Direct interaction between amphotericin B and ergosterol in lipid bilayers as revealed by 2H NMR spectroscopy

Article abstract of DOI:10.1021/ja9033473

Although amphotericin B (AmB) is thought to exert its antifungal activity by forming transmembrane ion-permeable self-assemblies together with ergosterol, no previous study has directly proven AmB-ergosterol interaction. To establish the interaction, we m

Full text of DOI:10.1021/ja9033473

Published on Web 07/31/2009
Direct Interaction between Amphotericin B and Ergosterol in
2
Lipid Bilayers As Revealed by H NMR Spectroscopy
Nobuaki Matsumori,* Kazuaki Tahara, Hiroko Yamamoto, Atsushi Morooka,
Mototsugu Doi, Tohru Oishi, and Michio Murata*
Department of Chemistry, Graduate School of Science, Osaka UniVersity, Toyonaka,
Osaka 560-0043, Japan
Received April 25, 2009; E-mail: matsmori@chem.sci.osaka-u.ac.jp; murata@chem.sci.osaka-u.ac.jp
Abstract: Although amphotericin B (AmB) is thought to exert its antifungal activity by forming transmembrane
ion-permeable self-assemblies together with ergosterol, no previous study has directly proven AmB-ergosterol
interaction. To establish the interaction, we measured 2H NMR using deuterium-labeled sterols and AmB.
The 2H NMR spectra of deuterated ergosterol in palmitoyloleoylphosphatidylcholine (POPC) bilayers showed
that fast axial diffusion of erogosterol was almost completely inhibited by the coexistence of AmB. Conversely,
cholesterol mobility in POPC membrane was essentially unchanged with or without AmB. These results
unequivocally demonstrate that ergosterol has significant interaction with AmB in POPC bilayers. In addition,
we examined the mobility of AmB using deuterium-labeled AmB, and found that, although AmB is almost
immobilized in sterol-free and cholesterol-containing POPC membranes, a certain ratio of AmB molecules
acquires mobility in the presence of ergosterol. The similar mobility of AmB and ergosterol in POPC bilayers
confirmed the idea of the direct intermolecular interaction between ergosterol and AmB.
Amphotericin B (AmB, Chart 1) has been a standard drug
for treatment of deep-seated systemic fungal infections for nearly
50 years.1-3 For lack of better alternatives, as well as the rare
occurrence of resistant strains,4 the clinical importance of AmB
has remained unchanged. It is presently widely accepted that
AmB molecules in lipid bilayers form a barrel-stave channel,
with their polyhydroxy side pointing inward and their lipophilic
heptaene part directing outward.5,6 The occurrence of such
molecular assemblies in fungal cell membranes increases ion
permeability and alters membrane potentials, ultimately leading
to cell death.6-9 The pharmacological action of AmB is based
on its selective toxicity against fungi over mammalians, which
is thought to result from its stronger interaction to ergosterol
(Chart 1), an abundant sterol in fungal membranes, than to
cholesterol(Chart1),themajorsterolinmammalianmembranes.10,11
The interaction between AmB and sterol, particularly ergosterol,
has also been supported by a number of experimental results:
(a) higher affinity of AmB to ergosterol-containing membranes
than to sterol-free and cholesterol membranes;10,12,13 (b) more
potent ion-permeability of AmB in ergosterol-membranes;8,14-16
and (c) spectroscopic changes of AmB in sterol-containing
membranes, as seen in UV,17-19 CD,8,20-22 and IR spectra.23,24
None of these studies, however, have directly proven the
AmB-sterol interaction in membrane environments. Although
we have recently demonstrated close contacts between polyene
carbon atoms of AmB and a fluorine atom of 6-F-ergosterol
using an AmB-6-F-ergosterol covalent conjugate by a solid state
NMR technique,25 it might not be so straightforward to
(12) Readio, O. D.; Bittman, R. Biochim. Biophys. Acta 1982, 685, 219–
224.
(13) Mouri, R.; Konoki, K.; Matsumori, N.; Oishi, T.; Murata, M.
Biochemistry 2008, 47, 7807–7815.
(14) Herve, M.; Cybulska, B.; Gary-Bobo, C. M. Eur. Biophys. J. 1985,
12, 121–128.
(15) Herve, M.; Debouzy, J. C.; Borowski, E.; Cybulska, B.; Gary-Bobo,
C. M. Biochim. Biophys. Acta 1989, 980, 261–272.
(16) Brutyan, R. A.; McPhie, P. J. Gen. Physiol. 1996, 107, 69–78.
(17) Fujii, G.; Chang, J. E.; Coley, T.; Steere, B. Biochemistry 1997, 36,
4959–4968.
(1) Gallis, H. A.; Drew, R. H.; Pickard, W. W. ReV. Infect. Dis. 1990,
12, 308–329.
(18) Millie, P.; Langlet, J.; Berges, J.; Caillet, J.; Demaret, J.-P. J. Phys.
Chem. B 1999, 103, 10883–10891.
(2) Hartsel, S. C.; Bolard, J. Trends Pharmacol. Sci. 1996, 12, 445–449.
(3) Ablordeppey, S. Y.; Fan, P. C.; Ablordeppey, J. H.; Mardenborough,
L. Curr. Med. Chem. 1999, 6, 1151–1195.
(19) Gagooe, M.; Koper, R.; Gruszecki, W. I. Biochim. Biophys. Acta,
Biomembr. 2001, 1511, 90–98.
(20) Milhaud, J.; Ponsinet, V.; Takashi, M.; Michels, B. Biochim. Biophys.
Acta, Biomembr. 2002, 1558, 95–108.
(4) Vanden-Bossche, H.; Dromer, F.; Improvisi, I.; LozanoChiu, M.; Rex,
J. H.; Sanglard, D. Med. Mycol. 1998, 36, 119–128.
(5) Andreoli, T. E. Ann. N.Y. Acad. Sci. 1974, 235, 448–468.
(6) De-Kruijff, B.; Demel, R. A. Biochim. Biophys. Acta, Biomembranes
1974, 339, 57–70.
(21) Bolard, J.; Seigneuret, M.; Boudet, G. Biochim. Biophys. Acta 1980,
599, 280–293.
(22) Bolard, J.; Legrand, P.; Heitz, F.; Cybulska, B. Biochemistry 1991,
30, 5707–5715.
(7) Cohen, B. E. Biochim. Biophys. Acta 1986, 857, 117–122.
(8) Vertut-Croquin, A.; Bolard, J.; Chabbert, M.; Gary-Bobo, C. Bio-
chemistry 1983, 22, 2939–2944.
(23) Fournier, I.; Barwicz, J.; Auger, M.; Tancrede, P. Chem. Phys. Lipids
2008, 151, 41–50.
(24) Paquet, M.-J.; Fournier, I.; Barwicz, J.; Tancrede, P.; Auger, M. Chem.
Phys. Lipids 2002, 119, 1–11.
(9) Van-Hoogevest, P.; De-Kruijff, B. Biochim. Biophys. Acta 1978, 511,
397–407.
(25) Kasai, Y.; Matsumori, N.; Umegawa, Y.; Matsuoka, S.; Ueno, H.;
Ikeuchi, H.; Oishi, T.; Murata, M. Chem.-Eur. J. 2008, 14, 1178–
1185.
(10) Witzke, N. M.; Bittman, R. Biochemistry 1984, 23, 1668–1674.
(11) Baran, M.; Mazerski, J. Biophys. Chem. 2002, 95, 125–133.
9
10.1021/ja9033473 CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 11855–11860 11855

A R T I C L E S
Matsumori et al.
Chart 1. Structures of AmB, Ergosterol, Cholesterol and Their
Deuterium-Labeled Counterparts
of ergosterol in lipid bilayers are affected by the presence of
AmB. We also investigated the motional property of AmB in
the presence or absence of sterols. The merit of applying the
2H NMR method is that the deuterium labeling in molecules of
interest is essentially nonperturbing. The findings provided the
first experimental evidence of the intermolecular interactions
between AmB and ergosterol. Additionally, we have found that
the strength of AmB-ergosterol interactions varies depending
on the type of phospholipid used.
Results and Discussion
The preparation of erogosterol-d1 (Chart 1) commenced with
protection of the 5,7-diene of ergosterol with 4-phenyl-1,2,4-
triazoline-3,5-dione (PTAD) (Scheme 1). Oxidation of alcohol
1 to ketone 2 followed by reduction with NaBD4 afforded a
mixture of the desired alcohol 3 and its C3-epimer, which were
separable by silica gel column chromatography. Removal of
the PTAD group of 3 using diisobutylaluminum hydride
(DIBAL), however, unexpectedly led to replacement of the
deuterium on C3 by a hydrogen atom, probably via a
Meerwein-Ponndorf-Verley reaction. Thus, prior to removal
of the PTAD group, the alcohol group of 3 was protected by a
t-butyldimethylsilyl (TBDMS) group to produce 4. Removal of
the PTAD group of 4 with DIBAL gave rise to 5, which was
further treated with tetrabutylammonium fluoride (TBAF) to
provide the desired ergosterol-d1. Cholesterol-d136 (Chart 1) was
also prepared as previously reported.37
Chart 2. Structures of Phospholipids
Deuterium-labeled AmB (AmB-d6, Chart 1), in which hy-
drogen atoms on C39 and C40 methyl groups were substituted
with deuterium atoms, was biosynthetically prepared with a
slightly modified procedure for producing [39,40,41-13C3]-
AmB.38 Briefly, the AmB-producing actinomycete Streptomyces
nodosus (ATCC14899) was cultured in a modified FCA medium
in which the nutrient contents were reduced to half the amount
of the original medium to decrease hydrocarbon sources. To
the medium was added sodium [3-2H3]propionate to produce
site-specifically 39- and 40-2H-labeled AmB-d6. From 50 mL
of the culture, about 15 mg of AmB-d6 was obtained with 30%
deuterium enrichment at C39 and C40 positions.
With deuterium-labeled molecules available, we first mea-
sured the solid-state 2H NMR of ergosterol-d1 and cholesterol-
d1 in a membrane to observe the influence of AmB on dynamic
properties of sterols. In fast axially rotating substances such as
sterol molecules in lipid bilayers, quadrupolar splittings observed
in 2H NMR spectra depend on the tilt angle of the C-2H bond
and the wobbling of a molecule with respect to the rotation
axis. As described previously, Dufourc et al. showed that the
2H NMR spectra of deuterated cholesterol in DMPC bilayers
were substantially unchanged in the presence of AmB.35 Hence,
we carried out similar experiments using deuterium-labeled
ergosterol in place of cholesterol. Figure 1 shows the 2H NMR
spectra of ergosterol-d1 in DMPC bilayers in the presence or
absence of AmB. The spectra show typical Pake doublet patterns
with quadrupolar splittings of ca. 40 kHz, which is significantly
reduced from the rigid-limit quadrupolar splitting value (ca. 114
kHz). This value is almost equivalent to that of cholesterol in
extrapolate the data of the conjugate to the native AmB-sterol
interaction, because the conjugate can be structurally constrained
by the linkage. Moreover, AmB turned out to have ion
permeability without sterols under certain situations, such as
higher concentrations of AmB, osmotic gradient, and gel phase
membranes.9,26-32 Therefore, there is another notion that sterol
enhances AmB channel activity by changing structures and
properties of membranes rather than by directly participating
in AmB ion-channel complex.33,34
Meanwhile, detecting sterol dynamics affected by the pres-
ence of AmB would provide a direct clue to the AmB-sterol
interaction. Until now, however, very few studies have been
conducted on the influence of AmB on dynamic properties of
sterols; the only example has been reported by Dufourc et al.,35
but they did not detect the change of cholesterol dynamics in
the presence of AmB upon measuring the 2H NMR of
deuterium-labeled cholesterol in dimyristoylphosphatidylcholine
(DMPC, Chart 2) bilayers. Hence, in the present study, we also
2
used H NMR and examined whether the dynamic properties
(26) Milhaud, J.; Hartmann, M.; Bolard, J. Biochimie 1989, 71, 49–56.
(27) Whyte, B. S.; Peterson, R. P.; Hartsel, S. C. Biochem. Biophys. Res.
Commun. 1989, 164, 609–614.
(28) Hartsel, S. C.; Benz, S. K.; Peterson, R. P.; Whyte, B. S. Biochemistry
1991, 30, 77–82.
(29) Cohen, B. E. Biochim. Biophys. Acta 1992, 1108, 49–58.
(30) Wolf, B. D.; Hartsel, S. C. Biochim. Biophys. Acta 1995, 1238, 156–
162.
(31) Cohen, B. E. Int. J. Pharm. 1998, 162, 95–106.
(32) Ruckwardt, T.; Scott, A.; Scott, J.; Mikulecky, P.; Hartsel, S. C.
Biochim. Biophys. Acta 1998, 1372, 283–288.
(36) Rosenfeld, R. S.; Fukushima, D. K.; Hellman, L.; Gallagher, T. F.
J. Biol. Chem. 1954, 211, 301–311.
(33) Cotero, B. V.; Rebolledo-Antunez, S.; Ortega-Blake, I. Biochim.
Biophys. Acta, Biomembr. 1998, 1375, 43–51.
(37) Matsumori, N.; Kasai, Y.; Oishi, T.; Murata, M.; Nomura, K. J. Am.
Chem. Soc. 2008, 130, 4757–4766.
(34) Venegas, B.; Gonzalez-Damian, J.; Celis, H.; Ortega-Blake, I. Biophys.
J. 2003, 85, 2323–2332.
(38) McNamara, C. M.; Box, S.; Crawforth, J. M.; Hickman, B. S.;
Norwood, T. J.; Rawlings, B. J. J. Chem. Soc., Perkin Trans. 1998,
1, 83–87.
(35) Dufourc, E. J.; Smith, I. C. P.; Jarrell, H. C. Biochim. Biophys. Acta,
Biomembr. 1984, 776, 317–329.
9
11856 J. AM. CHEM. SOC. VOL. 131, NO. 33, 2009

Direct Interaction between Amphotericin B and Ergosterol
A R T I C L E S
Scheme 1. Preparation of Ergosterol-d1
the DMPC bilayers,37 suggesting that ergosterol also undergoes
fast rotational averaging in the DMPC membrane as is the case
with cholesterol. Unexpectedly, the doublet pattern is still clearly
observable in the presence of AmB (Figure 1B), while the peak
intensity seems to be slightly reduced. This indicates that a
considerable portion of ergosterol does not interact with AmB
in the DMPC bilayers, and gives rise to the splitting peaks in
Figure 1B. Figure 1 also shows large isotropic components in
spectra of sterols in the membrane. Figure 2 shows the spectra
of ergosterol-d1 in POPC bilayers with increasing ratios of AmB.
In the absence of AmB (Figure 2A), ergosterol-d1 shows the
splitting pattern (ca. 39 kHz) as seen in the DMPC membranes,
thus, indicating the fast axial rotation of ergosterol in POPC
bilayers. Unlike the results in DMPC bilayers, the splitting signal
of ergosterol-d1 broadened and almost disappeared with the
increase in AmB ratios (Figure 2B,C). The vanishing of the
doublet signal and the peak broadening are generally character-
ized by intermediate molecular motion with correlation times
2
the H NMR spectra, which are most likely due to coexisting
small vesicles tumbling rapidly in water, although the samples
also contain a small amount of residual deuterium from HDO
in the water.
Since it is known that saturated phospholipids such as DMPC
and dipalmytolylphosphatidylcholine (DPPC) have strong in-
teractions with AmB,24,39 probably via hydrophobic contacts
between the saturated acyl chains of PC and the polyene portion
of AmB, we postulate that the strong AmB-DMPC interaction
rather impedes the interaction between AmB and ergosterol.
On the other hand, we have recently shown by surface plasmon
resonance (SPR) experiments that AmB exhibits much higher
affinity for sterol-containing palmitoyloleoylphosphatidylcholine
(POPC, Chart 2) membranes than those without sterol.13 Hence,
to selectively observe the spectral changes of deuterated sterols
in the presence or absence of AmB, we altered the membrane
2
constituent from DMPC to POPC, and measured the H NMR
Figure 1. 2H NMR spectrum of ergosterol-d1 in DMPC bilayers in the
absence (A) and presence (B) of AmB. The molar ratios of ergosterol-d1/
AmB/DMPC are 1:0:18 (A) and 1:2:18 (B). Spectra were recorded at 30
°C. The quadrupolar splitting value is 40 kHz. The isotropic center peaks
are likely to be attributed to coexisting small vesicles with faster tumbling
in water.
Figure 2. 2H NMR spectra of ergosterol-d1 in POPC with increasing ratios
of AmB. The molar ratios of ergosterol-d1/AmB/POPC are 1:0:18 (A), 1:1:
18 (B), and 1:2:18 (C). Spectra were recorded at 30 °C. The isotropic center
peaks are likely to be attributed to coexisting small vesicles with faster
tumbling in water.
9
J. AM. CHEM. SOC. VOL. 131, NO. 33, 2009 11857

A R T I C L E S
Matsumori et al.
Figure 3. 2H NMR spectra of cholesterol-d1 in POPC with increasing ratios
of AmB. The molar ratios of cholesterol-d1/AmB/POPC are 1:0:18 (A),
1:1:18 (B), and 1:2:18 (C). Spectra were recorded at 30 °C. The isotropic
center peaks are likely to be attributed to coexisting small vesicles with
faster tumbling in water.
of 10-4-10-5 s.40 Therefore, Figure 2C shows that the fast
motional averaging of ergosterol is inhibited by the presence
of AmB, indicating that ergosterol has a significant interaction
with AmB. In contrast, 2H NMR spectra of cholesterol-d1 hardly
change in the presence or absence of AmB (Figure 3), indicative
of weaker intermolecular interaction between AmB and cho-
lesterol. These data strongly support the notion that AmB
directly interacts with ergosterol, and further, confirm that
ergosterol has much stronger affinity for AmB than does
cholesterol.
Figure 4. 2H NMR spectra of AmB-d6 in powder state (A), in sterol-free
POPC bilayers (B), in cholesterol-containing POPC bilayers (C), and in
ergosterol-containing POPC bilayers (D). The molar ratios of AmB-d6/sterol/
POPC are 1:0:20 (B) and 1:2:18 (C, and D). Spectra were recorded at 30
°C. The isotropic components of panels B-D are likely to be attributed to
coexisting small vesicles with faster tumbling in water.
deuterated at the 39- and 40-methyl groups. Although Schaefer’s
group41 and our group42 have reported the dynamics of AmB
in membranes using its amide derivatives, this is the first report
2
using virtually nonderivatized AmB. Figure 4A shows the H
The above results can be interpreted in terms of the equilib-
rium of sterol molecules between freely rotating state (unbound
to AmB) and AmB-bound state. Although it is difficult to
quantitatively evaluate the equilibrium constant between AmB
and sterols in POPC membranes from the current 2H NMR data,
we can safely conclude that in case of cholesterol the equilibrium
leans much to the free state, which gives no detectable change
in 2H NMR spectra (Figure 3), whereas ergosterol is mostly in
the interacting state, thus, giving the broadened spectra as shown
in Figure 2.
NMR spectrum of AmB-d6 in the powder state, where the
quadrupolar splitting is intrinsically reduced to 37 kHz due to
the methyl rotation. In sterol-free and cholesterol-containing
POPC bilayers (Figure 4B,C), the quadrupolar splittings of AmB
are clearly observed and the values (37 kHz) are almost identical
to that in the powder state, demonstrating that AmB molecules
are mostly immobilized in those membrane systems. This
corresponds to formation of large aggregates of AmB in these
membranes.42 Meanwhile, the 2H NMR of AmB-d6 in ergosterol-
containing POPC membrane shows that the split signal, albeit
still observed, is markedly reduced (Figure 4D), although we
used the same amount of AmB-d6 and almost equalized the
number of accumulations for measuring Figure 4B-D spectra.
Next, to further characterize the AmB-ergosterol interaction
from the standpoint of AmB, we examined the motional
properties of AmB in POPC membranes using AmB-d6 that was
(39) Dufourc, E. J.; Smith, I. C. P.; Jarrell, H. C. Biochim. Biophys. Acta,
Biomembr. 1984, 778, 435–442.
(41) Hing, A. W.; Schaefer, J.; Kobayashi, G. S. Biochim. Biophys. Acta,
Biomembr. 2000, 1463, 323–332.
(40) Duer, M. J. Introduction to Solid-State NMR Spectroscopy; Blackwell:
Oxford, 2004; pp 322-335.
(42) Matsumori, N.; Sawada, Y.; Murata, M. J. Am. Chem. Soc. 2006, 128,
11977–11984.
9
11858 J. AM. CHEM. SOC. VOL. 131, NO. 33, 2009

Direct Interaction between Amphotericin B and Ergosterol
A R T I C L E S
Table 1. Mobility of AmB, Ergosterol, and Cholesterol in POPC
conjugated triene on the sterol skeleton is unexpectedly weaker
than that in the presence of sterols having a conjugated diene
system such as ergosterol or 7-dehydrocholesterol (to be
published in due course). This probably suggests that the
rigidification of sterol nucleus by conjugated double bonds is
not the only determinant of the interaction with AmB. To gain
further insights into the molecular recognition between AmB
and sterol, measurements of intermolecular distances between
AmB and ergosterol by solid-state NMR are now in progress.
Bilayers
molecular mobility
ergosterol
AmB
cholesterol
single substance in
POPC bilayers
high
immobile
high
AmB-ergosterol
coexisting system
AmB-cholesterol
coexisting system
medium
medium
immobile
high
Conclusions
2
In H NMR spectra, not only the signal broadening, but also
the peak intensity is an indicator of molecular mobility.
Therefore, the reduction of signal intensity in ergosterol-
containing POPC membrane as shown in Figure 4D suggests
that, although immobilized aggregates of AmB partly exist, a
significant proportion of AmB molecules are in motion on a
time scale of quadrupolar coupling (10-4-10-5 s). AmB is,
therefore, considered to gain some mobility in the ergosterol
membrane in contrast to the immobile AmB in cholesterol-
containing and sterol-free POPC membranes.
We have achieved the first direct observation of the inter-
molecular AmB-ergosterol interaction in POPC membranes
using deuterated sterols and AmB. The similar mobility of AmB
2
and ergosterol in POPC bilayers as shown in the H NMR
spectra confirms their direct complexation. As mentioned above,
there exists another notion that sterol enhances AmB channel
activity by changing structures and properties of membranes
rather than by directly participating in AmB ion-channel
complex.33,34 Therefore, this study settles the dispute about the
effect of ergosterol on AmB channel. In addition, we demon-
strated that a certain type of phospholipids interfered with the
AmB-ergosterol interaction; DMPC, which has high affinity
to AmB through the strong binding between polyene carbons
of AmB and saturated acyl chains of DMPC, relatively weakens
the AmB-ergosterol interaction, while POPC, which should
have weaker interaction with AmB because of the flexible unsat-
urated acyl chain, allows for the AmB-ergosterol interaction.
To understand the mode of action of AmB in biomembranes,
it is necessary to take account of three intermolecular interac-
tions: AmB-AmB, AmB-sterol, and AmB-phospholipid. It
is more likely that these interactions are not independent but
interdependent; therefore, one interaction could promote and/
or prevent the other two. As shown in this study, strong
AmB-phospholipid interaction may impede AmB-sterol vicin-
ity. On the other hand, AmB-ergosterol interaction could loosen
AmB-AmB contact to prevent the formation of large AmB
aggregates in membranes. Previously, a number of research
groups had attempted to elucidate the mechanism of the AmB
channel, but results were sometimes conflicting and confusing.
This is probably because they used different phospholipids,
which significantly influenced AmB-AmB and AmB-sterol
interactions. The present results clearly show that POPC is more
suitable than saturated phospholipids for examining the
AmB-sterol interaction. Moreover, since most biological
membranes, including fungal ones, contain unsaturated phos-
pholipids as a major constituent, the findings obtained in this
study using POPC membranes should largely hold true for the
biological activity of AmB.
Table 1 summarizes the mobility of AmB, ergosterol, and
2
cholesterol in POPC membranes revealed by the current H
NMR measurements. Interestingly, the coexistence of ergosterol
and AmB makes the mobility of ergosterol (Figure 2C) and
AmB (Figure 4D) comparable; fast axial diffusion of ergosterol
decelerates upon addition of AmB, while immobilized AmB
begins to move in the presence of ergosterol. These findings
unequivocally demonstrate that AmB and ergosterol interact
closely to form a channel complex and move together in POPC
bilayers.
We have recently reported that the AmB-AmB intermo-
lecular distance is elongated by 2 Å in an ergosterol-containing
POPC membrane as compared with the distance in sterol-free
membranes on the basis of rotational-echo double resonance
(REDOR) experiments.43 Those results together with the current
findings allow us to propose a simple model on the function of
ergosterol. Without ergosterol, AmB molecules are largely in
aggregated forms with low mobility due to the tight AmB-AmB
complexation. AmB is known to induce slight ion permeation
even in the absence of sterol,9 which might be attributed to
membrane defects formed at the phase boundary between lipid
bilayers and AmB aggregates.28 Meanwhile, in the presence of
ergosterol, the AmB-AmB binding should be loosened by the
interaction with ergosterol, which consequently enhances the
mobility of the AmB aggregates and increases the intermolecular
distances between AmB molecules.43 These effects of ergosterol
may facilitate the formation of ion channels in the membranes.
Nonetheless, it is still unclear how AmB discriminates the
minute difference between ergosterol and cholesterol and
interacts preferentially with ergosterol. Previous studies showed
that the conjugated double bonds on the sterol skeleton are of
critical importance in interacting with AmB,44,45 and the
conjugated diene on ergosterol is thought to contribute to
rigidifying the sterol backbone, thus, allowing the molecular
proximity necessary to maximize the hydrophobic interactions
with AmB.45 Our recent study, however, revealed that the ion
permeability of AmB in the presence of sterols that have a
Experimental Section
Experimental procedures for the preparation of AmB-d6 and
ergosterol-d1 are described in the Supporting Information.
2
2
Sample Preparation for H NMR. For measurements of H
NMR spectra of deuterated sterols in DMPC or POPC membranes,
5.0 µmol of ergosterol-d1 or cholesterol-d1, 90.0 µmol of phospho-
choline (DMPC or POPC), and AmB (0, 5, or 10 µmol) were
dissolved in CHCl3-MeOH, and the solvent was removed in vacuo
for 12 h. The dried membrane film was hydrated with 1 mL of
water. After a few minutes of sonication, the suspension was
freeze-thawed three times and vortexed to make multilamellar
vesicles. The vesicle solution was lyophilized, rehydrated with
deuterium-depleted water (50% w/w), and freeze-thawed three
(43) Umegawa, Y.; Matsumori, N.; Oishi, T.; Murata, M. Biochemistry
2008, 47, 13463–13469.
(44) Clejan, S.; Bittman, R. J. Biol. Chem. 1985, 260, 2884–2889.
(45) Charbonneau, C.; Fournier, I.; Dufresne, S.; Barwicz, J.; Tancrede,
P. Biophys. Chem. 2001, 91, 125–133.
9
J. AM. CHEM. SOC. VOL. 131, NO. 33, 2009 11859

A R T I C L E S
Matsumori et al.
times. The sample was transferred into a 5-mm glass tube (Wilmad),
which was sealed with epoxy glue.
Acknowledgment. We are grateful to Dr. Yuichi Umegawa
for his advice on streptomycete culture. This work was supported
by Grant-In-Aids for Young Scientists (A) (No. 17681027), for
Scientific Research (B) (No. 20310132) and (S) (No. 18101010)
from MEXT, Japan.
For measuring of 2H NMR spectra of AmB-d6 in POPC
membranes, a combination of 3.25 µmol of AmB-d6 and 58.44 µmol
of POPC (molar ratio 1:18), or that of 3.25 µmol of AmB-d6, 58.44.
µmol of POPC, and 6.5 µmol of cholesterol or ergosterol (molar
ration 1:18:2), was mixed in CHCl3-MeOH. The following proce-
dure is the same as above.
Supporting Information Available: Experimental details of
preparation of AmB-d6 and ergosterol-d1. This material is
available free of charge via the Internet at http://pubs.acs.
org.
2H NMR Measurements. All the 2H NMR spectra were recorded
on a 300 MHz CMX300 spectrometer (Chemagnetics, Varian, Palo
2
Alto, CA). Spectra were recorded at 30 °C with a 5-mm H static
JA9033473
probe (Otsuka Electronics, Osaka, Japan) using a quadrupolar echo
sequence.46 The 90° pulse width was 2 µs, interpulse delay was
30 µs, and repetition rate was 0.5 s. The sweep width was 200
kHz, and the number of scans was around 100 000.
(46) Davis, J. H.; Jeffrey, K. R.; Bloom, M.; Valic, M. I.; Higgs, T. P.
Chem. Phys. Lett. 1976, 42, 390–394.
9
11860 J. AM. CHEM. SOC. VOL. 131, NO. 33, 2009