Effect of crown ether ring size on binding and fluorescence response to saxitoxin in anthracylmethyl monoazacrown ether chemosensors

Article abstract of DOI:10.1139/V06-093

Convenient macrocyclization synthetic routes for the preparation of different-sized monoaza anthracylmethyl crown ether chemosensors (15-crown-5, 18-crown-6, 21-crown-7, 24-crown-8, and 27-crown-9) are described. Evaluation of these crowns as chemosensors for saxitoxin revealed that the larger crowns have moderately higher binding constants, with the 27-crown-9 chemosensor having the largest binding constant (2.29 × 10⁵ (mol/L)^-1). Fluorescence enhancements of 100% were observed at saxitoxin concentrations of 5 μmol/L, which is close to the detection limit in mouse bioassay.

Full text of DOI:10.1139/V06-093

1273
Effect of crown ether ring size on binding and
fluorescence response to saxitoxin in
anthracylmethyl monoazacrown ether
chemosensors1
Hua Mao, John B. Thorne, Jennifer S. Pharr, and Robert E. Gawley
Abstract: Convenient macrocyclization synthetic routes for the preparation of different-sized monoaza anthracylmethyl
crown ether chemosensors (15-crown-5, 18-crown-6, 21-crown-7, 24-crown-8, and 27-crown-9) are described. Evalua-
tion of these crowns as chemosensors for saxitoxin revealed that the larger crowns have moderately higher binding
constants, with the 27-crown-9 chemosensor having the largest binding constant (2.29 × 105 (mol/L)–1). Fluorescence
enhancements of 100% were observed at saxitoxin concentrations of 5 µmol/L, which is close to the detection limit in
mouse bioassay.
Key words: anthracene, crown ethers, saxitoxin, paralytic shellfish poison (PSP), binding constants; chemosensors.
Résumé : On décrit des voies de synthèse pratiques à base de macrocyclisations pour la préparation de senseurs
chimiques à base des monoaza anthracylméthyl éthers couronnes de tailles différentes, 15-courone-5, 18-couronne-6,
21-couronne-7, 24-couronne-8 et 27-couronne-9. L’évaluation de ces couronnes comme senseurs chimiques pour la
saxitoxine révèle que les couronnes les plus grosses présentent les constantes de fixation les plus élevées, alors que
le senseur chimique à base de 27-couronne-9 est associée à la constante de fixation la plus élevée, soit 2,29 × 10–5 (mol/L)–1.
On a observé des augmentations de la fluorescence de 100 % à des concentrations de 5 µmol/L, valeur qui est proche
de la limite de détection dans les bioessais à l’aide de souris.
Mots clés : anthracène, éthers couronnes, saxitonine, intoxication paralysante par les mollusques, constantes de fixation,
senseurs chimiques.
[Traduit par la Rédaction] Mao et al. 1279
Harmful algal blooms (HABs) are linked to many cases of
human poisoning each year. Economic losses and costs to
the fishing industry, public health, and tourism are estimated
to be $40 000 000US annually as of the late 1990s (1). The
toxins produced by HABs, as well as many other analytes of
interest to members of the biomedical community, are
“small molecules”. Saxitoxin (STX) (Fig. 1) is such a small
molecule because of its neurotoxicity and is the most toxic
component of the paralytic shellfish poisons (PSPs) (2). Its
activity is manifested through the binding of the toxin to
voltage-gated sodium channels and blocking sodium ion
transport across neuronal membranes. In 1996, saxitoxin
was included on the United States government’s list of
“Select Agents” (potential terrorist weapons). It is one of
only three small molecules on the list; the rest are viruses
(e.g., Ebola), bacteria (e.g., Yersenia pestis), and proteins
(e.g., ricin, abrin).
Mouse bioassay is the current method used by govern-
ment agencies to detect saxitoxin and its derivatives (3), but
for both ethical and economic reasons, an alternative would
be highly beneficial. We have been working for several years
to develop fluorescent chemosensors for the detection of
saxitoxin and have recently made significant advances. Spe-
cifically, we have shown that arylmethylcrowns are selective
for the detection of saxitoxin over sodium, potassium, and
calcium ions (4), as well as several organic analytes (5), in-
cluding tetrodotoxin (Fig. 1) (6). The latter point is clinically
relevant since saxitoxin and tetrodotoxin bind, competitively,
to the same site on voltage-gated sodium channels and pro-
duce the same clinical symptoms (7, 8).
Received 11 November 2005. Accepted 24 January 2006. Published on the NRC Research Press Web site at http://canjchem.nrc.ca
on 20 July 2006.
This paper is dedicated to Alfred Bader in recognition of his many contributions, both personal and professional, to the chemical
enterprise.
H. Mao, J.B. Thorne, J.S. Pharr, and R.E. Gawley.2 Department of Chemistry and Biochemistry, University of Arkansas,
Fayetteville, AR 72701, USA.
1This article is part of a Special Issue dedicated to Dr. Alfred Bader.
2Corresponding author (e-mail: bgawley@uark.edu).
Can. J. Chem. 84: 1273–1279 (2006)
doi:10.1139/V06-093
© 2006 NRC Canada

1274
Can. J. Chem. Vol. 84, 2006
Fig. 1.
H2N
O
O–
O
H
N
1
2
3
4
5
, n = 1, 15-crown-5
, n = 2, 18-crown-6
, n = 3, 21-crown-7
, n = 4, 24-crown-8
, n = 5, 27-crown-9
OH
H NH2
O
HN
O
O
O
+
NH2
+
HO
+H2N
N
N
H
OH
N
N
N
O
H
HO
HO
O
HO
OH
Saxitoxin, STX
Scheme 1.
n
Tetrodotoxin, TTX
n = 1, 2, 3, 4, 5
TsCl, Et3N
O
OTs
O
OH
TsO
O
HO
O
n
n
DMAP, CH 2Cl2
12, n = 3, 97%
13, n = 4, 90%
14, n = 5, 97%
9, n = 3
10, n = 4
11, n = 5
Bn
NaH, THF
65 °C, 1 day
N
HO
OH
15
Cl
O
O
O
O
3, n = 3, 89%
4, n = 4, 84%
5, n = 5, 78%
H2, Pd-C
MeOH
O
n
O
NH
N
Bn
Cs2CO3
MeCN
80 °C, 2 h
O
O
n
16, n = 3, 33%
17, n = 4, 26%
18, n = 5, 25%
6, n = 3, 93%
7, n = 4, 95%
8, n = 5, 91%
Until now we have focused on the 18-crown-6 monoaza-
or diaza-crown ethers, with anthracene, coumarin, or
acridine fluorophores. In early work, Cram and co-workers
showed that 27-crown-9 crown ether is the optimum size to
host a guanidinium guest (9, 10). Since saxitoxin (Fig. 1) is
a bisguanidinium dication, we explored larger crown sizes to
determine the effect on binding, using the anthracene
fluorophore as sensor. Thus, crown ethers 1–5 were prepared
and evaluated for fluorescence response to saxitoxin: 15-
crown-5 (1, n = 1), 18-crown-6 (2, n = 2), 21-crown-7 (3,
n = 3), 24-crown-8 (4, n = 4), and 27-crown-9 (5, n = 5)
(Fig. 1).
available. Diol 11 was prepared in three steps from
diethylene glycol and triethylene glycol according to a liter-
ature procedure (12) with some modifications (using benzyl
instead of allyl as the protecting group). Ditosylation of
polyethylene glycols 9–11 afforded ditosylates 12 (13), 13
(14), and 14, respectively, which were cyclized with N-
benzyldiol 15 using sodium hydride to afford the N-benzyl
crowns 16–18 in modest yields. Hydrogenolysis of 16–18 to
crowns 6–8 proceeded smoothly and alkylation with
chloromethyl anthracene afforded the sensors 3–5. The syn-
thetic steps all proceeded in excellent yields with the excep-
tion of the macrocyclization, which is typically low-yielding
(10).3
Synthesis
Titrations
Anthracylmethyl crown ethers 1 and 2 were originally
prepared by de Silva and co-workers (11). They were made
by alkylation of monoazacrowns with 9-(chloro-
methyl)anthracene. Anthracylmethyl crowns 3–5 were simi-
larly prepared by alkylation of monoazacrowns 6–8, as
shown in Scheme 1. The diols 9 and 10 are commercially
A solution of each crown ether in methanol was titrated
against saxitoxin using the fluorescence response of
anthracene at an excitation wavelength of 372 nm and an
emission wavelength of 420 nm. The concentration of the
crown ethers was held constant at 10–6 mol/L, while the
3 Supplementary data (1H and 13C NMR spectra for compounds 3–8 and 12–18 and typical binding isotherms for the titration of crowns 1–4)
for this article are available on the journal Web site (http://canjchem.nrc.ca) or may be purchased from the Depository of Unpublished Data,
Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada. DUD 5060. For more information on ob-
taining material refer to http://cisti-icist.nrc-cnrc.gc.ca/irm/unpub_e.shtml.
© 2006 NRC Canada

Mao et al.
1275
Fig. 2. Typical binding isotherm for titration of saxitoxin from
concentration of the toxin was varied by successive dilutions
with a 10–6 mol/L solution of the crown ether. Beginning
with 150 µL of a solution of 1 µmol/L crown sensor (1–5)
and a starting STX concentration of 100 µmol/L, binding
isotherms were obtained by successive removal of 50 µL
aliquots and replacement with 50 µL of a 1 µmol/L crown
solution. Fluorescence measurements for each dilution were
made after a 4 min equilibration. Binding constants for each
anthracylmethyl crown ether were obtained by curve fitting
of the intensity at 420 nm to a standard rectangular hyper-
bolic equation for 1:1 binding (15). The binding constant
was determined according to the equation
10–4 mol/L to 5×10–7 mol/L vs. anthracylmethyl 27-crown-9
chemosensor 5 (10–6 mol/L). A 50% fluorescence enhancement
is observed at 1 µmol/L [STX] and 100% enhancement is ob-
served at 5 µmol/L [STX].
700
600
500
5 µmol/L
400





k11
1 +
K [STX]
11
k
crown 
1 µmol/L
F
F
300
=
1 + K11[STX]
0
F0
1.0 × 10–5
200
2.0 × 10–5
3.0 × 10–5
4.0 × 10–5
5.0 × 10–5
where F and F0 are the observed fluorescence intensities in
the presence and absence of STX, respectively, kcrown and k11
are constants related to fluorescence intensities of the crown
and the 1:1 crown·STX complex, respectively, K11 is the
binding constant for the 1:1 complex, and [STX] is the equi-
librium concentration of unbound saxitoxin (15). The equa-
tion describes a hyperbolic binding isotherm and the binding
constant (K11) was obtained by a nonlinear least-squares
curve-fitting program. One of the isotherms for saxitoxin
binding to 5 is shown in Fig. 2. The binding constants (K11)
for the five crown sensors are listed in Table 1. Each value is
the average of at least two runs and each run had a correla-
tion coefficient for the least-squares fit of ≥0.98.
0.0
[STX] (mol/L)
Table 1. Binding constants of saxitoxin to crown sensors 1–5 in
methanol.
Crown sensor
Binding constant (Kb, (mol/L)–1)
1
2
3
4
5
4.93×104
5.3×104
4.68×104
1.07×105
2.29×105
Note: Each K11 is the average of at least two runs. Correlation coeffi-
cients of ≥0.98 were calculated for each fit.
Discussion
The sensing mechanism is illustrated by the following
equilibrium
Fig. 3. Molecular mechanics model of STX docked to anthracyl-
methyl 18-crown-6 (global minimum) (5). Note the absence of a
hydrogen bond to the benzylic nitrogen.
K11
Toxin + Crown
Toxin·Crown
where K11 is the binding (equilibrium) constant for forma-
tion of the 1:1 Toxin·Crown complex. In sensors such as 1–
5, the fluorescent chromophore having a benzylic nitrogen is
only weakly fluorescent owing to photoinduced electron
transfer (PET) (16). Upon complexation of the toxin, PET is
turned off and the chromophore fluoresces normally. The
usual mechanism invoked for this type of PET quenching is
complexation of a ligand to the benzylic nitrogen lone pair.
The binding model that we believe is operative in saxitoxin
sensing involves hydrogen bonding of the crown ether to one
of the guanidiniums (C-8, Fig. 1), as shown in Fig. 3 (5).
Monte Carlo searching of possible docked structures failed
to identify a low energy structure having a hydrogen bond to
the benzylic nitrogen (5). Note that the second guanidinium
appears to π stack with the anthracene fluorophore. We sug-
gest that this π stacking perturbs the relative energies of the
chromophore HOMO and the nitrogen lone pair, thus “turn-
ing off” the PET (5). This hypothesis is supported by the
fact that sodium, potassium, calcium (4), guanidinium (5,
17), and ammonium ions (5, 17), all of which are known to
bind to crown ethers, produce no fluorescence enhancement
with these sensors in alcohol solvents containing small
amounts of water. Further support is found in the failure of
tetrodotoxin (Fig. 1), a toxin having many hydrogen bond
donors but only one guanidinium and which binds competi-
tively to saxitoxin in sodium channels to produce any fluo-
rescence enhancement (6).
Three factors can contribute to an increase in fluorescence
intensity upon binding of the toxin: (1) a large binding con-
stant, which increases the relative concentration of the fluo-
rescing Toxin·Crown complex over the unbound Crown; (2)
a chromophore having a high molar absorptivity (extinction
coefficient); and (3) a high fluorescence quantum yield, such
that a high fraction of incident light is absorbed and a high
fraction of absorbed light is emitted when PET is turned off.
Anthracene has one of the highest fluorescence quantum
© 2006 NRC Canada

1276
Can. J. Chem. Vol. 84, 2006
yields known (-0.4–0.8, depending on conditions). The fluo-
rescence enhancement by saxitoxin in these sensors has not
shown significant dependence on the chromophore
(anthracene (5, 17), coumarin (4), or acridine (6)), so our ef-
forts have focussed primarily on increasing the binding con-
stant by making modifications to the crown ether.
Methanol, acetonitrile, and ethyl acetate were used as re-
ceived. The hexane used in column chromatography was
distilled before use. Water refers to high purity water that
was obtained from the Milli-Q purification system. All reac-
tions were performed under a nitrogen atmosphere.
In a full paper (5), we reported the synthesis and binding
constants of 11 anthracylmethyl 18-crown-6 ethers, 10 of
which were diazacrowns with additional substituents added
to the crown ring — opposite the anthracylmethyl group —
in the hope of increasing the binding constant. Although
replacement of one oxygen with a nitrogen doubled the
binding constant, further substitution failed to produce any
improvement.
N-Benzyl monoaza-21-crown-7 (16)
O
Bn
N
O
O
O
O
O
In early work on the binding of guanidinium ions to
crown ethers, molecular models were used to postulate six
hydrogen bonds between guanidinium ion and 27-crown-9
ethers (9, 10). The hypothesis was supported by the solu-
bilization of (otherwise insoluble) guanidinium ion in chlo-
roform by benzo-27-crown-9. The binding model invoked
hydrogen bonding between the six guanidinium N-Hs and
six of the nine oxygens of the crown ether. Although this
binding model cannot be employed for saxitoxin, the large
number of hydrogen bond donors in the toxin prompted the
question whether a larger crown might show enhanced bind-
ing. Interestingly, neither ammonium ion nor guanidinium
ion produced any fluorescence enhancement in 2 in ethanol
(5, 17). The results of the current study reveal increased
binding as the size of the crown ring increases. A control
experiment shows that guanidinium ion fails to enhance the
fluorescence of 5; if anything, a slight suppression is observed
in methanol.
To a solution of pentaethylene glycol (5.00 g, 21.0 mmol)
in CH2Cl2 (120 mL) was added Et3N (8.8 mL, 63.0 mmol),
DMAP (1.28 g, 10.5 mmol), and TsCl (8.41 g, 44.1 mmol)
at 0 °C. After stirring at RT for 1 h, NH4Cl (satd. aq. sol.,
50 mL) was added. The reaction mixture was extracted three
times with CH2Cl2 (3 × 20 mL), the organic phase was dried
with MgSO4 and evaporated under reduced pressure. The
residue was purified on silica gel using EtOAc–hexane
(50%:50%), which gave 12 as an oil (11.2 g, 97%).
To a solution of bistosylate 12 (11.2 g, 20.5 mmol) in
anhydr. THF (150 mL) was added N-benzyl diol 15 (4.00 g,
20.5 mmol) and NaH (2.46 g, 61.4 mmol; 60% dispersion in
mineral oil, the oil was removed from the product
chromatographically). The mixture was heated at 65 °C for
18 h. After cooling to RT, NH4Cl (satd. aq. sol., 100 mL)
was added. The reaction mixture was extracted three times
with CH2Cl2 (3 × 30 mL). The organic phase was dried with
MgSO4, evaporated under reduced pressure, and the residue
was purified on alumina using EtOAc–hexane (50%:50%),
which gave an oil (2.69 g, 33%). IR (CHCl3, cm–1) νmax
:
Summary
1
3523, 2870, 1643, 1453, 1124. H NMR (300 MHz, CDCl3)
δ: 2.78 (t, 4H, J = 5.8 Hz, CH2), 3.57–3.68 (m, 26H, CH2),
7.19–7.33 (m, 5H, CH-Ar). 13C NMR (75 MHz, CDCl3) δ:
53.78 (CH2), 59.66 (CH2-Bn), 70.00, 70.63, 70.78, 70.83,
70.90 (CH2), 126.78, 128.11, 128.83 (CH-Ar), 139.71 (C-
Anthracylmethyl crown ether chemosensors show en-
hanced binding over smaller crowns. This may be due to the
fact that there are more heteroatoms in the crown and (or)
that the larger crown is simply more flexible. Either effect
could offer more sites for hydrogen bonding to the toxin.
Our results show that simple chemosensors such as these
could be developed for the detection of saxitoxin and possi-
bly other PSP toxins, at concentrations comparable to those
at which mice are sensitive.
Ar). MS m/e: 398 [M+
+ 1]. HRMS calcd. for
C21H36NO6 398.2542 [MH+]; found: 398.2531. Anal. calcd.
for C21H35NO6: C 63.45, H 8.87; found: C 63.29, H 8.88.
1-Aza-21-crown-7 (6)
O
Experimental section
O
HN
General methods
O
O
IR spectra were recorded as thin films between NaCl
plates or as KBr pellets. H and 13C NMR were recorded at
1
O
O
1
1
300 MHz for H and 75 MHz for 13C. H and 13C NMR
chemical shifts are reported in ppm relative to residual chlo-
roform; coupling constants are reported in Hz. ESI mass
spectra were obtained by flow injection on a quadrupole ion
trap mass spectrometer with methanol as the carrier solvent.
High-resolution ESI mass spectra (HRMS/ESI) were ob-
tained by using direct flow injection on a 9.4 T Fourier
transform mass spectrometer. Dry solvents were freshly
distilled before use: dichloromethane was distilled from
calcium hydride and THF from sodium–benzophenone.
A solution of 16 (2.69 g, 6.76 mmol) in MeOH (50 mL)
containing Pd-C (10%, 200 mg) was stirred under 1 atm H2
(balloon, 1 atm = 101.325 kPa) overnight. The catalyst was
filtered through Celite and washed with MeOH (5 × 10 mL).
After evaporation, the residue was purified on alumina using
MeOH–EtOAc (15%:85%), which gave a colorless oil
(1.94 g, 93%). IR (CHCl3, cm–1) νmax: 3504, 2874, 1648,
1
1460, 1353, 1110. H NMR (300 MHz, CDCl3) δ: 2.71
© 2006 NRC Canada

Mao et al.
1277
(t, 4H, J = 4.9 Hz, CH2), 2.90 (1H, br, NH), 3.51–3.59 (m,
24H, CH2). 13C NMR (75 MHz, CDCl3) δ: 49.25 (CH2),
70.53, 70.57, 70.69, 70.75 (CH2). MS m/e: 308 [M+ + 1].
HRMS calcd. for C14H30NO6: 308.2073 [MH+]; found:
308.2061.
70.50, 70.55, 70.66 (6 × CH2), 127.92 and 129.84 (8 × CH-
Ar), 132.90 and 144.84 (4 × C-Ar). MS m/e: 591 [M+ + 1].
Anal. calcd. for C26H38O11S2: C 52.87, H 6.48; found: C
52.75, H 6.59.
1-Aza-24-crown-8 (7)
1-(Anthracen-9-ylmethyl)-aza-21-crown-7 (3)
O
O
HN
O
O
O
O
N
O
O
O
O
O
O
O
To a solution of bistosylate 13 (19.56g, 33.1 mmol) in
anhydr. THF (150 mL) was added N-benzyl diol 15 (6.45 g,
33.1 mmol) and NaH (3.97 g, 99.3 mmol; 60% dispersion in
mineral oil, the oil was removed from the product
chromatographically). The mixture was heated at 65 °C for
18 h. After cooling to RT, NH4Cl (satd. aq. sol., 100 mL)
was added. The reaction mixture was extracted three times
with CH2Cl2 (3 × 30 mL) and the combined organic phase
was dried with MgSO4, condensed, and the residue was puri-
fied on alumina using EtOAc–hexane (80%:20%), which
gave compound 17 as an oil (3.85 g, 26%).
A solution of 17 (3.85 g, 8.71 mmol) in MeOH (50 mL)
containing Pd-C (10%, 200 mg) was stirred under 1 atm H2
(balloon) overnight. The catalyst was filtered through Celite
and washed with MeOH (5 × 10 mL). After evaporation, the
residue was purified on alumina using MeOH–EtOAc
(15%:85%), which gave a colorless oil (2.91 g, 95%). IR
(CHCl3, cm–1) νmax: 3490, 2872, 1649, 1461, 1352, 1109. 1H
NMR (300 MHz, CDCl3) δ: 2.39 (s, br, NH), 2.72 (t, 4H, J =
5.1 Hz, CH2), 3.51–3.60 (m, 28H, CH2). 13C NMR
(75 MHz, CDCl3) δ: 49.09 (CH2), 70.44, 70.56, 70.66,
70.67, 70.70, 70.80 (CH2). MS m/e: 352 [M+ + 1]. HRMS
calcd. for C16H34NO7: 352.2335 [MH+]; found: 352.2326.
To a solution of 16 (0.301 g, 0.976 mmol) in anhydr.
CH3CN (30 mL) was added 9-(chloromethyl)anthracene
(0.221 g, 0.976 mmol) and Cs2CO3 (0.954 g, 2.93 mmol).
The mixture was heated at 80 °C for 2 h. After cooling to
RT, NH4Cl (satd. aq. sol., 30 mL) was added. The reaction
mixture was extracted three times with CH2Cl2 (3 × 20 mL),
the organic phase was dried with MgSO4 and evaporated un-
der reduced pressure. The residue was purified on SiO2 us-
ing EtOAc–MeOH (90%:10%), to give an oil (0.432 g,
89%). IR (CHCl3, cm–1) νmax: 3487, 2868, 1630, 1451, 1117.
1H NMR (300 MHz, CDCl3) δ: 2.90 (t, 4H, J = 5.7 Hz,
CH2), 3.56–3.73 (m, 24H, CH2), 4.64 (s, 2H, CH2), 7.44–
7.55 (m, 4H, CH-Ar), 8.00 (d, 2H, J = 8.9 Hz, CH-Ar), 8.41
(s, 1H, CH-Ar), 8.60 (d, 2H, J = 8.9 Hz, CH-Ar). 13C NMR
(75 MHz, CDCl3) δ: 51.75 (CH2), 53.86 (CH2), 70.10, 70.44,
70.85 (CH2), 124.80, 125.35, 125.52, 127.41, 128.89 (CH-
Ar), 130.57, 131.41, 131.43 (C-Ar). MS m/e: 498 [M+ + 1].
HRMS calcd. for C29H40NO6: 498.2855 [MH+]; found:
498.2838.
Hexaethylene glycol ditosylate (13)
1-(Anthracen-9-ylmethyl)-aza-24-crown-8 (4)
O
O
O
O
O
O
N
O
To a solution of hexaethylene glycol (3.36 g, 11.9 mmol)
in CH2Cl2 (50 mL) at 0 °C was added Et3N (5.0 mL,
35.7 mmol), DMAP (1.45 g, 5.96 mmol), and TsCl (5.68 g,
29.78 mmol). After stirring at RT for 30 min, NH4Cl (satd.
aq. sol., 20 mL) was added. The reaction mixture was ex-
tracted three times with CH2Cl2 (3 × 20 mL) and the com-
bined organic phase was dried with MgSO4, condensed, and
the residue was purified on silica gel using EtOAc–hexane
(80%:20%) to EtOAc (100%), which gave a colorless oil
(6.34 g, 90%). IR (CHCl3, cm–1) νmax: 2957, 2871, 1596,
To a solution of 7 (0.837 g, 2.38 mmol) in anhydr.
CH3CN (50 mL) was added 9-(chloromethyl)anthracene
(0.450 g, 1.98 mmol) and Cs2CO3 (1.29 g, 3.96 mmol). The
mixture was heated at 80 °C for 2 h. After cooling to RT,
NH4Cl (satd. aq. sol., 30 mL) was added. The reaction mix-
ture was extracted three times with CH2Cl2 (3 × 20 mL) and
the combined organic phase was dried with MgSO4, evapo-
rated, and purified on alumina using EtOAc–hexane
(60%:40%), which gave an oil (0.903 g, 84%). IR
(CHCl3, cm–1) νmax: 3486, 2868, 1636, 1452, 1119. 1H NMR
(300 MHz, CDCl3) δ: 2.90 (t, 4H, J = 5.7 Hz, CH2), 3.55–
1
1450, 1361, 1181. H NMR (300 MHz, CDCl3) δ: 2.41 (s,
6H, CH3-Ts), 3.54 (s, 8H, CH2), 3.56–3.62 (m, 8H, CH2),
3.63–3.66 (m, 4H, CH2), 4.10–4.13 (m, 4H, CH2), 7.30–7.33
(m, 4H, CH-Ts), 7.74–7.77 (m, 4H, CH-Ts). 13C NMR
(75 MHz, CDCl3) δ: 21.61 (CH3-Ts), 68.60, 69.31, 70.45,
© 2006 NRC Canada

1278
Can. J. Chem. Vol. 84, 2006
3.69 (m, 28H, CH2), 4.64 (s, 2H, CH2), 7.43–7.54 (m, 4H,
CH-Ar), 8.00 (d, 2H, J = 8.9 Hz, CH-Ar), 8.40 (s, 1H, CH-
Ar), 8.58 (d, 2H, J = 8.9 Hz, CH-Ar). 13C NMR (75 MHz,
CDCl3) δ: 51.84 (CH2), 53.79 (CH2), 70.08, 70.46, 70.80,
70.82, 70.84, 70.87 (CH2), 124.81, 125.36, 125.52, 127.41,
128.89 (CH-Ar), 130.58, 131.41, 131.43 (C-Ar). MS m/e:
542 [M+ + 1]. HRMS calcd. for C31H44NO7: 542.3118 [MH+];
found: 542.3099.
a colorless oil (4.71 g, 96%). IR (CHCl3, cm–1) νmax: 3525,
2872, 1598, 1457, 1182. 1H NMR (300 MHz, CDCl3) δ: 2.43
(s, 6H, CH3-Ts), 3.56–3.68 (m, 24H, CH2), 4.12–4.15 (m,
4H, CH2), 7.32–7.34 (m, 4H, CH-Ts), 7.76–7.79 (m, 4H,
CH-Ts). 13C NMR (75 MHz, CDCl3) δ: 21.63 (CH3-Ts),
68.63, 69.28, 70.48, 70.52, 70.54, 70.58, 70.70 (CH2),
127.95 and 129.84 (CH-Ar), 132.95 and 144.82 (C-Ar). MS
m/e: 635 [M+ + 1]. Anal. calcd. for C28H42O12S2: C 52.98, H
6.67; found: C 53.16, H 6.59.
1-((2-(2-(2-(2-(2-(2-(2-(benzyloxy)ethoxy)ethoxy)-
ethoxy)ethoxy)ethoxy)ethoxy)ethoxy)methyl)benzene
N-Benzyl monoaza-27-crown-9 (18)
O
O
O
OBn
O
O
BnO
O
O
O
O
O
O
Bn
N
To a solution of triethylene glycol bistosylate (5.02 g,
10.93 mmol, synthesized according to literature procedure
(13) with some modifications) in anhydr. THF (100 mL) was
added 2-(2-(benzyloxy)ethoxy)ethanol (4.50 g, 22.96 mmol)
and NaH (2.20 g, 54.68 mmol; 60% dispersion in mineral
oil, the oil was removed from the product chromato-
graphically). The mixture was heated at 65 °C for 1.5 h. Af-
ter cooling to RT, NH4Cl (satd. aq. sol., 100 mL) was added.
The reaction mixture was extracted three times with CH2Cl2
(3 × 30 mL) and the combined organic phase was dried with
MgSO4, evaporated, and the residue was purified on SiO2
using EtOAc, which gave the product as an oil (4.825 g,
O
O
O
To a solution of bistosylate 14 (2.164 g, 3.41 mmol) in
anhydr. THF (50 mL) was added N-benzyl diol 15 (0.798 g,
4.09 mmol) and NaH (0.545 g, 13.6 mmol; 60% dispersion
in mineral oil, the oil was removed from the product
chromatographically). The mixture was heated at 65 °C for
18 h. After cooling to RT, NH4Cl (satd. aq. sol., 50 mL) was
added. The reaction mixture was extracted three times with
CH2Cl2 (3 × 30 mL) and the combined organic phase was
dried with MgSO4, evaporated, and the residue was purified
on alumina using EtOAc–hexane (80%:20%), which gave an
oil (0.414 g, 25%). IR (CHCl3, cm–1) νmax: 3525, 2871,
1
87%). IR (CHCl3, cm–1) νmax: 3572, 2870, 1454, 1106. H
NMR (300 MHz, CDCl3) δ: 3.62–3.71 (m, 28H, CH2), 4.58
(s, 4H, CH2), 7.26–7.34 (m, 10H, CH-Ar). 13C NMR
(75 MHz, CDCl3) δ: 69.44, 70.59, 70.61, 70.66, 73.23
(CH2), 127.59, 127.74, 128.36 (CH-Ar), 138.29 (C-Ar). MS
m/e: 507 [M+ + 1]. Anal. calcd. for C28H42O8: C 66.38, H
8.36; found: C 66.09, H 8.30.
1
1644, 1456, 1113. H NMR (300 MHz, CDCl3) δ: 2.79 (t,
4H, J = 5.9 Hz, CH2), 3.52–3.71 (m, 34H, CH2), 7.23–7.40
(m, 5H, CH-Ar). 13C NMR (75 MHz, CDCl3) δ: 53.79
(CH2), 59.85 (CH2-Bn), 69.93, 70.53, 70.75 (CH2), 126.84,
128.15, 128.88 (CH-Ar), 139.45 (C-Ar). MS m/e: 486 [M+ +
1]. HRMS calcd. for C25H44NO8: 486.3067 [MH+]; found:
486.3049. Anal. calcd. for C25H43NO8: C 61.83, H 8.93;
found: C 61.81, H 9.06.
Heptaethylene glycol ditosylate (14)
O
O
O
O
O
OTs
OTs
1-Aza-27-crown-9 (8)
O
O
O
O
O
The bisbenzyl ether prepared above (4.825 g, 9.52 mmol)
was dissolved in MeOH (50 mL) containing Pd-C (10%,
300 mg) and stirred under 1 atm H2 pressure (balloon) over-
night. The catalyst was filtered through Celite and washed
with MeOH (5 × 20 mL). After evaporation, the product
(heptaethylene glycol) was obtained as a colorless oil
(2.52 g, 81%) and used without further purification or char-
acterization.
To this heptaethylene glycol (2.52 g, 7.73 mmol) in
CH2Cl2 (50 mL) was added Et3N (3.2 mL, 23.2 mmol),
DMAP (0.472 g, 3.87 mmol), and TsCl (3.09 g,
16.23 mmol) at 0 °C. After stirring at RT for 1 h, NH4Cl
(satd. aq. sol., 20 mL) was added. The reaction mixture was
extracted three times with CH2Cl2 (3 × 30 mL) and the com-
bined organic phase was dried with MgSO4, evaporated, and
the residue was purified on silica gel using EtOAc to afford
HN
O
O
O
O
A solution of 18 (0.369 g, 0.759 mmol) in MeOH (15 mL)
containing Pd-C (10%, 50 mg) was stirred under 1 atm of H2
(balloon) overnight. The catalyst was filtered through Celite
and washed with MeOH (5 × 10 mL). After evaporation, the
residue was purified on alumina using MeOH–EtOAc
(15%:85%), which gave a colorless oil (0.274 g, 91%). IR
(CHCl3, cm–1) νmax: 3452, 2915, 1649, 1461, 1103. 1H NMR
(300 MHz, CDCl3) δ: 2.35 (s, br, NH), 2.80 (t, 4H, J =
5.0 Hz, CH2), 3.58–3.69 (m, 32H, CH2). 13C NMR
(75 MHz, CDCl3) δ: 49.16 (CH2), 70.52, 70.60, 70.71, 70.75
© 2006 NRC Canada

Mao et al.
1279
(CH2). MS m/e: 396 [M+ + 1]. HRMS calcd. for C18H38NO8:
396.2597 [MH+]; found: 396.2580. Anal. calcd. for
C18H37NO8: C 54.66, H 9.43; found: C 54.38, H 9.48.
References
1. J. Tibbets. Environ. Health Persp. 106, A326 (1998).
1-(Anthracen-9-ylmethyl)-aza-27-crown-9 (5)
2. L.M. Botana (Editor). Seafood and freshwater toxins: Pharma-
cology, physiology, and detection. Marcel Dekker, New York.
2000.
3. M.A. Quilliam. J. AOAC Int. 80, 131 (1997).
4. P. Kele, J. Orbulescu, T.L. Calhoun, R.E. Gawley, and R.M.
Leblanc. Tetrahedron Lett. 43, 4413 (2002).
O
O
O
O
N
5. R.E. Gawley, S. Pinet, C.M. Cardona, P.K. Datta, T. Ren, W.C.
Guida, J. Nydick, and R.M. Leblanc. J. Am. Chem. Soc. 124,
13448 (2002).
O
O
O
O
6. R.E. Gawley, M. Shanmugasundaram, J.B. Thorne, and R.M.
Tarkka. Toxicon, 45, 783 (2005); Erratum, 46, 477 (2005).
7. C.Y. Kao. Pharmacol. Rev. 18, 211 (1966).
8. C.Y. Kao. In Molecular basis of drug action. Edited by T.P.
Singer and R.N. Ondarza. Elsevier, Amsterdam. 1981.
pp. 283–297.
9. K. Madan and D.J. Cram. J. Chem. Soc. Chem. Commun. 427
(1975).
10. K.P. Kyba, R.C. Helgeson, K. Madan, G.W. Gokel, T.L.
Tarnowski, S.S. Moore, and D.J. Cram. J. Am. Chem. Soc. 99,
2564 (1977).
11. R.A. Bissell, E. Calle, A.P. de Silva, S.A. de Silva, H.Q.N.
Gunaratne, J.-L. Habib-Jiwan, S.L.A. Peiris, R.A.D.D.
Rupasinghe, T.K.S.D. Samarasinghe, K.R.A.S. Sandanayake,
and J.-P. Soumillion. J. Chem. Soc. Perkin Trans. 2, 1559
(1992).
12. B.A. Burkett and T.H. Chan. Synthesis, 1007 (2004).
13. H.M. Colquhoun, E.P. Goodings, J.M. Maud, J.F. Stoddart,
J.B. Wolstenholme, and D.J. Williams. J. Chem. Soc. Perkin
Trans. 2, 607 (1985).
To a solution of 8 (0.210 g, 0.530 mmol) in anhydr.
CH3CN (20 mL) was added 9-(chloromethyl)anthracene
(0.109 g, 0.482 mmol) and Cs2CO3 (0.314 g, 0.964 mmol).
The mixture was heated at 80 °C for 2 h. After cooling to
RT, NH4Cl (satd. aq. sol., 30 mL) was added. The reaction
mixture was extracted three times with CH2Cl2 (3 × 20 mL)
and the combined organic phase was dried with MgSO4,
evaporated, and the residue was purified on alumina using
EtOAc–hexane (60%:40%), which gave an oil (0.221 g,
1
78%). IR (CHCl3, cm–1) νmax: 3449, 2870, 1646, 1108. H
NMR (300 MHz, CDCl3) δ: 2.89 (t, 4H, J = 5.8 Hz, CH2),
3.50–3.67 (m, 32H, CH2), 4.64 (s, 2H, CH2), 7.44–7.54 (m,
4H, CH-Ar), 7.99 (d, 2H, J = 8.8 Hz, CH-Ar), 8.41 (s, 1H,
CH-Ar), 8.58 (d, 2H, J = 8.8 Hz, CH-Ar). 13C NMR
(75 MHz, CDCl3) δ: 51.94 (CH2), 53.79 (CH2), 70.02, 70.32,
70.51, 70.69 (CH2), 124.82, 125.33, 125.53, 127.42, 128.89
(CH-Ar), 130.52, 131.39, 131.42 (C-Ar). MS m/e: 586
[M+ + 1]. HRMS calcd. for C33H48NO8: 586.3380 [MH+];
found: 586.3354.
14. P. Bandyopadhyay, P. Bandyopadhyay, and S.L. Regen. J. Am.
Chem. Soc. 124, 11254 (2002).
15. K.A. Connors. In Binding constants: The measurement of mo-
lecular complex stability. Wiley-Interscience, New York. 1987.
pp. 339–343.
Acknowledgements
This work was supported by the US EPA Science to
Achieve Results (STAR) program (R82959902 and
GR83238201). We are also grateful to Dr. Sherwood Hall,
US Food and Drug Administration, for a gift of saxitoxin.
Core facilities were funded by the US National Institutes of
Health (P20 R15569) and the Arkansas Biosciences Institute.
16. R.A. Bissell, A.P. de Silva, H.Q.N. Gunaratne, P.L.M. Lynch,
G.E.M. Maguire, C.P. McCoy, and K.R.A.S. Sandanayake.
Top. Curr. Chem. 168, 223 (1993).
17. R.E. Gawley, Q. Zhang, P.I. Higgs, S. Wang, and R.M.
Leblanc. Tetrahedron Lett. 40, 5461 (1999); Erratum, 40, 6135
(1999).
© 2006 NRC Canada