A rational approach to the design of selective substrates and potent nontransportable inhibitors of the excitatory amino acid transporter EAAC1 (EAAT3). New glutamate and aspartate analogues as potential neuroprotective agents

Article abstract of DOI:10.1021/jm015509z

Two three-dimensional receptor interaction models for EAAT substrates and nontransportable inhibitors have been developed, and new glutamate (Glu) and aspartate (Asp) analogues have been synthesized. The analogues 1a and 3 represent novel lead compounds f

Full text of DOI:10.1021/jm015509z

J . Med. Chem. 2001, 44, 2507-2510
2507
Ch a r t 1. New EAAT Substrates (1a ,b and 2) and
Inhibitors (3, 4)
A Ra tion a l Ap p r oa ch to th e Design of
Selective Su bstr a tes a n d P oten t
Non tr a n sp or ta ble In h ibitor s of th e
Excita tor y Am in o Acid Tr a n sp or ter
EAAC1 (EAAT3). New Glu ta m a te a n d
Asp a r ta te An a logu es a s P oten tia l
Neu r op r otective Agen ts
Giuseppe Campiani,*,† Meri De Angelis,‡
Silvia Armaroli,‡ Caterina Fattorusso,|
Bruno Catalanotti,† Anna Ramunno,† Vito Nacci,‡
Ettore Novellino,| Christof Grewer,§ Diana Ionescu,§
Thomas Rauen,# Roger Griffiths, Colin Sinclair,
Elena Fumagalli,¥ and Tiziana Mennini¥
Dipartimento di Scienze Farmaceutiche, Universita’ degli
Studi di Salerno, via Ponte Don Melillo 11,
84084 Fisciano, Italy, Dipartimento Farmaco Chimico
Tecnologico, Universita’ degli Studi di Siena,
via Aldo Moro, 53100 Siena, Italy, Dipartimento di Chimica
delle Sostanze Naturali and Dipartimento di Chimica
Farmaceutica e Tossicologica, Universita’ degli Studi di
Napoli “Federico II”, via D. Montesano 49,
(EAAT1-5).4 These Glu/Asp uptake systems are located
at the pre- and postsynaptic levels in the CNS, in glial
cells (EAAT1 and EAAT2), in the blood brain barrier
(EAAT3),5 and in the presynaptic vesicles.6-8 The link
between impairment of transporter functions and exci-
totoxic concentrations of Glu suggests that transporter
dysfunction may contribute to neurodegenerative dis-
eases.4,9 It has been recently demonstrated that during
ischemia the outward-activated glutamate transporter
may be the major source of extracellular Glu.10 Fur-
thermore, under energy deprivation conditions, Glu can
be released from glial cells by reversed transport.11
Therefore, the design of “reversed transport inhibitors”
(nontransportable inhibitors or blockers), with very low
affinity for iGluRs, could represent a new strategy for
the development of potential neuroprotective agents.4
On the other hand, selective competitive substrates for
EAATs, with negligible affinity for iGluRs, could rep-
resent a new therapeutic approach to minimize the
progression of chronic neurodegenerative diseases in-
volving glutamate toxicity.4 In this Letter we describe
a comprehensive approach, involving molecular model-
ing, synthesis, and biology in an attempt to discover
such novel neuroprotective agents targeting EAATs.
This approach led to the synthesis of new Glu and Asp
analogues as selective high affinity substrates and
nontransportable inhibitors, respectively. The Glu ana-
logue 1a and the Asp analogue 3 represent new proto-
typic compounds for the development of possible neu-
roprotective agents (Chart 1).
Molecu la r Mod elin g Stu d ies. To design new and
potent EAAT ligands, we investigated the conforma-
tional and structural features of known substrates (L-
threo-3-OHAsp; L-threo-4-OHGlu, L-trans-2,4-PDC, cis-
ACBD; L-CCGIII; 2,4MPDC) and nontransported in-
hibitors (L-anti-endo-3,4-MPDC; L-trans-2,3-PDC; L-threo-
3-BnOAsp; kainic acid; dihydrokainic acid; L-threo-3-
BzAsp).12 By a combination of systematic conformational
search (SYBYL, Tripos)13 and energy minimization
methods (cvff force fields; Discover, MSI),14 we have
generated hypothetical binding modes for competitive
substrates and nontransportable inhibitors (Figure 1).
This study allowed the definition of those structural and
conformational parameters required for an EAAT ligand
80131 Napoli, Italy, Max-Planck Institut fur Biophysik,
Frankfurt/ Main, Germany, Institute fur Biochemie,
Westfalische-Wilhelms-Universitat Munster,
Wihelm-Klemm-Str 2, D48149 Munster, Germany,
Biomolecular Science Building, University of St. Andrews,
St. Andrews, Fife KY16 9ST, Scotland, U.K., and Istituto di
Ricerche Farmacologiche Mario Negri, via Eritrea 62,
20157 Milano Italy
Received February 5, 2001
Abstr a ct: Two three-dimensional receptor interaction models
for EAAT substrates and nontransportable inhibitors have
been developed, and new glutamate (Glu) and aspartate (Asp)
analogues have been synthesized. The analogues 1a and 3
represent novel lead compounds for the development of EAAT
substrates and nontransportable inhibitors, selective for EAATs
over iGluRs, as possible neuroprotective agents useful to
minimize the progression of chronic or acute neurodegenera-
tive diseases. The role played by the protonatable amine
function in the interaction with EAATs has been discussed.
In tr od u ction . L-Glutamate (Glu) is the major exci-
tatory neurotransmitter in the mammalian central
nervous system (CNS). The process of excitatory neu-
rotransmission is initiated by the presynaptic release
of glutamate into the synaptic cleft, where glutamate
activates ionotropic (iGluRs) and metabotropic (mGluRs)
receptors, while the high affinity electrogenic transport
systems (EAATs) contribute to the termination of the
excitatory signal and to the recycling of the transmitter
into the metabolic pool.1-3 To date, five different mam-
malian subtypes of transporters have been cloned
* To whom correspondence should be addressed (Dipartimento di
Scienze Farmaceutiche, Universita’ degli Studi di Salerno, via Ponte
Don Melillo 11/C, 84084 Fisciano, Salerno, Italy; tel 0039-089-964381;
fax 0039-089-962828; e-mail campiani@unisa.it).
† Universita’ degli Studi di Salerno.
‡ Universita’ degli Studi di Siena.
| Universita’ degli Studi di Napoli “Federico II”.
§ Max-Planck Institut fur Biophysik.
#
Westfalische-Wilhelms-Universitat Munster.
University of St. Andrews.
Istituto di Ricerche Farmacologiche Mario Negri.
¥
10.1021/jm015509z CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/12/2001

2508 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Letters
Sch em e 1. Synthesis of New EAAT Substratesa
Figu r e 1. Superimposition of (A) nontransportable inhibitorss
L-anti-endo-3,4-MPDC (white), l-trans-2,3-PDC (violet), L-threo-
3-BnOAsp (yellow), kainic acid (redorange), dihydrokainic acid
(purple), and L-threo-3-BzAsp (green)sand (B) competitive
substratessL-threo-3-OHAsp (magenta), L-threo-4-OHGlu
(green-blue), L-trans-2,4-PDC (cyan), cis-ACBD (green), L-
CCGIII (orange), and 2,4-MPDC (white). Superimposition was
obtained by fitting the protonatable nitrogen (blue) and the
two carboxylic carbons (oxygens in red). Hydrogens are omitted
for clarity.
a
(a) DCC, HOBT, TEA; (b) Burgess reagent; (c) H2/Pd; (d)
BrCCl3, DBU; (e) LiOH, THF/H2O, overnight.
Sch em e 2. Synthesis of New EAAT Blockersa
F igu r e 2. Fitting of newly designed blockers and substrates
within the exceeding volume for blockers, with respect to
substrates (cyan framework), and for substrates with respect
to blockers (white framework). (A) New selective blockers: 3
(yellow), 4 (green). (B) New selective substrates 1a (green-
blue), 1b (purple), 2 (red-orange). Superimposition was ob-
tained by fitting the protonatable nitrogen (blue) and the two
carboxylic carbons (oxygens in red). Hydrogens are omitted
for clarity.
a
(a) LHMDS, I2, -78 °C; (b) H2 (1 atm), Pd/C; (c) NiO2; (d)
LiOH, THF, H2O.
to behave as competitive substrate or blocker. The first
evidence was that the calculated distances between the
two carboxylic functions (see Table 1 in the Supporting
Information) could not discriminate between substrates
and nontransported inhibitors but between glutamate-
and aspartate-related derivatives, which are present in
both classes. Indeed, the main structural and confor-
mational feature that can discriminate between sub-
strates and blockers is the steric excess, as previously
hypothesized.12 From the superimpositions (Figure 1a,b)
it can be observed that a constrained folded glutamate-
like conformation (i.e., 2,3-PDC and 3,4-MPDC) forces
the volume excess to occupy the edge opposite to the
one where carboxylic groups are located (blockers), while
a constrained unfolded glutamate-like conformation (i.e.,
2,4-PDC and 2,4-MPDC) allows a quite uniform volume
distribution on both sides of the backbone (Figure 2)
(transportable substrates). On the other hand, the
distance between the carboxylic groups represents a
critical parameter responsible for improvement of se-
lectivity toward Glu transporters, with respect to
iGluRs. Accordingly, we designed a new class of poten-
tially selective molecules, predicted as competitive
substrates (1a ,b and 2), based on an oxazoline or oxazole
structure, bearing an alkyl chain as a flexible spacer
between the two carboxylic functionalities (Chart 1A,
Figure 2a), and the potentially selective EAATs blockers
3 and 4 (Chart 1A, Figure 2b) to investigate the
importance of the protonatable function (critical for
iGluRs interaction) for a selective binding to EAATs.
Despite the lack of the free amine function (3, 4), the
designed inhibitors preserve a group capable to establish
a hydrogen bonding interaction with the carrier, such
as the heterocyclic oxygen/nitrogen atoms of 3 and 4.
Ch em istr y. Synthesis of the new compounds was
accomplished as shown in Schemes 1 and 2. By conden-
sation of the R-methyl ester of N-Cbz L-glutamate 5a
and R-benzyl ester of N-Cbz L-aspartate (5b) or R-benzyl
ester of N-Cbz L-glutamate (5c) with the hydrochloride
salt of L-serine methyl and benzyl esters 6a ,b, the Ser-
Glu 7a ,c and Ser-Asp 7b dipeptidic compounds were
obtained. We converted 7a -c to the protected oxazoline

Letters
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16 2509
Ta ble 1. Ki (µM) Values for Novel Glu and Asp Analogues for
Ta ble 2. Transport of the New Competitive Substrates 1a ,b
and 2 by EAAC1 under Steady-State Conditionsa and Inhibition
of EAAC1 by 3 and 4 Compared to D,L-threo-3-BnOAsp under
Steady-State Conditions in the Absence and Presence of
L-Glutamic Acida
AMPA, Kainate, and NMDA Receptorsa
AMPA
KA
NMDA
compd
Ki
Ki
Ki
L-threo-3-OHAsp
NA
NA
NA
14 ( 0.6
8 ( 0.4
0.49 ( 0.01
NA
9 ( 0.9
3 ( 0.7
NA
NA
0.2 ( 0.05
Ki (µM)
1a
NA
Imax/Imax
Ki (µM)
Imax/Imax
(+20 µM
1b
103 ( 18
41 ( 5
NA
NA
1.1 ( 0.3
compd KS (µM) (1 mM Glu) (no L-Glu) (1 mM Glu)
L-Glu)
2
L-Glu 5.1 ( 0.2b
1
3
NA
1a
1b
2
31 ( 6
1.15 ( 0.03
730 ( 60 0.92 ( 0.03
100 ( 27 1.03 ( 0.10
4
NA
L-Glu
0.063 ( 0.007
a
Each value is the mean ( SEM of at least three separate
experiments, performed in triplicate, and represents the concen-
tration giving half-maximal inhibition of [3H]AMPA (AMPA),
[3H]kainic acid (KA), and [3H]CGP39653 (NMDA) binding to rat
3
14 ( 1
-0.29 ( 0.1
42 ( 3
(68)c
4
75 ( 25 -0.25 ( 0.11 530 ( 60
(360)c
0.6 ( 0.1b -0.24 ( 0.05b 4.9 ( 0.4
(3.5)b,c,d
D,L-threo-3-OAsp
cortex homogenate. NA ) not active at >10-4 M.
b
a
derivatives 8a -c with Burgess reagent,15 and then,
after deprotection of the carboxylic and amine functions
of 8b,c, derivatives 1a ,b were obtained. Oxidation with
BrCCl3/DBU16 of 8a followed by deprotection provided
the desired oxazole derivative 2. In the case of blockers
(Chart 1A), conversion of 9a to the oxazoline derivative
10a , by means of iodine and LHMDS,17 followed by
catalytic hydrogen-mediated deprotection afforded the
oxazoline analogue 3. Alternatively, nickel oxide oxida-
tion18a of 10b, in turn obtained from 10b, provided the
oxazole dimethyl ester 11, which after saponification
gave 4.
In Vitr o P h a r m a cology. In binding assays, all the
compounds showed negligible affinity for the glycine site
associated to NMDA receptor. The Asp analogues 3 and
4 showed negligible interaction with AMPA, kainate,
and NMDA recognition sites (Kis >10-4 M) (Table 1).
Compound 1a , which presents a longer distance be-
tween the two carboxylic groups with respect to Glu,
did not show detectable inhibitory effect on [3H]kainic
acid, [3H]AMPA, [3H]MDL 105,519, and [3H]CGP 39653
receptor binding (Kis >10-4 M), while its superior
homologues 1b and 2 showed micromolar affinity for
interaction with KA and NMDA receptors, whereas the
affinity of these two compounds for AMPA receptors was
in the high micromolar range (Table 1). Their relatively
weak affinity for iGluRs could be explained on the basis
of a favorable spatial arrangement of the longer alkyl
chain.
Presumably, for 3 and 4, the lack of a protonatable
nitrogen (for oxazole and oxazoline rings, pKa ) 0.8 and
5.0, respectively18b) and the steric excess shown in the
molecular modeling studies critically contribute to their
EAAT selectivity. The whole set of compounds was
subjected to electrophysiological studies. We used whole-
cell current recordings from glutamate transporter-
expressing mammalian cells as an assay for the bio-
logical activity of the new compounds.7 According to the
electrophysiological data, the new Glu and Asp ana-
logues can be divided into two classes: (i) substrates
that are transported by EAAC1 (1a ,b and 2) and (ii)
nontransportable competitive inhibitors (3 and 4), as
expected from the molecular modeling studies.
At saturating concentrations, the transported sub-
strates evoked similar currents in EAAC1 as glutamate
(Table 2, example trace for 2 shown in Figure 3A),
suggesting that their steady-state transport rate does
not differ significantly from that found for glutamate
(∼35 s-1 at Vm ) 0 mV).7 However, their apparent
EAAC1 affinity varies in a range from 30 µM (1a ) to
Imax/Imax (1 mM L-Glu) represents the fractional current
induced by saturating concentrations of tested compounds com-
pared to that induced by saturating concentrations of glutamate
in the presence of intracellular SCN-. From ref 7. c The values
in parentheses represent Ki(S) calculated with eq 2 (Supporting
Information), based on Ki (third column) and KS ) 5.1 µM. 25
µM L-Glu.
b
d
730 µM (1b), which is 100-fold greater than the appar-
ent Km of EAAC1 for glutamate (Table 2). These results
suggest that the chain length as well as the configura-
tion of the γ-carboxy-analogue group of the new sub-
strates is important for their EAAC1 binding but not
for the rate of transport. The nontransportable inhibi-
tors 3 and 4 have two effects on the glutamate trans-
porters: (i) they inhibit glutamate transport and (ii)
they inhibit the leak-anion conductance in the absence
of transported substrates, a behavior that is typical for
competitive glutamate transporter inhibitors.7 In the
presence of 20 µM Glu, the Asp derivatives 3 and 4
inhibit EAAC1 in the micromolar concentration range
(Table 2). In the absence of glutamate, the new blockers
inhibited EAAC1 with a higher apparent affinity (Table
2, Figure 3C), in line with the competitive mechanism
which was expected for such substances, as demon-
strated in Figure 3C. The most potent and selective
inhibitor for EAAC1 is 3 (examples of current recordings
shown in Figure 3A), which exhibits an apparent
inhibition constant of 14 ( 1 µM (Figure 3B). Interest-
ingly, despite the lack of the protonatable free amine
function, the oxazoline 3 retained significant affinity for
EAAC1. Therefore, we conclude that protonation of the
amino group is not necessary for high affinity interac-
tion of the compound with EAAC1. In the absence of
an electrochemical gradient for anions across the mem-
brane, the new inhibitors do not induce any measurable
current in EAAC1 (shown for 3 in the right panel of
Figure 3A), indicating that they are, in fact, not
transported by this glutamate transporter subtype.
Furthermore, the excitotoxic effect on cerebellar granule
cells of a subset of compounds was also evaluated.
Cerebellar granule cells at 7-DIV were exposed for 24
h to increasing concentrations (0.1-1000 µM) of either
L-glutamate, 1a , and 3 (selective EAAC1 ligands) or 1b
and 2, two EAAC1 substrates showing moderate affinity
for iGluRs, before cell viability was monitored by the
MTT assay. Glutamate exerted a concentration-depend-
ent cytotoxic action in 7-DIV neurons (EC50 ∼ 50 µM).
In contrast, 1a ,b, 2, and 3 were nontoxic in 7-DIV cells
even after 24 h continuous exposure to concentrations
up to 1 mM.

2510 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 16
Letters
design of selective EAAT substrates and nontransport-
able inhibitors as possible neuroprotective agents.
Ack n ow led gm en t. We thank the European Union
for financial support (CEC BIOTECHNOLOGY PRO-
GRAMME - Demostration Contract CT 98-0223). We
are grateful to E. Bamberg for continuous encourage-
ment and support and N. Watzke for help with the cell
culture and transfection.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails (chemistry, molecular modeling, and pharmacology (ref)),
eqs 1 and 2, Table 1, and figures that portray the determina-
tion of Km values for EAAC1 substrates. This information is
available free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Michaelis, E. K. Molecular Biology of Glutamate Receptors in
the Central Nervous System and Their Role in Excitotoxicity,
Oxidative Stress and Aging. Prog. Neurobiol. 1998, 54, 369-
415.
(2) Gegelashvili, G.; Schousboe, A. High Affinity Glutamate Trans-
porters: Regulation of Expression and Activity. Mol. Pharmacol.
1997, 52, 6-15.
(3) Takahashi, M.; Billups, B.; Rossi, D.; Sorentis, M.; Hamann, M.;
Attwell, D. The Role of Glutamate Transporters in Glutamate
Homeostasis in the Brain. J . Exp. Biol. 1997, 200, 401-409.
(4) Seal, R. P.; Amara, S. G. Excitatory Amino Acid Transporters:
A Family in Flux. Annu. Rev. Pharmacol. Toxicol. 1999, 39, 431-
456 and references cited therein.
(5) O’Kane, R. L.; Martinez-Lopez, I.; DeJ oseph, M. R.; Vina, J . R.;
Hawkins, R. A. Na+-dependent Glutamate Transporters EAAT1,
EAAT2, and EAAT3 of the Blood-Brain Barrier. J . Biol. Chem.
1999, 274, 31891-31895.
(6) Takamori, S.; Rhee, J . S.; Rosenmund, C.; J ahn, R. Identification
of a Vesicular Glutamate Transporter that Defines a Glutamater-
gic Phenotype in Neurons. Nature 2000, 407, 189-194.
(7) Grewer, C.; Watzke, N.; Wiessner, M.; Rauen, T. Glutamate
Translocation of the Neuronal Glutamate Transporter EAAC1
Occurs within Milliseconds. Proc. Natl. Acad. Sci. U.S.A. 2000,
97, 9706-9711.
(8) Slotboom, D. J .; Konings, W. N.; Lolkema, J . S. Structural
Features of the Glutamate Transporter Family. Microbiol., &
Mol. Biol. Rev. 1999, 63, 293-307.
(9) Coyle, J . T.; Puttfarcken, P. Oxidative Stress, Glutamate, and
Neurodegenerative Disorders. Science 1993, 262, 689-695.
(10) Rossi, D. J .; Oshima, T.; Attwell, D. Glutamate Release in Severe
Brain Ischaemia is Mainly by Reversed Uptake. Nature 2000,
403, 316-321.
(11) Koch, H. P.; Chamberlin, A. R.; Bridges, R. J . Non Transportable
Inhibitors Attenuate Reversal of Glutamate Uptake in Synap-
tosomes Following a Metabolic Insult. Mol. Pharmacol. 1999,
55, 1044-1048.
F igu r e 3. (A) Left panel: Compound 2 is a transported
substrate of EAAC1. Whole-cell currents were evoked by
application of 1 mM glutamate (light gray bar) or 1 mM 2
(white bar) (Vm ) 0 mV, KSCN internal, pH 7.4, T ) 22 °C).
Similar results were obtained for 1a and 1b (not shown).
Middle panel: Compound 3 generates an outward anion
current. The experiment is similar to that shown in the left
panel, but 200 µM 3 was applied (dark gray bar). The light
gray bar indicates application of 10 µM glutamate. Right
panel: Compound 3 is not transported by EAAC1. In the
absence of an electrochemical gradient for anions ([Cl-]i )
[Cl-]o, Vm ) 0 mV), application of 1 mM 3 does not generate
currents in EAAC1. (B): Dose-response relationships of
EAAC1 outward currents generated by compounds 3 (closed
circles) and 4 (triangles) in the absence of glutamate. (C):
Inhibition of EAAC1 by Asp analogues 3 and 4 is competitive.
The circles represent the ratio Ki(S)/Ki. Ki(S) was determined
at 20 µM glutamate. The line is the expected value of Ki(S)/Ki
calculated according to eq 2 (Supporting Information) for a
competitive mechanism with KS ) 5.1 µM.
(12) Koch, H. P.; Kavanaugh, M. N. P.; Esslinger, C. S.; Zerangue,
N.; Humphrey, J . M.; Amara, S. G.; Chamberlain, A. R.; Bridges,
R. J . Differentiation of Substrates and Nonsubstrates Inhibitors
of the High-Affinity, Sodium-Dependent Glutamate Transport-
ers. Mol. Pharmacol. 1999, 56, 1095-1104.
Con clu sion . In summary, we disclosed the rational
design of novel and potent glutamate and aspartate
analogues selectively interacting with electrogenic
glutamate transporters, as potent and selective EAAT
substrates and nontransportable inhibitors. The new
substrates (1a ,b and 2) described in this Letter potently
interact with EAAC1, they are transported with the
same turnover rate as Glu, and unlike all the other
known substrates, they are characterized by weak-to-
negligible affinity for iGluRs. On the other hand, this
rational approach led to the identification of 3, a potent
and selective oxazoline-based blocker, the first charac-
terized by the lack of a free amine function to be
protonated at physiological pH, providing further in-
sights on the role played by this function in the
interaction with the EAAT binding site. Starting from
our binding site interaction models and the new leads,
these studies may pave the way for future rational
(13) SYBYL Molecular Modeling System (version 6.6), Tripos Assoc.,
St. Louis, MO.
(14) Inight 98, Discover module, MSI, San Diego, CA.
(15) (a) Wipf P.; Xu W. Total Synthesis of the Antimitotic Marine
Natural Product (+)-Curacin A. J . Org. Chem. 1996, 61, 6556-
6562. (b) Wipf, P.; Miller, C. P. A Short, Stereospecific Synthesis
of Dihydrooxazoles from Serine and Threonine Derivatives.
Tetrahedron Lett. 1992, 33, 907-910.
(16) Williams, D. R.; Lowder, P. D.; Gu, Y.; Brooks, D. A. Studies of
Mild Dehydrogenations in Heterocyclic Systems. Tetrahedron
Lett. 1997, 38, 331-334.
(17) Cardillo, G.; Gentilucci, L.; Tolomelli, A.; Tomasini, C. A Practical
Method for the Synthesis of -Amino -Hydroxy Acids. Synthesis
of Enantiomerically Pure Hydroxyaspartic Acid and Isoserine.
Synth. Lett. 1999, 11, 1727-1730.
(18) (a) Evans, D. L.; Minstewr, D. K.; J ordis, U. Nickel Peroxide
Dehydrogenation of Oxygen-, Sulfur-, and Nitrogen- Containing
Heterocycles. J . Org. Chem. 1979, 44, 497-501. (b) Oh, K. S.;
Lee, C.-W.; Choi, S. H.; Lee, J . S.; Kim, K. S. Origin of the High
Affinity and Selectivity of Novel Receptors for NH4+ over K+:
Charged Hydrogen Bonds vs Cation-p Interaction. Org. Lett.
2000, 2, 2679-2681.
J M015509Z