Effects of chemical modification of sphingomyelin ammonium group on formation of liquid-ordered phase

Article abstract of DOI:10.1016/j.bmc.2012.05.015

Sphingomyelin (SM) and cholesterol form microdomains called lipid rafts in cellular membranes. To develop a versatile fluorescent lipid probe, chemical modifications to both the hydrophobic and hydrophilic portions of SM are essential. Few reports describing SM probes with a fluorophore at the polar head group have been published. This study examined the effect of substitution on an ammonium moiety of SM on the membrane properties of SM. Two SM analogs with small propargyl and allyl groups on the quaternary nitrogen atom were synthesized and subjected to analysis using differential scanning calorimetry, fluorescent anisotropy, detergent solubilization, surface pressure, and density measurements. Results demonstrated that the two SM analogs retained the membrane properties of SM, including formation of an ordered phase and the ability to interact with cholesterol. A dansyl-substituted SM was prepared for fluorescent measurements. Dansyl-SM showed less of a propensity to form microdomains. These findings imply the potential application of N-substituted SMs as a raft-specific molecular probe.

Full text of DOI:10.1016/j.bmc.2012.05.015

Bioorganic & Medicinal Chemistry 20 (2012) 4012–4019
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Effects of chemical modification of sphingomyelin ammonium group
on formation of liquid-ordered phase
Sarah A. Goretta a, Masanao Kinoshita b, Shoko Mori a, Hiroshi Tsuchikawa a, Nobuaki Matsumori a,
Michio Murata a,b,
⇑
a Department of Chemistry, Graduate School of Science, Osaka University, 1-1 Machikaneyama, Toyonaka, Osaka 560-0043, Japan
b JST ERATO, Lipid Active Structure Project, 1-1 Machikaneyama, Toyonaka, Osaka 560-0043, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Sphingomyelin (SM) and cholesterol form microdomains called lipid rafts in cellular membranes. To
develop a versatile fluorescent lipid probe, chemical modifications to both the hydrophobic and hydro-
philic portions of SM are essential. Few reports describing SM probes with a fluorophore at the polar head
group have been published. This study examined the effect of substitution on an ammonium moiety of
SM on the membrane properties of SM. Two SM analogs with small propargyl and allyl groups on the qua-
ternary nitrogen atom were synthesized and subjected to analysis using differential scanning calorime-
try, fluorescent anisotropy, detergent solubilization, surface pressure, and density measurements. Results
demonstrated that the two SM analogs retained the membrane properties of SM, including formation of
an ordered phase and the ability to interact with cholesterol. A dansyl-substituted SM was prepared for
fluorescent measurements. Dansyl-SM showed less of a propensity to form microdomains. These findings
imply the potential application of N-substituted SMs as a raft-specific molecular probe.
Received 27 March 2012
Revised 8 May 2012
Accepted 8 May 2012
Available online 15 May 2012
Keywords:
Lipid raft
Sphingomyelin
Cholesterol
Lipid ordered phase
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
part of the lipid that is recognized by SBD proteins. In addition, an
intrinsic fluorophore such as tryptophan is preferable for locating
Lipid rafts are microdomains enriched in sphingomyelin (SM),
cholesterol (Chol), and proteins, and form a liquid-ordered phase
that is resistant to solubilization by Triton X-100.1 These mem-
brane domains play important roles in signal transduction,2 and
so membrane biophysical studies require them to be visualized
either by using a fluorescent lipid probe or by fluorescent proteins
that interact specifically with sphingomyelin-binding domains
(SBD). However, current lipid probes that possess a fluorophore
in the hydrophobic acyl portions can alter the distribution and
dynamics of lipids,3 hence making the observation of lipid rafts a
challenge. Although a few fluorescent-labeled SMs in the polar
head have been reported,4 their membrane properties, particularly
those of raft formation, have not been characterized.
Detection of intermolecular interactions between SMs and
specific SBD-proteins could produce information essential for
understanding the cellular functions of lipid rafts. Fluorescent
resonance energy transfer (FRET) is a promising technique for
detecting such interactions occurring in nanoscale domains. To
observe the original protein–lipid interactions occurring on the
surface of biomembranes through FRET experiments,5 a relatively
small fluorescence label needs to be introduced into the hydrophilic
the residues responsible for lipid recognition.
This study describes the synthesis of SM analogs through mod-
ification of a trimethylammonium moiety, and characterizes their
raft-forming properties for evaluating the possibility of developing
a fluorescence-labeled SM probe of physiological relevance.
2. Results
2.1. Design and synthesis of SM probes
For conversion from L-serine to a protected sphingosine or its
equivalents, a method reported by Blot et al.6 (or partly by others7)
was basically adapted with some modifications (Scheme 1). Briefly,
the stereocontrolled condensation of 1-pentadecyne with Garner
aldehyde 5 was followed by Birch reduction as the key steps, lead-
ing to the protected sphingosine 6. Modification of the method by
Sandbhor4 provided a cyclic phospholane and subsequent ring
opening with N,N-dimethyl-N-propargylamine yielded the pro-
tected lyso-SM 8 in a single step. A possible improvement in this
step is formation of the head group under non-acidic conditions,
which simplifies the scheme considerably by eliminating the need
for several steps for the conversion to azide and regeneration of an
amino group; previously,4 they used a 2-azide derivative of 7 for
phosphocholine substitution reaction probably due to treatment
⇑
Corresponding author. Tel./fax: +81 66850 5774.
E-mail address: murata@chem.sci.osaka-u.ac.jp (M. Murata).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.05.015

S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
4013
with TMSOTf. In the present Scheme 1, N-Boc intermediate 7 di-
rectly derived from N-Boc-Lys could be subjected the reaction.
After removal of Boc and PMB protection groups, treatment with
stearic nitrophenyl ester led to progargyl-SM 2; synthesis of 2 from
sphingosine has been reported by the same authors.4 Use of a dan-
syl fluorophore that has an excitation wavelength of 340 nm was
motivated by the possibility to measure FRET with a tryptophan
residue that emits fluorescence ca. 340 nm in SBD proteins such
as lysenin. The azide functionality was introduced in the fluoro-
phore for click chemistry with the propargyl group of 2 using the
method of Kim et al. (Scheme 1).8 The final click coupling pro-
ceeded smoothly to furnish dansyl-SM 4. A similar click reaction
using 2 has been reported for another fluorophore by Sandbhor
et al.4 An allyl group was chosen as another tether group because
metathesis reactions, which usually involve terminal olefins spe-
cifically, can be used to introduce a labeling group. Use of allyl-
dimethylamine facilitated the introduction of a modified choline
moiety in step e of Scheme 1, as compared with its propargyl coun-
terpart. The SM analogs 2, 3, and 4 were efficiently prepared and
their membrane properties evaluated.
Fluorescent anisotropy experiments, done to evaluate the order
of membrane lipids, were conducted with membrane preparations
consisting of 100% SSM, 2 and 3 to determine the influence of
ammonium functionality modification on membrane properties
(Fig. 1). As with the Tm measurements, hydrated bilayers composed
entirely of 2 and 3 showed an order similar to that of SMM, as indi-
cated by temperature-dependent changes in the fluorescence
anisotropy r of membrane-bound diphenylhexatriene. Slight shifts
in the temperature for inflection points are due to the lower Tm val-
ues of 2 and 3 as compared with that of 1.
2.3. Detergent solubilization of SSM analogs
Detergent solubilizing assays were conducted for SM analogs 2,
3, and 4 and compared to that of SSM, 1. After treatment with 0.5%
Triton-X, formation of lipid microdomains was evaluated by per-
centage of retained light scattering, measured as optical density
at 400 nm. As shown in Figure 2, allyl- and propargyl-substituted
analogs 2 and 3 gave rise to percentages similar to that of SSM in
the presence of 50% cholesterol (Chol), which is supposed to form
an SM raft-like ordered phase. Upon addition of 33% palmitoyl
oleoyl phosphatidyl choline (POPC) and 33% Chol, SSM underwent
phase separation, leading to formation of an SM/Chol-rich Lo phase
and a POPC-rich Ld phase. Under these conditions, analogs 2 and 3
showed significant formation of insoluble domains as in the case
with an SSM/Chol system (Fig. 2). In contrast, dansyl-substituted
4 produced a significantly lower amount of detergent-insoluble do-
mains both in SM-Chol and SM-PC-Chol systems. However, as com-
pared with POPC-Chol systems, which usually produce values of
less than 5% scattering,9 4 retained some ability to form a lipid or-
dered phase.
2.2. Measurement of phase transition temperature and order of
membranes with analogs
Phase transition temperature Tm, a basic property of membrane
lipids, was measured by differential scanning calorimetry (DSC) for
stearoylSM (SSM) and analogs 2 and 3 as their hydrated bilayer
preparations. The Tm values of 1, 2, and 3 were 42.5, 40.6, and
39.5 °C, respectively (DSC traces are provided as Supplementary
data). Analogs 2 and 3 possessed Tm values similar to that of SSM
1, implying that a small substituent on the ammonium moiety
did not greatly alter the intermolecular interactions in a membrane
form. We attempted to prepare bilayer vesicles from dansyl ana-
logue 4, but it was found that 4 fails to form stable bilayers in
the absence of Chol. Thus, we carried out fluorescent anisotropy
and specific volume experiments using the bilayer preparations
of 1, 2 and 3 (not 4). Instead, we performed surface pressure mea-
surements using the monolayers of 4 in comparison with those of
1, 2 and 3.
2.4. Interactions between SM analogs and cholesterol
We examined intermolecular interaction of each SM analogue
with Chol. First, the surface pressure versus molecular area (p–A)
isotherms for pure SM, pure Chol and SM/Chol monolayers at
25.0 0.1 °C are shown in Figure 3. The isotherm for the pure 1
monolayer showed
a
low slope region (p = 10–20 mN/m)
OH
O
O
N
P
O
O
HN
O
1: StearoylSM (SSM)
O
O
O
O
N
P
N
P
O
O
O
O
2: Propagyl-SSM
3: Allyl-SSM
N
N
N
O
O
H
N
N
N
P
S
O
O
O
O
: stearoylceramide
4: Dansyl-SSM
Structures. SSM 1 and analogs 2, 3, and 4.

4014
S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
OH
OPMB
CHO
NBoc
a, b
c, d
CO2Me
NHBoc
O
C13H27
HO
HO
C13H27
O
NBoc
NHBoc
5
6
7
OH
O
OPMB
O
O
O
O
f
e
N
P
N
N
P
O
O
C13H27
O
O
C13H27
HN
C17H35
NHBoc
2
8
g
OH
O
N
N
O
O
O
H
N
N
N
P
O
C13H27
C17H35
S
O
O
HN
4
Scheme 1. Synthesis of propargyl SM 2 and dansyl SM 4 from Garner aldehyde. Reagents and conditions: (a) n-BuLi, 1-pentadecyne, THF, ꢀ20 °C, 3 h, 65%; (b) Li/NH3, THF,
ꢀ78 °C, 72 h, 66%; (c) NaH, PMBCl, TBAI, THF, 0 °C to rt, 20 h, 83%; (d) LiCl, AcOH/H2O, rt, 2 h, 82%; (e) (1) 2-chloro-2-oxo-1,2,3-dioxaphospholane, DMAP, Et3N, Tol, 0 °C,
45 min, (2) N,N-dimethyl-N-propargylamine, MeCN, 80 °C, 48 h, 40%; (f) (1) TFA, CH2Cl2, 0 °C, 30 min, (2) 4-nitrophenylstearate, Et3N, DMAP, THF, rt, 23 h, 66%; (g) N-(3-
azidopropyl)-dansyl amine, CuSO4ꢁ5H2O 10 mol %, sodium ascorbate 30 mol %, t-BuOH/H2O 4:1, rt, 16 h, 57%.
Figure 1. Temperature-dependent changes of fluorescent anisotropy (r) of SSM 1 (red, top) and its analogs 2 (green, middle), 3 (blue, bottom).
Figure 2. Detergent solubilization activities of SSM 1 and analogs 2, 3, and 4 as measured by percentage of retained light scattering (y-axis) compared to those prior to Triton-
X treatment. Ch: cholesterol; PC: POPC. Error bars were derived from three experiments.

S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
4015
Figure 3. The surface pressure versus molecular area isotherm (p–A isotherm) of (1) 1/Chol, (2) 2/Chol, (3) 3/Chol and (4) 4/Chol mixed monolayers at 25 0.1 °C. The xChol
values were directly described in the figure.
corresponding to the phase transition between liquid expanded
(LE) and liquid condensed (LC) phases (the solid line in Panel 1 of
Fig. 3). With increasing molar fraction of Chol (xChol), the isotherm
shifted toward the lower molecular area, changing its shape. The
mixing energy of ideal particles (
Gibbs energy of mixing ( Gmix) because
molecular species. In all SM/Chol mixtures,
to be negative in all composition range (see the bottom panel in
Fig. 4). Thus, SSM and SM analogues preferentially contact with
Chol to reduce the mixing energy. In addition, 3/Chol and 4/Chol
D
Gid) was subtracted from the
Gid is independent of
Gex values were found
D
D
D
p–A isotherms of 2/Chol, 3/Chol and 4/Chol mixtures showed sim-
ilar xChol-dependence to the 1/Chol mixtures (Panels 2 and 3 of
Fig. 3). It should be noted that the low slope region was not ob-
mixtures gave lowest DGex values at xChol = 0.5, which is slightly
served in the
p
–A isotherm of pure 4 monolayer (solid line in Panel
higher than that in 1/Chol mixtures (compare squares and crosses
with circles in Fig. 4). These results indicate that the 3 and 4 pro-
vide larger capacity for Chol than 1. It is speculated that the rela-
tively large head groups of these analogues enhance the umbrella
effect, providing the large capacity for Chol (shown as the molecu-
lar areas of pure components in Fig. 4). However, the mixing en-
ergy of 4/Chol mixtures is larger than that in the other SM/Chol
mixtures, indicating that the interaction of Chol with 4 is some-
what weaker than that with the other SMs (crosses in Fig. 4).
Apparently, these results are in line with the detergent solubility
test, which showed the slightly higher solubility of 4/Chol mixture
than the other SM/Chol mixtures (Fig. 2).
4 of Fig. 3). Probably the monolayer stably forms the LE phase over
all pressure range.
To analyze interaction between SM and Chol molecules we plot-
ted the mean molecular areas Ames as a function of xChol (Fig. 4).
Here, we focused on the surface pressure at 30 mN/m because
monolayers at this pressure have often been used as a model for
biomembranes.10 In the 1/Chol mixtures, the deviations of Ames
from area additivity expressed by Eq. 1 were negative in all compo-
sition range, indicating that attractive interaction (condensation)
occurs between 1 and Chol molecules (Panel 1 of Fig. 4). The 2/
Chol, 3/Chol and 4/Chol monolayers also showed smaller Ames val-
ues than the area additivity in almost all composition range (Panels
2–4 of Fig. 4). Thus, these SM analogues attractively interact with
Chol.
We next examined the properties of bilayers composed of 2/
Chol and 3/Chol in a fluid phase (50 °C) by measuring the density
of membrane preparations while changing the molar fraction of
Chol xChol from 0 to 0.5. Figure 5a shows that the specific volume
We next qualitatively evaluated the affinity between each SM
and Chol on the basis of the excess mixing energy (
value of SM/Chol mixture at given compositions was calculated by
integrating the A value from = 0–30 according to Eq. 3. The
D
Gex). The
D
Gex
(inverse density)
creases and reaches a plateau at xChol = 0.25. The xChol value at
which reaches a plateau is called the critical point. Both 2/Chol
v of the SSM/Chol mixture decreases as xChol in-
D
p
v

4016
S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
Figure 4. The mean molecular areas Ames of (1) 1/Chol, (2) 2/Chol, (3) 3/Chol and (4) 4/Chol mixtures as a function of xChol at 30 mN/m. The data were obtained from Fig. 3. The
dashed lines show the area additivity (Aid) shown by Eq. 1. Bottom panel shows the excess mixing energy Gex of (circles) 1/Chol, (triangles) 2/Chol, (squares) 3/Chol and
A(=AmesꢀAid) over the surface pressure ( = 0–30 mN/m) according to
D
(crosses) 4/Chol mixtures as a function of xChol at 30 mN/m. The data was calculated by integrating the
D
p
Eq. 3.
and 3/Chol produced the same results, although their critical point
(xChol = 0.30) was slightly greater than that of SSM/Chol (Fig. 5b and
c). To estimate the molecular volumes of SM in SM/Chol mixtures,
the mean volumes vmean of the membrane preparations were cal-
culated and plotted as a function of xChol (Fig. 5d–f). Using specific
SSM (xChol = 0.25). In addition, since the reductions of vSM in 2/Chol
and 3/Chol at the critical point are close to that in SSM/Chol, the
condensation effect of Chol on 2 or 3 should be similar to that on
SSM. The vSM of SSM is less than those of 3 and 4 across the exper-
imental range because the molecular volume of SSM is smaller
than those of the analogs.
volume
v, the vmean is expressed as
fxCholMChol ꢀ ð1 ꢀ xCholÞMSM
gmðxCholÞ
mmean ¼ ðxCholÞ ¼
3. Discussion
NA
This study provides the first example of chemical modification
at the SM polar head and its effect on membrane properties, partic-
ularly membrane ordering effects. As depicted in Figures 1 and 2,
the interactions of analogs 2 and 3 with Chol are very similar to
that of SSM, demonstrating that the ammonium group can be
potentially utilized for introduction of a fluorophore and other
functionalities. A slight difference in the Tm values and fluorescent
anisotropy curves (Fig. 1) of 1 and 2/3 may be due to increased vol-
ume of the head group. The loose packing of pure lipid bilayers
formed by 2 or 3 may weaken intermolecular attraction, leading
to Tm values lower than that of SSM.
where MChol and MSM are molecular weights of Chol and SM,
respectively, and NA is the Avogadro’s number. According to Fig-
ure 5a–c,
functions. The intersection between the linear function and
xChol = 0 is the partial molecular volume of SM SM, indicating the
vmean versus xChol for each SM/Chol mixture fit two linear
v
effective volume of the SM/Chol mixture. A detailed description
of partial molecular volume has been reported previously.11
0
In the SSM/Chol mixture, the vSM value of 1226 ÅA3 under the
0
lower xChol decreased to 1205 ÅA3 at xChol = 0.25 (circles in Fig. 6).
This decrease in vSM at the critical point is attributed to Lo phase
formation in the SSM/Chol bilayers.11 Both the 2/Chol and 3/Chol
Hydrophobic substituents such as a dansyl group in 4 may pre-
vent formation of stable ordered domains as shown by the deter-
mixtures produced similar results: the vSM values of 2 and0 3
0
(1265 and 1277 ÅA3, respectively) decreased to 1242 and 1245 ÅA3,
respectively, at xChol = 0.30 (triangles and squares in Fig. 6). Results
suggest that analogs 2 and 3 interact with Chol at a stoichiometry
of 2:1 (xChol = 0.30), which is approximately the same as that of
gent solubilizing assays and
p–A experiments (Figs. 2–4).
However, 4 interacts with Chol to a certain extent, and partly forms
detergent-resistant domains, implying that, even with a fluores-

S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
4017
Figure 5. Chol concentration dependence of specific volume
v
of: (a) SSM/Chol, (b) 2/Chol, and (c) 3/Chol mixtures at 50 °C (fluid phase). The
v
versus xChol plot for each SM/
Chol mixture fits two linear functions (solid lines), which intersect at critical point xChol = 0.25 for the SSM/Chol mixture and xChol = 0.30 for the 2/Chol and the 3/Chol
mixtures. Chol concentration dependence of the mean molecular volume vmean of: (d) SSM/Chol, (e) 2/Chol, and (f) 3/Chol mixtures at 50 °C. The mean value was estimated
from . According to the versus xChol plot, the vmean fit two linear functions (solid lines) as a function of xChol
v
v
v
.
Thus, a dansyl group or other fluorophore having an excitation
maximum ca. 340 nm is potential target for developing
a
fluorescent lipid probes. Results of this study indicate that 4 may
promote an ordered phase when added as a minor constituent,
such as in the case of a fluorescent-labeled molecular probe. We
thus attempted to observe the distribution of 4 in membrane
containing microdomains using a confocal microscope. Due to
the rapid bleaching of fluorescence of the dansyl group coupled
with its shorter excitation wavelength, however, it has not
been successful so far to observe the localization of 4 on bilayer
membranes. More hydrophilic fluorophores with longer excitation
wave length that could be tagged to the head group of SM or other
lipids via an ammonium group would be a potential raft-specific/
selective probe that could distinguish lipid rafts12 with different
lipid constituents on cellular membranes. Research on the prepara-
tion of these probes is now underway.
Figure 6. Partial molecular volumes of SMs vSM in: SSM/Chol (circles), 2/Chol
(triangles), and 3/Chol (squares) at 50 °C. The vSM value was obtained from the
intersection between linear functions and xChol = 0.
As reported by Yamakoshi et al.13 an alkyne group can be ob-
served selectively in living cells by Raman microscopy. Thus, the
triple-bond-bearing analog 2 may be utilized as a bioorthogonal
‘Raman tag’ for bioimaging. As often seen in membrane biology
and biophysics, a large fluorescent tag inherently disrupts the
physiological functions of biomolecules, particularly smaller mole-
cules such as lipids. Thus, small alkyne groups, including an ethy-
nyl (only 25 Da), may serve as a versatile labeling functionality.
cent group, N-substituted SM retained the ability to form microdo-
main with Chol. Dansyl group are of interest because of its useful-
ness in FRET experiments for examining interactions with a
tryptophan residue, which is thought to play an essential role in
SM membrane recognition. In addition, the aromatic residue exerts
an anchoring effect, which stabilizes membrane proteins so
they can assume the correct position in the bilayer membranes.

4018
S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
temperature. The samples were freeze-thawed three times by cy-
4. Experimental
cling the samples between liquid nitrogen and a water bath main-
tained at 65 °C. Fluorescence polarization measurements were
performed with a JASCO FP-6600 spectrofluorometer equipped
with a JASCO ADP-303 polarization accessory. Quartz cuvettes with
a path length of 1 cm were used. The excitation wavelength was set
at 358 nm and emissions were monitored at 430 nm. Excitation and
emission slits with bandpasses of 1 and 10 nm, respectively, were
used for all measurements. The smallest possible excitation slit
was used to minimize any photoisomerization of DPH during irra-
diation. Measurement temperature was raised from 30 °C to 70 °C
at 2 °C/min, and fluorescence was measured with a 2 s response.
Polarization values were calculated using Spectra Manager soft-
ware attached to the spectrofluorometer.
4.1. Synthesis of SM analogues 2, 3, and 4
See Supplementary data including NMR chemical shift data.
4.2. Materials
Palmitoryl oleoyl phosphatidylcholine (POPC) was purchased
from Avanti Polar Lipids (Alabaster, AL). Cholesterol and Triton
X-100 were purchased from Nakarai Tesque (Kyoto, Japan).
Diphenylhexatriene (DPH) was purchased from Aldrich (St. Louis,
MO). Bovine brain sphingomyelin (SM) was purchased from Avanti
Polar Lipids (Alabaster, AL), and 18-0 SM was purified from the
brain sphingomyelin (SM) using HPLC.
4.6. p–A isotherm measurements of monlayers
4.3. Differential scanning calorimetry (DSC)
Monolayers of SM/Chol mixtures were prepared on a computer-
controlled Langmuir-type film balance (USI System, Fukuoka, Ja-
pan) calibrated by stearic acid. The subphase was distilled and
freshly deionized water from Milli-Q system (Millipore Corp.).
The apparatus was covered with a vinyl sheet, which prohibited
dust from depositing on the water surface. Thirty micro-liters of li-
pid solution (1 mg/mL) were spread onto the aqueous subphase
(100 ꢂ 290 mm2) with a glass micropipette (Drummond Scientific
Company, Pennsylvania, USA). The monolayers were compressed
at a rate of 10–20 mm2/s after the initial delay period of 10 min
for evaporation of organic solvents. The subphase temperature
was controlled to be 25.0 0.1 °C. We repeated the measurements
several times under the same conditions to obtain reliable results.
These measurements gave the molecular areas at the correspond-
ing pressure within the error of 0.01 nm2.
The main transition temperatures of pure SSM and analog 2 and
3 bilayers were examined by DSC. Samples were prepared using a
conventional method. Briefly, the appropriate amount of SSM, 2, or
3 powder was dispersed in 10 mM HEPES buffer (pH 7.6) and incu-
bated at 55 °C for 10 min with intermittent vortexing. Final lipid
concentration of the sample was 13 wt %. Then, 10–15 lL of the
sample (1.3–2.0 mg of lipids total) were enclosed in the aluminum
pane 0219-0062 (Perkin–Elmer, California) and put into the DSC
instrument (Diamond, Perkin–Elmer, California) immediately be-
fore measurements were obtained. The DCS temperature was cali-
brated using representative phospholipid bilayers with known
transition temperatures: DMPC (24.5 °C), DPPC (41.5 °C), and DSPG
(53.5 °C). A scanning rate of 5.0 °C/min was used for all DSC
measurements.
We evaluated the intermolecular interaction between SM and
Chol at the surface pressure of 30 mN/m on the basis of the devia-
tions of experimentally obtained mean molecular areas (Ames) from
those of ideal mixtures (Aid);
4.4. Percent solubilization experiments
Sample preparation for solubility experiments involved drying a
MeOH-CHCl3 solution of 500 nmol of total lipid (SM and sterol) un-
der a stream of argon, followed by redissolution in chloroform, and
then drying again under a stream of argon. After the lipids were
dried further under high vacuum for at least 12 h, they were hy-
drated (swelled) by adding 0.95 mL PBS buffer (pH 7). To uniformly
disperse the lipids and form homogeneous multilamellar lipid ves-
icles, each sample was vigorously vortexed at a temperature of
65 °C, well above the phase transition temperature of SM, and then
cooled to room temperature. For measuring solubilization by the
loss of light scattering, the optical density of these samples was
measured at 400 nm on a Shimadzu UV-2500 spectrophotometer.
Aid ¼ ASMxSM þ ACholxChol
;
ð1Þ
where ASM and AChol are the molecular areas of pure SM and
Chol, and xSM and xChol are molar fractions of SM and Chol, respec-
tively. The value of Aid corresponds to the mean molecular area in
the mixture constituted of non-interactive or completely immisci-
ble molecules. The negative deviation of Ames from Aid
(
D
A ¼ Ames ꢀ Aid < 0) indicates attractive lateral intermolecular
interactions between two molecules.
We calculated the excess Gibbs energy of mixing,
DGex, to defin-
itively evaluate the affinity of the SM analogue and Chol.
Gex Gmix Gid
Gmix ꢀ RTðxSM ln xSM þ xChol ln xCholÞ;
where R is the gas constant and T is absolute temperature. The
mixing energy of ideal particles ( Gid) is subtracted from the Gibbs
energy of mixing ( Gmix) because
species. According to Goodrich14
D
¼
D
ꢀ
D
¼
D
ð2Þ
Then, 50 lL of 10% (w/v) Triton X-100/PBS was added. After mixing
and incubating at 23 °C for 2 h, the optical density was remea-
sured. The ratio of optical density (%OD) after Triton X-100 incuba-
tion (not corrected for dilution with Triton X-100 solution) to that
before the addition of Triton X-100, was then calculated.
D
D
D
D
Gid is independent of molecular
Gex was calculated as an integral
of the deviation,
D
A, over the surface pressure p;
Z
Z
4.5. Fluorescent polarization measurements
p
p
D
Gex
¼
ð
D
AÞd
p
¼
ðAmes ꢀ ASMxSM ꢀ ACholxCholÞd
p
ð3Þ
0
0
Multilamellar vesicles of lipids containing 2 mol % DPH were
prepared by drying a MeOH–CHCl3 solution containing 160 nmol
of total lipids (phospholipids and sterol) and 3.2 nmol of DPH under
a stream of argon. The residue was redissolved in CHCl3, and dried
again under a stream of argon. After the lipids were dried further
under high vacuum for at least 12 h, they were hydrated (swelled)
by adding 3 mL of PBS buffer (pH 7.0). To uniformly disperse the lip-
ids and form homogeneous multilamellar lipid vesicles, each sam-
ple was vigorously vortexed at a temperature of 65 °C, well above
the phase transition temperature of SM, and then cooled to room
4.7. Density measurements
According to Nagle and others, the neutral floatation method al-
lows the specific volume (inverse density) of a bilayer sample to be
estimated using H2O/D2O mixtures as a solvent.15 Theoretically,
the specific volumes of H2O and D2O can be defined as vH2O and
v
D2O, respectively. Thus, the specific volume of solvent vsol is de-
fined as:

S. A. Goretta et al. / Bioorg. Med. Chem. 20 (2012) 4012–4019
4019
vsol
¼
uvD2O þ ð1 ꢀ
uÞ
vH2O
;
References and notes
where is the mass fraction of D2O. If the specific volume of the
u
1. Simons, K.; Ikonen, E. Nature 1997, 387, 569.
lipid vesicle is larger than sol, the vesicle floats in the H2O/D2O
v
v
v
2. Some recent reviews; (a) Bagatolli, L. A.; Ipsen, J. H.; Simonsen, A. C.; Mouritsen,
O. G. Prog. Lipid Res. 2010, 49, 378; (b) Kusumi, A.; Shirai, Y. M.; Koyama-Honda,
I.; Suzuki, K. G. N.; Fujiwara, T. K. FEBS Lett. 1814, 2010, 584; (c) Levental, I.;
Grzybek, M.; Simons, K. Biochemistry 2010, 49, 6305; (d) Elson, E. L.; Fried, E.;
Dolbow, J. E.; Genin, G. M. Annu. Rev. Biophys. 2010, 39, 207.
3. (a) Eggeling, C.; Ringemann, C.; Medda, R.; Schwarzmann, G.; Sandhoff, K.;
Polyakova, S.; Belov, V. N.; Hein, B.; von Middendorff, C.; Schönle, A.; Hell, S. W.
Nature 2009, 457, 1159; (b) Rosseto, R.; Tcacenco, C. M.; Ranganathan, R.;
Hajdu, J. Tetrahedron Lett. 2008, 49, 3500.
solvent. If is smaller than sol, the vesicle sinks in the solvent.
v
The floating/sinking process can be facilitated by centrifugation.
When the sample is obtained in the supernatant with the specific
sup
sol
volume of
v
and the precipitate comes out of solution involving
v , the specific volume of sample v can be deter-
sol
pre
the solvent with
mined using the following equation:
4. Sandbhor, M. S.; Key, J. A.; Strelkov, I. S.; Cairo, C. W. J. Org. Chem. 2009, 74,
8669.
sup
sol
pre
sol
sup
sol
pre
sol
v
þ
v
v
ꢀ
v
v
¼
ꢃ
5. (a) Nyholm, T.; Nylund, M.; Soderholm, A.; Slotte, J. P. Biophys. J. 2003, 84, 987;
(b) Zhao, H.; Jutila, A.; Nurminen, T.; Wickstrom, S. A.; Keski-Oja, J.; Kinnunen,
P. K. J. Biochemistry 2005, 44, 2857; (c) Ran, Y.; Fanucci, G. E. Anal. Biochem.
2008, 382, 132; (d) Smirnova, I. N.; Kasho, V. N.; Kaback, H. R. Biochemistry
2006, 45, 15279; (e) Haldar, S.; Raghuraman, H.; Chattopadhyay, A. J. Phys.
Chem. B 2008, 112, 14075; (f) Ramirez, D. M. C.; Ogilvie, W. W.; Johnston, L. J.
Biochim. Biophys. Acta, Biomembr. 2010, 1798, 558; (g) Cai, Y.; Ling, C. C.; Bundle,
D. R. Org. Biomol. Chem. 2006, 4, 1140.
6. Blot, V.; Jacquemard, U.; Reissig, H.-U.; Kleser, B. Synthesis 2009, 759.
7. (a) Radunz, H. E.; Devant, R. M.; Eiermann, V. Liebigs Ann. Chem. 1988, 1103; (b)
Herold, P. Helv. Chim. Acta 1988, 71, 354; (c) Garner, P.; Park, J. M.; Malecki, E. J.
Org. Chem. 1988, 53, 4395; (d) Nimkar, S.; Menaldino, D.; Merrill, A. H.; Liotta,
D. Tetrahedron Lett. 1988, 29, 3037; (e) Soai, K.; Takahashi, K. J. Chem. Soc.,
Perkin Trans. 1 1994, 1257; (f) Azuma, H.; Tamagaki, S.; Ogino, K. J. Org. Chem.
2000, 65, 3538; (g) Murakami, T.; Furusawa, K. Tetrahedron 2002, 58, 9257;
Chun, J.; Li, G.; Byun, H.-S.; Bittman, R. Tetrahedron Lett. 2002, 43, 375; (i)
Suzuki, H.; Mori, M.; Shibakami, M. Synlett 2003, 2163; (j) Sawatzki, P.; Kolter,
T. Eur. J. Org. Chem. 2004, 3693; (k) Murakami, T.; Furusawa, K.; Tamai, T.;
Yoshikai, K.; Nishikawa, M. Bioorg. Chem. Med. Lett. 2005, 15, 1115; (l) Okamoto,
N.; Kanie, O.; Huang, Y.-Y.; Fujii, R.; Watanabe, H.; Shimamura, M. Chem. Biol.
2005, 12, 677; (m) Reference 6c.; (n) Ohshita, K.; Ishiyama, H.; Takahashi, Y.;
Ito, J.; Mikami, Y.; Kobayashi, J. Bioorg. Med. Chem. 2007, 15, 4910.
8. (a) Kim, H.-Y.; Sohn, J.; Wijewickrama, G. T.; Edirisinghe, P.; Gherezghiher, T.;
Hemachandra, M.; Lu, P.-Y.; Chandrasena, R. E.; Molloy, M. E.; Tonetti, D. A.;
Thatcher, G. R. J. Bioorg. Med. Chem. 2010, 18, 809; (b) Landi, F.; Johansson, C.
M.; Campopiano, D. J.; Hulme, A. N. Org. Biomol. Chem. 2009, 8, 56; (c) Zhang, X.;
Cao, B.; Yu, S.; Zhang, X. Angew. Chem., Int. Ed. 2010, 49, 4047.
9. Schnitzer, E.; Kozlov, M. M.; Lichtenberg, D. Chem. Phys. Lipids 2005, 135, 69.
10. (a) Demel, R. A.; Geurts van Kessel, W. S. M.; Zwaal, R. F. A.; Roelofsen, B.; Van
Deenen, L. L. M. Biochim. Biophys. Acta 1975, 406, 97; (b) Nagle, J. F. J. Membr.
Biol. 1976, 148, 997; (c) Blume, A. Biochim. Biophys. Acta 1979, 577, 32; (d) Feng,
S. S. Langmuir 1999, 15, 998; (e) Zhai, X.; Li, X. M.; Momsen, M. M.; Brockmann,
H. L.; Brown, R. E. Biophys. J. 2006, 91, 2490.
11. Greenwood, A. I.; Tristram-Nagle, S.; Nagle, J. F. Chem. Phys. Lipids 2006, 143, 1.
12. As to raft heterogeneity, for examples, (a) GomezMouton, C.; Abad, J. L.; Mira,
E.; Lacalle, R. A.; Gallardo, E.; Jimenez-Baranda, S.; Illa, I.; Bernad, A.; Manes, S.;
Martinez-A, C. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 9642; (b) Bagnat, M.; Simons,
K. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14183; (c) Pike, L. J. Biochem. J. 2004, 378,
281; (d) Kiyokawa, E.; Baba, T. K.; Otsuka, N.; Makino, A.; Ohno, S.; Kobayashi,
T. J. Biol. Chem. 2005, 280, 24072; (e) Lingwood, D.; Kaiser, H.-J.; Levental, I.;
Simons, K. Biochem. Soc. Trans. 2009, 37, 955.
2
2
Experimentally, a solution containing the appropriate amounts
of SM and Chol co-dissolved into MeOH/CHCl3 (1:4 v/v) (Wako
Pure Chemical Industries, Ltd., Osaka, Japan) was dried under nitro-
gen flow and stored under high vacuum for more than 24 h to re-
move all organic solvent. The dried sample was dispersed into H2O/
D2O (Wako Pure Chemical Industries, Ltd, Osaka, Japan) and incu-
bated at 50 °C for several minutes with intermittent vortexing. Fi-
nal lipid concentration was ꢄ5 mM.
The incubated sample was rapidly transferred into a centrifuge
(Kubota 1910, Kubota Co., Ltd, Tokyo, Japan) and centrifuged
(20,000ꢂg) for 5–20 min at 55 °C, a temperature at which all sam-
ples formed a fluid phase. Sample temperature (K) was measured
directly with a thermometer (AD-5602A Sansyo Industries, Ltd, To-
kyo, Japan) immediately after centrifugation. When the sample
was reused, it was heated and cooled several times around the
main transition temperature after addition of H2O or D2O. Because
voids or packing defects appear in the lipid membranes near the
transition temperature, the heating/cooling cycles achieve homo-
geneous vsol between the outer and the inner portions of a lipid
vesicle.
Acknowledgments
We are grateful to Professor J. Peter Slotte, Åbo Akademi Uni-
versity, Professor Mikiko Sodeoka, Dr. Jun Ando, Dr. Hiroyuki
Yamakoshi, Riken, Professor Tohru Oishi, Kyushu University, and
Dr. Fuminori Satou, this ERATO project for discussions. This work
was supported in part by Grant-In-Aids for Scientific Research (S)
(No. 18101010), and (B) (No. 20310132) from MEXT, Japan and
by a SUNBOR grant from the Suntory Institute for Bioorganic Re-
search. A research fellowship to S.A.G. from the Japan Society for
the Promotion of Science (JSPS) is greatly appreciated.
13. Yamakoshi, H.; Dodo, K.; Okada, M.; Ando, J.; Palonpon, A.; Fujita, K.; Kawata,
S.; Sodeoka, M. J. Am. Chem. Soc. 2010, 133, 6102.
14. Goodrich, F. C. Molecular interaction in mixed monolayer In Second
International Congress on Surface Activity; Schulman, J. H., Ed.; Butterworth &
Co.: London, 1957; Vol. I, p 85.
15. (a) Nagle, J. F.; Wilkinson, D. A. Biophys. J. 1978, 23, 159; (b) Wiener, M. C.;
Tristram-Nagle, S.; Wilkinson, D. A.; Campbell, L. E.; Nagle, J. F. Biochim.
Biophys. Acta 1998, 938, 135; (c) Koenig, B. W.; Gawrisch, K. Biochim. Biophys.
Acta 2005, 1715, 65.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.05.015.