Biochimica et Biophysica Acta 1848 (2015) 1963–1973

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbamem

Impact of two different saponins on the organization of model

lipid membranes
Beata Korchowiec a,⁎, Marcelina Gorczyca b, Kamila Wojszko a,c, Maria Janikowska b,d,
Max Henry c, Ewa Rogalska c,⁎
a
Department of Physical Chemistry and Electrochemistry, Faculty of Chemistry, Jagiellonian University, ul. R. Ingardena 3, 30-060 Krakow, Poland
b
Department of Theoretical Chemistry, Faculty of Chemistry, Jagiellonian University, ul. R. Ingardena 3, 30-060 Krakow, Poland
c
Structure et Réactivité des Systèmes Moléculaires Complexes, BP 239, CNRS/Université de Lorraine, 54506 Vandoeuvre-lès-Nancy cedex, France
d
Faculty of Physics, Astronomy, and Applied Computer Science, Jagiellonian University, ul. S. Lojasiewicza 11, 30-348 Krakow, Poland

a r t i c l e i n f o a b s t r a c t

Article history: Saponins, naturally occurring plant compounds are known for their biological and pharmacological activity. This
Received 23 February 2015 activity is strongly related to the amphiphilic character of saponins that allows them to aggregate in aqueous so-
Received in revised form 2 June 2015 lution and interact with membrane components. In this work, Langmuir monolayer techniques combined with
Accepted 4 June 2015
polarization modulation infrared reflection-absorption spectroscopy (PM-IRRAS) and Brewster angle microscopy
Available online 6 June 2015
were used to study the interaction of selected saponins with lipid model membranes. Two structurally different
Keywords:
saponins were used: digitonin and a commercial Merck Saponin. Membranes of different composition, namely,
Digitonin cholesterol, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine or 1,2-dipalmitoyl-sn-glycero-3-phospho-rac-(1-
Saponin–membrane interactions glycerol) were formed at the air/water and air/saponin solution interfaces. The saponin–lipid interaction was
Langmuir films characterized by changes in surface pressure, surface potential, surface morphology and PM-IRRAS signal. Both
Lipid monolayers saponins interact with model membranes and change the physical state of membranes by perturbing the lipid
acyl chain orientation. The changes in membrane fluidity were more significant upon the interaction with
Merck Saponin. A higher affinity of saponins for cholesterol than phosphatidylglycerols was observed. Moreover,
our results indicate that digitonin interacts strongly with cholesterol and solubilize the cholesterol monolayer at
higher surface pressures. It was shown, that digitonin easily penetrate to the cholesterol monolayer and forms a
hydrogen bond with the hydroxyl groups. These findings might be useful in further understanding of the saponin
action at the membrane interface and of the mechanism of membrane lysis.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction
Saponins are naturally occurring surface active glycosides, which
possess amphiphilic character originating from lipophilic aglycone (sa-
pogenin) and hydrophilic sugar moieties [1,2]. The hydrophobic part
may be steroid (27C) or triterpenoid (30C) [3]. Most known saponins
are monodesmosidic, which means that only one position of the agly-
cone is glycosylated. The sugar moiety is attached by an ether linkage
to the C3 hydroxyl group occurring in the majority of sapogenins [4,5].
Saponins can possess from one to three straight or branched sugar
moieties. The most often occurring sugar moieties are D-glucose, L-
rhamnose, D-galactose, D-glucuronic acid, L-arabinose, D-xylose or D-
fucose. The sugar chain contains from one to several monosaccharide
residues [3].
Saponins are widely distributed in plants, but also in lower marine
animals, such as starfish or sea cucumber [1,4,6]. Steroidal saponins
⁎ Corresponding authors.
E-mail addresses: bkorch@chemia.uj.edu.pl (B. Korchowiec),
ewa.rogalska@univ-lorraine.fr (E. Rogalska).
have been detected in oats, tomato seeds, alliums, asparagus, yam,
yucca, fenugreek, ginseng, capsicum peppers and aubergine, whereas
triterpenoid saponins have been found in many legumes and also in al-
liums, tea, spinach, sugar beet, liquorice, sunflower, ginseng, horse
chestnut and quinoa. Their main role is protection of organisms from
harmful effects of pathogens, thanks to their anti-fungal, anti-virial
and anti-bacterial characteristics [1,7].
The name ‘saponin’, reflects a wide-spread ability of saponins to
form soap-like foams in aqueous solutions. An important property of sa-
ponins is their lytic action on erythrocyte membranes as a result of affin-
ity of the aglycone moiety to membrane cholesterol [2]. An interesting
property of saponins is cytotoxity. Most studies concerning the latter
effect focus on the impact of saponins on cancer cell lines. The effect of
saponins on different human cancers such as leukemia, melanoma,
CNS, non-small cell lung, colon, ovarian, breast, renal and prostate is
described in the literature [3,7]. Saponins can act extracellularly or in-
tracellularly on tumor cells. The first way of interaction is direct inhibi-
tion of membrane proteins or changing membrane permeability.
Induction of apoptosis and cell cycle arrest are the most frequent intra-
cellular pathways. Nowadays, saponins are used in cancer therapy to

http://dx.doi.org/10.1016/j.bbamem.2015.06.007
0005-2736/© 2015 Elsevier B.V. All rights reserved.
1964 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973

inhibit angiogenesis and drug efflux. They also possess anti-metastatic
properties [7].
In this work, a commercial Merck Saponin and a digitonin, two types
of saponins with different structures, were studied for their ability to in-
teract with membrane lipids. The chemical structures of saponins are
shown in Fig. 1. Digitonin is a steroidal, monodesmosidic saponin
(Fig. 1A, B). It contains a pentasaccharide moiety consisting of two ga-
lactose, two glucose and one xylose monosaccharide residues. Digitonin
naturally occurs in a ubiquitous plant, Digitalis purpurea [8]. Like most of
saponins, digitonin increases permeability of biological membrane and
disperses membrane-bound proteins [9]. Digitonin may also act as a
competitive inhibitor for P-glycoprotein (P-gp), which is an important
protein in cancer therapy; P-gp is responsible for removing many for-
eign substances from the cell [10]. However, the most common proper-
ty of digitonin is its affinity for cholesterol. One possible application of
digitonin is detection of cholesterol in the cell. Raj and coworkers [11]
developed a method of detection of cholesterol with digitonin-
conjugated gold nanoparticles. Vermeer et al. [12] and Fornas et al.
[13] used digitonin as a protective agent for preparing liver tissue for
electron microscopy to avoid loss of cholesterol. Augustin and co-
workers [4] studied complexation of cholesterol with digitonin. Due to
the fact that aglycone part of digitonin is lipophilic, digitonin can
spontaneously incorporate into the membrane. Subsequently, the ste-
roidal glycoalkaloid-cholesterol 1:1 complexes would form and accu-
mulate in the membrane. The mechanism of this process is not
completely understood. However, Armah et al. [14] proposed that steric
properties of saponin–cholesterol complexes may cause membrane cur-
vature and, in consequence, may provoke vesiculation or pore forma-
tion in the cell membrane.
As a second compound, the commercial Merck Saponin was used
(Fig. 1C, D). Merck Saponin is a saponin fraction isolated from the
roots of Gypsophila paniculata L. and Gypsophila arrostii Guss. The chem-
ical structure of two major components of Merck Saponin was
established using NMR techniques [15,16]. The aglycone moiety was
identified as gypsogenin, while seven or eight monosaccharide units
were identified in the sugar moiety (Fig. 1). Due to its high hemolytic
properties, Merck Saponin was widely used as a standard for hemolytic
tests [17]. Its toxic effect is much lower when given orally, as it cannot
enter the blood stream by crossing the intestine, and the presence of
blood plasma reduces the hemolytic effect [18]. The effect of Gypsophila
saponins on plasma cholesterol concentration was studied by different
groups [19,20]. In all cases, the authors showed that in rats, ingestion
of saponins resulted in a significant reduction of blood cholesterol
concentration.

Fig. 1. Structures of digitonin (A, B) and two major components of Merck Saponin; R = H or β-D-xylopyranosyl (C, D). (A, C): chemical structures, and (B, D): 3-D structures obtained at HF/
6-31G(d) level of theory; in Merck Saponin panel D: R = H. Molecular volumes of Merck Saponin and digitonin (B, D) are, 6241 Å3 and 4435 Å3, respectively. Molecular volume was defined
by 0.001 a.u. isosurface of electron density of molecule. This calculations were performed at HF/6-31G(d) level of theory using Gaussian 0.9 software [52]. Color code: carbons in black,
oxygens in red, and hydrogens in gray.
 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973
Despite a number of studies dealing with saponin–membrane inter-
actions, very few involved Langmuir monolayers [14,21–23]. In the last
years, Langmuir lipid monolayers were extensively used to investigate
interactions with different compounds co-spread at the interface or dis-
solved in the subphase [24–27]. These model membranes are useful for
obtaining information on the mechanism of saponin–membrane inter-
actions. Here, to better understand the effect of Merck Saponin or digi-
tonin on the lipid model membranes, the Langmuir film technique
was used to simultaneously measure the surface pressure and surface
potential. Langmuir monolayers formed with cholesterol (Chol), 1,2-
dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-dipalmitoyl-sn-
glycero-3-phospho-rac-(1-glycerol) (DPPG), as well as DPPC/Chol or
DPPG/Chol mixtures were formed on subphases containing Merck Sa-
ponin or digitonin. Brewster angle microscopy was used to study the
morphological properties of pure lipid and mixed lipid/saponin mono-
layers. To examine the molecular organization and molecular interac-
tion in model membranes, polarization modulation infrared reflection
absorption spectroscopy was used.
2. Materials and methods
2.1. Materials and reagents
Digitonin (66% pure) and cholesterol (Chol, ~ 99% pure) were
from Sigma-Aldrich. 1,2-dipalmitoyl-sn-glycero-3-phosphocholine
(DPPC; 99% pure) and 1,2-dipalmitoyl-sn-glycero-3-phospho-rac-
(1-glycerol) sodium salt (DPPG; 99% pure) were from Avanti
Polar Lipids. The structures of two major saponins present in the
commercial Merck Saponin (Merck No. 7695; supply interrupted),
were determined by high-field gradient-enhanced and homo-
and heteronuclear 2D NMR methods [15,16]. The two major compo-
nents were identified as: 3-O-β-D -xylopyranosyl-(1 → 3)-[β-D -
galactopyranosyl-(1 → 2)]-β- D -glucuronopyranosyl gypsogenin
28-O-(6-deoxy-β-D-glucopyranosyl)-(1 → 4)-[β-D -xylopyranosyl-
(1 → 3)-β- D -xylopyranosyl-(1 → 4)]-α- L -rhamnopyranosyl-
(1 → 2)-β-D-fucopyranoside, and 3-O-β-D-xylopyranosyl-(1 → 3)-
[β-D-galactopyranosyl-(1 → 2)]-β-D-glucuronopyranosyl gypsogenin
28-O-β-D-glucopyranosyl-(1 → 3)-[β-D-xylopyranosyl-(1 → 4)]-α-L-
rhamnopyranosyl-(1 → 2)-β-D-fucopyranoside. The chemical struc-
tures of the saponins studied in this work are shown in Fig. 1.
The aqueous solutions of Merck Saponin and digitonin were pre-
pared using Milli-Q water, which had resistivity of 18 MΩ cm and sur-
face tension of 72.8 mN m−1 at 20 °C. Spectrophotometric grade
chloroform (Sigma-Aldrich) was used for preparing lipid solutions.
2.2. Compression isotherms and Brewster angle microscopy
The surface pressure (Π) and electric surface potential (ΔV) mea-
surements were carried out with a KSV 2000 Langmuir balance (KSV In-
struments, Helsinki). A Teflon trough (58 × 15 × 1 cm) with two
symmetrically moving, hydrophilic Delrin barriers was used in all ex-
periments. The surface pressure was measured by the Wilhelmy meth-
od using a platinum plate. Surface potential was monitored by means of
KSV Spot 1 with a vibrating electrode and a stainless steel as a counter
electrode immersed below the water surface. Before each measure-
ment, the subphase surface was cleaned by sweeping and suction. All
solvents used for cleaning the trough and the barriers were of analytical
grade. The stability of the surface potential signal was checked before
each experiment, after cleaning the water surface. After the ΔV signal
had reached the constant value it was zeroed, and the film was spread
on the subphase. The pure Chol, DPPC and DPPG as well as DPPC/Chol
and DPPG/Chol mixtures of 0.3 molar fraction of Chol, were spread
with a Hamilton syringe on water subphase or aqueous saponin solu-
tion and left for 15 min to allow solvent evaporation and to reach an
equilibrium state of the monolayer. All isotherms were recorded upon
symmetric compression of the monolayer at a constant barrier speed
1965
of 2.5 Å2 molecule− 1 min−1. A PC computer and KSV software were
used to control the experiments. Each compression isotherm was per-
formed at least three times. The Π–A and ΔV–A isotherms were record-
ed at 20 ± 0.1 °C. The standard deviation obtained from compression
isotherms was ±0.5 Å2 on molecular area (A), ±0.2 mN m−1 on surface
pressure and ±0.01 V on surface potential.
The morphology of the films was imaged with a computer-interfaced
KSV 2000 Langmuir balance combined with a Brewster angle microscope
(KSV Optrel BAM 300, Helsinki). The Teflon trough dimensions were
58 × 6.5 × 1 cm; other experimental conditions were as described above.
2.3. Adsorption kinetics measurements
Adsorption experiments were carried out in a Teflon dish with a
subphase volume of 20 mL. The pure Chol, DPPC, DPPG, and mixed
DPPC/Chol and DPPG/Chol solutions were spread at the air–water
interface using a Hamilton syringe to obtain an initial surface pressure,
Π = 30 mN m−1. This value corresponds to the lateral pressure in the
biological bilayers [28]. After 20 min of equilibration, a small volume
of the concentrated saponin solution was injected into the subphase be-
neath the lipid film. The subphase was gently stirred with a magnetic
stirrer for 30 s. The final saponin concentration in the subphase was
78.5 μg mL−1. The variation of the surface pressure was recorded as a
function of time using a KSV 2000 Langmuir balance (KSV Instruments,
Helsinki). The registration of adsorption kinetics started immediately
after the injection of saponin into the subphase. All measurements
were performed at 20 °C.
2.4. Polarization–modulation infrared reflection–absorption spectroscopy
(PM-IRRAS)
The PM-IRRAS spectra of pure Chol, DPPC, DPPG and mixed
DPPC/Chol or DPPG/Chol monolayers formed on pure water and sapo-
nin solution subphases were registered at 20 °C. The Teflon trough di-
mensions were 36.5 × 7.5 × 0.5 cm; other experimental conditions
were as described in the preceding paragraph. The PM-IRRAS measure-
ments were performed using a KSV PMI 550 instrument (KSV Instru-
ments, Helsinki). The PMI 550 instrument contains a compact Fourier
Transform IR-spectrometer equipped with a polarization–modulation
(PM) unit on one arm of a goniometer, and a MCT detector on the
other arm. The incident angle of the light beam can be freely chosen be-
tween 40° and 90°; here the incident angle was 79°. The spectrometer
and the PM unit operate at different frequencies, allowing separation
of the two signals at the detector. The PM unit consists of a photoelastic
modulator (PEM), which is an IR-transparent ZnSe piezoelectric lens.
The incoming light is continuously modulated between s- and p-
polarization at a frequency of 74 kHz. This allows simultaneous mea-
surement of spectra for the two polarizations, the difference providing
surface specific information and the sum providing the reference spec-
trum. As the spectra are measured simultaneously, the effect of water
vapor is largely removed. The PM-IRRAS spectra of the film-covered sur-
face, S(f), as well as that of the subphase, S(w), were measured and the
normalized difference ΔS/S = [S(f) − S(w)]/S(w) is reported. 6000 in-
terferogram scans (10 scans per second) have been acquired for each
spectrum. In the mid-IR region, the wavenumber at which the half-
wave retardation takes place can be freely selected. Here, the maximum
of PEM efficiency was set to either 1500 or 2900 cm−1 for analyzing the
different regions of the spectra. The spectral range of the device is
800–4000 cm−1 and the resolution is 8 cm−1.
3. Results and discussion
3.1. Compression isotherms and Brewster angle microscopy
Merck Saponin and digitonin are water soluble saponins. They lower
the surface tension of aqueous solutions and form micelles when
1966 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973
reaching the critical micelle concentration (CMC). Since the CMC of
Merck Saponin was unknown, the surface tension (γ)–bulk concentra-
tion (c) dependency was established (Fig. 2A). This was done by
collecting the equilibrium surface pressure (П) data for a number of
Merck Saponin solutions with different bulk concentrations. П was
monitored as a function of time (Fig. 2B), to ensure that the calculated
values of γ used to plot the γ-log c dependencies correspond to the ther-
modynamic equilibria of the solutions. These results show, that the ad-
sorption time increases with the decreasing concentration of Merck
Saponin. For the high bulk concentrations reaching the equilibrium is
of the order of minutes (Fig. 2B, black line), whereas for the low bulk
concentrations the time to attain the equilibrium is above 20 h (yellow
line). The CMC value determined from the inflection point on the γ-log c
plot is 2.3 mg mL−1. In the case of digitonin, the CMC value provided by
Sigma-Aldrich [29] is approximately 0.6 mg mL−1.
The interaction of Merck Saponin or digitonin with lipids was stud-
ied using monolayers spread on low concentration solutions below
the CMC, to avoid saponin aggregation; concentration of 78.5 μg mL−1
was used with both saponins in all experiments.
In order to investigate the affinity of Merck Saponin and digitonin to
lipid Langmuir monolayers, compression isotherms of the monolayers
formed with Chol, DPPC or DPPG were registered. The lipid monolayers
were spread on pure water, as well as on Merck Saponin or digitonin
aqueous solutions. The Π–A and ΔV–A isotherms obtained upon com-
pression of the Chol, DPPC or DPPG monolayers at 20 °C are shown in
Fig. 3A, C and E, respectively. In Fig. 3B, D and F, the compressibility
modulus–surface pressure (C−1 S –Π) plots are presented.
The Π–A isotherms of all three pure lipids spread on the aqueous so-
lutions of Merck Saponin and digitonin differ considerably compared to
the pure water subphase. Indeed, the isotherms registered in the
presence of Merck Saponin are strongly shifted to high surface pressures;
the shift is lower in the case of digitonin. This indicates that the saponins
penetrate from the subphase into the phospholipid film. It has to be no-
ticed, both in the case of Π–A and ΔV–A dependencies, that saponins in-
sert in the low and high surface pressures lipid film. The only exception
is Chol/digitonin system for which Π–A isotherm illustrates the insertion
of digitonin at low surface pressures and a loss of the film forming mate-
rial at higher surface pressures. It suggests, that at low surface pressures
digitonin adsorbed to the Chol film and was maintained in the film by in-
teraction with Chol, while at high surface pressures both Chol and digito-
nin were gradually expelled from the film, shown by the low slope of
90
40
A B
Π (mN m-1)
80 30
20
70
γ (mN m-1)
10
0
60 0 5 10 15 20 25
Time (h)
50
40
30
-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0
log c
Fig. 2. Surface activity of Merck Saponin. (A) the adsorption isotherm of Merck Saponin at
the air/water interface. (B) Π (t) adsorption curves obtained for different concentrations
of saponin in aqueous solutions: 17.7 μg mL−1 (yellow), 78.5 μg mL−1 (navy),
157.0 μg mL−1 (gray), 392.5 μg mL−1 (orange), 785.0 μg mL−1 (magenta), 1.3 mg mL−1
(cyan), 2.7 mg mL−1 (blue), 9.2 mg mL−1 (green), 13.1 mg mL−1 (red) and
30.5 mg mL−1 (black).
Chol/digitonin surface pressure isotherm and lack of the collapse point
on the isotherm. This effect correlates well with ΔV–A results, which
show relatively small changes of the surface potential values, indicating
almost unchanged orientation of molecules upon compression of the
film. In the case of the other lipid/saponin systems, the ΔV–A dependen-
cies are modified in the presence of the saponin in the subphase com-
pared to pure water. Indeed, at high molecular areas the ΔV values
show lower fluctuations and in the most cases are higher compared to
the pure lipid films; this indicates the monolayer stabilization and order-
ing. On the other hand, the lower ΔV values observed in the most con-
densed phase regions with the Chol/Merck Saponin, Chol/digitonin,
DPPC/Merck Saponin, DPPC/digitonin and DPPG/digitonin compared to
the pure lipid films may indicate interactions between the saponin pres-
ent in the subphase and the lipid polar heads leading to some conforma-
tional disordering of the lipid molecules. The increase of the surface
potential observed in the case of the DPPG/Merck Saponin film com-
pared to the pure water subphase indicates that the DPPG hydrocarbon
chains reorient and acquire a more vertical orientation in the presence
of Merck Saponin.
It can be also observed that the compressibility modulus (C− 1
S ;
−1
CS = −A(∂Π/∂A)T [30]) values calculated from the Π–A dependency
(Fig. 3B, D and F) are lower for the monolayers spread on the Merck Sa-
ponin or digitonin solutions compared to pure water, indicating a
higher fluidity of the films containing either molecule. This result to-
gether with the surface potential results suggest that the saponins
inserted in the film modify the orientation of the lipids spread at the
air–water interface. The presence of Merck Saponin in the subphase in-
duces a more liquid-like character of the films compared with digitonin.
The C−1S values obtained from the isotherms at the most condensed
states of the monolayers are much lower for saponin solutions than
for pure water subphase. The latter is in accordance with the other re-
sults showing the presence of the saponin in the highly condensed
Chol, DPPC and DPPG monolayers. Interestingly, the C−1 S value obtained
from the Chol/digitonin isotherm at the most condensed state of the
monolayer is very low (89 mN m−1) and supports the conjecture that
cholesterol and digitonin were squeezed out from the more packed film.
The impact of saponins on the film structure can be observed using
BAM. The BAM snapshots of the Chol, DPPC and DPPG monolayers on
the pure water subphase and aqueous solutions of Merck Saponin and
digitonin are presented in Fig. 4. The images of the lipid film spread on
three different subphases were recorded at 45 Å2 for Chol and 55 Å2
for DPPC or DPPG. At the chosen molecular area the Chol film formed
on water and Merck Saponin solution is isotropic, as shown in Fig. 4A
and B respectively. However, it has to be noticed, that the image obtain-
ed on Merck Saponin subphase is brighter compared to that of the pure
water subphase. This effect, which shows as well in the analysis of the
isotherms, indicates a higher density of the film due to the incorporation
of Merck Saponin. On the contrary, some anisotropy can be observed at
this molecular area in the Chol film spread on the digitonin solution
(Fig. 4C); the bright pattern observed with BAM indicates that digitonin
is present in the monolayer and forms mixed Chol/digitonin structures.
Formation of the digitonin-cholesterol structures may favor removing
of the latter to the aqueous subphase from the pure, condensed Chol
film, while at low surface pressures, the adducts can be accumulated
in the film. Overall, the results obtained in our work are in accordance
with the literature. Indeed, while cholesterol alone is insoluble in
water, the cholesterol/digitonin complexes solubility is 0.6 mg L−1 [12].
The images corresponding to the LE–LC phase transition in the pure
DPPC and in DPPC/Merck Saponin and DPPC/digitonin films are present-
ed in Fig. 4D, E and F, respectively. It can be observed that the condensed
phase domains formed in the DPPC film spread on the digitonin solution
(Fig. 4F) are smaller and less condensed compared to the pure water
subphase. The latter indicates that digitonin is present in the monolayer
and alters the density of the monolayer. In the case of the DPPC/Merck
Saponin film, the condensed phase domains are not observed. Instead,
irregular bright spots can be seen (Fig. 4E), indicating formation of a
 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973 1967

A 0.5 B 1000
70
0.4
60 0.3 800

CS (mN m )
50 0.2

-1
Π (mN m )
-1

600

ΔV (V)
40 0.1
0.0
30 400

-1
-0.1
20
-0.2 200
10 -0.3
0 -0.4 0
30 40 50 60 70 80 90 0 10 20 30 40 50
2 -1
A (Å ) Π (mN m )
C D 400
90 0.6
0.5 350
80
0.4 300
70
0.3

CS (mN m )
Π (mN m )

250

-1
60
-1

0.2

ΔV (V)
50 0.1 200
40 0.0
150

-1
30 -0.1
-0.2 100
20
-0.3
10 50
-0.4
0 -0.5 0
30 40 50 60 70 80 90 100 110 0 10 20 30 40 50 60 70
2 -1
A (Å ) Π (mN m )
E F
1000
90 0.4
80 0.3
800
70 0.2
CS (mN m )

0.1
-1

60
Π (mN m )
-1

0.0 600
ΔV (V)

50
-0.1
40 400
-0.2
-1

30 -0.3
20 -0.4 200
10 -0.5
0 -0.6 0
30 40 50 60 70 80 90 100 110 0 10 20 30 40 50 60 70
2 -1
A (Å ) Π (mN m )

Fig. 3. Compression isotherms of Chol (A), DPPC (C) and DPPG (E) films. Solid lines, Π–A isotherms; dashed lines, ΔV–A isotherms. (B, D and F) C−1
S –Π dependency for Chol (B), DPPC
(D) and DPPG (F) films. Subphases: pure water (black); Merck Saponin (blue); digitonin (red).

mixed, phospholipid/Merck Saponin film. Indeed, the LE–LC phase tran-
sition was not observed in the corresponding Π–A isotherm, which is
strongly shifted towards the higher surface pressures compared to
those formed on water subphase.
BAM images of pure DPPG, DPPG/Merck Saponin and DPPG/digitonin
films (Fig. 4G, H and I, respectively) show that DPPG in the presence of
digitonin forms a homogenous, isotropic film, while in the presence of
Merck Saponin reveals an anisotropic morphology similar to the pure
DPPG monolayer. A brighter shade of the DPPG/Merck Saponin and
DPPG/digitonin film snapshots compared to those of the pure DPPG
film suggests adsorption or penetration of saponins to the DPPG mono-
layers and, consequently, a higher interfacial molecule concentration.
To better understand the interaction between saponins and lipid
model membranes, the properties of mixed DPPC/Chol and DPPG/Chol
monolayers upon interaction with Merck Saponin and digitonin were
investigated. The compression isotherms obtained with mixed films
containing 30 mol% of cholesterol, the biologically relevant content
[31,32], are shown in Fig. 5A and C. The plots of C−1 S –Π dependency
are presented in Fig. 5B and D.
In the case of the DPPC/Chol and DPPG/Chol mixed monolayers
(Fig. 5A and C, respectively), the shift of the Π–A isotherm to higher mo-
lecular areas in the presence of digitonin in the subphase is more signif-
icant than that observed with pure DPPC and DPPG monolayers. For
example, at the surface area of 55 Å2 the DPPC isotherm shifts to
14 mN m−1, while that of mixed DPPC/Chol film to 45 mN m−1. The lat-
ter suggests that a higher number of the digitonin molecules is incorpo-
rated into the mixed DPPC/Chol film compared to the pure DPPC. As
previously, the isotherms of the pure DPPG and mixed DPPG/Chol
monolayers shift respectively to 15 and 16 mN m−1. The smaller shift
of mixed DPPG/Chol isotherm compared to the DPPC/Chol indicates
that the affinity of digitonin for the mixed DPPG/Chol film is much
lower compared to DPPC/Chol film. A strong penetration and interac-
tion of digitonin with the lipid molecules was also confirmed by the
collapse area values. As shown in Fig. 5A and C, the isotherms
1968 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973

Fig. 4. Brewster angle micrographs of Chol (A–C), DPPC (D–F) and DPPG (G–I) monolayers spread on pure water (A, D, G) and on the subphases containing Merck Saponin (B, E, H) and
digitonin (C, F, I). The micrographs (A–C) and (D–I) were taken at 45 Å2 and 55 Å2, respectively. Scale: the width of the snapshots corresponds to 400 μm.

corresponding to the mixed films formed on digitonin solution (red The latter suggests that the digitonin molecules incorporated into the
lines) show higher molecular area values at the collapse point, com- mixed lipid films were not squeezed out from the highly compressed
pared to the isotherm obtained on pure water subphase (black lines). monolayers. This effect is more important for the zwitterionic DPPC

A B
90 0.6 500
80
70 0.4 400
CS (mN m )
-1
Π (mN m )

60 0.2
-1

300
50
ΔV (V)

0.0
40
200
-1

30 -0.2
20 100
-0.4
10
0 -0.6 0
30 40 50 60 70 80 90 100 0 10 20 30 40 50 60
2 -1
A (Å ) Π (mN m )
C D
600
80 0.4
70 500
0.2
60
CS (mN m )

400
Π (mN m )

-1
-1

50 0.0
ΔV (V)

40 300
-0.2
30
-1

-0.4 200
20
-0.6 100
10
0 -0.8 0
30 40 50 60 70 80 90 100 0 10 20 30 40 50 60
2 -1
A (Å ) Π (mN m )

Fig. 5. Compression isotherms of DPPC/Chol (A) and DPPG/Chol (B) mixed films. Solid lines, Π–A isotherms; dashed lines, ΔV–A isotherms. (C, D) C−1
S –Π dependency for DPPC/Chol
(C) and DPPG/Chol (D) films. Subphases: pure water (black); Merck Saponin (blue); digitonin (red). T = 20 °C.
 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973
compared to the negatively charged DPPG. Moreover, the results
obtained for DPPG film spread on a Merck Saponin solution are in accor-
dance with those obtained for digitonin. At the surface area of 55 Å2 the
shift of the pure and mixed DPPG isotherms is 31 and 34 mN m−1, re-
spectively, showing a stronger penetration of Merck Saponin to the
mixed system compared to the pure one. On the contrary, the penetra-
tion of Merck Saponin to the DPPC/Chol film (22 mN m−1 shift) is lower
compared to DPPC film (32 mN m−1 shift).
The properties of the mixed phospholipid/Chol films were evaluated
based on their compressibility (Fig. 5B and D). It can be seen that intro-
ducing of cholesterol into the zwitterionic DPPC film results in film con-
densation (Fig. 5B). The compressibility modulus, calculated at the most
condensed state of DPPC/Chol mixed film spread on a digitonin solution
is C−1
S = 429 mN m− 1, which is higher than that for the DPPC film
(C−1
S = 256 mN m−1). Similarly to DPPC, the increase of condensation
was also observed in the case of DPPG/Chol/digitonin system (Fig. 5D).
The maximum value for DPPG/Chol mixed film C−1 S = 387 mN m−1,
−1 −1
while for the pure DPPG CS = 354 mN m . Interestingly, the opposite
effect was observed in the case of mixed films formed on Merck Saponin
solution; an important decrease in the C− S
1
values, namely 94 and
−1
173 mN m , respectively, for DPPC and DPPG, indicate a more fluid-
like character of the mixed films. These results may suggest modifica-
tion of the conformation of both lipids upon the interaction with
Merck Saponin.
The run of ΔV–A isotherms of the mixed systems corresponds to that
described earlier for the pure lipid films. The higher ΔV values, observed
in the presence of Merck Saponin, can be explained by a more upright
orientation of the phospholipid hydrocarbon chains.
The BAM snapshots of the DPPC/Chol and DPPG/Chol mixed films
presented in Fig. 6 complete the compression isotherm results. The
brightness of the images in Fig. 6C and F reflects a higher condensation
of monolayers formed on digitonin solution compared to the respective
monolayers on Merck Saponin solution (Fig. 6B and E).
3.2. Adsorption kinetics
In order to compare the effect of Merck Saponin and digitonin on the
lipid membranes, adsorption kinetics were determined. First, a pure
lipid or mixed solution was spread at the water–air interface to obtain
an initial surface pressure of approximately 30 mN m−1. Then a small
volume of the concentrated saponin solution was injected into the
water subphase, beneath the lipid film, to reach a final concentration
of 78.5 μg mL−1. The adsorption of Merck Saponin and digitonin to
the pure Chol, DPPC and DPPG films, as well as to the mixed DPPC/Chol
and DPPG/Chol films was followed by recording the surface pressure
Fig. 6. Brewster angle micrographs of DPPC/Chol (A–C) and DPPG/Chol (D–F) monolayers spread on pure water (A, D) and on the subphase containing Merck Saponin (B, E) and digitonin
(C, F). The micrographs were taken at 55 Å2. Scale: the width of the snapshots corresponds to 400 μm.
1969
variations (at a constant film area) as a function of time. The corre-
sponding adsorption isotherm profiles are shown in Fig. 7.
The results show that both saponins penetrate from the subphase
into the condensed lipid monolayers and lower their surface tension.
However, the ΔΠ effect observed upon the penetration of Merck
Saponin to the lipid layers (Fig. 7A) is significantly lower compared to
digitonin (Fig. 7B). For example, the maximum ΔΠ effect for Chol/
Merck Saponin and Chol/digitonin systems is observed at 14 and
31 mN m− 1, respectively, showing a higher affinity of digitonin
for Chol, compared to Merck Saponin. A higher ΔΠ effect was also ob-
served for DPPC/Chol/digitonin or DPPG/Chol/digitonin compared to
DPPC/Chol/Merck Saponin or DPPC/Chol/Merck Saponin, respectively.
It should be also noted that the ΔΠ–t dependencies of pure phospho-
lipids and their mixtures with Chol (red, green, magenta and blue
lines) registered for digitonin reach the equilibrium surface pressure
value within few seconds after injecting the digitonin into the water
subphase. This process is significantly lower in the case of Merck Sapo-
nin. These results indicate that adsorption/penetration of saponin into
the lipid film is correlated with the molecular structure of the saponins.
3.3. PM-IRRAS
The PM-IRRAS technique [33–35] was used to analyze changes in the
orientation of lipid chains induced by saponins, as well as the molecular
interactions between saponins and lipid membranes. The PM-IRRAS
spectra recorded at surface pressure of 30 mN m−1 are shown in
Fig. 8; the values of the characteristic vibrations of the spectra are
given in Table 1.
Changes in the lipid acyl chains packing and conformation can be de-
tected based on lipid CH2 antisymmetric and symmetric stretching vi-
brations located in the spectral region from 3000 to 2800 cm−1. In the
absence of saponins, two dominant bands were observed in the spec-
trum of Chol, namely, at 2912 and 2847 cm−1 (Table 1), which were
assigned to the all-trans conformation of acyl chains [36–38]. The shift
of these bands to higher frequencies in the presence of saponins reflects
the increased number of gauche conformers of the acyl chains. The shift
of νas (CH2) is much more pronounced for digitonin than for Merck Sa-
ponin, showing that the incorporation of digitonin to the Chol layer
leads to a marked decrease in conformational order of the acyl chains.
This is not surprising, since digitonin easily penetrates to the Chol
monolayers which results in the formation of less ordered films. On
the other hand, the bands due to ν (C–O) stretching appear in the fin-
gerprint region of the PM-IRRAS spectrum of Chol. This band is sensitive
to hydrogen bonding and shifts to lower frequency upon group hydra-
tion [39,40]. According to our results, the ν (C–O) band is observed at
1970 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973

A 35 B 35
30 30

ΔΠ (mN m )

ΔΠ (mN m -1)
25 25
-1
20 20
15 15
10 10
5 5
0 0
0 1200 2400 3600 4800 6000 7200 0 1200 2400 3600 4800 6000 7200
Time (s) Time (s)
Fig. 7. Adsorption kinetics of Merck Saponin (A) and digitonin (B) to different lipid monolayers at initial surface pressure Π = 30 mN m−1. ΔΠ is the difference between the final and initial
surface pressures. Monolayer lipids: Chol (black); DPPC (red), DPPG (green), DPPC/Chol (magenta); DPPG/Chol (blue). Final saponin concentration in the subphase was 78.5 μg mL−1.
Temperature: 20 °C.

1061, 1052 and 1047 cm−1, respectively for Chol, Chol/Merck Saponin
and Chol/digitonin system. These results can be explained by an
increased hydrogen bonding of the cholesterol –OH groups in the pres-
ence of both saponins; the effect is more important in the case of digito-
nin. Indeed, H-bond formation between the bulky sugar headgroups of
digitonin and hydroxyl groups of cholesterol as well as the hydrophobic
interaction may result in formation of more soluble and less stable Chol/
digitonin complex, which can be squeezed out into the aqueous
subphase at higher surface pressures. This effect of cholesterol–sugar
headgroup interaction was reported in our previous paper. Indeed,
cholesterol and glycolipids can establish a hydrogen bond network
linking the cholesterol hydroxyl group and the maltooligosaccharide
headgroup of the glycolipid [41].
The spectra of the pure DPPC and DPPG monolayers exhibit charac-
teristic vibrations of phospholipids, namely methylene and carbonyl
ester stretching vibrations [42–44]. While no significant influence of
the saponins on the organization of the hydrocarbon chains in DPPG
can be observed, the shift of the νas (CH2) and νs (CH2) to higher
wavenumbers in the case of DPPC indicates lower conformational or-
dering and higher freedom of acyl chains in the presence of saponins.
This effect, more significant for digitonin, is in accordance with the ΔV
and penetration results, showing molecule disordering in the monolay-
er and a more fluid-like character of the phospholipid/saponin mono-
layers. Moreover, the impact of saponins on the polar heads of lipids
was monitored via the frequencies of the stretching vibration of the car-
bonyl ester group ν (C_O) observed in the PM-IRRAS spectra at ap-
proximately 1730 cm−1. As seen in Fig. 8D and F (respectively for
DPPC and DPPG), the presence of saponins in the subphase induces a
blue-shift of ν (C_O) band, indicating dehydration of the ester groups.
The latter effect is larger in the presence of digitonin and may result
from an increasing interaction with the digitonin sugar headgroups.
It is interesting to compare the PM-IRRAS spectra of the pure phos-
pholipids with their mixtures with cholesterol. According to our results
obtained on a pure water subphase, the position of methylene
stretching bands in the DPPC/Chol spectrum indicates a greater role of
trans conformers in the arrangement of molecules in the mixed film,
compared to the pure DPPC. Moreover, the analysis of the C_O band
frequencies reveals that the addition of cholesterol leads to a small in-
crease in the number of hydrated carbonyl ester groups. These results
are consistent with the literature reports [45–47], which show a reduc-
tion in the number of gauche conformers and a larger contribution of
the hydrated ester groups of DPPC when mixed with Chol. No such
effects were observed in the case of DPPG. The values of the ν (CH2)
are generally similar for pure and mixed DPPG indicating that the addi-
tion of Chol does not affect the conformation of the phospholipid acyl
chains. The blue-shift of the ν (C_O) band visible upon incorporation
of Chol (from 1732 to 1738 cm−1) indicates a dehydration of carbonyl
ester groups in the negatively charged phospholipid. The impact of sa-
ponins on the mixed model membranes is also noticeable. The νas
(CH2) signals are blue-shifted in the PM-IRRAS spectra registered on sa-
ponin subphases. The small shift of 1 and 2 cm−1 of the νas (CH2) band
observed with DPPG/Chol in the presence of Merck Saponin and digito-
nin, respectively, indicates a limited penetration of saponin into the hy-
drophobic part of the negatively charged phosphoglyceride, while the
important shift for DPPC/Chol mixture (8 and 18 cm− 1, respectively
for Merck Saponin and digitonin) shows that both saponins penetrate
into the hydrophobic region of the monolayer. It can be observed that
Chol favors hydration of the phospholipid carbonyl groups in the pres-
ence of saponins, as shown by the red shift of the ν (C_O) bands. We
propose that saponin–cholesterol interaction results in a higher expo-
sure of the phospholipid carbonyl ester groups to water and, conse-
quently, a higher hydration of these groups compared to saponin-free
systems.
4. Conclusions
Here, two different saponin preparations, namely Merck Saponin
and digitonin were used to improve our understanding of the interac-
tion between amphiphilic glycosides and cell membranes. The results
obtained show that molecules present in both preparations penetrate
into the lipid monolayers at a concentration below the CMC, and sig-
nificantly increase the fluidity of the monolayers. It was found that
the saponins used here induce a conformational disorder of the
lipid acyl chains. The fluidizing effect was more significant in the
case of Merck Saponin bearing a large glycone moiety. It can be ob-
served that molecular volumes of Merck Saponin and digitonin are,
6241 Å3 and 4435 Å3, respectively; this volume difference can play a
role in membrane fluidization.
Both saponins used in this study show a higher affinity for cholesterol
compared to phosphatidylglycerols. It has to be noted that saponins con-
tain structural motifs that are similar to membrane sterols. The structural
likeness between the aglycone part of these glycosides and the sterol
part of cholesterol may be the main reason for a facilitated interaction.
Moreover, the saponin sugar residues may interact with the cholesterol

Fig. 8. PM-IRRAS spectra of Chol (A, B), DPPC (C, D), DPPG (E, F), DPPC/Chol (G, H) and DPPG/Chol (I, J) monolayer in the CH2 (A, C, E, G, I), C–O (B) and C_O (D, F, H, J) stretching regions.
Subphases: pure water (black); Merck Saponin (blue); digitonin (red). Saponins were injected beneath the lipid monolayer compressed to 30 mN m−1. T = 20 °C. Dashed lines:
deconvoluted bands.
 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973 1971

A 0.008 B
0.008
0.006
0.006

0.004
0.004

0.002 0.002

0.000 0.000
3000 2960 2920 2880 2840 2800 1100 1080 1060 1040 1020
-1 -1
Wavenumber (cm ) Wavenumber (cm )

C 0.010 D 0.006

0.008 0.005

0.004
0.006
0.003
0.004
0.002
0.002 0.001

0.000 0.000
3000 2960 2920 2880 2840 2800 1800 1780 1760 1740 1720 1700
-1 -1
Wavenumber (cm ) Wavenumber (cm )

E 0.008 F
0.006
0.006

0.004
0.004

0.002 0.002

0.000 0.000
3000 2960 2920 2880 2840 2800 1800 1780 1760 1740 1720 1700
-1 -1
Wavenumber (cm ) Wavenumber (cm )

G 0.008 H
0.008
0.006
0.006

0.004
0.004

0.002 0.002

0.000 0.000
3000 2960 2920 2880 2840 2800 1800 1780 1760 1740 1720 1700
-1 -1
Wavenumber (cm ) Wavenumber (cm )
I J 0.010
0.008

0.008
0.006
0.006
0.004
0.004
0.002
0.002

0.000 0.000
3000 2960 2920 2880 2840 2800 1800 1780 1760 1740 1720 1700
-1 -1
Wavenumber (cm ) Wawenumber (cm )
1972 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973

Table 1 significantly affect the interaction of OSW-1 with its putative target in
Characteristic vibrational wavenumbers of lipid bands obtained from PM-IRRAS spectra. the cells.
νas (CH2) νs (CH2) ν (C–O) While our results do not allow identifying a precise site in the sapo-
(cm−1) (cm−1) (cm−1) nins studied, which are responsible for the interaction with the mem-
Chol/H2O 2912 2847 1061 brane lipids, it may be proposed that the interaction with cholesterol
Chol/Merck Saponin 2932 2863 1052 results in a higher exposure of the phospholipid carbonyl ester groups
Chol/digitonin 2937 2859 1047 to water and, consequently, a higher hydration of these groups com-
pared to saponin-free systems. This effect may be the driving force for
νas (CH2) νs (CH2) ν (C_O)
(cm−1) (cm−1) (cm−1) the cholesterol-dependent modification of cell membranes by saponins.
To determine the role of cholesterol in the mechanism of digitonin action
DPPC/H2O 2920 2852 1728
DPPC/Merck Saponin 2928 2854 1738
at an atomic level further systematic studies using chemically pure sapo-
DPPC/digitonin 2930 2855 1742 nins are required.
DPPG/H2O 2919 2843 1732
DPPG/Merck Saponin 2922 2851 1733
Acknowledgments
DPPG/digitonin 2920 2851 1737
DPPC/Chol/H2O 2914 2847 1726
DPPC/Chol/Merck Saponin 2922 2865 1735 M. G. acknowledges the financial support from an Interdisciplinary
DPPC/Chol/digitonin 2932 2856 1732 PhD Studies project entitled “Molecular Sciences for Medicine” co-
DPPG/Chol/H2O 2919 2851 1738 financed by the European Social Fund within the Human Capital Opera-
DPPG/Chol/Merck Saponin 2920 2850 1731
DPPG/Chol/digitonin 2921 2850 1730
tional Programme. M. G. thanks the Université de Lorraine for facilitating
her stay at the laboratory of “Structure et Réactivité des Systèmes
Moléculaires Complexes”.
hydroxyl group. Interestingly, a higher affinity for the lipid monolayers
was observed with digitonin having a higher surface activity. According References
to the results obtained, the number of the digitonin molecules incorpo-
rated into the phospholipid films was higher compared to Merck Sapo- [1] G. Francis, Z. Kerem, H.P.S. Makkar, K. Becker, The biological action of saponins in an-
imal systems: a review, Br. J. Nutr. 88 (2002) 587–605.
nin. Digitonin had as well a more pronounced impact on the lipid acyl
[2] S. Bottger, K. Hofmann, M.F. Melzig, Saponins can perturb biologic membranes and
chain reorganization and on the carbonyl group dehydration. This allows reduce the surface tension of aqueous solutions: A correlation? Bioorg. Med. Chem.
us to propose that digitonin penetrates more easily into the lipid film 20 (2012) 2822–2828.
[3] I. Podolak, A. Galanty, D. Sobolewska, Saponins as cytotoxic agents: a review,
and inserts between the lipid acyl chains and the polar heads of the
Phytochem. Rev. 9 (2010) 425–474.
monolayer. It has to be noted that spatial organization of digitonin and [4] J.M. Augustin, V. Kuzina, S.B. Andersen, S. Bak, Molecular activities, biosynthesis and
Merck Saponin is significantly different from the steric point of view evolution of triterpenoid saponins, Phytochemistry 72 (2011) 435–457.
(Fig. 1B, D). Indeed, Merck Saponin bearing seven or eight monosaccha- [5] J.P. Vincken, L. Heng, A. de Groot, H. Gruppen, Saponins, classification and occur-
rence in the plant kingdom, Phytochemistry 68 (2007) 275–297.
ride units is more bulky compared to digitonin, containing a pentasac- [6] R. Riguera, Isolating bioactive compounds from marine organisms, J. Mar.
charide moiety. The bulky, hydrophilic saccharide part in Merck Biotechnol. 5 (1997) 187–193.
Saponin may hinder its insertion in the membrane, while favoring inter- [7] H. Fuchs, D. Bachran, H. Panjideh, N. Schellmann, A. Weng, M.F. Melzig, M.
Sutherland, C. Bachran, Saponins as tool for improved targeted tumor therapies,
action with water. Curr. Drug Targets 10 (2009) 140–151.
Furthermore, our results show that digitonin interacts directly [8] J.K. Hellmann, S. Munter, M. Wink, F. Frischknecht, Synergistic and additive effects of
with cholesterol and disrupts the Chol monolayer. It can be supposed epigallocatechin gallate and digitonin on plasmodium sporozoite survival and mo-
tility, PLoS One 5 (2010).
that the hydrophobic interaction between the rigid rings of digitonin [9] Y. Matsubayashi, L. Iwai, T. Toda, Q.R. Lu, H. Kawasaki, Immunostaining for
and cholesterol, as well as the hydrophilic interaction leading to the oligodendrocyte-specific galactosphingolipids in fixed brain sections using the
formation of the hydrogen bond network between the sugar moie- cholesterol-selective detergent digitonin, J. Neurosci. Methods 178 (2009) 87–98.
[10] S.Y. Eid, M.Z. El-Readi, E. Eldin, S.H. Fatani, M. Wink, Influence of combinations of
ties and cholesterol hydroxyl group are responsible for the monolayer
digitonin with selected phenolics, terpenoids, and alkaloids on the expression and
disruption. On the other hand, Chol/digitonin complex was not formed activity of P-glycoprotein in leukaemia and colon cancer cells, Phytomedicine 21
in the case of the mixed phospholipid/Chol/digitonin systems. It can be (2013) 47–61.
[11] V. Raj, R. Jaime, D. Astruc, K. Sreenivasan, Detection of cholesterol by digitonin con-
concluded, that the interaction between phospholipid and Chol are too
jugated gold nanoparticles, Biosens. Bioelectron. 27 (2011) 197–200.
strong for digitonin to remove Chol from the interface. Our findings sug- [12] B.J. Vermeer, C.M. Vangent, W.C. Debruijn, T. Boonders, Effect of digitonin-
gest that membrane activity of digitonin is cholesterol dependent. In- containing fixatives on retention of free cholesterol and cholesterol esters,
deed, membrane damage or membrane fluidity is induced at high or Histochem. J. 10 (1978) 287–298.
[13] E. Fornas, J. Renaupiqueras, A. Fortea, F. Mayordomo, E. Alborch, Cholesterol and 25-
low digitonin contents, respectively. These results are in accordance hydroxycholesterol retention in specimens of liver and aorta prepared for Electron-
with the literature. Indeed, the lytic activity of digitonin was revealed Microscopy. 2. Effect of filipin, osmium, digitonin and saponin, Lipids 28 (1993)
using cell membranes rich in cholesterol. Moreover, the antiviral activ- 929–935.
[14] C.N. Armah, A.R. Mackie, C. Roy, K. Price, A.E. Osbourn, P. Bowyer, S. Ladha, The
ity of digitonin may be connected with liquefaction of the viral envelope membrane-permeabilizing effect of avenacin A-1 involves the reorganization of bi-
indicating high rigidity due to the presence of cholesterol [48,49]. layer cholesterol, Biophys. J. 76 (1999) 281–290.
It should be mentioned that different biological effects depend on [15] C. Delay, J.A. Gavin, A. Aumelas, P.A. Bonnet, C. Roumestand, Isolation and structure
elucidation of a highly haemolytic saponin from the Merck saponin extract using
defined molecular interactions between particular chemical groups of high-field gradient-enhanced NMR techniques, Carbohydr. Res. 302 (1997) 67–78.
the saponins and their targets [50]. While the latter are not general [16] D. Frechet, B. Christ, B.M. Dusorbier, H. Fischer, M. Vuilhorgne, 4 triterpenoid sapo-
effects but specific for selected saponins with defined chemical proper- nins from dried roots of Gypsophila species, Phytochemistry 30 (1991) 927–931.
[17] R. Atta ur (Ed.), Bioactive Natural Products, Part H, Stud. Nat. Prod. Chem., 27,
ties, pinpointing the moieties involved in the interaction is a challenge.
Elsevier Science B.V., 2002
Recently, a biotinylated derivative of a highly potent anticancer natural [18] A.J. George, Legal status and toxicity of saponins, Food Cosmet. Toxicol. 3 (1965)
saponin OSW-1 was synthesized with the aim to determine its cellular 85–91.
[19] S. Southon, I.T. Johnson, J.M. Gee, K.R. Price, The effect of Gypsophila saponins in the
target(s) [51]. The biotinylated OSW-1 offers the feature of a reversible
diet on mineral status and plasma-cholesterol concentration in the rat, Br. J. Nutr. 59
solid-support reagent when combined with monomeric-avidin sepha- (1988) 49–55.
rose or avidin-coated solid surface. This approach permitted a facilitated [20] K.R. Price, S. Southon, G.R. Fenwick, The effect of saponins on iron and zinc availabil-
evaluation of the anti-proliferative effect of this chemically pure ity, Spec. Publ. R. Soc. Chem. 72 (1989) 155–157.
[21] S. Sakamoto, H. Nakahara, T. Uto, Y. Shoyama, O. Shibata, Investigation of interfacial
molecule in several human cancer cell lines. It was concluded that the behavior of glycyrrhizin with a lipid raft model via a Langmuir monolayer study,
attachment of biotin to the C-3 position of the aglycone does not Biochim. Biophys. Acta Biomembr. 1828 (2013) 1271–1283.
 B. Korchowiec et al. / Biochimica et Biophysica Acta 1848 (2015) 1963–1973
[22] K.J. Stine, R.K. Hercules, J.D. Duff, B.W. Walker, Interaction of the glycoalkaloid
tomatine with DMPC and sterol monolayers studied by surface pressure measure-
ments and Brewster angle microscopy, J. Phys. Chem. B 110 (2006) 22220–22229.
[23] K. Wojciechowski, M. Orczyk, T. Gutberlet, M. Trapp, K. Marcinkowski, T. Kobiela, T.
Geue, Unusual penetration of phospholipid mono- and bilayers by Quillaja bark sa-
ponin biosurfactant, Biochim. Biophys. Acta Biomembr. 1838 (2014) 1931–1940.
[24] C. Stefaniu, G. Brezesinski, H. Mohwald, Langmuir monolayers as models to
study processes at membrane surfaces, Adv. Colloid Interface Sci. 208 (2014)
197–213.
[25] W. Barzyk, S. Campagna, K. Wieclaw, B. Korchowiec, E. Rogalska, The affinity of two
antimicrobial peptides derived from bovine milk proteins for model lipid mem-
branes, Colloids Surf. A 343 (2009) 104–110.
[26] Y. Corvis, B. Korchowiec, G. Brezesinski, S. Follot, E. Rogalska, Impact of aluminum on
the oxidation of lipids and enzymatic lipolysis in monomolecular films at the air/
water interface, Langmuir 23 (2007) 3338–3348.
[27] B. Korchowiec, M. Gorczyca, A.B. Salem, J.-B. Regnouf de Vains, E. Rogalska, Interac-
tion of a β-lactam calixarene derivative with a model eukaryotic membrane affects
the activity of PLA2, Colloids Surf. B 103 (2013) 217–222.
[28] D. Marsh, Lateral pressure in membranes, Biochim. Biophys. Acta Rev. Biomembr.
1286 (1996) 183–223.
[29] www.sigmaaldrich.com/catalog/product/sigma/d141 (in).
[30] J.T. Davies, E.K. Rideal, Interfacial phenomena, 2nd ed. Academic Press, 1963.
[31] P.L. Yeagle, Cholesterol and the cell membrane, Biochim. Biophys. Acta Rev.
Biomembr. 822 (1985) 267–287.
[32] F. Schroeder, J.K. Woodford, J. Kavecansky, W.G. Wood, C. Joiner, Cholesterol do-
mains in biological membranes, Mol. Membr. Biol. 12 (1995) 113–119.
[33] D. Blaudez, T. Buffeteau, J.C. Cornut, B. Desbat, N. Escafre, M. Pezolet, J.M. Turlet,
Polarization-modulated FT-IR spectroscopy of a spread monolayer at the air/water
interface, Appl. Spectrosc. 47 (1993) 869–874.
[34] D. Blaudez, J.-M. Turlet, J. Dufourcq, D. Bard, T. Buffeteau, B. Desbat, Investigations at
the air/water interface using polarization modulation IR spectroscopy, J. Chem. Soc.
Faraday Trans. 92 (1996) 525–530.
[35] L. Dziri, B. Desbat, R.M. Leblanc, Polarization-modulated FT-IR spectroscopy studies
of acetylcholinesterase secondary structure at the air–water interface, J. Am.
Chem. Soc. 121 (1999) 9618–9625.
[36] X. Bi, C.R. Flach, J. Perez-Gil, I. Plasencia, D. Andreu, E. Oliveira, R. Mendelsohn, Sec-
ondary structure and lipid interactions of the N-terminal segment of pulmonary
surfactant SP-C in Langmuir films: IR reflection–absorption spectroscopy and sur-
face pressure studies, Biochemistry 41 (2002) 8385–8395.
[37] M. Dyck, A. Kerth, A. Blume, M. Loesche, Interaction of the neurotransmitter, neuro-
peptide Y, with phospholipid membranes: infrared spectroscopic characterization at
the air/water interface, J. Phys. Chem. B 110 (2006) 22152–22159.
[38] Z. Zhang, A.L. Verma, M. Yoneyama, K. Nakashima, K. Iriyama, Y. Ozaki, Molecular ori-
entation and aggregation in Langmuir–Blodgett films of 5-(4-n-octadecylpyridyl)-
10,15,20-tri-p-tolylporphyrin studied by ultraviolet–visible and infrared spectros-
copies, Langmuir 13 (1997) 4422–4427.
[39] J.M. Vanderkooi, J.L. Dashnau, B. Zelent, Temperature excursion infrared (TEIR) spec-
troscopy used to study hydrogen bonding between water and biomolecules,
Biochim. Biophys. Acta Protein Proteomics 1749 (2005) 214–233.
1973
[40] R.N.A.H. Lewis, R.N. McElhaney, W. Pohle, H.H. Mantsch, Components of the carbon-
yl stretching band in the infrared spectra of hydrated 1,2-diacylglycerolipid bilayers:
a reevaluation, Biophys. J. 67 (1994) 2367–2375.
[41] B. Korchowiec, J. Korchowiec, M. Hato, E. Rogalska, Glycolipid-cholesterol mono-
layers: towards a better understanding of the interaction between the membrane
components, Biochim. Biophys. Acta Biomembr. 1808 (2011) 2466–2476.
[42] K. Czapla, B. Korchowiec, M. Orlof, J.R. Magnieto, E. Rogalska, Enzymatic probing of
model lipid membranes: phospholipase A2 activity toward monolayers modified
by oxicam NSAIDs, J. Phys. Chem. B 115 (2011) 9290–9298.
[43] J. Rubio-Magnieto, S.V. Luis, M. Orlof, B. Korchowiec, G. Sautrey, E. Rogalska, Effects
of gemini amphiphilic pseudopeptides on model lipid membranes: a Langmuir
monolayer study, Colloids Surf. B 102 (2013) 659–666.
[44] G. Sautrey, M. Orlof, B. Korchowiec, J.B.R. de Vains, E. Rogalska, Membrane activity of
tetra-p-guanidinoethylcalix[4]arene as a possible reason for its antibacterial proper-
ties, J. Phys. Chem. B 115 (2011) 15002–15012.
[45] O. Reis, R. Winter, T.W. Zerda, The effect of high external pressure on DPPC-
cholesterol multilamellar vesicles: a pressure-tuning Fourier transform infrared
spectroscopy study, Biochim. Biophys. Acta Biomembr. 1279 (1996) 5–16.
[46] D.A. Mannock, M.Y.T. Lee, R.N.A.H. Lewis, R.N. McElhaney, Comparative calorimetric
and spectroscopic studies of the effects of cholesterol and epicholesterol on the
thermotropic phase behaviour of dipalmitoylphosphatidylcholine bilayer mem-
branes, Biochim. Biophys. Acta Biomembr. 1778 (2008) 2191–2202.
[47] N.V. Venkatatraman, S. Vasudevan, Cholesterol binding to the alkyl chains of an in-
tercalated surfactant bilayer, J. Phys. Chem. B 107 (2003) 10119–10126.
[48] S. Harada, K. Yusa, K. Monde, T. Akaike, Y. Maeda, Influence of membrane fluidity on
human immunodeficiency virus type 1 entry, Biochem. Biophys. Res. Commun. 329
(2005) 480–486.
[49] S. Harada, The broad anti-viral agent glycyrrhizin directly modulates the fluidity of
plasma membrane and HIV-1 envelope, Biochem. J. 392 (2005) 191–199.
[50] V.R. Netala, S.B. Ghosh, P. Bobbu, D. Anitha, V. Tartte, Triterpenoid saponins: a re-
view on biosynthesis, applications and mechanism of their action, Int. J. Pharm.
Pharm. Sci. 7 (2015) 24–28.
[51] Y. Kang, C. Lou, K.B.R. Ahmed, P. Huang, Z. Jin, Synthesis of biotinylated OSW-1,
Bioorg. Med. Chem. Lett. 19 (2009) 5166–5168.
[52] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark,
J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi,
N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli,
J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,
J.J.G. Dannenber, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J.
Cioslowski, D.J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford CT, 2013.