Facilitated catecholamine transport through bulk and polymer-supported liquid membranes

Article abstract of DOI:10.1021/ja9615076

A series of crown boronic acids, 1-4, were synthesized and studied as carriers for catecholamine transport through bulk liquid membranes (BLMs) and supported liquid membranes (SLMs). Carrier 1 greatly facilitated the transport of primary catecholamines through BLMs; whereas, the more lipophilic analogues 3 and 4 were less effective. A combination of kinetic, mass spectral, and NMR evidence suggests that the transported species in BLMs is the cyclic, zwitterionic, 1:1 complex 7. The SLM transport studies used a liquid membrane of 2-nitrophenyl octyl ether supported by a thin, flat sheet of porous polypropylene. In the absence of carrier there was neglible dopamine transport (<5 x 10^-9 mol/m²·s) at pH 7.2. When the membrane contained carrier 3 (33 mM or about 2% wt), facilitated catecholamine transport was observed in the order of dopamine (5 x 10^-7 mol/m²·s) > epinephrine (1.5 x 10^-7 mol/m²·s) > norepinephrine (0.8 x 10^-7 mol/m²·s). SLMs containing carrier 3 were stable, implying that carrier 3 is a very good candidate for transport mechanism studies. Crown boronic acid 4 was an even better transport carrier of primary catecholamines with a transport order of norepinephrine (4.7 x 10^-6 mol/m²·s) > dopamine (3.5 x 10^-6 mol/m²·s) >> epinephrine (3 x 10^-8 mol/m²·s). It is 10 times more effective than an equimolar mixture of boronic acid 5 and crown 6, which is one of the best examples of ditopic cooperativity yet observed in SLM transport. SLMs containing 4, however, did not exhibit long-term stability. Overall, it is possible that a device based on SLMs containing crown boronic acid carriers can be developed to selectively extract catecholamines from clinical samples.

Full text of DOI:10.1021/ja9615076

9820
J. Am. Chem. Soc. 1996, 118, 9820-9825
Facilitated Catecholamine Transport through Bulk and
Polymer-Supported Liquid Membranes
Marie-France Paugam,† Jeffrey T. Bien,† Bradley D. Smith,*,†
Lysander A. J. Chrisstoffels,‡ Feike de Jong,*,‡ and David N. Reinhoudt*,‡
Contribution from the Department of Chemistry and Biochemistry, UniVersity of Notre Dame,
Notre Dame, Indiana 46556, and Department of Chemical Technology, UniVersity of Twente,
7500 AE Enschede, The Netherlands
ReceiVed May 6, 1996X
Abstract: A series of crown boronic acids, 1-4, were synthesized and studied as carriers for catecholamine transport
through bulk liquid membranes (BLMs) and supported liquid membranes (SLMs). Carrier 1 greatly facilitated the
transport of primary catecholamines through BLMs; whereas, the more lipophilic analogues 3 and 4 were less effective.
A combination of kinetic, mass spectral, and NMR evidence suggests that the transported species in BLMs is the
cyclic, zwitterionic, 1:1 complex 7. The SLM transport studies used a liquid membrane of 2-nitrophenyl octyl ether
supported by a thin, flat sheet of porous polypropylene. In the absence of carrier there was neglible dopamine
transport (<5 × 10-9 mol/m2‚s) at pH 7.2. When the membrane contained carrier 3 (33 mM or about 2% wt),
facilitated catecholamine transport was observed in the order of dopamine (5 × 10-7 mol/m2‚s) > epinephrine (1.5
× 10-7 mol/m2‚s) > norepinephrine (0.8 × 10-7 mol/m2‚s). SLMs containing carrier 3 were stable, implying that
carrier 3 is a very good candidate for transport mechanism studies. Crown boronic acid 4 was an even better transport
carrier of primary catecholamines with a transport order of norepinephrine (4.7 × 10-6 mol/m2‚s) > dopamine (3.5
× 10-6 mol/m2‚s) . epinephrine (3 × 10-8 mol/m2‚s). It is 10 times more effective than an equimolar mixture of
boronic acid 5 and crown 6, which is one of best examples of ditopic cooperativity yet observed in SLM transport.
SLMs containing 4, however, did not exibit long-term stability. Overall, it is possible that a device based on SLMs
containing crown boronic acid carriers can be developed to selectively extract catecholamines from clinical samples.
Introduction
automated method of extracting catecholamines from clinical
samples would be desirable.
Catecholamines play an important role in health and disease.
Changes in catecholamine levels have been correlated with
stress, heart disease, changes in blood pressure and thyroid
hormone levels, catecholamine-secreting tumors, neuromuscular
disorders, and various types of mental illness.1,2 Numerous
methods have been proposed for the determination of catechol-
amines and their metabolites in biological fluids such as urine
and plasma. The most widely used are spectrometric, fluoro-
metric, and radioenzymatic methods and HPLC with electro-
chemical detection.1,3 The metabolites are often the primary
assay target because of their higher concentrations and greater
stability. However, in certain cases the concentration levels of
the parent catecholamines are more diagnostic. This means
highly sensitive, specific, and reliable methods are required for
measuring the low amounts of dopamine, norepinephrine, and
epinephrine. In most cases, a preliminary extraction and
purification of the biological sample is necessary. The most
common pretreatments involve alumina adsorption, ion-
exchange, and/or boric acid chromatography.1,3 These methods
are labor-intensive and require a fair amount of technical
expertize to ensure reliable results. A more straightforward and
We are attempting to develop a membrane-based catechol-
amine purification system. Our approach is to design carrier
compounds that can selectively and actively transport catechol-
amines through liquid organic membranes.4 Two types of liquid
membranes were initially considered: (i) bulk liquid membranes
(BLMs), where the aqueous source and receiving phases are
separated by an immiscible organic liquid, and (ii) supported
liquid membranes (SLMs), which have essentially the same
configuration as BLMs but the organic liquid is contained within
the pores of a thin porous polymer.5
Our preliminary studies used BLMs, which were chosen for
a number of reasons.4 BLM transport experiments are com-
paratively cheap and easy to conduct. Also, there are few
restrictions on carrier structure, e.g., BLM carriers are not
required to be highly lipophilic. BLMs, however, are not
practical in an industrial setting and are primarily used as a
screening assay for carrier candidates. Since the final device
* Author to whom correspondence should be addressed.
† University of Notre Dame.
‡ University of Twente.
X Abstract published in AdVance ACS Abstracts, October 1, 1996.
(1) Whitley, R. J.; Meikle, A. W.; Watts, N. B. In Tietz Textbook of
Clinical Chemistry, 2nd ed.; Burtis, C. A., Ashwood, E. R., Eds.;
Saunders: Philadelphia, 1994; pp 1739-1764.
(4) Paugam, M.-F.; Valencia, L. S.; Boggess, B.; Smith, B. D. J. Am.
Chem. Soc. 1994, 116, 11203-11204.
(5) (a) Visser, H. C.; Reinhoudt, D. N.; de Jong, F. Chem. ReV. 1994,
23, 75-82. (b) Liquid membranes: Chemical Applications; Araki, T.,
Tsukube, H., Eds.; CRC Press: Boca Raton, FL, 1990. (c) Fyles, T. M. In
Inclusion Aspects of Membrane Chemistry; Osa, T., Atwood, J. L., Eds.;
Kluwer: Boston, 1991.
(2) Neumeyer, J. L.; Booth, R. G. In Principles of Medicinal Chemistry,
4th ed.; Foye, W. O., Lemke, T. L., Williams, D. A., Eds.; Lea and
Febiger: Philadelphia, 1995; Chapter 13.
(3) (a) Rosano, T. G.; Swift, T. A.; Hayes, L. W. Clin. Chem. 1991, 37,
1854-1867. (b) QuantitatiVe Analysis of Catecholamines and Related
Compounds; Krstulovic, A. M., Ed.; Ellis Horwood: Chichester, 1986.
S0002-7863(96)01507-7 CCC: $12.00 © 1996 American Chemical Society

Facilitated Catecholamine Transport
J. Am. Chem. Soc., Vol. 118, No. 41, 1996 9821
Scheme 1
Scheme 2
was envisioned to be small, robust, and automated, it was
decided that an SLM-based system would be more appropriate.
The large surface areas, well-defined diffusional pathway, and
sturdy membranes associated with SLMs ensure fast and
reproducible transport rates. Furthermore, rates can be con-
trolled by modifying the shape of the polymer support, e.g.,
changing from flat sheets to hollow fibers increases the effective
surface area and consequently raises the total transport rate.5
An initial communication of this work focused on the
remarkable selectivity of carrier 1 (Scheme 1) to transport
dopamine through a BLM.4 In this paper, we present additional
BLM transport studies, as well as a structural elucidation of
the transported species in BLMs using probe compound 2. We
also describe various catecholamine carriers that operate ef-
ficiently in SLMs. Specifically, we have synthesized and
studied the transport abilities of carriers 3-6. This work is the
first example of facilitated catecholamine transport through
SLMs.
Carrier Design and Synthesis
The pKa values of the boronic acid residues in carriers 1-4
(Scheme 1) are all approximately 9.0, so that at neutral pH the
carriers are uncharged.6 Transport of a non-diol-containing
ammonium cation at neutral pH would require cotransport of
an accompanying anion. This is an energetically demanding
process which is very dependent on the lipophilicity of the
anion.7 Condensation of a boronic acid with a catechol unit,
however, produces a boronate ester of greater acidity than the
parent boronic acid, such that at neutral pH the tetrahedral
boronate anion is formed.8,9 The generic structure of a cyclic,
1:1 complex between any of the carriers 1-4 and a primary
catecholamine is the covalent adduct 7, a lipophilic zwitterionic
species that is able to diffuse through a nonpolar membrane.10
Thus, carriers 1-4 were predicted to selectively facilitate
catecholamine transport through a liquid organic membrane.
Carriers 3 and 4 are more lipophilic versions of 1 and were
designed to operate in SLMs. An examination of CPK models
indicated they could assume conformations capable of ditopic
binding with a catecholamine molecule. Compounds 5 and 6
were designed to be used together in an SLM as a carrier
admixture. Both of their structures incorporate (2-nitrophen-
oxy)octyl groups to ensure good solubility in the supported,
2-nitrophenyl octyl ether membrane.11
Carriers 1-4 were synthesized in a straightforward fashion
by metalation of their respective bromo precursors 8, 9, 11, and
12 with butyllithium and subsequent treatment with trimethyl
borate (Scheme 1). Intermediate 11 was prepared by alkylation
of crown 13 (Scheme 2), which in turn was made using the
method of Bartsch.12 Carriers 5 and 6 were synthesized by
alkylating the commercially available precursors with 1-bromo-
8-(2-nitrophenoxy)octane.11
(10) Other examples of synthetic ditopic receptors for catecholamines
include: (a) Imada, T.; Kijima, H.; Takeuchi, M.; Shinkai, S. Tetrahedron
1996, 52, 2817-2826. (b) Kimura, E.; Fujioka, H.; Kodama, M. J. Chem.
Soc., Chem. Commun. 1986, 1158-1159. (c) Saigo, K.; Kihara, N.;
Hashimoto, Y.; Lin, R.; Fujimura, H.; Suzuki, Y.; Hasegawa, M. J. Am.
Chem. Soc. 1990, 112, 1144-1150. (d) Dumont, B.; Schmitt, M.-F.; Joly,
J.-P. Tetrahedron Lett. 1994, 35, 4773-4776. (e) Sutherland, I. O. In
AdVances in Supramolecular Chemistry; Gokel, G. W., Ed.; JAI Press:
Greenwhich, CT, 1990; Vol. 1.
(11) Visser, H. C.; Vink, R.; Snellink-Ruel, B. H. M.; Kokhuis, S. B.
M.; Harkema, S.; de Jong, F.; Reinhoudt, D. N. Recl. TraV. Chim. Pays-
Bas 1995, 114, 285-294.
(12) Charewicz, W. A.; Heo, G. S.; Bartsch, R. A. Anal. Chem. 1982,
54, 2094-2097. Heo, G. S.; Bartsch, R. A.; Schlobohm, L. L.; Lee, J. G.
J. Org. Chem. 1981, 46, 3575-3576.
(6) Torsell, K. In Progress in Boron Chemistry; Steinberg, H., McClos-
key, A. L., Eds.; Pergamon: New York, 1964; p 385.
(7) Lamb, J. D.; Christensen, J. J.; Izatt, S. R.; Bedke, K.; Astin, M. S.;
Izatt, R. M. J. Am. Chem. Soc. 1980, 102, 3399-3403.
(8) (a) Smith, B. D. Supramol. Chem. 1996, 7, 55-60. (b) James, T. D.;
Samankumara Sandanayake, K. R. A.; Shinkai, S. Supramol. Chem. 1995,
6, 141-157.
(9) London, R. E.; Gabel, S. A. J. Am. Chem. Soc. 1994, 116, 2562-
2569.

9822 J. Am. Chem. Soc., Vol. 118, No. 41, 1996
Paugam et al.
dopamine the carrier boron atom has become sp3 hybridized in
both aqueous and organic phases. This is consistent with
structure 17, the fluoro-substituted analogue of 16.
Table 1. Initial Fluxes for Dopamine Transport through BLMs
Containing Different Carriers
fluxb
carriera
(10-7 mol/m2‚s)
1
none
phenylboronic acid
18
0.1
0.3
0.2
6
dicyclohexyl-18-crown-6
dicyclohexyl-18-crown-6 + phenylboronic acid
18-crown-6 + phenylboronic acid
3
3 + dicyclohexyl-18-crown-6
13 + phenylboronic acid
dibenzo-18-crown-6 + phenylboronic acid
4
4
1.6
1.7
0.1
0.1
5
6 + phenylboronic acid
1.4
a Source phase: sodium phosphate (100 mM, pH 7.4), sodium
dithionite (10 mM), dopamine (41 mM). Organic phase: carrier(s) (1
mM) in chloroform. Receiving phase: sodium phosphate (100 mM,
pH 7.4), sodium dithionite (10 mM). Temperature: 293 K. b Initial
dopamine flux extrapolated to t ) 0, (15%.
Transport Studies Using SLMs
The apparatus used in the SLM transport studies has been
described before.14 The liquid membrane was 2-nitrophenyl
octyl ether supported by a thin, flat sheet (15 cm2) of porous
polypropylene (Accurel). Although a flat sheet is not the most
efficient membrane geometry (hollow fibers provide a greater
surface area), its surface area can be measured accurately, which
greatly facilitates the mechanistic interpretation. Typically, the
source and receiving phases (50 mL) contained sodium phos-
phate (100 mM) buffered at pH 7.2 and sodium dithionite (1
mM) as an antioxidant. At the beginning of the transport
experiment the source phase also contained 50 mM catechol-
amine. The initial catecholamine flux in the receiving phase
was determined from the change in UV absorption extrapolated
to t ) 0.
Transport Studies Using BLMs
The dopamine fluxes through a BLM (chloroform) containing
various carrier mixtures are summarized in Table 1. Qualitative
aqueous/chloroform partition measurements with carrier 1
indicate that it is only moderately lipophilic. Nonetheless, the
more lipophilic carriers 3 and 4 are less efficient at promoting
dopamine transport. It appears that in the BLM experiment
there is an advantage to having the carrier partially dissolved
in the source phase where additional complexation can occur
followed by movement of the complexed carrier into the BLM.
The results of the SLM transport studies are shown in Table
2. Control experiments showed no detectable leakage of the
liquid 2-nitrophenyl octyl ether or any of the carriers 3-6 into
the receiving phase. In the absence of carrier there was neglible
dopamine transport (<5 × 10-9 mol/m2‚s) through the SLM.
When the liquid membrane contained crown 6 (33 mM or about
2% wt), a small dopamine flux (4 × 10-8 mol/m2‚s) was
observed which increased 10-fold (4 × 10-7 mol/m2‚s) when
an equimolar mixture of crown 6 and boronic acid 5 was used.
A similar dopamine flux (5 × 10-7 mol/m2‚s) was observed
with carrier 3, which combines the boronic acid and crown
binding groups into a single molecule. While this was an
encouraging result, producing a transport enhancement of >100
times the rate of background diffusion, an even higher flux was
expected as the strategy of covalently attaching two monotopic
carriers and forming a single ditopic carrier should in principle
give rise to improved binding and transport.15 It appeared that
the problem was due to the 19-crown-6 group in 3, which was
likely a poor ammonium binder because of the weakened
Identification of Transported Species in BLMs
Our initial attempts to elucidate the structure of the transported
species in BLMs relied heavily on kinetic and mass spectral
evidence. A combination of FAB and electrospray mass
spectrometry produced strong evidence for structure 16 as the
transported species when carrier 1 was used to transport
dopamine through a layer of chloroform (e.g., negative ion FAB
spectrum of aqueous and organic phases showed m/z ) 564
[16 - H]-).4 Attempts to corroborate this structural assignment
1
using NMR spectroscopy were initially unclear. A H NMR
spectrum of the organic layer produced ambiguous results due
to broadened and overlapping signals, and the solutions were
too dilute for 13C and 11B analyses. Thus, we resorted to a 19
F
NMR method using the fluoro-substituted carrier 2. The method
was based on observations made by London and Gabel that the
19F signal for 4-fluorobezeneboronic acid moves about 7 ppm
upfield when the boronic acid forms a tetrahedral boronate
complex with a catechol derivative.9 This upfield shift reflects
the change in boron hybridization from sp2 to sp3, which is
transmitted to the para-substituted fluorine by resonance de-
localization. The 19F NMR signal for carrier 2 in aqueous
solution (4 mM), buffered at pH 7.4, occurred at -39.1 ppm
relative to external CF3CH2OH.13 Addition of 40 mM dopamine
moved the signal upfield to -45.6 ppm. Extraction of this
solution with chloroform and subsequent 19F NMR analysis of
(14) Stolwijk, T. B.; Sudho¨lter, E. J. R.; Reinhoudt, D. N. J. Am. Chem.
Soc. 1987, 109, 7042-7047. van Straaten-Nijenhuis, W.; de Jong, F.;
Reinhoudt, D. N. Recl. TraV. Chim. Pays-Bas 1993, 112, 317-324.
(15) While there have been many studies of ditopic membrane carriers,
surprisingly few have unambiguously demonstrated an increase in ditopic
cooperativity due to covalent conjugation, i.e., the ditopic receptor is a more
efficient transporter than an equimolar mixture of two appropriate monotopic
half-receptors. See, for example: Rudkevich, D. M.; Mercer-Chalmers, J.
D.; Verboom, W.; Ungaro, R.; de Jong, F.; Reinhoudt, D. N. J. Am. Chem.
Soc. 1995, 117, 6124-6125. Reetz, M. T.; Huff, J.; Rudolph, J.; To¨llner,
K.; Deege, A.; Goddard, R. J. Am. Chem. Soc. 1994, 116, 11588-11589.
Mohler, L. K.; Czarnik, A. W. J. Am. Chem. Soc. 1993, 115, 2998-2999.
Andreu, C.; Gala´n, A.; Kobiro, K.; de Mendoza, J.; Park, T. K.; Rebek, J.;
Salmero´n, A.; Usman, N. 1994, 116, 5501-5502. Kra´l, V.; Andrievsky,
A.; Sesslor, J. L. J. Chem. Soc., Chem. Commun. 1995, 2349-2351. Sesslor,
J. L.; Furuta, H.; Kra´l, V. Supramol. Chem. 1993, 1, 209-220. Schwabacher,
A. W.; Lee, J.; Lei, H. J. Am. Chem. Soc. 1992, 114, 7597-7598. Seel, A.;
Vo¨gtle, F. Angew. Chem., Int. Ed. Engl. 1991, 30, 442-444. Aoyama, Y.;
Asakawa, M.; Yamagishi, A.; Toi, H.; Ogoshi, H. J. Am. Chem. Soc. 1990,
112, 3145-3151.
the organic phase also produced a signal at -45.6 ppm (all 19
NMR spectra are reproduced as supporting information). This
upfield change in 19F chemical shift correlates with the results
of London and Gabel and indicates that after treatment with
F
(13) The sample of carrier 2 used in the 19F NMR studies was
contaminated with 8% of the protiodeboronated analogue 10. This impurity
was not expected to prejudice the outcome of the experiment and in fact
was a benefit because it acted as an internal reference signal (see Supporting
Information).

Facilitated Catecholamine Transport
J. Am. Chem. Soc., Vol. 118, No. 41, 1996 9823
Figure 1. Uphill transport of dopamine from (O) source phase at pH 7.4 or (b) receiving phase at pH 5.5. Initially all aqueous layers contained
0.25 mM dopamine, 0.1 M sodium phosphate buffer, and 1 mM sodium dithionite. (a) BLM experiment with chloroform membrane containing 1
(1 mM). (b) SLM experiment with 2-nitrophenyl octyl ether membrane containing 4 (33 mM). Note the aqueous phase volumes in the SLM
experiment are approximately 15 times greater than those in the BLM experiment.
Table 2. Initial Fluxes for Catecholamine Transport through SLMs
mechanistic studies. Unfortunately, the more efficient carrier
4 did not display good long-term stability, as repeated runs with
the same membrane showed a decline in observed fluxes
(approximately 15% decline with each repetition). As the
transport experiment progressed, the physical appearance of the
SLM containing 4 was observed to change from translucent to
white. HPLC studies showed no evidence of leaching of the
carrier into the aqueous phases. Instead, it seemed that the
carrier-catecholamine complex was slowly precipitating within
the 2-nitrophenyl octyl ether.17
An eventual application of this technology will most likely
require that the catecholamine transport be not only rapid and
selective but also active, i.e., it must be capable of concentrating
the catecholamine in the receiving phase. Moreover, due to
the susceptibility of catecholamines to decompose under basic
conditions, the most practical active transport system would be
one driven by a neutral to acid pH gradient. The binding to
produce complex 7 is an acid-producing equilibrium. Thus, a
neutral to acid pH gradient is predicted to drive catecholamine
transport uphill. Uphill transport studies were conducted using
a BLM containing carrier 1 and an SLM containing carrier 4.
In both cases, uphill transport from a departure phase at pH 7.4
into a receiving phase at pH 5.5 was readily achieved (Figure
1).
Containing Different Carriersa
solute
carriera
fluxb (10-7 mol/m2‚s)
dopamine
none
none
none
3
<0.05
<0.05
0.2
5
norepinephrine
epinephrine
dopamine
norepinephrine
epinephrine
dopamine
norepinephrine
epinephrine
uridine
dopamine
dopamine
dopamine
norepinephrine
epinephrine
3
3
4
4
4
4
5
6
0.8
1.5
35
47
0.3
<0.1
<0.1
0.4
4
3
0.8
5 + 6
5 + 6
5 + 6
a Transport through a liquid membrane supported by a thin, flat sheet
of Accurel. Source phase: sodium phosphate (100 mM, pH 7.2), sodium
dithionite (1 mM), dopamine (50 mM). Liquid membrane: carrier(s)
(33 mM or about 2% wt) dissolved in 2-nitrophenyl octyl ether.
Receiving phase: sodium phosphate (100 mM, pH 7.2), sodium
dithionite (1 mM). Temperature: 298 K. b Initial catecholamine flux
extrapolated to t ) 0, (10%.
basicity of its four phenoxy oxygens.16 Consequently, we were
very pleased to find that ditopic carrier 4, which contains an
18-crown-6 group, induced an impressive dopamine flux of 3.5
× 10-6 mol/m2‚s. This is 1 order of magnitude more effective
than the 5/6 mixture and is one of best examples of ditopic
cooperativity yet observed in SLM transport.15
The abilities of 3, 4, and 5/6 to transport epinephrine and
norepinephrine were also determined (Table 2). The transport
order with carrier 4 and with carrier mixture 5/6 was nor-
epinephrine ∼ dopamine . epinephrine. This is in reasonable
agreement with the order of BLM transport rates found with
carrier 1 and was expected because 18-crown-6 groups are
selective for primary ammonium cations over secondary am-
monium cations.4 The transport order with the less efficient
carrier 3 was dopamine > epinephrine > norepinephrine. The
change in selectivity is attributed to the weaker ammonium
binding ability of the 19-crown-6 moiety in 3, which means
the lipophilicity of the catecholamine plays a more important
role in controlling transport permeability (as reflected by the
background fluxes in Table 1, epinephrine is more lipophilic
than norepinephrine).
Conclusions
1. Crown boronic acid 1 greatly enhances the transport of
primary catecholamines through BLMs;4 whereas, the more
lipophilic carriers 3 and 4 are less effective. It appears that in
the BLM transport experiment there is an advantage to having
the carrier partially dissolved in the source phase where
additional complexation can occur followed by movement of
the complexed carrier into the BLM.
2. A combination of kinetic, mass spectral, and NMR
evidence strongly suggests that the transported species in a
chloroform membrane is the cyclic, 1:1 complex 7.17
3. Crown boronic acid 3 greatly facilitates the transport of
catecholamines through SLMs, with a transport order of
dopamine > epinephrine > norepinephrine. SLMs containing
carrier 3 are stable, implying that carrier 3 is a very good
candidate for transport mechanism studies.
4. Crown boronic acid 4 is an outstanding carrier of primary
catecholamines with a transport order of norepinephrine >
(17) The ditopic recognition sites in 1 are essentially identical to those
in 4. Thus, it is reasonable to expect that 1 and 4 would have the same
general transported structure, i.e., the 1:1 cyclic zwitterion 7. One important
difference, however, is that the membrane concentration of 4 in the SLM
study was 33 times higher than the concentration of 1 in the BLM study.
Although there is strong evidence that 1 forms structure 7, it is possible
that 4 may be form a larger oligomeric analogue(s), which may explain its
propensity to precipitate from the membrane.
The stabilities of the various SLMs were investigated.
Membranes containing carrier 3 appeared to be quite stable as
repeated runs with the same membrane produced little change
in flux. Thus carrier 3 is a very good candidate for further
(16) Aldag, R.; Scro¨der, G. Liebigs Ann. Chem. 1984, 1036-1048.

9824 J. Am. Chem. Soc., Vol. 118, No. 41, 1996
Paugam et al.
overnight, the mixture was quenched with 10% HCl (5 mL). The
solution was extracted with methylene chloride (3 × 10 mL). The
organic layers were dried (MgSO4), filtered, and evaporated. Subse-
quent purification by flash chromatography (silica gel, ethyl acetate as
eluent until the first fractions were collected and then 4:1 ethyl acetate/
methanol) gave 3 as a mixture of three isomers in 48% yield: 1H NMR
(300 MHz, acetone-d6/10% D2O) δ 7.90 (s, 1H), 7.77 (d, 1H, J ) 7.0
Hz), 7.47 (d, 1H, J ) 7.0 Hz), 7.30 (t, 1H, J ) 7.0 Hz), 6.88 (m, 8H),
4.81 (br s, 2H), 4.15 (m, 9H), 3.80 (m, 4H), 3.68 (br s, 4H), 1.26, 1.23
(s, 18 H) ppm; 13C NMR (75 MHz, acetone-d6/10% D2O) δ 149.9,
149.8, 149.2, 149.1, 148.3, 148.2, 147.4, 146.1, 145.0, 138.5, 138.4,
134.4, 134.3, 134.2, 134.0, 130.4, 130.5, 128.0, 119.7, 119.6, 118.5,
117.6, 117.3, 116.1, 115.9, 114.4, 112.7, 77.8, 77.7, 77.6, 72.6, 71.8,
71.5, 71.1, 70.1, 69.7, 69.5, 34.7, 34.5, 31.6 ppm; HRMS (FAB, glycerol
matrix) m/z 692.3707, calcd for C39O10H53B [M + glycerol]+ 692.3739.
1-Bromo-3,5-bis(bromomethyl)benzene (14). A mixture of 5-bromo-
m-xylene (28.6 mmol), N-bromosuccinimide (58.0 mmol), and R,R′-
azobisisobutyronitrile (150 mg) was heated at reflux in dry CCl4 (200
mL) for 18 h. The solution was allowed to cool, then filtered, and
evaporated. The desired compound was crystallized from hexanes as
dopamine . epinephrine. It is initially 10 times more effective
than an equimolar mixture of boronic acid 5 and crown 6, which
is one of best examples of ditopic cooperativity yet observed
in SLM transport.15 SLMs containing 4, however, do not exibit
long-term stability. Future work needs to focus on developing
analogues of 4 that maintain its impressive transport perfor-
mance and have high SLM stability.
5. It is possible that a device based on SLMs containing
crown boronic acid carriers can be developed to selectively
extract catecholamines from clinical samples.
Experimental Section
The materials and general experimental methods have been described
before.14,18
Crown Boronic Acid 1. Compound 8 (0.86 mmol) was treated
with butyllithium (0.90 mmol) in THF (100 mL) at -100 °C. The
reaction was stirred for 1 h at -100 °C, treated with B(OMe)3 (5.3
mmol), and allowed to warm up to room temperature overnight. The
solution was poured onto a mixture of crushed ice (50 g)/concentrated
hydrochloric acid (30 mL). The aqueous solution was extracted with
diethyl ether (3 × 20 mL), which removed all impurities. Subsequent
extraction with chloroform (4 × 20 mL) gave compound 1 in 32%
yield: 1H NMR (300 MHz, acetone-d6) δ 7.83 (s, 1H), 7.60 (d, 1H, J
) 7.5 Hz), 7.38 (d, 1H, J ) 7.5 Hz), 7.30 (t, 1H, J ) 7.5 Hz), 7.10 (s,
exchanged with D2O), 4.51 (s, 2H), 3.72 (m, 3H), 3.57 (m, 20 H),
3.29 (m, 2H) ppm; 13C NMR (10% D2O/acetone-d6) δ 138.7, 134.9,
134.8, 131.2, 129.7, 128.9, 79.2, 74.4, 72.3, 71.6, 71.4, 71.3, 70.8, 70.3
ppm; HRMS (FAB) m/z 485.2558, calcd for C23O10H37B [M + H]+
485.2558.
Crown Bromofluorobenzene 9. A solution of 2-(hydroxymethyl)-
18-crown-6 (0.53 mmol) in THF (1 mL) was added dropwise to a
solution of NaH (0.80 mmol) in THF (50 mL) at room temperature.
The mixture was stirred at room temperature for 3 h and then treated
with 5-bromo-2-fluorobenzyl bromide (0.53 mmol, prepared by NBS
bromination of 5-bromo-2-fluorotoluene) in THF (5 mL). The mixture
was heated at reflux for 4 h, cooled, filtered, and evaporated. The
residue was purified by flash chromatography (silica gel, 30:1 dichlo-
romethane/methanol) to afford the desired material as a tan oil in 90%
yield: 1H NMR (300 MHz, CDCl3) δ 7.56 (1H, dd, J ) 6.0, 2.0 Hz),
7.35 (1H, m), 6.91 (1H, t, J ) 9.0 Hz), 4.57 (2H, s), 3.82 (3H, m),
3.67 (20H, s), 3.58 (2H, m) ppm; 13C NMR (CDCl3, 75 MHz) δ 160.4
(d, J ) 255.0 Hz), 132.9 (d, J ) 52.0 Hz), 128.7 (d, J ) 22.0 Hz),
117.5 (d, J ) 22.0 Hz), 78.8, 78.7, 71.4, 71.3, 71.2, 71.1, 70.3, 66.5
ppm; HRMS (FAB) m/z 503.1027, calcd for C20H30O7BrFNa [M +
Na]+ 503.1057.
Crown Fluorobenzeneboronic Acid 2. Precursor 9 was converted
to the boronic acid using the same procedure described for 1. The
crude product was 90% 2 and 10% 10 as proved by comparison with
authentic material. Purification by flash chromatography followed by
reverse phase HPLC reduced this contaminant to 8%. The final yield
of 2 was 10%: 1H NMR (500 MHz, acetone-d6/10% D2O) δ 7.97 (1H,
dd, J ) 8.0, 2.0 Hz), 7.82 (1H, ddd, J ) 14.0, 8.0, 2.0 Hz), 7.06 (1H,
dd, J ) 11.0, 8.0 Hz), 4.58 (2H, s), 3.73 (3H, m), 3.57 (22H, s) ppm;
HRMS (FAB) m/z 469.1990, calcd for C20H32O9BFNa [M + Na]+
469.2025.
Crown Bromobenzene 11. 3-Bromobenzyl bromide (1.5 mmol)
was alkylated with crown 1312 (1.5 mmol) according to the procedure
described for 9. The crude material was purified by flash chromatog-
raphy (silica gel, 1:3 ethyl acetate/hexanes) to give 11 in 98% yield.
1H NMR (300 MHz, CDCl3) δ 7.56 (s, 1H), 7.38 (d, 1H, J ) 8.0 Hz),
7.29 (m, 1H), 7.19 (t, 1H, J ) 8.0 Hz), 6.85 (m, 12H), 4.81 (m, 2H),
4.21 (m, 4H), 3.86 (m, 4H), 3.78 (m, 4H), 1.26, 1.28 (s, 18H) ppm;
MS (FAB) m/z 673/671 [M]+.
1
white needles: yield 33%; H NMR (300 MHz, CDCl3 ) δ 7.48 (s,
2H), 7.37 (s, 1H), 4.42 (s, 4H) ppm; MS (EI) m/z 344/342 [M]+.
Crown Bromobenzene 12. Octadecanol (4.5 mmol) was alkylated
with 14 (4.5 mmol) according to the procedure described for 9. The
crude product was partially purified by chromatography (silica gel,
petroleum ether until the first fractions were collected and then 8:1
petroleum ether/ethyl acetate) to give a mixture of 14, 15, and 1-bromo-
3,5-[bis(1-octadecanoxy)methyl]benzene, 18. The residue was recrys-
tallized from 4:1 ethanol/methylene chloride at 0 °C to give a mixture
of 15 and 18 (2:1). The mixture (1.38 mmol of 15) was alkylated
with 2-(hydroxymethyl)-18-crown-6 (1.38 mmol) according to the
procedure described for 9. The crude product was purified by flash
chromatography (alumina, 1:3 petroleum ether/ethyl acetate, Rf ) 0.43)
to give 12 as an oil in 77% yield: 1H NMR (300 MHz, CDCl3) δ 7.40
(s, 2H), 7.18 (s, 1H), 4.50 (s, 2H), 4.45 (s, 2H), 3.80 (m, 3H), 3.67 (br
s, 20H), 3.55 (m, 2H), 3.46 (t, 2H, J ) 6.5 Hz), 1.61 (p, 2H, J ) 7.0
Hz), 1.25 (s, 30H), 0.88 (t, 3H, J ) 7.0 Hz); IR (KBr) 2920, 2850,
1670, 1450, 1350, 1300, 1100 cm-1; MS (FAB) m/z 769/767 [M +
H]+.
Crown Benzeneboronic Acid 4. Compound 12 (0.17 mmol) was
treated with butyllithium (0.35 mmol) in THF (7 mL) at -75 °C. The
reaction mixture was gradually brought to -55 °C over 1 h and stirred
an additional 6 h. The solution was recooled to -75 °C and treated
with B(OMe)3 (1 mmol). After quenching and workup similar to the
procedure described for 3, the crude material was purified by flash
chromatography (alumina, ethyl acetate/methanol gradient) to give
compound 4 as an oil in 55% yield. 1H NMR (300 MHz, acetone-d6/
10% D2O) δ 7.41 (s, 1H), 7.37 (s, 1H), 7.15 (s, 1H), 4.51 (s, 2H), 4.55
(s, 2H), 3.72 (m, 3H), 3.57 (br s, 20H), 3.53 (m, 2H), 3.44 (t, 2H, J )
6.5 Hz), 1.56 (p, 2H, J ) 7.0 Hz), 1.25 (s, 30H), 0.84 (t, J ) 7.0 Hz)
ppm; 13C NMR (75 MHz, acetone-d6/10% D2O) δ 136.6, 136.1, 131.7,
131.6, 127.5, 76.4, 71.8, 71.3, 69.6, 69.0, 68.8, 68.7, 68.6, 68.5, 67.9,
67.3, 30.5, 28.2, 24.8, 21.2 ppm; HRMS (FAB) m/z 789.5297, calcd
for C42H75O11BNa [M + Na]+ 789.5308.
4-[[[8-(2-Nitrophenoxy)octyl]oxy]carbonyl]benzeneboronic Acid
(5). A suspension of 4-carboxybenzeneboronic acid (3.0 mmol) and
potassium hydrogen carbonate (6.0 mmol) in N,N-dimethylformamide
(25 mL) was treated with 1-bromo-8-(2-nitrophenoxy)octane11 (3.1
mmol) and stirred at 65 °C for 48 h. The solution was allowed to
cool, acidified (0.33 M sulfuric acid, pH ) 2-3), and extracted with
methylene chloride (3 × 20 mL), which was evaporated to afford an
oil. Subsequent purification by MPLC (ethyl acetate/methanol gradient)
gave 5 in 65% yield: 1H NMR (300 MHz, acetone-d6/10% D2O) δ
7.94 (s, 4H), 7.78 (d, 1H, J ) 8.0 Hz), 7.59 (t, 1H, J ) 8.0 Hz), 7.28
(d, 1H, J ) 8.0 Hz), 7.07 (t, 1H, J ) 8.0 Hz), 4.29 (t, 1H, J ) 6.5 Hz),
4.15 (t, 1H, J ) 6.5 Hz), 1.70 (m, 4H), 1.30 (m, 8H) ppm; 13C NMR
(75 MHz, acetone-d6/10% D2O) δ 211.2, 167.0, 152.5, 141.0, 134.9,
134.7, 128.9, 125.5, 120.9, 115.5, 70.0, 65.5, 30.0, 29.7, 29.6, 29.5,
29.4 ppm; HRMS (FAB, glycerol matrix) m/z 472.2122, calcd for
C24H31O8NB [M + glycerol]+ 472.2143.
Crown Boronic Acid 3. Compound 11 (0.54 mmol) was treated
with butyllithium (1.1 mmol) in THF (10 mL) at -100 °C. The reaction
mixture was stirred at -100 °C for 1 h and then brought to -40 °C
over 2 h. The solution was then recooled to -100 °C and treated with
B(OMe)3 (2.7 mmol). After being warmed to room temperature
2-[[[8-(2-Nitrophenoxy)octyl]oxy]methyl]-18-crown-6 (6). 1-Bromo-
(18) Morin, G. T.; Paugam, M.-F.; Hughes, M. P.; Smith, B. D. J. Org.
Chem. 1994, 59, 2724-2728.
8-(2-nitrophenoxy)octane11 (1.01 mmol) was alkylated with 2-(hy-

Facilitated Catecholamine Transport
J. Am. Chem. Soc., Vol. 118, No. 41, 1996 9825
droxymethyl)-18-crown-6 (1.0 mmol) according to the procedure
described for 9. The residue was purified by flash chromatography
(alumina, 1:1 ethyl acetate/hexanes) to give 6 as a yellow oil in 33%
yield: 1H NMR (300 MHz, CDCl3) δ 7.80 (d, 1H, J ) 8.0 Hz), 7.48
(d, 1H, J ) 8.0 Hz), 7.04 (d, 1H, J ) 8.0 Hz), 6.98 (t, 1H, J ) 8.0
Hz), 4.07 (t, 2H, J ) 6.5 Hz), 3.79 (t, 2H, J ) 5.0 Hz), 3.64 (m, 21H),
3.46 (m, 4H), 1.78 (p, 2H, J ) 8.0 Hz), 1.45 (m, 8H) ppm; 13C NMR
(75 MHz, CDCl3) δ 152.3, 139.9, 133.8, 125.4, 119.9, 114.3, 78.3,
71.7, 71.5, 70.7, 70.6, 70.5, 64.5, 62.8, 32.6, 29.5, 29.2, 29.1, 28.5,
25.9, 25.7, 25.5 ppm; HRMS (FAB) m/z 566.2952, calcd for C27H45O10-
NNa [M + Na]+ 566.2947.
Transport of Catecholamines through BLMs. The apparatus and
methodology used in the BLM transport experiments have been
described in detail.18 Briefly, a solution of sodium phosphate buffer
(0.1 M, pH 7.4, 7 mL) containing sodium dithionite (Na2S2O4, 0.01
M) as an antioxidant and a solution of carrier in chloroform (1 mM, 7
mL) were shaken in a separatory funnel. The organic layer was then
placed in the bottom of a U-tube apparatus (internal diameter 1.20 cm,
height 10 cm, and 2.5 cm between the arms) and the aqueous phase
added in halves to the two arms of the appartaus. Only the organic
layer was stirred (475 rpm). The transport experiment was initiated
by adding a small aliquot of concentrated catecholamine solution to
the source phase. The appearance of catecholamine in the receiving
phase was monitored spectroscopically at 279 nm. All transport runs
were repeated at least once, and the observed fluxes were always within
(15% of each other.
close to 15 cm2. The exact dimensions were taken into consideration
when computing fluxes. The source and receiving phases contained
sodium phosphate (100 mM) buffered at pH 7.2 and sodium dithionite
(1 mM) as an antioxidant. The source phase also contained 50 mM
catecholamine. The membrane was a thin, flat sheet of Accurel that
had been soaked in 2-nitrophenyl octyl ether containing 33 mM (about
2% wt) of carrier. The temperature was maintained at 298 K. The
appearance of catecholamine in the receiving phase was determined
from the change in UV absorption at 279 nm. Continuous measurement
of the receiving phase was achieved by cycling through a flow-through
UV cell. Initial fluxes were determined by curve-fitting analysis after
50 min of transport. All transport runs were repeated at least once,
and the observed fluxes were always within (10% of each other.
19F NMR Studies. An aqueous solution of 2 (4 mL, 4 mM) and
dopamine (40 mM), buffered at pH ) 7.4, was shaken and equilibrated
with an equal volume of chloroform. Both layers were analyzed by
19F NMR relative to external CF3CH2OH.
Acknowledgment. This work was supported by a NATO
collaborative reseach grant and the U.S. National Science
Foundation. B.D.S. acknowledges a Cottrell Scholar award of
Research Corporation.
Supporting Information Available: 1H and 13C NMR
spectra of carriers 1, 3, 4, and 5 and 19F NMR spectra of carrier
2 with and without dopmaine (9 pages). See any current
masthead page for ordering and Internet access instructions.
Transport of Catecholamines through SLMs. The design of the
SLM transport cells has been described before.14 A series of near-
identical, water-jacketed cells were used, each with source and receiving
phase volumes of approximately 50 mL and membrane surface areas
JA9615076