The effect of midpolar regime mimics on anion transport mediated by amphiphilic heptapeptides

Article abstract of DOI:10.1039/b705179b

Nine amphiphilic heptapeptides, synthetic anion transporters (SATs) of the form (C₁₈H₃₇)₂N-Y-(Gly)₃-Pro-(Gly) ₃-OCH₂Ph were prepared. The unit (OH₂)Y represents the diacids succi

Full text of DOI:10.1039/b705179b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
The effect of midpolar regime mimics on anion transport mediated by
amphiphilic heptapeptides
a
a
b
b
Robert Pajewski, Jolanta Pajewska, Ruiqiong Li, Megan M. Daschbach,
a
ac
Elizabeth A. Fowler and George W. Gokel*
Received (in Durham, UK) 4th April 2007, Accepted 22nd June 2007
First published as an Advance Article on the web 1st August 2007
DOI: 10.1039/b705179b
Nine amphiphilic heptapeptides, synthetic anion transporters (SATs) of the form
18 37 2 3 3 2 2
(C H ) N–Y–(Gly) -Pro-(Gly) –OCH Ph were prepared. The unit (OH )Y represents the diacids
succinic, glutaric, diglycolic, 3-thiaglutaric, N-methyliminodiglycine, isophthalic, and terephthalic
acids. Additionally, Y was absent or present as acetic acid affording the structure
(
C H ) N–(Gly) -Pro-(Gly) –OCH Ph or (C H ) N–(Gly) -Pro-(Gly) –OCH Ph. The
18 37 2 3 3 2 18 37 2 4 3 2
diglycolic acid derivative was reported previously but the remaining compounds are new. These
ꢀ
nine peptides mediated release of Cl from DOPC/DOPA vesicles with varying efficacy. Chloride
release diminished for the Y-containing amphiphilic heptapeptides in the order glutaric, succinic
4
3-thiaglutaric 4 terephthalic 4 acetic, N-methyliminodiacetic, 4 no Y, isophthalic. The
ꢀ
release of Cl was generally exponential over time but the curve shapes were distinctly sigmoidal
for the more flexible diacids. Computational studies were undertaken to assess differences in
ꢀ
conformation. Overall, it appeared that Cl release for those SATs that could adopt a linear
conformation on the N-terminal side of proline correlated best with the polarity of the diacid.
were constructed of phospholipids or, indeed, if they were even
Introduction
bilayers. Nevertheless, our design began from the general
character of a modern phospholipid. We recognized that the
The efficiency and complexity of modern ion channel proteins
has led chemists to emulate the transport function with various
reaction of (C18
H
37
)
2
NH with the anhydride of diglycolic acid
should directly give (C18 NCO-
COOH. This single step reaction afforded twin
1
–6
model systems.
Synthetic models of cation-conducting
[O(CH
2
COOH)
2
]
37 2
H )
channels have been developed and studied in a number of
laboratories. Notable examples have been reported by
CH
2
OCH
2
alkyl chains akin to those present in many phospholipids.
Further, the pattern of polar oxygen atoms arising from
diglycolic acid corresponded well to that present in the mid-
polar regime. This single step reaction would, in principle,
afford two of the modules critical to emulate a phospholipid
structure.
7
Tabushi, Fyles, Kobuke, Ghadiri,
8
9
10
11
12
Voyer,
Koert,
1
3
Matile, and their co-workers. We have also reported exten-
1
4
sive studies in the cation channel area. The important efforts
of others are included in the reviews cited as refs. 1–6.
1
5
Just as structural studies of protein chloride transporters
1
6
lagged behind those of cation channels, model systems for
anion conductors have only recently been developed. Several
models have been reported including modified natural pep-
We devised a heptapeptide headgroup based on ClC chan-
nel protein sequences and the known GxxP hinge-bend
27,28
sequence.
Taken together, the simplest peptide sequence
-Pro-(Gly) B. The
heptapeptide was capped at its C-terminus with a benzyl ester.
2
The resulting structure was (C18 NCOCH OCH -
1
7
tides and the amphiphilic oligopeptide systems reported
that seemed to hold promise was B(Gly)
3
3
1
here. Additional anion transporters have been reported by
8
1
9
20
21
Davis, Smith, and Gin and their co-workers. There is
ambiguity in the mechanism(s) by which these model anion
transporters function as there is for the natural ClC family of
H
37
)
2
2
18
CO–(Gly)
3
-Pro-(Gly)
ꢀ
3
–OCH
2
Ph. We found that this com-
ꢀ
pound mediated Cl transport, it was selective for Cl over
+
K , and transport showed the open-close behavior typical of
2
2–26
‘
‘channel’’ proteins.
Our anion channel design began from the notion that the
ion channels.
earliest channels must have resembled the monomers of
membranes in which they functioned. It is, of course, un-
known whether or not the earliest encapsulating membranes
In subsequent studies, we explored the effect on ion trans-
29
port of changing the C-terminal ester, varying the alkyl chain
30
lengths, changing proline to pipecolic acid, and varying the
31
3
2
number of amino acids in the peptide. It was also found that
a
Departments of Molecular Biology & Pharmacology, Washington
University, St. Louis, MO 63130, USA
Department of Chemistry, Washington University, St. Louis, MO
63130, USA
Departments of Chemistry & Biochemistry and Biology, Center for
Nanoscience, University of Missouri—St. Louis, St. Louis, MO
33
these compounds exhibited biological activity. An aspect of
b
the system that remained unstudied was the subunit that
connects the dialkylamine residue to the heptapeptide. In a
phospholipid monomer, we have referred to the fragment
linking the fatty acids to the phosphate (i.e., the glyceryl unit)
as the ‘‘midpolar regime.’’ The goal of this study was to
c
63121, USA. E-mail: gokelg@umsl.edu; Fax: 314/516-5342;
Tel: 314/516-5321
1
960 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
evaluate changes in the structure of the midpolar regime
mimic and the effect of such alterations on ion transport
efficacy.
remaining ester-terminated peptide added on. Such a proce-
dure was used to construct the first example of this family and
3
1a
the approach has been described in detail.
In the present work, three coupling approaches have been
taken. The first parallels the initial synthetic approach, re-
Results and discussion
3
1a
ported in detail for 5.
In this case, diglycolic acid was
The compounds studied are variants of the general structure
converted to diglycolic anhydride by heating with acetic
anhydride. Diglycolic anhydride was then heated with diocta-
(
C H ) N–acid–CO–(Gly) -Pro-(Gly) –OCH Ph. A struc-
18 37 2 3 3 2
ture of the synthetic anion transporter (SAT) is shown. Its
essential elements are identified on the top structure. In order
to gain insight into the effect on transport of variations in the
midpolar regime mimic, compounds were prepared that varied
in flexibility, in chain length, in the presence or absence of
donor groups, and the presence of arenes. The structures used
in this study are shown as 1–9. Compound 1 is a monoacid
that reacts to incorporate an acetic acid amide (glycine) unit
into the SAT. Compounds 2–8 are diacids, ultimately incor-
porating two amide linkages.
decylamine to form the half-amide acid (C18
CH OCH COOH, precursor to 5. An essentially identical
approach was used to form (C18 NCOCH CH COOH,
(C18 NCOCH CH CH COOH, NCOCH
)CH COOH and (C18 NCOCH COOH, the
precursors to 3, 4, 6 and 7, respectively.
37 2
H ) NCO-
2
2
H
37
)
2
2
2
H
37
)
2
2
2
2
(C18
H
37
)
2
2
-
N(CH
3
2
H
37
)
2
2 2
SCH
The dialkylamido acid module required for 2 was prepared
by first converting bromoacetic acid into its ethyl ester.
Heating BrCH
COOEt with (C18
H
37
)
NH and Na
2
CO in
2 3
2
CH CH CH CN gave (C18
2
3
3
H
37
)
NCH COOEt in 95%
2 2
yield. The precursors to 8 and 9, which have aromatic
midpolar mimics, were prepared by treating dioctadecylamine
with an excess of either isophthaloyl chloride (-8) or
terephthaloyl chloride (-9). The approach taken in the latter
Syntheses
Conceptually, the preparation of the various (C18
37 2
H ) -
N–Y–CO–(Gly) -Pro-(Gly) –OCH Ph molecules requires
2
3
3
assembly of three modules. The midpolar element needs to
be connected to the hydrocarbon chains comprising the
N-terminus and originating from dioctadecylamine. The hep-
tapeptide must be assembled and esterified at its C-terminal
end. This peptide module must then be connected to the
37 2
two cases afforded (C18H ) NCO–Ar–COOH in 41 and 37%
yields, respectively. The lower yields obtained in these con-
densations were compensated by the ready availability of the
starting materials and the ease of purification.
We note that our efforts to prepare the corresponding
derivatives of maleic and butynedioic acids failed. In the
former case, a product was obtained but no double bond
R
2
N–Y–COOH fragment. In practice, the synthesis requires
a peptide fragment to be connected to R N–Y–COOH and the
2
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1960–1972 | 1961

View Article Online
could be detected by NMR suggesting a possible Michael
addition. Only a tarry residue was obtained from reactions
using butynedioic acid as the starting material.
and data recording commenced. The vesicles were lysed after
1 h by addition of Triton X-100. The values for ion release
were then normalized to the value determined after lysis. The
ꢀ
Once the dialkylamido acids were in hand, construction of
3
graphs shown below plot the fraction of Cl released over
0b
the heptapeptide proceeded as previously described.
In
time, typically 1500 s. The reproducibility of these determina-
tions is confirmed by the fact that each release curve shown is
the average of at least three independent experiments.
short, the acid was coupled to triglycine, which was then
linked to proline. The three additional glycines were then
coupled to the existing unit to afford the entire structure:
Chloride ion release mediated by diglycolic acid derivative 5
18 37 2 3 3 2
(C H ) NCO–Y–CO–(Gly) -Pro-(Gly) –OCH Ph. Details
of the syntheses are reported in the Experimental section.
The SATs were typically waxy, white solids that were poorly
soluble in water. Attempts to crystallize them have thus far
failed to produce samples suitable for X-ray analysis.
ꢀ
In previously reported studies, we found that release of Cl
from liposomes was mediated in a concentration dependent
fashion by (C H ) NCOCH OCH CO–(Gly) -Pro-(Gly) –
1
8
37 2
2
2
3
3
2
OCH Ph. Replacing proline with leucine changed the fourth
amino acid from cyclic to acyclic while changing the
carbon atom number by only one. When the structure
The process of ion release
37 2 2 2 3
was thus changed to (C18H ) N–COCH OCH CO–(Gly) -
The bilayer membranes of liposomes prepared from phospho-
+
ꢀ
Leu-(Gly)
3
–OCH
2
Ph, chloride ion transport was dramatically
lipid monomers do not pass ions such as Na or Cl .
Compounds 1–9 are amphiphilic peptides that must insert
into the liposomal bilayer and form a stable pathway through
the membrane. We surmise that the peptides in aqueous
suspension associate with the outer leaflet of the liposomal
membrane and then the twin hydrocarbon tails insert into it.
The various amphiphilic peptide monomers present in the
outer leaflet diffuse laterally, presumably in a random pattern.
When two (or more) of the monomers come into proximity,
they form an opening through which water and ions may pass.
We surmise that monomer headgroups in the lower leaflet of
the bilayer reorganize to form the lower part of the conduction
pathway. There is precedence for this and transverse relaxa-
tion (‘‘flip-flop’’) of the amphiphilic peptide is expected to be
slow.
ꢀ
reduced. Significant differences in Cl transport were also
apparent when proline was changed to pipecolic acid, its six-
membered ring homolog, although every other feature re-
3
1a,34
mained identical.
In the studies below, structures 3–9
may all be described by the semi-schematic formula
(
C H ) N–diacid–(Gly) -Pro-(Gly) –OCH Ph. Compound
18 37 2 3 3 2
2
is (C18
H
37
)
2
N–(Gly)
-Pro-(Gly)
and 2 would serve as controls or benchmarks for the
4
-Pro-(Gly)
3
–OCH
2
Ph and
1
is
(
C
18
H
37
)
2
N–(Gly)
3
3
–OCH
2
Ph. It was intended that
1
remaining compounds, whose structures were systematically
varied.
Overview of transport efficacy
Data for the release of chloride ion from vesicles by closely
related structures are illustrated in several graphs in the
sections below. Because different compounds give different
Once the conducting pore is formed, the vesicle is expected
to empty its contents quite rapidly. The open-close behaviors
of certain amphiphilic peptides of this type have been recorded
ꢀ
Cl release profiles, we have compared 1–9 in the table below.
1
8
and reported previously. The bilayer voltage clamp experi-
ment is an excellent analytical technique but it requires both
special equipment and expertise. In the present case, we have
chosen to use ion selective electrodes (ISEs) to record the
release of ions from a collection of vesicles. The method
comprises a reproducible assay of overall efficacy so that
comparisons among different compounds can be made on a
quantitative basis.
The release values presented are for 1500 s, an arbitrarily
selected time point. Such data are useful to gage the overall
effect of midpolar regime changes on transport. The values
shown were obtained by averaging the release values observed
over a 20 s interval (1490–1510 s). Data points were collected
each second during the experiment.
Replacement of ether oxygen in 5 by NMe (6) or S (7)
ꢀ
Fig. 1 shows the fractional release of Cl mediated by
Analysis of ion release
(
C
18
H
37
)
2
N–Y–(Gly)
3 3 2
Pro(Gly) –OCH Ph structures in which
The compounds studied for the present report were designed
ꢀ
Y comprises BCO(CH ) COB (4), O(CH COB) (5),
2
3
2
2
to be Cl transporters. In order to test their efficacy in that
CH N(CH COB) (6), or S(CH COB) (7). The release
3
2
2
2
2
role, we developed a technique involving a chloride selective
traces for the heteroatom containing diacids (5–7) generally
1
8
electrode. Thus, vesicles were prepared from a 7 : 3 w/w (2 : 1
show similar exponential shapes while the ion release curve for
ꢀ
mol/mol) mixture of 1,2-dioleoyl-sn-glycero-3-phosphocholine
glutaric acid derivative 4 is distinctly sigmoidal. When Cl
(
DOPC) and 1,2-dioleoyl-sn-glycero-3-phosphate (DOPA).
transport is compared at 1500 s, the fraction of entrapped ion
released in each case is as follows: 4, 0.53; 5, 0.30; 6, 0.33; and
7, 0.49.
The vesicles were prepared in a solution containing 600 mM
KCl and HEPES buffer (pH 7). After brief sonication, the
suspension was passed through a 200 nm filter membrane and
a Sephadex G25 column. The resulting vesicles were found, by
using laser light scattering, to be B200 nm.
An Accumet chloride combination electrode was introduced
into a vesicle suspension (2 mL) and allowed to equilibrate for
B5 min. The ionophore was added (B9 mM in 2-propanol)
ꢀ
The initial Cl release rate is presumably related to the
kinetics of insertion and pore formation. If the nitrogen atom
is protonated at physiological pH, it would be attracted to the
vesicle surface, which incorporates a significant concentration
of phosphoric acid headgroups. We were unable to find a
literature pK value for the nitrogen of N-methyldiglycine. We
A
1
962 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
Fig. 1 Chloride anion release from 7 : 3 w/w DOPC/DOPA vesicles
mediated by 4, 5, 6 and 7. Ion release was detected by ion selective
electrode methods (see text for details).
Fig. 2 Chloride anion release from 7 : 3 w/w DOPC/DOPA vesicles
mediated by 1, 2, 3 and 4. Ion release was detected by ion selective
electrode methods (see text for details).
therefore estimated the nitrogen basicity from the pK
A
of
Two points seem apparent
trimethylamine: 9.80. A nitrogen having this pK will be fully
A
Compound 1, which lacks a midpolar regime mimic, shows
poorer transport than do 2–7, which incorporate such a
module. This observation supports the initial design criteria.
Second, even a simple acetic acid residue (2) is superior to no
midpolar mimic, although it is inferior to the original design
protonated at pH 7.4 or lower. At a minimum, the nitrogen of
N-methyldiglycine in the SAT must be substantially
protonated.
ꢀ
Diminished Cl transport was observed with 5 and 6,
which are structurally similar to, but more polar than, 4 and
(5) and most of the variants explored here. An additional
7. The similarity in curve shape suggests that these
conclusion that can be drawn is that although there is little
three compounds can be reasonably compared. Divalent sulfur
is somewhat, but not dramatically, larger than either
O or N. The difference in electronegativity is in the order O
difference in ultimate transport by the succinic (3) and glutaric
(4) acid derivatives, it seems that an efficacious conductance
pathway is achieved more rapidly in less flexible 3 than in 4.
4
3
N 4 S E C. The Pauling electronegativity values are
.5, 3.0 and 2.5, respectively. These observations suggest
Aromatic diacids
that a lower polarity midpolar regime may be more efficacious
in this structural framework. A lower polarity conduction path
is expected to be more favorable than one in which inter-
actions can take place between it and the fluxional ion.
Repulsion would obviously be unfavorable but so would a
strongly stabilizing contact that favored complexation over
transport.
Isophthalic (8) and terephthalic (9) acids differ only in the
ꢀ
relative positions of their carboxylate groups. Fractional Cl
release mediated by 1, 8 and 9 is compared in Fig. 3.
Compound 8, in which isophthalic acid comprises the mid-
ꢀ
polar element, is a very poor Cl transporter. Its isomer, 9,
which incorporates terephthalic acid, is similar in transport
efficacy to the derivatives containing succinic (3), glutaric (4)
and 3-thiaglutaric acids (7).
The curve shape is significantly different for glutaric acid
derivative 4 than for the other ionophores in this series. Initial
insertion leads to ion transport by 4 that is as rapid as that
observed for diglycolic acid derivative 5. It may be that many
of the 4 monomers require a conformational adjustment
before they can insert and function effectively. The three
adjacent methylene units of glutaric acid present a more
complex conformational situation than the two oxygen-sepa-
rated methylenes of diglycolic acid in 5.
Terephthalic acid is not a flexible residue but 9 shows a
ꢀ
slightly sigmoidal Cl release curve. This suggests that it
experiences some of the same organizational issues apparent
in the release curves for the succinic (3) and glutaric (4) acid
derivatives. Such a profile was not apparent in the ionophores
incorporating heteroatom-containing diacids, i.e., 5–7. Since
the origin of the sigmoidal shape is not known, any compar-
ison is speculative. Still, it appears that greater conformational
variability correlates to a delayed onset of transport activity.
For this reason, conformational minima were calculated for
each of the diacids as discussed below.
Transport by succinic and glutaric acid derivatives
The graph of Fig. 2 compares compounds 1–4. In 1, there is no
midpolar regime mimic. The twin octadecyl chains are
attached directly to the N-terminal glycine. In 2, the peptide
Acid conformations
sequence is (Gly)
4
-Pro-(Gly)
3
. The N-terminal glycine may be
The Gaussian 03 program suite was used to calculate mini-
mum energy conformations for the acid components of 2–9.
Dimethylamino, rather than dioctadecylamino, comprised the
N-terminal amide. The C-terminal carboxyl was considered
as an N-methylamide, to represent the peptide chain’s
considered an acetic acid midpolar regime mimic in this case,
making the sequence R N–CH CO–(Gly) -Pro-(Gly) –
2
2
3
3
OCH Ph. In 3 and 4, the midpolar regime mimics are succinic
2
and glutaric acids, respectively.
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1960–1972 | 1963

View Article Online
indicated by an arrow and the number of the compound that
incorporates the particular diacid.
The fourth line of Fig. 4 shows the calculated overlays of
compounds 3 and 4 (lower left panel), 5–7 (lower center panel)
and 8 and 9 (lower right panel). Differences in the orientations
of the carboxylic acid residues are clearly discernible in the
comparisons of succinic (3) vs. glutaric (4) acids and in the
meta- (8) and para-substitution (9) patterns of dicarboxyben-
zene. When isophthalic acid is the midpolar element, the SAT
ꢀ
shows about half the Cl release activity of the compound
incorporating terephthalic acid.
The results obtained for (CH
3
)
2
NCO–CH
2
–Het–CH
2
–
CO–NHCH , where Het is O (5), S (7) or NCH
3
3
(6) vary
more than expected. Of course, the more acute angles of the
C–S–C bond were anticipated, but the large conformational
differences shown in Fig. 4 were not. Admittedly, these
systems may be too truncated to adequately reflect the am-
phiphile’s conformation. An additional attempt was made to
gain insight into this question. Thus, the calculations were
Fig. 3 Chloride anion release from 7 : 3 w/w DOPC/DOPA vesicles
mediated by 1, 8 and 9. Ion release was detected by ion selective
electrode methods (see text for details).
N-terminus. Calculations that included the 36 additional
carbons of the two octadecyl chains would have required an
inordinate amount of computation. Similar reasoning led to
replacement of the heptapeptide by a simple methylamide
3 2 3 3
repeated for the (CH ) NCO–Y–CO–(Gly) –OCH analogs of
3, 4, 5, 8 and 9 (data not shown). In all cases, the diacid
conformations were similar or identical to those shown in
Fig. 4 and the tripeptide was linear, extended, and essentially
invariant in each calculation.
residue. Thus, each (C18
Gly) –OCH Ph was reduced to (CH
for the purpose of comparison. The calculated
CH ) NCO–Y–CO–NHCH structures are shown on the first
H
37
)
2
NCO–Y–CO–(Gly)
3
-Pro-
(
3
2
3
)
2
NCO–Y–CO–NHCH
3
Overall comparisons
(
Table 1 (above) shows that compounds 1–9 mediate the release
ꢀ
3
2
3
three lines of Fig. 4. The ultimate product in each case is
of Cl to extents ranging from 21 to 53% of the total enclosed
Fig. 4 Calculated (Gaussian 03) gas phase conformations for acids or diacids incorporated into 2–9 (first three lines). Bottom line:
superimposition of (left) calculated structures of precursors to 3 and 4; (middle) 5, 6 and 7; and (bottom) 8 and 9 (see text for details).
1
964 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
within phospholipid vesicles. The two poorest midpolar ele-
ments are none (1) and isophthalic acid (8). It seems likely that
the N-terminal glycine in 1 plays the role of a midpolar
element but this reduces the peptide component from G PG
tographic grade, AX 611) or with Kieselgel 60 (70–230 mesh).
Chromatotron chromatography was performed on a Harrison
Research Model 7924 Chromatotron with 2 mm thick circular
plates prepared from Kieselgel 60 PF-254.
3
3
to G PG . We have previously found that smaller peptides are
3
All reactions were conducted under dry N unless otherwise
2
2
ꢀ
32
less efficacious Cl transporters. Compound 9 possesses a
midpolar element but its poor transport efficacy is likely due to
the orientation of its carboxyl groups. Not only are they held
rigidly at a 1201 angle, but if the two amide carbonyls exist in
the anti conformation (see Fig. 4), the alkyl chains and peptide
are oriented nearly at a right angle to each other.
stated. All reagents were the best (non-LC) grade commer-
cially available and were distilled, recrystallized, or used with-
out further purification, as appropriate. Molecular distillation
temperatures refer to the oven temperature of a Kugelrohr
apparatus. Combustion analyses were performed by Atlantic
Microlab, Inc., Atlanta, GA, and are reported as percentages.
Where water is factored into the analytical data, spectral
evidence is presented for its presence. Samples suitable for
single crystal X-ray analysis were grown by slow cooling of a
warm saturated solution, by evaporation of a saturated solu-
tion, or by vapor diffusion.
There is little structural difference expected in the closely
related SATs 5–7. Despite the more acute C–S–C bond angle
in 7, Fig. 4 shows that the heteroatoms of 5, 6 and 7 are nearly
overlapped. Since these molecules are flexible, it appears that
the greater polarity of oxygen or methylamino is the key factor
in their diminished transport efficacy. Glutaric acid derivative
Preparation of BocPGGG–OCH Ph
2
4
replaces the heteroatoms of 5–7 by a methylene. This
increases the number of possible conformations for the mole-
cule and likely contributes to the oddly shaped ion release
curve (see Fig. 1 and 2). Still, the polarity is low and transport
by both 4 and 7 is identical within experimental error at 1500 s.
The transport efficacy of 4 (and 3) is consistent with the low
polarity hypothesis.
2
Boc-L-Proline (1.43 g, 6.7 mmol), TsOH–GGG–O–CH Ph
(
3.0 g, 6.7 mmol), and Et
3
N (2.80 mL) were dissolved in
CH Cl (40 mL) and cooled to 5 1C. EDCI (1.34 g, 7 mmol)
2
2
was added and the reaction was stirred at rt for 3 days. Solvent
was evaporated and the residue was dissolved in EtOAc (50
mL) and washed with aq. NH
dried over MgSO and evaporated. The crude, oily product
was chromatographed (SiO , 5% MeOH–CH Cl ) and af-
forded colorless crystals (2.25 g, 71%, mp 54–55 1C).
NMR: 1.42 (9H, s, C(CH ), 1.80–2.20 (4H, m, Pro
NCH CH CH ), 3.35–3.55 (2H, m, Pro NCH CH CH ),
.85–4.20 (7H, m, Gly NCH , Pro NCH), 5.15 (2H, s,
PhCH O), 7.05 (2H, br s, Gly CONH), 7.30–7.35 (5H, m,
HAr), 7.80 (1H, br s, Gly CONH). C NMR: 24.6, 28.3, 29.4,
4
Cl (25 mL) and brine (25 mL),
4
Experimental
2
2
2
1
H
General
3 3
)
1
3
H NMR spectra were recorded at 300 MHz in CDCl solvent
2
2
2
2
2
2
3
2
and are reported in ppm (d) downfield from internal (CH ) Si.
3
4
1
3
C NMR spectra were recorded at 75 MHz in CDCl unless
3
2
1
3
otherwise stated. Infrared spectra were recorded on a Perkin-
Elmer 1710 Fourier Transform infrared spectrophotometer
4
1
2
1
1.1, 43.0, 43.3, 47.2, 60.7, 66.9, 80.9, 128.4, 128.5, 128.7,
35.4, 155.8, 169.6, 170.0, 173.9. IR (KBr): 3310, 3066, 2976,
933, 1753, 1667, 1540, 1455, 1408, 1366, 1245, 1174, 1129,
ꢀ
1
and were calibrated against the 1601 cm band of polystyr-
ene. Optical rotations were measured on a Perkin-Elmer
Model 241 Polarimeter in a glass microcell (100 mm path
length, 1 mL volume) with a Na gas discharge lamp as the light
source. Melting points were determined on a Thomas Hoover
apparatus in open capillaries and are uncorrected. Thin layer
chromatographic (TLC) analyses were performed on alumi-
num oxide 60 F-254 neutral (Type E) with a 0.2 mm layer
thickness or on silica gel 60 F-254 with a 0.2 mm layer
thickness. Preparative chromatography columns were packed
with activated aluminium oxide (MCB 80–325 mesh, chroma-
ꢀ
1
031, 974, 912, 773, 739, 698 cm . Anal. Calc. for
: C, 57.97; H, 6.77; N, 11.76%. Found: C,
7.87; H, 6.76; N, 11.39%.
23 32 4 7
C H N O
5
.
2
Preparation of PGGG–OCH Ph HCl
Boc–PGGG–OCH Ph (0.2 g, 0.42 mmol) was dissolved in 4 M
2
HCl in dioxane (10 mL) at 5 1C and the reaction mixture was
stirred for 1 h. The solvent was evaporated in vacuo and the
2
residue was crystallized from MeOH–Et O (0.18 g, 100%) to
1
give a colorless solid (mp 145–146 1C). H NMR (CD
.00–2.25 (4H, m, Pro NCH CH CH ), 3.35–3.45 (2H, m, Pro
), 3.90–4.05 (6H, m, Gly NCH ), 4.30–4.40
O), 7.30–7.40 (5H,
3
OD):
Table 1 Comparison of chloride release from phospholipid vesicles
mediated by 1–9
2
2
2
2
NCH
2
CH
2
CH
2
2
a
b
(
1H, m, Pro NCH), 5.18 (2H, s, PhCH
2
No.
Midpolar mimic
Release
1
3
m, HAr). C NMR: 25.2, 30.9, 42.1, 43.3, 43.7, 47.6, 61.4,
1
2
3
4
5
6
7
8
9
a
None (G PG
3
3
)
0.21
0.33
0.53
0.53
0.30
0.33
0.49
0.22
0.44
6
8.1, 129.5, 129.6, 129.9, 137.5, 170.9, 171.4, 171.8, 172.4.
2
4
3
CH CO (G PG )
Succinic
Glutaric
Diglycolic
N-Methyliminodiacetic
3-Thiaglutaric
Isophthalic
Compound 1: (2-{2-[(1-{2-[2-(2-dioctadecylamino-acetylamino)-
acetylamino]acetyl}pyrrolidine-2-carbonyl)amino]acetylamino}-
acetylamino)acetic acid benzyl ester
1
2
8 NCH
2 2 2 2
CO–GG–OCH Ph. To 18 NCH COOH (0.39 g,
Terephthalic
0
.63 mmol) dissolved in CH Cl (20 mL), PyBroP (0.34 g, 0.73
2
2
b
ꢀ
Acid connecting dialkylamino and peptide chains. Fractional Cl
release at 1500 s.
mmol) and HOBt (0.1 g, 0.74 mmol) were added and the
mixture was stirred at rt. After 0.5 h, TsOH–GGG–OCH Ph
2
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1960–1972 | 1965

View Article Online
1
3
(
0.27 g, 0.68 mmol) and DIEA (0.27 mL) were added and the
(2H, q, J = 7.2 Hz, OCH
2 3
CH ). C NMR: 14.1, 14.3, 22.7,
reaction mixture was stirred at rt for 2 days. The solvent was
27.4, 27.5, 29.3, 29.5, 29.6, 29.7, 31.9, 54.5, 55.3, 60.2, 171.7.
evaporated and the residue chromatographed (SiO , 2%
2
1
Dioctadecylaminoacetic acid (C H ) NCH COOH.
1
A
8
37 2
2
MeOH–CHCl ) to give a colorless oil (0.44 g, 89%).
3
H
solution of (C18
H
37
)
2
NCH
2
COOCH
2
CH
3
(2.0 g, 3.3 mmol)
NMR: 0.88 (6H, t,
J
=
6.9 Hz, ꢀCH CH ), 1.25
2
3
in EtOH (30 mL) containing 2 N NaOH (20 mL) was refluxed
overnight. After cooling, the pH was adjusted to 6 and solids
were removed by filtration. The solid material was dissolved in
CHCl3 (50 mL), filtered, and the solvent evaporated. The
(
60H, pseudo-s, CH (CH ) CH CH N), 1.43 (2H, br s, CH -
3
3
2 15
2
2
(
CH
2
)
15CH
2
CH
.48 (4H, br s, CH
.98 (2H, d, J = 5.7 Hz, Gly NCH
), 5.18 (2H, s, PhCH
CONH), 7.30–7.40 (5H, m, HAr), 7.96 (1H, br s, Gly CONH).
2
N), 1.64 (2H, br s, CH
(CH N), 3.10 (2H, br s, NCH
), 4.07 (2H, d, J = 5.7 Hz,
O), 6.66 (1H, br s, Gly
3
(CH
2
)
15CH
2
CH
2
N),
2
3
3
2
)
16CH
2
2
CO),
2
product was dried (high vacuum) to give a white solid (1.9 g,
1
Gly NCH
2
2
1
00%), mp 102–103 1C. H NMR: 0.88 (6H, t, J = 6.6 Hz,
CH (CH ) CH CH N), 1.25 (60H, pseudo-s, CH (CH ) -
1
3
3
2 15
2
2
3
2 15
C NMR: 14.1, 22.7, 27.4, 29.3, 29.5, 29.7, 31.9, 41.3, 43.1,
5.5, 67.2, 128.4, 128.6, 128.7, 135.4, 169.2, 169.4, 173.5.
CH
2
CH
2
N), 1.70 (4H, br s, CH
(CH
CO), 9.95 (1H, br s, COOH). C NMR: 14.1, 22.7,
4.0, 26.8, 29.2, 29.3, 29.4, 29.5, 29.7, 31.9, 54.2, 55.2, 167.6.
3
(CH
2
)
15CH
2
CH
2
N), 3.13
5
(
4H, t, J = 7.8 Hz, CH
3
2
)
15CH
2
3
CH
2
N), 3.58 (2H, s,
1
NCH
2
1
8 NCH CO–GGPGGG–OCH Ph,
2
1.
18 NCH CO–
2 2
2
2
2
GG–OCH Ph (0.40 g, 0.51 mmol) was dissolved in EtOH
2
(
30 mL), 10% Pd/C (0.14 g) was added, and this mixture was
1
8
2
NCH
mmol), EDCI (0.14 g, 0.72 mmol), and HOBt (0.1 g, 0.74
mmol) were dissolved in CH Cl (20 mL) and stirred at rt.
2 2 2 2
CO–GGG–OCH Ph. 18 NCH COOH (0.4 g, 0.65
shaken (60 psi H ) for 3 h. The mixture was heated to reflux,
2
filtered (Celite pad), and the solvent was evaporated to give a
white solid (0.35 g, 100%). This product was used in the
2
2
After 0.5 h, TsOH–GGG–OCH Ph (0.3 g, 0.65 mmol) and
2
next step without further purification. 18
0.35 g, 0.5 mmol) was suspended in CH
EDCI (0.11 g, 0.57 mmol), PGGG–OCH
.52 mmol), HOBt (0.08 g, 0.57 mmol) and DIEA (0.1 mL) were
2
NCH
Cl (30 mL) at 0 1C.
Ph ꢂ HCl (0.22 g,
2
CO–GG–OH
DIEA (0.26 mL) were added and the reaction mixture was
stirred at rt for 2 days. The solvent was evaporated and the
residue chromatographed (SiO , 2% MeOH–CHCl ) to give a
(
2
2
2
2
3
0
1
colorless oil (0.42 g, 76%). H NMR: 0.88 (6H, t, J = 6.9 Hz,
CH CH ), 1.25 (60H, pseudo-s, CH (CH CH N), 1.45
4H, br s, CH (CH CH N), 2.53 (4H, br s,
CH (CH N), 3.16 (2H, br s, NCH CO), 3.95 (2H, d,
J = 5.7 Hz, Gly NCH ), 3.99 (2H, d, J = 5.7 Hz, Gly NCH ),
.06 (2H, d, J = 5.7 Hz, Gly NCH ), 5.16 (2H, s, PhCH O),
.96 (1H, br s, Gly CONH), 7.05 (1H, br s, Gly CONH),
added and the reaction was stirred for 48 h at rt. The solvent was
2
3
3
2
)
15CH
2
2
evaporated, the residue was chromatographed (SiO , 2%
2
(
3
2
)
15CH
2
2
MeOH–CHCl ), and crystallized (MeOH) to give a white solid
3
3
2
)
16CH
2
2
1
(
0.29 g, 51%), mp 116–117 1C. H NMR: 0.87 (6H, t, J = 6.3
2
2
Hz, CH (CH ) CH CH N), 1.25 (60H, m, CH (CH ) -
3
2 15
2
2
3
2 15
4
6
7
2
2
CH
2H, br s, CH
NCH CH CH ), 2.47 (4H, br s, CH
2H, br s, NCH CO), 3.30–3.55 (2H, m, Pro NCH
.60–4.15 (10H, overlapping signals due to Gly CH
Ph), 7.30–7.35
5H, m, HAr), 7.60 (1H, br s, NH), 7.81 (2H, br s, NH), 8.00
2
CH
2
N), 1.46 (2H, br s, CH
(CH CH N), 1.80–2.20 (4H, m, Pro
(CH N), 3.07
CH CH ),
), 4.35
3 2 2 2
(CH )15CH CH N), 1.65
(
3
2
)
15CH
2
2
1
3
.30–7.35 (5H, m, HAr), 8.05 (1H, br s, Gly CONH).
C
2
2
2
3
2 2
)16CH
NMR: 14.1, 22.7, 27.3, 29.4, 29.5, 29.6, 31.9, 41.2, 41.8, 43.1,
5.5, 67.2, 128.4, 128.6, 128.7, 135.4, 169.2, 169.4, 173.5.
(
2
2
2
2
5
3
2
(
(
(
1H, t, J = 6.3 Hz, Pro CH), 5.15 (2H, m, OCH
2
1
8 NCH CO–GGGPGGG–OCH Ph.
2
18 NCH CO–
2 2
2
2
GGG–OCH Ph (0.4 g, 0.47 mmol) and 10% Pd/C (0.13 g)
2
1
3
1H, br s, NH). C NMR: 14.0, 22.7, 25.1, 26.9, 27.4, 28.9,
were dissolved in EtOH (30 mL) shaken (60 psi H ) for 3 h.
2
2
5
1
1
9.3, 29.5, 29.7, 31.9, 41.3, 42.1, 42.6, 42.8, 43.6, 46.9, 55.2,
8.3, 61.1, 67.1, 128.2, 128.4, 128.6, 135.3, 168.7, 169.4, 169.8,
The mixture was heated to reflux, filtered (Celite pad), and the
solvent evaporated to give a white solid (0.25 g, 72%). This
product was used in the next step without purification.
70.0, 172.9, 173.7. Anal. Calc. for C H N O : C 68.47, H
6
0
105
7
8
0.06, N 9.32%. Found: C 67.10, H 9.53, N 9.59%.
1
8
2
NCH
in CH Cl
PGGG–OCH
.36 mmol) and DIEA (0.06 mL) were added and the mixture
2
CO–GGG–OH (0.25 g, 0.33 mmol) was suspended
(30 mL) at 0 1C. EDCI (0.07 g, 0.36 mmol),
Ph ꢂ HCl (0.14 g, 0.33 mmol), HOBt (0.05 g,
2
2
Compound 2: [2-(2-{[1-(2-{2-[2-(2-dioctadecylaminoacetylamino)-
acetylamino]acetylamino}acetyl)pyrrolidine-2-carbonyl]amino}
acetylamino)acetylamino]acetic acid benzyl ester
2
0
was stirred for 48 h at rt. The solvent was evaporated, the
residue was chromatographed (SiO , 5% MeOH–CHCl ), and
Dioctadecylaminoacetic acid ethyl ester (C H ) -
1
8
37 2
2
3
NCH COOCH CH . A solution of dioctadecylamine (5.20 g,
2
then crystallized from MeOH to give a white solid (0.29 g,
1
2
3
1
0.0 mmol), ethyl bromoacetate (1.43 g, 8.5 mmol), Na CO
2
78%), mp 137–138 1C. H NMR: 0.87 (6H, t, J = 6.3 Hz,
CH (CH ) CH CH N), 1.25 (60H, m, CH (CH ) -
3
(
1.6 g, 10.0 mmol), and KI (30 mg, 0.2 mmol) in n-PrCN (75
3
2 15
2
2
3
2 15
mL) was heated at reflux for 2 h. The mixture was cooled and
filtered to remove salts. This insoluble material was washed
CH
2
CH
2
N), 1.50 (2H, br s, CH
3
(CH
), 2.47 (4H, br s,
CO), 3.30–3.55
), 3.60–4.15 (12H, overlapping
), 4.32 (1H, t, J = 6.3 Hz, Pro CH),
Ph), 7.30–7.35 (5H, m, HAr), 7.42 (1H, br s,
NH), 7.51 (2H, br s, NH), 7.80 (1H, br s, HAr), 7.94 (1H, br s,
2 2 2
)15CH CH N),
1.80–2.20 (4H, m, Pro NCH
CH (CH
(2H, m, Pro NCH
signals due to Gly CH
5.14 (2H, m, OCH
2
CH CH
2
2
with CHCl
washed with brine (50 mL) and concentrated in vacuo. Chro-
matography (SiO , 10% hexanes in CHCl then 100% CHCl
gave 4.94 g (95%) of a slightly yellow oil. H NMR: 0.87 (6H,
t, J = 6.6 Hz, CH (CH CH N), 1.26 (63H, overlap-
ping signals due to CH (CH ) CH CH N and OCH CH ),
3
(2 ꢃ 20 mL) and the combined organic layer was
3
2
)
16CH
2
N), 3.07 (2H, br s, NCH
CH CH
2
2
2
2
2
3
1
3
)
2
2
3
2
)
15CH
2
2
1
3
NH), 8.00 (1H, br s, NH). C NMR: 14.1, 22.7, 25.1, 26.9,
27.4, 29.0, 29.3, 29.5, 29.7, 31.9, 41.3, 41.8, 42.7, 43.5, 46.9,
55.1, 58.4, 61.3, 67.1, 128.2, 128.4, 128.6, 135.3, 168.7, 169.6,
3
2 15
2
2
2
3
1
.43 (4H, br s, CH (CH ) CH CH N), 2.54 (4H, t, J = 6.6
3 2 15 2 2
Hz, CH (CH ) CH CH N), 3.30 (2H, s, NCH CO), 4.16
3
2 15
2
2
2
1
966 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
1
70.0, 170.1, 170.3, 170.1, 173.3. Anal. Calc. for
10: C 67.11, H 9.81, N 10.10%. Found: C 67.13,
NMR (CD
3
OD): 0.90 (6H, t, J = 6.9 Hz, CH
(CH
3
(CH
CH
2
15
) -
C
62
H
108
8
N O
CH CH N), 1.29 (60H, pseudo-s, CH
2
2
3
2
)
15CH
2
2
N),
H 9.79, N 10.30%.
1.40–1.60 (4H, br s, CH (CH ) CH CH N), 2.52 (2H, t, J
3 2 15 2 2
=
6.3 Hz, COCH ), 2.74 (2H, t, J = 6.3 Hz, COCH ),
2 2
Compound 3: [2-(2-{[1-(2-{2-[2-(3-dioctadecylcarbamoylpropionyl-
amino)acetylamino]acetylamino}acetyl)pyrrolidine-2-carbonyl]-
amino}acetylamino)acetylamino]acetic acid benzyl ester, 3
3
.80–3.90 (6H, overlapping signals due to Gly NCH2).
1
2
8 COCH
2
CH
2
CO–GGGPGGG–OCH
CO–GGG–OH (0.66 g, 0.83 mmol) was sus-
Cl (30 mL) at 0 1C. EDCI (0.17 g, 0.91 mmol),
Ph ꢂ HCl (0.34 g, 0.83 mmol), HOBt (0.12 g,
.91 mmol) and Et N (0.16 mL) were added and the mixture
was stirred for 48 h at rt. The solvent was evaporated, the
residue was chromatographed (SiO , 10–20% MeOH–CHCl ),
and then crystallized from MeOH to give a white solid (0.57 g,
2
Ph.
1
8
2
COCH
CH
2 2
1
8
2
2 2
COCH CH COOH. A solution of dioctadecylamine
pended in CH
PGGG–OCH
2
2
(
4.0 g, 7.6 mmol) and succinic anhydride (0.76 g, 7.6 mmol)
was refluxed in toluene (100 mL) for 48 h. The solvent was
2
0
3
evaporated and the residue was crystallized from Et O to give
2
1
the product as a white solid (3.88 g, 82%), mp 63–64 1C. H
NMR: 0.88 (6H, t, J = 6.6 Hz, CH (CH ) CH CH N), 1.25
2
3
3
2 15
2
2
(
60H, pseudo-s, CH
CH (CH CH
J = 7.8 Hz, CH (CH
CH (CH N). C NMR: 14.1, 22.7, 26.9, 27.0, 27.6,
8.0, 28.8, 29.3, 29.7, 30.5, 31.9, 46.7, 48.3, 172.3, 174.8.
3
(CH
2
)
15CH
2
CH
2
N), 1.54 (4H, br s,
), 3.22 (2H, t,
N), 3.31 (2H, t, J = 7.8 Hz,
1
82%), mp 150–152 1C. H NMR: 0.87 (6H, t, J = 6.3 Hz,
CH (CH ) CH CH N), 1.25 (60H, m, CH (CH ) -
3
2
)
15CH
2
2 2
N), 2.68 (4H, s, COCH
3
2
)16CH
2
3
2 15
2
2
3
2 15
1
3
CH CH N), 1.42 (2H, br s, CH (CH ) CH CH N), 1.55
2
3
2
)
16CH
2
2
2
3
2 15
2
(
2H, br s, CH
NCH CH CH ), 2.40–2.80 (4H, m, COCH
(4H, br s, CH (CH ) CH N), 3.25–4.40 (15H, overlapping
3
(CH
2
2
)15CH CH
2
N), 1.80–2.40 (4H, m, Pro
2
2
2
2
2
CH CO), 3.20
2
TsOH–GGG–OCH Ph. GGG (3.0 g, 15.9 mmol) and p-
2
3
2 16
2
toluenesulfonic acid monohydrate (3.6 g, 18.9 mmol) were
added to a mixture of benzyl alcohol (20 mL) and toluene (30
mL). The mixture was dehydrated by heating under reflux with
a Dean–Stark trap (B8 h). The mixture was cooled to rt,
signals due to Pro CH and NCH
.15 (2H, m, OCH Ph), 7.30–7.40 (5H, m, HAr), 7.46 (2H, br s,
NH), 7.73 (1H, br s, NH), 7.90 (1H, br s, HAr), 8.14 (2H, br s,
2 2 2 2
CH CH , and Gly CH ),
5
2
1
3
NH). C NMR: 14.1, 22.7, 25.3, 27.0, 27.1, 27.8, 28.7, 28.9,
diluted with Et
h. The solid p-toluenesulfonate of GGG–OCH
collected, washed with Et O (50 mL) and dried. After crystal-
lization from MeOH–Et O the salt (5.5 g, 77%) melted at
76–177 1C. H NMR (DMSO-d ): 2.34 (3H, s, CH Ph), 3.74
2H, s, Gly NCH ), 3.97 (4H, s, Gly NCH ), 5.14 (2H, s,
2
O (50 mL), and cooled in an ice water bath for
2
6
1
9.3, 29.7, 30.9, 31.9, 41.4, 42.6, 43.3, 43.9, 46.2, 46.7, 48.1,
1.4, 67.4, 128.3, 128.4, 128.6, 135.3, 168.7, 169.7, 170.2, 170.4,
71.4, 171.7, 171.9, 173.8, 174.5. Anal. Calc. for
2
2
Ph was
2
2
C H N O : C 66.75, H 9.63, N 9.73%. Found: C 67.00,
6
4
110
8
10
1
1
6
3
H 9.78, N 9.68%.
(
2
2
PhCH O), 7.21 (2H, d, J = 8.4 Hz, Tosyl H ), 7.30–7.35 (5H,
2
Compound 4: [2-(2-{[1-(2-{2-[2-(4-dioctadecylcarbamoylbutyryl-
amino)acetylamino]acetylamino}acetyl)pyrrolidine-2-carbonyl]-
amino}acetylamino)acetylamino]acetic acid benzyl ester
Ar
1
m, Ph HAr), 7.69 (2H, d, J = 8.4 Hz, Tosyl HAr). C NMR:
3
2
1
1.4, 41.7, 42.1, 43.2, 68.1, 127.2, 129.5, 129.6, 129.9, 130.2,
37.5, 142.1, 143.7, 168.4, 171.4, 172.2.
1
8 COCH CH CH COOH. Dioctadecylamine (3.0 g, 5.75
2 2 2 2
mmol) and glutaric anhydride (0.85 g, 7.5 mmol) in THF (40
mL) was heated at reflux for 48 h. The solvent was evaporated
1
8
2
COCH
2
CH
2
CO–GGG–OCH
2
Ph.
2 2
18 COCH CH2-
COOH (1.87 g, 3.0 mmol) and EDCI (0.62 g, 3.2 mmol) were
and the product was crystallized from Et O to give a white
2
stirred in CH Cl2 (60 mL) at rt. After 0.5 h, TsOH–
2
1
solid (3.38 g, 92%), mp 62–63 1C. H NMR: 0.88 (6H, t, J =
GGG–OCH Ph (1.32 g, 3.0 mmol) and Et N (1.2 mL) were
2
3
6
.6 Hz, CH
.53 (4H, br s, CH
CH CH CO), 2.44 (4H, m, COCH
3.20 (2H, t, J = 7.8 Hz, CH (CH ) CH N), 3.29 (2H, t,
3
), 1.25 (60H, pseudo-s, CH
3
(CH
N), 1.97 (2H, m,
CH CH CO),
2 2 2
)15CH CH N),
added and the mixture was stirred at rt overnight. Evaporation
1
3
(CH CH
2
)
15CH
2
2
of the solvent followed by chromatography (SiO , 5%
2
COCH
2
2
2
2
2
2
3
MeOH–CHCl ) gave a white solid (2.26 g, 85%), mp 72–
1
7
3 1C. H NMR: 0.88 (6H, t, J = 6.9 Hz, CH
CH CH N), 1.24 (60H, pseudo-s, CH (CH
.40–1.55 (4H, br s, CH (CH CH
J = 5.6 Hz, COCH ), 2.78 (2H, t, J = 5.6 Hz, COCH
.10–3.25 (4H, overlapping signals due to CH (CH N),
.93 (4H, d, J = 5.7 Hz, Gly NCH ), 4.03 (2H, d, J = 5.7 Hz,
3
(CH
2
)
15
-
3
2 16
2
1
3
J = 7.8 Hz, CH
3
(CH
6.9, 27.1, 27.7, 28.0, 29.1, 29.3, 29.4, 29.6, 29.7, 31.7, 31.9,
3.5, 46.2, 48.1, 172.2.
2 2
)16CH N). C NMR: 14.1, 20.5, 22.7,
2
2
3
2
)
15CH
2
CH
2
N),
2
3
1
3
2
)
15CH
2
2
N), 2.41 (2H, t,
),
2
2
3
3
3
2 2
)16CH
1
8 COCH CH CH CO–GGG–OCH Ph.
2 2 2 2 2
2
1
2
8 COCH CH CH COOH (1.5 g, 2.36 mmol), EDCI (0.5 g,
2 2 2 2
Gly NCH ), 5.14 (2H, s, PhCH O), 6.58 (1H, br s, Gly
2
2
.6 mmol), and HOBt (0.35 g, 2.6 mmol) were dissolved in
Cl
(60 mL) and stirred at rt. After 0.5 h, TsOH ꢂ GG-
G–OCH Ph (1.06 g, 2.36 mmol) and Et N (1.0 mL) were
added and the mixture was stirred at rt for 2 days. The solvent
was evaporated and the residue was chromatographed (SiO
% MeOH–CHCl ) to give a white solid (1.82 g, 87%), mp
81–82 1C. H NMR: 0.88 (6H, t, J = 6.9 Hz, –CH CH ), 1.24
(60H, pseudo-s, CH (CH CH N), 1.40–1.55 (4H, br s,
CH (CH ) CH CH N), 1.95 (2H, m, COCH CH CH CO),
CONH), 7.16 (1H, t, J = 5.7 Hz, Gly CONH), 7.30–7.35
1
3
5H, m, HAr), 8.17 (1H, t, J = 5.7 Hz, Gly CONH). C NMR:
CH
2
2
(
2
3
1
3
1
4.1, 22.7, 27.0, 27.1, 27.7, 28.7, 29.0, 29.3, 29.6, 29.7, 31.7,
1.9, 41.0, 43.1, 43.9, 46.5, 48.2, 66.8, 128.1, 128.3, 128.6,
35.4, 169.4, 169.6, 169.9, 171.9, 174.6.
2
,
2
3
1
1
8
2
COCH
2
CH
2
CO–GGG–OH.
18
2
COCH
2
CH
2
CO–
2
3
GGG–OCH
2
Ph (2.2 g, 2.5 mmol) and 10% Pd/C (0.2 g) in
3
2
)15CH
2
2
EtOH (100 mL) were shaken (60 psi H ) for 3 h, heated to
2
3
2 15
2
2
2
2
2
reflux, and then filtered (Celite pad). Evaporation of the
1
solvent gave a white solid (1.86 g, 94%), mp 151–152 1C. H
2.25–2.40 (4H, m, COCH CH CH CO), 3.10–3.25 (4H, over-
2 2 2
lapping signals due to CH (CH ) CH N), 3.92 (2H, d,
3 2 16 2
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1960–1972 | 1967

View Article Online
J = 5.7 Hz, Gly NCH
2
), 3.96 (2H, d, J = 5.7 Hz, Gly NCH
2
),
The mixture was cooled to rt, solvent evaporated, the residue
4
7
7
.02 (2H, d, J = 5.7 Hz, Gly NCH ), 5.13 (2H, s, PhCH O),
2
2
3 2 3
was dissolved in CHCl (40 mL), washed with aq. Na CO
.17 (1H, t, J = 5.7 Hz, Gly CONH), 7.30–7.35 (5H, m, HAr),
.39 (1H, t, J = 5.7 Hz, Gly CONH), 7.74 (1H, t, J = 5.7 Hz,
(2 ꢃ 25 mL), brine (25 mL), dried over MgSO , and evapo-
4
rated. The oily product was chromatographed (SiO , 5–15%
2
1
3
Gly CONH). C NMR: 14.0, 21.2, 22.6, 26.9, 27.1, 27.8, 29.1,
ethyl acetate–hexanes) to give a slightly yellow oil (2.98 g,
1
52%). H NMR: 0.88 (6H, t, J = 6.9 Hz, CH (CH ) -
2
6
1
9.3, 29.4, 29.5, 29.6, 31.9, 35.2, 41.2, 43.0, 43.4, 46.0, 48.2,
7.0, 128.2, 128.3, 128.5, 135.3, 169.6, 169.7, 170.1, 172.1,
73.9.
3
2 15
CH
1.50 (4H, br s, CH
.20–3.30 (4H, overlapping signals due to CH
2
CH
2
N), 1.25 (60H, pseudo-s, CH
(CH CH N), 2.50 (3H, s, CH
2
(CH )16CH
3
(CH
2
)
15CH
2
CH
2
N),
3
N),
2
N),
3
2
)
15CH
2
2
3
3
1
8 COCH CH CH CO–GGGPGGG–OCH Ph.
2 2 2 2 2
2 2
3.45 (2H, s, COCH ), 3.53 (2H, s, COCH ), 5.15 (2H, s,
13
PhCH O), 7.30-7.35 (5H, m, HAr). C NMR: 14.1, 22.7,
2
1
8 COCH CH CH CO–GGG–OCH Ph (1.17 g, 1.31 mmol)
2
2
2
2
2
and 10% Pd/C (0.2 g) in EtOH (40 mL) were shaken (60 psi
2
6.9, 27.1, 27.7, 29.0, 29.3, 29.4, 29.6, 29.7, 31.9, 42.5, 45.8,
H ) for 3 h. The mixture was heated to reflux, filtered (Celite
2
47.4, 57.6, 57.9, 66.1, 128.1, 128.2, 128.5, 135.9, 169.2, 170.9.
pad), and the solvent was evaporated to afford a white solid
(
0.99 g, 93%). This product was used in the next step without
further purification. 18 COCH CH CH CO–GGG–OH (0.52 g,
.65 mmol) was suspended in CH Cl
0.14 g, 0.72 mmol), PGGG–OCH
18 COCH (CH )NCH CO–GGG–OCH Ph.
2
2
2
3
2
2
2
2
2
2 2 3 2 2
18 COCH (CH )NCH COOCH Ph (2.98 g, 4.02 mmol) and
0
2
2
(30 mL) at 0 1C. EDCI
Ph ꢂ HCl (0.27 g, 0.65
10% Pd/C (0.2 g) in EtOH (30 mL) were shaken (60 psi H ) for
2
(
2
3 h. The mixture was heated to reflux, filtered (Celite pad), and
mmol), HOBt (0.1 g, 0.72 mmol) and Et
3
N (0.3 mL) were
the solvent was evaporated to give a white solid (2.45 g, 93%).
This product was used without further purification.
added and the mixture was stirred for 48 h at rt. The solvent
was evaporated, the residue was chromatographed (SiO , 10%
2
1
8
2
COCH
2 3 2
(CH )NCH COOH. (1.0 g, 1.53 mmol) was
MeOH–CHCl ) and then crystallized from MeOH to give a
3
1
suspended in CH Cl (30 mL) at 0 1C. EDCI (0.31 g, 1.62
white solid (0.4 g, 53%), mp 160–161 1C. H NMR: 0.87
2
2
mmol), TsOH ꢂ GGG–OCH
2
Ph (0.69 g, 1.53 mmol), HOBt
(
6H, t, J = 6.3 Hz, CH (CH ) CH CH N), 1.25 (60H, m,
3 2 15 2 2
(
0.22 g, 1.63 mmol) and DIEA (0.9 mL) were added and the
CH
3
(CH
CH
CH
4H, m, CH
signals due to Pro CH and NCH
.14 (2H, m, OCH Ph), 7.30–7.40 (5H, m, HAr), 7.47 (1H, br s,
NH), 7.57 (1H, br s, NH), 7.77 (1H, br s, HAr), 8.12 (1H, br s,
2
)
15CH
2
CH
N), 1.80–2.40 (10H, overlapping signals due to
CH CO and Pro NCH CH CH ), 3.10–3.20
(CH N), 3.25–4.30 (15H, overlapping
CH CH , and Gly CH ),
2 3 2 15
N), 1.50 (4H, br s, CH (CH ) -
reaction was stirred for 48 h at rt. The solvent was evaporated
and the residue was chromatographed (SiO2, 2–5%
CH
2
2
COCH
2
2
2
2
2
2
MeOH–CHCl ) to give a white solid (1.21 g, 87%), mp
3
(
3
2
)
16CH
2
1
6
3–64 1C.
CH
H
NMR: 0.88 (6H, t,
CH N), 1.25 (60H, m, CH
N), 1.40–1.60 (4H, m, CH (CH CH
J
=
6.3 Hz,
(CH
N), 2.50
2
2
2
2
(CH
CH
2
)15CH
2
2
3
2 15
) -
5
2
3
CH
2
2
3
2
)
15CH
2
2
1
3
3
(3H, s, CH N), 3.10–3.15 (4H, overlapping signals due to
NH), 8.28 (1H, br s, NH). C NMR: 14.0, 21.2, 22.7, 25.2,
CH
COCH
signals due to Gly 4CH
J = 5.7 Hz, NH), 7.30–7.35 (5H, m, HAr), 8.18 (1H, t, J = 5.7
3
(CH
), 3.41 (2H, s, COCH
), 5.14 (2H, s, OCH
2
)
16CH
2
N and COCH
), 3.90–4.10 (6H, overlapping
Ph), 6.99 (1H, t,
2
), 3.22 (2H, t, J = 6.3 Hz,
2
4
1
1
9
7.0, 27.1, 27.8, 29.0, 29.1, 29.3, 29.4, 29.6, 29.7, 31.9, 35.2,
1.4, 41.7, 43.0, 46.2, 46.8, 48.3, 61.3, 67.2, 128.2, 128.4, 128.6,
28.8, 135.4, 168.6, 170.1, 170.2, 170.6, 171.3, 172.4, 173.9,
2
2
2
2
74.0. Anal. Calc. for C65
112 8
H N O10: C 66.98, H 9.68, N
1
3
Hz, NH), 8.59 (1H, t, J = 5.7 Hz, NH). C NMR: 14.1, 22.7,
6.9, 27.1, 27.6, 28.9, 29.3, 29.4, 29.7, 31.9, 41.2, 43.3, 44.9,
.61%. Found: C 66.70, H 9.70, N 9.74%.
2
Compound 5: [2-(2-{[1-(2-{2-[2-(2-dioctadecylcarbamoylmethoxy-
acetylamino)acetylamino]acetylamino}acetyl)pyrrolidine2-carbonyl]-
amino}acetylamino)acetylamino]acetic acid benzyl ester, 5
46.6, 47.3, 58.9, 61.0, 66.9, 128.2, 128.4, 128.6, 135.4, 169.2,
169.5, 170.3, 172.9.
1
8 COCH (CH )NCH CO–GGG–OH.
2 2 3 2
A detailed preparation of this compound has previously
1a
appeared.
1
8 COCH (CH )NCH CO–GGG–OCH Ph (1.2 g, 1.31
2 2 3 2 2
3
mmol) and 10% Pd/C (0.12 g) in EtOH (30 mL) was shaken
60 psi H ) for 3 h. The reaction mixture was heated to reflux,
(
2
Compound 6: {2-[2-({1-[2-(2-{2-[2-(dioctadecylcarbamoylmethyl-
methylamino)acetylamino]acetylamino}acetylamino)acetyl]-
pyrrolidine-2-carbonyl}amino)acetylamino]acetylamino}acetic
acid benzyl ester
filtered (Celite pad), and the solvent was evaporated to afford
a white solid (1.06 g, 98%). This product was used without
1
further purification. H NMR: 0.87 (6H, t, J = 6.3 Hz,
3 2 2 2 3 2 15
CH (CH )15CH CH N), 1.25 (60H, m, CH (CH ) -
1
8 COCH (CH )NCH COOCH Ph. Methyliminodiacetic
2 2 3 2 2
CH CH N), 1.51 (4H, m, CH (CH ) CH CH N), 2.66 (3H,
2
2
2
3
2 15
2
acid 1.15 g, 7.82 mmol) was suspended in acetic anhydride
5.5 g) and heated to reflux for 0.5 h. The mixture was cooled
s, CH
3
N), 3.15 (2H, t, J = 6.3 Hz, CH
(CH
.72 (2H, s, COCH ), 3.80–4.10 (6H, overlapping signals due
3
(CH
2
)
16CH
2
N), 3.26
(
(
2H, t, J = 6.3 Hz, CH
3
)16CH N), 3.59 (2H, s, COCH ),
2 2 2
and the solvent removed in vacuo. The residue was dissolved in
THF (40 mL), dioctadecylamine (4.0 g) was added, and the
reaction was heated to reflux overnight. The hot solution was
filtered (Celite pad). The solvent was evaporated and the
residue dissolved in toluene (50 mL). Benzyl alcohol (1.92 g,
3
2
to Gly CH ), 7.67 (1H, br s, NH), 8.14 (1H, br s, NH), 8.60
2
13
1H, br s, NH). C NMR: 14.1, 22.7, 26.9, 27.1, 27.7, 28.9,
(
2
1
9.3, 29.6, 29.7, 31.9, 42.2, 43.0, 43.7, 46.6, 47.5, 57.8, 59.5,
67.4, 170.2, 170.3, 173.3.
1
1
7.7 mmol) and p-toluenesulfonic acid monohydrate (3.38 g,
7.8 mmol) were added. The mixture was heated to reflux and
18 COCH (CH )NCH CO–GGGPGGG–OCH Ph.
2
2
3
2
2
water was removed by using a Dean–Stark trap (overnight).
18 COCH (CH )NCH CO–GGG–OH (1.03 g, 1.25 mmol)
2 2 3 2
1
968 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
was suspended in CH
mmol), PGGG–OCH
0.18 g, 1.33 mmol) and DIEA (0.5 mL) were added and the
2
Cl
2
(30 mL) at 0 1C. EDCI (0.25 g, 1.30
67.0, 128.2, 128.4, 128.6, 135.4, 168.7, 169.4, 169.6, 169.7,
170.1.
2
Ph ꢂ HCl (0.52 g, 1.25 mmol), HOBt
(
1
2
8 COCH
2
SCH
2
CO–GGGPGGG–OCH
2
Ph.
reaction was stirred for 48 h at rt. The solvent was evaporated
and the residue chromatographed (SiO2, 5–10%
1
8 COCH SCH
2 2
2
CO–GGG–OCH
2
Ph (2.0 g, 2.2 mmol) was
dissolved in MeOH (10 mL) and 15% NaOH (30 mL) was
added. The mixture was heated to reflux overnight, then
cooled and the pH was adjusted to 2 (conc. HCl). The mixture
was filtered and the precipitate was dried in vacuo to give a
white solid (1.7 g, 94%). This product was used without
further purification.
MeOH–CHCl ) and recrystallized from MeOH to give a white
3
1
solid (1.23 g, 83%), mp 106–107 1C. H NMR: 0.87 (6H, t,
J
=
6.3 Hz, CH
(CH CH N), 1.51 (4H, m, CH
CH N), 1.85–2.20 (4H, m, Pro NCH CH CH
N), 3.13 (2H, t, J = 6.3 Hz, CH (CH
.26 (2H, t, J = 6.3 Hz, CH (CH N), 3.50–4.20 (18H,
CH CH , Gly CH
COCH ), 4.36 (1H, m, Pro CH), 5.14 (2H, m, OCH Ph),
3
(CH
2
)
15CH
2
CH
2
N), 1.25 (60H, m,
(CH
), 2.77
N),
CH
3
2
2
)
15CH
2
2
3
2 15
) -
CH
2
2
2
2
(
3H, s, CH
3
3
2 2
)16CH
3
3
2 2
)16CH
1
8
2
COCH
suspended in CH
mmol), PGGG–OCH
0.23 g, 1.70 mmol) and DIEA (0.3 mL) were added and the
reaction was stirred for 48 h at rt. The solvent was evaporated
and the residue chromatographed (SiO 5–10%
MeOH–CHCl ) to give a white solid (1.58 g, 83%), mp
9–60 1C. H NMR: 0.87 (6H, t, J = 6.3 Hz, CH (CH ) -
2
SCH
2
CO–GGG–OH. (1.33 g, 1.61 mmol) was
Cl (30 mL) at 0 1C. EDCI (0.33 g, 1.72
Ph ꢂ HCl (0.67 g, 1.61 mmol), HOBt
overlapping signals due to Pro NCH
2
2
2
2
,
2
2
2
2
2
7
.30–7.35 (5H, m, HAr), 7.64 (1H, br s, NH), 7.80 (1H, br s,
(
NH), 7.93 (1H, br s, NH), 8.09 (1H, br s, NH), 8.31 (1H, br s,
1
NH), 8.78 (1H, br s, NH). C NMR: 14.1, 22.7, 25.1, 26.7,
3
2
,
2
4
1
7.1, 27.7, 28.9, 29.1, 29.3, 29.4, 29.6, 29.7, 31.9, 41.3, 41.9,
2.9, 43.1, 43.4, 43.9, 46.5, 47.0, 47.4, 57.9, 61.4, 67.2, 128.2,
28.4, 128.6, 135.3, 169.2, 170.2, 170.3, 170.4, 171.0, 173.4.
3
1
5
3
2 15
CH CH N), 1.25 (60H, m, CH (CH ) CH CH N), 1.40–1.60
2
2
2
3
2 15
2
113 9
Anal. Calc. for C65H N O10: C 66.12, H 9.65, N 10.68%.
(
4H, m, CH (CH ) CH CH N), 1.80–2.10 (4H, m, Pro
3 2 15 2 2
Found: C 65.93, H 9.61, N 10.48%.
NCH CH CH ), 3.00–4.10 (22H, overlapping signals due to
2
2
2
CH
3
(CH
2
)
15CH
2
CH
2
N, Pro NCH
2
CH
2
CH
2
,
Gly CH
Ph),
.30–7.35 (5H, m, HAr), 7.70 (2H, br s, NH), 8.48 (1H, t,
2
,
Compound 7: [2-(2-{[1-(2-{2-[2-(2-dioctadecylcarbamoylmethyl-
sulfanylacetylamino)acetylamino]acetylamino}acetyl)pyrroli-
dine-2-carbonyl]amino}acetylamino)acetylamino]acetic acid
benzyl ester, 7
2 2
COCH S), 4.38 (1H, m, Pro CH), 5.14 (2H, s, OCH
7
13
J = 5.7 Hz, NH). C NMR: 14.1, 22.7, 25.0, 26.9, 27.0, 27.6,
2
4
1
8.9, 29.1, 29.4, 29.7, 31.7, 31.9, 34.1, 41.1, 42.6, 42.9, 43.6,
6.4, 47.1, 48.8, 61.3, 66.9, 128.2, 128.3, 128.5, 135.4, 168.7,
69.0, 169.8, 169.9, 170.0, 170.2, 173.0. Anal. Calc. for
2 2 2
18 COCH SCH COOH. Dioctadecylamine (7.9 g, 15.1
mmol) and thiodiacetic anhydride (2.0 g, 15.1 mmol) in
THF (100 mL) was heated at reflux for 48 h. The solvent
C H N O S ꢂ H O: C 63.97, H 9.39, N 9.32%. Found: C
64 110
8
10
2
6
3.87, H 9.37, N 9.12%.
was evaporated and the product was crystallized (Et O) to give
2
1
the amido acid as a white solid (9.9 g, 70%), mp 59–61 1C. H
NMR: 0.88 (6H, t, J = 6.6 Hz, CH (CH ) CH CH N), 1.25
Compound 8: [2-(2-{[1-(2-{2-[2-(3-dioctadecylcarbamoylbenzoyl-
amino)acetylamino]acetylamino}acetyl)pyrrolidine-2-carbonyl]-
amino}acetylamino)acetylamino]acetic acid benzyl ester
3
2 15
2
2
(
60H, pseudo-s, CH
3
(CH
2
)
15CH
2
CH
2
N), 1.54 (4H, br s,
7.8 Hz,
N), 3.30–3.40 (4H, overlapping signals due
N and COCH S), 3.48 (2H, s, COCH S).
CH
CH
3
(CH
(CH
2
)
)
15CH
16CH
2
CH
2
N), 3.26 (2H, t,
J
=
N,N-Dioctadecylisophthalamic acid. Isophthaloyl dichloride
3
2
2
(
(
4.25 g, 20.9 mmol) was dissolved in a mixture of dry DMF
150 mL) and CH Cl (300 mL). Dioctadecylamine (2.0 g, 3.9
to CH
3
3
(CH
2
)
16CH
2
2
2
1
C NMR: 14.1, 22.7, 26.9, 27.6, 29.1, 29.2, 29.3, 29.5, 29.7,
1.9, 32.45, 34.6, 47.1, 49.0, 169.8, 170.2.
2
2
mmol) and Et N (6.6 mL) were dissolved in a mixture of dry
3
3
DMF (150 mL) and CH Cl (300 mL) and added dropwise to
2
2
the solution of isophthaloyl dichloride during 15 min. The
1
8
2
COCH
2
SCH CO–GGG–OCH
2
2
Ph. To 18 COCH
2
2
-
mixture was stirred at rt for 40 min and evaporated in vacuo.
SCH
2
COOH (2.0 g, 3.0 mmol) dissolved in CH
2
Cl (30 mL),
2
The product was dissolved in CH
with dilute HCl (3 ꢃ 50 mL), brine (25 mL), dried over MgSO
and evaporated. The oily product was chromatographed
SiO , 25% EtOAc–hexanes) to give a white solid (1.07 g,
2 2
Cl (150 mL) and washed
EDCI (0.62 g, 3.2 mmol) and HOBt (0.43 g, 3.2 mmol) were
added and the mixture was stirred at rt. After 0.5 h,
4
TsOH ꢂ GGG–OCH Ph (1.32 g, 3.0 mmol) and DIEA
2
(
2
(
1.7 mL) were added and the reaction mixture was stirred at
1
4
1%, mp 71–72 1C). H NMR: 0.87 (6H, t, J = 6.6 Hz, CH
.25 (60H, pseudo-s, CH (CH ) CH CH N), 1.50 (2H, br s,
3
),
rt overnight. The solvent was evaporated and the residue
chromatographed (SiO , 2% MeOH–CHCl ) to give a color-
1
less oil (2.20 g, 79%). H NMR: 0.87 (6H, t, J = 6.9 Hz,
1
3
2 15
2
2
2
3
CH (CH ) CH CH N), 1.66 (2H, br s, CH (CH ) -
2 15
3
2 15
2
2
3
CH CH N), 3.17 (2H, t, J = 7.8 Hz, CH (CH ) CH N),
2
2
2
3
2 16
CH
3
(CH
CH
2
)
15CH
2
CH
2
N), 1.25 (60H, pseudo-s, CH
(CH
3
(CH
CH
2
15
) -
3
.49 (2H, t, J = 7.8 Hz, CH (CH ) CH N), 7.40–7.45 (1H,
3 2 16 2
CH
2
2
N), 1.40–1.60 (4H, br s, CH
3
2
)
15CH
2
2
N),
13
m, HAr), 7.50–7.57 (1H, m, HAr), 8.00–8.10 (2H, m, HAr).
NMR: 14.1, 22.7, 26.5, 27.1, 27.5, 29.3, 29.7, 31.9, 45.1, 49.2,
28.2, 128.8, 129.4, 130.7, 131.8, 137.7, 170.1, 170.4.
C
3
.20–3.35 (6H, overlapping signals due to CH
and COCH S), 3.44 (2H, s, COCH S), 3.90–4.10 (6H, over-
lapping signals due to Gly NCH ), 5.15 (2H, s, PhCH O), 7.11
1H, t, J = 5.7 Hz, Gly CONH), 7.30–7.35 (5H, m, HAr), 7.74
1H, t, J = 5.7 Hz, Gly CONH), 8.07 (1H, t, J = 5.7 Hz, Gly
3 2 2
(CH )16CH N
2
2
1
2
2
(
(
2 6 4 2
18 CO–m-C H –CO–GGG–OCH Ph. To N,N-dioctadecyl-
isophthalamic acid (0.48 g, 0.7 mmol) was dissolved
in CH Cl (25 mL), EDCI (0.15 g, 0.77 mmol) and HOBt
1
3
CONH). C NMR: 14.0, 22.6, 26.9, 27.0, 27.7, 29.2, 29.3,
9.5, 29.6, 29.7, 31.9, 33.1, 35.7, 41.2, 43.1, 44.0, 46.6, 48.8,
2
2
2
(0.11 g, 0.77 mmol) were added and the mixture was stirred at
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1960–1972 | 1969

View Article Online
rt. After 0.5 h, TsOH ꢂ GGG–OCH
2
Ph (0.32 g, 0.7 mmol) and
was stirred at rt for 40 min and the solvent was removed in
Et N (0.3 mL) were added and the reaction mixture was
3
2 2
vacuo. The product was dissolved in CH Cl (150 mL) and
stirred at rt for 48 h. The solvent was evaporated and the
residue chromatographed (SiO , 2% MeOH–CHCl ) to give a
washed with 10% HCl (3 ꢃ 50 mL), brine (25 mL), dried over
MgSO , and evaporated. The oily product was chromato-
4
2
3
1
white solid (0.43 g, 66%, mp 48–49 1C). H NMR: 0.87 (6H, t,
J = 6.9 Hz, –CH CH ), 1.25 (60H, pseudo-s, CH (CH ) -
graphed (SiO , 25% EtOAc–hexanes) to afford a white solid
2
1
(1.07 g, 37%, mp 51–52 1C). H NMR: 0.87 (6H, t, J = 6.6
2
3
3
2 15
CH
br s, CH
CH (CH
2
CH
2
N), 1.54 (2H, br s, CH
(CH CH N), 3.13 (2H, t, J = 7.8 Hz,
N), 3.44 (2H, t, 7.8 Hz, CH
N), 3.83 (2H, d, J = 5.7 Hz, Gly NCH ), 3.80–
.00 (4H, overlapping signals due to Gly NCH ), 5.13 (2H, s,
O), 7.30–7.50 (10H, m, HAr), 7.87 (2H, s, HAr), 7.96
3
(CH
2
)
15CH
2
CH
2
N), 1.63 (2H,
Hz, CH
1.45–1.70 (4H, m, CH
7.8 Hz, CH (CH
CH (CH N) 7.44 (2H, d, J = 8.1 Hz, HAr), 8.13 (2H,
3
), 1.25 (60H, pseudo-s, CH
(CH CH N), 3.13 (2H, t, J =
N), 3.49 (2H, t, J = 7.8 Hz,
3 2 2 2
(CH )15CH CH N),
3
2
)
15CH
2
2
3
2
)
15CH
2
2
3
2
)
16CH
2
J
=
3
-
3
2 2
)16CH
(
CH
2
)
16CH
2
2
3
2 2
)16CH
1
3
4
2
d, J = 8.1 Hz, HAr). C NMR: 14.1, 22.7, 26.5, 27.1, 27.5,
27.7, 28.7, 29.1, 29.3, 29.7, 31.9, 44.9, 49.0, 126.6, 130.0, 130.3,
142.3, 170.5, 170.6.
PhCH
1H, br s, Gly CONH). C NMR: 14.1, 22.7, 26.6, 27.1, 27.5,
2
1
3
(
2
4
1
8.9, 29.1, 29.3, 29.4, 29.6, 29.7, 31.9, 41.2, 42.7, 43.8, 45.2,
9.3, 67.2, 125.4, 128.2, 128.4, 128.6, 128.7, 129.3, 134.0, 135.2,
37.2, 167.5, 169.6, 169.9, 170.0, 170.9.
18 CO–p-C H –CO–GGG–OCH Ph. To N,N-dioctadecyl-
2
2
6
4
terephthalamic acid (0.38 g, 0.55 mmol) dissolved in CH Cl
2
2
(
25 mL), EDCI (0.12 g, 0.6 mmol) and HOBt (0.08 g, 0.6
1
8
2
CO–m-C
-CO–GGG–OCH
Pd/C (0.14 g) in EtOH (30 mL) were shaken (60 psi H
6
H
4
–CO–GGGPGGG–OCH
Ph (0.85 g, 0.91 mmol) and 10%
) for 3 h.
2
Ph.
18
2
CO–
mmol) were added and the mixture was stirred at rt. After
m-C
6
H
4
2
0.5 h, TsOH ꢂ GGG–OCH Ph (0.25 g, 0.55 mmol) and Et N
2
3
2
(0.23 mL) were added and the reaction mixture was stirred at
rt for 48 h. The solvent was evaporated and the residue
chromatographed (SiO , 2% MeOH–CHCl ) to give a white
The reaction mixture was heated to reflux, filtered, and
evaporated to afford a white solid (0.68 g, 90%). This product
was used without further purification.
2
3
1
solid (0.42 g, 83%, mp 83–84 1C). H NMR: 0.87 (6H, t, J =
.9 Hz, -CH CH ), 1.25 (60H, pseudo-s, CH (CH ) -
6
2
3
3
2 15
1
8
2
CO–m-C
6 4
H –CO–GGG–OH. (0.5 g, 0.59 mmol) was
CH CH N), 1.40–1.65 (4H, br s, CH (CH ) CH CH N),
2
2
2
3
2 15
2
suspended in CH Cl (30 mL) at 0 1C. EDCI (0.12 g, 0.65
2
2
3.12 (2H, t, J = 7.8 Hz, CH (CH ) CH N), 3.46 (2H, t,
3 2 16 2
mmol), PGGG–OCH Ph ꢂ HCl (0.24 g, 0.59 mmol), HOBt
2
J = 7.8 Hz, CH (CH ) CH N), 3.87 (2H, d, J = 5.7 Hz, Gly
3 2 16 2
(
0.09 g, 0.65 mmol) and Et N (0.25 mL) were added and the
3
NCH ), 4.00–4.10 (4H, overlapping signals due to Gly NCH ),
2
2
reaction was stirred for 48 h at rt. The solvent was evaporated
and the residue chromatographed (SiO , 10% MeOH–CHCl
to give 8 as a white solid (0.34 g, 48%), mp 138–139 1C. H
NMR: 0.87 (6H, t, J = 6.3 Hz, CH (CH CH N), 1.25
60H, m, CH (CH CH N), 1.54 (2H, br s,
CH N), 1.63 (2H, br s, CH (CH
N), 1.80–2.10 (4H, m, Pro NCH CH CH ), 3.10
(CH N), 3.10–4.20 (17H,
overlapping signals due to CH (CH ) CH CH N, Pro CH
5
.15 (2H, s, PhCH
2
O), 7.16 (2H, m, Gly CONH), 7.34 (5H,
2
3
)
pseudo-s, HAr), 7.37 (2H, d, J = 8.1 Hz, HAr), 7.65 (1H, m,
1
Gly CONH) 7.87 (2H, d, J = 8.1 Hz, HAr). C NMR: 14.0,
1
3
3
2
)
15CH
2
2
2
4
1
2.6, 27.1, 27.5, 28.8, 29.3, 29.6, 29.7, 31.9, 41.3, 42.8, 43.7,
5.0, 49.1, 67.2, 126.5, 127.7, 128.2, 128.5, 128.6, 134.2, 135.2,
40.5, 167.6, 169.4, 169.7, 170.7.
(
3
2
)
15CH
2
2
CH
3
(CH
CH
2
)
15CH
2
2
3
2 15
) -
CH
2
2
2
2
2
(
2H, t, J = 7.8 Hz, CH
3
2 2
)16CH
1
8 CO–p-C H -CO–GGGPGGG–OCH Ph.
2 6 4 2
3
2 15
2
2
18 CO–pC H –CO–GGG–OCH Ph (0.7 g, 0.75 mmol) and
2 6 4 2
and NCH CH CH , and Gly CH ), 5.09 (2H, m, OCH Ph),
2
2
2
2
2
10% Pd/C (0.12 g) in EtOH (30 mL) was shaken (60 psi H2)
for 3 h. The mixture was heated to reflux, filtered (Celite pad)
and the solvent was evaporated to afford a white solid (0.61 g,
7
.30–7.40 (10H, m, HAr), 7.55 (1H, br s, NH), 7.68 (1H, br s,
NH), 7.85–7.95 (2H, m, HAr), 8.30 (1H, br s, NH), 8.38 (1H, br
1
s, NH). C NMR: 14.1, 22.6, 25.2, 26.6, 27.1, 27.5, 28.6, 28.9,
3
9
1
7%). This product was used in without further purification.
CO–pC -CO–GGG–OH (0.4 g, 0.48 mmol) was sus-
Cl (30 mL) at 0 1C. EDCI (0.1 g, 0.53 mmol),
Ph ꢂ HCl (0.2 g, 0.48 mmol), HOBt (0.07 g, 0.53
N (0.2 mL) were added and the reaction was
2
4
1
1
6
9.1, 29.4, 29.6, 29.7, 31.9, 41.2, 41.7, 42.6, 42.7, 43.4, 44.1,
5.2, 46.8, 49.3, 61.0, 67.2, 125.8, 128.2, 128.4, 128.6, 128.8,
29.1, 133.8, 135.3, 137.2, 167.2, 168.3, 170.0, 170.1, 170.3,
8
2
6 4
H
pended in CH
2
2
PGGG–OCH
mmol) and Et
2
70.5, 171.1, 171.3, 173.9. Anal. Calc. for C68
110 8
H N O10: C
3
8.08, H 9.24, N 9.34%. Found: C 67.87, H 9.17, N 9.48%.
stirred for 48 h at rt. The solvent was evaporated and the
residue was chromatographed (SiO , 5% MeOH–CHCl ) to
2
3
Compound 9: [2-(2-{[1-(2-{2-[2-(4-dioctadecylcarbamoylbenzoyl-
amino)acetylamino]acetylamino}acetyl)pyrrolidine-2-carbonyl]-
amino}acetylaminoacetylamino]acetic acid benzyl ester
1
give 9 as a white solid (0.34 g, 59%), mp 150–151 1C.
NMR: 0.87 (6H, t, J = 6.3 Hz, CH (CH ) CH CH N), 1.25
H
3
2 15
2
2
(
60H, m, CH
(CH CH
CH ), 3.07 (2H, t, J = 7.8 Hz, CH
3.10–4.25 (17H, overlapping signals due to CH
CH CH N, Pro CH and NCH CH CH
5.14 (2H, m, OCH
J = 8.1 Hz, HAr), 7.54 (2H, br s, NH), 7.95 (2H, d, J = 8.1
3
(CH
2
)
15CH
2
CH
2
N), 1.40–1.60 (4H, m, CH
3
-
N,N-Dioctadecylterephthalamic acid. Terephthalic acid
2
)
15CH
2
2
N), 1.80–2.10 (4H, m, Pro NCH
2
-
(
3.8 g, 22.8 mmol) was dissolved in toluene (20 mL) and
2
CH
2
3
(CH
2
2
)16CH N),
SOCl (15 mL) was added in one portion. The reaction
2
3
(CH
2
)
15
-
mixture was heated at 40 1C overnight and then evaporated
in vacuo. The resulting yellow oil was dissolved in dry DMF
2
2
2
2
2
, and Gly CH
2
),
2
Ph), 7.30–7.35 (5H, m, HAr), 7.38 (2H, d,
(
2 2
150 mL) and CH Cl (300 mL). Dioctadecylamine (2.2 g, 4.2
1
3
mmol) and Et N (7.2 mL) were dissolved in dry DMF (150
3
Hz, HAr), 8.04 (1H, br s, NH), 8.52 (2H, br s, NH). C NMR:
14.0, 22.6, 25.1, 26.6, 27.1, 27.5, 28.7, 28.9, 29.2, 29.3, 29.4,
29.7, 31.9, 41.3, 41.5, 42.6, 42.8, 43.4, 43.6, 45.0, 46.7, 49.2,
mL) and CH Cl (300 mL) and added dropwise to the solution
2
2
of terephthaloyl chloride during 15 min. The reaction mixture
1
970 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
6
1
1.0, 67.3, 126.4, 127.9, 128.2, 128.4, 128.6, 134.4, 135.3, 140.1,
67.7, 168.0, 170.0, 170.3, 170.4, 170.5, 171.0, 171.6, 174.2.
incorporate and to confirm analytically in the product. We
conclude from the studies described here that 3-thiaglutaric
acid is equally convenient to incorporate and gives a SAT that
Anal. Calc. for C H N O : C 68.08, H 9.24, N 9.34%.
6
8
110
8
10
ꢀ
Found: C 67.86, H 9.27, N 9.59%.
typically has higher Cl transport efficacy. Terephthalic acid is
also a candidate for a midpolar regime mimic but it leads to a
somewhat less efficacious SAT (9) than 7. The location of the
carboxyl groups in terephthalic acid prevents formation of a
cyclic anhydride, making the diacid less convenient to incor-
porate. Future designs will incorporate 3-thiaglutaric acid.
Polar elements and very high flexibility will be avoided in
designing new synthetic anion transporters of this class.
Formation of vesicles. 1,2-Dioleoyl-sn-glycero-3-phospho-
choline (DOPC) and 1,2-dioleoyl-sn-glycero-3-phosphate
(
DOPA) were obtained from Avanti Polar Lipids as CHCl
3
solutions (vials with 25 mg in 2.5 mL). Dry lipid films of
DOPC-DOPA (20 mg, 7 : 3 w/w, i.e., 2 : 1 mol/mol) were
dissolved in Et O (0.5 mL) and then 0.5 mL of internal buffer
2
0
(
600 mM KCl : 10 mM N-2-hydroxyethylpiperazine-N -2-
ethanesulfonic acid (HEPES), pH adjusted to 7.0) were added.
The mixture was sonicated for approximately 10 s to give an
opalescent dispersion. The organic solvent was then removed
by evaporation under mild vacuum at 30 1C. The suspension
was filtered through a 200 nm filter membrane (five times)
using a mini extruder and passed through a Sephadex G25
column, which had been equilibrated with external buffer
Acknowledgements
We thank the NIH for grants (GM 36262 and GM 63190) that
supported this work.
References
(
2 4
400 mM K SO : 10 mM HEPES, pH adjusted to 7.0). The
1 G. W. Gokel and O. Murillo, Acc. Chem. Res., 1996, 29, 425–432.
2 T. M. Fyles, Curr. Opin. Chem. Biol., 1997, 1, 497–505.
vesicles collected were subsequently characterized as B200 nm
using laser light scattering. The final lipids concentration was
assessed through colorimetric determination of the phospho-
lipid-ammonium ferrothiocyanate complex.
3
4
G. W. Gokel and A. Mukhopadhyay, Chem. Soc. Rev., 2001, 30,
74–286.
G. W. Gokel, P. H. Schlesinger, N. K. Djedovic, R. Ferdani, E. C.
Harder, J. Hu, W. M. Leevy, J. Pajewska, R. Pajewski and M. E.
Weber, Bioorg. Med. Chem., 2004, 12, 1291–304.
2
Studies of chloride release from liposomes. Chloride release
was assayed directly on B200 nm, 0.23 mM phospholipid
vesicles by using a chloride selective electrode (Accumet
Chloride Combination Electrode). The electrode was intro-
duced in a 2 mL vesicle solution and allowed to equilibrate.
The voltage output was recorded, and after 5 min, aliquots of
the compound solution (B9 mM in 2-propanol) were added.
The amount of 2-propanol was limited to 20 mL to minimize
its effect on the liposomes in any experiment. Complete lysis of
the vesicles was induced by addition of a 2% aqueous solution
of Triton X-100 (100 mL) and the data collected (DigiData
5 (a) N. Sakai, J. Mareda and S. Matile, Acc. Chem. Res., 2005, 38,
9–87; (b) A. L. Sisson, M. R. Shah, S. Bhosale and S. Matile,
Chem. Soc. Rev., 2006, 35, 1269–1286.
7
6
(a) B. McNally and W. M. Leevy, Supramol. Chem., 2007, 19,
29–37; (b) B. D. Smith, A. P. Davis and D. Shephard, Chem. Soc.
Rev., 2007, 36, 348–357.
7
8
9
I. Tabushi, Y. Kuroda and K. Yokota, Tetrahedron Lett., 1982, 23,
4
V. E. Carmichael, P. J. Dutton, T. M. Fyles, T. D. James, J. A.
Swan and M. Zojaji, J. Am. Chem. Soc., 1989, 111, 767–769.
Y. Kobuke, K. Ueda and M. Sokabe, J. Am. Chem. Soc., 1992,
601–4604.
1
14, 7618–7622.
0 M. R. Ghadiri, J. R. Granja and L. K. Buehler, Nature, 1994, 369,
01–304.
11 N. Voyer and M. Robataille, J. Am. Chem. Soc., 1995, 117,
599–6600.
1
3
1
322A series interface and Axoscope 9.0 software) were
6
normalized to this value.
1
2 H. Wagner, K. Harms, U. Koert, S. Meder and G. Boheim, Angew.
Chem., Int. Ed. Engl., 1996, 35, 2643–2646.
Computational details. All calculations were performed
using the Gaussian 03 suite of programs. The B3LYP as one
of Density Functional Theory methods was used for all
geometry optimization. The method combines Beck’s three-
1
3 N. Sakai, K. C. Brennan, L. A. Weiss and S. Matile, J. Am. Chem.
Soc., 1997, 119, 8726–8727.
14 (a) A. Nakano, Q. Xie, J. V. Mallen, L. Echegoyen and G. W.
Gokel, J. Am. Chem. Soc., 1990, 112, 1287–9; (b) O. Murillo, S.
Watanabe, A. Nakano and G. W. Gokel, J. Am. Chem. Soc., 1995,
117, 7665–7679; (c) G. W. Gokel, Chem. Commun., 2000, 1–9.
15 R. Dutzler, E. B. Campbell, M. Cadene, B. T. Chait and R.
3
5
parameter functional, with the nonlocal correlation provided
3
6
by the correlation functional of Lee, Yang, and Parr, which
has been shown to be suitable for vibrational frequency
calculations and geometry optimization. We performed the
restricted B3LYP with 6-31+G* basis set. For each molecule,
several reasonable conformations were calculated to compare
the final energies of stable conformations and the one having
the lowest energy was chosen as the final structure. For the
chosen structure, all geometries were completely optimized in
gas phase.
MacKinnon, Nature, 2002, 415, 287–294.
6 D. A. Doyle, J. M. Cabral, R. A. Pfuetzner, A. Kuo, J. M. Gulbis,
S. L. Cohen, B. T. Chait and R. MacKinnon, Science, 1998, 280,
1
6
17 D. P. Wallace, J. M. Tomich, J. W. Eppler, T. Iwamoto, J. J.
9–77.
Grantham and L. P. Sullivan, Biochim. Biophys. Acta, 2000, 1464,
6
9–82.
1
8 P. H. Schlesinger, R. Ferdani, J. Liu, J. Pajewska, R. Pajewski, M.
Saito, H. Shabany and G. W. Gokel, J. Am. Chem. Soc., 2002, 124,
1848–9.
1
9 V. Sidorov, F. W. Kotch, G. Abdrakhmanova, R. Mizani, J. C.
Fettinger and J. T. Davis, J. Am. Chem. Soc., 2002, 124,
2
267–2278.
0 J. M. Mahoney, G. U. Nawaratna, A. M. Beatty, P. J. Duggan and
Conclusions
2
2
2
1
8
The original SAT design incorporated diglycolic acid for two
reasons. First, its overall size and polarity corresponded well
to the glyceryl unit present in phospholipid monomers. Sec-
ond, it was a convenient synthetic module that was easy to
B. D. Smith, Inorg. Chem., 2004, 43, 5902–7.
1 N. Madhavan, E. C. Robert and M. S. Gin, Angew. Chem., Int.
Ed., 2005, 44, 7584–7587.
2 M. Maduke, C. Miller and J. A. Mindell, Annu. Rev. Biophys.
Biomol. Struct., 2000, 29, 411–438.
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1960–1972 | 1971

View Article Online
2
2
3 O. Scheel, A. A. Zdebik, S. Lourdel and T. J. Jentsch, Nature,
005, 436, 424–7.
4 A. Accardi, M. Walden, W. Nguitragool, H. Jayaram, C. Williams
and C. Miller, J. Gen. Physiol., 2005, 126, 563–70.
5 C. Miller, Science, 2006, 312, 534–5.
6 C. Miller, Nature, 2006, 440, 484–9.
7 D. P. Tieleman, I. H. Shrivastava, M. R. Ulmschneider and M. S.
Sansom, Proteins, 2001, 44, 63–72.
8 O. Beckstein, P. C. Biggin, P. Bond, J. N. Bright, C. Domene, A.
Grottesi, J. Holyoake and M. S. Sansom, FEBS Lett., 2003, 555,
N. Djedovic, R. Ferdani, E. Harder, J. Pajewska, R. Pajewski, M.
E. Weber, P. H. Schlesinger and G. W. Gokel, New J. Chem., 2005,
29, 291–305.
2
31 (a) P. H. Schlesinger, R. Ferdani, J. Pajewska, R. Pajewski and G.
W. Gokel, New J. Chem., 2003, 27, 60–67; (b) R. Ferdani, R.
Pajewski, N. Djedovic, J. Pajewska, P. H. Schlesinger and G. W.
Gokel, New J. Chem., 2005, 29, 673–680.
32 R. Ferdani, R. Pajewski, J. Pajewska, P. H. Schlesinger and G. W.
Gokel, Chem. Commun., 2006, 439–441.
33 R. Pajewski, R. Garcia-Medina, S. L. Brody, W. M. Leevy,
P. H. Schlesinger and G. W. Gokel, Chem. Commun., 2006,
329–331.
34 R. Pajewski, R. Ferdani, J. Pajewska, R. Li and G. W. Gokel
Cation, J. Am. Chem. Soc., 2005, 127, 18281–18295.
35 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
36 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785–789.
2
2
2
2
2
3
85–90.
9 N. Djedovic, R. Ferdani, E. Harder, J. Pajewska, R. Pajewski, P.
H. Schlesinger and G. W. Gokel, Chem. Commun., 2003,
2862–2863.
0 (a) P. H. Schlesinger, N. K. Djedovic, R. Ferdani, J. Pajewska, R.
Pajewski and G. W. Gokel, Chem. Commun., 2003, 308–309; (b)
1
972 | New J. Chem., 2007, 31, 1960–1972
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007