Benzophenone-containing fatty acids and their related photosensitive fluorescent new probes: Design, physico-chemical properties and preliminary functional investigations

Article abstract of DOI:10.1016/j.bmc.2011.10.043

Hydrophobic photoaffinity labeling is a powerful strategy to identify hydrophobic segments within molecules, in particular membrane proteins. Here we report the design and synthesis of a novel family of fluorescent and photosensitive lipid tools, which have a common amino acid scaffold functionalized by three groups: (i) a first fatty acid chain grafted with a photoactivatable benzophenone moiety (Fatty Acid BenzoPhenone, FABP), (ii) a second fatty acid chain to ensure anchoring into a half-bilayer or hydrophobic environment, and (iii) a fluorescent carboxytetramethylrhodamine headgroup (CTMR) to detect the photolabeled compound. We present data of the synthesis and characterization of three lipid tools whose benzophenone ring is situated at various distances from the central scaffold. We could therefore establish structure/properties relationships dependent upon the depth of insertion of benzophenone into the membrane. Our lipid tools were extensively characterized both physico- and bio-chemically, and we assessed their functionality in vitro using bacterioRhodopsin (bR). We thus provide the scientific community with novel and reliable tools for the identification and study of hydrophobic regions in proteins.

Full text of DOI:10.1016/j.bmc.2011.10.043

Bioorganic & Medicinal Chemistry 19 (2011) 7464–7473
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Benzophenone-containing fatty acids and their related photosensitive
fluorescent new probes: Design, physico-chemical properties
and preliminary functional investigations
Benoît Hilbold a, Marie Perrault b, Christophe Ehret a, Song-Lin Niu a, Benoît Frisch a,
a,
Eve-Isabelle Pécheur b, , Line Bourel-Bonnet
⇑
⇑
a UMR 7199 CNRS/Université de Strasbourg, Equipe de Biovectorologie, Faculté de Pharmacie, 74 route du Rhin, 67401 Illkirch Cedex, France
b UMR 5086 CNRS/Université Lyon 1; IBCP, 7 passage du Vercors, 69367 Lyon Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Hydrophobic photoaffinity labeling is a powerful strategy to identify hydrophobic segments within mol-
ecules, in particular membrane proteins. Here we report the design and synthesis of a novel family of
fluorescent and photosensitive lipid tools, which have a common amino acid scaffold functionalized by
three groups: (i) a first fatty acid chain grafted with a photoactivatable benzophenone moiety (Fatty Acid
BenzoPhenone, FABP), (ii) a second fatty acid chain to ensure anchoring into a half-bilayer or hydrophobic
environment, and (iii) a fluorescent carboxytetramethylrhodamine headgroup (CTMR) to detect the
photolabeled compound. We present data of the synthesis and characterization of three lipid tools whose
benzophenone ring is situated at various distances from the central scaffold. We could therefore establish
structure/properties relationships dependent upon the depth of insertion of benzophenone into the
membrane. Our lipid tools were extensively characterized both physico- and bio-chemically, and we
assessed their functionality in vitro using bacterioRhodopsin (bR). We thus provide the scientific commu-
nity with novel and reliable tools for the identification and study of hydrophobic regions in proteins.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 29 August 2011
Revised 14 October 2011
Accepted 14 October 2011
Available online 20 October 2011
Keywords:
Fatty acid
Benzophenone
Photolabeling
Fluorescence
Membrane protein
1. Introduction
signed and their functionality studied. These tools have a common
amino acid scaffold derivatized by three groups: (i) a first fatty acid
Among all photoreactive chemical functions,1 benzophenone
(BP) is one of the most popular, especially in the field of protein/pro-
tein and protein/lipid interaction studies.2–4 BP activation wave-
length is usually around 365 nm, which limits protein degradation
and enables studies on cell cultures or other living systems. More-
over, the BP double ring has an unambiguous photoreactivity. Corre-
sponding activated radicals react in a covalent manner and high
yields with C–H groups5,6 present in their hydrophobic surround-
ings. This allows for a precise identification of the chemical environ-
ment. Finally BP is chemically stable in moderate ambient light and
compatible with particular synthesis conditions, like those encoun-
tered in solid-phase chemistry.7–9 As a consequence, BP groups are
widely used in numerous applications such as for example, photo-
affinity probes,10 photolithography11 or photocrosslinking.12
En route toward a study of protein/lipid interaction, a novel
family of fluorescent lipid tools aimed at photoaffinity labeling of
hydrophobic peptides within transmembrane proteins was de-
chain containing a photoactivatable benzophenone group (FABP),
(ii) a second fatty chain to ensure a stable anchoring into half-
bilayers or micelles, and (iii) a fluorescent carboxytetramethyl-
rhodamine headgroup (CTMR).
We designed a family of benzophenone-containing fatty acids
(FABPs) as building blocks displaying two features essential for
our study: (i) they are long enough to allow the probe insertion
into a lipid half-bilayer; (ii) they are photoactivatable, thereby
reacting with amino-acids present in their immediate vicinity. In
this context, the position of the BP double ring within the fatty
chain is of crucial importance.
Three lipid tools could be obtained in only few steps, within
overall yields considered as good for this kind of lipid compounds
(6% for 12 steps to 37% for four steps). Their fluorescent properties
were fully characterized and their photoreactivity was precisely
tuned. A relationship between photolabeling efficiency and depth
of membrane insertion of the lipid tools could then be established
using bacterioRhodopsin (bR) as a model membrane protein. More
generally, the reactivity and the good recovery of the photolabeled
protein indicate that this method can be transposed to other rele-
vant targets for the identification and study of ill-defined hydro-
phobic regions of proteins.
⇑
Corresponding authors. Tel.: +33 (0)3 6885 4143; fax: +33 (0)3 6885 4306 (L.B.-B.);
tel.: +33 (0)4 7272 2644; fax: +33 (0)4 7272 2604 (E.-I.P.).
E-mail addresses: e.pecheur@ibcp.fr (E.-I. Pécheur), lbourel@unistra.fr
(L. Bourel-Bonnet).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.10.043

B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
7465
Fatty chain
Fluorophore
Platform
Photoreactive
group
Figure 1. General structure of the lipid tools.
This choice is in accordance with the chemical partners involved
and the following of the synthesis.
O
H(CH2)m
2.1.2. Benzophenone
O
Benzophenone was chosen for the above-exposed reasons.
Benzophenone-bearing carboxylic acids were first synthesized by
Breslow et al. to study steroids17 and membrane structure.18
Hydrogen abstraction from lipids by triplet states of derivatized
BP photosensitizers was performed by Patterson and co-workers,19
notably in linoleate micelles.20 Lala and coworkers were among the
first to perform depth-dependent analysis of membranes21 by
designing suitable fluorescent and photoactivatable lipid tools con-
taining a FABP.22 More recently, some chemical routes to FABPs
were also reported, especially by Spencer and coworkers,23 to
study cellular cholesterol efflux and HDL formation by the mean
of analogs of cholesterol24 or phospholipids.25
(CH2)nC OH
d
Figure 2. General structure of the benzophenone-containing fatty acids (FABPs)
building blocks. m and n indicate the number of methylene. d is the distance (Å)
between the cross-linking site (–CO) and the oxygen atoms of the hydroxyl group in
the carboxylic acid (–COOH).
2. Bioorganic chemistry part
2.1. General conception of lipid tools with three functionalities
2.1.3. Rhodamine
2.1.1. The central platform and the chemical bonds
Detection of the photoaffinity labeling by fluorescence was pre-
ferred to detection by radioactivity as a safer and more ubiquitous
strategy. In order to follow the homogeneity of the bilayer inser-
tion of the lipid tool and to track the adducts formed after photo-
activation in a fine and sensitive way, we chose a fluorescent ring26
such as carboxytetramethylrhodamine. This residue has a good
stability in the envisaged conditions of synthesis; moreover rhodol
derivatives have a very good photostability and are relatively
insensitive to pH changes.
In the present case, it was necessary to design lipid tools able to
insert into lipid membranes where hydrophobic proteins are usu-
ally embedded. The partition of biomolecules modified by long al-
kyl chains between membranes and solutions has been extensively
studied. Epand13 showed that a stable insertion requires at least
two alkyl chains spatially close to each other. Moreover, the lipid
tools had to be photoreactive and detectable. Thus, at least three
chemical functionalities were required on a common platform
(Fig. 1). As we previously designed it,14 we chose stable chemical
bonds leading to efficient coupling, such as amides to assemble
the expected tools. For convenience, we chose a commercially
2.2. Preparation of the FABPs building blocks
available alpha amino-acid platform, that is the Na-Boc-L-2,3-dia-
2.2.1. Medium FABP, C6-BP-C5-COOH (1a)
minopropionic acid (Boc-Dpr-OH). It has the advantage of display-
ing three different functions capable of reacting selectively and
successively via a peptidic bond with three different partners. In
the past, we had already chosen an amino-acid as a platform for
such compounds, and demonstrated that functionality was pre-
served, without inducing any perturbation to the lipid bilayers.15,16
En route toward a depth-dependent exploration, we aimed at the
synthesis of a first FABP, where BP is situated in the middle of the
fatty chain as starting material. We applied Lala’s method27 toobtain
6-[4-(4-n-hexylbenzoyl)phenyl]hexanoic acid (‘C6-BP-C5-COOH’)
1a within a comparable overall yield. To implement the family, other
analogs and isomers were required, particularly one where BP was
O
O
Tolu¸ ne
Cl
AlCl3
75 %
Br
Br
(2)
(3)
O
O
MeOH
1. 9-BBN, THF
O
OH
H2SO4
93 %
2. Pd(dppf)Cl2, Cs2CO3, AsPh3
DMF, H2O, THF
69 %
O
O
O
(4)
O
KOH
95 % EtOH
93 %
OH
O
(1b)
Scheme 1. Convergent synthesis of distal FABP 1b.

7466
B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
O
Br
Cl
OH
1. MeOH, DMAP
2. LiAlH4, Et2O
3. Swern
DHP, TsOH, 92%
or
TIPSCl, imidazole, 99 %
77 %
O
H
OH
Br
nBuLi, -78 ºC
7a : 19 %
7b : 70 %
OR
OR
AcOH/THF/H2O
(5)
(6a-b)
(7a-b)
or
Bu4N-F+, THF, 73 %
OH
OH
OH
Jones
91 %
(8)
O
O
(1c)
Scheme 2. Convergent synthesis of proximal FABP 1c.
far from the carboxylic acid function (‘C1-BP-C10-COOH’, 1b, a distal
FABP), and another one where BP was close to the carboxylic acid
function (‘C10-BP-C1-COOH’, 1c, a proximal FABP).
high yield (77%), as a colorless oil, in a three-step procedure starting
from commercially available
4-decylbenzoyl chloride.
Briefly, the acyl chloride was converted in the corresponding
methyl ester, that was fully reduced into the corresponding
4-n-decylbenzylic alcohol (Scheme 2). The primary alcohol was oxi-
dized into 5 according to the procedure first described by Omura and
Swern.29 Though an alternate route was described by Rosemund,30
this route conducted successfully to the expected compound. In
parallel, we developed two strategies to protect the commercially
available 2-(4-bromophenyl)ethanol. First, dihydropyrane (DHP)
was employed31 and over 6 h, the corresponding acetal 6a was ob-
tained in 92% yield (Table 1). Nevertheless, it was tricky to condense
6a with aldehyde 5 and only 19% of the expected secondary alcohol
7a could be obtained.
An alternate strategy using triisopropylsilyl chloride (TIPSCl) in
the presence of imidazole allowed the conversion of 2-(4-bromo-
phenyl)ethanol into the corresponding ether 6b in high yield
(99%) over 2 h. Since our attempt to condense the protected bromo
compound 6b onto aldehyde 5 under Grignard conditions failed,
compound 6b was instead treated with n-butyl-lithium at ꢀ78 °C
2.2.2. Distal FABP, C1-BP-C10-COOH (1b)
The preparation of distal FABP (where m = 1 and n = 10, Figure 2)
followed the already reported protocol by Spencer24 with slight vari-
ations. Briefly, 4-bromo-40-methylbenzophenone was prepared
according to Nakatani et al.,28 by a Friedel–Crafts reaction of
4-bromobenzoyl chloride on toluene in the presence of aluminium
chloride. This electrophilic aromatic substitution occurred preferen-
tially in para position and gave the expected 4-bromo-40-methyl-
benzophenone within good yields (75%). In parallel, undecanoic
acid was first protected by esterification by methanol in the pres-
ence of a catalytic amount of sulfuric acid (93%), in order to avoid
side-reactions in the further steps. A Suzuki coupling was then per-
formed between the two reagents, with the help of 9-borabicy-
clo[3.3.1]nonane (9-BBN) in THF, in the presence of Pd0 generated
in situ to get the expected ketoester within 96% yield. The final
saponification followed by a neutralization led to the free carboxylic
acid within96% yield. Finally, in our hands, two steps were necessary
to get the expected 1-(4-(40-methylbenzophenone))-10-undecanoic
acid 1b (‘C1-BP-C10-COOH’) within 92% yield (starting from
4-bromo-40-methyl benzophenone), and four steps within 64% over-
all yield in a convergent route starting from the bromobenzoyl chlo-
ride (Scheme 1).
Table 1
Protection of the 2-(4-bromophenyl)ethanol by either DHP (a) or TIPSCl (b) and
recovery of the protected secondary alcohol 7a or 7b.
Protecting group (R) ?
6 (yield) (%) ?
7 (yield) (%)
2.2.3. Proximal FABP, C10-BP-C1-COOH (1c)
Conversely to what was obtained for the distal FABP, the synthe-
sis of theexpected proximal FABP(where m = 10 and n = 1,Fig. 2) was
tricky. Lala’s method remained unsuccessful. To circumvent the use
of oxone already reported, we designed an alternate route to synthe-
size (4-(4-decylbenzoyl)phenyl)ethanoic acid (‘C10-BP-C1-COOH’,
1c). Thus, the precursor 4-n-decylbenzaldehyde 5 was prepared in
a
92
19
O
b
99
70
Si

B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
7467
O
O
H
1. PyBOP, DIEA
O
N
O
OH
O
2.
H
N
O
OH
O
N
OH
(1a)
(9a)
H
O
NH2
O
Boc-L-Dpr-OH
83%
O
O
O
H
N
O
H2N
N
N
H
TFA, CH2Cl2
84%
14
14
1. PyBOP, DIEA
H
O
N
N
2. H2N
(10a)
(11a)
14
H
H
78%
O
O
O
O
N
O
N
N
O
O
H
O
N
CTMR, PyBOP, DIEA
67%
N
H
O
O
O
O
O
O
N
NH
(12b)
HN
O
N
H
14
N
O
N
N
O
H
(12a)
O
O
O
H
O
N
N
H
O
O
NH
O
(12c)
O
Scheme 3. Total synthesis of lipid tool 12a. In square: structures of 12b and 12c in comparison.
Table 2
Structures of benzophenone-containing fatty acids (FABPs) with predicted distances (d, Å) from the hydroxyl group
oxygen atom to the keto carboxyl group atom (inspired by Spencer and coworkers23
)
Name
Formula
d (Å)
O
C1-BP-C10-COOH
Distal FABP (1b)
18.7
OH
O
O
C6-BP-C5-COOH
Medium FABP (1a)
O
O
12.7
7.9
OH
OH
O
C
10-BP-C1-COOH
Proximal FABP (1c)
and coupled to 5 to give the secondary alcohol 7b as a colorless oil
in 70% yield.
reaction allowed to obtain (4-(4-decylbenzoyl)phenyl)ethanoic
acid (‘C10-BP-C1-COOH’, 1c) in better overall yield than what was
reported for a lower number of steps starting from the aldehyde.
This improvement paves the way to the synthesis of new FABP
analogs.
After that, tetra-n-butylammonium fluoride (TBAF) was used to
deprotect 7b and primary alcohol 8 was obtained in 73% yield. Fur-
ther, 8 was doubly oxidized by chromium trioxide in concentrated
sulfuric acid aqueous solution to give 1c in 91% yield, as a white
solid. All compounds (final and intermediate) were characterized
by MS and NMR. Starting form the 4-n-decylbenzaldehyde, the
yield of the synthesis was as high as 46%, in comparison to the
22% previously reported. Finally, the overall yield of the total syn-
thesis was 35%, seven steps were needed to get the final product
from commercially available 4-decylbenzoyl chloride following a
convergent strategy (Scheme 3). Thus, the fine tuning of the
2.3. Distance prediction
In the FABPs described here, the cross-linking site (–CO) is situ-
ated at defined distances (d in Scheme 1) from the oxygen atoms
of the hydroxyl group in the carboxylic acid (ꢀCOOH). These dis-
tances had previously been calculated by Spencer and coworkers
using the Spartan program.23 For our three compounds, these

7468
B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
Table 3
2.4. Assembly of the probes: general procedure
Yield of each step in the synthesis of the lipid tools. m and n are the number of
methylene link in the fatty chain (see Fig. 1)
Three FABPs (1a–c) were thus synthesized and engaged into the
preparation of three novel lipid tools. Each experiment was per-
formed away from light. In order to easily monitor the synthesis
progress by UV, we chose to introduce the FABPs at the first step
of the synthesis. Each FABP was activated in the presence of DIEA
by PyBOP. Then Boc-Dpr-OH was added to the solution. After
work-up, Boc-Dpr(FABP)-OH 9a–c were isolated and activated by
PyBOP in the presence of DIEA before the addition of hexadecyl-
amine. Boc-Dpr(FABP)-NH-C16H33 10a–c were then isolated and
m
n
9 (%)
10 (%)
11 (%)
12 (%)
a (medium BP)
b (distal BP)
c (proximal BP)
6
1
10
5
10
1
83
90
50
78
64
54
84
86
95
67
74
66
distances are now ranging from 7.9 Å (‘C10-BP-C1-COOH’) to 18.7 Å
(‘C1-BP-C10-COOH’), and the middle value for ‘C6-BP-C5-COOH’ can
thus be estimated to 12.7 Å (Table 2).
Figure 3. Fluorescence characteristics of lipid tools 12a, 12b and 12c. Liposomes composed of PC/chol/lipid tool (65:30:5 molar ratio) were prepared in PBS buffer at pH 7.4 or
pH 5.0. Excitation and emission spectra (50
M final lipid composition) were recorded at 25 °C on an Infinite 1000 TecanÒ spectrofluorimeter. (A) Excitation spectra (emission
l
set at 590 nm) of 12a (black), 12b (red) and 12c (blue), at pH 7.4 (solid lines) or 5.0 (dotted lines). (B) Emission spectra (excitation set at 560 nm) of 12a, 12b and 12c (similar
symbols and color codes as in A). (C and D) Emission spectra of 12a at pH 7.4 (C) and 5.0 (D), after no. (0), 20, 45 or 90 s of UV irradiation. (E and F) Emission spectra of R18 at
pH 7.4 (E) and 5.0 (F), after no. (0), 20, 45 or 90 s of UV irradiation.

B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
7469
12a
12b
12c
bR
12a
12b
12c
bR
1
2
3
4
5
6
7
8
1
B
2
3
4
5
6
7
8
250
75
50
A
25
bR (27 kDa)
bR (27 kDa)
15
10
Figure 4. Photolabeling of bR with lipid tools (12a–c). (A) Coomassie coloration compared to all blue standard (left), (B) Fluorimager analysis. 10 lg of bR are used in each
experiment (for each lane). Lane 1, molecular mass markers; protein standards in kDa are indicated on the left side. Lane 2, bR + 12a not irradiated; lane 3, bR + 12a irradiated;
lane 4, bR + 12b not irradiated; lane 5, bR + 12b irradiated; lane 6, bR + 12c not irradiated; lane 7, bR + 12c irradiated; lane 8, bR alone.
deprotected by a mixture of TFA and dichloromethane (1/1, v/v).
After work-up, H-Dpr(FABP)-NH-C16H33 11a–c were isolated. Fi-
nally CTMR was the last building block to be introduced. After its
activation by PyBOP in the presence of DIEA, H-Dpr(FABP)-NH-
(Fig. 3E and F). Excitation (data not shown) and emission maxima
remained unaffected by UV irradiation ( Fig. 3E for pH 7.4 and
Fig. 3F for pH 5.0). Taken together these data clearly demonstrate
that the observed stability of rhodamine fluorescence to pH and
UV irradiation is an intrinsic property of rhodamine, whatever
the chemical lipid skeleton it is associated to.
C16H33 11a–c were added. The reaction progress was followed by
TLC. After work-up, Rhod-Dpr(FABP)-NH-C16H33 12a–c were iso-
lated as purple powders (Scheme 3 for 12a synthesis).
Intermediate yields to each lipid tool were comparable (Table 3).
Every compound (final and intermediate) was extensively charac-
terized by MS, TLC and NMR whenever possible. Note that CTMR
was used as the commercially available mixture of the 5- and 6-
carboxytetramethylrhodamine isomers. The two regioisomers were
sometimes detectable as two spots by TLC, but could not be further
separated. The separation was not required for the biological use of
the lipid tool.
3.2. Hydrophobic photoaffinity labeling of bacterioRhodopsin
To check the capability of lipid tools 12a, 12b and 12c to label
hydrophobic protein, we chose bacterioRhodopsin as a convenient
membrane protein model. Indeed its three-dimensional structure
is known,36,37 revealing an arrangement with seven alpha-helicoidal
transmembrane domains. Moreover it is commercially available,
which renders its use very easy. BacterioRhodopsin resuspended
in water was incubated with micelles containing lipid tools (12a,
12b or 12c). The micelles obtained were submitted to 45 s UV irradi-
ation at 365 nm. The suspension was denaturated and submitted to
SDS–PAGE protein gels (Fig. 4) followed by a Coomassie coloration
(A), compared to all blue standard (left) and Fluorimager analysis
(B). Interestingly when visualizing the gel under Fluorimager, we
first noticed fluorescent labeling for a molecular weight correspond-
ing to bacterioRhodopsin, which indicates that the covalent photola-
beling procedure with our lipid tool was successful (Fig. 4B lanes 3, 5
and 7). Most importantly fluorescent labeling was not observed
when the protein/lipid tool mixture was not irradiated (Fig. 4B lanes
2, 4 and 6). Also note that the protein was not damaged by the irra-
diation (Fig. 4A) and the protein alone did not give any fluorescent
signal (Fig. 4 lane 8). When bR was heated to 95 °C before irradiation,
we noticed the appearance of high-order oligomers that could be
photolabeled (data not shown). This tends to indicate that bR treat-
ment by heat leads to the formation of multimers hydrophobic by
nature, and that photolabeling with our lipid tools is not dependent
upon a specific protein conformation.
3. Biochemistry part
3.1. Fluorescence and absorbance properties of the lipid tools
We characterized the spectral properties of lipid tools 12a, 12b
and 12c by fluorescence spectroscopy in liposomes. This was done
using a chloroformic solution of lipid tool added to chloroformic
solutions of phosphatidylcholine and cholesterol (PC/chol/lipid
tool 65:30:5 molar ratio). Fluorescence spectra were then recorded
at pH 7.4 and pH 5.0. Depending on the lipid tool considered, the
influence of pH on its fluorescence remained low to negligible on
both excitation (Fig. 3A) and emission spectra (Fig. 3B), in agree-
ment with the well-documented stability of rhodol derivatives to-
ward pH variations.32 Excitation and emission maxima were found
at 560 2 nm and 585 2 nm for both pH, respectively. These are
therefore typical spectral properties of a rhodamine derivative.
This indicates that CTMR grafted onto these novel lipid tools be-
haves as a regular rhodamine moiety.
Figure 4B shows that the efficacy of the photo-crosslinking is
dependent on the distance between the benzophenone double ring
and the CO. In this particular case, the longer the distance (d), the
more intense the crosslinking.
Interestingly when comparing spectra obtained from liposomes
containing our lipid tools, submitted or not to a 20 s, 45 s or 90 s—
UV irradiation, we noticed a remarkable stability of the rhodamine
moiety to UV light exposure for both pH tested (Fig. 3C and D for
12a; data not shown for 12b and 12c). Not only the quantum yields
of fluorescence were comparable, but also were the maxima of
both excitation (data not shown) and emission spectra. This agrees
well with the previously reported photostability of rhodamine
present in other fluorophores.33,34
To check whether this stability under our specific experimental
conditions was intrinsic to the rhodamine moiety or correlated to
the chemical formula of lipid tools, we prepared PC/chol liposomes
containing 5 mol % of octadecyl Rhodamine B chloride (R18), a
well-documented lipid compound used in fluorescence spectros-
copy, notably in viral fusion assays.35,17 The behavior of R18 into
PC/chol liposomes was comparable to that observed with 12a,
12b and 12c, with respect to pH and irradiation conditions
4. Conclusion
An alternate expedient chemical route is now available to reach
functional FABPs. We therefore synthesized a novel generation of
lipid tools for hydrophobic photoaffinity labeling, based upon a
fluorescent detection via a rhodamine dye. This approach relies
on the use of lipid tools containing a photoactivatable agent and
a fluorescent probe for the detection of adducts of the reaction.
The corresponding lipid tools offer numerous advantages: their
synthesis is relatively easy, they are stable in membranes due to
their double fatty acid chain structure, photolabeling is performed
at an ultraviolet wavelength which prevents degradation of the

7470
B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
proteins studied. After photoactivation, the benzophenone ring is
expected to induce a covalent cross-linking reaction with its
immediate environment in the membrane. The CTMR headgroup
allows to detect and follow the cross-linking reaction products.
Preliminary biochemical experiments efficiently proved the func-
tionality of our new probes. We thus provide the scientific commu-
nity with new, robust and efficient tools of study and identification
of hydrophobic zones of proteins.38
overnight to give 5.00 g (93%) of 3 as a yellow oil. TLC: Rf = 0.4
(9:1 Cyclohexane/EtOAc); NMR 1H (400 MHz; CDCl3) d 5.83–5.77
(ddt, 1H, J = 17.1;10.1;6.8), 5.01–4.91 (m, 2H), 3.66 (s, 3H), 2.29
(t, 2H, J = 7.6), 2.04 (q, 2H, J = 7.0), 1.61 (quint, 2H, J = 7.3),
1.38–1.29 (m, 10H); NMR 13C (100 MHz; CDCl3) d 174.9, 139.8,
114.8, 52.0, 34.8, 34.4, 29.8, 25.6. MS calcd for C12H23O2 = 199.2,
found 199.2.
5.1.1.3. Methyl 11-(4-(4-methylbenzoyl)phenyl)undecano-
ate (4).
To a stirred solution of 181 mg (0.91 mmol) of 2
5. Experimental procedures
5.1. Chemistry
in 5 mL of THF at 0 °C under argon, 4 mL of a 0.5 M solution
of 9-BBN in THF was added. The resulting mixture was stirred
at room temperature for 4 h. To
a solution of 890 mg
(2.73 mmol) of cesium carbonate in 2 mL of DMF, 3 mL of
THF and 0.6 mL of water, 55 mg (0.18 mmol) of triphenylar-
sine, 131.7 mg (0.18 mmol) of Pd(dppf)Cl2 and 250 mg
(0.91 mmol) of 3 were added. The first mixture mentioned
was added, via a cannula, to the second one under argon.
The resulting solution was stirred at room temperature for
22 h and poured into 15 mL of water and 15 mL of EtOAc.
The organic layer was washed with water, with a saturated
NaHCO3 aqueous solution and with brine. The solution was fil-
trated through Celite and washed with EtOAc. The organic ex-
tract was dried over MgSO4, filtered and concentrated under
vacuum. The crude product was purified on silica gel with
100:0–100:10 cyclohexane/ethyl acetate v/v to give 247 mg
(69%) of 4 as a yellow oil. TLC: Rf = 0.4 (9:1 Cyclohexane/
EtOAc). NMR 1H (300 MHz; CDCl3) d 7.73–7.71 (m, 4H),
7.37–7.29 (d, 4H, J = 7.4), 3.67 (s, 3H), 2.69 (t, 2H, J = 7.7),
2.45 (s, 3H), 2.31 (t, 2H, J = 7.5), 1.67–1.60 (m, 4H), 1.36–
1.24 (m, 12H); NMR 13C (75 MHz; CDCl3) d 196.3, 174.3,
147.9, 142.9, 135.4, 135.3, 130.2, 128.9, 128.3, 51.4, 36.0,
34.1, 31.2, 29.7, 29.5, 29.4, 29.3, 29.2, 29.2, 24.9, 21.6. MS
calcd for C26H35O3 = 395.2, found 395.2.
5.1.1. Generals
The amino-acid Boc-L-Dpr-OH was purchased from Iris Biotech
GmbH. The ninhydrin, PMA and TFA were from Sigma-Aldrich. The
CTMR and PyBOP were purchased from Novabiochem. The
hexadecylamine was from Aldrich and the DIEA from Alfa Aesar.
6-(4-(4-n-hexylbenzoyl) phenyl) hexanoic acid was synthesized
following an already reported protocol.39 Chemical reagents were
used without further purification. CH2Cl2 and THF were distilled
under argon atmosphere and dried over CaH2 and sodium (in the
presence of benzophenone) respectively. Chromatography was car-
ried out on silica gel 60 (40–63 lm): powder and aluminium-sup-
ported sheets for TLC plates were purchased from Merck. All 1H
and 13C NMR spectra were acquired on Brucker 300 MHz or
400 MHz spectrometers as mentioned. 1H chemical shifts (d) are
reported in ppm as referenced to residual undeuterated solvent
peak and coupling constants (J) are reported in Hz. Mass spectra
(ES-MS) were recorded on an ESI/TOF Mariner mass spectrometer.
When a compound is described for the first time, a high resolution
mass result (HR-MS) acquired on a GC/TOF Agilent technology is
provided. Melting points are measured on a Büchi B-540 apparatus.
Analytical HPLC analyses were performed on an analytical Jupiter
5
l
C4 300 Å column (150 ꢁ 4.6 mm) from Phenomenex with a
5.1.1.4. 11-(4-(4-methylbenzoyl)phenyl)undecanoic acid
flow rate of 1 mL minꢀ1 using a 10 min linear gradient from 0%
CH3CN in water (0.1% TFA) to 100% CH3CN (0.1% TFA) followed
by 5 min at 100% CH3CN (0.1% TFA). Retention times (Rt) from ana-
lytical RP-HPLC are reported in minutes.
(1b).
To a solution of 200 mg (0.51 mmol) of 4 in 10 mL
of EtOH 95%, 85 mg (1.52 mmol) of sodium hydroxide was
added. The resulting mixture was stirred overnight at room
temperature. The solution was acidified to pH 1 with HCl 1 N
aqueous solution. The aqueous layer was extracted with
EtOAc. The combined organic extracts were washed with brine
and dried over MgSO4, filtered and concentrated under vac-
uum. The crude product was purified by recristallisation in
petroleum ether to give 180 mg (93%) of 1b as a white solid.
NMR 1H (300 MHz; CDCl3) d 7.74–7.71 (dd, 4H, J = 8.1, 2.7),
7.29–7.27 (d, 4H, J = 7.4), 2.69 (t, 2H, J = 7.7), 2.45 (s, 3H),
2.36 (t, 2H, J = 7.5), 1.65–1.61 (m, 4H), 1.41–1.22 (m, 12H);
NMR 13C (50 MHz; CDCl3) d 196.3, 179.1, 147.9, 142.9, 135.4,
135.2, 130.2, 128.9, 128.3, 36.0, 33.9, 31.2, 29.5, 29.4, 29.4,
29.3, 29.2, 29.0, 24.7, 21.6. MS calcd for C25H33O3 = 381.2,
found 381.2.
5.1.1.1. 4-Bromo-40-methylbenzophenone (2).
To a solution
of 500 mg (2.28 mmol) of 4-bromobenzoyl chloride in 5 mL of tol-
uene was added, in one portion, 456 mg (3.42 mmol) of AlCl3. The
resulting mixture was stirred at reflux for 2 h. The solution was
cooled at room temperature and poured into an ice concentrated
HCl solution. The aqueous layer was extracted with EtOAc. The
combined organic extracts were washed with 1 N HCl aqueous
solution, with water and brine and dried over MgSO4, filtered
and concentrated under vacuum. The crude product was purified
on silica gel with 38:2–37:3 cyclohexane/ethyl acetate v/v to give,
after recristallisation in n-hexane, 470 mg (75%) of 2 as a white so-
lid. TLC: Rf = 0.4 (9:1 Cyclohexane/EtOAc); NMR 1H (300 MHz;
CDCl3) d 7.70–7.60 (m, 6H), 7.30–7.27 (m, 2H), 2.44 (s, 3H). NMR
13C (75 MHz; CDCl3) d 195.9, 144.2, 137.3, 135.0, 132.1, 132.0,
130.9, 129.7, 127.8, 22.3. MS calcd for C14H11BrO = 274.0, found
274.0.
5.1.1.4.1. Methyl 4-decylbenzoate40
.
To a solution of 1.00 g
(3.56 mmol) of 4-decylbenzoyl chloride in 10 mL of dry methanol,
43 mg (0.36 mmol) of dimethylaminopyridine were added. The
resulting mixture was stirred at room temperature for 1 h and then
concentrated to dryness. The crude solution was dissolved with
20 mL of CH2Cl2 and washed with water (2 ꢁ 20 mL). The aqueous
layer was extracted with CH2Cl2 (3 ꢁ 100 mL). The organic layers
were collected and dried over MgSO4, filtered and concentrated
under vacuum. The new product was dried under vacuum over-
night to give 0.97 g (99%) of the expected compound as a colorless
oil. NMR 1H (400 MHz; CDCl3): d 7.96 (d, 2H, J = 8.2 Hz), 7.24 (d, 2H,
J = 8.2 Hz), 3.90 (s, 3H), 2.66 (t, 2H, J = 7.8 Hz), 1.63 (m, 2H), 1.32–
1.27 (m, 14H), 0.89 (t, 3H, J = 6.7 Hz). NMR 13C (100 MHz; CDCl3): d
167.8, 149.1, 130.3, 129.0, 128.3, 52.5, 36.7, 32.6, 31.8, 30.2, 30.2,
5.1.1.2. Methyl 10-undecenoate (3).
To a solution of 5.00 g
(27.1 mmol) of 10-undecenoic acid in 100 mL of methanol, 1 mL
of concentrated H2SO4 was added. The resulting mixture was stir-
red at reflux overnight. The solution was cooled to room tempera-
ture and 200 mL of petroleum ether was added. The combined
organic extracts were washed with a saturated NaHCO3 aqueous
solution, with brine and dried over MgSO4, filtered and concen-
trated under vacuum. The new product was dried under vacuum

B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
7471
30.1, 30.0, 29.9, 23.3, 14.7. MS calcd for C18H29O2 = 277.2, found
277.2.
aqueous layer was extracted with ethyl acetate. The combined or-
ganic extracts were washed with brine and dried over MgSO4, fil-
tered and concentrated under vacuum. The crude product was
purified on silica gel with 100:0–100:7 cyclohexane/ethyl acetate
v/v to give 0.89 g (70%) of 7b as a colorless oil. NMR 1H (400 MHz;
CDCl3): d 7.32–7.13 (m, 8H), 5.81 (s, 1H), 3.89 (dt, 2H, J = 10.6,
7.2 Hz), 2.86 (dt, 2H, J = 10.6, 7.2 Hz), 2.59 (t, 2H, J = 7.8 Hz), 1.59
(m, 2H), 1.30 (m, 14H), 1.05 (m, 21H), 0.89 (t, 3H, J = 6.7 Hz). NMR
13C (100 MHz; CDCl3): d 142.9, 142.5, 141.9, 139.2, 129.9, 129.1,
127.1, 76.7, 65.5, 40.1, 36.3, 32.6, 32.2, 30.3, 30.2, 30.0, 23.4, 18.6,
14.8, 12.7. HRMScalcd for C31H49O2Si = 481.3502 (oneisopropylres-
idue is lost), found 481.3496.
5.1.1.4.2. 4-Decylphenyl methanol41
.
A solution of 40 mg
(1.03 mmol) of lithium aluminum hydride in 15 mL of ethylic ether
was placed in a two-necked flask equipped with reflux condenser
and a dropping funnel. Through the dropping funnel, 0.95 g
(1.03 mmol) of methyl 4-decylbenzoate was introduced at a rate
such as to produce a gentle reflux. The resulting mixture was stir-
red at room temperature for another 1 h. Then 20 mL of water were
added dropwise and cautiously, the solution was poured onto ice
and a solution of 10% of H2SO4 was added. The aqueous layer
was extracted with ethylic ether. The combined organic extracts
were dried over MgSO4, filtered and concentrated under vacuum.
The crude product was purified on silica gel with 95:5–80:20
cyclohexane/ethyl acetate v/v to give 0.77 g (90%) of 4-decylphenyl
methanol as a white solid. NMR 1H (300 MHz; CDCl3): d 7.29 (d, 2H,
J = 7.9 Hz), 7.19 (d, 2H, J = 7.9 Hz), 4.65 (s, 2H), 2.63 (t, 2H,
J = 7.8 Hz), 1.63 (m, 2H), 1.29 (m, 14H), 0.91 (t, 3H, J = 6.8 Hz).
NMR 13C (75 MHz; CDCl3): d 143.1, 138.8, 129.2, 127.8, 65.9,
36.3, 32.6, 32.2, 30.3, 30.2, 30.0, 23.3, 14.7. MS calcd for
5.1.1.8. 2-(4-((4-decylphenyl)(hydroxy)methyl)phenyl)ethanol
(8).
To a solution of 0.84 g (1.60 mmol) of 7b in 10 mL of
THF, 3 mL (3 mmol) of n-tetrabutylammonium fluoride were
added. The resulting mixture was stirred at room temperature
for 1.5 h and was quenched by a saturated NaHCO3 aqueous solu-
tion (20 mL). The aqueous layer was extracted with EtOAc. The
combined organic extracts were washed with brine and dried over
MgSO4, filtered and concentrated under vacuum. The crude prod-
uct was purified on silica gel with 100:0–98:2 cyclohexane/ethyl
acetate v/v to give 0.43 g (73%) of 8 as a white solid. NMR 1H
(300 MHz; CDCl3): d 7.36–7.14 (m, 8H), 5.82 (s, 1H), 3.85 (m,
2H), 2.86 (t, 2H, J = 6.5 Hz), 2.58 (t, 2H, J = 7.8 Hz), 1.59 (m, 2H),
1.26 (m, 14H), 0.89 (t, 3H, J = 6.7 Hz). NMR 13C (75 MHz; CDCl3):
d 143.0, 142.9, 141.8, 138.3, 129.7, 129.2, 127.4, 127.1, 76.6, 64.2,
39.5, 36.3, 32.6, 32.1, 30.3, 30.2, 30.0, 23.4, 14.8. HRMS calcd for
C17H28O = 248.2, found 248.2.
5.1.1.5. 4-Decylbenzaldehyde (5).
To a solution of 390
lL
(4.63 mmol)ofoxalylchlorideatꢀ78 °C, 460
l
L(6.49 mmol)ofdim-
ethylsulfoxyde in 10 mL of CH2Cl2 were cautiously added. The
resulting mixture was stirred for 10 min and a solution of 0.77 g
(3.09 mmol) of 4-decylphenyl methanol solubilized in 10 mL of
CH2Cl2 was added cautiously at ꢀ78 °C. The resulting solution was
stirred for another 20 min and 1.73 mL (12.4 mmol) of triethylamine
was added and the solution was allowed to reach room temperature
over 1.5 h. 20 mL of a saturated NaHCO3 aqueous solution were
added. The aqueous layer was extracted with ethylic ether
(1 ꢁ 50 mL). The combined organic extracts were washed with sat-
urated NaHCO3 aqueous solution, with brine and dried over MgSO4,
filtered and concentrated under vacuum. The new product was dried
overnight to give 0.66 g (87%) of 5 as a colorless oil. NMR 1H
(300 MHz; CDCl3): d 9.98 (s, 1H), 7.80 (d, 2H, J = 8.1 Hz), 7.34 (d,
2H, J = 8.1 Hz), 2.69 (t, 2H, J = 7.8 Hz), 1.65 (m, 2H), 1.27 (m, 14H),
0.89 (t, 3H, J = 6.6 Hz). NMR 13C (75 MHz; CDCl3): d 192.5, 151.1,
135.1, 130.5, 129.7, 36.9, 32.6, 31.7, 30.2, 30.1, 30.0, 23.3, 14.7. HRMS
calcd for C17H26O = 246.1984, found 246.2004.
C25H36O2 = 368.2715, found 368.2738.
5.1.1.9.
2-(4-((4-decylphenyl)(hydroxy)methyl)phenyl)acetic
acid (1c).
To a solution of 0.41 g (1.11 mmol) of 8 in 10 mL
of acetone was added, drop by drop, 1 mL of Jones reagent
(26.72 g of CrO3 in 77 mL of H2O and 23 mL of concentrated
H2SO4). The resulting mixture was stirred at room temperature
for 2 h and quenched with 2-propanol. Water (20 mL) was added
to the solution and most of the solvent was removed under re-
duced pressure. The aqueous layer was extracted with CH2Cl2.
The combined organic extracts were dried over MgSO4, filtered
and concentrated under vacuum. The crude product was purified
on silica gel with 95:5–85:15 cyclohexane/ethyl acetate v/v to give
0.38 g (91%) of 1c as a white solid. TLC: Rf = 0.4 (9:1 CH2Cl2/MeOH);
mp 71.0–72.5 °C; NMR 1H (300 MHz; CDCl3): d 7.77 (dd, 4H, J = 8.1,
12.8 Hz), 7.35 (dd, 4H, J = 8.1, 38.7 Hz), 3.76 (s, 2H), 2.69 (t, 2H,
J = 7.7 Hz), 1.64 (m, 2H), 1.28 (m, 14H), 0.89 (t, 3H, J = 6.4 Hz).
NMR 13C (75 MHz; CDCl3): d 196.8, 177.4, 149.0, 138.3, 137.6,
135.6, 131.0, 130.0, 129.0, 41.6, 36.7, 32.6, 31.8, 30.2, 30.0, 23.3,
14.8. MS calcd for C25H33O3 = 381.2, found 381.2.
5.1.1.6. 4-(Bromophenethoxy)triisopropylsilane (6b)42
.
To
a mixture of 0.50 g (2.48 mmol) of 4-bromophenethyl alcohol in
10 mL of CH2Cl2 was added a solution of 0.72 g (3.73 mmol) of tri-
isopropylsilyl chloride and 0.51 g (7.44 mmol) of imidazole. The
resulting mixture was stirred at room temperature for 2 h. The or-
ganic layer was washed with water (10 mL) and brine (10 mL). The
aqueous layer was extracted with CH2Cl2 (3 ꢁ 100 mL). The com-
bined organic extracts were dried with MgSO4, filtered and evapo-
rated. The crude product was purified on silica gel with 100:0 to
100:10 cyclohexane/ethyl acetate v/v to give 0.88 g (99%) of 6b
as a colorless oil. NMR 1H (300 MHz; CDCl3): d 7.40 (m, 2H), 7.11
(m, 2H), 3.87 (t, 1H, J = 6.9 Hz), 2.81 (t, 1H, J = 6.9 Hz), 1.05 (m,
21H). NMR 13C (75 MHz; CDCl3): d 139.0, 131.9, 131.6, 120.6,
65.1, 39.8, 18.6, 12.7. HRMS calcd for C14H22BrOSi = 313.0623 (with
79Br), found 313.0638.
5.1.2. General procedure to synthesize lipid tools
5.1.2.1. Synthesis of Boc-
L
-Dpr(FABP)-OH (9).
To a solution
of FABP (1 equiv) in anhydrous dichloromethane, DIEA (5 equiv)
and PyBOP (1 equiv) were added. The resulting mixture was stirred
for 10 min (acid activation). Then Boc-L-Dpr-OH (1 equiv) solubi-
lized in a mixture of anhydrous dichloromethane and methanol
(3:2 v/v), were added to the above mentioned activated solution.
The resulting mixture was stirred at room temperature protected
from the light. The organic layer was washed with saturated NaH-
CO3 aqueous solution and the aqueous layer was extracted with
CH2Cl2. The combined organic extracts were dried over MgSO4, fil-
tered and concentrated under vacuum. The crude product was
purified on silica gel with 100:0–100:15 CH2Cl2/MeOH v/v eluent.
5.1.1.7. (4-Decylphenyl)(4-(2-5-triisopropylsilyloxy) ethyl)phe-
nyl) methanol (7b).
To a solution of 0.96 g (2.67 mmol) of 6b
in 10 mL of dry THF at ꢀ78 °C under argon, 1.8 mL (2.92 mmol) of
n-butyllithium (1.6 M in hexane) was added dropwise with a syr-
inge. After 1 h at ꢀ78 °C, a solution of 0.60 g (2.43 mmol) of 5 in
dry THF was added drop by drop. The resulting mixture was allowed
to reach room temperature over 1.5 h and kept stirring over night. It
was quenched by a saturated NaHCO3 aqueous solution (20 mL). The
5.1.2.2. Synthesis of Boc-
solution of Boc- -Dpr(FABP)-OH (1 equiv) in anhydrous dichloro-
methane, DIEA (5 equiv) and PyBOP (1.3 equiv) were added. The
L
-Dpr(FABP)-NH-C16H33 (10).
To a
L

7472
B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
resulting mixture was stirred for 10 min (acid activation). Then
hexadecylamine (1 equiv) was added. The resulting mixture was
stirred at room temperature protected from the light. The organic
layer was washed with saturated NaHCO3 aqueous solution and
the aqueous layer was extracted with CH2Cl2. The combined organ-
ic extracts were dried over MgSO4, filtered and concentrated under
vacuum. The crude product was purified on silica gel with
100:0–100:4 CH2Cl2/MeOH v/v eluent.
6H). RMN 13C (125 MHz; CDCl3) d 196.9, 176.0, 175.9, 170.7, 170.5,
169.0, 167.5, 167.3, 156.0, 155.6, 148.6, 148.0, 136.2, 136.0, 130.8,
128.9, 126.2, 125.3, 118.5, 112.5, 111.8, 97.8, 56.3, 54.3, 51.5, 44.3,
39.8, 37.2, 36.6, 36.4, 32.6, 32.3, 31.8, 31.4, 30.3–29.5, 27.6, 26.1,
23.3, 14.7. HRMS calcd for
C69H92N5O7 = 1102.6997, found
1102.6961; MS calcd for C69H92N5O7 = 1102.7 found 1102.9.
5.1.3.5. (S)-2-(tert-Butoxycarbonylamino)-3-(11-(4-(4-methylb
enzoyl)phenyl)undecanamido) propanoic acid (9b).
Reac-
5.1.2.3. Synthesis of H-
L
-Dpr(FABP)-NH-C16H33 (11).
Boc-
L
-
tion time: 2.5 h; TLC: Rf = 0.3 (9:1 CH2Cl2/MeOH); NMR 1H
(300 MHz; CDCl3) d 7.69 (m, 4H), 7.27 (m, 4H), 4.14 (m, 1H), 3.74
(q, 1H, J = 7.0), 2.65 (m, 2H), 2.42 (s, 3H), 2.17 (m, 2H), 1.62–1.21
(m, 25H). MS calcd for C33H47N2O6 = 567.3, found 567.3.
Dpr(FABP)-NH-C16H33 was dissolved in a mixture of CH2Cl2/TFA 1:1
v/v. The resulting mixture was stirred for 1 h at room temperature
protected from the light. The solution was diluted with CH2Cl2.
The organic layer was washed with saturated NaHCO3 aqueous solu-
tion and the aqueous layer was extracted with CH2Cl2. The com-
bined organic extracts were dried over MgSO4, filtered and
concentrated under vacuum.
5.1.3.6. (S)-tert-Butyl 1-(hexadecylamino)-3-(11-(4-(4-meth-
ylbenzoyl)phenyl)undecanamido)-1-oxopropan-2-ylcarbamate
(10b).
Reaction time: 5 h; TLC: Rf = 0.8 (9:1 CH2Cl2/MeOH);
NMR 1H (300 MHz; CDCl3) d 7.73 (dd, 4H, J = 8.1;2.7), 7,29 (d, 4H,
J = 7.4), 6.82 (s, 1H), 6.27 (s, 1H), 5.99 (s, 1H), 4.16 (m, 1H), 3.74
(m, 1H), 3.51 (m, 1H), 3.23 (m, 2H), 2.69 (t, 2H, J = 7.6), 2.45 (s,
3Ha), 2.19 (t, 2H, J = 7.7), 1.67–1.25 (m, 53H), 0.89 (m, 3H); NMR
13C (75 MHz; CDCl3) d 196.9, 148.5, 143.6, 136.1, 135.9, 130.8,
129.6, 128.9, 81.0, 55.9, 42.7, 40.2, 37.3, 36.7, 32.6, 31.8, 30.4,
30.2, 30.2, 30.1, 30.0, 29.9, 29.0, 27.5, 26.4, 23.4, 22.3, 14.8. MS
calcd for C49H80N3O5 = 790.6, found 790.6.
5.1.2.4. Synthesis of Rhod-Dpr(FABP)-NH-C16H33 (12).
To a
solution of CTMR (1.3 equiv) in anhydrous dichloromethane, DIEA
(5 equiv) and PyBOP (1.3 equiv) were added. The resulting mixture
was stirred for 10 min (acid activation). Then H-L-Dpr(FABP)-NH-
C16H33 (1 equiv) solubilized in anhydrous dichloromethane was
added to the above mentioned activated solution. The solution
was concentrated under vacuum. The crude product was purified
on silica gel with 100:0–100:15 CH2Cl2/MeOH v/v eluent.
5.1.3.7. (S)-N-(2-Amino-3-(hexadecylamino)-3-oxopropyl)-11-
5.1.3. Specific characterizations
5.1.3.1. (S)-2-(tert-Butoxycarbonylamino)-3-(6-(4-(4-hexylben
(4-(4-methylbenzoyl)phenyl) undecanamide (11b).
Reac-
tion time: 2 h; TLC: Rf = 0.4 (9:1 CH2Cl2/MeOH); NMR 1H
(400 MHz; CDCl3) d 7.72 (dd, 4H, J = 8.1, 2.7), 7.53 (t, 1H, J = 5.6),
7.29 (d, 4H, J = 7.4), 6.31 (m, 1H), 3.65 (m, 1H), 3.47 (m, 2H), 3.24
(q, 2H, J = 6.7), 2.68 (t, 2H, J = 7.7), 2.45 (s, 3H), 2.18 (t, 2H,
J = 7.6), 1.63 (m, 4H), 1.51 (m, 2H), 1.28 (m, 40H), 0.89 (t, 3Hs,
J = 6.8); NMR 13C (100 MHz; CDCl3) d 196.9, 175.0, 173.9, 148.5,
143.6, 136.1, 135.9, 130.9, 129.6, 128.9, 55.9, 44.3, 39.9, 37.5,
36.7, 32.6, 31.8, 30.4, 30.3, 30.2, 30.1, 30.0, 29.9, 27.7, 26.4, 23.4,
22.3, 14.8. MS calcd for C44H72N3O3 = 690.5, found 690.5.
zoyl)phenyl)hexanamido) propanoic acid (9a).
Reaction
time: 12 h; TLC: Rf = 0.3 (9:1 CH2Cl2/MeOH); NMR 1H (300 MHz;
CDCl3) d 7.67 (m, 4H), 7.24 (m, 4H), 4.14 (m, 1H), 3.73 (q, 1H,
J = 7.0), 3.14 (q, 1H, J = 7.0), 2.65 (m, 4H), 2.17 (m, 2H), 1.62–1.21
(m, 23H), 0.87 (m, 3H). MS calcd for C33H47N2O6 = 567.3, found
567.4.
5.1.3.2.
ylbenzoyl)phenyl)hexanamido)-1-oxopropan-2-ylcarbamate
(10a). Reaction time: 6 h; TLC: Rf = 0.4 (9:2 CH2Cl2/MeOH);
(S)-tert-Butyl 1-(hexadecylamino)-3-(6-(4-(4-hex-
5.1.3.8. Rhod-Dpr(CO-C10-BP-C1)-NH-C16H33 (12b).
Reaction
NMR 1H (300 MHz; CDCl3) d 7.73 (m, 4H), 7.27 (m, 4H), 6.86 (s,
1H), 6.35 (s, 1H), 5.98 (d, 1H, J = 6.0), 4.16 (m, 1H), 3.71 (m, 1H),
3.51 (m, 1H), 3.23 (m, 2H), 2.69 (t, 4H, J = 7.5), 2.20 (t, 2H,
J = 7.5), 1.70–1.25 (m, 51H), 0.88 (m, 6H); NMR 13C (75 MHz ;
CDCl3) d 196.9, 175.5, 171.1, 148.6, 148.0, 136.3, 136.0, 130.9,
128.9, 81.0, 42.7, 40.2, 37.0, 36.7, 36.4, 32.6, 32.4, 31.8, 31.5,
30.4–29.5, 29.0, 27.5, 26.1, 23.3, 14.7. MS calcd for
time: 24 h; TLC: Rf = [0.8; 0.7] (8.5:1.5 CH2Cl2/MeOH); HPLC:
Rt = 13.6 min. RMN 1H (500 MHz; CDCl3) d 8.79 (m,1H), 8.65
(m,1H), 8.09 (m,1H), 7.75 (m, 2H), 7.72 (d, 4H, J = 8,1), 7.56 (m,
2H), 7.26 (m, 4H), 6.91 (m, 1H), 6.56 (m, 2H), 4.11 (m, 1H), 3.90
(m, 2H), 3.25–3.13 (m, 14H), 2.65 (m, 2H), 2.44 (s, 3H), 2.19 (m,
2H), 1.61 (m, 4H), 1.28 (m, 40H), 0.87 (m, 3H). RMN 13C
(125 MHz; CDCl3) d 197.0, 176.4, 170.7, 167.8, 156.5, 156.0,
148.6, 143.6, 142.9, 136.0, 135.8, 130.9, 129.6, 128.9, 126.2,
125.3, 118.7, 111.6, 97.6, 64.7, 54.4, 44.2, 42.7, 41.3, 40.5, 37.5,
36.7, 32.6, 31.8, 30.4, 30.3, 30.2, 30.1, 30.0, 29.9, 27.7, 26.4, 23.4,
C49H80N3O5 = 790.7, found 790.6.
5.1.3.3. (S)-N-(2-Amino-3-(hexadecylamino)-3-oxopropyl)-6-(4-
(4-hexylbenzoyl)phenyl) hexanamide (11a). Reaction time:
22.3, 14.8. HRMS calcd for
1102.6951.
C69H92N5O7 = 1102.6997 found
1 h; TLC: Rf = 0.4 (9:1 CH2Cl2/MeOH); NMR 1H (300 MHz ; CDCl3)
d 7.73 (m, 4H), 7.50 (m, 1H), 7.27 (m, 4H), 6.29 (m, 1H), 3.65 (m,
1H), 3.47–3.37 (m, 2H), 3.23 (m, 2H), 2.69 (t, 4H, J = 7.5), 2.20 (t,
2H, J = 7.5), 1.66-1.25 (m, 42H), 0.88 (m, 6H); NMR 13C (75 MHz ;
CDCl3) d 196.9, 174.8, 173.9, 148.6, 148.0, 136.2, 136.0, 130.8,
128.9, 55.9, 44.3, 39.8, 37.2, 36.6, 36.4, 32.6, 32.3, 31.8, 31.4,
30.3–29.5, 27.6, 26.1, 23.3, 14.7. MS calcd for C44H72N3O3 = 690.5,
found 690.5.
5.1.3.9.
(S)-2-(tert-Butoxycarbonylamino)-3-(2-(4-(4-decylb
enzoyl)phenyl)acetamido)propanoic acid (9c).
Reaction
time: 12 h; Used for the next step as such.
5.1.3.10. (S)-tert-butyl 3-(2-(4-(4-Decylbenzoyl)phenyl)acetam-
ido)-1-(hexadecylamino)-1-oxo propan-2-ylcarbamate (10c)
.
Reaction time: 12 h; TLC: Rf = 0.5 (9:1 CH2Cl2/MeOH); NMR
5.1.3.4. Rhod-Dpr(CO-C5-BP-C6)-NH-C16H33 (12a).
Reaction
1H (400 MHz ; CDCl3) d 7.77 (dd, 4H, J = 17.5, 8.1), 7.33 (m, 4H),
6.68 (s, 1H), 6.37 (s, 1H), 5.83 (d, 1H, J = 6.5), 4.17 (m, 1H), 3.72
(m, 1H), 3.64 (s, 2H), 3.50 (m, 1H), 3.21 (q, 2H, J = 6.5), 2.69 (t,
2H, J = 7.6), 1.67–1.25 (m, 53H), 0.89 (t, 6H, J = 6.6); NMR 13C
(100 MHz; CDCl3) d 196.5, 172.6, 170.9, 149.0, 139.5, 137.8,
135.6, 131.3, 130.9, 129.8, 129.1, 81.2, 44.2, 40.3, 36.7, 31.9, 30.4,
time: 24 h; TLC: Rf = [0.7 ; 0.5] (9:2 CH2Cl2/MeOH); HPLC:
Rt = 13,6 min. RMN 1H (500 MHz; CDCl3) d 11.05 (s, 1H), 8.90 (m,
1H), 8.15 (m, 2H), 7.69 (m, 4H), 7.23 (m, 4H), 6.86 (m, 2H), 6.57
(m, 4H), 3.75 (m, 1H), 4.72–4.62 (m, 2H), 3.18 (m, 2H), 3.10
(m,12H), 2.68 (m, 4H), 2.17 (m, 2H), 1.64–1.22 (m, 42H), 0.88 (m,

B. Hilbold et al. / Bioorg. Med. Chem. 19 (2011) 7464–7473
7473
30.3, 30.2, 30.1, 30.0, 29.0, 27.5, 23.4, 14.8. MS calcd for
44H72N3O3 = 690.5, found 690.4.
beled mixture, followed by a 15 min-incubation at room tempera-
ture and analysis by SDS–PAGE on 15% acrylamide/20% glycerol
gels. After migration, the gel was first analyzed for rhodamine fluo-
rescence with a Fluorimager Typhoon 8600 equipped with a rhoda-
mine filter set, and second stained with Coomassie blue to
visualize all protein bands.
C
5.1.3.11. (S)-2-Amino-3-(2-(4-(4-decylbenzoyl)phenyl)acetam-
ido)-N-hexadecylpropanamide (11c). Reaction time: 5 h;
TLC: Rf = 0.4 (9:1 CH2Cl2/MeOH); NMR 1H (400 MHz; CDCl3) d 7.75
(dd, 4H, J = 17.5, 8.1), 7.44 (s, 1H), 7.34 (m, 4H), 6.52 (s, 1H), 3.63 (s,
2H), 3.45 (m, 2H), 3.21 (m, 2H), 2.69 (t, 2H, J = 7.6), 1.65 (m, 2H),
1.48 (m, 2H), 1.25 (m, 42H), 0.88 (t, 6H, J = 6.6); NMR 13C (100 MHz;
CDCl3) d 196.5, 172.6, 149.0, 139.5, 137.8, 135.6, 131.3, 130.9, 129.8,
129.1, 55.7, 44.2, 40.3, 36.7, 31.9, 30.4, 30.3, 30.2, 30.1, 30.0, 27.5,
23.4, 14.8. MS calcd for C44H72N3O3 = 690.5, found 690.5.
Acknowledgments
This work was supported by a MRE fellowship allocated to B.H.
and L.B.-B. by the President of the University de Strasbourg (UDS).
We gratefully acknowledge the financial support from the Univer-
sity of Strasbourg, University of Lyon 1, the Centre National de la
Recherche Scientifique (CNRS) and the Région Rhône Alpes (Cluster
Infectiologie).
5.1.3.12. Rhod-Dpr(CO-C1-BP-C10)-NH-C16H33 (12c).
Reac-
tion time: 24 h; TLC: Rfs = 0.5 and 0.6 (8.5:1.5 CH2Cl2/MeOH);
HPLC: Rts = 14.1 and 14.3 min. RMN 1H (500 MHz; CDCl3) d 11.01
(s, 1H), 8.78 (m, 1H), 8.12 (m, 1H), 7.78 (m, 2H), 7.61 (m, 4H),
7.27 (m, 4H), 7.10 (m, 2H), 6.66 (m, 1H), 6.52 (m, 2H), 3.63 (s,
2H), 4.75 (m, 2H), 3.18–3.07 (m, 14H), 2.69 (m, 2H), 1.60–1.18
(m, 44H), 0.87 (t, 6H, J = 6,5 Hz). RMN 13C (125 MHz; CDCl3) d
196.8, 170.6, 157.1, 156.6, 149.3, 148.8, 142.7, 137.8, 135.6,
131.3, 130.9, 129.8, 129.1, 126.4, 125.4, 118.7, 111.6, 97.2, 54.3,
44.2, 42.6, 40.3, 36.7, 31.9, 30.4, 30.3, 30.2, 30.1, 30.0, 27.5, 23.4,
14.8. HRMS calcd for C69H92N5O7 = 1102.6997 found 1102.6946;
MS calcd for C69H92N5O7 = 1102.7, found 1102.6.
References and notes
1. Kotzyba-Hibert, F.; Kapfer, I.; Goeldner, M. Angew. Chem., Int. Ed. Engl. 1995, 34,
1296.
2. Brunner, J. Annu. Rev. Biochem. 1993, 62, 483.
3. Dorman, G.; Prestwich, G. D. Biochemistry 1994, 33, 5661.
4. Dorman, G.; Prestwich, G. D. TIB Tech 2000, 18, 64.
5. Kage, R.; Leeman, S. E.; Krause, J. E.; Costello, C. E.; Boyd, N. D. J. Biol. Chem.
1996, 271, 25797.
6. Sachon, E.; Bolbach, G.; Chassaing, G.; Lavielle, S.; Sagan, S. J. Biol. Chem. 2002,
277, 50409.
7. Ayoub, M.; Chassaing, G.; Loffet, A.; Lavielle, S. Tetrahedron Lett. 1995, 23, 4069.
8. Karoyan, P.; Sagan, S.; Clodic, G.; Lavielle, S.; Chassaing, G. Bioorg. Med. Chem.
Lett. 1998, 8, 1369.
5.2. Biochemistry
9. Ponthieux, S.; Cabot, J.; Mouillac, B.; Seyer, R.; Barberis, C.; Carnazzi, E. J. Med.
Chem. 2005, 48, 3379.
5.2.1. Generals
10. Li, L.; Tang, W.; Zhao, Z. K. Bioorg. Med. Chem. Lett. 2009, 19, 4824.
11. Toh, C. R.; Fraterman, T. A.; Walker, D. A.; Bailey, R. C. Langmuir 2009, 25, 8894.
12. Yan, P.; Wang, T.; Newton, G. J.; Knyushko, T. V.; Xiong, Y.; Bigelow, D. J.;
Squier, T. C.; Mayer, M. U. Chem. Bio. Chem. 2009, 10, 1507.
13. Epand, R. M. Biopolymers 1997, 43, 15.
14. Jolimaitre, P.; Roux, A.; Blanpain, A.; Leduc, C.; Bassereau, P.; Bourel-Bonnet, L.
Chem. Phys. Lipids 2005, 133, 215.
15. Bourel-Bonnet, L.; Pécheur, E. I.; Grandjean, C.; Blanpain, A.; Baust, T.; Melnyk,
O.; Hoflack, B.; Gras-Masse, H. Bioconjugate Chem. 2005, 16, 450.
16. Jolimaitre, P.; Poirier, C.; Richard, A.; Blanpain, A.; Delord, B.; Roux, D.; Bourel-
Bonnet, L. Eur. J. Med. Chem. 2007, 42, 114.
17. Breslow, R.; Balwin, S.; Flechtner, T.; Kalicky, P.; Liu, S.; Washburn, W. J. Am.
Chem. Soc. 1973, 95, 3251.
18. Czarniecki, M. F.; Breslow, R. J. Am. Chem. Soc. 1979, 101, 3675.
19. Markovic, D. Z.; Durand, T.; Patterson, L. K. Photochem. Photobiol. 1990, 51, 389.
20. Markovic, D. Z.; Patterson, L. K. Photochem. Photobiol. 1993, 58, 329.
21. Lala, A. K. Chem. Phys. Lipids 2002, 116, 177.
22. John, B.; Kumar, E. R.; Lala, A. K. Biophys. Chem. 2000, 87, 37.
23. Gan, Y.; Wang, P.; Spencer, T. A. J. Org. Chem. 2006, 71, 9487.
24. Gan, Y.; Blank, D. H.; Ney, J. E.; Spencer, T. A. J. Org. Chem. 2006, 71, 5864.
25. Wang, P.; Blank, D. H.; Spencer, T. A. J. Org. Chem. 2004, 69, 2693.
26. Blumenthal, R.; Gallo, S. A.; Viard, M.; Raviv, Y.; Puri, A. Chem. Phys. Lipids 2002,
116, 39.
Dodecylphosphocholine (DPC) and lyophilized bacteriorhodop-
sin from Halobacterium salinarum were purchased from Sigma. Egg
yolk phosphatidylcholine (PC) and cholesterol (chol, 99% pure)
were purchased from Avanti Polar Lipids (Alabaster, USA). Octa-
decyl Rhodamine B chloride (R18) was purchased from Fluoprobes.
5.2.2. Liposome preparation
Phospholipid (PC), cholesterol and lipid tool (12a–c) (65:30:5
molar ratio) dissolved in chloroform were mixed in a test tube.
After solvent evaporation by heat (45 min at 37 °C), the dried lipid
film was hydrated by addition of 300 lL of PBS pH 7.4 to obtain a
final concentration of 5 mM of lipids. The mixture was vortexed
and the resulting suspension was heated (37 °C) and cooled (liquid
nitrogen) five times and passed through an extruder equipped with
a 100 nm polycarbonate filter (30–50 times).
5.2.3. Fluorescence spectroscopy
27. Lala, A. K.; Kumar, E. R. J. Am. Chem. Soc. 1993, 115, 3982.
28. Nakatani, K.; Dohno, C.; Nakamura, T.; Saito, I. Tetrahedron Lett. 1998, 39, 2779.
29. Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651.
30. Bernardy, K. F.; Floyd, M. B.; Poletto, J. F.; Weiss, M. J. J. Org. Chem. 1979, 44, 1438.
31. Harding, K. E.; May, L. M.; Dick, K. F. J. Org. Chem. 1975, 40, 1664.
32. Whitaker, J. E.; Haugland, R. P.; Ryan, D.; Hewitt, P. C.; Prendergast, F. G. Anal.
Biochem. 1992, 207, 267.
33. Longmire, M. R.; Ogawa, M.; Hama, Y.; Kosaka, N.; Regino, C. A.; Choyke, P. L.;
Kobayashi, H. Bioconjugate Chem. 2008, 19, 1735.
34. Benchaib, M.; Delorme, R.; Pluvinage, M.; Bryon, P. A.; Souchier, C. Histochem.
Cell Biol. 1996, 106, 253.
35. Hoekstra, D.; de Boer, T.; Klappe, K.; Wilschut, J. Biochemistry 1984, 23, 5675.
36. Henderson, R. J. Mol. Biol. 1975, 93, 123.
37. Pebay-Peyroula, E.; Rummel, G.; Rosenbusch, J. P.; Landau, E. M. Science 1997,
277, 1676.
38. Pécheur, E. I., Hilbold, B., Bourel-Bonnet, L. French Patent Application 1152498,
2011.
39. John, B.; Kumar, E. R.; Lala, A. K. Biophys. Chem. 2000, 87, 37.
40. Matsushita, Y. I.; Sakamoto, K.; Murakami, T.; Matsui, T. Synth. Commun. 1994,
24, 3307.
Excitation and emission spectra of lipid tools in liposomes were
recorded at 32 °C on a TECAN Infinite M1000 spectrofluorimeter.
The liposome composition, unless otherwise indicated, was PC/
chol/lipid tools (65:30:5 molar ratio). Liposomes were 100 nm uni-
lamellar vesicles and were prepared as described earlier.43 Blank
liposomes were composed of PC/chol (70:30), and their back-
ground fluorescence was substracted from the recorded spectra.
5.2.4. BacterioRhodopsin photolabeling
To a solution of lipid tool in chloroform/methanol 2/1 v/v, DPC
(molar ratio lipid tool/detergent 1:50) was added. After evapora-
tion under high vacuum (2 h), the dried lipid film was hydrated
by addition of 50 lL of water and 50 lL of Tris–HCl pH 8.0 buffer
solution. BacterioRhodopsin suspended in H2O was incubated with
lipid tools in DPC micelles at 1:5 molar protein-to-lipid ratios. The
mixture was then left 1 h in the dark at 4 °C, and submitted to 45 s
irradiation by UV light with a Flood Dymax lamp-house equipped
with a 400 W short-arc high pressure mercury lamp (Equipements
Scientifiques SA, Garches), at a 10 cm-distance from the UV source
in a quartz cell. Laëmmli buffer was then added to the photola-
41. Brechbuehler, H.; Buechi, H.; Hatz, E.; Schreiber, J.; Eschenmoser, A. Helv. Chim.
Acta 1965, 48, 1746.
42. Bastian, J. A. et al. U.S. Patent 6025382, 2000.
43. Lavillette, D.; Bartosch, B.; Nourrisson, D.; Verney, G.; Cosset, F. L.; Penin, F.;
Pécheur, E. I. J. Biol. Chem. 2006, 281, 3909.