Synthesis of symmetrical, single-chain, phenylene/biphenylenemodified bolaamphiphiles

Article abstract of DOI:10.1007/s00706-012-0833-2

Two new, symmetrical, phenylene- or biphenylene- modified bolaamphiphiles bearing two phosphocholine headgroups and an alkyl spacer chain length of 32 and 36 carbon atoms, respectively, have been synthesised. The key step was the Cu(II)-catalysed Grignard reaction used either as a simultaneous bis-coupling procedure or in a stepwise homocoupling. Particularly with the use of the homo-coupling, we were able to separate the phenylene-free by-products from the desired products. This homo-coupling additionally offered the possibility of preparing unsymmetrical bolaamphiphiles. Conversion of the diols into bipolar phospholipids was achieved by bis-phosphorylation with β- bromoethylphosphoric acid dichloride and subsequent quarternisation with trimethylamine. Unlike previous studies with aliphatic bolaamphiphiles that formed flexible nanofibres in aqueous suspension, the bolaamphiphiles of the present study, containing phenylene- and biphenylene groups in the middle part of the alkyl spacer chain, formed small ellipsoidal aggregates at ambient temperature. Springer-Verlag 2012.

Full text of DOI:10.1007/s00706-012-0833-2

Monatsh Chem
DOI 10.1007/s00706-012-0833-2
ORIGINAL PAPER
Synthesis of symmetrical, single-chain, phenylene/biphenylene-
modified bolaamphiphiles
•
Simon Drescher Stefan Sonnenberger
•
•
Annette Meister Alfred Blume Bodo Dobner
•
Received: 29 March 2012 / Accepted: 7 August 2012
Ó Springer-Verlag 2012
Abstract Two new, symmetrical, phenylene- or biphenyl-
ene-modified bolaamphiphiles bearing two phosphocholine
headgroups and an alkyl spacer chain length of 32 and 36
carbon atoms, respectively, have been synthesised. The key
step was the Cu(II)-catalysed Grignard reaction usedeitheras a
simultaneous bis-coupling procedure or in a stepwise homo-
coupling. Particularly with the use of the homo-coupling, we
were able to separate the phenylene-free by-products from the
desired products. This homo-coupling additionally offered the
possibility of preparing unsymmetrical bolaamphiphiles.
Conversion of the diols into bipolar phospholipids was
achieved by bis-phosphorylation with b-bromoethylphos-
phoric acid dichloride and subsequent quarternisation with
trimethylamine. Unlike previous studies with aliphatic bola-
amphiphiles that formed flexible nanofibres in aqueous
suspension, the bolaamphiphiles of the present study, con-
taining phenylene- and biphenylene groups in the middle part
of the alkyl spacer chain, formed small ellipsoidal aggregates
at ambient temperature.
Introduction
Bipolar amphiphiles (bolaamphiphiles, BAs) [1] are natu-
rally found in the cell membranes of certain species of
archaebacteria, e.g. methanogenic and thermoacidophilic
Archaea. These membrane lipids are the consequence of
adaptations to the extreme living conditions of the Archaea,
such as high temperatures or low pH values. Owing to their
lipid composition, membranes of certain Archaea effi-
ciently protect the cytoplasm while maintaining the fluid
state that is important for transport and signal transduction.
This stabilisation is achieved by virtue of the substantially
different chemical structure of the membrane constituents
compared to common phospholipids: they are composed of
two hydrophilic headgroups, arranged on both sides of the
cell membrane [2, 3], connected to one or two lipophilic
membrane-spanning alkyl chains. Fluidisation of Archaea’s
cell membranes is tuned by the addition of several methyl
branches or various numbers of cyclopentane rings within
the alkyl chain of the BAs. The resulting chemical and
thermal stability as well as the tendency to form closed
vesicles make this class of amphiphiles applicable in bio-
technology, material science and pharmacy [4–9]; e.g.
inherently stable archaeosomes are an alternative to con-
ventional liposomes for use as drug delivery systems [10].
In addition, it has been shown that natural and synthetic
BAs have a stabilizing effect on phospholipid vesicles [11].
Because of the laborious isolation procedures of archae-
bacterial lipids, generally resulting in mixtures of lipids
with different alkyl chain lengths, and the time-consuming
synthesis of natural BAs, less complex and well-defined
model substances are required for systematic studies on the
interactions of these BAs with conventional phospholipids.
Over the last decade, considerable efforts have been
devoted to the synthesis of novel bipolar archaebacterial
Keywords Grignard reactions ꢀ Lipids ꢀ
Bolaamphiphiles ꢀ Nanostructures ꢀ Phosphates
Electronic supplementary material The online version of this
article (doi:10.1007/s00706-012-0833-2) contains supplementary
material, which is available to authorized users.
S. Drescher (&) ꢀ S. Sonnenberger ꢀ B. Dobner
Institute of Pharmacy, Martin-Luther-University
Halle-Wittenberg, Wolfgang-Langenbeck-Str. 4,
06120 Halle (Saale), Germany
e-mail: simon.drescher@pharmazie.uni-halle.de
A. Meister ꢀ A. Blume
Institute of Chemistry, Martin-Luther-University
Halle-Wittenberg, von-Danckelmann-Platz 4,
06120 Halle (Saale), Germany
123

S. Drescher et al.
model compounds [11–13]. Recently, we reported the
synthesis and temperature-dependent aggregation behav-
iour of symmetrical, single-chain, polymethylene-1,x-
bis(phosphocholine)s (PC-Cn-PC)—a very simple archae-
bacterial model lipid with hydrocarbon chain lengths (n) of
22–32 carbon atoms and two phosphocholine (PC) head-
groups attached at both ends [14]. It has been shown that
this class of single-chain BAs self-assembles into well-
defined long nanofibres and micelles, depending on tem-
perature and concentration [14–16]. The thickness of the
fibres corresponds roughly to the molecular length of the
BAs. Within the fibres, the bola molecules are aligned side
by side, but the bulky headgroups induce a twisted
arrangement, causing a helical superstructure of the fibre,
which could be imaged recently by atomic force micros-
copy (AFM) experiments [17].
O
P
N
O
O
O
O
O
n
O
P
N
n
m
O
Fig. 1 Chemical structure of phenylene/biphenylene-modified bola-
amphiphiles presented in this work (1, m = 1, n = 14; 2, m = 2,
n = 10)
We present herein two synthetic approaches to sym-
metrical, single-chain, phenylene/biphenylene-modified
BAs (1, 2), using a copper(II)-catalysed Grignard reaction
as the key step as well as the first investigations into the
aggregation behaviour of these compounds by transmission
electron microscopy (TEM).
Following the concept of archaebacterial lipids, we firstly
tried to incorporate the single-chain BA PC-C32-PC into
vesicles of common phospholipids, such as 1,2-dipalmitoyl-
sn-glycero-3-phosphocholine (DPPC), 1,2-dimyristoyl-sn-
glycero-3-phosphocholine (DMPC) or 1-palmitoyl-2-oleolyl-
sn-glycero-3-phosphocholine (POPC). However, only a
small amount of the bolalipids could be incorporated into
bilayers [18]. The reasons for this behaviour are packing
problems caused by the larger space requirement of the PC
headgroup in comparison to the small cross-sectional area of
the long alkyl chain [18]. Consequently, a membrane-span-
ning conformation of the PC-C32-PC in a lipid bilayer is
energetically unfavourable, because of the resulting void
volume which cannot be filled by the alkyl chains of the
bilayer-forming lipid. To overcome these packing problems
two structural changes of the BA are conceivable: (a) one
can minimise the space requirement of the headgroup by
a stepwise replacement of the methyl groups leading to
phosphodimethylethanolamines (Me2PE-Cn-Me2PE), which
form nanofibres [19, 20] or square lamellae [21], and
the phosphomonomethylethanolamines (MePE-C32-MePE)
which self-assemble into sheet-like structures [22]; (b) one
can expand the cross-sectional area of the alkyl chains in
order to fill the void volume. The concept of our work pre-
sented herein adapts the latter idea, namely the insertion of a
phenylene and biphenylene ring system [23, 24], respectively,
into the middle part of the long alkyl chain while keeping the
headgroup unchanged and comparable to the headgroups of
common phospholipids, such as 1,2-distearoyl-sn-glycero-3-
phosphocholine(DSPC), DPPCorPOPC(exceptthepresence
of the glycerol moiety). Figure 1 depicts the target structures
of the BAs synthesised in this work, which include an alkyl
chain with an overall length of 36 or 32 carbon atoms and two
phosphocholine headgroups. These spacer lengths were cho-
sen in order to approximately span bilayer membranes
comprising DSPC (two times C18) and DPPC (two times
C16), respectively.
Results and discussion
The preparation of long-chain BAs depends on an effective
synthetic approach for the corresponding diols which rep-
resent the hydrophobic part of the bola molecules.
Unmodified 1,x-diols with 22 or more carbon atoms have
already been synthesised by well-known multistep proce-
dures including (1) bis-acylation of cyclic enamines with
dicarboxylic acid dichlorides, (2) hydrolysis of enamino
ketones and subsequent ring opening, (3) Wolff–Kishner
reduction of bis(oxoacid)s and subsequent reduction with
LiAlH4, or (4) double Wittig reaction with bis(phospho-
rylide)s and x-functionalised aldehydes [25, 26]. Our
synthetic approaches use the Grignard reaction under
catalysis by Li2CuCl4 [27], which can be designed as a
homo-coupling [28, 29] or bis-coupling reaction [14–16].
The latter ones use the Grignard reagent and 1,x-dibro-
mides in a molar ratio of 2:1 (Fig. 2a, b).
Grignard bis-coupling
For the preparation of the phenylene-modified diols we adapted
the Grignard bis-coupling [14] replacing the 1,x-dibromoalk-
anes by phenylene- or biphenylene dihalides. The synthesis
started from tetrahydropyranyl (thp)-protected x-bromoalco-
hols 3, which were transformed into the corresponding
Grignard reagent by treatment with magnesium in dry THF at
50 °C. After removing the excess of magnesium the Grignard
solution reacts with 1,4-bis(chloromethyl)benzene (4a), 1,4-
bis(bromomethyl)-benzene (4b) and 4,40-bis(chloromethyl)-
1,10-biphenyl (4c), respectively, under Li2CuC4 catalysis
yielding either 2,20-[1,4-phenylenebis(hexadecane-16,1-diyl-
oxy)]bis(tetrahydro-2H-pyran) (5a) or 2,20-[biphenyl-4,40-
diylbis(dodecane-12,1-diyloxy)]bis(tetrahydro-2H-pyran)
(5b; Scheme 1).
123

Synthesis of symmetrical, single-chain, phenylene/biphenylene-modified bolaamphiphiles
Fig. 2 Schematic overview of Grignard coupling reactions: a Grig-
nard bis-coupling [14] and b Grignard homo-coupling [28, 29] of
long-chain alkyl bromides; c Grignard bis-coupling and d stepwise
Grignard homo-coupling for the synthesis of phenylene/biphenylene-
modified bolaamphiphiles presented in this work
Scheme 1
i, ii
X
O
O
+
n
Br
X
O
O
O
O
n
m
m
n
3a, n = 14
3b, n = 10
4a, m = 1, X = Cl
4b, m = 1, X = Br
4c, m = 2, X = Cl
4d, m = 2, X = Br
5a, m = 1, n = 14
5b, m = 2, n = 10
iii
28-40% (overall)
HO
n
OH
n
m
i) 3a or 3b, Mg, THF, 50 °C; ii) THF, Li2CuCl4, 0 °C, 4a-4d;
[6a*, m = 1, n = 14]
[6b, m = 2, n = 10
iii) MeOH, PPTS, reflux
One general drawback of these bis-coupling reactions is
the formation of ‘short-chain’ by-products (Fig. 2a)
resulting from the reaction of the Grignard reagent of
compound 3a, 3b with unreacted bromide 3a, 3b [14]. In
the present case, we obtained the bis(tetrahydro-2H-pyr-
an)s of triacontane-1,30-diol and docosane-1,22-diol,
respectively, depending on the bromides used. The sepa-
ration of these by-products from the target compounds 5
failed because of their very similar chromatographic
properties and, hence, we used the crude bis(tetrahydro-
2H-pyran)s 5 without further purification. Unexpectedly,
we did not detect the formation of ‘long-chain’ by-products
in any case. These ‘long-chain’ by-products (Fig. 2a) result
from dimerization of the 1:1 product of the Grignard reagent
and the dihalide [14] by transmetallation, as described in the
workbyTamuraandKochi [27]. Onecanhypothesisethat the
transfer of magnesium from the alkyl Grignard reagent to
the aryl dihalide occurred as well, but the subsequent reaction
of this aryl Grignard reagent with other bromides was sup-
pressed because the aryl Grignard reagent is less reactive
compared to the alkyl counterpart. Moreover, this transmet-
allation might also explain the large amount of ‘short-chain’
by-product as the transmetallation once again generated thp-
protected x-bromoalcohols (3), which produced further
123

S. Drescher et al.
6b could be purified in several steps of recrystallization and
chromatography.
Table 1 Experimental conditions for Grignard bis-coupling reactions
Entry Bromide
Dihalide
Li2CuCl4 Temp./°C
Comparing the yields of the Grignard bis-coupling
reactions one might hypothesise that dibromides react
better than dichlorides. Therefore, we decided to use the
4,40-bis(bromomethyl)-1,10-biphenyl (4d) instead of its
chlorine counterpart 4c to enhance the yield of 6b. The
synthesis of 4d, starting from biphenyl-4,40-dicarboxylic
acid, and the analytical data of the intermediates are
described in detail in the Electronic Supplementary Infor-
mation. Finally, the dibromide 4d was used for the
Grignard bis-coupling reaction with 3b as pointed out in
Scheme 1. The resulting yield of the diol 6b is slightly
higher while using 4d instead of 4c (40 vs. 28 %).
a
1
2
3
4
5
3a (3.13 g,
8 mmol)
4a (0.61 g,
3.5 mmol)
?
?
?
-
?
-45 to
-35
a
3a (3.13 g,
8 mmol)
4b (0.61 g,
3.5 mmol)
-10 to
-5
b
3a (3.13 g,
8 mmol)
4b (0.61 g,
3.5 mmol)
-45 to
-35
3a (12.9 g,
33 mmol)
4b (3.96 g,
15 mmol)
-45 to
-35
a
3b (6.0 g,
18 mmol)
4c (2.0 g,
8 mmol)
-10 to
-5
a
Addition of the catalyst after the addition of the dihalide
b
Incubation of the Grignard reagent with the catalyst prior to the
addition of the dihalide
Stepwise Grignard homo-coupling
‘short-chain’ by-products as described above (Fig. 2c).
Several attempts were made to reduce or, at best, to avoid this
side reaction. Table 1 summarises the different experimental
conditions of the Grignard bis-coupling reactions.
To circumvent the unfeasible purification process of the
phenylene-modified diol 6a* a stepwise Grignard coupling
using an unsymmetrical xylene derivative as starting
material was applied (Fig. 2d). In a first attempt, we used
benzene-1,4-dicarboxylic acid, which was esterified with
MeOH and reduced to the corresponding 1,4-phenylene-
dimethanol. The following protection of only one hydroxy
moiety using 3,4-dihydro-2H-pyran (DHP) and PPTS
failed because of purification problems. Secondly, we
attempted to reduce the commercially available 4-(bro-
momethyl)benzoic acid methyl ester by treating with
LiAlH4 in Et2O at -10 °C. However, this reaction failed
owing to the elimination of the bromine atom during the
reduction, in line observations by Reuter et al. [30]. The
third attempt started from [4-(chloromethyl)phenyl]meth-
anol (7), which already contains the halogen atom required
for the intended Grignard coupling and a hydroxy moiety.
Firstly, the alcohol group was protected as its thp ether
using DHP and PPTS. The resulting 2-[[4-(chloro-
methyl)benzyl]oxy]tetrahydro-2H-pyran (8) was obtained
in 84 % yield after column chromatography. The simple
and high-yielding substitution into bromides, necessary for
the further Grignard coupling, is an advantage of this thp
residue.
All attempts explored, e.g. reaction at lower tempera-
tures, preliminary incubation of the Grignard reagent with
the catalyst, reaction without copper(II) catalyst, or a
changed sequence of addition of the reactants, had no
effect on the final product composition. We detected the
formation of the ‘short-chain’ by-product in all cases and,
hence, we had to use the crude bis(tetrahydro-2H-pyran)s 5
for subsequent reactions. In addition, the overall Grignard
yields (including compound 5 and the ‘short-chain’ by-
product) are slightly higher using the bromide 4b instead of
the chlorides 4a, 4c.
In the next step the thp-protecting groups of the bis(tet-
rahydro-2H-pyran)s 5 were cleaved in dry MeOH with
catalytic amounts of pyridinium p-toluenesulfonate (PPTS)
to yield the diols 6. But, the 16,160-(1,4-phenylene)bis(hexa-
decan-1-ol) (6a*) still contained small amounts of the
by-product triacontane-1,30-diol (MW = 454.81 g/mol, see
Fig. S1, ElectronicSupplementaryMaterial), whichcouldnot
be separated by recrystallization or column chromatography
(the asterisk indicates contamination with by-product). NMR
measurements of the mixture of 6a* and its by-product (entry
3 in Table 1) and the comparison of the peak areas of the
signalsofthemethylenegroupsnexttothe hydroxygroupand
to the phenyl ring, respectively, allow for a rough estimation
of the composition: 64 % of compound 6a* is accompanied
by 36 % of the by-product (Fig. S2, Electronic Supplemen-
tary Material). Since the yield of the Grignard reagent of
compound 3a was determined to be 93 %, the yield of the
‘short-chain’ by-product should not exceed 14 % and, hence,
the high percentage of the by-product is caused by trans-
metallation duringthecoupling reaction. This phenomenon is
a general drawback of Grignard bis-coupling reactions.
Nevertheless, the corresponding biphenylene-modified diol
In the next step, the first elongation via Grignard coupling
was performed. Therefore, the benzyl 15-bromopentadecyl
ether was transformed into the corresponding Grignard
reagent by treatment with magnesium in dry THF. The sub-
sequent copper(II)-catalysed homo-coupling with compound
8 resulted in the formation of 2-[[4-[16-(benzyloxy)-
hexadecyl]benzyl]oxy]tetrahydro-2H-pyran (9; Scheme 2).
The marginal yields of 25 % after column chromatography
are caused by the lower reactivity of the chlorine atom in
Grignard reactions. Attempts to convert thechloro compound
8 (or its precursor compound 7) into the analogous bromo
derivative by heating with LiBr in dry acetone failed owing to
123

Synthesis of symmetrical, single-chain, phenylene/biphenylene-modified bolaamphiphiles
Scheme 2
i
Cl
Cl
OH
O
O
84%
7
8
ii, iii
25%
O
iv
90%
O
14
14
Br
O
O
10
v, vi
9
O
14
O
R
14
11
, R = thp
40 %
vii
(with respect to 10)
HO
viii
14
12, R = H
OH
95%
14
6a
i) DHP, CH2Cl2, PPTS, rt; ii) benzyl 15-bromopentadecyl ether, Mg, THF, 50 °C; iii) THF, LiCuCl4, 0 °C, 8;
iv) PPh3, Br2, CH2Cl2, rt; v) 3a, Mg, THF, 50 °C; vi) THF, Li2CuCl4, 0 °C, 10; vii) PPTS, MeOH, reflux;
viii) H2, Pd/C (10%), heptane-EtOH-ethyl acetate
the instability of the thp residue and the hydroxy moiety
EtOH, and ethyl acetate as eluent to yield 16,160-(1,4-phe-
nylene)bis(hexadecan-1-ol) (6a) without any by-products.
Although the procedure with two subsequently (step-
wise) performed Grignard homo-coupling reactions
required more synthetic steps, this pathway resulted in pure
phenylene-modified 1,x-diols without ‘phenylene-free’ by-
products and, moreover, it can be used for the preparation
of unsymmetrical 1,x-diols and also BAs with different
headgroups and/or unequal alkyl chain lengths substituted
at the phenyl ring.
against LiBr, and 1-(bromomethyl)-4-(chloromethyl)ben-
zene and 1,4-bis(bromomethyl)benzene (4b) are formed
instead.
In thefollowingreaction thethp residueof compound9 was
substituted by bromide [31] quantitatively resulting in 1-[16-
(benzyloxy)hexadecyl]-4-(bromomethyl)benzene (10). After-
wards, this bromide can be used for the second Grignard
coupling: compound 3a was converted to the Grignard reagent
and subsequently coupled with compound 10 to yield 2-[[16-
[4-[16-(benzyloxy)hexadecyl]phenyl]hexadecyl]oxy]tetra-
hydro-2H-pyran (11), a phenylene-modified long-chain 1,x-
diol with orthogonally cleavable protecting groups. Since the
chromatographic properties of the desired compound 11 and
the by-product, the bis(tetrahydro-2H-pyran) of the tria-
contane-1,30-diol, are very similar, a failure in purification
was anticipated and we decided to use the crude product of 11
for further synthesis instead. The cleavage of the thp residue
using the procedure mentioned above gave the mono-pro-
tected 16-[4-[16-(benzyloxy)hexadecyl]phenyl]hexadecan-
1-ol (12) in an acceptable yield of 40 % with respect to
bromide 10 after chromatography. Finally, the benzyl moiety
of compound 12 was removed by hydrogenolysis using
palladium on carbon as catalyst and a mixture of heptane,
Phosphorylation and quarternisation
For the final phosphorylation step we used common
reagents, such as b-bromoethylphosphoric acid dichloride,
according to the method described by Eibl and Nicksch
[32]. The subsequent quarternisation reaction was carried
out with Me3N in a solvent mixture of CHCl3, CH3CN, and
EtOHtoprovidethebis(phosphocholine)s1 and2 (Scheme 3)
in 42–45 % isolated yields.
Regarding the purification of the biphenylene-modified
BA 2, it is noteworthy that even if we used the ‘short-
chain’ by-product-contaminated diol 6b for the phosphor-
ylation and quarternisation reaction, a separation of the
123

S. Drescher et al.
Scheme 3
O
P
i, ii
N
O
O
6a, 6b
O
O
P
n
42-45%
O
O
N
n
m
O
1, m = 1, n = 14
2
, m = 2, n = 10
i) CHCl3, -bromoethylphosphoric acid dichloride, 50 °C; then THF-H2O, rt, 1.5 h; ii) CHCl3, CH3CN,
EtOH, N(CH3)3, 50 °C, 24 h; then 3-4 days rt
resulting BA 2 from its ‘biphenylene-free’ by-product PC-
C22-PC could be achieved. However, this purification
failed if we used 6a* for the phosphorylation and quar-
ternisation reaction owing to the similar chromatographic
properties of the phenylene-modified BA 1 and its ‘phe-
nylene-free’ by-product PC-C30-PC. Regarding the latter
phenomenon, we conclude that (1) the synthesis of phe-
nylene-modified BAs is impossible using the Grignard
bis-coupling reaction, hence, the stepwise Grignard homo-
coupling should be used and (2) the difference in the chain
length of the bola molecules should be at least two phenyl
rings, corresponding to eight carbon atoms, for a successful
purification of the final BAs.
Aggregation behaviour
The aggregation behaviour of both novel phenylene- and
biphenylene-modified BAs was studied in a preliminary
investigation using TEM. As mentioned before, the BA
without any modification (e.g. PC-C32-PC) aggregates in
water into long and flexible nanofibres. To visualise the
aggregates formed in suspension, TEM images were taken
of samples prepared at 25 °C. Figure 3 shows the images
of the negatively stained samples: Small worm-like
aggregates are visible. In the case of BA 1 (Fig. 3a) the
aggregates have a size of about 8–18 nm in length and
3–4 nm in width. Some of the aggregates are clustered
together. The shape of the aggregates of BA 2 (Fig. 3b) is
more irregular compared to BA 1, but they also exhibit the
form of small ellipsoidal micelles with slightly smaller
dimensions. However, the formation of long fibrous
aggregates, as was previously shown for PC-C32-PC sus-
pensions [15, 16], was not observed for the samples at
ambient temperatures. Since the self-assembly process of
the PC-C32-PC into long nanofibres is exclusively driven
by van der Waals interactions of the long alkyl chain
[14–16], perturbations such as phenylene- or biphenylene
rings within the polymethylene chain provoke decreased
interactions of the long alkyl chains and, hence, resulted in
a destabilization of the fibre aggregates [33]. This phe-
nomenon is also observed if other chemical modifications,
Fig. 3 TEM images of aqueous uranyl acetate stained suspensions of
bolaamphiphiles 1 (a) and 2 (b) at a concentration of 0.05 mg/cm3.
The samples were prepared at 25 °C. The bar corresponds to 50 nm
such as sulphur or oxygen atoms [34], diacetylene groups
[35, 36] and methyl branches [33], respectively, were
inserted into the alkyl chain.
123

Synthesis of symmetrical, single-chain, phenylene/biphenylene-modified bolaamphiphiles
Summary and outlook
recorded on an AMD 402 (70 eV) spectrometer (EI–MS).
High-resolution mass spectra (HR–MS) were recorded on a
Thermo Fisher Scientific LTQ-Orbitrap mass spectrometer
with static nano-electrospray ionisation. Elemental analy-
ses (C, H, N) were conducted using a Leco CHNS-932;
results were in good agreement ( 0.4 %) with calculated
values.
The synthesis of biphenylene-modified BAs is feasible
using the Grignard bis-coupling reaction. The preparation
of the corresponding phenylene-modified BA failed
because of purification problems of the resulting bola
molecules. The alternative route using an unsymmetrical
starting material and a stepwise Grignard homo-coupling
led to the desired phenylene-modified BA in acceptable
yields. During this homo-coupling, the ‘short-chain’ phe-
nylene-free by-products, whose formation is a general
drawback of the mentioned Grignard bis-coupling, can be
successfully separated from the desired products. More-
over, the Grignard homo-coupling is also adaptable to the
preparation of unsymmetrical bola molecules with unequal
alkyl chain lengths substituted at the phenyl ring and/or
different headgroups. The new phenylene/biphenylene-
modified BAs do not self-assemble into long nanofibres in
aqueous suspension as previously observed for bolalipids
with long alkyl chains, but aggregate at room temperature
into small ellipsoidal aggregates.
General procedure for the preparation of phenylene/
biphenylene-modified 1,x-diols using Grignard
bis-coupling
A solution of 3.13–12.90 g 3a (8–33 mmol) or 6.0 g 3b
(18 mmol), respectively, in 20–100 cm3 dry THF was
added dropwise to 0.29–1.22 g magnesium turnings
(12–50 mmol) under argon atmosphere while stirring.
Afterwards, the mixture was heated to 50 °C for 2 h. The
excess Mg was removed under argon atmosphere and the
Grignard solution was cooled to 0 °C. Then 1–3 cm3
Li2CuCl4 (0.1 M in THF) was added with stirring followed
by a solution of 0.61 g 4a (3.5 mmol), 0.92–3.96 g 4b
(3.5–15 mmol), 2.0 g 4c (8 mmol), or 2.7 g 4d (8 mmol)
in 20–50 cm3 dry THF. The stirring was continued for a
further 3 h at 0 °C. For the work-up 100–200 cm3 Et2O
was added and the resulting mixture was poured into
100–200 cm3 of a cold saturated solution of NH4Cl. The
organic layer was separated and the aqueous phase was
extracted with 100–200 cm3 Et2O (29). The combined
organic phases were washed with 100–200 cm3 H2O (29),
dried over Na2SO4 and concentrated to dryness in vacuo.
The white and waxy crude products 5a and 5b were used
for the cleavage of the thp-protecting group without further
purification: the crude bis(tetrahydro-2H-pyran)s 5 were
dissolved in 50 cm3 dry MeOH and heated under reflux for
3 h with catalytic amounts of PPTS. The hot suspension
was filtered and the white residue was recrystallized from
heptane or purified by column chromatography using the
gradient technique and CHCl3–Et2O as eluent.
The preparation of phenylene-modified BAs with longer
alkyl chains including other types of C–C coupling reac-
tions, a detailed characterisation of the novel aggregates as
well as investigations regarding the mixing behaviour of
these new BAs with common phospholipids are underway.
Preliminary experiments on the incorporation of pheny-
lene-modified BAs into phospholipid bilayers showed
promising results. The results of these investigations will
be published in a separate paper.
Experimental
All chemicals were purchased from Sigma-Aldrich Co. and
used without further purification. b-Bromoethylphosphoric
acid dichloride was prepared according to the literature
[37]. All solvents were dried and distilled before use. The
purity of all compounds was checked by thin-layer chro-
matography (TLC) using silica gel 60 F254 plates (Merck)
and the following mobile phases: A = CHCl3–heptane
(3:2, v/v), B = CHCl3–heptane (2:3, v/v), C = CHCl3,
D = CHCl3–Et2O (1:1, v/v), E = CHCl3–MeOH–NH3
(10:10:3, v/v/v). Silica gel (Merck, 0.063–0.200 mm) was
used for column chromatography of the products. The
purification of the final BAs was carried out by medium
pressure liquid chromatography (MPLC, Bu¨chi) on silica
gel (Merck, 0.032–0.060 mm). Melting points were deter-
16,160-(1,4-Phenylene)bis(hexadecan-1-ol) (6a*)
The product could not be separated from the by-product
triacontane-1,30-diol using recrystallization or chromato-
graphy.
12,120-(Biphenyl-4,40-diyl)bis(dodecan-1-ol)
(6b, C38H58O2)
Compound 6b was obtained as a white solid in 28 % yield
(1.17 g) from reaction of 6.0 g 3b (18 mmol) with 2.0 g 4c
(8 mmol) and in 40 % yield from reaction of 6.0 g 3b
(18 mmol) with 2.7 g 4d (8 mmol) according to the
general procedure described above. M.p.: 126–128 °C;
1H NMR (400 MHz, CDCl3): d = 1.25–1.31 (m, 32H, 29
–(CH2)2(CH2)8(CH2)2O–), 1.54–1.62 (m, 8H, 29
–CH2CH2(CH2)8CH2CH2O–), 2.61 (t, 3J = 7.8 Hz, 4H,
1
mined with a Boetius apparatus. H and 13C NMR spectra
were recorded on a Varian Gemini 2000 spectrometer or a
Varian Inova 500 with the use of CDCl3 or CD3OD as
internal standard. Mass spectrometric data were obtained
with a Finnigan MAT SSQ 710 C (ESI–MS) or were
123

S. Drescher et al.
29 –CH2(CH2)11O–), 3.62 (t, 3J = 6.6 Hz, 4H, 29 –(CH2)11
CH2O–), 7.20–7.48 (m, 8H, –C6H4C6H4–) ppm; 13C NMR
(100 MHz, CDCl3, 45 °C): d = 25.74 (–CH2(CH2)2OH),
29.37, 29.54, 29.60, 31.35 (–CH2CH2C6H4C6H4CH2CH2–),
32.81 (–CH2CH2OH), 35.55 (–CH2C6H4C6H4CH2–), 63.05
(–CH2OH), 126.71 (2,20,6,60-C, –C6H4C6H4–), 128.63 (3,30,
5,50-C, –C6H4C6H4–), 138.51 (1,10-C, –C6H4C6H4–), 141.68
(4,40-C, –C6H4C6H4–) ppm; MS (70 eV): m/z (%) = 522
(100) [M?], 504 (5) [M?-H2O].
100 cm3 of a cold saturated solution of NH4Cl. The organic
layer was separated and the aqueous phase was extracted
with 60 cm3 Et2O (29). The combined organic phases
were washed with 50 cm3 brine and 50 cm3 H2O, dried
over Na2SO4 and concentrated to dryness in vacuo. The
residue was purified by column chromatography using the
gradient technique and heptane–Et2O and TEA (0.5 %) as
eluent to afford 0.91 g (25 %) of 9 as white crystalline
substance. Rf = 0.42 (solvent B); 1H NMR (500 MHz,
CDCl3): d = 1.24–1.36 (m, 24H, –(CH2)2(CH2)12-
(CH2)2O–), 1.50–1.89 (m, 10H, –CH2CH2(CH2)12-
CH2CH2O–, –OCH(CH2)3CH2O–), 2.57 (t, 3J = 7.5 Hz,
2H, –CH2C6H4CH2O–), 3.45 (t, 3J = 6.7 Hz, 2H,
–CH2OCH2C6H5), 3.51–3.56 (m, 1H, –OCH(CH2)3-
CH2O–), 3.88–3.94 (m, 1H, –OCH(CH2)3CH2O–), 4.49
(d, 2J = 12.2 Hz, 1H, –CH2C6H4CH2O–), 4.57 (s, 2H,
–OCH2C6H5), 4.68–4.70 (m, 1H, –OCHO–), 4.77 (d,
2J = 12.2 Hz, 1H, –CH2C6H4CH2O–), 7.13–7.15 (m, 2H,
2,6-H –C6H4–), 7.25–7.27 (m, 2H, 3,5-H –C6H4–),
7.32–7.35 (m, 5H, –C6H5) ppm; 13C NMR (125 MHz,
CDCl3): d = 19.32 (CHCH2CH2(CH2)2O–), 25.45
(CH(CH2)2CH2CH2O–), 26.19 (–CH2(CH2)2OCH2C6H5),
29.33, 29.43, 29.51, 29.59, 29.60, 29.65, 29.66, 29.77
(–CH2–), 30.54 (CHCH2(CH2)3O–), 31.51 (–CH2CH2C6
H4CH2O–), 35.69 (–CH2C6H4CH2O–), 62.07 (CH(CH2)3
CH2O–), 68.34 (–C6H4CH2O–), 70.53 (–OCH2C6H5),
72.83 (–CH2OCH2C6H5), 97.78 (–OCHO–), 127.41 (4-C,
–C6H5), 127.58 (2,6-C, –C6H5), 127.89 (2,6-C, –C6H4–),
128.30 and 128.37 (3,5-C, –C6H5 and –C6H4–), 138.70 and
138.73 (1-C, –C6H5 and –C6H4–), 142.28 (4-C, –C6H4–)
ppm; MS (70 eV): m/z (%) = 522 (49) [M?].
General procedure for the preparation
of phenylene-modified 1,x-diols using stepwise
Grignard homo-coupling
2-[[4-(Chloromethyl)benzyl]oxy]tetrahydro-2H-pyran
(8, C13H17ClO2)
A solution of 5.0 g 7 (32 mmol) and 3.8 g DHP (45 mmol) in
100 cm3 dryCHCl3 wasstirred at25 °C for24 h. Afterwards,
100 cm3 water was added and the organic layer was
separated. The aqueous residue was extracted with 50 cm3
CHCl3 (29). Thecombined organic phaseswerewashed with
50 cm3 water, dried over Na2SO4, evaporated, and the
residue was purified by column chromatography using the
gradient technique and heptane–Et2O and TEA (0.5 %) as
eluent to afford 6.5 g (84 %) of 8 as colourless oil. Rf = 0.44
(solventA);1HNMR(400 MHz,CDCl3):d = 1.46–1.81(m,
6H, –OCH(CH2)3CH2O–), 3.43–3.49 (m, 1H, –OCH(CH2)3-
CH2O–), 3.78–3.86 (m, 1H, –OCH(CH2)3CH2O–), 4.44 (d,
2J = 12.2 Hz, 1H, –OCH2C6H4–), 4.50 (s, 2H, –CH2Cl),
2
4.63–4.64 (m, 1H, –OCHO–), 4.71 (d, J = 12.2 Hz, 1H,
–OCH2C6H4–), 7.29 (s, 4H, –C6H4–) ppm; 13C NMR
(100 MHz, CDCl3): d = 19.27 (CHCH2CH2(CH2)2O–),
25.43 (CH(CH2)2CH2CH2O–), 30.49 (CHCH2(CH2)3O–),
45.92 (ClCH2–), 61.99 (CH(CH2)3CH2O–), 68.25
(–OCH2C6H4–), 97.69 (–OCHO–), 127.88 and 128.44
(2,3,5,6-C, –C6H4–), 136.54 and 138.60 (1,4-C, –C6H4–)
ppm; MS (70 eV): m/z (%) = 240 (14) [M?], 205 (7)
[M?-Cl], 139 (100) [M?-C5H9O2], 104 (59) [M?-
C5H9O2Cl], 91 (15) [C7H7?], 85 (39) [C5H9O?].
1-[16-(Benzyloxy)hexadecyl]-4-(bromomethyl)benzene
(10, C30H45BrO)
A solution of 0.19 cm3 bromine (3.8 mmol), diluted in 5 cm3
CH2Cl2, was added dropwise into a solution of 1.0 g
triphenylphosphane (3.8 mmol) in 10 cm3 dry CH2Cl2 whilst
stirringat 0 °C. A solutionof 0.9 g 9 (1.7 mmol), dissolvedin
3 cm3 CH2Cl2, was added and the mixture was stirred for
16 h at 25 °C. Afterwards, the organic layer was washed with
15 cm3 brine (29), dried over Na2SO4, evaporated, and the
crude product was purified by column chromatography with
heptane as eluent to afford 0.77 g (90 %) of 10 as white
crystalline substance. Rf = 0.53 (solvent A); 1H NMR
(400 MHz, CDCl3): d = 1.23–1.36 (m, 24H, –(CH2)2
(CH2)12(CH2)2O–), 1.54–1.63 (m, 4H, –CH2CH2
(CH2)12CH2CH2O–), 2.57 (t, 3J = 7.8 Hz, 2H, –CH2C6
H4CH2Br), 3.45 (t, 3J = 6.7 Hz, 2H, –CH2OCH2
C6H5), 4.47 (s, 2H, –CH2Br), 4.48 (s, 2H, –OCH2C6H5),
7.12–7.16 (m, 2H, 2,6-H –C6H4–), 7.25–7.29 (m, 2H, 3,5-H,
–C6H4–), 7.32–7.33 (m, 5H, –C6H5) ppm; 13C NMR
(100 MHz, CDCl3): d = 26.20 (–CH2(CH2)2OCH2C6H5),
29.30, 29.31, 29.49, 29.57, 29.60, 29.61, 29.66, 29.67, 29.78
2-[[4-[16-(Benzyloxy)hexadecyl]benzyl]oxy]-
tetrahydro-2H-pyran (9, C32H54O3)
A solution of 2.66 g benzyl 15-bromopentadecyl ether
(6.7 mmol) in 25 cm3 dry THF was added dropwise to
0.24 g magnesium turnings (10 mmol) under argon atmo-
sphere while stirring. Afterwards, the mixture was heated
to 50 °C for 3 h. The excess magnesium was removed
under argon atmosphere and the Grignard solution was
cooled to 0 °C. Then 1.0 cm3 Li2CuCl4 (0.1 M in THF)
was added with stirring followed by a solution of 1.6 g 8
(6.6 mmol) in 10 cm3 dry THF. The stirring was continued
for a further 2 h at 0 °C. For the work-up 100 cm3 Et2O
was added and the resulting mixture was poured into
123

Synthesis of symmetrical, single-chain, phenylene/biphenylene-modified bolaamphiphiles
(–CH2–), 31.38 (–CH2CH2C6H4CH2O–), 33.79 (–CH2Br),
35.67 (–CH2C6H4CH2O–), 70.54 (–OCH2C6H5), 72.85
(–CH2OCH2C6H5), 127.43 (4-C, –C6H5), 127.59 (2,6-C,
–C6H5), 128.32 (3,5-C, –C6H5), 128.52 (3,5-C, –C6H4–),
128.76 (2,6-C, –C6H4–), 134.96 (1-C, –C6H4–), 138.74 (1-C,
–C6H5), 143.38 (4-C, –C6H4–) ppm; MS (70 eV):
m/z (%) = 501 (2) [M?], 421 (40) [M?-Br], 329 (10)
[M?-C7H7Br], 91 (100) [C7H7?].
16,160-(1,4-Phenylene)bis(hexadecan-1-ol)
(6a, C38H70O2)
A mixture of 0.15 g 12 (0.23 mmol), dissolved in 50 cm3
heptane–EtOH–ethyl acetate (3:1:1, v/v/v), and 35 mg
palladium (10 % on carbon) was stirred under hydrogen
(5 atm) at 25 °C for 18 h. The catalyst was removed by
filtration and washed with warm CHCl3 several times. The
combined organic layers were evaporated and the white
residue was recrystallized from heptane to afford 0.12 g
(95 %) of 6a. M.p.: 102–103 °C; Rf = 0.27 (solvent D); 1H
NMR (500 MHz, CDCl3): d = 1.23–1.33 (m, 48H,
29 –(CH2)2(CH2)12(CH2)2OH), 1.51–1.60 (m, 8H, 29
16-[4-[16-(Benzyloxy)hexadecyl]phenyl]hexadecan-1-ol
(12, C45H76O2)
A solution of 1.1 g 3a (2.7 mmol) in 6 cm3 dry THF was
added dropwise to 0.1 g magnesium turnings (4 mmol)
under argon atmosphere while stirring. Afterwards, the
mixture was heated to 50 °C for 3 h. The excess magne-
sium was removed under argon atmosphere and the
Grignard solution was cooled to 0 °C. Then 0.5 cm3
Li2CuCl4 (0.1 M in THF) was added with stirring followed
by a solution of 0.5 g 10 (1 mmol) in 5 cm3 dry THF. The
stirring was continued for a further 3 h at 0 °C. For the
work-up 15 cm3 Et2O was added and the resulting mixture
was poured into 25 cm3 of a cold saturated solution of
NH4Cl. The organic layer was separated and the aqueous
phase was extracted again with 30 cm3 Et2O (29). The
combined organic phases were washed with 30 cm3 brine
and 30 cm3 water, dried over Na2SO4 and concentrated to
dryness in vacuo. The white and waxy crude product 11
was used for the cleavage of the thp-protecting groups
without further purification. Therefore, compound 11 was
dissolved in 20 cm3 dry MeOH and heated under reflux for
2 h with catalytic amounts of PPTS. The reaction was then
allowed to stand overnight at 25 °C. The cold suspension was
filtered and the white residue was purified by column
chromatography using CHCl3–MeOH (95:5, v/v) as eluent to
afford 0.26 g (40 % with respect to 10) of 12 as white
crystalline substance. M.p.: 82–83 °C; Rf = 0.12 (solvent A),
3
–CH2CH2(CH2)12CH2CH2OH), 2.52 (t, J = 7.7 Hz, 4H,
–CH2C6H4CH2–), 3.62 (t, 3J = 6.6 Hz, 4H, 29 –CH2OH),
7.06 (s, 4H, –C6H4–) ppm; 13C NMR (125 MHz, CDCl3):
d = 25.74 (–CH2(CH2)2OH), 29.39, 29.43, 29.53, 29.59,
29.61, 29.66, 31.58 (–CH2CH2C6H4CH2CH2–), 32.82
(–CH2CH2OH), 35.57 (–CH2C6H4CH2–), 63.11 (–CH2
OH), 128.20 (2,3,5,6-C, –C6H4–), 140.07 (1,4-C, –C6H4–)
ppm; MS (70 eV): m/z (%) = 558 (100) [M?], 540 (58)
[M?-H2O].
General procedure for phosphorylation
and quarternisation reaction
The reaction procedure is based on the general synthesis of
polymethylene-1,x-diyl-bis(phosphocholines) described
previously [14]: 0.96 g b-bromoethylphosphoric acid
dichloride (4 mmol) was poured into 20 cm3 dry CHCl3
under cooling with ice–water. A mixture of 1.0 cm3 dry
TEA (7 mmol) in 20 cm3 dry CHCl3 was added slowly
with stirring, which was continued for 30 min at 0 °C. The
diols 6 (0.5 mmol, in solid form) were poured in one
portion into the mixture. To completely dissolve the solid
diols, the reaction mixture was gently warmed until a clear
solution appeared. Afterwards, the mixture was stirred at
25 °C for 24 h. After the complete conversion of the diols,
40 cm3 crushed ice was added to the solution and the
mixture was stirred vigorously for a further 2 h. The
organic layer was separated and the aqueous phase was
extracted with 20 cm3 CHCl3 (39). The combined organic
layers were evaporated and the residue was dissolved in
10 cm3 THF–H2O (9:1, v/v). After 1.5 h, the solvent was
evaporated and the crude bromoesters were transferred into
a mixture of 10 cm3 CHCl3, 10 cm3 CH3CN and an etha-
nolic solution of Me3N (2 mmol, 4.2 M). The mixture was
kept in a closed tube at 50 °C for 24 h. Thereafter, the clear
solution was allowed to stand for 3–4 days at 25 °C.
Afterwards, the reaction mixture was evaporated to dryness
and the residue was purified by MPLC using CHCl3–
MeOH–H2O as eluent and the gradient technique to afford
0.18–0.20 g (42–45 %) of the phenylene/biphenylene-
modified BAs 1 and 2.
1
0.27 (solvent C), 0.57 (solvent D); H NMR (500 MHz,
CDCl3): d = 1.24–1.35 (m, 48H, 29 –(CH2)2(CH2)12
(CH2)2O–), 1.52–1.62 (m, 8H, 29 –CH2CH2(CH2)12
CH2CH2O–), 2.54 (t, 3J = 7.7 Hz, 4H, –CH2C6H4CH2–),
3.45 (t, 3J = 6.6 Hz, 2H, –CH2OCH2C6H5), 3.62 (t,
3J = 6.6 Hz, 2H, –CH2OH), 4.48 (s, 2H, –OCH2C6H5), 7.06
(s, 4H, –C6H4–), 7.25–7.33 (m, 5H, –C6H5) ppm; 13C NMR
(125 MHz, CDCl3): d = 25.73 (–CH2(CH2)2OH), 26.19
(–CH2(CH2)2OCH2C6H5), 29.39, 29.43, 29.48, 29.53, 29.60,
29.66, 29.77, 31.58 (–CH2CH2C6H4CH2CH2–), 32.82
(–CH2CH2OH), 35.56 (–CH2C6H4CH2–), 63.10 (–CH2OH),
70.54 (–CH2C6H5), 72.84 (–CH2OCH2C6H5), 127.43 (4-C,
–C6H5), 127.60 (2,6-C, –C6H5), 128.20 (2,3,5,6-C, –C6H4–),
128.32 (3,5-C, –C6H5), 138.74 (1-C, –C6H5), 140.07 (1,4-C,
–C6H4–) ppm; MS (70 eV): m/z (%) = 648 (4) [M?], 630 (11)
[M?-H2O], 557 (66) [M?-C7H7], 539 (100) [M?-C7H9O],
521 (74) [M?-C7H11O2].
123

S. Drescher et al.
16,160-(1,4-Phenylene)bis[hexadec-1-yl-
[2-(trimethylammonio)ethyl-
phosphate]] (1, C48H94N2O8P2)
Sample preparation
The appropriate amount of bolalipid was suspended in
water (ultrapure water from a Milli-Q A10 system, Milli-
pore GmbH, Schwalbach, Germany) to a concentration of
1.0 mg/cm3. To achieve a homogeneous suspension, the
samples were heated above 80 °C three times and vortexed.
Prior to the TEM measurements, the bola stock suspensions
were diluted with ultrapure water to a concentration of
0.05 mg/cm3 and allowed to stand at 25 °C for at least
24 h.
Compound 1 was obtained as a white solid in 45 % yield
(0.20 g) from the phosphorylation reaction of 0.28 g 6a
(0.5 mmol) and subsequent quarternisation with Me3N
following the general procedure described above.
Rf = 0.26 (solvent E); 1H NMR (500 MHz, CDCl3–
CD3OD): d = 1.08–1.20 (m, 48H, 29 –(CH2)2(CH2)12
(CH2)2O–), 1.38–1.47 (m, 8H, 29 –CH2CH2(CH2)12
3
CH2CH2O–), 2.38 (t, J = 7.7 Hz, 4H, –CH2C6H4CH2–),
3.03 (s, 18H, 69 –CH3), 3.39–3.41(m, 4H, 29 NCH2CH2O–),
3
3.68 (q, J = 6.6 Hz, 4H, 29 –(CH2)15CH2O–), 4.02–4.06
Transmission electron microscopy (TEM)
(m, 4H, 29 NCH2CH2O–), 6.90 (s, 4H, –C6H4–) ppm; 13
C
NMR (125 MHz, CDCl3–CD3OD): d = 25.47 (–(CH2)13
CH2(CH2)2O–), 29.03, 29.10, 29.20, 29.28, 29.33, 29.35,
Samples were prepared by spreading 5 mm3 of the BA
suspension (c = 0.05 mg/cm3) onto a Cu grid coated with
a formvar film. After 1 min, excess liquid was blotted off
with filter paper and 5 mm3 of 1 % aqueous uranyl acetate
solution was placed onto the grid and drained off after
1 min. The dried specimens were examined with a Zeiss
EM 900 transmission electron microscope.
3
29.36, 29.38, 30.46 (d, JC,P = 7.7 Hz, 29 –(CH2)14
CH2CH2O–), 31.28 (–CH2CH2C6H4CH2CH2–), 35.22
(–CH2C6H4CH2–), 53.88 (t, J = 3.8 Hz, 69 –CH3), 58.46
2
(d, JC,P = 5.3 Hz, 29 NCH2CH2O–), 65.65 (d, JC,P
2
=
6.2 Hz, 29 –(CH2)15CH2O–), 66.27 (b, 29 NCH2CH2O–),
127.89 (2,3,5,6-C, –C6H4–), 139.77 (1,4-C, –C6H4–) ppm;
ESI–MS: m/z = 467.39 [M2??2Na], 889.47 [M??H],
911.36 [M??Na], 923.42 [M-?Cl]; HR-MS: calcd. for
C48H95N2O8P2 [M??H]: 899.6558, found 889.6573; calcd.
for C48H94N2O8P2Na [M??Na]: 911.6378, found 911.6389.
Acknowledgments This work was supported by grants from the
Deutsche Forschungsgemeinschaft (projects Bl 182/19-3 and Do
463/4-2). We thank Prof. Andrea Sinz and Dr. Christian Ihling
(Department of Pharmaceutical Chemistry and Bioanalytics, Martin-
Luther-University Halle-Wittenberg) for the high resolution mass
spectrometry analysis. The support of Dr. Gerd Hause (Biocenter,
Martin-Luther-University Halle-Wittenberg) by providing us access
to the electron microscope facility is greatly appreciated. Finally, S.D.
and B.D. thank Dipl.-Pharm. Susann Althaus and Dr. Beate Stiebitz
for the help in the synthesis.
12,120-(Biphenyl-4,40-diyl)bis[dodec-1-yl-
[2-(trimethylammonio)ethyl-
phosphate]] (2, C46H82N2O8P2)
Compound 2 was obtained as a white solid in 42 % yield
(0.18 g) from the phosphorylation reaction of 0.26 g 6b
(0.5 mmol) and subsequent quarternisation with Me3N
following the general procedure described above.
Rf = 0.24 (solvent E); 1H NMR (500 MHz, CDCl3–
CD3OD): d = 1.19–1.29 (m, 32H, 29 –(CH2)2(CH2)8
(CH2)2O–), 1.52–1.62 (m, 8H, 29 –CH2CH2(CH2)8
CH2CH2O–), 2.58 (t, 3J = 7.6 Hz, 4H, 29 –CH2
(CH2)11O–), 3.17 (s, 18H, 69 CH3), 3.56–3.57 (m, 4H,
29 NCH2CH2O–), 3.79 (q, J = 6.6 Hz, 4H, 29 –(CH2)11
CH2O–), 4.15–4.20 (m, 4H, 29 NCH2CH2O–), 7.10–7.46 (m,
8H, –C6H4C6H4–) ppm; 13C NMR (125 MHz, CDCl3–
CD3OD): d = 25.62 (–(CH2)9CH2(CH2)2O–), 28.87, 29.23,
29.34, 29.39, 29.43, 29.62, 30.68 (–(CH2)10CH2CH2O–),
31.19 (–CH2CH2(CH2)10O–), 35.37 (–CH2(CH2)11O–), 54.35
References
1. Fuhrhop J-H, Wang T (2004) Chem Rev 104:2901
2. De Rosa M, Esposito E, Gambacorta A, Nicolaus B, Bu’Lock JD
(1980) Phytochemistry 19:827
3. Langworthy TA (1977) Biochim Biophys Acta Lipids Lipid
Metab 487:37
4. Bakowsky U, Rothe U, Antonopoulos E, Martini T, Henkel L,
Freisleben HJ (2000) Chem Phys Lipids 105:31
5. Benvegnu T, Rethore G, Brard M, Richter W, Plusquellec D
(2005) Chem Commun 5536
6. Brown DA, Venegas B, Cooke PH, English V, Chong PL-G
(2009) Chem Phys Lipids 159:95
7. Cornell BA, Braach-Maksvytis VBL, King LG, Osmann PDJ,
Raguse B, Wieczorek L, Pace RJ (1997) Nature 387:580
8. Patel GP, Ponce A, Zhou H, Chen W (2008) Int J Toxicol 27:329
´
´
2
´
9. Rethore G, Montier T, Le Gall T, Delepine P, Cammas-Marion S,
Lemiegre L, Lehn P, Benvegnu T (2007) Chem Commun 2054
´
´
(b, –CH3), 58.80 (NCH2CH2O–), 64.09 (d, JC,P = 7.7 Hz,
`
–(CH2)11CH2O–), 65.83 (b, NCH2CH2O–), 126.64 (2,20,6,60-
C, –C6H4C6H4–), 128.65 and 128.77 (3,30,5,50-C, –C6H4
C6H4–), 138.32(1,10-C, –C6H4C6H4–), 141.64(4,40-C, –C6H4
C6H4–) ppm; ESI–MS: m/z = 853.5 [M??H], 875.4
[M??Na]; HR-MS: calcd. for C46H83N2O8P2 [M??H]:
853.5625, found 853.5622; calcd. for C46H82N2O8P2Na
[M??Na]: 875.5444, found 875.5441.
10. Sprott GD, Tolson DL, Patel GB (1997) FEMS Microbiol Lett
154:17
11. Benvegnu T, Brard M, Plusquellec D (2004) Curr Opin Colloid
Interface Sci 8:469
12. Meister A, Blume A (2007) Curr Opin Colloid Interface Sci
12:138
13. Meister A, Blume A (2012) Adv Planar Lipid Bilayers Lipo-
somes. doi:10.1016/B978-0-12-396534-9.00004-0
123

Synthesis of symmetrical, single-chain, phenylene/biphenylene-modified bolaamphiphiles
14. Drescher S, Meister A, Blume A, Karlsson G, Almgren M,
Dobner B (2007) Chem Eur J 13:5300
23. Zou B, Wang M, Qiu D, Zhang X, Chi L, Fuchs H (2002) Chem
Commun 1008
¨
¨
15. Kohler K, Forster G, Hauser A, Dobner B, Heiser UF, Ziethe F,
Richter W, Steiniger F, Drechsler M, Stettin H, Blume A (2004) J
Am Chem Soc 126:16804
24. Wu G, Verwilst P, Xu J, Xu H, Wang R, Smet M, Dehaen W,
Faul CFJ, Wang Z, Zhang X (2012) Langmuir 28:5023
25. Grechishnikova I, Johansson LB-A, Molotkovsky JG (1996)
Chem Phys Lipids 81:87
26. McKiernan R, Gido SP, Penelle J (2002) Polymer 43:3007
27. Tamura M, Kochi J (1971) Synthesis 303
¨
¨
16. Kohler K, Forster G, Hauser A, Dobner B, Heiser UF, Ziethe F,
Richter W, Steiniger F, Drechsler M, Stettin H, Blume A (2004)
Angew Chem Int Ed 43:245
17. Meister A, Drescher S, Mey I, Wahab M, Graf G, Garamus VM,
28. Heiser UF, Dobner B (1997) J Chem Soc Perkin Trans 1:809
29. Heiser UF, Dobner B (1996) Chem Commun 2025
30. Kathiresan M, Walder L, Ye F, Reuter H (2010) Tetrahedron Lett
51:2188
31. Schwarz M, Oliver JE, Sonnet PE (1975) J Org Chem 40:2410
32. Eibl H, Nicksch A (1975) Ger Offen DE 2345057 A1 19750327;
(1975) Chem Abstr 83:27617
33. Blume A, Drescher S, Meister A, Graf G, Dobner B (2012)
Faraday Disc. doi:10.1039/C2FD20102J
34. Drescher S, Meister A, Graf G, Hause G, Blume A, Dobner B
(2008) Chem Eur J 14:6796
¨
Hause G, Mogel H-J, Janshoff A, Dobner B, Blume A (2008) J
Phys Chem B 112:4506
¨
18. Meister A, Kohler K, Drescher S, Dobner B, Karlsson G,
Edwards K, Hause G, Blume A (2007) Soft Matter 3:1025
¨
¨
19. Kohler K, Meister A, Forster G, Dobner B, Drescher S, Ziethe F,
Richter W, Steiniger F, Drechsler M, Hause G, Blume A (2006)
Soft Matter 2:77
20. Meister A, Bastrop M, Koschoreck S, Garamus VM, Sinemus T,
Hempel G, Drescher S, Dobner B, Richtering W, Huber K,
Blume A (2007) Langmuir 23:7715
21. Meister A, Drescher S, Karlsson G, Hause G, Baumeister U,
Hempel G, Garamus VM, Dobner B, Blume A (2010) Soft Matter
6:1317
22. Meister A, Drescher S, Garamus VM, Karlsson G, Graf G,
Dobner B, Blume A (2008) Langmuir 24:6238
35. Drescher S, Helmis K, Langner A, Dobner B (2010) Monatsh
Chem 141:339
36. Graf G (2011) Ph.D. thesis. Martin-Luther-University, Halle
37. Hirt G, Berechthold R (1958) Pharm Acta Helv 33:349
123