Phospholipid-bound molecular rotors: Synthesis and characterization

Article abstract of DOI:10.1016/S0968-0896(02)00240-7

Molecular rotors are fluorescent molecules with a viscosity-sensitive quantum yield that are often used to measure viscosity changes in cell membranes and liposomes. However, commercially available molecular rotors, such as DCVJ (1) do not localize in cell membranes but rapidly migrate into the cytoplasm leading to unreliable measurements of cell membrane viscosity. To overcome this problem, we synthesized molecular rotors covalently attached to a phospholipid scaffold. Attaching the rotor group to the hydrophobic end of phosphatidylcholine (PC) did not affect the rotor's viscosity sensitivity and allowed adequate integration into artificial bilayers as well as complete localization in the plasma membrane of an endothelial cell line. Moreover, these new rotors enabled the monitoring of phospholipid transition temperature. However, attachment of the rotor groups to the hydrophilic head of the phospholipid led to a partial loss of viscosity sensitivity. The improved sensitivity and exclusive localization in the cell plasma membrane exhibited by the phospholipid-bound molecular rotors suggest that these probes can be used for the study of membrane microviscosity.

Full text of DOI:10.1016/S0968-0896(02)00240-7

Bioorganic & Medicinal Chemistry 10 (2002) 3627–3636
Phospholipid-Bound Molecular Rotors:
Synthesis and Characterization
Mark A. Haidekker,a Thomas Brady,b Ke Wen,b Cliff Okada,a Hazel Y. Stevens,a
Jeniffer M. Snell,a John A. Frangosa and Emmanuel A. Theodorakisb,*
aDepartment of Bioengineering, University of California, San Diego, La Jolla, CA 92093-0412, USA
bDepartment of Chemistry and Biochemistry, University of California, San Diego, La Jolla, CA 92093-0358, USA
Received 28 March 2002; accepted 26 April 2002
Abstract—Molecular rotors are fluorescent molecules with a viscosity-sensitive quantum yield that are often used to measure visc-
osity changes in cell membranes and liposomes. However, commercially available molecular rotors, such as DCVJ( 1) do not
localize in cell membranes but rapidly migrate into the cytoplasm leading to unreliable measurements of cell membrane viscosity.
To overcome this problem, we synthesized molecular rotors covalently attached to a phospholipid scaffold. Attaching the rotor
group to the hydrophobic end of phosphatidylcholine (PC) did not affect the rotor’s viscosity sensitivity and allowed adequate
integration into artificial bilayers as well as complete localization in the plasma membrane of an endothelial cell line. Moreover,
these new rotors enabled the monitoring of phospholipid transition temperature. However, attachment of the rotor groups to the
hydrophilic head of the phospholipid led to a partial loss of viscosity sensitivity. The improved sensitivity and exclusive localization
in the cell plasma membrane exhibited by the phospholipid-bound molecular rotors suggest that these probes can be used for the
study of membrane microviscosity.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
based methods3 that have been used to assess the local
Cellular membranes are remarkable molecular assem-
blies, composed mainly of phospholipids and proteins,
that play a central role in both the structure and the
function of cells.1 Perhaps the best way to visualize a
membrane is to consider the ‘fluid mosaic model’ first
described by Singer and Nicolson.2 In this model the
membrane is pictured as a fluid-like phospholipid
bilayer, into which are embedded proteins. An impor-
tant element of this model is the recognition that the
membrane is a dynamic structure and its fluidity influ-
ences the structure and position of proteins, thereby
affecting the physiological function of cells. Conse-
quently, changes in membrane fluidity have been linked
with alterations in physiological processes, such as car-
rier-mediated transport, activities of membrane-bound
enzymes and receptor binding which, in turn, are asso-
ciated with aging and disease pathogenesis. The impor-
tance of membrane viscosity in cellular biology and
physiology led to the development of several fluorescence-
viscosity of both cell membranes and model membranes
(such as liposomes and vesicles). Among them are inclu-
ded: fluorescence recovery after photobleaching (FRAP),4
fluorescence anisotropy (or fluorescence polarization)5
and the use of environment-sensitive fluorescent probes.6
In the latter category are included compounds such as
9-(dicyanovinyl)-julolidine (DCVJ, 1) (Fig. 1) with a
viscosity-dependent fluorescence quantum yield.
The intriguing fluorescent properties of DCVJ( 1) rest
upon its ability to lose the excited state either by radia-
tion or by intramolecular rotation, the ratio of which
depends on the free-volume of the environment.7 This
ability has defined a new class of molecules commonly
referred to as molecular rotor,8 that have been used as
membrane viscosity sensors in both chemical and bio-
logical processes. While DCVJallows the probing of the
cell membrane viscosity it also localizes in the cell
interior, showing a particular affinity to tubulin struc-
tures.9 Cognizant of this limitation, we developed
recently a more lipophilic analogue of DCVJ, referred
to as FCVJ( 2), in which a farnesyl side chain was
attached to the viscosity-sensitive julolidine fragment.10
*Corresponding author. Tel.: +1-858-822-0456; fax: +1-858-822-
0386; e-mail: etheodor@ucsd.edu
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00240-7

3628
M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
the synthesis of probe 28 in which the julolidine fragment
is connected at the hydrophilic end of a phospholipid.
The synthesis of rotor 22 commenced with esterification
of commercially available 6-bromohexanoyl chloride (3,
n=1) using p-methoxybenzyl alcohol (5) in the presence
of DMA to afford ester 6 in 82% yield (Scheme 1). Treat-
ment of bromide 6 with NaN3 followed by reduction of
the resulting azide 8 with Ph3P gave rise to amino ester
10 in 83% combined yield.11 DCC-induced condensa-
tion of amine 10 with cyanoacetic acid (12) afforded
ester 13 which upon ireatrnent with aldehyde 15 in the
presence of DBU produced compound 16 (two steps,
31% combined yield). Deprotection of the ester func-
tionality of 16 with TFA gave rise to acid 18 in 59%
yield12 and set the stage for attachment to lysophos-
pholipids 20 and 21. This condensation was performed
in the presence of DCC and DMAP and produced
rotors 22 and 23 in 54 and 49% yield, respectively. This
synthetic approach was also used to produce fluorescent
rotor 24 (n=6, m=12).
Figure 1. Structures of DCVJ( 1) and FCVJ( 2).
Although this modification led to a 20-fold improved
signal resolution as a response to a biomechanical stim-
ulus (shear stress) as compared to 1, it did not fully
exclude migration into the cell interior. To further
increase the membrane localization profile of this type
of probes, we synthesized a new family of molecular
rotors, in which the julolidine rotor was attached via a
polymethylene linker to the hydrophobic or hydrophilic
end of a phospholipid structure. Herein, we report the
synthesis, physical and biochemical characterization of
such probes in artificial bilayers and cultured cells.
The synthesis of rotor 28 is shown in Scheme 2. Com-
mercially available dimyristoyl-l-a-phosphatidyl etha-
nolamine (C14:0) (25) was treated with cyanoester 26 in
the presence of DMAP to produce compound 27 which
upon condensation with 15 afforded compound 28
(n=10) in 38% overall yield.
Results
Synthesis of phospholipid-based rotors
To construct the designed probes in which the julolidine
rotor is attached to either end of a phospholipid back-
bone we devised two synthetic strategies as shown in
Schemes 1 and 2. In Scheme 1 is presented the synthesis
of compounds 22, 23, and 24 having the julolidine frag-
ment connected at the hydrophobic end of a phospho-
lipid structure. On the other hand, Scheme 2 illustrates
Physical properties of the synthesized probes
All probes were soluble in various solvents, including
chloroform and methanol. When dissolved in methanol,
these compounds exhibit triple excitation maxima at
320, 396, and 470 nm with a single emission maximum at
Scheme 1. Synthesis of probes 22, 23, and 24 in which the rotor is attached at the hydrophobic end of a phospholipid.

M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
3629
Figure 3. Assessment of the rotor behavior of compounds 22, 23, 24,
and 28. All molecules except compound 28 show a power-law increase
of the measured intensity (lex=470 nm, lem=490 nm) with the visc-
osity of the solvent. With the exception of compound 28, the slopes of
the fitted lines were found to be in the range 0.48–0.53, which is similar
to DCVJ(slope of 0.6).
compound 28) showed the profile of a fluorescent rotor,
characterized by an increase in the intensity of fluores-
cence emission by increasing the viscosity of the solvent
(Fig. 3).12 The results are in agreement with a power law
relationship of quantum yield (ꢀ) and viscosity (Z), as
indicated by the Forster–Hoffmann equation (eq 1).13
Scheme 2. Synthesis of probe 28 in which the rotor is attached at the
hydrophilic end of a phospholipid.
Log ꢀ ¼ C þ x ꢀ log Z
ð1Þ
The constant x in eq 1 is a dye-dependent constant and
was found to be 0.52Æ0.01 for 22, 0.53Æ0.02 for 23,
0.48Æ0.029 for 24, and 0.21Æ0.12 for 28; data given as
meanÆstandard deviation from three independent
experiments. This indicates that probes 22, 23, and 24
(but not 28) exhibit a rotor behavior similar to DCVJ
(1), where the constant x is known to be 0.6.10
Integration of phospholipid-bound molecular rotors in
artificial membranes
Figure 2. Excitation and emission spectra of the phospholipid-bound
molecular rotor 22. Three excitation maxima at 320, 396, and 470 nm
have been observed. All three excitation wavelengths lead to an emis-
sion maximum at 490 nm. The small peak at 620 nm, which is only
visible for 320 nm excitation, is caused by the solvent, methanol.
Compounds 23, 24, and 28 show very similar spectra.
To examine if our designed probes retain their rotor
properties when embedded in cellular membranes or
model membranes, we studied their fluorescent emission
as a function of temperature in liposome bilayers. It is
known that DCVJ( 1), when integrated into an artificial
bilayer consisting of dimyristoyl-phosphatidylcholine
(DMPC), shows an emission intensity which is inversely
related to the temperature.14 This may be explained by
considering that increase of temperature can lead to
both an increase of intramolecular rotation of the rotor
and a decrease of phospholipid viscosity. Similar effects
were observed for compounds 22, 23, and 24. Above the
transition temperature of DMPC at 23 ꢀC, the rate of
decrease of the temperature-dependent intensity is
markedly reduced (Fig. 4, top panel). Between 23 and
25 ꢀC we observed a transition with a steeper gradient:
À0.17% as opposed to 13%, observed below 23 ꢀC and
À4%, observed above 25 ꢀC (slopes have been normal-
ized to the absolute intensity at 40 ꢀC). When using 23,
490 nm. The emission intensity is highest for excitation at
470 nm. Fig. 2 shows the excitation spectrum as well as the
emission spectra for all three excitation maxima for com-
pound 22. Similar spectra were obtained for compounds
23, 24, and 28. All subsequent experiments were per-
formed at lex=470 nm and lem=490 nm.
To verify the rotor properties of the above compounds
we examined the relationship between their quantum
yield of fluorescence emission and solvent viscosity. This
was accomplished using mixtures of ethylene glycol and
glycerol as solvents. Compounds 22, 23, and 24 (but not

3630
M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
gested by the cell detachment (see Fig. 5, top row).
Nonetheless, this detachment is not observed with com-
pounds 23, 24, and 28, indicating that these probes can
be used in cell-based studies.
Discussion
Driven by the need to understand cell membrane func-
tion, various fluorescent-based methods to measure cell
membrane viscosity have been developed. These meth-
ods include: FRAP (fluorescence recovery after photo-
bleaching);5 and fluorescence anisotropy (or fluorescence
polarization).6 Although both methods allow a direct,
quantitative measurement of local viscosity they have,
nonetheless, distinct disadvantages. FRAP requires a
relatively long time for the measurement to allow suffi-
cient fluorescence recovery, limiting both the spatial and
temporal resolution.16 Also, the photobleaching laser
pulse introduces power densities up to 1 MW/cm2 to the
membrane, leading to generation of free radicals by
photolysis and to local heating which in turn damages
the surrounding proteins and decreases the apparent
viscosity due to protein cross-linking.18 On the other
hand, measurement of fluorescence anisotropy is mostly
limited by the need for polarization filters, which absorb
a large amount of the emitted light, thereby reducing
sensitivity.17 Due to photobleaching effects, excitation
light cannot be increased to fully compensate for polar-
izer absorption. In addition, the mathematical treat-
ment of anisotropy results, when used to compute
viscosity, has been subject to controversy.18,19 A third
fluorescence-based method used to assess membrane
viscosity relies on the use of environment-sensitive
fluorescent probes. In this category are included fluor-
escent molecules, such as DCVJ( 1) that exhibit a visc-
osity-dependent fluorescence quantum yield. These
compounds, often referred to as ‘molecular rotors’, are
particularly attractive to use since by simply measuring
their fluorescence intensity we can obtain information
on viscosity changes. Nonetheless, the major dis-
advantage of currently available molecular rotors is
their affinity to cellular organelles other than the plasma
membrane. For example, a strong affinity of DCVJ( 1)
for tubulin leads to areas of strong intracellular fluores-
cence, which reduce measurement sensitivity through
inactive background fluorescence.
Figure 4. Decrease of emission intensity of DCVJ( 1) and the phos-
pholipid-bound rotor 23 incorporated in phosphatidylcholine bilayers.
The transition temperature of pure phosphatidylcholine is 23 ꢀC.
While at this temperature 1 merely changes its slope, compound 23
shows a distinct inversion of the slope, clearly marking the transition
from the gel- to the liquid-crystal phase. The thin lines indicate
regression lines fitted into the data within the three distinctly different
phases: Gel (green), transition (black) and liquid-crystal (blue).
an inversion of the slope becomes visible between 23
and 27 ꢀC. Normalized to the intensity at 40 ꢀC, the
slopes were À9% below 23 ꢀC, +1% between 23 and
27 ꢀC, and À5% above 27 ꢀC. Above 27 ꢀC, the nor-
malized slopes of compounds 1 and 23 are similar.
Compound 28 does not show this inverted slope. Below
23 ꢀC, a slope of À10% was found, between 23 and
27 ꢀC À12%, and above 27 ꢀC À6%.
Localization of the probes in cultured cells
To determine the extent of cell membrane localization
of our synthesized probes we compared their staining to
that of commercially available probe Dil-C18, a dye
known to localize exclusively in plasma membrane.15
Cells of the endothelial cell line, ECV-304, plated onto
coverslips, were dual-labeled with DiI-C18 and one of
the compounds 22, 23, 24, or 28, and the staining was
examined using an epifluorescent microscope. In all four
cases, a clear co-localization of DiI with the phospho-
lipid-bound rotor could be seen (Fig. 5). A darker spot
indicating nuclear or nucleolar staining, which would
indicate internalization, was not observed. This indi-
cates that the phospholipid-bound molecular rotors
localize exclusively in the plasma membrane. It is worth
noting that dye 22 exhibits some cytotoxicity, as sug-
To improve the membrane localization profile of such
rotors we sought to attach them to a phospholipid
backbone using a polymethylene linker. Two strategies
were developed allowing attachment of the julolidine
scaffold at either the non-polar or the polar end of a
phospholipid (compounds 22, 23, 24, and compound 28,
respectively).
The molecular rotors synthesized in this study exhibit a
complex fluorescence pattern with triple excitation max-
ima, but a single emission maximum. The most wide-
spread molecular rotor, DCVJ( 1), shows only two single
excitation maxima.19 The emission intensity of our syn-
thesized rotors depends on the viscosity of the solvent, in
a manner similar to DCVJ. Examination of compounds

M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
3631
Figure 5. Dual stained cells of the ECV-304 line. The first column shows DiI-staining, while the second column shows fluorescence generated by the
probes 22 (top row), 23 (second row), 24 (third row), and 28 (bottom row). The false-color image in the third column shows DiI-staining in red and
the staining with the phospholipid-bound rotor in green. Yellow hues appear where both dyes are present. DiI and all phospholipid-bound probes
show similar staining patterns indicating that these dyes localize in the membrane. Cell detachment caused by 22 can be seen (top row) but is not
observed with the other probes.
22, 23, and 24 indicated that they have a slightly lower
10
interactions exist within the phospholipid-dye molecule
that inhibit free interaction with the environment, leading
to reduced viscosity sensitivity.
slope (ca. 0.5) than DCVJand FCVJ(ca. 0.6).
None-
theless, this decrease of slope does not translate to a sig-
nificant loss of viscosity sensitivity, particularly when the
probe is localized in cell membranes. Moreover, the visc-
osity sensitivity of compounds 22, 23 and 24 suggests that
the rotor–environment interaction is not affected if the
rotor is attached to the hydrophobic end of the phos-
pholipid. However, compound 28 (having the rotor
attached at the polar end of a phospholipid) was found
to have a significantly reduced exponent x in the For-
ster–Hoffmann relationship (eq 1). This suggests that
It has been observed that molecular rotors can be used
to monitor the phase transition of phospholipids.16 We
used this property to establish the behavior of phos-
pholipid-bound rotors in phospholipid bilayers. As
expected, we found changes of the gradient (intensity
change over temperature change) at the transition tem-
perature of phosphatidylcholine liposomes using DCVJ
(1) as a probe.16 The transition phase is even more

3632
M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
clearly marked by a temporary increase in intensity
when using phospholipid-bound rotor 23 (see Fig. 4,
bottom panel). The behavior of phospholipid bilayers at
the transition temperature is not fully understood.
Studies suggest that in the liquid-crystal state close to
the transition temperature, local gel-phase domains exist,
and furthermore, surface viscosity increases sig-
nificantly.20 This would be consistent with the increase in
intensity observed with molecular rotors bound to phos-
pholipids (such as 23). It is worth noting that this increase
is not observed with DCVJ. This, in turn, suggests that
the synthesized phospholipid-bound rotors reflect more
accurately changes in phospholipid transition phase of
liposomes and related membrane models.
these, rotor 24 exhibited an identical localization profile
with DiI. Compound 28, in which the rotor was
attached to the polar end of the phospholipid was
devoid of such properties.
Materials and Methods
General techniques
Lysophospholipids 20, 21, and 25 were purchased from
Avanti, Polar Lipids, Inc., AL, USA and used without
any additional purification. All reagents were commer-
cially obtained (Aldrich, Acros) at highest commercial
quality and used without further purification except
where noted. Air- and moisture-sensitive liquids and
solutions were transferred via syringe or stainless steel
cannula. Organic solutions were concentrated by rotary
evaporation below 45 ꢀC at about 20 mmHg. All non-
aqueous reactions were carried out using flame-dried
glassware, under an argon atmosphere in dry, freshly
distilled solvents under anhydrous conditions, unless
otherwise noted. Tetrahydrofuran (THF) and diethyl
ether (Et2O) were distilled from sodium/benzophenone;
dichloromethane (CH2Cl2) and toluene from calcium
hydride; and benzene from potassium. N,N-Diisopropyl-
ethylamine, diisopropylamine, pyridine, triethylamine
and boron trifluoride etherate were distilled from calcium
hydride prior to use. Yields refer to chromatographically
and spectroscopically (1H NMR) homogeneous materials,
unless otherwise stated. Reactions were monitored by
thin-layer chromatography carried out on 0.25 mm E.
Merck silica gel plates (60F-254) using UV light as
visualizing agent and 7% ethanolic phosphomolybdic
acid, or p-anisaldehyde solution and heat as developing
agents. E. Merck silica gel (60, particle size 0.040–0.063
mm) was used for flash chromatography. Preparative
thin-layer chromatography separations were carried out
on 0.25 or 0.50 mm E. Merck silica gel plates (60F-254).
NMR spectra were recorded on Varian Mercury 300,
400 and/or Unity 500 MHz instruments and calibrated
using residual undeuterated solvent as an internal refer-
ence. The following abbreviations were used to explain
the multiplicities: s=singlet; d=doublet, t=triplet,
q=quartet, m=multiplet, b=broad. IR spectra were
recorded on a Nicolet 320 Avatar FT-IR spectrometer
and values are reported in cmÀ1 units. High resolution
mass spectra (HRMS) were recorded on a VG 7070 HS
mass spectrometer under chemical ionization (CI) con-
ditions or on a VG ZAB-ZSE mass spectrometer under
fast atom bombardment (FAB) conditions.
When the rotor group is attached to the hydrophilic end
(see probe 28), the gradient inversion observed with 22,
23, and 24 is not visible. Failure of compound 28 to
exhibit viscosity sensitivity may be due to the loss of
zwitterionic character of the phospholipids structure and
the presence of a phosphate anion close to the julolidine
part. The difference between the slopes in the gel- and the
liquid-crystal phase is even lower than with DCVJ, which
is consistent with the diminished viscosity sensitivity
stated earlier. We therefore consider compound 28 not
to be useful as a membrane probe.
The excellent localization of all synthesized phospho-
lipid bound rotors in cell plasma membrane, is illu-
strated in Figure 5. The co-localization study with the
known membrane dye DiI-C18 clearly shows similar
staining patterns for both dyes. A darker area at the
location of the cell nucleus, as observed with DCVJ
and—to a lesser extent—the farnesyl derivative FCVJ
(2) was not observed with 22, 23, 24, or 28. These results
suggest that all phospholipid-bound rotors remain
attached to the plasma membrane with negligible trans-
fer into the interior of the cell. However, it was also
observed that under the experimental conditions of this
study, compound 22 leads to cell detachment, an indi-
cation of cytotoxicity. This effect may be due to the
introduction of a short (C-8) phosphatidylcholine con-
jugate that interferes with the stucture of the cell mem-
brane. Such cell detachment was not observed with the
longer phospholipid conjugates 23 and 24. Based on the
false-color images (Fig. 5) we can conclude that phos-
pholipid-bound rotor 24 shows identical localization
profile with DiI-C18, rendering it a probe of choice for
studies of cell membrane viscosity.
Conclusion
Ester 6. To a solution of 5.7 g of 4-methoxy benzyl
alcohol (5) (28.9 mmol) and 3.5 g of DMAP (29 mmol)
in 50 mL of dichloromethane under argon at 0 ꢀC were
added dropwise 5.7 g of 6-bromo hexanoyl chloride (27
mmol). The reaction mixture was then allowed to warm
up to 25 ꢀC, where it was stirred for 2 h. The DMAP-
HCl salt was filtered off by gravity filtration and the
mixture was concentrated under reduced pressure. The
resulting oil was filtered through a short silica plug with
hexane. The solvent was then removed to give 7.5 g of
benzyl ester 6 (23.8 mmol, 82%), which was used in the
To improve the membrane localization of DCVJand
related fluorescent rotors, we attached the julolidine
fragment to either end of a phospholipid structure. All
compounds exhibited the desired membrane localization
profile, which was comparable to that of the commonly
used membrane dye DiI. Only compounds in which the
rotor was attached to the non-polar end of the phospho-
lipid (such as 22, 23, and 24) displayed the characteristic
property of a fluorescent rotor, which is the viscosity
dependent fluorescence emission quantum yield. Among

M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
3633
next reaction without further purification. 6: colorless
liquid; IR (film) nmax 1732; 1H NMR (400 MHz, CDCl3)
d 7.39 (2H, d, J=6.4), 6.89 (2H, d, J=6.4), 5.05 (2H, s),
3.81 (3H, s), 3.38 (2H, t, J=6.8), 2.34 (2H, t, J=7.2),
1.85 (2H, m), 1.65 (2H, m), 1.45 (2H, m); 13C NMR
(100 MHz, CDCl3) d 173.1, 159.3, 130.0, 127.9, 113.8,
66.0, 55.3, 34.1, 33.6, 32.4, 27.7, 24.1; HRMS, calcd for
C14H19BrO3 (M+Cs+) 446.9568, found 446.9572.
(3H, s), 2.59 (2H, t, J =6.8), 2.27 (2H, t, J =7.2), 1 60–
1.54 (4H, m), 1.40–1.36 (2H, m), 1.29–1.25 (2H, m); 13C
NMR (400 MHz, CDCl3) d 173.5, 159.5, 132.0, 129.9,
128.5, 113.8, 65.8, 55.1, 41.7, 34.1, 26.3, 24.7; HRMS,
calcd for C14H21NO3 (M+H+) 252.1594, found 252.1595.
Amino ester 11. Preparation of this compound was
accomplished in 91% yield following the procedure
described above for the synthesis of amino ester 10. 11:
1
Ester 7. Preparation of this compound was accom-
plished in 61% yield following the procedure described
above for the synthesis of ester 6. 7: white solid; IR
white solid; H NMR (400 MHz, CDCl3) d 7.30 (2H, d,
J=6.8), 6.89 (2H, d, J =6.8), 5.04 (2H, s), 3.81 (3H, s),
2.69 (2H, t, J =6.8), 2.33 (2H, t, J =7.6), 1.60 (2H, m),
1.42 (2H, m), 1.27 (12H, s), 1.13 (2H, s); 13C NMR
(100 MHz, CDCl3) d 173.6, 159.4, 129.9, 128.1, 113.8,
113.7, 65.8, 55.2, 41.0, 34.3, 30.7, 29.6, 29.4, 29.3, 29.1,
26.8, 24.9; HRMS, calcd for C19H31NO3 (M+H+)
322.2382, found 322.2390.
1
(film) nmax 1733; H NMR (400 MHz, CDCl3) d 7.30
(2H, d, J =8.8), 6.90 (2H, d, J=8.8), 5.05 (2H, s), 3.81
(3H, s), 3.42 (2H, t, J =6.8), 2.34 (2H, t, J =8), 1.85
(2H, m), 1.62 (2H, m), 1.27 (10H, s); 13C NMR
(100 MHz, CDCl3) d 173.7, 159.4, 129.9, 128.1, 113.8,
113.8, 65.8, 55.2, 34.3, 34.0, 32.8, 29.3, 29.3, 29.1, 29.0,
28.7, 28.1, 24.9; HRMS, calcd for C19H29BrO3
(M+Cs+) 517.0351, found 517.0372.
Cyano ester 13. To a solution of 2 g of amine 9 (8
mmol) and 0.68 g of cyano acetic acid (12) (8 mmol) in
15 mL of dichloromethane at 0 ꢀC under argon was
added dropwise a solution of 1.6 g DCC (8 mmol) in 5
mL of dichloromethane. When the addition was com-
plete the ice bath was removed and the reaction was
stirred at 25 ꢀC for 3 h. 20 mL of hexane was then added
to the reaction mixture and the DCU side-product was
filtered off. The reaction mixture was concentrated
under reduced pressure and the residue chromato-
graphed (silica, 0–10% methanol in dichloromethane)
to yield 1.3 g of compound 13 (4.1 mmol, 41%). 13:
colorless oil; IR (film) nmax 2305, 1725; 1H NMR
(400 MHz, CDCl3) d 7.29 (2H, d, J =8.4), 6.89 (2H, d, J
=8.4), 6.29 (1H, bs), 5.05 (2H, s), 3.81(3H, s), 3.35 (2H,
s), 3.28 (2H, q, J =7.2), 2.34 (2H, t, J=7.6), 1.65 (2H,
m, J=8.0), 154 (2H, m, J =7.2), 1.34 (2H, m, J =7.2);
13C NMR (100 MHz, CDCl3) 173.2, 160.6, 159.4, 129.9,
127.9, 114.7, 114.4, 66.1, 55.3, 40.1, 34.0, 28.8, 26.2,
25.9, 24.3; HRMS, calcd for C17H22N2O4 (M+Cs+)
451.0630, found 451.0619.
Azide 8. A solution of 7.4 g of ester 6 (23.5 mol) and 1.7
g of sodium azide (26 mmol) in DMF (50 mL) was stir-
red at 25 ꢀC for 12 h. The reaction mixture was diluted
with water (200 mL) and extracted with ether (3Â100
mL). The solvent was removed under reduced pressure
and the resulting oil was chromatographed (silica, 10–
30% ether in hexanes) to yield 6.1 g of compound 8 (22
mmol, 94%). 8: colorless liquid; IR (film) nmax 2095,
1732; 1H NMR (400 MHz, CDCl3) d 7.29 (2H, d, J
=6.4), 6.89 (2H, d, J =6.4), 5.05 (2H, s), 3.80 (3H, s),
3.25 (2H, t, J =6.8), 2.34 (2H, t, J =7.2), 1.68–1.54
(4H, m), 1.40 (2H, m); 13C NMR (100 MHz CDCl3) d
173.3, 159.6, 130.0, 128.1, 113.9, 66.0, 55.2, 51.2, 34.0,
28.5, 26.1, 24.3; HRMS, calcd for C14H19 N3O3
(M+Cs+) 410.0477, found 410.0491.
Azide 9. Preparation of this compound was accom-
plished in 92% yield following the procedure described
above for the synthesis of azide 8. 9: colorless liquid; IR
1
(film) nmax 2095, 1732; H NMR (400 MHz, CDCl3) d
Cyano ester 14. Preparation of this compound was
accomplished in 47% yield following the procedure
described above for the synthesis of compound 13. 14:
white solid; IR (film) nmax 2307, 1724; 1H NMR
(400 MHz, CDCl3) d 7.29 (2H, d, J =8.8), 6.88 (2H, d, J
=8.8), 6.15 (1H, s), 5.02 (2H, s), 3.79 (3H, s), 3.26 (2H,
s), 3.29 (2H, q, J =6.4), 2.32 (2H, t, J=8), 1.60 (2H, m),
1.51 (2H, m), 1.24 (32H, s); 13C NMR (100 MHz,
CDCl3) d 173.8, 160.6, 159.5, 130.0, 128.1, 113.8, 113.8,
65.9, 55.3, 40.4, 34.3, 29.3, 29.2, 29.1, 29.1, 29.0, 26.7,
25.8, 24.9; HRMS, calcd for C22H32N2O4 (M+Cs+)
521.1413, found 521.1429.
7.28 (2H, d, J =8.8), 6.88 (2H, d, J =8.8), 5.03 (2H, s),
3.79 (3H, s), 3.23 (2H, t, J =7.2), 2.30 (2H, t, J =7.6),
1.59 (4H, m), 1.25 (12H, s); 13C NMR (100 MHz,
CDCl3) d 173.3, 159.3, 129.7, 128.0, 113.6, 65.6, 54.9,
51.2, 34.1, 29.2, 29.1, 29.0, 28.9, 28.9, 28.6, 26.5, 24.7;
HRMS, calcd for C19H29N3O3 (M+Cs+) 480.1260,
found 480.1272.
Amino ester 10. To a solution of 3.0 g of azide 8 (9.4
mmol) in THF (50 mL) were added 2.62 g of triphenyl-
phosphine (10 mmol) and 1 mL of H2O. After stirring at
25 ꢀC for 12 h the reaction mixture was concentrated
under reduced pressure and the remaining residue was
dissolved in 75 mL of dichloromethane and washed with
aqueous saturated sodium carbonate (2Â50 mL). The
organic layer was collected, dried over MgSO4 and
concentrated. The residue was filtered through a short
plug of silica (0–5% methanol in dichloromethane) to
yield 2.1 g of the amino ester 10 (8.3 mmol, 88%) which
was taken directly to the next step as it was prone to
Ester 16. A solution of 2.0 g of cyano ester 13 (6.3
mmol) and 1.3 g of aldehyde 15 (6.3 mmol) in 20 mL of
THF were treated with 0.96 g of DBU (6.3 mmol). The
reaction mixture was then warmed to 50 ꢀC and stirred
for 12 h. The solvent was removed under reduced pres-
sure and the crude residue was chromatographed (silica,
10–40% ether in hexanes) to yield 2.3 g of compound 16
(4.6 mmol, 76%). 16: yellow solid; IR (film) nmax 3054,
2986, 2305, 2197, 1730, 1666, 1516; 1H NMR (400 MHz,
CDCl3) d 8.00 (IH, s), 7.43 (2H, s), 7.30 (2H, d, J =8.8),
1
decomposition. 10: H NMR (400 MHz, CDCl3) d 7.23
(2H, d, J =6.4), 6.81 (2H, d, J =6.4), 4.98 (2H, s), 3.74

3634
M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
6.88 (2H, d, J =8.8), 6.21 (1H, t, J =5.2), 5.04 (2H, s),
3.80 (3H, s), 3.36 (2H, q, J =6.8), 3.31 (4H, t, J =6.0),
2.74 (4H, t, J =7.6), 2.34 (2H, t, J=7.6), 1.95 (4H, m),
1.70–1.54 (4H, m) 1.38 (2H, m); 13C NMR (100 MHz,
CDCl3) 173.1, 162.2, 159.3, 152.1, 146.8, 130.8, 130.0,
128.0, 120.5, 119.4, 118.4, 113.8, 93.2, 65.9, 55.3, 50.1,
40.1, 34.2, 29.4, 27.6, 26.4, 24.6, 21.2; HRMS, calcd for
C30H35N3O4 (M+Cs+) 634.1678, found 634.1693.
CDCl3) d 7.97 (1H, s), 7.42 (2H, s), 6.44 (1H, t, J =5.6),
5.21–5.19 (IH, m), 4.38–4.23 (3H, m), 4.15–4.10 (1H,
m), 3.98 (2H, t, J =6.4), 3.83 (2H, s), 3.38 (11H, s), 3.29
(4H, t, J=5.6), 2.73 (4H, t, J=6.4), 2.34–2.25 (4H, m),
1.97–1.91 (4H, m), 1.65–1.52 (4H, m), 1.4–1.35 (2H, m),
1.23 (16H, s), 0.86 (3H, t, J =6.8); 13C NMR (100 MHz,
CDCl3) d 173.3, 172.7, 162.3, 152.9, 146.8, 130.8, 120.5,
119.3, 118.4, 93.4, 70.6, 70.4, 66.3, 63.5, 62.9, 59.3, 54.5,
50.1, 40.1, 34.1, 32.0, 29.7, 29.7, 29.6, 29.4, 29.4,
29.3,29.2, 27.7, 26.4, 26.3, 24.9, 24.6, 22.7, 21.3, 14.2;
HRMS m/z calcd for C42H67N4O9P (M+H+) 803.4718,
found 803.4682.
Ester 17. Preparation of this compound was accom-
plished in 72% yield following the procedure described
above for the synthesis of compound 16. 17: yellow
solid; IR (film) nmax 3054, 2986, 2305, 2197, 1730, 1666,
1
1516; H NMR (400 MHz, CDCl3) d 8.00 (1H, s), 7.42
Phospholipid 23. Preparation of this compound was
accomplished following the procedure described above
for the synthesis of compound 22. 23: IR (film) nmax
3377, 2924, 2852, 2200, 1735, 1664; 1H NMR (400 MHz,
CDCl3) d 7.95 (1H, s), 7.42 (2H, s), 6.44 (1H, t, J=5.6),
5.20–5.17 (1H, m), 4.38–4.34 (3H, m), 4.15–4.10 (1H,
m), 3.98 (2H, t, J =6.4), 3.84 (2H, s), 3.38 (11H, s), 3.29
(4H, t, J=5.6), 2.73 (4H, t, J=6.4), 2.34–2.25 (4H, m),
1.97–1.91 (4H, m), 1.65–1.52 (4H, m), 1.4–1.35 (2H, m),
1.23 (26H, s), 0.86 (3H, t, J =6.8); 13C NMR (100 MHz,
CDCl3) d 173.3, 172.7, 162.3, 152.9, 146.8, 130.8, 120.5,
119.3, 118.2, 93.4, 70.6, 70.5, 66.3, 63.6, 62.8, 59.4, 54.4,
50.1, 40.1, 34.1, 32.0, 29.7, 29.6, 29.4, 29.3, 29.2, 27.6,
26.3, 24.9, 24.6, 22.7, 21.2, 14.2; HRMS m/z calcd for
C46H75N4O9P (M+H+) 859.3544, found 859.5322.
(2H, s), 7.28 (2H, s), 6.88 (2H, d, J =8.4), 6.22 (1H, s),
5.03 (2H, s), 3.78 (3H, s), 3.37 (2H, q, J =6.8), 3.29
(4H, t, J=5.6), 2.73 (4H, t, J =6.0), 2.39 (2H, t,
J=7.2), 1.94 (4H, m), 1.62–1.56 (4H, m), 1.25 (12H, s);
13C NMR (100MHz, CDCl3) d 173.6, 162.3, 159.4, 152.1,
146.8; 130.8 129.9, 128.2, 120.6, 119.4, 118.5, 113.8, 113.7,
93.4, 65.7, 55.2, 49.9, 40.27, 34.3, 29.5, 29.31, 29.26, 29.2,
29.1, 29.0, 27.5, 26.8, 24.9, 21.1; HRMS, calcd for
C35H45N3O4 (M+H+) 572.3487, found 572.3491.
Acid 18. 2.2 g of ester 16 (4.4 mmol) were dissolved in
10 mL of a 4:1 mixture of trifluoroacetic acid and ani-
sole and stirred at 25 ꢀC for 10 min. The solvents were
initially removed under reduced pressure, and sub-
sequently azeotropically removed using benzene (3Â10
mL). The residue was then chromatographed (silica, 0–
10% methanol in dichloromethane) to produce 1.5 g of
carboxylic acid 18 (3.9 mmol, 59%). 18: red solid; IR
Phospholipid 24. Preparation of this compound was
accomplished in 48% yield following the procedure
described above for the synthesis of compound 22. 24:
1
1
(film) nmax 2199, 1720, 1647, 1515; H NMR (400 MHz,
red solid; H NMR (400 MHz, CDCl3) d 7.98 (1H, s),
DMSO-d6) d 7.94 (1H, t, J=5.6), 7.77 (1H, s), 7.40 (2H,
s), 3.27 (4H, m), 3.15 (2H, m), 2.66 (4H, m), 2.19 (2H, t,
J =7.6), 1.85 (4H, m, J =5.6), 1.48 (4H, m), 1.26 (2H,
m); 13C NMR (100 MHz, DMSO-d6) d 174.4, 162.2,
150.0, 146.4, 130.1, 120.3, 118.5, 117.7, 94.9, 49.3, 33.6,
28.8, 27.1, 25.9, 24.2, 20.7; HRMS, calcd for
C22H27N3O3 (M+C+) 514.1103, found 514.1130.
7.41 (2H, s), 6.22 (1H, t, J=5.2), 5.17 (1H, m), 4.38–
4.28 (3H, m), 4.12–4.08 (1H, m), 3.93 (2H, d, J=6.0),
3.78 (2H, s), 3.37 (11H, s), 4.29 (4H, t, J=5.2), 2.72
(4H, t, J=6.0), 2.28–2.23 (5H, m), 1.96–1.90 (4H, m),
1.55 (7H, s), 1.23 (40H, s), 0.85 (3H, t, J=6.4); 13C
NMR (100 MHz, CDCl3) d 173.4, 173.1, 162.0, 152.1,
146,87, 130.9, 120.6, 119.4, 118.6, 93.5, 70.5, 70.4, 66.3,
63.3, 62.9, 59.2, 54.4, 50.0, 40.3, 34.3, 34.1, 31.9, 29.7,
29.6, 29.6, 29.5, 29.4, 29.4, 29.3, 29.3, 29.2, 29.1, 29.1,
27.6, 26.9, 24.9, 24.8, 22.6, 21.2, 14.1; HRMS m/z calcd
for C51H85N4O9P (M+H+) 929.4327, found 929.4339.
Acid 19. Preparation of this compound was accom-
plished following the procedure described above for the
1
synthesis of compound 18. 19: brown solid; 73%; H
NMR (400 MHz, CDCl3) d 8.02 (1H, s), 7.44 (2H, s),
6.27 (1H, t, J=5.6), 3.45 (4H, m), 3.29 (2H, m), 2.83
(4H, t, J =6.4), 2.43 (2H, t, J =7.6) 2.04 (4H, t, J =6),
1.64–1.53 (4H, m), 1.28 (12H, s); 13C NMR (100 MHz,
CDCl3) d 179.0, 162.5, 152.3, 146.9, 130.9, 120.6, 119.5,
118.5, 93.2, 50.0, 40.3, 33.9, 29.5, 29.2, 29.3, 29.1, 29.1,
28.9, 27.6, 26.8, 24.6, 21.2; HRMS, calcd for
C27H37N3O3 (M+Cs+) 584.1886, found 584.1899.
Compound 28. To a solution of 51 mg cyano acetic
p-nitro phenolate (0.25 mmol) in 0.5 mL CDCl3, was
added 160 mg (0.25 mmol) dimyristoyl-l-a-phosphati-
dyl ethanolamine (C14:0) and 31 mg (0.25 mmol)
DMAP and the reaction was stirred at 25 ꢀC for 12 h.
The solvent was then removed under reduced pressure
and the residue was dissolved in THF (1.5 mL) and
treated with 42 mg of DBU (0.25 mmol) and 52 mg of
aldehyde 15 (0.28 mmol). The reaction mixture was
stirred at 50 ꢀC for 12 h. The solvent was removed under
reduced pressure and the residue was purified on silica
(0–8% methanol in dichloromethane) to yield 85 mg of
phospholipid 28 (38%). 28: bright yellow liquid; IR
(film) nmax 3370, 2923, 2852, 2201, 1738, 1516; 1H NMR
(400 MHz, CD3OD) d 7.76 (1H, s), 7.34 (2H, s), 5.10
(1H, m), 4.38–4.28 (1H, m), 4.10–4.02 (1H, m), 3.90–
3.79 (3H, m), 3.21–3.09 (4H, m), 2.63–2.56 (4H, m), 1.91
(2H, m, J=5.6), 1.84 (2H, m, J=6.0), 1.70–1.41 (12H,
Phospholipid 22. To a solution of 138 mg of acid 18
(0.33 mmol) in 3 mL of chloroform were added 107 mg
of lysophospholipid 20 (0.33 mmol), 64 mg of EDC
(0.33 mmol), and 40 mg of DMAP (0.33 mmol) and the
reaction was stirred for 18 h at 25 ꢀC. The solvent was
removed under reduced pressure and the resulting resi-
due was purified by reversed phase flash chromato-
graphy (0–100% acetonitrile/water) to give 157 mg of
phospholipid 22 (54%). 22: red liquid; IR (film) nmax
3377, 2924, 2852, 2200, 1735, 1664; 1H NMR (400 MHz,

M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
3635
m), 1.15 (40H, s), 0.78 (6H, t, J=5.2); 13C NMR
(100 MHz, CDCl3) d 172.5, 172.3, 165.5, 150.7, 146.4,
130.2, 120.22, 118.3, 117.9, 71.4, 71.4, 65.6, 65.1, 63.7,
55.6, 51.3, 49.7, 48.3, 37.6, 33.9, 33.7, 31.7, 31.6, 29.4,
29.3, 29.2, 29.1, 28.9, 28.7, 27.3, 26.5, 24.6, 23.7, 22.4,
20.9, 19.2, 13.3; HRMS m/z calcd for C49H80N3O9P
(MÀH+) 884.5559, found 884.5518.
CHCl3 and methanol (2:1), then adding 100 mL of the
probe stock solution (10 mM in CHCl3). The fluid was
dried under nitrogen on ice, then resuspended in 10 mL
phosphate-buffered saline (PBS) and sonicated for 10
min.
Staining solution for DiI-C18 was prepared by adding 5
mL Vybrant DiI labeling solution (Molecular Probes,
Eugene, OR, USA) to 1 mL PBS. Cell staining was
performed by washing the slides in PBS, then covering
the slide surface with 1 mL of the DiI staining medium.
After 20 min incubation in the dark at 37 ꢀC, the slides
were washed in PBS and the micelle suspension was
added in a similar manner with an incubation time of 10
min. After the incubation period, the slides were washed
again and immediately examined under the microscope
(Nikon Diaphot TMD with DVC-1310 CCD camera
acquisition system). Fluorescent images were obtained
by acquiring one image with the G1B filter set (dyes 22,
23, 24, and 28) and a second image by switching to the
Texas Red filter set without moving the sample. The
two images were merged using the red channel for DiI
and the green channel for 22, 23, 24, and 28.
Determination of physical properties
Stock solutions of 22, 23, 24, 28, and 1 were prepared at
a concentration of 10 mM in chloroform. For the
determination of the relationship between quantum
yield and viscosity of the medium, mixtures of ethylene
glycol and glycerol were prepared. Different viscosities
were achieved by different volume/volume mixture
ratios of ethylene glycol/glycerol as follows: 7:3 (49 cP),
5:5 (115 cP), 4:6 (163 cP), 3:7 (245 cP), 2:8 (391 cP) fol-
lowing an experiment described previously.12,21 3.5 mL
of each of these mixtures was placed into a spectro-
scopic cuvette. Then, 10 mL of probe stock solution was
added. The emission intensity of each of the five samples
was acquired using a Spex FluoroMax 3 fluoro-
photometer (Jobin-Yvon, Edison, NJ, USA). Excitation
wavelength was set to 470 nm and an emission scan
performed from 480 to 550 nm. The maximum was
recorded and plotted against the viscosity in double-
logarithmic scale. A straight line was fitted to the loga-
rithmized data points using the least-squares method,
and the slope obtained.
Acknowledgements
This research was partially supported by the NIH (HL-
40696 and CA-85600).
References and Notes
Preparation and measurement of stained liposomes
1. For Selected Accounts On Cell Membrane-Related Topics,
see: Morre, D. J., Howell, K. E., Bergeron, J. J. M., Eds.
Molecular Mechanisms of Membrane Traffic. Springer: New
York, 1993. Rohrbach, D. H., Timpl, R., Eds. Molecular and
Cellular Aspects of Basement Membranes. Academic: San
Diego, 1993. Yeagle, P. L., Ed. The Membranes of Cells. Aca-
demic: San Diego, 1993. Haber, E., Ed. The Cell Membrane:
Its Role in Interaction withthe Outside World . Plenum: New
York, 1984.
Dimyristoylphosphatidylcholine (DMPC), dissolved in
chloroform at 25 mg/mL, was purchased from Avanti
Polar Lipids. To 400 mL of this solution was added 20
mL of the 10 mM probe stock solution. In a glass vial,
the chloroform was evaporated under nitrogen at 0 ꢀC.
8 mL phosphate-buffered saline (PBS, Irvine Scientific)
was added. After sonication, liposomes were created by
extruding the mixture through a 400-nm filter at 37 ꢀC.
The final liposome suspension was placed into a spec-
troscopic glass cuvette and placed in a Shimadzu RF-
1501 spectrophotometer (Shimadzu, Kyoto, Japan)
equipped with a temperature-controlled cuvette holder
and a time-based acquisition program. The liposome
suspension was cooled to 10 ꢀC in the fluoro-
spectrometer. At fixed wavelengths lex=470 nm and
2. (a) Singer, S. J.; Nicolson, G. L. Science 1972, 175, 720. (b)
Singer, S. J. Ann. Rev. Biochem. 1974, 43, 805.
3. Lakowicz, J. R. Ed. Principles of Fluorescence Spectro-
scopy. Kluver Academic/Plenum: New York, 1999 (2nd ed.).
4. Axelrod, D.; Koppel, D. E.; Schlessinger, J.; Elson, E.;
Webb, W. W. Biophys. J. 1976, 16, 1055. Swaminathan, R.;
Bicknese, S.; Periasamy, N.; Verkman, A. S. Biophys. J. 1996,
71, 1140.
l
em=490 am, emission intensity was recorded while the
5. Shinitzky, M.; Barenholz, Y. Biochim. Biophys. Acta 1978,
515, 367.
temperature was raised from 10 to 40 ꢀC over a period
of 5 min. Temperature and intensity were recorded
simultaneously.
6. Parasassi, T.; Di Stefano, M.; Loiero, M.; Ravagnan, G.;
Gratton, E. Biophys. J. 1994, 66, 763. Rettig, W., Lapouyade,
R. Fluorescence Probes Based on Twisted Intramolecular
Charge Transfer (TICT) States and Other Adiabatic Photo-
reactions. In Topics in Fluorescence Spectroscopy, Vol. 4,
Probe Design and Chemical Sensing: Lakowicz, J. R., Ed.;
Plenum: New York, 1994; p 109.
7. Loutfy, R. O. Pure Appl. Chem. 1986, 58, 1239. Williams,
M. L.; Landel, R. F.; Ferry, J. D. J. Am. Chem. Soc. 1955, 84,
2803. Kasatai, K.; Kawasaki, M.; Sato, H. Chem. Phys. 1984,
83, 461. Iwaki, T.; Torigoe, C.; Noji, M.; Nakanishi, M. Bio-
chemistry 1993, 32, 7589. Haidekker, M. A.; L’Heureux, N.;
Frangos, J. A. Am. J. Physiol. Heart Circ. Physiol. 2000, 278,
H1401.
Preparation and microscopy of stained cells
Immortalized endothelial cells of the cell line ECV-304
(ATCC) were seeded onto glass coverslips at a density
of 800,000 mLÀ1 and grown to confluence in M199,
supplemented with 10% fetal calf serum (FCS) within
24 h. The dyes, 22, 23, 24, and 28, were delivered to the
cells in the form of micelles as follows:22 Micelles were
prepared by dissolving 0.25 mg sodium deoxycholate
(Sigma, St Louis, MO, USA) in 5 mL of a mixture of

3636
M. A. Haidekker et al. / Bioorg. Med. Chem. 10 (2002) 3627–3636
8. Viriot, M. L.; Carre, M. C.; Geoffroy-Chapotot, C.; Brem-
billa, A.; Muller, S.; Stoltz, J.-F. Clin Hemorheol. Microcirc.
1998, 19, 151.
9. Kung, C. E.; Reed, J. K. Biochemistry 1989, 28, 6678.
Sawada, S.; Iio, T.; Hayashi, Y.; Takahashi, S. Anal. Biochem.
1992, 204, 110. Iio, T.; Takahashi, S.; Sawada, S. J. Biochem.
1993, 113, 196.
10. Haidekker, M. A.; Ling, T.; Anglo, M.; Stevens, H. Y.;
Frangos, J. A.; Theodorakis, E. A. Chem. Biol. 2001, 8, 123.
11. Yokokawa, F.; Toshinobu, A.; Takayuki, S. Org. Lett.
2000, 2, 4169.
12. Sigeru, T.; Hideo, T.; Masatoshi, T.; Yutaka, K.; Michio,
S.; Takashi, S.; Ryo, K.; Ichiro, K.; Akihiro, S.; Shigemitsu,
N. J. Org. Chem. 1991, 56, 3633.
13. Forster, Th.; Hoffmann, G. Z. Phys. Chem. 1971, 75, 63.
14. Lukac, S. J. Am. Chem. Soc. 1984, 106, 4386.
15. Honig, M. G.; Hume, R. I. Cell Biol. 1986, 103, 171. Wolf,
D. E.; Scott, B. K.; Millette, C. F. J. Cell Biol. 1986, 103, 1745.
Ledley, F. D.; Soriao, H. E.; O’Malley, B. W., Jr.; Lewis, D.;
Darligto, G. J.; Finegold, M. Biotechniques 1992, 580, 584.
Axelrod, D. Biophys. J. 1979, 26, 557.
16. Edidin, M. Fluorescence Photobleaching and Recovery,
FPR, in the Analysis of Membrane Structure and Dynamics.
In Mobility and Proximity in Biological Membranes; Damja-
novich, S., Edidin, M., Eds.; CRC: Boca Raton, 1994 p 109.
17. Abdul-Rahim, H.; Bouchy, M. J. Photochem. Photobio. B:
Biol. 1998, 47, 95.
18. (a) van Ginkel, G.; van Langen, H.; Levine, Y. K. Bio-
chimie 1989, 71, 23. (b) Mulders, F.; van Langen, H.; van
Ginkel, G.; Levine, Y. K. Biochim. Biophys. Acta 1986, 859,
209. Chapman, D. Biomembrane Fluidity: The Concept and
Its Development; In Membrane Fluidity in Biology; Aloia, R.
C., Ed.; Academic: New York, London, 1983; p 5.
19. Loufty, R. O.; Law, K. Y. J. Phys. Chem. 1980, 84,
2803. Sundstrom, V.; Gillbro, T. J. Phys. Chem. 1982, 86,
1788.
20. Dimova, R.; Pouligny, B.; Dietrich, C. Biophys. J. 2000,
79, 340.
21. Iwaki, T.; Torigoe, C.; Noji, M.; Nakanishi, M. Biochem-
istry 1993, 32, 7589.
22. Began, G.; Sudharsan, E.; Appu Rao, A. G. Biochemistry
1999, 38, 13920.