The influence of aromatic residues in hydraphile spacer units: Assay by ion selective electrode methods and in bacteria

Article abstract of DOI:10.1016/j.bmc.2005.01.064

A small library of hydraphiles has been prepared that incorporates either 1,4-phenylenedioxy or 2,6-naphthalenedioxy within the spacer chains. The side chains attached to the distal macrocycles in these tris(macrocyclic) compounds are either n-dodecyl or

Full text of DOI:10.1016/j.bmc.2005.01.064

Bioorganic & Medicinal Chemistry 13 (2005) 3321–3327
The influence of aromatic residues in hydraphile spacer units:
assay by ion selective electrode methods and in bacteria
Adam E. Meyer, W. Matthew Leevy, Robert Pajewski, Iwao Suzuki,
Michelle E. Weber and George W. Gokel*
Departments of Chemistry and Molecular Biology & Pharmacology, Washington University School of Medicine,
660 S.Euclid Avenue, Campus Box 8103, St.Louis, MO 63110, USA
Received 2 December 2004; revised 25 January 2005; accepted 25 January 2005
Abstract—A small library of hydraphiles has been prepared that incorporates either 1,4-phenylenedioxy or 2,6-naphthalenedioxy
within the spacer chains. The side chains attached to the distal macrocycles in these tris(macrocyclic) compounds are either n-dodec-
yl or benzyl. The presence of the arenes subunits significantly affect sodium cation release from vesicles. The efficacy of ion transport
is paralleled by the toxicity of these compounds to Bacillus subtilis.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
In the present study, we sought to examine the possibil-
ity of cation–pi interactions between alkali metal cations
and arenes. Our extensive work in this area during the
past few years10 has confirmed that these ion–dipole
interactions are formidable and predictable.11 The plan
was to incorporate arenes within the hydraphile spacer
units. This would place the arenes within the ion path
inthe regionof lowest polarity. This, inturn, should
maximize the cation–pi interaction, if present. We report
here the preparationof a small library of compounds to
examine this hypothesis. The efficacy of these com-
pounds has been assayed in vitro using phospholipid lipo-
somes and in the bacterium Bacillus subtilis.
The hydraphiles1 are a family of synthetic ion-conduct-
ing channels that function in phospholipid bilayers.
Extensive studies have confirmed their placement and
overall conformation in liposomal bilayers.2 As a result
of these studies, the roles of the spacers, the head
groups, and of the side chains were confirmed, at least
in general terms. It was particularly gratifying when
the function of our central relay3 was confirmed by the
structure of the KcsA protein channel isolated from
Streptomyces lividans,4 a feat that earned MacKinnon
5
the 2003 Nobel Prize inChemistry.
Before the KcsA potassium selective proteinchanel
structure was reported, Kumpf and Dougherty put forth
a hypothesis6 that cation–pi interactions controlled Na+/
K+ selection. This seemed plausible in light of the con-
served GYG sequence that is present in the channelꢀs
selectivity filter. MacKinnon and co-workers showed
by site-directed mutagenesis experiments that the tyro-
sine implicated by Dougherty was not serving this func-
tion.7 Even so, examples of cation–pi interactions have
mounted dramatically during the past decade8 and its
potential as a supramolecular force in biology is
enormous.9
2. Results and discussion
2.1. Compounds used in this study
Six hydraphile channels were used in the present study.
Compounds 1 and 2 have beenpreviously reported and
extensively studied. The remaining four compounds
have not previously been reported. The general strat-
egy12 for the preparationof tris(macrocycles) has been
to monoalkylate 4,13-diaza-18-crown-6 (HhN18NiH).13
The synthesis of 2 begins with the reaction of
HhN18NiH and PhCH2Cl to give PhCH2hN18NiH.
Reactionof this with Br(CH 2)12hN18Ni(CH2)12Br
give the tris(macrocycle) PhCH2(hN18Ni(CH2)12)2-
hN18NiCH2Ph, 2. Compounds 3–6 require construction
*
Corresponding author. Tel.: +1 314 362 9297; fax: +1 314 362 9298/
7058; e-mail: ggokel@molecool.wustl.edu
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.01.064

3322
A.E.Meyer et al./ Bioorg.Med.Chem.13 (2005) 3321–3327
of the inner spacer chains that contain the aromatic
residues.
2.2. Assay of hydraphile efficacy
The iontransport ability of hydraphiles
1–6 was as-
The preparationof 5, for example, was accomplished as
follows. Commercially available 2,6-dihydroxynaphthal-
ene was converted into 2,6-bis(3-bromopropoxy)naph-
thalene by reaction with excess 1,3-dibromopropane
(MeCN, K2CO3, 64%). The O-alkylated naphthol was
contaminated by 2-allyloxy-6-(3-bromopropoxy)naph-
thalene (ꢀ6%, NMR). The mixture was used inthe next
step. 4,13-Diaza-18-crown-6 was then treated with an
excess of the dibromopropoxynaphthalene in the pres-
ence of Na2CO3 and catalytic KI. Br(CH2)3O-
C10H6O(CH2)3hN18Ni(CH2)3OC10H6O(CH2)3Br was
obtained in 40% yield. This dibromide was treated with
2 equiv of N-dodecyl-4,13-diaza-18-crown-6 to give 5
(13%) as a nearly colorless solid, mp 103–104.5 ꢁC. De-
tails are recorded inSection4 and the structures of 1–6
are shown.
sessed by measuring Na+ release from phospholipid ves-
icles. Inearlier studies, sodium cationexchange was
monitored by using the 23Na NMR method of Riddell
and Hayer.14 Recently, we have developed an ion selec-
tive electrode (ISE) methodology that was used inthe
present study.15 The iontransport results were com-
pared with those obtained from biological testing and
the data obtained from the two sources was found to
be remarkably consistent.
2.3. Ion selective electrode assay in phospholipid vesicles
Ionselective electrodes have beenused for many years
to assay the complexationbehavior of crownethers.
The applicationof these methods to vesicular efflux is
more recent. Breukink et al. used a potassium selective
electrode to assay the effect of the antibiotic nisin on
membranes.16 Similar studies were reported by Silber-
17
steinet al. The method we have reported and used in
these studies relies ona sodium selective electrode to
quantitate ion channel activity.
O
O
O
O
Briefly, the vesicles used inthe studies reported here
were prepared by sonication of an aqueous suspension
of 1,2-dioleoyl-sn-phosphocholine followed by filtra-
tionthrough a 0.2 mm filter. The aqueous suspension
contained 750 mM NaCl/15 mM N-2-hydroxyethylpip-
erazine-N0-2-ethanesulfonic acid (HEPES) at pH 7.0
so [Na+]internal = 750 mM. Exchange of the external
buffer solutionwas accomplished by passing the lipo-
somes over a Sephadex G25 column, equilibrated
with sodium-free buffer (750 mM cholineCl/15 mM
HEPES, pH 7.0). Lipid concentration was meas-
ured as reported18 and vesicle size was confirmed
N
N
N
N
N
N
N
N
O
O
O
O
O
O
O
O
1
O
O
O
O
N
N
N
N
O
O
O
O
O
O
O
O
2
3
by using
analyzer.
a
Coulter N4MD submicron particle
O
O
O
O
O
O
O
O
O
O
N
N
N
N
N
N
Sodium release was measured by inserting a combina-
tionNa /pH microelectrode ina disposable beaker that
O
O
O
O
O
O
+
O
O
contained the buffered liposome suspension (lipid con-
centration 0.4 mM, total volume 2 mL). After obtaining
a baseline recording (5 min), the hydraphile was added
as a 2-propanol solution. Voltage changes reflecting
sodium efflux were recorded for 25 min. The vesicles
were lysed by treatment with n-octylglucoside to achieve
a final sodium release value, which was used to norm-
alize the data. Additional details are recorded in
Section4.
N
N
O
O
O
O
O
O
N
N
N
N
O
O
O
O
O
O
O
O
4
5
O
O
O
O
O
O
O
2.4. Assessment of sodium transport rates
N
N
N
N
O
O
N
N
N
N
O
O
O
O
O
O
O
O
O
O
We have previously reported the activity profile for 2,
which shows maximal concentration-dependent release
at [2] = 20 lM inabout 900 s. The release profile for 1
(see Fig. 1) is similar to that previously observed for 2.
Inthe case of 1, however, 72 lM channel (rather than
20 lM) is required to achieve full Na+ release. Com-
pounds 1 and 2 differ only in the identity of the side
arm, but this small structural change significantly
impacts the iontransport ability.
O
O
O
O
O
N
N
N
N
O
O
O
O
O
O
O
O
6

A.E.Meyer et al./ Bioorg.Med.Chem.13 (2005) 3321–3327
3323
rapid and goes to approximately 70% completion in
1500 s. The other channels, 1 and 3–6, show <10% re-
lease inthe presence of 12 lM hydraphile. The bottom
panel of Figure 2 shows Na+ release for 1 and 3–6 at
a fivefold higher concentration of 60 lM.
Compounds 4 and 6 have arenes in both the spacer units
and the side arms. While 4 remains inactive at 60 lM, 6
shows a total release of ꢀ20% at 1500 s. As shownin
Figure 1, hydraphile 1 (no arene subunits) releases
ꢀ95% of Na+ at [1] = 60 lM. Channels 3 and 5 (arene
subunits in the spacer chains only) go to complete re-
lease immediately after addition of the channel solution
to the liposome suspension. Compounds 3 and 5 are sig-
nificantly more active than 1. Release of Na+ was mea-
sured for 1, 3, an d5 at a channel concentration of
36 lM (data not shown) and 3 and 5 demonstrated higher
transport efficacy at this intermediate concentration as
well.
Figure 1. Fractional sodium ion release from phospholipid vesicles
mediated by 1.
The concentrations are, from bottom to top, 12, 24, 36,
48, 60, and 72 lM. In order to better understand the
impact on hydraphile channel activity of arene subunits
within the structure, we monitored the transport activity
of 2–6. The top panel of Figure 2 shows the Na+ release
of 12 lM of 1–6, as monitored by the ISE. Clearly, benz-
yl channel 2 is by far the most active of this small library
of compounds. At this concentration, sodium release is
Qualitatively, the Na+-release efficacy of compounds 1–
6 may be summarized as follows: 2 ꢁ 1, 3–6. Whenonly
1
and 3–6 are compared, the activity order is
3 ꢀ 5 > 1 > 4 ꢀ 6. From data obtained at the higher
concentration, it is clear that activity is affected more
by side-arm identity than by the arene within the spacer
chain. Thus, 3 and 5 have benzene and naphthalene sub-
units, respectively, in their spacer chains but they both
have dodecyl side arms. Likewise, 4 and 6 have benzene
and naphthalene subunits, respectively, in their spacer
chains but they both have benzyl side arms. When there
is no arene within the spacer chains, hydraphile 2, which
has benzyl side arms, is more active than 1, which con-
tains only aliphatic components. This contrasts with the
higher activity of dodecyl-side-armed 3 and 5, which are
more active than the benzyl-side-armed counterparts, 4
and 6.
Hydraphile 1 contains dodecyl units in both the side arm
and backbone, and is expected to be the most flexible of
these three systems. Incorporation of the arene back-
bone, in the form of a benzyl group (3) or a naphthalene
group (5) will decrease the channelꢀs conformational
flexibility. It seems reasonable to expect the more flexi-
ble molecule to better adapt to the membrane environ-
ment. On the other hand, when an ion-conducting
conformation is achieved within the bilayer, the more
rigid arene spacers should help to maintain it. We infer
from the experimentally determined transport rates that
maintaining an active channel-forming conformation is
more important for ion conduction than is the inherent
flexibility of the compound.
Because benzyl-side-armed 2 is more active thandodec-
yl-side-armed 1, we expected 4 and 6 to be more active
than 3 and 5. The evidence suggests that the pair of aro-
matic systems oneach side of the distal macrocycle
inhibits ion transport. Corey–Pauling–Koltun (CPK)
molecular models show that there is sufficient conforma-
tional flexibility in either 4 or 6 to form either a pi–pi
stack betweenthe arenes or a pi-stacked ꢁsandwichꢀ that
includes a cation. The models suggest that 6 is less flex-
ible thanis 4, which may explainthe modest difference in
Figure 2. Comparisonof Na +-release activity from liposomes medi-
ated by 1–6 at concentrations of 12 lM (upper panel) and 60 lM.

3324
A.E.Meyer et al./ Bioorg.Med.Chem.13 (2005) 3321–3327
Table 1. Biological activity of hydraphiles to B.subtilis
recorded at 75 MHz inCDCl unless otherwise stated.
3
Infrared spectra were recorded on a Perkin–Elmer
1710 Fourier Transform Infrared Spectrophotometer
and were calibrated against the 1601 cmꢂ1 band of poly-
styrene. Melting points were determined on a Thomas
Hoover apparatus in open capillaries and are uncor-
rected. Thinlayer chromatographic (TLC) analyses were
performed onaluminum 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 chromato-
graphy columns were packed with activated aluminum
oxide (MCB 80–325 mesh, chromatographic grade, AX
611) or with Kieselgel 60 (70–230 mesh). Chromatotron
chromatography was performed ona HarrisonResearch
Model 7924 Chromatotronwith 2 mm thick circular
plates prepared from Kieselgel 60 PF-254.
Compd No.
MIC (lM)
3
4
5
6
1
8.9
0.5
4.3
activity. Specifically, 6 shows ꢀ20% Na+ release at
1500 s compared to ꢀ2% for 4. Naphthalene is larger
and the spacer containing it is less flexible than the benz-
ene-containing spacer. Thus, 6 should be less flexible
and less able to form an obstructive pi-stacking
interaction.
2.5. Biological assays
All reactions were conducted under dry N2 unless other-
wise stated. All reagents were the best (non-LC) grade
commercially available and were distilled, recrystallized,
or used without further purification, as appropriate.
Molecular distillationtemperatures refer to the oven
temperature of a Kugelrohr apparatus. Combustion
analyses were performed by Atlantic Microlab, Inc.,
Atlanta, GA, and are reported as percents. Where water
is factored into the analytical data, spectral evidence is
presented for its presence.
In previous studies, we noted that many of the hydra-
philes showed toxicity to various organisms including
the bacterium B.subtilis . We have noted an excellent
correlationbetweentrasnport efficacy as assessed in
vitro and toxicity to bacteria cells in vivo.19 We therefore
surveyed the activity of 3–6 to see if a similar correlation
was apparent. The minimum inhibitory concentration
(MIC) data (in lM) are recorded in Table 1.
Remarkably, the same trends in activity as observed in
the ISE transport data can be seen in the toxicity profiles
of 3–6 to B.subtilis . Compounds 3 and 5, which have
arene spacer units and dodecyl side arms, were deter-
mined to have MIC values of 1 and 0.5 mM, respec-
tively, which are comparable to 1. However, whenthe
hydraphile has an arene present in both the spacer unit
and the side arm, the MIC values are at least eightfold
higher.
1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC) was
purchased from Avanti Polar Lipids as chloroform solu-
tions. Hydroxyethylpiperazine-N0-2-ethanesulfonic acid
(HEPES), and the inorganic salts NaCl and cholineCl
were all purchased from Sigma–Aldrich. The water that
was used for all buffer preparationwas of Milli-Q Plus
quality, which is essential to avoid salt contamination
inthe buffer systems. N-Octylglucoside was purchased
from CalBioChem.
3. Conclusions
4.2. N,N0-Bis{12-[N-(N0-dodecyl)-diaza-18-crown-6]-
dodecyl}-diaza-18-crown-6, C12H25hN18Ni(CH2)12-
hN18Ni(CH2)12hN18NiC12H25 (1)
We have incorporated arenes into the spacer chains of a
small library of hydraphiles. Specifically, the arenes are
1,4-phenylenedioxy (3, 4) or 2,6-naphthalenedioxy (5,
6). The side arms are either n-dodecyl (3, 5) or benzyl
(4, 6). The presence of the arenes gives mixed effects
on ion transport but both increases and decreases in
Na+ release from liposomes is paralleled by changes in
toxicity to B.subtilis .
Compound 1 was prepared as previously reported.2a
4.3. N,N0-Bis{N-[12-(N0-benzyldiaza-18-crown-6)dode-
cyl]}-diaza-18-crown-6, PhCH2hN18Ni(CH2)12-
hN18Ni(CH2)12hN18NiCH2Ph (2)
Compound 2 was prepared as previously reported.2a
Apparently, some rigidification of the ion path is bene-
ficial to transport but when arenes are present in both
the spacer and side chain, interactions apparently
between them significantly diminish transport and
toxicity.
4.4. C12H25hN18Ni(CH2)4OC6H4O(CH2)4hN18Ni-
(CH2)4OC6H4O(CH2)4hN18NiC12H25 (3)
4.4.1. 1,4-Bis-(4-chlorobutoxy)benzene. A solutionof
1,4-dichlorobutane (63.50 g, 0.5 mol), hydroquinone
(2.75 g, 25 mmol), K2CO3 (8.30 g, 0.6 mol), in n-PrCN
(100 mL) was heated at reflux for 48 h and then cooled
and filtered. The solids were washed with CHCl3
(2 · 20 mL) and the combined organic material was
4. Experimental
4.1. General
evaporated invacuo. Columnchromatography (SiO
,
2
1H NMR were recorded at 300 MHz inCDCl solvents
and are reported in ppm downfield from internal
95% hexanes/ethyl acetate) gave 1,4-bis(4-chlorobut-
oxy)benzene (6.84 g, 94%) as a colorless solid, mp 78–
79 ꢁC.
3
(CH3)3Si unless otherwise noted. 13C NMR were

A.E.Meyer et al./ Bioorg.Med.Chem.13 (2005) 3321–3327
3325
1H NMR: 1.85–2.00 (8H, m, OCH2CH2CH2CH2Cl),
3.62 (4H, t, J = 6.0 Hz, OCH2CH2CH2CH2Cl), 3.94
(4H, t, J = 6.0 Hz, OCH2CH2CH2CH2Cl), 6.82 (4H, s,
HAr). 13C NMR: 26.7, 29.3, 44.7, 67.5, 115.4, 153.0.
IR (KBr): 2957, 2922, 2878, 1510, 1470, 1455, 1440,
1420, 1400, 1352, 1302, 1283, 1235, 1116, 1049, 1015,
and the combined organics were washed with brine
(50 mL) and concentrated in vacuo. Column chroma-
tography (SiO2, 0–2% Et3N/acetone) gave PhCh2-
hN18Ni(CH2)4OC6H4O(CH2)4hN18Ni(CH2)4OC6H4O-
(CH2)4hN18NiCH2Ph (1.14 g, 70%) as a slightly yellow
1
oil. H NMR CDCl3: 1.52–1.64 (8H, m, NCH2CH2),
823, 774, 738, 722 cmꢂ1
.
1.66–1.80 (8H, m, CH2CH2OPh), 2.55 (8H, m,
NCH2CH2), 2.72–2.84 (24H, m, NCH2CH2O), 3.56–
3.62 (48H, m, NCH2CH2OCH2), 3.65 (4H, s, CH2Ph),
3.88 (8H, t, J = 6.0 Hz, CH2CH2OPh), 6.80 (8H, s,
HAr), 7.20–7.40 (10H, m, CH2PhHAr). 13C NMR:
23.9, 27.3, 29.3, 53.8, 54.0, 55.6, 60.0, 70.0, 70.7, 115.4,
126.8, 128.1, 128.8, 139.7, 153.2. IR (neat): 2942, 2866,
1704, 1509, 1472, 1455, 1353, 1353, 1294, 1230, 1126,
1061, 853, 736, 700. Anal. Calcd for C78H126N6O16: C,
66.73; H, 9.05; N, 5.99. Found: C, 66.52; H, 9.05; N,
6.12.
4.4.2. Cl(CH2)4OC6H4O(CH2)4hN18Ni(CH2)4OC6H4O-
(CH2)4Cl. A solution of 1,4-bis(4-chlorobutoxy)benzene
(4.40 g, 15.1 mmol), 4,13-diaza-18-crown-6 (1.00 g,
3.8 mmol), Na2CO3 (2.12 g, 20 mmol), and KI (30 mg,
0.2 mmol) in n-PrCN (30 mL) was heated to reflux for
72 h, cooled, and filtered. The solids were washed with
CHCl3 (2 · 20 mL) and the combined organic material
was washed with brine (50 mL) and concentrated
invacuo. Columnchromatography (SiO 2, 0–1% Et3N
in50%
heexsa/anceet)o,n gave
Cl(CH
2)4O-
C6H4O(CH2)4hN18Ni(CH2)4OC6H4O(CH2)4Cl (1.20 g,
4.6. C12H25hN18Ni(CH2)3OC10H6O(CH2)3-
1
41%) as a slightly yellow oil. H NMR CDCl3: 1.52–
hN18Ni(CH2)3OC10H6O(CH2)3hN18NiC12H25 (5)
1.64 (4H, m, NCH2CH2), 1.66–1.74 (4H, m,
CH2CH2Cl), 1.82–1.98 (8H, m, CH2CH2OPh), 2.55
(4H, t, J = 6 Hz, NCH2CH2), 2.77 (8H, t, J = 6 Hz,
NCH2CH2O), 3.50–3.65 (20H, overlapping signals due
to CH2CH2Cl and NCH2CH2OCH2), 3.85–3.95 (8H,
m, CH2CH2OPh), 6.80 (8H, s, HAr). 13C NMR: 23.9,
26.7, 27.2, 29.3, 44.7, 54.0, 55.6, 67.6, 68.4, 70.0, 70.7,
115.4, 152.9, 153.3. IR (neat): 2943, 2868, 1509, 1473,
1391, 1353, 1284, 1230, 1122, 1059, 826, 728.
4.6.1. 2,6-Bis(3-bromopropoxy)naphthalene (5A). A mix-
ture of 2,6-dihydroxynaphthalene (3.20 g, 20 mmol) and
K2CO3 (13.8 g, 0.1 mol) inMeCN (300 mL) was heated
at reflux for 1 h and 1,3-dibromopropane (40.4 g,
0.2 mol) was added at once. After 8 h heating, the hot
mixture was filtered and reduced to about 30% of the
previous volume. After a second filtration, CH2Cl2
(100 mL) was added, the mixture was filtered again,
and then evaporated to dryness. Crystallization from
EtOH gave 5A (5.12 g, 64%) as a light orange powder.
1H NMR indicated that this isolated materials con-
tained 6% of allyloxy residue, and used to the next step
without further purification. 1H NMR: 2.378 (4H, quin-
tet, J = 6.0 Hz, BrCH2CH2CH2O), 3.653 (4H, t, J =
6.6 Hz, BrCH2CH2CH2O), 4.202 (4H, t, J = 6.0 Hz,
BrCH2CH2CH2O), 7.10–7.14 (4H, m, aromatics),
7.638 (2H, d, J = 9.6 Hz, aromatics). IR (KBr disk):
3059, 2951, 2932, 2908, 1706, 1605, 1508, 1469, 1433,
1417, 1397, 1337, 1286, 1272, 1256, 1234, 1201, 1166,
1149, 1115, 1026, 964, 914, 805, 785, 852, 785, 693,
4.4.3. Preparation of 3. A mixture of HhN18Ni-
(CH2)11CH3 (0.56 g, 1.30 mmol), [Cl(CH2)4(OC6H4O)-
(0.40 g,
(CH2)4]hN18Ni[(CH2)4(OC6H4O)(CH2)4Cl]
0.518 mmol), NaCO3 (0.55 g, 5.18 mmol), KI (cat.),
and 20 mL of butyronitrile were heated to reflux for
4 days, cooled, filtered (Celite), and evaporated. The resi-
due was dissolved inCHCl (50 mL), washed with 10%
3
aq Na2CO3 (3 · 75 mL), brine (75 mL), and evaporated
(high vacuum). Columnchromatography (SiO 2, 2%
Et3N inacetone) gave 3 (120 mg, 15%) as a waxy, tan
solid, mp 40.5–41.5 ꢁC. 1H NMR: 0.87 (6H, t,
J = 6.6 Hz, CH3), 1.25 (38H, m, CH2(CH2)9CH3), 1.61
(8H, m, NCH2CH2CH2), 1.74 (8H, m, OCH2CH2CH2),
2.56 (12H, m, NCH2CH2CH2), 2.78 (24H, t, J = 6.0 Hz,
NCH2CH2O), 3.6 (48H, t, J = 6.0 Hz, NCH2CH2O,
OCH2CH2O), 3.89 (8H, t, J = 6.0 Hz, OCH2CH2),
6.79 (8H, s, OPhO). 13C NMR: 14.3, 22.9, 24.0,
27.2, 27.4, 27.7, 29.5, 29.8, 32.1, 54.1, 55.7, 56.2, 68.5,
70.1, 70.9, 115.5, 153.3. IR (CHCl3): 730, 770, 825,
925, 1072, 1127, 1230, 1289, 1352, 1469, 1508, 1589,
1675, 2855, 2924, 3045, 3369. Elem. Anal. Calcd for
C88H164N6O17ÆH2O: C, 66.97; H, 10.47; N, 5.32. Found:
C, 67.05; H, 10.34; N, 5.36.
663, 625, 559, 473 cmꢂ1
.
4.6.2. N,N0-Bis[6-(3-bromopropoxy(2-naphthoxy)prop-
yl)]-4,13-diaza-18-crown-6(5B). To a mixture of 4,13-
diaza-18-crown-6 (525 mg, 2.0 mmol), Na2CO3 (7.32 g,
69.1 mmol), and KI (1.65 mg, 0.01 mmol), a warmed
solutionof 5A (4.42 g, 11.0 mmol) in n-PrCN (25 mL)
was added and the resultant suspension was heated at
reflux for 3.5 h, cooled, filtered, and concentrated. Tolu-
ene (5 mL) was added and evaporated to assure the
complete removal of n-PrCN. Residual solid was dis-
solved inmiinmal CH 2Cl2. Columnchromatography
(Al2O3, eluant CH2Cl2, then2% i-PrOH–CH2Cl2) gave
5B (0.73 g, 40%, NMR purity ꢀ80%) as a light yellow
solid. This material was used directly inthe next step
without further purification. 1H NMR: 1.949 (4H, quin-
tet, J = 6.7 Hz, NCH2CH2CH2O), 2.365 (4H, quintet,
4.5. C6H5CH2hN18Ni(CH2)4OC6H4O(CH2)4hN18Ni-
(CH2)4OC6H4O(CH2)4hN18NiCH2C6H5 (4)
A
solutionof Cl(CH 2)4OC6H4O(CH2)4hN18Ni-
(CH2)4OC6H4O(CH2)4Cl (0.60 g, 0.78 mmol), N-benz-
yl-4,13-diaza-18-crown-6 (0.74 g, 2.10 mmol), Na2CO3
(0.8 g, 7.5 mmol), and KI (30 mg, 0.2 mmol) in n-PrCN
(20 mL) was heated at reflux for 72 h, cooled, and fil-
tered. The salts were washed with CHCl3 (2 · 20 mL)
J = 6.1 Hz,
BrCH2CH2CH2O),
2.707
(4H,
t,
J = 7.0 Hz, NCH2CH2CH2O), 2.777 (8H, t, J = 5.8 Hz,
NCH2CH2O), 3.50–3.70 (20H, m, BrCH2CH2,
CH2OCH2), 4.084 (4H, t, J = 6.2 Hz, NCH2CH2CH2O),

3326
A.E.Meyer et al./ Bioorg.Med.Chem.13 (2005) 3321–3327
4.185 (4H, t, J = 5.6 Hz, BrCH2CH2CH2O), 7.06–7.14
(8H, m, aromatics), 7.621 (4H, d, J = 10.0 Hz, aromat-
ics). IR (KBr disk): 3067, 3030, 2927, 2849, 2821,
1605, 1510, 1466, 1396, 1351, 1291, 1270, 1234, 1161,
1143, 1119, 1074, 1030, 967, 936, 853, 816, 692, 625,
4.8. Vesicle preparation
Vesicles were prepared from 1,2-dioleoyl-sn-glycero-3-
phosphocholine by using the reverse evaporation method
of Szoka and Papahadjopoulos.20 The phospholipid
were obtained as CHCl3 solutions, which were dried to
lipid films and stored under vacuum at ambient tempera-
ture. The vesicles were prepared by dissolving a dry lipid
film in0.3 mL diethyl ether and 0.3 mL buffer (750 mM
NaCl/15 mM N-2-hydroxyethylpiperazine-N0-2-ethane-
sulfonic acid (HEPES), pH 7.0). The mixture was soni-
cated for ꢀ20 s to give anopaque solutio.n The
organic solvent was removed under reduced pressure
and the solution was passed through a mini-extruder
containing a 0.2 mm polycarbonate membrane filter.
The residual, external buffer solution was exchanged
for a sodium-free buffer (750 mM cholineCl/15 mM
HEPES, pH 7.0) via passage over a Sephadex G25 col-
umn. Vesicle concentration was measured as reported18
and vesicle size was measured by a Coulter N4MD sub-
micron particle analyzer. The vesicles used in the trans-
port studies had diameters of ꢀ200–250 nm depending
onlipid tail length.
573, 474 cmꢂ1
.
4.6.3. 4,13-Bis{4-[6-(13-dodecyl-4,13-diaza-18-crown-6)-
4-yl-1-oxabutyl]naphth-2-yl-4-oxabutyl}-4,13-diaza-18-
crown-6(5). A mixture of 5B (0.62 g, 0.69 mmol),
N-dodecyl-4,13-diaza-18-crown-6 (0.61 g, 1.41 mmol),
Na2CO3 (2.99 g, 28.1 mmol), and KI (20 mg, 0.12
mmol) in n-PrCN (20 mL) was heated at reflux for
28 h. After cooling and filtering, the solids were ex-
tracted (CH2Cl2, 30 mL), and the combined organic
solutions were concentrated. Toluene was added and
evaporated to assure complete removal of n-PrCN. This
procedure was repeated twice. The solids thus obtained
were takenup inCH 2Cl2 washed with H2O (3 · 15 mL),
dried (MgSO4), and the solvent was evaporated. Col-
umnchromatography (Al O3, CH2Cl2–i-PrOH–MeOH,
2
96:2.4:1.6). Fractions containing 5 were collected and
solvent was evaporated. Alternate crystallization from
95% EtOH or AcOEt afforded pure 5 (143 mg, 13%)
1
4.9. Sodium efflux measurements
Mp 103–104.5 ꢁC; H NMR: 0.878 (6H, t, J = 6.6 Hz,
methyl), 1.18–1.34 (36H, m, alkyl), 1.36–1.48 (4H, br,
NCH2CH2CH2CH2), 1.90–2.00 (8H, br, NCH2-
CH2CH2O), 2.461 (4H, t, J = 6.7 Hz, NCH2CH2CH2-
CH2), 2.65–2.85 (32H, m, NCH2CH2), 3.55–3.75 (48H,
m, CH2OCH2), 4.05–4.10 (8H, m, NCH2CH2CH2O),
7.05–7.15 (8H, m, aromatics), 7.603 (4H, pseudo-d,
J = 8.7 Hz, aromatics). IR (KBr disk): 2918, 2850,
1604, 1510, 1466, 1450, 1396, 1381, 1351, 1292, 1267,
1234, 1160, 1143, 1119, 1073, 1023, 989, 970, 935, 854,
722, 691, 624, 577, 476. Anal. Calcd for
C92H158N6O16: C, 68.88; H, 9.93; N, 5.24. Found: C,
68.97; H, 9.88; N, 5.27.
Sodium cationefflux from liposomes was measured
using a pH/Na+ combination microelectrode (Thermo-
Orion). The electrode was equilibrated in sodium-free
buffer (external buffer) in a small, disposable beaker
while stirring. Vesicles suspended in external buffer were
added to achieve a total lipid concentration of 0.4 mM
and a total solution volume of 2.0 mL. A baseline mea-
surement was recorded for 5 min. The channel solution
was introduced as a 2-propanol solution and Na+ efflux
was monitored for 25 min. The vesicles were lysed with a
10% solutionof n-octylglucoside to achieve total Na+ re-
lease. This final release value was used to normalize all
data (converted from mV to units of concentration).
4.7. C6H5CH2hN18Ni(CH2)3OC10H6O(CH2)3hN18Ni-
(CH2)3OC10H6O(CH2)3hN18NiCH2C6H5 (6)
4.10. Determination of minimum inhibitory concentration
A mixture of HhN18NiCH2Ph (0.43 g, 0.475 mmol),
[Br(CH2)3OC10H6O(CH2)3]hN18Ni[(CH2)3OC10H6O-
(CH2)3Br] (0.59 g, 1.66 mmol), NaCO3 (1.01 g, 9.5
mmol), KI (cat.), and butyronitrile (20 mL) was refluxed
for 2 days. The reactionwas thencooled, filtered (Cel-
ite), and evaporated. Column chromatography (SiO2,
2% Et3N inacetone) followed by crystallizationfrom
acetone gave 6 (98 mg, 14%) as a pink granular solid,
mp 90–91 ꢁC. 1H NMR: 1.89 (8H, d, J = 6.0 Hz,
NCH2CH2CH2O), 2.72 (32H, m, NCH2), 3.51 (52H,
m, CH2OCH2CH2OCH2, NCH2Bz), 4.01 (8H, d, 6.0
Hz, NCH2CH2CH2O), 7.03 (8H, d, J = 9 Hz,
OCCHCHC), 7.23 (10H, m, CH2Ph), 7.51 (4H, d, J =
9 Hz, OCCHC). 13C NMR: 27.5, 52.4, 53.9, 53.9, 54.4,
60.1, 66.2, 70.2, 70.9, 107.2, 119.3, 127.0, 128.3, 129.0,
129.9, 155.6. IR (CHCl3): 623, 661, 699, 734, 801, 850,
912, 973, 1115, 1161, 1234, 1296, 1353, 1394, 1453,
1509, 1604, 2242, 2865, 3027, 3059, 3371 cmꢂ1. Anal.
Calcd for C82H122N6O17ÆH2O: C, 67.19; H, 8.53; N,
5.73. Found: C, 67.22; H, 8.38; N, 5.51.
We determined the minimum inhibitory concentration
(inM) as the lowest twofold dilutionof hydraphile that
prevented bacterial growth, as outlined by the
NCCLS.21 Inshort, 5 · 105 colony forming units of B.
subtilis (JH642 WT) cells were grownat 30 ꢁC in 2 mL
of Luria Bertani (LB) Miller media (10 g/L peptone,
5 g/L yeast extract, 10 g/L NaCl, 100 g/mL ampicillin)
that were twofold serially diluted with hydraphile test
compound. The MIC was taken as the lowest hydraphile
concentration that inhibited growth after 24 h as judged
by visual turbidity.
Acknowledgements
We thank the NIH for a grant (GM-36262) that sup-
ported this work and for Training Grant Fellowships
under grants T32-GM08785 and T32-GM04892 to
M.E.W. and W.M.L., respectively.

A.E.Meyer et al./ Bioorg.Med.Chem.13 (2005) 3321–3327
3327
11. (a) De Wall, S. L.; Meadows, E. S.; Barbour, L. J.; Gokel,
G. W. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 6271–6276;
(b) Meadows, E. S.; De Wall, S. L.; Barbour, L. J.; Gokel,
G. W. J.Am.Chem.Soc. 2001, 123, 3092–3107; (c) Hu, J.;
Barbour, L. J.; Gokel, G. W. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 5121–5126.
12. Murillo, O.; Watanabe, S.; Nakano, A.; Gokel, G. W.
J.Am.Chem.Soc. 1995, 117, 7665–7679.
13. Hernandez, J. C.; Trafton, J. E.; Gokel, G. W. Tetrahe-
dron Lett. 1991, 6269–6272.
References and notes
1. Gokel, G. W. Chem.Commun. 2000, 1–9.
2. (a) Murillo, O.; Abel, E.; Maguire, G. E. M.; Gokel, G.
W. J.Chem.Soc,. Chem.Commun.
1996, 2147–2148; (b)
Murillo, O.; Suzuki, I.; Abel, E.; Gokel, G. W. J.Am.
Chem.Soc. 1996, 118, 7628–7629; (c) Abel, E.; Maguire,
G. E. M.; Meadows, E. S.; Murillo, O.; Jin, T.; Gokel, G.
W. J.Am.Chem.Soc. 1997, 119, 9061–9062.
3. Murray, C. L.; Shabany, H.; Gokel, G. W. Chem.
Commun. 2000, 2371–2372.
4. Doyle, D. A.; Cabral, J. M.; Pfuetzner, R. A.; Kuo, A.;
Gulbis, J. M.; Cohen, S. L.; Chait, B. T.; MacKinnon, R.
Science 1998, 280, 69–77.
14. Riddell, F. G.; Hayer, M. K. Biochem.Biophys.Acta 1985,
817, 313–317.
15. Weber, M. E.; Schlesinger, P. H.; Gokel, G. W. J.Am.
Chem.Soc. 2004, 126, 636–642.
16. Breukink, E.; Kraaij, C. v.; Dernel, R. A.; Siezen, R. J.;
Kuipers, O. P.; Kruijff, B. D. Biochemistry 1997, 36, 6968–
6976.
5. MacKinnon, R. Angew.Chem,. Int.Ed.Engl.
4265–4277.
6. Kumpf, R. A.; Dougherty, D. A. Science 1993, 261, 1708–
1710.
7. Heginbotham, L.; Lu, Z.; Abramson, T.; MacKinnon, R.
Biophys.J. 1994, 66, 1061–1067.
2004, 43,
17. Silberstein, A.; Mirzabekov, T.; Anderson, W. F.; Rozen-
1999, 1461, 103–
berg, Y. Biochim.Biophys.Acta
112.
18. Stewart, J. C. Anal.Biochem. 1980, 104, 10–14.
19. Leevy, W. M.; Weber, M. E.; Schlesinger, P. H.; Gokel,
G. W. Chem.Commun. 2004, 89–91.
8. (a) Ma, J. C.; Dougherty, D. A. Chem.Rev. 1997, 97,
1303–1324; (b) Gallivan, J. P.; Dougherty, D. A. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 9459–9464.
20. Szoka, F.; Papahadjopoulos, D. Proc.Natl.Acad.Sci.
U.S.A. 1978, 75, 4194–4198.
21. National Committee for Clinical Laboratory Standards.
In Methods for Dilution Antimicrobial Susceptibility Tests
for Bacteria that grow Aerobically, 5th ed.; M7-A5
(NCCLS, Wayne, Pennsylvania, 2000).
9. (a) Gokel, G. W.; Wall, S. L. D.; Meadows, E. S. Eur.J.
Org.Chem. 2000, 2967–2978; (b) Gokel, G. W.; Barbour,
L. J.; De Wall, S. L.; Meadows, S. L. Coord.Chem.Rev.
2001, 222, 127–154; (c) Gokel, G. W.; Barbour, L. J.;
Ferdani, R.; Hu, J. Acc.Chem.Res. 2002, 35, 878–886.
10. Gokel, G. W. Chem.Commun. 2003, 2847–2852.