspensions of SPC liposomes (0.60 mM) in the presence of BSE after
1 h of incubation at 37 C. The inset corresponds to low concentrations
of BSE.

Fig. 1 shows the behaviour of SPC vesicles when incubated with
BSE for 1 h at 37 C (the absorbance remained constant aer
20 minutes of incubation at each BSE concentration, indicating
that all the mixtures reached a steady state). The results of
absorbance reected the morphological changes of liposomes
caused by the interaction with BSE. The absorbance curve
illustrates the well-known three-stage model for the interaction
between liposomes and surfactants:31 (I) bile salt molecules
interact with the membranes without disrupting the vesicles (0–
2 mM of BSE); (II) aer saturation of the bilayers, vesicles are
progressively solubilised and, concomitantly, mixed micelles

This journal is ? The Royal Society of Chemistry 2014

previously obtained characteristic points. The results parallel
the general behaviour of liposome/surfactant systems17–19,23 in
which, from each family of characteristic points, a linear ansatz
can be made using the following general equation:
[detergent]T ¼ [detergent]w + Re[Lip]T (1)
where, when applied to the case of BSE, [BSE]T is the total BSE
concentration, [BSE]w is the BSE concentration in the bulk,
Fig. 2 Solubilisation curves of SPC liposomes (circle: 3.33mM; square:
2.33 mM; up triangle: 1.75 mM; down triangle: 0.88 mM; diamond:
0.60 mM) obtained in the presence of BSE after 20 min at 37 C. Inset:
the phase diagram of SPC/BSE mixtures obtained from the solubili-
sation curves. (*: Point rejected).

Soft Matter, 2014, xx, 1–9 | 3

[Lip]T is the total lipid concentration, and Re is the eective BSE
to the lipid ratio, that is, the ratio of the total detergent
concentration that is bound to the lipids in the dierent types of
mixed aggregates ([BSE]agg/[Lip]agg). Each one of the previous
parameters are used to establish the dierent phase boundaries
of the system. From the previous equation, it is also possible to
calculate the molar fraction of BSE in the mixed aggregates,
xBSEagg , that is:23,32,33
xBSEagg ¼
½BSEagg
½BSE þ ½Lip ¼
Re
Re þ 1
(2)
of the sample had to be raised above the phase transition
temperature of HSPC (52.2 0.1 C, n ¼ 4).
The absorbance results obtained at 55 C (Fig. 6) for the two
higher phospholipid concentrations (3.3 and 2.1 mM) were
quite similar to the curves of SPC liposomes at 37 C. They
showed an initial decrease of the absorbance and a subsequent
increase before the onset of solubilisation. Then, in those cases
and as for SPC, the vesicular domain of the uid bilayers of
HSPC exhibited the stages Ia and Ib in the presence of BSE. But
for lower phospholipid concentrations the shape of the curves
was not totally maintained: The portion of the vesicular domain
corresponding to stage Ia gradually decreased with decreasing
s obtained by incubation with BSE
Csat Csol
[BSE]satw
(mM) Rsate xBSEagg
[BSE]solw
(mM) Rsole xBSE,solagg
1.24 0.14 0.39 0.07 0.28 4.58 0.12 1.60 0.06 0.62
3.55 0.74 1.58 0.35 0.61 7.37 0.43 5.07 0.8 0.84
1.1 2.3 22.9 3.6 0.96 1.24 1.46 43.0 2.6 0.98
Soft Matter Paper
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

agg agg
The values of the parameters obtained for the relationships
are shown in Table 1 and the corresponding curves set the
limits of the four dierent stages mentioned previously (Ia, Ib, II
and III). As can be observed (inset Fig. 2), in the studied range
the CIab points do not give a curve with a signicant slope and,
consequently, a mean value of 0.41 0.23 mM is obtained for
[BSE]Iabw . Due to the deviation from linearity, the absorbance at a
lower lipid concentration was rejected for the calculation of the
saturation phase boundary. Fig. 3 shows the variation of the
mean diameter of SPC vesicles as a function of the total BSE
concentration. The results indicate a gradual decrease of the
vesicle size (initial diameter 539 150 nm) when BSE is added.
By the end of the solubilisation process the diameter was
consistent with the size of mixed micelles (about 10 nm). As the
diameter is expressed in volume percentage, it can be inferred
that all the aggregates detected corresponded to micelles, and
no other kind of vesicles was present.
3.2 HSPC Liposomes
The absorbance variation of HSPC liposomes as a function of
the BSE concentration aer 1 h incubation at 37 C is shown in
Fig. 4. Results show a dierent behaviour than that of SPC
liposomes. As can be observed, no stage Ia was detected and a
high positive slope of the absorbance was obtained at low BSE
concentrations. Subsequently, and in a similar way to SPC
vesicles, a BSE concentration which corresponded to the onset
of solubilisation was achieved (about 3.5 mM). A further
increase in the BSE concentration caused a decrease of the
absorbance but, contrary to SPC liposomes, it did not lead to a
zero value of the absorbance. Instead, the absorbance decreased
to a nal constant value (about 0.5) that was higher than the
initial value (0.2) obtained in the absence of BSE. In the uo-
rescence assay (Fig. 4), the release of the dye took place at very
low concentrations of BSE and there was an almost a complete
Table 1 Solubilisation parameters of SPC, HSPC and HSPC/CHOL liposome
Composition
Temperature
(C)
Inc.
time
CIab
[BSE]Iabw
(mM) RIabe xBSE,Iabagg
SPC 37 20 min 0.41 0.23 0 0
HSPC 55 1 h 3.1 1.3 0 0
HSPC/CHOL 65 1–2 h — — —

4 | Soft Matter, 2014, xx, 1–9

leakage (90–95%) at a BSE concentration of 0.8 mM (although
no vesicle disruption took place). Similar absorbance changes
were observed at dierent phospholipid concentrations (Fig. 5)
and full solubilisation of liposomes was not attained even at the
highest BSE to the phospholipid ratio used. In regard to the size
measurements performed at a phospholipid concentration of
1.2 mM (Fig. 3), it can be observed that their evolution was
concomitant to the absorbance changes at the same concen-
tration and no micelle size was achieved (initial diameter 800
240 nm; nal diameter 1800 260 nm).
In order to achieve full vesicle solubilisation the temperature
Fig. 3 Eect of the BSE concentration on the mean diameter
(%volume distribution) after 1 hour of incubation of SPC (circle:
1.8 mM, 37 C), HSPC (dark square: 1.2 mM, 37 C; white square:
3.3 mM, 55 C), and HSPC/CHOL liposomes (dark up triangle: 1.1 mM,
37 C; white up triangle: 1 mM, 65 C). The mean diameter of HSPC/
CHOL liposomes incubated with 56 mM BSE at 65 C is also expressed
as number distribution (white down triangle).

This journal is ? The Royal Society of Chemistry 2014

Fig. 6 Solubilisation curves of HSPC liposomes (circle: 3.33 mM;
Paper Soft Matter
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15

Fig. 4 Absorbance (circle) and uorescence (square) changes of
suspensions of HSPC liposomes (0.67 mM) in the presence of BSE after
1 h of incubation at 37 C.

HSPC concentration and, consequently, it was not possible to
fully assess the behaviour in the vesicular stage. The boundary
phase parameters were calculated from the characteristic points
(Table 1) and the phase diagram obtained (Fig. 6). Due to its
negative deviation, the absorbance corresponding to a lipid
concentration of 0.4 mM was rejected. This change in the
expected value has been described by other authors34,35 at a very
low lipid concentration using pure bile salts, and taking into
account the nite size of mixed micelles and the end-cap eect
of cylindrical micelles. Size analysis (Fig. 3) also reects the
complete solubilisation of HSPC liposomes at 55 C: A mean
diameter close to 10 nm was observed at the end of the process,
indicating the mere existence of mixed micelles.
3.3 HSPC/CHOL liposomes
DSC thermograms of HSPC/CHOL liposomes showed the
abolition of the phase transition temperature (data not shown)
Fig. 5 Eect of the BSE concentration after 1 hour of incubation at
37 C on the absorbance of HSPC liposomes (circle: 3.33 mM; red
square: 2.13 mM; up triangle: 1.24 mM; down triangle: 0.67 mM;
diamond: 0.40 mM).
square: 2.13 mM; up triangle: 1.24 mM; down triangle: 0.67 mM; dia-

This journal is ? The Royal Society of Chemistry 2014
20
25
30
35

measured in pure HSPC samples. This fact indicates that a
homogeneous liquid-ordered membrane was obtained. Fig. 7
corresponds to the incubation of HSPC/CHOL liposomes with
BSE. It can be observed that there was a sharp initial decrease of
the absorbance at low BSE concentrations, and a subsequent
constant value (0.22) at half the initial absorbance. There was a
concomitant decrease in the uorescence (Fig. 7), and the
minimum value (32% of the initial value) was reached at a BSE
concentration of about 2 mM. Consequently, and similar to SPC
and HSPC, the interaction of bile salts with HSPC/CHOL lipo-
mond: 0.40 mM) obtained in the presence of BSE after 1 h at 55 C.
Inset: the phase diagram of HSPC/BSE mixtures obtained from the
solubilisation curves. (*: Point rejected).

somes at low BSE to lipid ratios increased the bilayer perme-
ability of the vesicles. Both the constant values of the
absorbance and the diameter evolution (Fig. 3) obtained at
higher BSE concentrations reveal that liposomes were not
Fig. 7 Absorbance (circle) and uorescence (square) changes of
suspensions of HSPC/CHOL liposomes (0.5 mM) in the presence of
BSE after 1 h of incubation at 37 C.
Soft Matter, 2014, xx, 1–9 | 5
40
45
50
55

solubilised, as described for HSPC vesicles. Surprisingly, and
contrary to the incubations performed with SPC and HSPC
liposomes, a residual uorescence was maintained once the
absorbance reached a plateau (about 50% of the initial uo-
rescence). This remarkable resistance of HSPC/CHOL lipo-
somes to solubilisation is evident in Fig. 8: Not even diluted
liposomes (0.25 mM total lipid) at the highest BSE concentra-
tion (56 mM) were solubilised during the incubation, only
decreasing the absorbance to 50% of the initial value. The
curves corresponding to phospholipid concentrations of 0.76
and 1 mM (Fig. 8) clearly show the stage at which pyranine was
initially released. In this zone, there was an initial decrease of
the absorbance upon increasing the BSE concentration which
was concomitant with a vesicle size diminution (Fig. 3), and the
subsequent absorbance increase was also associated with a
diameter increase. This behaviour parallels that observed with
SPC liposomes, and could be indicative of similar mechanisms
of interaction of BSE with the two types of bilayers, that is, the
existence of Ia and Ib stages.
In order to achieve full solubilisation of HSPC/CHOL lipo-
somes in 1–2 hours the temperature had to be raised to 65 C.
The results (Fig. 9) show the complex behaviour of the system,
was 20 nm. These results provide evidence of the heterogeneity
of the sample at this stage, although both the absorbance value
and the diameter obtained from the number distribution
indicate that a large quantity of the initial liposomes was
solubilised.
4 Discussion
The experimental design used in the present work combines, on
the one hand, absorbance and size measurements that can
explain the changes that take place during the interaction of
BSE with liposomes. On the other hand, the uorescence
experiments provide information on how the membrane
permeability/integrity is aected during this process. As it is
known, the in vivo eectiveness of the liposomes depends on
the interaction with bile salts, on lipid digestion, and also on
the solubilisation of the incorporated drug, and possible
enhancement of permeability. Consequently, our study
provides specic information which could partially explain the
in vivo eectiveness of liposomes as oral drug delivery systems.
From the results obtained by the incubation of SPC vesicles
with a natural BSE for 1 h at 37 C, three primary conclusions
may be reached: (a) SPC liposomes can be eciently solubilised
under these conditions; (b) their aqueous content can be
Soft Matter Paper
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

with some consecutive absorbance peaks before the onset of
solubilisation. Consequently, unlike that which occurs with SPC
and HSPC, more than two vesicular sub-domains can be
considered for HSPC/CHOL liposomes when they interact with
BSE if all the break points of the absorbance plots are taken into
account. Unfortunately, the proles of the curves were similar
only for the higher lipid concentrations used, thus it was not
possible to obtain the equivalent characteristic points for all the
lipid concentrations.
This fact conditioned the calculation of the boundary phases
and explains the existence of a negative value of [BSE]satw (Table
1) which is, obviously, a consequence of the experimental error
of the extrapolation. For this reason Table 1 shows the data
Fig. 8 Eect of the BSE concentration on the absorbance of HSPC/
CHOL liposomes (black circle: 1.04 mM; square: 0.76 mM; up triangle:
0.5 mM; down triangle: 0.36 mM; diamond: 0.25 mM) after 1 hour of
incubation at 37 C.

6 | Soft Matter, 2014, xx, 1–9

corresponding to the stages II and III which are complemented
in the inset of Fig. 9 with some other points of the vesicular
domain. Note that, as for the cases of SPC and HSPC, some
characteristics points corresponding to vesicular stages at low
phospholipid concentrations showed deviations from linearity
and were excluded in the linear regression t. With regard to the
vesicle size analysis, the mean diameter determined aer full
solubilisation was about 200 nm (% volume average). However,
the measured mean diameter expressed as % number average
Fig. 9 Solubilisation curves of HSPC/CHOL liposomes (circle:
1.04 mM; square: 0.76 mM; up triangle: 0.5 mM; down triangle:
0.36 mM; diamond: 0.25 mM) obtained in the presence of BSE after 1 h
at 65 C. Inset: the phase diagram of HSPC/CHOL/BSE mixtures
obtained from the solubilisation curves. (*: Points rejected).

This journal is ? The Royal Society of Chemistry 2014

Paper Soft Matter
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

completely released to the bulk, even if no destruction of the
molecular structure of the vesicle has taken place and; (c) at very
low BSE to lipid ratios (stage Ia) the aqueous content is main-
tained, although an interaction between BSE and liposomes is
detected.
The observed eect of BSE on phosphatidylcholine lipo-
somes (Fig. 1–3) gave similar absorbance results to that
obtained by Andrieux et al.23 and Elsayed and Cevc21 using pure
bile salts. These authors performed a continuous slow addition
of taurocholate (TC) or worked under equilibrium conditions
using cholate (C), respectively. In our case, before the onset of
solubilisation (that is, in the vesicular domain) BSE rst causes
a strong decrease of the absorbance and then a small increase.
This behaviour parallels the previously mentioned studies and,
therefore, a similar explanation of the vesicular changes could
be proposed. Elsayed and Cevc21 explained these absorbance
results by a vesicle-apparent shrinkage caused by bilayer uc-
tuations induced by the surfactant (which would produce a
decrease in the absorbance) and an expansion of the membrane
and an increase of the vesicle size caused, merely, by its inser-
tion into the membrane (which would cause an increase in the
absorbance). Consequently, as they said, there is a counter-play
between both phenomena, which would explain the negative
and positive slopes of the absorbances observed at stages Ia and
Ib, respectively.
If we now consider our uorescence results, it can be
demonstrated that SPC bilayers remain stable while they
interact with the BSE during stage Ia (more than 80% of the
uorescence is retained). As the molar fraction xBSE,Iabagg is 0 mM,
it can be concluded that in stage Ia there is no insertion of BSE
into the bilayers. This scenario is compatible with the hypoth-
esis of Andrieux et al.23 who proposed that, in the rst range of
the vesicular domain, TC monomers are located at the phos-
pholipid–water interface with no interaction with the hydro-
phobic core of the membrane. The results observed in our work
suggest that this arrangement can be also applied to all the BSE
components in the Ia stage: the non-existence of surfactant
insertion during the incubation ensures the retention of the
probe by preserving the permeability of the liposomal
membrane.
However, it is obvious that the changes in absorbance and
the mean diameter measured in the entire vesicular domain
provide evidence that there are variations of the liposome shape
and/or size. Such variations only aect membrane permeability
during stage Ib (pyranine is released), that is, when the bilayer
expansion caused by the insertion of the bile salts causes the
positive slope of the absorbance. In a recent work, Niu et al.36
evaluated the eectiveness of insulin-loaded liposomes that
contained several bile salts using a phospholipid (SPC) to bile
salt molar ratio of 4 : 1. The increased transport of insulin
observed by the authors in Caco-2 cell cultures is compatible
with the existence of liposomes in the Ib stage, that is, the
vesicular structure is maintained, and at the same time they
gradually release their content into the medium.
As can be seen in Table 1, BSE exhibits an increased eec-
tiveness of solubilisation compared with some pure bile salts.
The [detergent]satw values obtained by other authors using

This journal is ? The Royal Society of Chemistry 2014

phosphatidylcholine (PC) liposomes in saline medium are 1.4
mM for deoxycholate (DC) and 6 mM for C (large unilamellar
liposomes, 30 C),37 and 3 mM for TC (small unilamellar lipo-
somes, 25 C).23 In our experiments, a smaller value was
obtained (1.24 0.14 mM), and as [detergent]satw increases with
temperature,37 it can be shown that at 37 C BSE mixed micelles
are formed at a lower concentration than the mentioned pure
bile salts. The Rsate , value obtained with BSE (0.39 0.07) is
similar to that achieved with DC (0.35), C (0.33) and TC (0.29) by
the same previous authors.23,37 Thus, at a given lipid concen-
tration, the maximum quantity of BSE, DC, C and TC in the
liposome membranes before the formation of mixed micelles is
approximately the same. As expected, and instead of being
equal, [BSE]solw >[BSE]satw as usually occurs with bile salts due to
intermicellar interactions and its value (4.58 0.12 mM) is
located between that of TC (4 mM),23 C (from 5 to 8 mM)21,37 and
DC (2 mM).37 With reference to Rsole , the value obtained for BSE
(1.60 0.06) is higher than that obtained with the previous bile
salts. This fact means that the minimum molar fraction of BSE
necessary to transfer all the PC into mixed micelles (0.62) about
30% higher than that of C (0.47),21 which, in turn, is higher than
that of DC and TC.
The observed eect of BSE on PC liposomes is not conclusive
when compared to in vitro digestions carried out with simulated
complex intestinal uids instead of pure bile salts. Literature
shows contradictory results14,38 which could be caused by the
dierent grade of purity of the PC used, as well as the dierent
lipid digestion models.
Taking into account the previous facts, it can be concluded
that PC liposomes, characterised by their uid state membrane,
are greatly susceptible to release all their entrapped aqueous
content during intestinal digestion due to the eect of bile salts.
The very small range of stage Ia hardly ensures the maintenance
of the water soluble entrapped molecules. By contrast, stage Ib
and (obviously) partial and full liposome solubilisation lead to
the total release of the aqueous core to the medium. This
behaviour can be an advantage if lipophilic substances are
incorporated into the liposome bilayer. In these cases, if the
membrane uidity is not altered, vesicles will be solubilised
into mixed micelles which would increase the intestinal uptake.
The behaviour of HSPC liposomes upon incubation is totally
dierent from that of SPC vesicles. The results shown in Fig. 3–5
indicate that stage Ib is the main one in the vesicular domain
and that full solubilisation of liposomes is not achieved under
the reported conditions. Consequently, large mixed structures
are formed instead of micelles, even at high BSE to HSPC ratios.
The uorescence and absorbance results are compatible with
those observed by Andrieux et al.20 in DSC experiments per-
formed with dipalmitoylphosphatidylcholine (DPPC) liposomes
under the continuous addition of TC. They detected a decrease
in the fusion enthalpy of membranes as a consequence of TC
insertion in the vesicular domain, and concluded that under
those conditions (before the phase transition temperature) the
DPPC gel-phase exhibits a disordered state when TC is present
in the bilayer. In our case, this proposed pre-transition disor-
dered state, caused by a rapid insertion of bile salts into
the membranes, is consistent with both the dramatic

Soft Matter, 2014, xx, 1–9 | 7

Soft Matter Paper
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

morphological changes evidenced by the absorbance changes
and the intense pyranine leakage. In contrast, when the
concentration of TC is increased, Andrieux et al.20 achieve a full
liposome solubilisation at 37 C, that is, below the phase tran-
sition temperature of DPPC (Tm ¼ 41 C), a phenomenon that is
not observed in our HSPC liposomes. This dierent behaviour
could be caused by the transition temperature of HSPC (Tm ¼
51.8 0.2 C, n ¼ 4). This hypothesis is compatible with the
results obtained by Kokkona et al.39 They performed incubation
experiments (1 h, 37 C) with 10 mM C or TC and 2.5 mM DPPC
or distearoylphosphatidylcholine (DSPC, Tm¼ 55 C) liposomes
and found that around 15% and 20% of the entrapped dye
remained in DPPC and DSPC vesicles, respectively. Therefore, C
and TC highly modied bilayer permeability but did not solu-
bilise the liposomes. In our case, the eect of BSE on bilayer
permeability is higher than that caused by TC and C alone, due
to the fact that the dye is completely released to the bulk. But
the high transition temperature of HSPC liposomes, similar to
that of DSPC, could explain their resistance to solubilisation.
Solubilisation of HSPC liposomes by bile salts was found to
be dependent on the temperature. Full vesicle solubilisation
can be achieved at 55 C (Fig. 3 and 6), that is, when
the membrane is in the uid state. Under these conditions
[BSE]satw (3.55 0.74 mM) and [BSE]solw (7.37 0.43 mM) are
similar to that obtained by Hildebrand et al.34 with DPPC lipo-
somes, in saline, with C and at 60 C (5.6 mM and 6.6 mM
respectively). On the other hand, the slopes of BSE (Table 1) are
one order of magnitude higher than that of C. This fact could
be a consequence of the transition temperature of the phos-
pholipid. It is known that solubilisation of dimyr-
istoylphosphatidylcholine (Tm ¼ 24 C) liposomes with C at
30 C shows slopes 10 times smaller than that of DPPC.34
In light of what has been previously said, HSPC liposomes
used in oral formulations would release all the entrapped water-
soluble substances during their interaction with bile salts. In
the presence of BSE, and although they cannot be solubilised by
it, the organisation of the gel-state bilayers is drastically altered,
resulting in an increased permeability and, nally, mixed nano/
microstructures. For this reason, if lipophilic substances were
included in the bilayers, they would not be as eciently trans-
ferred to mixed micelles, as in the case of SPC liposomes.
From the results obtained with HSPC/CHOL liposomes
(Fig. 3, 7 and 8) it can be inferred that the incubations per-
formed at 37 C and with high BSE concentrations resulted in
mixed vesicles, which retain part of their initial entrapped
aqueous volume. Despite the signicant morphological
changes, they retained a high percentage of the entrapped
uorescent dye. This behaviour markedly contrasts with that of
uid (SPC) and gel-state (HSPC) liposomes at 37 C and provides
evidence that the liquid-ordered bilayer of HSPC/CHOL vesicles
can undergo a very particular interaction with the BSE bile salts
during the intestinal digestion.
In this regard, the morphological changes of HSPC/CHOL
liposomes that cause a decrease in the absorbance (Fig. 8)
and mean diameter (Fig. 3) (characteristic events of the Ia stage)
are correlated with a leakage of the entrapped dye (Fig. 7). Note
that this was not the behaviour of SPC liposomes during the Ia

8 | Soft Matter, 2014, xx, 1–9

stage, where practically no release of HTPS nor insertion of the
surfactants into the bilayers were detected. This fact suggests
that, during the vesicle-apparent shrinkage of the HSPC/CHOL
liposomes, a surfactant insertion into the liquid-ordered bila-
yers takes place. At higher BSE concentrations, when a subse-
quent increase of the vesicle size is produced (stage Ib), the loss
of the uorescent dye ends, just the opposite of that which
occurs with SPC and HSPC liposomes. Therefore, there is a
critical inserted-surfactant to the lipid ratio that leads to a
strong stabilisation of the HSPC/CHOL membrane (Fig. 8) and
prevents the loss of the entrapped aqueous medium. Accord-
ingly, HSPC/CHOL liposomes were denitely the most resistant
vesicles to a complex natural mixture of bile salts, as demon-
strated by the fact that they were only solubilised at high bile
extract to lipid ratios and elevated temperatures (Table 1 and
Fig. 3 and 9). As can be observed in the inset of Fig. 9, at low
lipid concentrations there were large deviations from linearity,
and this fact prevented the precise determination of the vesic-
ular sub-domains.
Consequently, if HSPC/CHOL liposomes are used in oral
applications, it is expected that during the intestinal digestion
they would maintain a part of their entrapped aqueous material
and would preserve, partially, their vesicular structure. This
particular behaviour has to be due to their liquid-ordered
molecular organisation, induced by cholesterol, and, there-
fore, could be also caused by other lipophilic molecules that, at
the same time, could be of pharmacological interest. This could
be the case, for example, of phytosterols, which have a very
similar structure to cholesterol. In this sense, it has been shown
that, in mixtures with DPPC, some of these vegetable sterols
cause a similar phase behaviour to that caused by cholesterol,
and also induced, in dierent degrees, an increase of the bilayer
thickness.40
5 Conclusions
The present work shows that each one of the three liposome
preparations assayed show dierent types of interaction with
natural BSE at the human physiological concentration range,
that is, from 4 mM to 11 mM, pH 6.5 and 37 C. As all the
vesicles contained a phospholipid with a common headgroup
(choline) the dierent behaviour can be attributed mainly to the
dierence in their membrane status (uid, gel-state and liquid-
ordered bilayers). These particular responses to physiological
bile salts should be taken into account when designing lipo-
some formulations as drug carriers. They also indicate that each
type of vesicle oers singular features that can be useful in oral
delivery systems. Thus, our ndings may be useful for investi-
gators developing liposomal delivery systems in the oral form,
which is our main objective.
Acknowledgements
This research was supported (PTR95-0480-OP) by the
Ministry of Science and Technology of Spain. The authors
would like to thank Dr Josep Cladera for helpful comments
and suggestions.

This journal is ? The Royal Society of Chemistry 2014

References
1 R. R. Sawant and V. P. Torchilin, So Matter, 2010, 6, 4026–
4044.
2 M. Malmsten, So Matter, 2006, 2, 760–769.
3 G. Fricker, T. Kromp, A. Wendel, A. Blume, J. Zirkel,
H. Rebmann, C. Setzer, R. O. Quinkert, F. Martin and
K. Mu¨ller-Goymann, Pharm. Res., 2010, 27, 1469–1486.
4 P. Guan, Y. Lu, J. Qi, M. Niu, R. Lian, F. Hu and W. Wu, Int. J.
Nanomed., 2011, 6, 965–974.
5 S. K. Bobbala and P. R. Veerareddy, J. Liposome Res., 2012, 22,
285–294.
6 P. N. Gupta and S. P. Vyas, Colloids Surf., B, 2011, 82, 118–
21 M. M. Elsayed and G. Cevc, Biochim. Biophys. Acta, 2011,
1808, 140–153.
22 E. Schnitzer, M. M. Kozlov and D. Lichtenberg, Chem. Phys.
Lipids, 2005, 135, 69–82.
23 K. Andrieux, L. Forte, S. Lesieur, M. Paternostre, M. Ollivon
and C. Grabielle-Madelmont, Pharm. Res., 2004, 21, 1505–
1516.
24 A. Hildebrand, K. Beyer, R. Neubert, P. Garidel and A. Blume,
J. Colloid Interface Sci., 2004, 279, 559–571.
25 D. Wustner, A. Herrmann and P. Muller, J. Lipid Res., 2000,
41, 395–404.
26 M. R. Vist and J. H. Davis, Biochemistry, 1990, 29, 451–464.
27 The structure of biological membranes, ed. P. Yeagle, CRC
Press, 1992.
28 U. Gustafsson, S. Sahlin and C. Einarsson, World J.
Gastroenterol., 2003, 9, 1576–1579.
Paper Soft Matter
1
5
10
15
20
25
30
35
40
45
50
55
1
5
10
15
20
25
30
35
40
45
50
55

125.
7 W. Liu, A. Ye, W. Liu, C. Liu and H. Singh, J. Dairy Sci., 2013,
96, 2061–2070.
8 Y. Dai, R. Zhou, L. Liu, Y. Lu, J. Qi and W. Wu, Int. J.
Nanomed., 2013, 8, 1921–1933.
9 A. Macierzanka, N. M. Rigby, A. P. Coreld, N. Wellner,
F. Bo¨ttger, E. N. C. Millsa and A. R. Mackie, So Matter,
2011, 7, 8077–8084.
10 J. S. Dempe, R. K. Scheerle, E. Pfeier and M. Metzler, Mol.
Nutr. Food Res., 2013, 57, 1543–1549.
11 L. G. Hermida, A. Roig, C. Bregni, M. Sab´es-Xaman´ı and
R. Barnadas-Rodr´ıguez, J. Liposome Res., 2011, 21, 203–212.
12 A. Patel, Y. Hu, J. K. Tiwari and K. P. Velikov, So Matter,
2010, 6, 6192–6199.
13 R. P. Glahn, C. Lai, J. Hsu, J. F. Thompson, M. Guo and
D. R. van Campen, J. Nutr., 1998, 128, 257–264.
14 L. G. Hermida, M. Sab´es-Xaman´ı and R. Barnadas-
Rodr´ıguez, J. Liposome Res., 2009, 19, 207–219.
15 M. Jovan´ı, R. Barbera´, R. Farre´ and E. Mart´ın-de-Aguilera, J.
Agric. Food Chem., 2001, 49, 3480–3485.
16 R. P. Glahn, O. A. Lee, A. Yeung, M. I. Goldman and
D. D. Miller, J. Nutr., 1998, 128, 1555–1561.
17 V. N. Ngassam, M. C. Howland, A. Sapuri-Butti, N. Rosidic
and A. N. Parikh, So Matter, 2012, 8, 3734–3738.
18 U. Kragh-Hansen, M. le Maire and J. V. Moller, Biophys. J.,
1998, 75, 2932–2946.
19 D. Levy, A. Gulik, M. Seigneuret and J. L. Rigaud,
Biochemistry, 1990, 29, 9480–9488.
20 K. Andrieux, L. Forte, S. Lesieur, M. Paternostre, M. Ollivon
and C. Grabielle-Madelmont, Eur. J. Pharm. Biopharm., 2009,
71, 346–355.

This journal is ? The Royal Society of Chemistry 2014

29 M. D. Yago, V. Gonza´lez, P. Serrano, R. Calpena,
M. A. Mart´ınez, E. Mart´ınez-Victoria and M. Man˜as,
Nutrition, 2005, 21, 339–347.
30 R. Barnadas-Rodr´ıguez and M. Sab´es-Xaman´ı, Methods
Enzymol., 2003, 367, 28–46.
31 D. Lichtenberg, R. J. Robson and E. A. . Dennis, Biochim.
Biophys. Acta, 1983, 737, 285–304.
32 M. Paternostre, O. Meyer, C. Grabielle-Madelmont,
S. Lesieur, M. Ghanam and M. Ollivon, Biophys. J., 1995,
69, 2476–2488.
33 M. Ueno, Biochemistry, 1989, 28, 5631–5634.
34 A. Hildebrand, R. Neubert, P. Garidel and A. Blume,
Langmuir, 2002, 18, 2836–2847.
35 Y. Roth, E. Opatowski, D. Lichtenberg and M. M. Kozlov,
Langmuir, 2000, 16, 2052–2061.
36 M. Niu, Y. Tan, P. Guan, L. Hovgaard, Y. Lu, J. Qi, R. Lian,
X. Li and W. Wua, Int. J. Pharm., 2014, 460, 119–130.
37 A. Hildebrand, K. Beyer, R. Neubert, P. Garidel and A. Blume,
Colloids Surf., B, 2003, 32, 335–351.
38 W. Liu, A. Ye, C. Liu, W. Liu and H. Singh, Food Res. Int.,
2012, 48, 499–506.
39 M. Kokkona, P. Kallinteri, D. Fatouros and S. G. Antimisiaris,
Eur. J. Pharm. Sci., 2000, 9, 245–252.
40 M. Arif Kamal and V. A. Raghunathan, So Matter, 2012, 8,
8952–8958.

Soft Matter, 2014, xx, 1–9 | 9