Synthesis and fluorescence properties of novel transmembrane probes and determination of their orientation within vesicles

Article abstract of DOI:10.1002/1522-2675(20000906)83:9<2464::AID-HLCA2464>3.0.CO;2-T

Two novel transmembrane fluorescent diester probes D and E bearing an anthracenediyl moiety in the middle of the molecule have been synthesized. Their absorption and fluorescence spectra in CHCl₃ solution as well as their fluorescence characteristics in dimyristoylphosphatidylcholine (DMPC) large unilamellar vesicles were determined. Although their absorption spectra (first transition, S₀ → S₁) present a good overlap with the fluorescence spectrum of tryptophan, only probe E could be a good acceptor for the energy-transfer experiments, since a strong overlap exists between the absorption spectrum of tryptophan and the second transition (S₀ → S₂) of the absorption spectrum of probe D. The Forster critical distance R₀ for energy transfer between tryptophan (donor) and probe E (acceptor) is found to be 23-24 A. Finally, linear-dichroism studies on shear-deformed DMPC vesicles show the incorporated probe E to lie essentially perpendicular to the bilayer plane. These results establish that probe E could be useful in the study of membrane-bound protein topography by the fluorescence-energy-transfer method.

Full text of DOI:10.1002/1522-2675(20000906)83:9<2464::AID-HLCA2464>3.0.CO;2-T

2464
Helvetica Chimica Acta ± Vol. 83(2000)
Synthesis and Fluorescence Properties of Novel Transmembrane Probes and
Determination of Their Orientation within Vesicles1)
by Ernesto Quesadaa), Malin Ardhammarb), Bengt NordeÂnb), Michel Mieschc), Guy Duportaild),
Yvonne Bonzi-Coulibalye), Yoichí Nakatani*e), and Guy Ourissone)
a
Â
) Instituto de Química Organica General C.S.I.C., Juan de la Cierva 3, E-28006 Madrid
b) Department of Physical Chemistry, Chalmers University of Technology, SE-41296 Göteborg
c
Â
Â
) Laboratoire de Chimie Organique Synthetique, associe au CNRS, Institut de Chimie,
Â
Universite Louis Pasteur, 1 rue Blaise Pascal, F-67084 Strasbourg
d
Â
) Laboratoire de Pharmacologie et Physicochimie, UMR 7034 du CNRS, Faculte de Pharmacie,
Â
Universite Louis Pasteur, F-67401 Illkirch
e
Â
) Laboratoire de Chimie Organique des Substances Naturelles, associe au CNRS, Centre de Neurochimie,
Â
Universite Louis Pasteur, 5 rue Blaise Pascal, F-67084 Strasbourg (tel: (‡ 33) 388456732;
fax: (‡ 33) 388607620; e-mail: nakatani@chimie.u-strasbg.fr)
This paper is dedicated to a 45 yearsꢀ friendship with Albert Eschenmoser, not enough made use of. In our next
life, I shall better make use of you, Albert!
G. Ourisson
Two novel transmembrane fluorescent diester probes D and E bearing an anthracenediyl moiety in the
middle of the molecule have been synthesized. Their absorption and fluorescence spectra in CHCl3 solution as
well as their fluorescence characteristics in dimyristoylphosphatidylcholine (DMPC) large unilamellar vesicles
were determined. Although their absorption spectra (first transition, S0 ! S1) present a good overlap with the
fluorescence spectrum of tryptophan, only probe E could be a good acceptor for the energy-transfer
experiments, since a strong overlap exists between the absorption spectrum of tryptophan and the second
transition (S0 ! S2) of the absorption spectrum of probe D. The Förster critical distance R0 for energy transfer
between tryptophan (donor) and probe E (acceptor) is found to be 23± 24 Š. Finally, linear-dichroism studies on
shear-deformed DMPC vesicles show the incorporated probe E to lie essentially perpendicular to the bilayer
plane. These results establish that probe E could be useful in the study of membrane-bound protein topography
by the fluorescence-energy-transfer method.
Introduction. ± Although there has been considerable progress in crystallizing
membrane proteins (for a recent example, see [1]), the determination of protein
organization within a membrane is still quite difficult. The in situ determination of the
topography (localization, orientation, and conformation) of molecules present in
biomembranes calls for specific methods involving, e.g., photochemical labelling [2],
fluorescence-energy transfer [3], or fluorescence quenching [4]. The fluorescence
studies are especially useful to define the position of tryptophan residues of membrane
proteins. However, just like for the photolabelling methods, a limitation of the
fluorescence methods lies in the difficulty of avoiding extensive disorder of the probe in
the liquid-crystalline phase of the phospholipid matrix, leading to a loss of depth
selectivity. The concept of transmembrane immobilization of the probe might provide a
solution to this problem [5 ± 7].
1
)
Preliminary results of this work were presented at the XVII International IUPAC Meeting on
Photochemistry, Sitges, Barcelona, Spain, July, 1998.

Helvetica Chimica Acta ± Vol. 83(2000)
2465
We have previously developed the photoactivable membrane probe A designed to
maintain the photosensitive group near the middle of the membrane [5]. This is a
phospholipid characterized by a double-headed system, with the photoactivable group
(in our case, a benzophenone) localized in the center of a transmembrane chain. It is
easily incorporated into dimyristoylphosphatidylcholine (DMPC) vesicles where, at
least in the presence of a high concentration of cholesterol, it spans the membrane. It is
also highly regioselective for targets lying near the center of the membrane. We had
extended the same principle to fluorescence studies with the novel membrane probes B
and C of similar overall length [8] (Fig. 1). It was surprising to find that not only the
diphospholipidic fluorene probe B, but also the fluorene diester C, carrying distal ester
groups, are apparently, as judged by their fluorescence properties, well oriented
perpendicularly to the bilayer [9].
We have planned to incorporate probes B or C into membranes so that, if they
approached tryptophan residues of a protein, there could be fluorescence-energy
transfer between the tryptophan residue and the probe (Fig. 2). However, we found
that there is an important overlap between the absorption spectra of a tryptophan
Fig. 1. Structure of transmembrane photosensitive or fluorescent probes

Fig. 2. Schematic representation of how to determine the topography of proteins in the membrane: photolabelling method (left) and fluorescence-energy-transfer method
(right)

Helvetica Chimica Acta ± Vol. 83(2000)
2467
(donor) and the diacetylenic fluorene chromophore (acceptor) of both B and C. This
overlap would prevent an accurate evaluation of the distance between donor and
acceptor groups. To circumvent this problem, we have now synthesized the anthracene
diesters D and E. We show that probe E has favorable properties for fluorescence-
energy transfer in membranes. Moreover, linear-dichroism experiments demonstrate
that probe E is inserted perpendicularly to the bilayer plane. Therefore, this probe
might enable us to quantify the donor-acceptor distance and to determine the location
of tryptophan residues in membranes. While the corresponding diacids might have
been better anchored in the water compartments on both sides of the membranes, it
was found unnecessary to study them, as the diesters were apparently already
transverse, and furthermore, the diacids were insoluble.
Results and Discussion. ± Synthesis. We used 2,6-diaminoanthraquinone (1) as the
starting material to prepare both probes D and E (Scheme 1). Thus, quinone 1 was
transformed into 2,6-dibromoanthracene (2) [10], and the two unbranched acyl chains
were introduced using a Sonogashira-Hagihara-type cross-coupling [11] ([Pd(Ph3P)2Cl2]/
CuI catalysis, Et2NH) with the acetylenic derivative 3. Probe D was isolated as bright
yellow crystals in 70% yield. Reduction of both triple bonds of D was best achieved by a
slurry of activated Raney-Ni, leading to probe E in 55% yield.
Scheme 1. Synthesis of the Fluorescent Probe E
i
a) 1. CuBr2, MeCN, 808; 2. NaBH4, MeOH; 3. 5m HCl, 908, 6 h; 4. NaBH4, PrOH. b) [Pd(Ph3P)2Cl2], CuI,
Et2NH, reflux, 4 h. c) Raney-Ni, THF, r.t.
Alternatively, we ran first a Sonogashira-Hagihara-type cross-coupling between 2,6-
dibromoanthraquinone (4) and the acetylenic derivative 3 (Scheme 2). Compound 5 was
obtained in 73% yield. The reduction was carried out in AcOH/CHCl3 1:1 in the presence
of PtO2 as catalyst under H2 at 10 bar; this gave the dihydro derivative 6 in 31% yield.
Finally, a DDQ (2,3-dichloro-5,6-dicyano-p-benzoquinone ˆ 4,5-dichloro-3,6-dioxocy-
clohexadiene-1,2-dicarbonitrile) oxidation of compound 6 led to probe E in 50% yield.

2468
Helvetica Chimica Acta ± Vol. 83(2000)
Scheme 2. Alternative Synthesis of the Fluorescent Probe E
a) [Pd(Ph3P)2Cl2], CuI, Et2NH, reflux, 8 h. b) H2, PtO2, AcOH/CHCl3, 10 bar, r.t. c) DDQ, dioxane, reflux, 6 h.
Absorption and Fluorescence Characterization. Spectra were first measured in
CHCl3 solution, and comparison was systematically made with anthracene, considered
as a reference compound for this kind of derivatives. Absorption spectra are shown in
Fig. 3 for the first absorption band (S0 ! S1) and in Fig. 4 for the second absorption
band (S0 ! S2). Concerning the first absorption band, both probes show properties
Fig. 3. Absorption spectra (S0 ! S1 electronic transition) of anthracene ( ´´´´ ) and probes D and E in CHCl3

Helvetica Chimica Acta ± Vol. 83(2000)
2469
Fig. 4. Absorption spectra (S0 ! S2 electronic transition) of anthracene ( ´´´´ ) and probes D and E in CHCl3
similar to those of anthracene, with four well-defined vibronic peaks (although
somewhat smoothed for probe E). Compared to anthracene, an overall red-shift is
observed, more pronounced for D (ca. 15 nm) than for E (ca. 6 nm), and accompanied,
in the case of probe D, by a strong hyperchromism. Both absorption spectra present a
good overlap with the fluorescence-emission spectrum of tryptophan.
Concerning the second absorption band, it is important to note the strong
absorbance of probe D in the range 275 ± 295 nm, with a huge maximum of the molar
extinction coefficient (2.86 ´ 105 molÀ1 cmÀ1 at 288 nm). This wavelength region,
unfortunately, corresponds to the absorption spectrum of tryptophan, which would
preclude any use of this probe as an acceptor in fluorescence-energy-transfer
experiments with tryptophan-containing peptides or proteins, since donor (tryptophan)
and acceptor would be excited by the same wavelengths. Contrary to probe D, probe E
presents only a weak absorbance in this wavelength range, and consequently could be a
good acceptor for tryptophan fluorescence.
Fluorescence-emission spectra (Fig. 5) were obtained by exciting solutions of
identical optical densities (OD ˆ 0.1) at the excitation wavelength (360 nm), so that
these spectra show the relative fluorescence quantum yields of both probes and
anthracene in CHCl3. By considering the areas under these emission spectra (with the
scale expressed in wavenumbers rather than in wavelengths), the fluorescence
quantum yield of probe D is 2.1 times higher than the anthracene one, while that of
probe E is only 1.4 times higher. The absolute quantum yields, determined with quinine
hydrogen sulfate as a reference (in 0.05m H2SO4, quantum yield ˆ 0.51) [12] are found

2470
Helvetica Chimica Acta ± Vol. 83(2000)
Fig. 5. Fluorescence-emission spectra of anthracene ( ´´´´ ) and probes D and E in CHCl3 (excitation wavelength
360 nm; optical density 0.1)
to be 0.28 for D and 0.19 for E. Like for the first-band absorption spectra, the emission
spectra show four well-defined vibronic peaks, with an overall red-shift of ca. 24 nm for
probe D but only 13nm for probe E. The higher quantum yield and red-shift observed
for probe D are probably due to the presence of the triple bonds, leading to a stronger
conjugation of the molecular orbitals of the excited molecules. Only a single
fluorescence lifetime was measured for both probes in CHCl3, namely 3.1 ns for probe
D and 3.9 ns for probe E.
To appraise the suitability of probe E for its use as an energy-transfer acceptor and
to investigate spatial relationships with tryptophan residues of membrane proteins, we
evaluated the critical distance R0 between a tryptophan fluorophore and probe E, accord-
ing to Försterꢀs equation [13] (see Eqn. 1), where R0 is the distance for which 50% of the
energy transfer is efficient, k2 the orientation factor between donor and acceptor (averaged
as 2 :3), FD the absolute quantum yield of the donor in the absence of acceptor, and n
the refraction index of the medium. J is the overlap integral between donor and
acceptor, according to Eqn. 2, where FD is the peak-normalized emission spectrum of
the donor and eA(l) the molar extinction coefficient vs. wavelength of the acceptor.
R0 ˆ 8.785 ´ 10À5k2FDJ/n4
(1)
(2)
6
„
J ˆ FD(l) eA(l) l4 dl

Helvetica Chimica Acta ± Vol. 83(2000)
2471
R0 was determined for two limiting cases, namely a buried (in our case inside the
membrane) or an exposed tryptophan (at the lipid-water interface). These limiting
cases may be mimicked by N-acetyl-l-tryptophanamide either dissolved in dioxane
(buried; blue-shifted emission spectrum, FD ˆ 0.30) or in H2O (exposed; red-shifted
emission spectrum, FD ˆ 0.14). Very close values of R0 were found: 24 Š for a buried
tryptophan and 23Š for an exposed one. Such values indicate that probe E should be
able to localize a tryptophan moiety whatever its position within the membrane bilayer.
Fluorescence characteristics of probes D and E embedded in large unilamellar
vesicles of DMPC (dimyristoylphosphatidylcholine) were also determined. Excitation
and emission fluorescence spectra obtained in the La liquid-crystalline phase are
reported in Fig. 6, and are similar to those obtained in CHCl3; quite identical spectra
are obtained in the Lb gel phase. Absolute quantum yields and fluorescence-decay
parameters are reported in Table 1. Both quantum yields and average decay times are
found to be higher than in CHCl3 solution; these fluorescence characteristics do not
appear to be very sensitive to the phase transition of DMPC vesicles, which is consistent
with the hypothesis of a transmembrane topography of such diesters, as was previously
shown with the fluorene probe C [9].
Linear-Dichroism Measurements on Shear-Deformed Liposomes. To obtain clear
indications concerning the transmembrane localization of probes D and E, we
performed linear-dichroism (LD) experiments on shear-deformed DMPC unilamellar
vesicles. We chose to examine the linear dichroism of the long axis of the compounds,
the short-axis transition generally being too weak to be observable at the concen-
trations used here (see Exper. Part). All compounds show negative values of the linear
dichroism (LD) and reduced linear dichroism (LDr) (Table 2, Fig. 7). The diacetylenic
diesters C and D have the lowest values of the reduced linear dichroism, followed by
the free anthracene and fluorene molecules, whereas the saturated diester E has a
reduced linear dichroism about five times as large as the LDr of the other compounds.
In contrast, the diesters all show high orientation parameters in stretched poly-
ethylene film (Szz ꢀ 0.7), whereas the free anthracene and fluorene have poorer orien-
tation.
The results in stretched polyethylene films showed that all three diesters behave as
rod-shaped molecules with high orientation parameters [14]. In bilayers, all compounds
exhibit negative linear dichroism in the region of the long-axis transition, and are thus
oriented with their long axes perpendicular to the bilayer plane [15]. The saturated
diester E has a much better orientation than the diacetylenic diesters C and D, which
are only oriented to the same degree as the free fluorene and anthracene molecules.
There are several possible explanations for the behavior of probe D in the bilayer. The
anthracene is substituted in 2 and 6 positions, imposing a certain tilt on the long axis of
the anthracene, especially in the case of the rigid diacetylenic diester D, and thus
lowering its LDr. However, such a tilt should lower the LDr only by 25 ± 30%. Rather,
the answer could be that the diester D with triple bonds is slightly too short to span the
DMPC membrane adequately. In this case, the moderately hydrophilic ester groups,
which are intended to have an anchoring effect, could not anchor the probe with its two
heads at the interface. Hence, one end of the diester may be drifting inside the bilayer,
making it possible for the molecule not to remain perpendicularly oriented and thereby
lowering the LD.

2472
Helvetica Chimica Acta ± Vol. 83(2000)
Fig. 6. Excitation and emission fluorescence spectra of probes D and E in DMPC large unilamellar vesicles, with
a lipid-to-probe ratio of 100 :1
These results lead to the conclusion that the saturated diester E, possessing
appropriate fluorescence properties to be an acceptor in fluorescence-energy-transfer
experiments with tryptophan as a donor, presents a far better transmembrane
orientation within the membrane than the unsaturated diesters C and D. To the best
of our knowledge, this is the first time that a simple long-chain molecule like E,
terminated by ester groups, is shown to be oriented perpendicularly to the bilayer plane
with its fluorophore positioned near the center of the bilayer. Therefore, this probe

Helvetica Chimica Acta ± Vol. 83(2000)
2473
Table 1. Absolute Quantum Yields (F; Æ0.005) and Fluorescence Lifetimes (t) for Probes D and E Embedded
in DMPC Large Unilamellar Vesicles in La (308) and Lb (158) Phases, at a Lipid-to-Probe Ratio of 100 : 1
b
b
c
T [8C]
F
t1a) [ns]
a1
)
t2a) [ns]
a2
)
hti ) [ns]
D
D
E
E
15
30
15
30
0.36
0.35
0.35
0.34
6.60
6.50
7.95
7.45
0.92
0.93
0.53
0.60
1.85
1.60
1.45
1.45
0.08
0.07
0.47
0.40
6.20
6.15
4.90
5.05
a) Fluorescence lifetimes ti (Æ 0.05) were obtained by the phase method, with frequencies of the exciting light
scanned from 3to 90 MHz. Diphenylhexatriene in hexane was chosen as fluorescence-lifetime reference
(t 6.75 ns) [17]. b) ai (Æ 0.001): Relative contribution of each lifetime component. c) hti: Average fluorescence
lifetime.
Table 2. Linear Dichroism (LD) of Compounds Dissolved in the Bilayer of Shear-Deformed Vesicles at 358 and
3100 sÀ1 and Orientation Parameters Szz in Stretched Polyethylene Film
Long-axis Linear di- Absorp- Reduced
Szz in stretched
transition chroism
tion Aiso linear dichroisma) polyethylene film
LDr ´ 103
[nm]
LD ´ 103
Anthracene
Diacetylenic anthracene diester D 285
Sat. anthracene diester E
Fluorene
Diacetylenic fluorene diester C
252
À 4.8
À 2.1
À 2.1
À 2.9
À 2.0
0.68
0.37
0.048
0.35
0.74
À 7.1
À 5.7
0.3
0.7
0.6
260
266
330
À 44
À 8.30.22
À 2.7
0.7
a) Reduced linear dichroism LDr ˆ LD/Aiso
.
might enable us to quantify the donor-acceptor distance and to determine the location
of tryptophan residues in membranes.
Professor P. Hodges is gratefully thanked for providing us the experimental procedure leading to the
Â
synthesis of 2,6-dibromoanthracene. E. Quesada is grateful for grants from the Ministerio de Educacion y
Â
Â
Â
Â
Cultura of Spain and the Societe des Amis des Sciences of France. Y. Bonzi-Coulibaly is indebted to the Societe
Â
des Amis des Sciences of France for a grant. We also wish to thank Prof. H. Bouas-Laurent, Universite de
Â
Â
Bordeaux 1, and Dr. L. Desaubry, Universite Louis Pasteur, for stimulating discussions. This work was
Â
Â
Â
supported in part by the Direccion General de Investigacion Científica y Tecnica of Spain (projects PB93-0126
and PB96-0852) and by the European Union (DG XII, contract PL950990).
Experimental Part
General. All solvents were freshly distilled prior to use. THF and Et2O were distilled from sodium metal-
benzophenone ketyl, and MeOH, EtOH, iPrOH, MeCN and 1,4-dioxane from CaH2. Air- or moisture-sensitive
reactions were conducted in dried glassware and under a positive pressure of dry Ar. All reagents were
commercial products purchased from Aldrich, Fluka, Lancaster, Acros, or other suppliers. 2,6-Dibromoan-
thracene (2), methyl undec-10-ynoate (3), and 2,6-dibromoanthraquinone (4) were prepared according to
recently published procedures [8][10]. All org. extracts were dried (MgSO4) and concentrated in a rotary
evaporator. TLC: Precoated aluminium silica-gel plates 60 F 254 (Merck, 0.25 mm); detection under UV lamps
at 254 or 360 nm, by I2 vapors, or by dipping into a soln. of vanillin (1 g) in EtOH/H2SO4 95 :5 (1 l) or
phosphomolybdic acid (5 g) in EtOH (250 ml), followed by heating on a hot plate. Prep. medium-pressure
column chromatography (CC): pressure 0.5 ± 1.1 bar, silica-gel columns (40 ± 63mm, Merck); according to the
standard procedure [16]. M.p.: Reichert-Kofler hot-stage microscope; uncorrected. IR Spectra: KBr pellets;
Perkin-Elmer-881 or -681 IR spectrophotometers; in cmÀ1 1H- and 13C-NMR Spectra: Bruker-SY-200
.

2474
Helvetica Chimica Acta ± Vol. 83(2000)
Fig. 7. Linear dichroism (LD) of shear-deformed DMPC vesicles at 358 and 3100 sÀ1: DMPC vesicles (Ð),
DMPC vesicles with probe E ( ´´´´ ) and DMPC vesicles with probe D (---). Probe/lipid ratio 1:100. The blank
DMPC vesicles exhibit strong scattering with increasing positive LD when passing from 500 to 250 nm; the
negative LD at shorter wavelength is due to lipid absorption. The two peaks at 285 and 260 nm show that their
anthracenediyl long axes are preferentially oriented along the lipid chains (for theory of interpretation of shear
LD of liposomes, see [14]).
(200 MHz), Bruker-AM-200 (200 MHz), Bruker-AM-400 (400 MHz), Varian-Gemini-200 (200 MHz), Varian-
INOVA (300 MHz), or Varian-INOVA-400 (400 MHz) spectrometers; signal of the residual non-deuterated
solvent as internal reference: d(H) 7.26 or 2.49 ppm and d(C) 77.0 or 39.7 ppm for spectra in CDCl3 or
(D6)DMSO, resp., J values in Hz; 13C-assignments by 13C,1H polarization transfer (DEPT) and 13C,1H
heteronuclear single quantum correlation (HSQC) experiments. MS: Direct injection on a Trio-Fisons or a
Hewlett-Packard-5973 spectrometer, electron impact (EI) technique; in m/z (% of basis peak). Elemental
analyses were carried out on a Carlo Erba 1108 elemental analyzer.
Dimethyl 11,11'-(Anthracene-2,6-diyl)bis[undec-10-ynoate] (D). To freshly crystallized 2,6-dibromoan-
thracene (2; 100 mg, 0.298 mmol, 1 equiv.), methyl undec-10-ynoate (3; 175.2 mg, 0.893mmol, 3equiv.),
[Pd(Ph3P)2Cl2] (41.8 mg, 59.6 mmol, 0.20 equiv.), and CuI (5.7 g, 29.8 mmol, 0.10 equiv.) in a Schlenk-type flask
under continuous Ar stream, freshly distilled deoxygenated Et2NH (15 ml) was added, and the mixture was
stirred and heated under reflux for 4 h. The brown crude soln. was cooled and filtered with CH2Cl2 through a
short Celite pad, and the soln. was evaporated. The residual solid was washed with pentane (5 Â 25 ml portions)
until disappearance of non-fluorescent by-products (TLC), and further purified by CC (hexane/AcOEt 9 :1,
then hexane/Et2O 3:2). The crude product was recrystallized from acetone: 118 mg (70%) of D. Bright yellow
crystals. M.p. 101 ± 1038. Rf 0.25 (hexane/AcOEt 9 :1), 0.5 (hexane/Et2O 3:2). IR (KBr): 2890, 2810, 2182, 1712,
1605, 1448, 1395, 1358, 1300, 1262, 1227, 1187, 1150, 879, 779. 1H-NMR (400 MHz, CDCl3): 1.35 (br. s, 12 H,
CH2); 1.54 (m, 4 H, CH2); 1.63( m, 4 H, CH2); 1.65 (m, 4 H, CH2); 2.31 (t, 3J ˆ 7.32, 4 H, CH2); 2.47 (t, 3J ˆ 7.17,
4 H, CH2); 3.66 (s, 2 Me); 7.38 (dd, 4J ˆ 1.31, 3J ˆ 8.94, HÀC(3), HÀC(7)); 7.86 (d, 3J ˆ 8.9, HÀC(4), HÀC(8));
8.03( s, HÀC(1), HÀC(5)); 8.27 (s, HÀC(9), HÀC(10)). 13C-NMR (50 MHz, CDCl3): 19.57; 24.91; 28.75;
28.88; 28.94; 29.11 (2Â); 34.09; 51.40; 81.19; 91.65; 121.06; 125.81; 128.04; 128.37; 130.82; 131.19; 131.41; 174.26.

Helvetica Chimica Acta ± Vol. 83(2000)
2475
‡
EI-MS (70 eV): 566 (100, M ), 409 (12), 295 (6), 279 (15), 265 (17), 252 (36), 239 (16), 229 (20), 215 (11), 74
(24). Anal. calc. for C38H46O4: C 80.53, H 8.18; found: C 80.64, H 8.21.
Dimethyl 11,11'-(Anthracene-2,6-diyl)bis[undecanoate] (E). Procedure A: Triple-Bond Reduction of D by
Raney-Ni Catalysis. Caution: The catalyst must never be dried and must be handled carefully. A slurry of
activated (ꢁready-to useꢀ) Raney-Ni in H2O was successively washed and decanted with Milli-Q water, EtOH,
i
and PrOH (4Â) in an ultrasound bath. The Raney-Ni slurry was then added to D (51 mg, 90 mmol) in THF
(15 ml) under Ar at r.t. under efficient stirring (TLC (hexane/Et2O 3:2; det. 360 nm) monitoring and color
change from yellow (D) to colorless (E); the catalyst surface taking a characteristic blue color). After ca. 1 h,
the stirring was stopped and the soln. decanted. The remaining solid catalyst was washed with CH2Cl2 (3 Â
20 ml) and the combined decanted phase filtered through a Celite pad, dried (MgSO4), and evaporated to afford
a white-yellowish solid (50 mg) which was purified by recrystallization from acetone (208): 28 mg (55%) of E.
Procedure B: DDQ Oxidation of 9,10-Dihydro Derivative 6.
A soln. of dimethyl 11,11'-(9,10-
dihydroanthracene-2,6-diyl)bis[undecanoate] (6, 8 mg, 14 mmol) in dry 1,4-dioxane (3ml) was added through
a canula into a flask containing DDQ (freshly recrystallized from CHCl3; 4.5 mg, 20 mmol) under Ar, with
protection from light, and heating under reflux for 6 h. The solid DDQH2 formed was separated by filtration
through a Celite pad (CH2Cl2 as eluent), the filtrate evaporated, and the crude solid purified by CC (hexane/
AcOEt 3:2) yielding E (7.6 mg, 94%) as a yellowish solid. Further recrystallization from acetone afforded anal.
pure E (4.2 mg, 50%). M.p. 108 ± 1108. Rf 0.55 (hexane/Et2O 3 :2). IR (KBr): 3400, 2887, 2818, 1721, 1638, 1502,
1438, 1394, 1212, 1181, 1146, 887, 865. 1H-NMR (400 MHz, CDCl3): 1.28 (br. s, 20 H, CH2); 1.35 (m, 4 H, CH2);
1.61 (quint. 3J ˆ 7.6, 4 H, CH2); 1.73( quint., 3J ˆ 7.6, 4 H, CH2); 2.29 (t, 3J ˆ 7.47, 4 H, CH2); 2.78 (t, 3J ˆ 7.47, 4 H,
CH2); 3.66 (s, 2 Me); 7.30 (dd, 4J ˆ 1.47, 3J ˆ 8.64, HÀC(3), HÀC(7)); 7.72 (s, HÀC(1), HÀC(5)); 7.90 (d, 3J ˆ
8.64, HÀC(4), HÀC(8)); 8.29 (s, HÀC(9), HÀC(10)). 13C-NMR (50 MHz, CDCl3): 24.93; 29.12; 29.23; 29.33;
29.41; 29.49 (2Â); 30.98; 34.09; 36.24; 51.41; 125.00; 125.75; 127.37; 127.85; 130.65; 131.50; 139.29; 174.33. EI-MS
‡
(70 eV): 574 (100; M ); 389 (13), 231 (6), 217 (18), 204 (33), 191 (17), 74 (16), 69 (6). Anal. calc. for C38H54O4:
C 79.45, H 9.47; found: C 79.28, H 9.28.
Dimethyl 11,11'-(9,10-Dioxoanthracene-2,6-diyl)bis[undec-10-ynoate] (5). A mixture of 2,6-dibromoan-
thraquinone (4) (500 mg, 1.37 mmol), 3 (805 mg, 4.1 mmol, 3equiv.), [Pd(Ph 3P)2Cl2] (96 mg, 0.137 mmol,
0.1 equiv.), CuI (26.1 mg, 0.137 mmol, 0.1 equiv.), and degassed Et2NH (50 ml) in a Schlenk-type flask was
heated under reflux and under Ar for 8 h. The crude mixture was then cooled to r.t. and filtered with CH2Cl2
through a short Celite pad, and the soln. was evaporated. The crude solid was washed with pentane (5 Â 25 ml)
and the remaining solid purified by CC (hexane/Et2O/CH2Cl2 3:2 :0.2 ! 1 :4 :0.5). The obtained yellow solid
was washed with pentane and dried: 5 (598 mg, 73%). Light yellow crystals. M.p. 111 ± 1138. Rf 0.45 (hexane/
Et2O/CH2Cl2 3:2 :0.2). IR (KBr): 3420, 2890, 2820, 2190, 1715, 1658, 1578, 1452, 1409, 1289, 1245, 1210, 1150,
959, 903, 897, 831, 719. 1H-NMR (300 MHz, CDCl3): 1.34 (br. s, 12 H, CH2); 1.46 (m, 4 H, CH2); 1.63
(quint., 3J ˆ 6.9, 8 H, CH2); 2.31 (t, 3J ˆ 7.5, 4 H, CH2); 2.46 (t, 3J ˆ 6.9, 4 H, CH2); 3.66 (s, 2 Me); 7.75 (dd, 4J ˆ
1.8, 3J ˆ 8.1, HÀC(3), HÀC(7)); 8.22 (d, 3J ˆ 8.1, HÀC(4), HÀC(8)); 8.27 (d, 3J ˆ 1.5, HÀC(1), HÀC(5)).
13C-NMR (50 MHz, CDCl3): 19.57; 24.91; 28.42; 28.85; 28.91; 29.08; 34.06; 51.40; 79.75; 96.37; 127.25; 130.17;
‡
130.70; 131.81; 133.41; 136.59; 174.22; 182.16. EI-MS (70 eV): 596 (53, M ), 564 (34), 532 (12), 504 (8), 476 (7),
311 (11), 297 (11), 259 (8), 226 (16), 123 (10), 95 (29), 55 (100). Anal. calc. for C38H44O6: C 76.48, H 7.43; found:
C 76.70, H 7.24.
Dimethyl 11,11'-(9,10-Dihydroanthracene-2,6-diyl)bis[undecanoate] (6). Through a soln. of 5 (108 mg,
0.19 mmol) AcOH/CHCl3 1 :1 (10 ml) in a high-pressure hydrogenation flask, Ar was bubbled for 15 min. Then,
PtO2 (49 mg, 0.21 mmol) was added under Ar, and the system was closed. After 3vacuum/H 2 cycles, the mixture
was stirred under H2 (10 bar) at r.t. for 16 h. The resulting org. layer was washed with H2O, dried (MgSO4), and
evaporated. The white solid (90 mg) was further purified by CC (hexane/Et2O 3:2): pure 6 (34.4 mg, 31%).
Colorless crystals. M.p. 65 ± 678. Rf 0.59 (hexane/Et2O 3 :2). IR (KBr): 3400, 2885, 2818, 1730, 1452, 1400, 1247,
1217, 1179, 1151, 812, 781. 1H-NMR (300 MHz, CDCl3): 1.27 (br. s, 24 H, CH2); 1.61 (quint., 3J ˆ 7.2, 4 H, CH2);
2.30 (t, 3J ˆ 7.5, 4 H, CH2); 2.57 (t, 3J ˆ 7.8, 4 H, CH2); 3.66 (s, 2 Me); 3.88 (s, 4 H, CH2); 6.99 (dd, 4J ˆ 1.8, 3J ˆ 7.5,
HÀC(3), HÀC(7)); 7.10 (d, 4J ˆ 1.8, HÀC(1), HÀC(5)); 7.19 (d, 3J ˆ 7.5, HÀC(4), HÀC(8)). 13C-NMR
(50 MHz, CDCl3): 24.96; 26.98; 29.15; 29.22; 29.33; 29.47; 31.69; 34.12; 35.63; 35.8; 36.61; 51.39; 126.04; 127.21;
‡
127.42; 133.94; 136.72; 140.67; 174.30. EI-MS (70 eV): 576 (50, M ), 544 (68), 516 (87), 345 (36), 205 (56), 191
(100), 179 (33), 69 (16).
Vesicle Preparations for Fluorescence and Linear-Dichroism Experiments. Large unilamellar vesicles
(LUV) were prepared by the extrusion method: DMPC and the various probes (diesters, anthracene, or
fluorene) were dissolved in CHCl3 and mixed in appropriate proportions (50 :1 to 100 :1). The CHCl3 was then
evaporated. The liquid film thus obtained was hydrated with an aqueous buffer (2 mm, TES ( ˆ 2-

2476
Helvetica Chimica Acta ± Vol. 83(2000)
{[tris(hydroxymethyl)methyl]amino}ethane-1-sulfonic acid), 2 mm l-histidine, 0.1m NaCl, pH 7.4 for fluores-
cence experiments; 0.15m PBS for linear-dichroism experiments) and vortexed vigorously for 5 min to give a
soln. of multilamellar vesicles (MLV). LUV were obtained by extrusion of the previous MLV soln. through
polycarbonate filters (Nucleopore) at 3 58 by means of a thermostated extruder (Lipex Biomembranes). For
fluorescence experiments (concentration of lipids ca. 1 mm), a first extrusion was carried out on 200 nm filters (7
passages) and a second one on 100-nm filters (11 passages); for linear-dichroism experiments (concentration ca.
10 mm), only the extrusion step on 100-nm filters (11 passages) was carried out, but preceded by freeze-thawing
in liq. N2 (5 cycles) to homogenize the MLV soln. The method provided homogeneous vesicles (110 ± 120 nm
diameter), as checked by light scattering with a N4SD Coultronics Nanosizer.
Absorption and Fluorescence Spectra. UV Spectra: Kontron-Uvikon-810 or Cary-4 UV/VIS spectropho-
tometer. Fluorescence-emission spectra: SLM-48000 spectrofluorimeter in the steady-state mode; fluorescence
lifetimes by the phase and modulation method in the time-resolved mode.
Linear-Dichroism Measurements. Linear-dichroism (LD) measurements were performed at 358 in a
Couette cell with a Jasco-500 instrument. A baseline was recorded at low shear gradient, typically 100 sÀ1, while
the measurement was performed with a shear gradient of 3100 sÀ1. The isotropic absorption (Aiso) of the soln.
was measured, and the reduced linear dichroism (LDr) could be calculated as LD/Aiso. Stretched polyethylene
films were prepared following the standard procedure: after stretching the film to five times its initial length, it
was allowed to swell with CHCl3, and baselines were collected for the absorption parallel and perpendicular to
the stretching direction. Anthracene or fluorene diesters in CHCl3 were then applied onto the film, the CHCl3
was evaporated, and excess sample was washed off the film with EtOH. Polarized absorption spectra ± parallel
and perpendicular to the stretching direction ± were collected. The reduced linear dichroism could be calculated
as LDr ˆ 3 ´ (Apar À Aperp)/(Apar ‡ 2 ´ Aperp). For a pure long-axis-polarized transition, LDr ˆ 3 ´ Szz, where Szz is the
orientation parameter [15].
REFERENCES
[1] E. M. Landau, J. P. Rosenbusch, Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 14532.
[2] H. G. Khorana, Bioorg. Chem. 1980, 9, 363; J. Brunner, Annu. Rev. Biochem. 1993, 62, 483; L. Peng, M. L.
Alcaraz, P. Klotz, F. Kotzyba-Hibert, M. Goeldner, FEBS Lett. 1994, 346, 127.
[3] A. M. Kleinfeld, Biochemistry 1985, 24, 1874; A. M. Kleinfeld, M. F. Lukacovic, Biochemistry 1985, 24,
1883.
[4] T. Markellov, A. Zlotnick, J. Everett, J. Tennyson, P. W. Holloway, Biochemistry 1985, 24, 2895; E. Blatt,
W. H. Sawyer, Biochim. Biophys. Acta 1985, 822, 43; T. J. McIntosh, P. W. Holloway, Biochemistry 1987, 26,
1783; M. C. Wiener, J. H. White, Biochemistry 1991, 30, 6797.
[5] M. Yamamoto, W. Warnock, A. Milon, Y. Nakatani, G. Ourisson, Angew. Chem. 1993, 105, 302; Angew.
Â
Chem., Int. Ed. 1993, 32, 229; M. Yamamoto, V. Dolle, W. Warnock, Y. Diyizou, M. Yamada, Y. Nakatani,
G. Ourisson, Bull. Soc. Chim. Fr. 1994, 131, 317; Y. Nakatani, M. Yamamoto, Y. Diyizou, W. Warnock, V.
Â
Dolle, W. Hahn, A. Milon, G. Ourisson, Chem. Eur. J. 1996, 2, 129.
[6] J. M. Delfino, S. L. Schreiber, F. M. Richards, J. Am. Chem. Soc. 1993, 115, 3458.
[7] B. J. Gaffney, G. L. Willingham, R. S. Schepp, Biochemistry 1983, 22, 881; G. L. Willingham, B. J. Gaffney,
Biochemistry 1983, 22, 892.
[8] J. P. Starck, Y. Nakatani, G. Ourisson, Tetrahedron 1995, 51, 2629.
[9] J. P. Starck, Y. Nakatani, G. Ourisson, D. J. Cowley, G. Duportail, New J. Chem. 1996, 20, 1293.
[10] P. Hodge, G. A. Powers, M. A. Rabjohns, Chem. Commun. 1997, 73 .
[11] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 50, 4467; S. Takahashi, Y. Kuroyama, K.
Sonogashira, N. Hagihara, Synthesis 1980, 627.
[12] R. A. Velapoldi, K. D. Mielenz, ꢁNBS Special Publication 260-64: Standard Reference Materials: A
Fluorescence Standard Reference Material: Quinine Sulfate Dihydrateꢀ, U.S. Department of Commerce,
Washington, 1980.
[13] P. Wu, L. Brand, Anal. Biochem. 1994, 218, 1.
Â
[14] A. Rodger, B. Norden, ꢁCircular Dichroism and Linear Dichroismꢀ, Oxford University Press, Oxford-New
York, 1997.
Â
[15] M. Ardhammar, N. Mikati, B. Norden, J. Am. Chem. Soc. 1998, 120, 9957.
[16] W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923.
[17] D. A. Barrow, B. R. Lentz, J. Biochem. Biophys. Methods 1983, 7, 217.
Received June 2, 2000