Synthesis of 6-F-ergosterol and its influence on membrane-permeabilization of amphotericin B and amphidinol 3

Article abstract of DOI:10.1039/c0ob00685h

Two well-known antifungals, amphotericin B (AmB) and amphodinol 3 (AM3), are thought to exert antifungal activity by forming ion-permeable channels or pores together with sterol molecules. However, detailed molecular recognitions for AmB-sterol and AM3-sterol in lipid bilayers have yet to be determined. Toward ¹⁹F NMR-based investigation of the molecular recognition underlying their potent antifungal activity, we synthesized 6-fluoro-ergosterol in five steps via ring opening of (5α,6α)-epoxide of ergosterol acetate with using novel combination of TiF₄ and n-Bu ₄N⁺Ph₃SiF₂^-. Then we evaluated its activity of promoting pore formation of AmB and AM3, and found that pore formation of AmB was barely promoted by 6-F-ergosterol in clear contrast to the dramatic promotion effect of unmodified ergosterol, whereas AM3 activity was markedly enhanced in the presence of 6-F-ergosterol, which was comparable to that of unmodified ergosterol. These results indicate that the introduction of an F atom at C6 position of ergosterol plays an inhibitory role in interacting with AmB, but it is not the case with AM3. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c0ob00685h

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 1437
www.rsc.org/obc
PAPER
Synthesis of 6-F-ergosterol and its influence on membrane-permeabilization of
amphotericin B and amphidinol 3
a
a
a
a
a
a
Yusuke Kasai, Nobuaki Matsumori,* Hiroyuki Ueno, Kenichi Nonomura, Shinya Yano, Murata Michio*
and Tohru Oishib
Received 7th September 2010, Accepted 22nd November 2010
DOI: 10.1039/c0ob00685h
Two well-known antifungals, amphotericin B (AmB) and amphodinol 3 (AM3), are thought to exert
antifungal activity by forming ion-permeable channels or pores together with sterol molecules.
However, detailed molecular recognitions for AmB-sterol and AM3-sterol in lipid bilayers have yet to
1
9
be determined. Toward F NMR-based investigation of the molecular recognition underlying their
potent antifungal activity, we synthesized 6-fluoro-ergosterol in five steps via ring opening of
+
SiF
-.
(
5a,6a)-epoxide of ergosterol acetate with using novel combination of TiF
4
and n-Bu
4
N Ph
3
2
Then we evaluated its activity of promoting pore formation of AmB and AM3, and found that pore
formation of AmB was barely promoted by 6-F-ergosterol in clear contrast to the dramatic promotion
effect of unmodified ergosterol, whereas AM3 activity was markedly enhanced in the presence of
6
-F-ergosterol, which was comparable to that of unmodified ergosterol. These results indicate that the
introduction of an F atom at C6 position of ergosterol plays an inhibitory role in interacting with AmB,
but it is not the case with AM3.
Introduction
AM3 has much higher affinity to sterol-containing membranes
than sterol-free one. As in the case of AmB, however, the mecha-
9
Amphotericin B (AmB, Fig. 1) has been a standard drug for
treatment of deep-seated systemic fungal infections for nearly
nism of sterol’s promoting AM3 activity is far from understanding.
19
Meanwhile, F NMR has been of interest for many years in
1–3
5
0 years. For lack of better alternatives, as well as the rare
investigating biological systems due in large part to the attractive
4
occurrence of resistant strains, the clinical importance of AmB
has remained unchanged. It is widely accepted that AmB exerts its
antifungal activity by forming transmembrane ion-permeable self-
assemblies together with ergosterol (Fig. 1), an abundant sterol in
fungal membranes. The selective toxicity of AmB is accounted
for by its higher affinity to ergosterol than cholesterol (Fig.
19
properties of F, which include 100% natural abundance, spin
I = 1/2, large magnetogyric ratio, and low background signals in
19
biological samples. In particular, solid-state F NMR has become
an important tool for characterizing molecular interactions in
lipid bilayers. In the course of our structural studies on the
10
ion-channel formed by AmB and sterol molecules, particularly
1
), a component of mammalian membrane. However, detailed
11
12
the molecular recognitions for AmB–AmB and AmB–sterol,
molecular recognition between AmB and ergosterol in lipid
bilayers has yet to be determined.
Similarly, amphidinol 3 (AM3, Fig. 1) also shows membrane
permeabilizing activity. AM3 was isolated from marine di-
noflagellate Amphidinium krebsii as a potent antifungal, hemolytic
and cytotoxic substance, and then shown to form pores across
biological membranes. Recently, we have revealed that the pore-
formation of AM3 is dramatically promoted in the presence of
13,14
we have prepared several fluorinated derivatives of AmB
and fluorinated cholesterol (Fig. 1), and used them for solid-
15
16
19
state NMR measurements. In this study, toward the F NMR
investigation of AmB-ergosterol and AM3-ergosterol interactions
in membrane, we developed a novel preparation method of a
fluorinated derivative of ergosterol, 6-F-ergosterol (Fig. 1). We
then assessed its activity of promoting membrane permeabilization
by AmB and AM3.
5
–7
6,8
5
,6
7
cholesterol and ergosterol, which is presumably related to the fact
revealed by surface plasmon resonance (SPR) experiments that
Results
a
Department of Chemistry, Graduate School of Science, Osaka
Preparation of 6-F-ergosterol
University, 1-1 Machikaneyama, Toyonaka, Osaka, 560-0043, Japan.
E-mail:
matsmori@chem.sci.osaka-u.ac.jp,
murata@chem.sci.osaka-
Synthesis of 6-F-ergosterol is shown in Scheme 1. Although
u.ac.jp; Fax: +81 6 6850-5790 +81 6 6850-5774; Tel: +81 6 6850-5789 +81
synthesis of 6-F-ergosteryl acetate 4 from ergosteryl acetate 1 was
6
6850-5774
17
b
reported by Barrett et al. by treatment with chromyl fluoride
Department of Chemistry, Graduate School of Sciences, Kyushu University,
18
6
-10-1 Hakozaki, Higashi-ku, Fukuoka, 812-8581, Japan
(CrO
2
F
2
)
giving fluorohydrin 3 followed by dehydration with
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1437–1442 | 1437

View Article Online
Fig. 1 Structures of sterols including 6-F-ergosterol, AmB, and AM3.
Epoxidation of ergosteryl acetate 1 with MCPBA in the presence
of Na CO gave (5a,6a)-epoxide 2 as a single diastereomer
in 88% for two steps from ergosterol. Treatment of 2 with
2
3
19
2
0
triethylamine trihydrofluoride (Et
3
N·3HF) resulted in the for-
mation of desired 6a,5a-fluorohydrin 3 and 6b-epimer 5 with
concomitant formation of conjugated diene 6 in a 15 : 10 : 75
ratio (Table 1, entry 1). The structures of these compounds
1
7,19,21
were determined by NMR experiments.
Presumably, the
fluorohydrins 3 and 5 were formed via allylic cation gener-
ated from acid catalyzed ring opening of the vinyl epoxide 2,
followed by nucleophilic attack of fluoride anion at the less
hindered C6 position (Fig. 2A, path a), whereas deprotonation
at C14 resulted in the formation of diene 6 (Fig. 2A, path
b). We next carried out the reaction using TiF
4
and fluo-
+
-
rinating reagents: tetra-n-butylammonium fluoride n-Bu
4
N F
TBAF, entry 2 in Table 1), tris(dimethylamino)sulfonium
22
(
+
-
difluorotrimethylsilicate (Me
2
N)
3
S Me
3
SiF
2
(TASF, entry 3 in
23
Table 1), and tetra-n-butylammonium difluorotriphenylsilicate
+
-
24
n-Bu
4
N Ph
3
SiF
2
(TBAT, entry 4 in Table 1), to give compa-
rable results as entry 1. After considerable experimentation, we
found that the reaction using novel combination of reagents,
TiF
= 43 : 0 : 57, entry 5 in Table 1) to afford fluorohydrin 3 as a
single diastereomer, presumably via activation of the epoxide with
4
/TBAT in dichloromethane, gave the best result (3 : 5 :
6
Scheme 1 Synthesis of 6-F-ergosterol. Reagents and conditions: a) Ac
2
O,
25
titanium species and subsequent internal fluoride ion delivery
(Fig. 2B). Purification by flash column chromatography afforded
3 in 38% yield which was isolated from diene 6 (45%). Finally,
DMAP, pyridine, CH
8% (2 steps); c) TiF , nBu
d) SOCl , pyridine, CH Cl
rt, 1 h, 42% (2 steps).
2
Cl
2
, rt, 1 h; b) MCPBA, Na
2
CO
3
, CH
2
Cl
2
, rt, 1 h,
◦
8
4
4
NPh
3
SiF
2
, CH
2
Cl
2
, 0 C, 40 min, 38% (6: 45%);
◦
◦
2
2
2
, -30 C to 0 C, 1 h; e) K
2
CO
3
, MeOH, THF,
17
dehydration of cis-fluorohydrin 4 with SOCl /pyridine and
2
methanolysis of the acetate furnished 6-F-ergosterol in 42% for
SOCl
generated from CrO
stream of nitrogen gas. Therefore, we developed an alternative
method using conventional synthetic operations with tractable
reagents.
2
, careful handling is required for hazardous chromyl fluoride
◦
two steps. In contrast to 3, dehydration of trans-fluorohydrin
3
and CoF
3
by heating at 450 C under the
26
5
using SOCl /pyridine or Burgess reagent was unsuccessful
2
to give a mixture of byproducts containing C4-C5 unsaturated
counterpart.
1
438 | Org. Biomol. Chem., 2011, 9, 1437–1442
This journal is © The Royal Society of Chemistry 2011

View Article Online
Table 1 Epoxide opening reaction
a
Entry
Conditions
Et
Ratio (3 : 5 : 6)
◦
1
2
3
4
5
3
N·3HF, CH
2
Cl
2
, 0 C, 1 h
15 : 10 : 75
12 : 22 : 66
19 : 15 : 66
17 : 14 : 69
43 : 0 : 57
◦
TiF
TiF
TiF
TiF
4
4
4
4
, TBAF, THF, 0 C to rt, 9 h
, TASF, CH
, TBAT, THF, 0 C to rt, 2 d
, TBAT, CH
◦
◦
2
Cl
2
, -50 C to -30 C, 2 h
◦
◦
2
Cl
2
, 0 C, 40 min
a
1
Ratio was determined by H NMR of the crude product.
+
+
Fig. 3 AmB-induced K /H exchange monitored by liposome-encap-
sulated pH dependent fluorescent dye. Liposomes were prepared from
POPC containing 10 mol% ergosterol (filled square), 10 mol% 6-F-er-
gosterol (open square), 10 mol% cholesterol (filled circle), and 10 mol%
6
-F-cholesterol (open circle). The r values of sterol-free POPC membrane
stayed near 0 over the R values tested. Three experiments were done for
each point, and average values are shown in the figure with error bars. Inlet:
the enlarged graph for 6-F-erogosterol, 6-F-cholesterol, and cholesterol.
C6 of the sterol skeleton largely eliminates the ergosterol selectivity
of channel formation by AmB.
Fig. 2 Plausible reaction pathways of the epoxide opening.
+
K Flux assay of AmB in the presence of 6-F-ergosterol
Calcein leakage assay of AM3 in the presence of 6-F-ergosterol
To examine the effect of 6-F-ergosterol on the ion channel
formation of AmB molecules, we then assessed the membrane-
permeabilizing activity of AmB using artificial liposomes
composed of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine
Then we examined the membrane-permeabilizing activity of AM3
in the presence of fluorinated and unmodified sterols. Although
AM3 also forms pores in membranes as AmB does, its pore size
28
was reported to be around 2 nm in diameter, which is about three
27
+
29
(
POPC). The liposomes possessed a higher external K con-
times larger that of AmB (0.7 nm in diameter). Due to its larger
+
centration and trans-membrane pH gradient with pH 5.8 inside
pore size, the K -influx assay used for AmB was not applicable
+
and pH 7.8 outside. K influx into the liposomes through
to the evaluation of AM3 activity, because pH-dependent dye
BCECF, which should be entrapped into liposomes even after pore
formation, leaks from the AM3 pore. Therefore, the membrane-
permeabilizing activity of AM3 was evaluated by the leakage of
+
AmB ion channels elicited H leakages by a proton-transporter
carbonyl cyanid-p-trifluoro-methoxyphenyl hydrazone (FCCP),
consequently raising the inner pH. The pH change in the liposomes
was monitored by the increase in the fluorescent intensity of
liposome-entrapped BCECF, a pH dependent fluorescent dye. In
7
a fluorescent dye, calcein, entrapped in POPC liposomes. Since
the fluorescence of liposome-encapsulated calcein is very weak
due to self-quenching while that of calcein outside liposomes is
strong, the AM3 pore formation can be evaluated by the increase
in fluorescent intensity of calcein. On the contrary, the calcein
leakage assay is not suitable to the evaluation of AmB activity
because AmB pores are not large enough to pass through calcein
molecules.
+
+
this method, K /H exchange rate r, which grossly corresponds
to the conductance of AmB-induced ion channels, was calculated
from the initial rate of fluorescence intensity change. The r value
+
was expressed as percentage of [K ] entering into liposomes per
+
second with respect to the initial [K ] difference between outside
and inside of the liposomes.
Fig. 3 shows the relation between AmB concentration and
r value. We used POPC liposomes containing 10 mol% of 6-
F- and unmodified ergosterols, as well as 6-F- and unmodified
cholesterols for comparison. As shown in Fig. 3, while ergos-
terol naturally promoted membrane-permeabilization of AmB,
the activity of 6-F-ergosterol was unexpectedly diminished as
compared with unmodified ergosterol. These results indicate that
the introduction of an F atom at C6 position of ergosterol plays
an inhibitory role in interacting with AmB. Intriguingly, the
promoting activity of 6-F-cholesterol is comparable to that of 6-
F-ergosterol, which implies that introduction of a fluorine atom at
Table 2 lists t90 values that are elapsed times until 90% of
liposome-entrapped calcein was leaked at 2.5 mM of AM3.
As reported previously, AM3 showed no activity to sterol-free
7
liposomes, whereas addition of 5% cholesterol or ergosterol to
liposomes dramatically elevated the activity of AM3 (Table 2).
Interestingly, calcein was leaked more rapidly from ergosterol-
containing liposomes than from cholesterol-liposomes. This is
consistent to our recent finding in SPR experiments that AM3
has higher affinity to ergosterol-membrane than cholesterol-
9
membrane. Noteworthy is that fluorinated sterols show compa-
rable or even higher promoting effects than unmodified ones; the
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1437–1442 | 1439

View Article Online
Table 2 Times for 90% calcein leakage from liposomes by 2.5 mM of
recognition between AM3 and sterol predominantly takes place
around the A-ring of sterol skeleton including 3-OH group, and
therefore the introduction of F atom on the B-ring has small
influence on the AM3 activity.
In summary, the different response of AmB and AM3 to 6-
F-ergosterol probably reflects the difference in pore formation
mechanism of AmB and AM3, and appears to be consistent with
the proposed pore models, barrel-stave pore for AmB and toroidal
one for AM3.
AM3
a
POPC liposomes
6
-F-
6-F-
Sterol-
b
b
b
b
Ergosterol Ergosterol Cholesterol Cholesterol free
c
d
t
90 (sec) 56
64
203
22
n.d.
a
b
POPC concentration of liposome suspensions was set to 27 mM. Sterols
c
were contained at 5 mol% of POPC. Times for 90% calcein leakage
from liposomes at 2.5 mM of AM3. Leakage was not observed at all for
1
d
5 min.
Conclusion
In conclusion, 6-F-ergosterol was synthesized from ergosterol
in 14% overall yield for five steps through epoxide opening
effect of 6-F-ergosterol was almost identical to that of unmodified
ergosterol, and 6-F-cholesterol showed even higher promoting
activity than unmodified cholesterol. These data suggest that, in
the case of AM3, the introduction of a fluorine atom at C6 of
sterol skeleton does not hinder, or in some cases rather enhances,
the AM3-sterol interactions.
+
-
reaction using a novel combination of TiF
4
and n-Bu
4
N Ph
3
SiF
2
in dichloromethane. We then assessed its activity of promot-
ing membrane-permeabilization of AmB and AM3. The assays
showed that the activity of AmB was barely promoted by 6-
F-ergosterol, while it promoted AM3 activity as efficacious as
unmodified ergosterol. This suggests that the introduction of a
F atom at C6 position of ergosterol plays an inhibitory role in
interacting with AmB, but it is not the case with AM3. The
different response of AmB and AM3 to 6-F-ergosterol is likely
attributed to the difference in the pore formation mechanism
of AmB and AM3. For comparison, we also assayed 6-F- and
unmodified cholesterols, and found that 6-F-cholesterol is more
effective than cholesterol in promoting AM3 activity. The current
study demonstrates that the fluorinated derivative of ergosterol
Discussion
In this study, we first developed a practical route to synthesize
6
-F-erogosterol, and examined its activity of promoting pore
formations of AmB and AM3. The results unexpectedly showed
that the activity of AmB was barely promoted by 6-F-ergosterol,
while it promoted AM3 activity as efficacious as unmodified
ergosterol. Here, we try to explain this different response of AmB
and AM3 to 6-F-ergosterol from the viewpoint of pore formation
mechanisms of AmB and AM3.
1
9
and cholesterol should serve as powerful molecular probes for
F
NMR studies on the detailed mechanism of AM3 pore formation.
30
AmB is thought to form barrel-stave type pores, in which
AmB molecules penetrate cell membranes while interacting with
Experimental
1
2b
ergosterol in a parallel manner.
Accordingly, in the pore
31
complex, AmB recognizes multiple points of ergosterol molecule,
Preparation of 6-F-ergosterol
such as 3-OH group, steroid skeleton including conjugated double
bonds and the terminal side chain. In particular, the conjugated
double bonds of ergosterol are thought to be of critical importance
Chemistry. Anhydrous methylene chloride (CH
tetrahydrofuran (THF) were purchased from Kanto Chemical Co.
Inc., and used without further drying. TiF was purchased from
was from TCI. All other
chemicals were obtained from local venders, and used as supplied
unless otherwise stated. Analytical thin-layer chromatography
2
Cl
2
) and
31,32
4
for interacting with the heptaene portion of AmB.
In this
+
-
Sigma-Aldrich, and n-Bu
4
N Ph
3
SiF
2
context, the introduction of electronegative fluorine atom on C6
of ergosterol should considerably affect the electronic state of
the conjugated diene in ergosterol, which probably weakens the
interaction with the polyene portion of AmB and consequently
reduces the AmB activity.
(TLC) was performed using E. Merck silica gel 60 F254 pre-coated
plates (0.25-mm thickness). For column chromatography, Kanto
silica gel 60 N (spherical, neutral, 100–210 mm) was used. Optical
rotations were recorded on a JASCO P-1010 polarimeter. IR
spectra were recorded on a JASCO FT-IR-300E Fourier transform
On the other hand, AM3 is assumed to form toroidal-type
7,33
pores,
in which AM3 polyol chains always interact with
lipid polar headgroups, and consequently the lipid monolayer
continuously curves from the outer leaflet to the inner in the
1
19
infrared spectrometer. H and F NMR spectra were recorded on
a JEOL GSX500 or ECA500 spectrometer. ESI-MS spectra were
measured on an LCQ-deca (Thermo Finnigan).
33
fashion of a toroidal hole. The aforementioned larger pore size
of AM3 as compared with AmB is consistent to the toroidal
pore formation, because toroidal pores are generally said to be
+
-
6-F-ergosterol. To a solution of n-Bu
4
N Ph
3
SiF
2
(642 mg,
34
larger than barrel-stave pores. If the toroidal pore formation is
true for AM3, it is likely that the interaction of AM3 and sterol
predominantly occurs between the polar regions. In fact, in the
1.19 mmol, 2.0 eq) and TiF (147 mg, 1.19 mmol, 2.0 eq) in
4
CH
2
Cl
2
(1.0 mL) was added a solution of epoxide 2 (271 mg,
◦
0.60 mmol) in CH
2
Cl
2
(2.0 mL) at 0 C. After 40 min of stirring, the
7
previous paper we have revealed that esterification of 3-OH of
reaction mixture was quenched with saturated aqueous NaHCO
and extracted with AcOEt. The organic layer was washed with
saturated aqueous NaCl, dried over anhydrous MgSO , filtered
3
,
cholesterol almost eliminate the AM3 activity. The paper also
reported that the AM3 activity was efficaciously enhanced not only
4
7
by cholesterol and ergosterol but also by 25-hydroxycholesterol,
and concentrated. The residue was purified by flash column
chromatography (hexane–AcOEt = 15/1) to afford fluorohydrin
3 (108 mg, 0.228 mmol, 38%) and conjugated diene 6 (121 mg,
which commonly share the A-ring structure in steroid skeleton.
Taken together, it can be reasonably assumed that the molecular
1
440 | Org. Biomol. Chem., 2011, 9, 1437–1442
This journal is © The Royal Society of Chemistry 2011

View Article Online
0
.265 mmol, 45%). To a solution of 3 (17.8 mg, 37.5 mmol)
(10 mm ¥ 150 mm, GE Healthcare Bio-Sciences AB, Uppsala,
Sweden) with 0.15 M potassium phosphate buffer (pH 5.8).
The lipid concentration in liposome suspension was determined
by HPLC (SCL-10AVP, Shimadzu) with an evaporative light
scattering detector (ELSD-LT, Shimadzu). Liposome suspension
was diluted with the same buffer to lipid concentration of 0.5 mM.
To monitor the potassium–proton exchange, AmB in DMSO–
MeOH and a 0.5 ml aliquot of 1 mM FCCP was added to 200 ml
and pyridine (0.3 mL, 3.75 mmol, 100 eq) in CH Cl (1.0 mL)
was added a solution of SOCl
CH
1
Et
and washed with 0.2 M HCl, saturated aqueous NaHCO
saturated aqueous NaCl, dried over anhydrous MgSO , filtered
and concentrated. The residue containing 4 was used for next
step without further purification. A solution of the crude 4 in
2
2
2
(13.7 mL, 18.7 mmol, 5.0 eq) in
(0.5 mL) at -30 C and allowed to warm to 0 C. After
h of stirring at 0 C, the reaction mixture was diluted with
◦
◦
2
Cl
2
◦
2
O and quenched with H
2
O. The organic layer was separated
3
, and
4
◦
of liposome suspension, which was then incubated for 3 h at 6 C.
The suspension (200 ml) was added to a cuvette containing 1.8 ml
THF (0.5 mL) and MeOH (0.5 mL) was added solid K
2
CO
3
of K
2
HPO (0.15 M) that was set to a fluorescence spectrometer
4
(
10.4 mg, 75 mmol, 2.0 eq), and stirred for 1 h at room temperature.
in advance, and the time course of pH change in the liposomes
was monitored at 6 C by the increase in the fluorescent intensity
(excitation 500 nm and emission 535 nm). The initial pH of the
liposome suspension was pH 7.8 while that of liposome lumen was
◦
The solvent was removed under reduced pressure and the residue
was purified by silica gel column chromatography to afford 6-F-
ergosterol (6.5 mg, 15.7 mmol, 42% for two steps): pale yellow
amorphous solid; m.p. 158.0–159.0 C; R
5
◦
f
0.18 (hexane–AcOEt =
pH 5.8. The background fluorescence change (I
the extrapolation of the tangent of the chart. The exchange rate r
(%/s) was calculated from the tangent as r = (I - I
- I ) t where t is the time after the addition of the liposomes (sec)
and the I is the fluorescent intensity at t. After the time-course
0
) was obtained by
2
1
/1); [a]D -119.1 (c 0.42, C
6
H
6
); IR (KBr) n 3434, 2956, 2871,
-
1
1
677, 1459, 1369, 1255, 1128, 1058, 1008, 995, 970, 844, 806 cm ;
t
0
) ¥ 100/(I100
1
H NMR (500 MHz, CDCl
.20 (1H, dd, J = 15.2, 7.5 Hz), 5.15 (1H, dd, J = 15.2, 7.5 Hz), 3.59
1H, dddd, J = 11.2, 11.2, 4.9, 4.9 Hz), 2.96 (1H, ddd, J = 14.6, 4.8,
3
) d 5.38 (1H, ddd, J = 8.6, 2.6, 2.6 Hz),
0
5
(
1
1
0
t
measurements, a valinomycin aqueous solution (10 ml, 5 mM)
was added to the suspension to obtain the fluorescent intensity
corresponding to the 100% exchange (I100).
.7 Hz), 2.07–1.95 (3H, m), 1.90–1.80 (5H, m), 1.78–1.56 (6H, m),
.50–1.42 (2H, m), 1.40–1.21 (5H, m), 1.01 (3H, d, J = 6.9 Hz),
.95 (3H, s), 0.90 (3H, d, J = 6.9 Hz), 0.81 (3H, d, J = 7.1 Hz), 0.80
1
3
Calcein leakage experiments of AM3. Large unilamellar vesi-
cles (LUVs) were prepared as follows. POPC (20 mg) with or
(
3H, d, J = 7.1 Hz), 0.60 (3H, s); C NMR (125 Hz, CDCl
3
) d
1
3
1
1
4
2
51.9 (d, JCF = 241.5 Hz), 145.1 (d, JCF = 10.4 Hz), 135.2, 132.0,
13.1 (d, JCF = 9.3 Hz), 112.6 (d, JCF = 36.4 Hz), 69.6, 55.6, 54.5,
6.2, 42.8, 40.4, 38.8, 38.5, 36.9, 33.1, 31.4, 30.4 (d, JCF = 4.7 Hz),
8.2, 22.8, 21.1, 20.9, 20.0, 19.6, 17.6, 16.7, 12.1; F NMR (470.4
2
2
without 5 mol% sterol was dissolved in 3 mL of CHCl . After
3
◦
3
evaporation of CHCl at 30 C under vacuum for 2 h, it was
3
1
9
dried in vacuo overnight. The lipid film was suspended in 3 mL
of 1 mM Tris–HCl (pH 7.5) containing 60 mM calcein and
agitated for 30 min. A freeze-thaw cycle was repeated three times to
obtain multilamellar vesicles (MLV). Subsequently, the suspension
was passed through a polycarbonate membrane filter (pore size,
MHz, DMSO-d
6
) d -126.2 (br d, J = 8.0 Hz); MS (ESI) m/z 415
+
[
M+H] .
Assays
2
00 nm) 19 times using a Liposofast extruder (Avestin Inc.,
Chemistry. AmB, carbonyl cyanid-p-trifluoro-methoxy-
phenyl hydrazone (FCCP), 20,70-bis(carboxyethyl)-4 or 5-
carboxyfluorescein (BCECF), valinomycin, and calcein were
purchased from Nacalai Tesque (Kyoto, Japan). 1-Palmitoyl-
Ottawa, Canada) above 5 degree of T . The resultant calcein-
m
entrapping LUVs were separated from the excess amount of
calcein by gel filtration using Sepharose 4B (GE Healthcare Bio-
Sciences AB, Uppsala, Sweden) with 10 mM Tris–HCl buffer
2
-oleoyl-sn-glycero-3-phosphocholine (POPC) was obtained
(pH 7.5) containing 1 mM EDTA and 150 mM NaCl. The lipid
from Avanti Polar Lipids, Inc. (Alabaster, AL). Cholesterol and
ergosterol were purchased from Sigma–Aldrich (St. Louis, MO).
and cholesterol concentration in the LUV fraction were measured
using a phospholipid C-Test (Wako Pure Chemical Industries, Ltd,
Osaka, Japan) and cholesterol E-Test Wako (Wako Pure Chemical
Industries, Ltd, Osaka, Japan), respectively. The resulting stock
6
AM3 was isolated as previously reported. All the other chemicals
were standard and analytical quality reagents. Polycarbonate
filters were from Nuclepore (Pleasanton, CA). Fluorescence
spectra were recorded on a JASCO FP-6600 fluorophotometer.
◦
solution was stored at 4 C under nitrogen gas. To monitor calcein
leakage from LUV, 20 ml of the LUV suspension in a cuvette was
diluted with 980 ml of the same buffer to give a lipid concentration
of 27 mM. A 20-ml aliquot of AM3 in MeOH was added to the LUV
suspension. The time courses of liposome leakage were measured
on a spectrofluorometer with excitation at 490 nm (slit 1.5 nm),
+
K Flux assays of AmB. POPC and sterol were dissolved in
chloroform or in methanol to prepare stock solutions. A series of
liposomes was prepared by adding aliquots of the stock solutions
into round-bottom flasks. The solvent was evaporated to form
lipid films at the bottom of flask, which were then left under
vacuum for 6 h to completely remove the solvent. Lipid films thus
obtained were hydrated with phosphate buffer (0.15 M potassium
◦
emission at 517 nm (slit 5 nm) at 25 C in a final volume of 1 ml
buffer with vesicles. Triton X-100 (10%, 20 ml) was added to obtain
the condition corresponding to the 100% leakage.
-
1
phosphate and 40 mg ml BCECF, pH 5.8) by sonication, vortex
mixing, and subsequently five cycles of frozen/thawed to furnish
large vesicles. After the sizing of the liposomes using Liposofast
extruder (Avestin Inc., Ottawa, Canada) by filtering 31 times
through a polycarbonate filter of 200 nm pore size, the resultant
BCECF-entrapping LUV liposomes were separated from the
excess amount of BCECF by gel filtration using Sepharose 4B
Acknowledgements
We are grateful to Dr Respati T. Swasono of our laboratory
for providing AM3 sample and to Mr. Tetsuro Takano of our
laboratory for his cooperation in the AmB assay. This work
was supported by Grants-in-Aid for Scientific Research (S) (No.
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 1437–1442 | 1441

View Article Online
1
8101010), for Priority Area (A) (No. 16073211), for Young
Scientists (A) (No. 17681027), and for Scientific Research (B) (No.
0310132) from MEXT, Japan.
13 N Matsumori, Y. Umegawa, T. Oishi and M. Murata, Bioorg. Med.
Chem. Lett., 2005, 15, 3565–3567.
1
1
1
4 H. Tsuchikawa, N. Matsushita, N. Matsumori, M. Murata and T. Oishi,
Tetrahedron Lett., 2006, 47, 6187–6191.
5 N. Matsumori, Y. Kasai, T. Oishi, M. Murata and K. Nomura, J. Am.
Chem. Soc., 2008, 130, 4757–4766.
6 Y. Umegawa, N. Matsumori, T. Oishi and M. Murata, Biochemistry,
2008, 47, 13463–13469.
2
References
1
G. A. Gallis, R. H. Drew and W. W. Pickard, ReV. Infect. Dis., 1990,
2, 308–329.
1
17 A. G. M. Barrett, D. H. R. Barton and T. Tsushima, J. Chem. Soc.,
Perkin Trans. 1, 1980, 639–642.
18 G. D. Flesch and H. J. Svec, J. Am. Chem. Soc., 1958, 80, 3189–3191.
19 D. P. Michaud, N. T. Nashed and D. M. Jerina, J. Org. Chem., 1985,
50, 1835–1840.
2
3
S. C. Hartsel and J. Bolard, Trends Pharmacol. Sci., 1996, 12, 445–449.
S. Y. Ablordeppey, P. C. Fan, J. H. Ablordeppey and L.
Mardenborough, Curr. Med. Chem., 1999, 6, 1151–1195.
H. Vanden-Bossche, F. Dromer, I. Improvisi, M. LozanoChiu, J. H.
Rex and D. Sanglard, Med. Mycol., 1998, 36, 119–128.
G. K. Paul, N. Matsumori, K. Konoki, M. Sasaki, M. Murata and
K. Tachibana, In Harmful and Toxic Algal Blooms, Proceedings of
the seventh International Conference on Toxic Phytoplankton, 503,
Sendai, July, 1995.
4
5
20 A. Hedhli and A. Baklouti, J. Fluorine Chem., 1995, 70, 141–144.
21 (a) Y. S. Choe and J. A. Katzenellenbogen, Steroids, 1995, 60, 414–422;
(b) B. Ruan, W. K. Wilson, J. Pang and G. J. Schroepfer, Jr., Steroids,
2000, 65, 29–39.
22 K. Mikami, S. Ohba and H. Ohmura, J. Organomet. Chem., 2002, 662,
6
7
8
9
G. K. Paul, N. Matsumori, K. Konoki, M. Murata and K. Tachibana,
J. Mar. Biotechnol., 1997, 5, 124–128.
N. Morsy, T. Houdai, K. Konoki, N. Matsumori, T. Oishi and M.
Murata, Bioorg. Med. Chem., 2008, 16, 3084–3090.
M. Murata, S. Matsuoka, N. Matsumori, G. K. Paul and K. Tachibana,
J. Am. Chem. Soc., 1999, 121, 870–871.
R. T. Swasono, R. Mouri, N. Morsy, N. Matsumori, T. Oishi and M.
Murata, Bioorg. Med. Chem. Lett., 2010, 20, 2215–2218.
77–82.
23 (a) B. Doboszewski, G. W. Hay and W. A. Szarek, Can. J. Chem., 1987,
65, 412–419; (b) S. V. Ley, M. Parra, A. J. Redgrave, F. Sternfeld and
A. Vidal, Tetrahedron Lett., 1989, 30, 3557–3560.
24 A. S. Pilcher, H. L. Ammon and P. DeShong, J. Am. Chem. Soc., 1995,
117, 5166–5167.
25 Y. Itoh, S. Jang, S. Ohba and K. Mikami, Chem. Lett., 2004, 33, 776–
777.
1
1
0 N. Matsumori, K. Tahara, H. Yamamoto, A. Morooka, M. Doi, T.
Oishi and M. Murata, J. Am. Chem. Soc., 2009, 131, 11855–11860.
1 (a) N. Matsumori, N Yamaji, S. Matsuoka, T. Oishi and M. Murata, J.
Am. Chem. Soc., 2002, 124, 4180–4181; (b) N. Yamaji, N Matsumori,
S. Matsuoka, T. Oishi and M. Murata, Org. Lett., 2002, 4, 2087–2089;
26 E. M. Burgess, H. R. Jr, Penton and E. A. Taylor, J. Org. Chem., 1973,
38, 26–31.
27 T. Takano, K. Konoki, N. Matsumori and M. Murata, Bioorg. Med.
Chem., 2009, 17, 6301–6304.
28 T. Houdai, S. Matsuoka, N. Matsumori and M. Murata, Biochim.
Biophys. Acta, Biomembr., 2004, 1667, 91–100.
29 T. Katsu, S. Okada, T. Imamura, K. Komagoe, K. Masuda, T. Inoue
and S. Nakao, Anal. Sci., 2008, 24, 1551–1556.
30 B. DeKruijff and R. A. Demel, Biochim. Biophys. Acta, Biomembr.,
1974, 339, 57–70.
31 S. Clejan and R. Bittman, J. Biol. Chem., 1985, 250, 2884–2889.
32 C. Charbonneau, I. Fournier, S. Dufresne, J. Barwicz and P. Tancrede,
Biophys. Chem., 2001, 91, 125–133.
(
2
c) N. Matsumori, R. Masuda and M. Murata, Chem. Biodiversity,
004, 1, 346–352; (d) Y. Umegawa, N. Matsumori, T. Oishi and M.
Murata, Tetrahedron Lett., 2007, 48, 3393–3396; (e) M. Murata, Y.
Kasai, Y Umegawa, N. Matsushita, H. Tsuchikawa, N. Matsumori
and T. Oishi, Pure Appl. Chem., 2009, 81, 1123–1129.
1
2 (a) N. Matsumori, N Eiraku, S. Matsuoka, T. Oishi, M. Murata, T.
Aoki and T Ide, Chem. Biol., 2004, 11, 673–679; (b) N. Matsumori, Y.
Sawada and M. Murata, J. Am. Chem. Soc., 2005, 127, 10667–10675;
(
1
c) N. Matsumori, Y. Sawada and M. Murata, J. Am. Chem. Soc., 2006,
28, 11977–11984; (d) R. Mouri, K. Konoki, N. Matsumori, T. Oishi
33 T Houdai, N. Matsumori and M Murata, Org. Lett., 2008, 10, 4191–
4194.
and M. Murata, Biochemistry, 2008, 47, 7807–7815; (e) Y. Kasai, N.
Matsumori, Y. Umegawa, S. Matsuoka, H. Ueno, H Ikeuchi, T. Oishi
and M. Murata, Chem.–Eur. J., 2008, 14, 1178–1185.
34 (a) L. Yang, T. A. Harroun, T. M. Weiss, L. Ding and H. W. Huang,
Biophys. J., 2001, 81, 1475–1485; (b) J. H. Lin and A. Baumgartner,
Biophys. J., 2000, 78, 1714–1724.
1
442 | Org. Biomol. Chem., 2011, 9, 1437–1442
This journal is © The Royal Society of Chemistry 2011