Biochimica et Biophysica Acta 1808 (2011) 2706–2713
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a m e m
Molecular organization of antibiotic amphotericin B in
dipalmitoylphosphatidylcholine monolayers induced by K + and Na + ions:
The Langmuir technique study
Marta Arczewska a, Mariusz Gagoś a, b,⁎
a
b
Department of Biophysics, University of Life Sciences in Lublin, 20–950 Lublin, Poland
Department of Cell Biology, Institute of Biology, Maria Curie-Skłodowska University, 20-033 Lublin, Poland
a r t i c l e
i n f o
Article history:
Received 4 May 2011
Received in revised form 18 July 2011
Accepted 19 July 2011
Available online 26 July 2011
Keywords:
Amphotericin B
Molecular organization
Langmuir monolayer
Na+ ion
K+ ion
a b s t r a c t
The effect of potassium (K +) and sodium (Na+) ions on the self-association of antibiotic amphotericin B
(AmB) in the lipid membrane was reported. Mixed Langmuir monolayers of AmB and dipalmitoylphosphatidylcholine (DPPC) were investigated by recording surface pressure–area isotherms spread on aqueous
buffers containing physiological concentration of K + and Na+ ions. The analyses of the π–A isotherms and
compressional modulus curves indicate the interactions in the AmB–DPPC system. The strength of the AmB–
DPPC interactions and the stability of the mixed monolayers were examined on the basis of the excess free
energy of mixing values. The obtained results proved a high affinity of AmB towards lipids induced by the
presence of K+ than Na + ions. The most stable monolayers in the presence of K+ and Na + ions were formed
by AmB and DPPC with the 1:1 and 2:1 stoichiometry. The understanding of the AmB aggregation processes at
the molecular level should contribute to elucidate the mechanisms of action and toxicity of this widely used
drug. The presented results are potentially valuable in respect to develop more efficient and less toxic AmB
formulations.
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
The amphiphilic polyene antibiotic amphotericin B (AmB) is
currently the drug of choice in the treatment of severe fungal infections despite its undesirable side effects. The toxicity of the drug
has been related to its low solubility due to the formulation of selfassociated species called molecular aggregates. According to the
general conviction, the biological action of the drug is based on the
formation of membrane pores that considerably affect physiological
ion transport, especially K + ions [1–3]. On the basis of commonly
accepted hypothesis — the disturbance of membrane barrier function is made by creating transmembrane pores or channels [4,5]. The
molecules of AmB adopt a quasi parallel orientation with their polar
sides (the polyhydroxyl groups) facing the interior of such formed
pores and the hydrophobic parts interacting with the lipids of the
cellular membrane [6,7]. Further this model was completed in many
experimental studies [8,9] both in theoretical studies on chemical
structure and molecular properties of these channels [10,11]. It was
postulated that the possibilities of chemotherapeutical activity of AmB
are the result of interaction with biological membrane and it is based
on higher affinity of the antibiotic towards ergosterol-containing
⁎ Corresponding author at: Department of Biophysics, University of Life Sciences in
Lublin, 20–950 Lublin, Poland. Tel.: + 48 81 4456899; fax: + 48 81 4456684.
E-mail address: [email protected] (M. Gagoś).
0005-2736/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbamem.2011.07.027
fungal cells compared to cholesterol-containing mammalian cells
[12–16]. The application of atomic force microscopy (AFM) studies
allowed illustrating the porous structures formed from AmB in a
monolayer containing 90 mol% of antibiotic and 10 mol% of DPPC. The
size of the pore was estimated at ~ 17 Å while the internal diameter
came out to ~ 6 Å [17].
Although there are numerous studies of interactions between
AmB and model phospholipid membranes, the complete interpretation of these mechanisms is still unknown. AmB might be entering
into bilayer membrane system which is dependent on its concentration and self-association but it is also possible in sterol-free membranes [18,19]. However, sterols seem to be necessary to stabilize
the channels [20,21]. AmB may strongly interact with phospholipid
molecules to form a stoichiometric complex 1:1. It has been postulated that there are interactions between the conjugated chain of
antibiotic and the methylene groups of lipid acyl chains, while the
sugar moiety interacts with the phosphate head group by the formation of a hydrogen bond [19,21]. The channel activity of the antibiotic
is caused not only by the presence of sterols but it is also connected
with the interaction of other membranes components. The possibility
of the interactions of AmB with the surface of phospholipid bilayer is
dependent on the properties of lipid polar heads, the length and the
degree of the saturation of PC acyl chains, as well [22,23].
Analyzing of the pressure–area isotherm of the mixed monolayers,
Minones et al. [23] reported the existence of stronger interactions
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
between AmB and the saturated phospholipid as compared to unsaturated one. The nature of these interactions revealed the influence of
apolar parts of the phospholipid chains on the stoichiometry of AmB–
lipid complexes [23,24]. The greatest interactions in the AmB–DPPC
system were found to occur for a 2:1 mixture and a stable complex was composed of two horizontally oriented AmB molecules and
one DPPC molecule in a vertical position [24]. The occurrence of
stronger interaction between AmB and phospholipids than between
sterols and membranes plays a key role in the biological activity of
the compound. It leads to diminish the effective concentration of
the antibiotics molecules in membranes which might interact with
membrane sterols [22,25].
Overall, our previous works presented that the K + and Na + ions
can affect the molecular organization of AmB in substantially different
ways. The obtained results with the application of different spectroscopic techniques, such as electronic absorption spectroscopy, Raman
and FTIR, indicated the influence of K + ions on the aggregation levels
of AmB molecules [26,27]. It was also proposed that AmB may have a
direct influence on the ATP–proton pump, in the case of the cells of
fungi, or it may inhibit the ATP (Na +–K +) activity in the animal cells
[28]. On the other hand, it was observed that the enrichment of
fungal cell cultures with K + cations exhibits protective properties
against the toxic activity of AmB [29]. The latest reports indicated that
high sodium intake (N 4 mEq/kg) per day might be associated with
lower nephrotoxicity in extremely premature infants treated with
AmB [30].
In this work we present results regarding the significant interaction in mixed monolayers between AmB and DPPC formed at the
air–water solution interface in the presence of the K + and Na + ions.
2. Experimental section
2.1. Materials and methods
Amphotericin B in crystalline form and dipalmitoylphosphatidylcholine (DPPC) were purchased from Sigma-Aldrich (Poland). The
AmB was dissolved in deionized (mQ) water, alkalized to pH 12 with
KOH or NaOH and then centrifuged for 15 min at 15,000 ×g in order to
remove micro-crystals of the drug still remaining in the sample. The
phospholipid has been dissolved in chloroform/methanol 9:1 v/v
mixture. AmB was further purified by means of HPLC on YMC C-30
(Europe GmbH, Germany) coated phase reversed column (length
250 mm, internal diameter 4.6 mm) with 40% 2-propanol in H2O as
a mobile phase. The final concentration of AmB was calculated from
the absorption spectra on the basis of the molar extinction coefficient
(0–0 absorption maximum at 408 nm) ε408 = 1.08 × 10 5 M − 1 cm − 1
and ε408 = 1 × 10 5 M − 1 cm − 1 in the cases of the sample dissolved
in alkalized water with KOH and NaOH, respectively. Hepes sodium
and potassium salt were purchased from Sigma Chem. Co. (analytical
grade). A buffer solution (subphase) of potassium and sodium
Hepes (10 mM) was adjusted to pH = 7.4 with 1 M HCl. The subphase
for each experiment was filtered through a set of membrane filter
(Millipore Express® Plus, 0.22 μm).
Monolayers were carried out with a Minitrough 2 (KSV Instruments
Ltd., Finland) placed in a laminar hood purged with N2 (relative
humidity of 80%). Monomolecular layers were formed in a Teflon trough
(282 mm × 75 mm) equipped with two symmetric, hydrophilic barriers
and a Wilhelmy platinum plate as surface pressure tensor. Monomolecular layers of mixture were formed at the air–water (buffer)
interface. After each measurement, the surface was precisely cleaned.
Spreading solutions were deposited onto 10 mM sodium and potassium
Hepes separately (at first AmB and then DPPC solution, respectively)
with the Hamilton microsyringe (precise to± 0.1 μl). The solvent was
allowed to evaporate for 15 min, and the monolayers were compressed
with the barrier speed of 75 cm2/min. Each of the π–A presented in this
work represents the average of three independent experiments. The
2707
subphase temperature (24 ± 1 °C) was controlled by a Poly Science
thermociculator through.
2.2. Data analysis
To better characterize the influence of lipid on the physical state of
antibiotic monolayer the compression modulus CS− 1 for the mixed
films has been investigated, defined according to (Eq. (1)) [31,32]
1
Cs ¼ Aðdπ=dAÞ
ð1Þ
where A is the area per molecule at the indicated surface pressure π.
The compression modulus is obtained by numerical calculation of the
first derivative from the π–A isotherm datapoints and plotted as a
function of surface pressure [33]. A characteristic minimum on the
CS− 1 = f(π) graph is used to identify the phase transition. The value of
surface pressure at CS− 1 = 0 indicates the occurrence of collapse πcoll.
The interactions between compounds in the mixed monolayers and
their miscibility have been also analyzed for their mutual miscibility
in accordance with the additivity rule [34].
Based on the mean molecular surface (A12) as a function of (X1,2) it
was possible to determine whether two components are immiscible
or ideally miscible, because deviations from the ideal mixture show
the nonlinearity dependence of A12 = f (X1,2) [22,34–36]. For the ideal
mixing, the mean area per molecule, A12 is defined according to
(Eq. (2)) [35]:
A12 ¼ A1 X1 þ A2 X2
ð2Þ
where X1, X2 are the molar fractions of components 1 and 2 in the
mixed monolayer, and A1, A2 are the molecular areas of single components at the same surface pressure.
The excess of free energy of mixing ΔG EXC is used as an indication
of intermolecular interactions and the stability of the mixed films. The
value of ΔG EXC has been calculated as the compression work difference between ideal and real monolayer mixtures directly from the π–
A isotherms using Eq. (3) as follows:
π
ΔGEXC = N ∫ ðA12 −X1 A1 −X2 A2 Þdπ
ð3Þ
0
where: N is Avogadro's number [34,35,37]. The negative sign of ΔG EXC
is considered as a criterion of monolayer's stability while a positive
value may suggest the phase separation in the monolayer [35].
3. Results and discussion
Due to amphiphilic properties of AmB, it creates monomolecular
layers at the air–water interface. Using the Langmuir monomolecular
technique makes it possible to carry out detailed analysis of molecular
organizations of AmB both in a monolayer model system and aqueous
solution. Fig. 1 presents the surface pressure–area (π–A) isotherms for
pure AmB monolayer spread on the 10 mM buffer subphase (pH 7.4,
potassium and sodium Hepes according to the type of experiments,
respectively). The shape of the AmB compression isotherm is very
close to the isotherms previously reported [17,38–43]. It exhibits a
typical plateau region in the range of molecular areas from 100 Å 2 to
40 Å 2, which is attributed to the reorientation of the AmB molecules
from a horizontal to a vertical position [38,39,43,44]. The extrapolation of the linear parts of the isotherms in the regions corresponding
to the expanded and condensed states of AmB monolayers allows to
know the values of the limiting area occupied by the molecules in
its horizontal (Ae) and vertical (Ac) orientation with relation to the
surface of the water. All these parameters together with the surface
pressures both at the beginning and at the end of the transition region
are shown in Table 1. The surface pressures of the plateaus in Fig. 1 are
2708
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
70
HO
CH 3
H3 C
O
60
AmB +
K+
Na+
CH3
O
B
OH
OH
50
HO
OH
π [mN/m]
OH
40
HO
O
O
O
HO
CH 3
HO
COO
-
NH3
+
OH
air
water
CH3
30
O
O
OH
H3C
NH3 +
OH
A
20
COO
CH3
HO
O
H3 C
OH
OH
OH
OH
O
O
OH
OH
OH
-
air
water
10
c
e
Ac
0
0
20
Ae
40
60
80
100
120
140
160
180
200
Area [Å2/molecule]
Fig. 1. The surface pressure–area isotherms of AmB spread on sodium buffer subphase (solid line) and potassium buffer subphase (dashed line). The linear fits to the linear portions
of the isotherms of compression extrapolated to zero surface pressure point to the specific molecular areas in a horizontal position (Ae) and in a vertical position (Ac). The inset
presents a model of reorientation of AmB molecules on the subphase. Temperature 24 °C.
in the range of 10–12 mN/m. The molecular reorientation from the
horizontal to vertical position in case of AmB is endothermic process
which requires a high energy [39]. In the horizontal orientation AmB
molecules are anchored to the interface by the hydroxylic groups.
During compression the hydrogen bonds are significantly reduced
and the molecules are oriented vertical (anchored by the polar
heads, the inset of Fig. 1B). The results obtained by Brewster angle
microscopy show that during monolayer compression their domainlike structure is transformed into a homogeneous image as a result of
the reorientation of the molecules [24]. Minones et al. [42] have
observed a 3-fold increase in the relative film thickness during compression the monolayer from expanded to a condensed state and
the ratio of the cross-sectional area of AmB in this two orientations
was also 3. On the other hand, Sykora et al. [40] have claimed that
the phenomenon which is responsible for occurring plateau region
might be also the aggregation process. When the monolayer is compressed the molecules of AmB are anchoring to the interface by the
polar head and such setting gives the possibility of chromophores
interaction and creating the H-aggregated forms (card pack) [45]. The
Table 1
The estimated value of limiting molecular areas at the beginning and the end of plateau.
The average SD ± of six experiments.
AmB + K+a
AmB + Na+b
AmBc
Ae
πe
Ac
πc
[Å2/molecule]
[mN/m]
[Å2/molecule]
[mN/m]
157.0 ± 4
148.0 ± 2
148.0
10.2 ± 0.4
10.3 ± 0.2
10
31.2 ± 0.6
27.0 ± 1.2
36.0
11.4 ± 0.7
11.2 ± 0.1
12
a
Amphotericin B before deposition onto buffer subphase (10 mM potassium Hepes,
pH 7.4) was dissolved at pH 12.6 with 1 M KOH.
b
Amphotericin B before deposition onto buffer subphase (10 mM sodium Hepes,
pH 7.4) was dissolved at pH 12.6 with 1 M NaOH.
c
Reported date, AmB before deposition onto water subphase was dissolved in
2–propanol/water (v:v, 4:6) [17,46].
limiting molecular areas obtained from fitting the isotherms in
extremely packed monolayers are as follows: 27.0 ± 1.2 Å 2/molecule
and 31.2 ± 0.6 Å 2/molecule for AmB in the presence of Na + and K +
ions, respectively (see Table 1). The values are comparable to previously published data although the slight differences are the result
of using different solvents as well as the rate of compression [39,40].
The most remarkable differences were observed in the horizontal position. In the presence of K + ions the limiting molecular area
(157 ± 4 Å 2/molecule) is bigger than in the Na + ions' environment
(148 ± 2 Å 2/molecule). Taking into account these values it can be
concluded that the AmB molecules create the aggregated forms on
the surface whose size is bigger in the presence of K + ions. The
orientation of AmB in the monolayer is connected with the interaction of the –COO − and –NH3+ groups as well as –OH groups in the
polyhydroxyl part of molecule [43]. The ionized –COO − group is
subjected to greater exposure to the aqueous phase than –NH3+ due to
stronger its hydration, which orients the macrolide ring closer to
the polar environment. With the increase in surface pressure it is also
possible to change the distance between these groups both in the
horizontal orientation and the vertical one [43]. Due to the changes in
the interactions between the K + and Na + ions and the –COO − group,
the molecules of AmB might interact differently with the surface. This
can result in different areas occupied by single molecules of AmB
in the presence of these ions. In the presence of K + or Na + ions, the
–COO − group may generate specific ionic interactions as a result of
weak ion pairs that was presented in our previous works [27,46].
The results of the experimental research relative to the molecular
dynamics indicated that the strength of ion pairing with the –COO −
group decreases in the sequence Na + N K +[47,48]. On the other
hand, we suggested a higher preference of K + over Na + to an
anionic site the –COO − group on the assumption that the binding
force comes mainly from electrostatic interaction and it is related to
the difference in the size of these two cations. In such a case, the ion
of smallest hydration radius (which corresponds to the greatest
radius of the non-hydrated ion) is able to attract the negative
attachment site so that the binding of K + will be stronger [46].
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
2709
60
DPPC
XAmB = 0.1
XAmB = 0.3
XAmB = 0.5
50
A
XAmB = 0.7
XAmB = 0.9
AmB
+ K+
B
+ K+
40
30
20
[mN/m]
10
0
DPPC
XAmB = 0.1
XAmB = 0.3
XAmB = 0.5
50
40
C
D
XAmB = 0.7
XAmB = 0.9
AmB
+ Na+
+ Na+
30
20
10
0
0
25
50
75
100
125
150
175
Area
0
25
50
75
100
125
150
175
200
[Å2/molecule]
Fig. 2. The surface pressure–area isotherms of mixed monolayers composed of AmB and DPPC formed at 10 mM potassium buffer subphase (A, B) and 10 mM sodium buffer
subphase (C, D). Temperature 24 °C.
Fig. 2 shows the π–A isotherms for the pure phospholipid, pure
AmB and the mixtures of AmB–DPPC which contain AmB in the molar
concentration from 10% to 90% (these values are equal with molar
fraction XAmB = 0.1–0.9). These monolayers were spread on potassium subphase (10 mM Hepes buffer solution with pH 7.4, Fig. 2A and
B) and sodium subphase (10 mM Hepes buffer solution with pH 7.4,
Fig. 2C and D). The formed monolayer of pure DPPC exhibits at the
surface pressure of 6 mN/m, a characteristic LE–LC phase transition evidenced as a plateau [38,39,44]. When the AmB is incorporated
into the lipid monolayer, the phase transition is gradually shifted in
the direction of the higher surface pressure (ca. 11 mN/m). When
the amount of AmB is higher (XAmB N 0.5), the biggest changes in the
plateau shape concern the monolayer compressed in the presence of
Na + ions. In the K + ions' environment they occur in similar range to
pure compound.
For a more detailed analysis of the physical state of the monolayer
and to know the influence of AmB on the condensation of model lipid
membranes, the values of compression modulus CS− 1 as a function of
surface pressure have been calculated (Eq. (1), see Fig. 3). According
to Davis et al. [31] these values were typical of monolayers in liquid
state. Two minima are observed in CS− 1 curves. The first minimum
(M0) shown in Fig. 3A and C for pure DPPC (curves 1) refers to the
above mentioned transition from liquid expanded phase to liquid
condensed phase. During the incorporation of AmB into lipid monolayer this minimum gradually disappears and a new one appears (M1)
at higher surface pressures close to 11 mN/m. A fairly distinguished
discontinuity at 16–20 mN/m is also observed for the mixtures of
XAmB = 0.5–0.7 both in the presence of K + and Na + ions. However,
this minimum in the Na + ions' environment is reduced to a small
inflection. It might be due to the reorientation of AmB molecules at
high surface pressures or its partial dissolution into the aqueous
subphase [24]. For mixtures of XAmB = 0.9 as well as the singlecomponent monolayers of pure AmB, three regions can be noticed
(Fig. 3B, curves 2 and 3). The first region corresponds to the
monolayers in the liquid expanded phase, the second, is connected
with the liquid condensed phase characterized by high value of Cs− 1.
The third region is represented by a wide plateau and is connected
with the LE–LC transition at the range of 10–11 mN/m [17,24]. For the
mixed monolayers spread on buffer containing Na + ions, the regions
mentioned above were not distinguished. The monolayer of pure
AmB exhibits also a small inflection in the range of 44–50 mN/m. The
presence of this point at such high surface pressure may be connected
with the disorganization of the molecular structures in a form of pores
[49]. AFM microscopy revealed the formation of cylindrical pore-like
structures in the topography of the two-component monolayers
composed of 90 mol% AmB and 10 mol% DPPC [17]. The structures
formed during compression become disordered upon further compression of the monolayer. It was suggested by Gagoś et al. [49] that
the symmetrically ordered structures of AmB are not stable at the
surface pressures above 20 mN/m and other membrane cell components (e.g. sterols) could play a role in the stabilization of molecular
pores. Finally, it may be stated that the incorporation of AmB into
the DPPC monolayers causes film expansion and the disappearance of
the phase transition for pure phospholipid [24]. It is probably
2710
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
150
1 DPPC
2 XAmB= 0.1
3 XAmB = 0.3
4 XAmB = 0.5
125
1 XAmB= 0.7
2 XAmB= 0.9
3 XAmB= 1
A
1
2
K+
70
K+
60
1
3
100
B
50
3
75
4
40
2
30
50
20
Cs-1[mN/m]
25
10
M0
0
125
M1
M1
1 DPPC
2 XAmB= 0.1
3 XAmB= 0.3
4 XAmB= 0.5
1 XAmB= 0.7
2 XAmB= 0.9
3 XAmB= 1
C
2
Na+
1
3
0
70
D
Na+
1
60
100
50
3
2
40
75
4
30
50
20
25
10
0
0
0
10
20
30
40
50
0
10
20
30
40
50
60
70
[mN/m]
Fig. 3. The compression modulus CS− 1 as a function of surface pressure for mixtures of AmB and DPPC obtained on the basis of the π–A isotherms presented in Fig. 2. The mixed
monolayers composed of AmB and DPPC formed at 10 mM potassium buffer subphase (A, B) and 10 mM sodium buffer subphase (C, D).
associated with an ordering effect of DPPC with respect to AmB
[45,50,51].
In Table 2, the surface pressures of phase transitions and the values
of the compression modules Cs− 1 obtained for the π–A isotherms,
shown in Fig. 2, were compiled. The increasing amounts of AmB in the
mixed monolayers cause the decrease in the value of Cs− 1 compared to
pure DPPC that means a lower rigidity of obtained monolayers. This
decrease is especially noticed for the mixed films at the range of
XAmB = 0.1–0.7. Interestingly, that high molar fractions of AmB
(90 mol%) decrease the value of Cs− 1compared to pure antibiotic.
The mentioned effect was distinctly noticed in the presence of the
K + ions.
Table 2
The surface pressures for the phase transitions (M0, M1) and the maximum value of the
compressional modules of Cs− 1max for two-component monolayers of AmB–DPPC at
different molar fractions spread on buffer subphase (10 mM Hepes, pH 7.4) containing
ions K+ and Na+ (data obtained from Figs. 2 and 3).
AmB (XAmB)
M0
+K
0 (DPPC)
0.1
0.3
0.5
0.7
0.9
1 (AmB)
6.4
9.3
7.3
Cs− 1max
M1
+
+ Na
6.4
9.5
8.3
+
+K
11.5
11.3
10.4
10.5
10.4
10.2
+
+ Na
10.7
11.0
10.8
11.0
10.6
10.2
+
+ K+
+ Na+
145.0
132.0
110.0
76.0
59.5
36.6
44.4
145.0
138.0
100.0
87.0
67.6
53.6
68.7
However, the mixed monolayers show the phase transition at higher
surface pressure than the phase transition of pure DPPC (~6 mN/m) and
they have got similar values like pure AmB (~11 mN/m). This behavior
indicates the existence of molecular interactions between the components and their miscibility at the air–water interface. It is also ascribed to
the molecular reorientation of antibiotic in the mixed films but this
process may be limited by the presence of phospholipid. The molecules
of AmB interact with the polar headgroups of DPPC and they may be
organized in this region [16,43]. In this case, applying the Crisp phase
rule [37] –F = C − P + 1 (where F is the number of degree of freedom
of the system, C is the number of components and P describes the
number of phases involved) to the transition region is helpful to
estimate the number of surface phases in equilibrium and the miscibility
or immiscibility of mixture components. According to this, there are
two different behaviors depending on the composition and surface
pressure of the mixed monolayers which are independent on the
presence of K + and Na + ions, see Fig. 4. In the range of composition
XAmB = 0–0.5 and at surface pressures b 10 mN/m the components are
miscible. Since the mixed monolayers consisting of two film-forming
component, C = 2, and the fact that the surface pressure corresponding
to the LE-LC phase transition varies with the composition of the mixed
system, F = 1 (see Fig. 4 and Table 2). Therefore, there are two surface
phases in equilibrium. The phase P1 (below the surface pressure corresponding the transition region) composed of AmB–DPPC complexes
with 1:1 stoichiometry, where both components are horizontally
oriented on the subphase, and by molecules of DPPC in excess, with
their hydrocarbon chains tilted towards the water surface. The phase P2
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
2711
160
24
+ K+
K+
Na+
5 mN/m
10 mN/m
20 mN/m
140
20
A
[mN/m]
120
16
100
M1
12
80
8
M0
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
XAmB
Fig. 4. The values of compression modulus CS− 1 vs. the molar fraction of AmB (XAmB)
for mixed monolayers of AmB and DPPC at different surface pressures: 12 mN/m
(A), 23 mN/m (B), 30 mN/m (C), and 45 mN/m (D). The mixed films spread on
potassium buffer subphase (●) and sodium buffer subphase (○). Plot of phase
transition pressures as a function of the composition of AmB–DPPC mixed monolayers.
M0 corresponds to the LE–LC phase transition for pure DPPC and M1 denotes the phase
transition of mixed monolayers.
A12 [Å2/molecule]
60
4
40
20
+ Na+
B
140
120
100
80
(above the surface pressure corresponding to that transition), formed
by the 1:1 “complex” and by DPPC molecules with their hydrocarbon
chains more or less vertically oriented on the water as a result of
the horizontal–vertical orientation change of molecules along the LE–
LC phase transition. In the range of composition XAmB = 0.5–1 was
observed only the minimum M1, where Cs− 1 and the surface pressure
values remain constant with the composition of the mixtures (see line
M1, Fig. 4), which is characteristic of mixed systems with immiscible
components. The application of the Crisp phase rule to the phase
transition of the AmB (M1) proves the existence of three phases in
equilibrium in this situation. Indeed, C = 2 and F = 0 because the surface pressure corresponding to this minimum does not vary with the
film-forming composition. So, in this case P = 3. The first one, P1, was
described above (at the surface pressure corresponding to the AmB
phase transition). There are also two other separate phases P3 and P4
which are formed by the segregated components of the complex, that is,
by phospholipid molecules, vertically oriented, and by AmB, also
vertically oriented.
The plots of the mean molecular area (A12) as a function of the
molar fractions XAmB of AmB (Eq. (2)) are presented in Fig. 5. At
surface pressures 5 and 10 mN/m, the deviations from the ideality are
positive for mixtures of XAmB = 0.1 and 0.7 in the presence of K + ions
(Fig. 5A). However, in the presence of Na + (Fig. 5B) the positive
deviations from the ideal behavior were observed when XAmB = 0.3. In
contrast, at high surface pressures (20 mN/m), the obtained results of
mean molecular area almost coincide with the theoretical value
calculated on the basis of the additivity rule (shown as dashed lines).
This fact shows that the components of the mixed films are only
partially miscible or immiscible. However, these values are uncertain
due to the increased likelihood of instability of the AmB monolayers
[24,44,52]. On the other hand, as can be seen from Fig. 5A, in the
presence of K + ions slight negative deviations from the additivity
rule were observed in the range of XAMB = 0.1–0.7 at higher surface
pressures (20 mN/m).
The results of the calculations based on Eq. (3), in the form of
ΔG EXC = f(XAmB) dependencies for two-component monolayers were
presented in Fig. 6. As can be seen, at lower surface pressures (5 and
10 mN/m), ΔG EXC is positive in the whole range of mixture compositions in the presence of K + ions in subphase (Fig. 5A). A different
tendency can be observed at higher surface pressures (20 mN/m)
60
40
20
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
XAmB
Fig. 5. Mean molecular area (A12) as a function of the molar fraction of AmB (XAmB) for
the mixed monolayers of AmB and DPPC. The mixed films spread on potassium buffer
subphase (A) and sodium buffer subphase (B). The dashed line indicates the direction
of the changes characteristic of an ideal system.
where the values of ΔG EXC were negative in the whole range of XAmB,
showing a minimum value when XAmB = 0.5 at ca. −0.545 kJ/mol
(Fig. 6A). The presence of this minimum proves that the most stable
mixed monolayers of AmB–DPPC were formed at 1:1 stoichiometry
and suggests strong interactions between molecules. Similar results
regarding the interactions between AmB and phospholipid were
also found by Balakrishnan et al. [19,25]. The compound might be
involved in the complex formation with DPPC via the intermolecular
hydrogen bonds with head groups of lipid and it is localized mostly in
this polar region [16]. The strength of the interaction between the
molecules of AmB and DPPC in the presence of K + ions is further
confirmed by the fact that in similar studies conducted by Hąc-Wydro
et al. [25] but for the mixed monolayer spread only on water
subphase, the value of ΔG EXC was higher and reached ca. −0.7 kJ/mol.
Analyzing the mixed monolayers spread on buffer subphase
containing Na + ions, the different tendency of ΔG EXC vs. XAmB dependencies was found (see Fig. 6B). In the monolayers of medium AmB
content (XAmB = 0.3–0.5), the excess free energy of mixing was
positive in the whole range of surface pressures. Stronger interactions associated with the appearance of minima in the analyzed
dependencies related to low content of AmB (XAmB = 0.1) and at
above 50 mol.% of antibiotic in the mixture. The most negative value
of ΔG EXC for a wide minimum is achieved at XAmB = 0.7 (ΔG EXC ca.
− 0.309 kJ/mol), this suggests that stronger attractions and the
stability of AmB–DPPC complex can be approximated to a 2:1
2712
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
1200
+ K+
5 mN/m
10 mN/m
20 mN/m
and Na + (Fig. 6B) reveals higher stability for the monolayers spread
on K +-containing subphase. The differences in the interactions of
AmB with phospholipid in the mixed monolayers spread on different
kinds of subphase under the same experimental conditions could
be also connected with the influence of aforementioned ions on
the formation of aggregated structures. Our previous findings which
carried out the application of spectroscopic methods have indicated
that monovalent ions can affect the molecular organization of AmB
in substantially different ways, and that the K + ions exhibited
stronger ionic binding affinity to the –COO − group of AmB relative to
Na +[26,27]. A considerably stronger effect of K + ions on the aggregation level of AmB is mainly related to the difference in the size of
these two cations [26]. Although the monolayer experiments for twocomponent system composed of AmB and DPPC have been already
reported [24,25], in none of these studies AmB was dissolved at pH 12
with KOH or NaOH. In the context of this, it may be proposed that the
spreading solvent as well as different subphase affecting the limiting
molecular areas determines the surface occupied by the single molecule of AmB and the stoichiometry of the antibiotic–phospholipid
complexes. Differences in the strength of interactions with DPPC exist
in the whole range of AmB concentrations and they are much stronger
in the presence of K + ions.
A
800
400
GEXC. [J/mol]
0
-400
600
+ Na+
B
4. Conclusions
400
The differences in the interactions between AmB and DPPC in the
presence of K + and Na + ions may result from the existence of different influences of these ions on the molecular organization of drug.
These effects can be associated with differences in the sizes of these
ions. The strength of these interactions is significantly weaker for the
mixed monolayers spread on Na +-containing subphase. The presence
of K + ions can be an important element in facilitating the interaction of molecules with the model lipid membranes and enhance the
efficiency of transmembrane transport of these ions without interaction with sterols. In light of this discussion, it seems likely that K + ions
bind more efficiently with AmB than Na + ions.
200
0
-200
Acknowledgements
-400
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
XAmB
Fig. 6. The excess of free energy of mixing (ΔG EXC) as a function of the molar fraction
of AmB (XAmB) for the mixed monolayers of AmB and DPPC at different surface
pressures. The mixed films spread on potassium buffer subphase (A) and sodium
buffer subphase (B).
stoichiometry. The presence of such a complex has been previously
proposed by a number of authors [24,41,53].
In summary, at surface pressures above the plateau in the case of
AmB–DPPC mixtures in the presence of K + ions,strong interactions
which provoke the existence of negative deviations from the theoretical value, calculated on the basis of the additivity rule, were
observed. On the other hand, experimental results at low surface
pressures (5–10 mN/m) indicate positive deviation from the ideality.
Saint-Pierre-Chazalet et al. [38] obtained similar results for a mixture
of AmB–cholesterol, for which the observed effects were also associated with the interaction between two components and prevented
the molecules of AmB desorption from the surface. There are a number of experimental researches that promote this kind of interpretation [44,52]. The strong interactions between the molecules of
phospholipid and antibiotic cause an increase in the average specific
molecular areas compared to those of AmB molecules which would
be occupied in the case of an ideal system. The comparison of ΔG EXC
vs. XAmB dependencies for mixtures in the presence of K + (Fig. 6A)
This article is dedicated to Prof. Jan Sielewiesiuk from the
Department of Biophysics, Maria Curie-Skłodowska University in Lublin.
This research was financed by the Ministry of Education and
Science of Poland from the budget funds for science in the years 2008–
2011 within the research project N N401 015035. M.G. would like to
express his thanks to Prof. P. Dynarowicz-Łątka from the Department
of General Chemistry, Faculty of Chemistry, Jagiellonian University
(Kraków, Poland) for helpful advice.
References
[1] R. Holz, A. Finkelstein, The water and nonelectrolyte permeability induced in thin
lipid membranes by the polyene antibiotics nystatin and amphotericin B, J. Gen.
Physiol. 56 (1970) 125–145.
[2] J. Brajtburg, W.G. Powderly, G.S. Kobayashi, G. Medoff, Amphotericin B: current
understanding of mechanisms of action, Antimicrob. Agents Chemother. 34
(1990) 183–188.
[3] S.C. Hartsel, S.K. Benz, R.P. Peterson, B.S. Whyte, Potasium-selective amphotericin
B channels are pre-dominant in vesicles regardless of sidedness, Biochemistry 30
(1991) 77–82.
[4] B. De Kruijff, W.J. Gerritsen, A. Oerlemans, R.A. Demel, L.L. van Deenen, Polyene
antibiotic-sterol interactions in membranes of Acholeplasma laidlawii cells and
lecithin liposomes. I. Specificity of the membrane permeability changes induced
by the polyene antibiotics, Biochim. Biophys. Acta 339 (1974) 30–43.
[5] M. Bagiński, Cybulska, B., Gruszecki, W. I., Interaction of polyene macrolide
antibiotics with lipid model membranes, in: A. Ottova-Liu (Ed.), Advances in
Planar Lipid Bilayers and Liposomes, vol. 3, Elsevier Science Publ., Amsterdam,
2006, pp. 269–329.
[6] L.N. Ermishkin, K.M. Kasumov, V.M. Potzeluyev, Single ionic channels induced in
lipid bilayers by polyene antibiotics amphotericin B and nystatine, Nature 262
(1976) 698–699.
M. Arczewska, M. Gagoś / Biochimica et Biophysica Acta 1808 (2011) 2706–2713
[7] T.E. Andreoli, The structure and function of amphotericin B–cholesterol pores in
lipid bilayer membranes, Ann. N. Y. Acad. Sci. 235 (1974) 448–468.
[8] S.C. Hartsel, C. Hatch, W. Ayenew, How does amphotericin B work?: Studies on
model membrane systems, J. Liposome Res. 3 (1993) 377–408.
[9] G. Fujii, J.-E. Chang, T. Coley, B. Steere, The formation of amphotericin B ion
channels in lipid bilayers, Biochemistry 36 (1997) 4959–4968.
[10] V. Khutorsky, Ion coordination in the amphotericin B channel, Biophys. J. 71
(1996) 2984–2995.
[11] H. Resat, F.A. Sungur, M. Baginski, E. Borowski, V. Aviyente, Conformational
properties of amphotericin B amide derivatives—impact on selective toxicity, J.
Comput. Aided Mol. Des. 14 (2000) 689–703.
[12] I. Gruda, P. Nadeau, J. Brajtburg, G. Medoff, Application of differential spectra in
the ultraviolet–visible region to study the formation of amphotericin B–sterol
complexes, Biochim. Biophys. Acta 602 (1980) 260–268.
[13] A. Vertut-Croquin, J. Bolard, M. Chabbert, C. Gary-Bobo, Differences in the interaction
of the polyene antibiotic amphotericin B with cholesterol- or ergosterol-containing
phospholipid vesicles. A circual dichroism and permeability study, Biochemistry 22
(1983) 2939–2944.
[14] J.D. Readio, R. Bittman, Equilibrium binding of amphotericin B and its methyl ester
and borate complex to sterols, Biochim. Biophys. Acta 685 (1982) 219–224.
[15] R.A. Brutyan, P. McPhee, On the one-sided action of amphotericin B on lipid
bilayer membranes, J. Gen. Physiol. 107 (1996) 69–78.
[16] M. Gagoś, J. Gabrielska, M. Dalla Serra, W.I. Gruszecki, Binding of antibiotic
amphotericin B to lipid membranes: monomolecular layer technique and linear
dichroism-FTIR studies, Mol. Membr. Biol. 22 (2005) 433–442.
[17] W.I. Gruszecki, M. Gagoś, P. Kernen, Polyene antibiotic amphotericin B in
monomolecular layers: spectrophotometric and scanning force microscopic
analysis, FEBS Lett. 524 (2002) 92–96.
[18] B.V. Cotero, S. Rebolledo-Antunez, I. Ortega-Blake, On the role of sterol in the
formation of the amphotericin B channel, Biochim. Biophys. Acta 1375 (1998) 43–51.
[19] A.R. Balakrishnan, K.R. Easwaran, Lipid–amphotericin B complex structure in
solution: a possible first step in the aggregation process in cell membranes,
Biochemistry 32 (1993) 4139–4144.
[20] M. Bagiński, H. Resat, J.A. McCammon, Molecular properties of amphotericin B
membrane channel: a molecular dynamics simulation, Mol. Pharmacol. 52 (1997)
560–570.
[21] M. Baginski, H. Resat, E. Borowski, Comparative molecular dynamics simulations
of amphotericin B–cholesterol/ergosterol membrane channels, Biochim. Biophys.
Acta 1567 (2002) 63–78.
[22] K. Hąc-Wydro, J. Kapusta, A. Jagoda, P. Wydro, P. Dynarowicz-Łątka, The influence
of phospholipid structure on the interactions with nystatin, a polyene antifungal
antibiotic A Langmuir monolayer study, Chem. Phys. Lipids 150 (2007) 125–135.
[23] J. Minones Jr., P. Dynarowicz-Łątka, O. Conde, J. Minones, E. Iribarnegaray, M. Cases,
Interactions of amphotericin B with saturated and unsaturated phoshatidylocholines
at the air/water interface, Colloids Surf. B Biointerfaces 29 (2003) 205–215.
[24] J. Miñones Jr., J. Miñones, O. Conde, J.M. Rodriguez Patino, P. Dynarowicz-Łątka,
Mixed monolayers of amphotericin B–dipalmitoyl phosphatidyl choline: study of
complex formation, Langmuir 18 (2002) 2817–2827.
[25] K. Hąc-Wydro, P. Dynarowicz-Łątka, J. Grzybowska, E. Borowski, N-(1piperidinepropionyl)amphotericin B methyl ester (PAME)—a new derivative
of the antifungal antibiotic amphotericin B: searching for the mechanism of its
reduced toxicity, J. Colloid Interface Sci. 287 (2005) 476–484.
[26] M. Gagoś, M. Arczewska, Influence of K(+) and Na(+) ions on the aggregation
processes of antibiotic amphotericin B: electronic absorption and FTIR spectroscopic
studies, J. Phys. Chem. B 115 (2011) 3185–3192.
[27] M. Gagoś, M. Arczewska, W.I. Gruszecki, Raman spectroscopic study of aggregation
process of antibiotic amphotericin B induced by H+, Na+ and K+ ions, J. Phys. Chem.
B (2011) doi.
[28] J. Brajtburg, J. Bolard, Carrier effects on biological activity of amphotericin B, Clin.
Microbiol. Rev. 9 (1996) 512-&.
[29] J. Brajtburg, G. Medoff, G. Kobayashi, S. Elberg, Influence of extracellular K+ or
Mg+ 2 on the stages of the antifungal effects of amphotericin B and filipin,
Antimicrob. Agents Chemother. 18 (1980) 593–597.
[30] R. Turcu, M.J. Patterson, S. Omar, Influence of sodium intake on amphotericin Binduced nephrotoxicity among extremely premature infants, Pediatr. Nephrol. 24
(2009) 497–505.
2713
[31] J.T. Davies, E.K. Rideal, Interfacial Phenomena, Academic Press, New York and
London, 1963.
[32] K. Hąc-Wydro, P. Dynarowicz-Łątka, Interaction between nystatin and
natural membrane lipids in Langmuir monolayers—the role of a phospholipid
in the mechanism of polyenes mode of action, Biophys. Chem. 123 (2006)
154–161.
[33] K. Hąc-Wydro, P. Dynarowicz-Łątka, J. Grzybowska, E. Borowski, How does the
N-acylation and esterification of amphotericin B molecule affect its interactions
with cellular membrane components—the Langmuir monolayer study, Colloids
Surf. B Biointerfaces 46 (2005) 7–19.
[34] I.S. Costin, G.T. Barnes, Two-component monolayers. II. Surface pressure–area
relations for the octadecanol–docosyl sulphate system, J. Colloid Interface Sci. 51
(1975) 106–121.
[35] P. Dynarowicz-Łątka, K. Kita, Molecular interaction in mixed monolayers at the
air/water interface, Adv. Colloid Interface Sci. 79 (1999) 1–17.
[36] P. Dynarowicz-Łątka, K. Kita, J.R. Seoane, J. Minones, C. Conde, Studying of the
molecular interaction in mixed monolayers at the air/water interface — application
in medical sciences, Acta Poloniae Pharmaceutica, Drug Res. 56 (1999) 3–10.
[37] G.L. Gaines, Insoluble Monolayers at the Liquid–Gas Interfaces, Interscience, New
York, 1966 chapter 6.
[38] M. Saint-Pierre-Chazalet, C. Thomas, M. Dupeyarat, C.M. Gary-Bobo, Amphotericin
B–sterol complex formation and competition with egg phosphatidylcholine: a
monolayer study, Biochim. Biophys. Acta 944 (1988) 477–486.
[39] J.R. Seoane, N. Vila Romeu, J. Miñones, O. Conde, P. Dynarowicz, M. Casas, The
behavior of amphotericin B monolayers at the air/water interface, Prog. Colloid
Polym. Sci. 105 (1997) 173–179.
[40] J.C. Sykora, W.C. Neely, V. Vodyanoy, Solvent effects on amphotericin B monolayers, J. Colloid Interface Sci. 269 (2004) 499–502.
[41] J.R. Seoane, J. Minones, O. Conde, M. Casas, E. Iribarnegaray, Molecular
organisation of amphotericin B at the air–water interface in the presence of
sterols: a monolayer study, Biochim. Biophys. Acta 1375 (1998) 73–83.
[42] J. Minones Jr., C. Carrera, P. Dynarowicz-Łątka, J. Minones, O. Conde, R. Seoane, J.M.
Rodriguez Patino, Orientational changes of amphotericin B in Langmuir monolayers
observed by Brewster Angle microscopy, Langmuir 17 (2001) 1477–1482.
[43] M. Gagoś, M. Arczewska, Spectroscopic studies of molecular organization of
antibiotic amphotericin B in monolayers and dipalmitoylphosphatidylcholine
lipid multibilayers, Biochim. Biophys. Acta 1798 (2010) 2124–2130.
[44] I. Rey-Gomez-Serranillos, P. Dynarowicz-Łątka, J. Minones Jr., R. Seoane,
Desorption of amphotericin B from mixed monolayers with cholesterol at the
air/water interface, J. Colloid Interface Sci. 234 (2001) 351–355.
[45] M. Gagoś, R. Koper, W.I. Gruszecki, Spectrophotometric analysis of organisation of
dipalmitoylphosphatylcholine bilayers containing the polyene antibiotic amphotericin B, Biochim. Biophys. Acta 1511 (2001) 90–98.
[46] M. Gagoś, M. Arczewska, Influence of K(+) and Na(+) ions on the aggregation
processes of antibiotic amphotericin B: electronic absorption and FTIR spectroscopic studies, J. Phys. Chem. B (2011), doi:10.1021/jp110543g.
[47] E.F. Aziz, N. Ottosson, S. Eisebitt, W. Eberhardt, B. Jagoda-Cwiklik, R. Vacha, P.
Jungwirth, B. Winter, Cation-specific interactions with carboxylate in amino acid
and acetate aqueous solutions: X-ray absorption and ab initio calculations, J. Phys.
Chem. B 112 (2008) 12567–12570.
[48] B. Hess, N.F. van der Vegt, Cation specific binding with protein surface charges,
Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 13296–13300.
[49] M. Gagoś, W.I. Gruszecki, Organization of polyene antibiotic amphotericin B at the
argon–water interface, Biophys. Chem. 137 (2008) 110–115.
[50] M. Herec, A. Islamov, A. Kuklin, M. Gagos, W.I. Gruszecki, Effect of antibiotic
amphotericin B on structural and dynamic properties of lipid membranes formed
with egg yolk phosphatidylcholine, Chem. Phys. Lipids 147 (2007) 78–86.
[51] M. Herec, H. Dziubinska, K. Trebacz, J.W. Morzycki, W.I. Gruszecki, An effect of
antibiotic amphotericin B on ion transport across model lipid membranes and
tonoplast membranes, Biochem. Pharmacol. 70 (2005) 668–675.
[52] J.J. Miñones, J. Miñones, O. Conde, R. Seoane, P. Dynarowicz-Lątka, Influence of a
spreading method on the properties of amphotericin B–dipalmitoyl phosphatidic
acid mixed films at the air/water interface, Langmuir 16 (2000) 5743–5748.
[53] J.R. Seoane, J. Miñones, O. Conde, M. Casas, E. Iribarnegaray, Interaction between
amphotericin B and sterols in monolayers. Mixed films of ergosterol–amphotericin B,
Langmuir 15 (1999) 3570–3573.