Design and synthesis of labeled analogs of PhTX-56, a potent and selective AMPA receptor antagonist

Article abstract of DOI:10.1016/j.bmc.2005.05.023

Polyamines and polyamine toxins are biologically important molecules, having modulatory effects on nucleotides and proteins. The wasp toxin, philanthotoxin-433 (PhTX-433), is a non-selective and uncompetitive antagonist of ionotropic receptors, such as ionotropic glutamate receptors and nicotinic acetylcholine receptors. Polyamine toxins are used for the characterization of subtypes of ionotropic glutamate receptors, the Ca²⁺-permeable AMPA and kainate receptors. A derivative of the native polyamine toxin, philanthotoxin-56 (PhTX-56), has recently been shown to be an exceptionally potent and selective antagonist of Ca²⁺-permeable AMPA receptors. PhTX-56 and its labeled derivatives are promising tools for structure-function studies of the ion channel of the AMPA receptor. We now describe the design and synthesis of ³H-, ¹³C-, and ¹⁵N-labeled derivatives of PhTX-56 for molecular level studies of AMPA receptors. [ ³H]PhTX-56 was prepared from a diiodo-precursor with high specific radioactivity, providing the first radiolabeled ligand binding to the pore-forming part of AMPA receptors. For advanced biological NMR studies, ¹³C and ¹⁵N-labeled PhTX-56 were synthesized using solid-phase synthesis. These analogs can provide detailed information on the ligand-receptor interaction. In conclusion, synthesis of labeled derivatives of PhTX-56 provides important tools for future studies of the pore-forming region of AMPA receptors.

Full text of DOI:10.1016/j.bmc.2005.05.023

Bioorganic & Medicinal Chemistry 13 (2005) 5104–5112
Design and synthesis of labeled analogs of PhTX-56, a potent
and selective AMPA receptor antagonist
Trine F. Andersen, Stine B. Vogensen, Lars S. Jensen,
Kolja M. Knapp and Kristian Strømgaard*
Department of Medicinal Chemistry, The Danish University of Pharmaceutical Sciences, Universitetsparken 2,
DK-2100 Copenhagen, Denmark
Received 29 March 2005; revised 19 May 2005; accepted 20 May 2005
Available online 28 June 2005
Dedicated to Professor Koji Nakanishi on receipt of the 2004 Tetrahedron Prize.
Abstract—Polyamines and polyamine toxins are biologically important molecules, having modulatory effects on nucleotides and
proteins. The wasp toxin, philanthotoxin-433 (PhTX-433), is a non-selective and uncompetitive antagonist of ionotropic receptors,
such as ionotropic glutamate receptors and nicotinic acetylcholine receptors. Polyamine toxins are used for the characterization of
subtypes of ionotropic glutamate receptors, the Ca2+-permeable AMPA and kainate receptors. A derivative of the native polyamine
toxin, philanthotoxin-56 (PhTX-56), has recently been shown to be an exceptionally potent and selective antagonist of Ca2+-perme-
able AMPA receptors. PhTX-56 and its labeled derivatives are promising tools for structure–function studies of the ion channel of
the AMPA receptor. We now describe the design and synthesis of 3H-, 13C-, and 15N-labeled derivatives of PhTX-56 for molecular
level studies of AMPA receptors. [3H]PhTX-56 was prepared from a diiodo-precursor with high specific radioactivity, providing
the first radiolabeled ligand binding to the pore-forming part of AMPA receptors. For advanced biological NMR studies, 13C
and 15N-labeled PhTX-56 were synthesized using solid-phase synthesis. These analogs can provide detailed information on the
ligand–receptor interaction. In conclusion, synthesis of labeled derivatives of PhTX-56 provides important tools for future studies
of the pore-forming region of AMPA receptors.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
and kainic acid. The iGlu receptors are particularly
interesting, as there is strong evidence to suggest that
glutamate-induced excessive Ca2+ entry under patho-
physiological conditions leads to a wide range of neuro-
logical insults including ischemia-induced brain damage,
epileptic seizures, and chronic neurodegenerative disor-
ders, such as AlzheimerÕs disease, as well as other neuro-
logical and psychiatric disorders.3,4 Moreover, iGlu
receptors are believed to be important for the basic cel-
lular and molecular mechanism by which memories are
formed and stored, as characterized by long-term poten-
tiation (LTP)5 and long-term depression (LTD).6
Endogenous polyamines are present in a vast majority
of cells. They play important roles in the synthesis of
proteins, cell division, and differentiation, and bind to
nucleic acids resulting in their condensation, thereby
affecting gene expression.1 Endogenous polyamines
and polyamine derivatives also bind and affect several
important ion channel proteins, such as potassium chan-
nels and ionotropic glutamate (iGlu) receptors.2
The iGlu receptors are important mediators of fast syn-
aptic transmission in the mammalian central nervous
system (CNS) and are classified into three subtypes
according to selective activation by the agonists, namely
N-methyl-D-aspartate (NMDA), (R,S)-2-amino-3-(3-
hydroxy-5-methylisoxazol-4-yl)propanoic acid (AMPA),
AMPA receptors are distributed ubiquitously through-
out the CNS and mediate fast excitatory neurotransmis-
sion. The involvement of AMPA receptors in LTP and
LTD, and the physical transport of AMPA receptors
into and out of the synaptic membrane have only re-
cently received widespread attention.7 AMPA receptors
are composed of four subunits designated GluR1–4, and
the receptor is formed by a tetrameric assembly of these
subunits.3 They can be classified according to their Ca2+
Keywords: Natural products; Polyamine toxins; Glutamate receptors;
AMPA receptors.
*
Corresponding author. Tel.: +45 3530 6114; fax: +45 3530 6040;
e-mail: krst@dfuni.dk
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.05.023

T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
5105
permeability; in GluR1, GluR3, and GluR4 subunits,
a glutamine (Q) residue is present at a site known as
the Q/R site that is located in the pore-forming region
of AMPA receptors (Fig. 2). This residue has been edi-
ted in virtually all GluR2 subunits to an arginine (R),
which determines the Ca2+ permeability of AMPA
receptors. Hence, Ca2+ permeability is inversely corre-
lated to the relative abundance of the GluR2 subunit.3
It has been suggested that Ca2+-permeable AMPA
receptors are important for the pathogenesis of neuro-
logical disorders.8 Thus, the use of AMPA receptor
antagonists is believed to be helpful in treating neurode-
generative disorders and/or reducing the amount of neu-
rological damage associated with these disorders.8,9
Polyamine toxins form a class of low molecular weight
compounds, isolated from the venom of spiders and
wasps, which are non-selective antagonists of ionotropic
receptors, such as iGlu and nicotinic acetylcholine
(nACh) receptors.15 Polyamine toxins are uncompetitive
antagonists, that is, they require the activation of the
receptor before antagonism can take place, probably
by binding to the pore-forming region of the receptor
(Fig. 2). This mode of action is interesting from a drug
discovery perspective, as recently an uncompetitive
antagonist of NMDA receptors, memantine (Ebixaꢁ),
was introduced for treating patients with moderate-
to-severe AlzheimerÕs disease.16
The polyamine wasp toxin philanthotoxin-433 [(S)-
PhTX-433 (1)] (Fig. 1), isolated from the venom of a
female digger wasp Philanthus triangulum, is a prototyp-
ical example.17 Polyamine toxin 1 and its synthetic
analog (S)-PhTX-343 (2) are antagonists of Ca2+-perme-
able AMPA and kainate receptors, which make them
useful pharmacological tools for determination of the
subunit composition of these receptors.18 The potential
of polyamine toxins, as drug candidates, has so far been
considered to be limited due to the non-specific action
on other ionotropic receptors. But, recently, it has been
demonstrated that modification of the polyamine por-
tion of PhTX-343 (2) led to selective antagonism on var-
ious classes of ionotropic receptors.19 Removal of one of
the inner basic sites in PhTX-343 (2) led to an enhanced
potency at AMPA receptors, whereas the activity at
other ionotropic receptors remained low.19c,d The most
striking example was that of philanthotoxin-56 [(S)-
PhTX-56 (3)] (Fig. 1), which inhibited AMPA receptors
that were composed of GluR1 subunits with a Ki value
of 3.3 0.78 nM, thus being the most potent polyamine
toxin analog ever described.19d PhTX-56 (3) showed
selectivity for Ca2+-permeable AMPA receptors, as it
was more than 1000-fold less potent at GluR2-contain-
ing AMPA receptors. Quite remarkably, PhTX-56 (3)
was also highly selective for AMPA, relative to kainate
A thorough understanding of the structure and function
of iGlu receptors in general and AMPA receptors, in
particular, is of major scientific interest and might have
pathbreaking consequences for the development of ther-
apeutics to treat neurological disorders. Gouaux and co-
workers have done a pioneering work in this field by
solving the X-ray crystallographic structure of a con-
struct of the extracellular ligand-binding domain.10 This
structure has been used to explain several important
functions of an AMPA receptor, such as a mechanism
for desensitization11 and a structural basis for partial
agonism.12 On the other hand, the membrane-spanning
(or pore) region of AMPA receptors, and iGlu receptors
in general, is much less well-defined.13 It is believed that
iGlu receptors are structurally and evolutionarily related
to K+-channels, and the discovery of the GluR0 chan-
nel/receptor, a hybrid between K+-channels and iGlu
receptors, has substantiated this belief.14 The current
hypothesis suggests that the pore-regions of iGlu recep-
tors are similar to inverted K+-channels,13b but still
properties, such as ion selectivity, are strikingly differ-
ent. Currently, very little is known about this region
and no concrete structural information is available.
Therefore, pharmacological tools that can allude to
some of these issues are of major interest.
Figure 1. Structures of polyamine toxins. PhTX-433 (1) is the native philanthotoxin isolated from the wasp P. triangulum, while PhTX-343 (2) and
PhTX-56 (3) are derivatives, the latter being a very potent and highly selective AMPA receptor antagonist.

5106
T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
receptors, as it was more than 500-fold less potent at
kainate receptors.19d
avoided and the use of protecting groups is reduced.15a,b
Generally, two approaches have been used for the
synthesis of polyamines on solid phase, either by
reductive amination or by alkylation. The latter strategy
has extensively adopted the Fukuyama amination,
where an amine is protected and activated as a nitroben-
zene sulfonamide, which is subsequently alkylated
either by conventional alkylation or by a Mitsunobu
reaction.15a,b
PhTX-56 is therefore an excellent tool to study ligand–
receptor interactions at the pore-forming part of AMPA
receptors. To the best of our knowledge, no radiolabeled
ligand exists that would allow binding studies with this
region; hence, we wanted to provide the first radioligand
that would allow such binding studies on the AMPA
receptor. Furthermore, we are interested in employing
advanced biological NMR spectroscopy to study li-
gand–receptor interactions at the molecular level, which
require isotope labeling of PhTX-56 with either 13C,
15N, or both. Recently, one example of 15N-labeling of
polyamine toxin derivatives has appeared; Bienz and
co-workers prepared 15N-labeled acylpentamines, to be
used as reference compounds to explain some unex-
pected observations during their mass spectrometric
investigations of novel polyamine toxins.20
We have previously shown how a small library of poly-
amine toxins could be synthesized by a sequential strat-
egy using the Fukuyama amination under Mitsunobu
conditions,22 and the methodology has since been used
for the synthesis of several philanthotoxin analogs.19d
Recently, a slightly modified version was presented
using N-protected aminoalkyl halides, rather than ami-
no alcohols, as a more convenient reaction sequence.23
Here, we describe the synthesis of a number of labeled
PhTX-56 derivatives, which, in general, were synthe-
sized by a solid-phase synthesis strategy.
Previously, photolabeling studies using analogs of phil-
anthotoxins have been carried out at the nicotinic ace-
tylcholine (nACh) receptor, showing that the aromatic
head-group of philanthotoxins binds to the lumen just
above the pore, but specific interactions of the poly-
amine portion remain elusive (Fig. 2).21 Recent studies
have shown that the polyamine moiety is crucial for
interaction with receptors, and this part of the molecule
determines selectivity, and in many cases the potency.19
Therefore, studying the interaction between the poly-
amine moiety of PhTX-56 (3) and the AMPA receptor
is of major interest, but has so far not been possible.
2. Results
Two principally different labeling strategies were carried
out, which in both cases left the native structure of
PhTX-56 unchanged; only the isotope labeling was
3
changed. In one case, a H-labeled version of PhTX-56
was synthesized by preparing a precursor on solid phase,
which was subsequently hydrogenated using tritium gas.
The other strategy involved labeling the polyamine
moiety of PhTX-56 with 13C and 15N, which could be
carried out by attaching labeled precursors to a solid
phase, and carrying out solid phase synthesis, as
previously described, providing 13C- and 15N-labeled
PhTX-56.
Biological interest in polyamines and polyamine toxins
has paved the way for an efficient synthesis of these com-
pounds, including native polyamines and polyamine
toxins, as well as their derivatives for the investigation
of structure–activity relationship (SAR) studies.15 The
use of solid-phase synthesis has greatly facilitated the
synthesis of these compounds, the major advantage
being that purification of highly polar intermediates is
2.1. [3H]PhTX-56
For the design and synthesis of a tritiated version of
PhTX-56, it was quite clear where to incorporate the tri-
tium label; one of the most frequently adopted methods
for the labeling of peptides is to convert a diiodo-tyrosine
precursor into a tritium-labeled tyrosine by simple hydro-
genation. Therefore, we decided to prepare diiodo-PhTX-
56 (4), as shown in Scheme 1. The polyamine portion was
synthesized, as outlined in Scheme 1, and as previously de-
scribed;19d,22 a resin-bound diamine (5) was reacted with
o-nitrobenzenesulfonyl chloride to give the correspond-
ing o-nitrobenzene sulfonamide 6. This was reacted with
a protected amino alcohol in a modified Mitsunobu reac-
tion, and the protecting group was removed by treatment
with tetrabutyl ammonium fluoride (TBAF) to give the
resin-bound polyamine precursor (7). The crucial step
was the coupling with N-Fmoc 3,5-diiodo-tyrosine, as
the tyrosine derivatives are generally protected at the phe-
nol group with a tert-butyl group. However, it was be-
lieved that steric hindrance induced by the two iodo-
atoms, as well as the ability of N-[(dimethylamino)-1H-
1,2,3-triazol[4,5-b]pyridin-1-yl-methylene]-N-methylme-
thanaminium hexafluorophosphate N-oxide (HATU) to
be less prone to reaction with phenolic groups, would be
Figure 2. Suggested mode of binding of PhTX-56 (3) to the pore-
forming region of AMPA receptors. Polyamine toxins, such as PhTX-
56 (3), are believed to antagonize ionotropic receptors by binding to
the pore-forming region, inside the ion channel. Antagonism of AMPA
receptors only takes place when receptors are composed of GluR1, 3,
and 4 subunits, whereas block is prevented if the GluR2 subunit is
present. This selectivity is explained by editing of the Q/R site; in
GluR1, 3, and 4 subunits glutamine (Q) resides, whereas virtually all
GluR2 subunits are edited and contain an arginine (R). Modified from
Ref. 10b.

T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
5107
Scheme 1. Reagents: (a) o-nitrobenzenesulfonyl chloride, DIEA; (b) 3-[N-[2-(trimethylsilyl)ethoxycarbonyl]amino]pentanol, Bu3P, ADDP; (c)
TBAF; (d) (S)-N-Fmoc-O-(tert-butyl)tyrosine, HATU, collidine; (e) 20% piperidine in DMF; (f) butyric acid, HATU, collidine; (g) HSCH2CH2OH,
DBU; (h) TFA, triisopropylsilane, CH2Cl2, H2O.
sufficient for a hasslefree reaction, and this turned out to
be true. Subsequently, the Fmoc group was removed to
give 8, which was reacted with butyric acid, followed by
the removal of the 2-nitrobenzene sulfonamide (NS)
group. Finally, the compound was cleaved from the resin
to give the desired diiodo-PhTX-56 (4). The crude prod-
uct was purified by preparative HPLC-MS, to give dii-
odo-PhTX-56 (4) in 35% yield.
evacuated three times. Tritium gas was then introduced
at atmospheric pressure, and the reaction mixture was
stirred overnight, which provided the crude product. Fol-
lowing purification by preparative HPLC, [3H]PhTX-56
(10) was obtained in high radioactive purity (>99.5%)
and with high specific activity (>50 Ci/mmol).24
2.2. 13C2, 15N2, and 13C2,15N2-labeled PhTX-56
With 4 in hand, conditions for reduction were investi-
gated; compound 4 was dissolved in DMF, Pd/C was
added, and the reaction mixture was stirred for 20 h un-
der an atmosphere of hydrogen providing PhTX-56 (3).
The same synthetic procedure was repeated, but using
deuterium, instead of hydrogen, gas, providing deuter-
ated PhTX-56 (9) (Scheme 2). The synthesis of com-
pound 9 confirmed the feasibility of the procedure and
showed that the iodine–deuterium exchange was quanti-
tative. Both PhTX-56 (3) and 9 were purified by pre-
parative HPLC-MS and were obtained in good yield
and high purity (>99% by UV).
In the design of 13C- and 15N-labeled derivatives of PhTX-
56 for advanced biological NMR studies, it was impor-
tant that the labels be placed in the polyamine portion
of the molecule since recent studies have shown that the
polyamine portion determines the selectivity and, to a cer-
tain extent, the potency of philanthotoxins.19 Moreover,
studies with photolabile PhTX-343 derivatives of nACh
receptors have indicated that the aromatic head-group
was located at the outer mouth of the ion channel.21
Hence, the head-group probably interacts with the recep-
tor in a non-specific manner, but plays an important role
in anchoring the molecule to the receptor.
For the preparation of tritium-labeled PhTX-56 and
[3H]PhTX-56 (10), similar reaction conditions were ap-
plied; a mixture of diiodo-precursor 4, diisopropylamine
(DIPA), and Pd/C in DMF was frozen at ꢀ80 ꢂC and
We wanted to incorporate the labels into the polyamine
moiety and envisaged that a synthetic approach similar
to that described above would be feasible. Since 1,6-dia-
minohexane is commercially available with both 13C2-
Scheme 2. Reagents: (a) Pd/C, hydrogen, deuterium or tritium gas.
Scheme 3. Reagents: (a) 18-crown-6; (b) LiAlH4; (c) 2-(trimethylsilyl)-ethyl 4-nitrophenyl carbonate, Et3N; (d) trityl chloride resin, DIEA; (e) TBAF.

5108
T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
and 15N2-labels, we have used them. However, as an
alternative approach we employed [13C], [15N], and
[13C,15N]potassium cyanide, which is much cheaper, as
outlined in Scheme 3. Starting from 1,4-dibromobutane
and potassium cyanide, 1,4-dicyanobutane (11) was
prepared in high yield by refluxing in acetonitrile in
the presence of 18-crown-6, and 11 was reduced to
1,6-diaminohexane (12) by treatment with LiAlH4.
3. Discussion
In the present work, we have prepared a range of labeled
derivatives of PhTX-56 (3), which itself is an interesting
pharmacological tool, as it is believed to bind to the
pore-forming region of AMPA receptors (Fig. 2). It is
highly sensitive to editing at the Q/R site and does not
antagonize AMPA receptors with arginine in this posi-
tion, that is, GluR2-containing receptors. PhTX-56 (3)
binds with very high affinity (Ki = 3.3 0.78 nM) and
has a unique selectivity, as it is 500-fold more potent
at AMPA than at kainate, receptors. There are a num-
ber of possibilities when labeling PhTX-56; the classical
Unprotected diamines are normally reacted with trityl
chloride resin, using a 10-fold excess of the diamine
however; in this case we were interested in minimizing
the use of diamines. We investigated the efficiency of
loading 1,6-diaminohexane to the resin, by comparing
the reaction with the trityl chloride resin with either 5
or 10 equivalents of 1,6-diaminohexane or 1 equivalent
of 1-N-(trimethylsilyl)ethoxycarbonyl-6-aminohexane.
The latter was treated with TBAF, and all three resins
were reacted with butyric acid in the presence of HATU
and collidine. The products were cleaved from the resins
and were subsequently analyzed by HPLC-MS, which
showed that the yields using either 5 or 10 equivalents
of 1,6-diaminohexane or 1 equivalent of 1-N-(trimethyl-
silyl)ethoxycarbonyl-6-aminohexane were 40%, 68%,
and 87%, respectively. This underlines the increased effi-
ciency by using mono-protected diaminohexane; hence,
the diamines were mono-protected by the reaction with
2-(trimethylsilyl)-ethyl 4-nitrophenyl carbonate to give
1-N-(trimethylsilyl)ethoxycarbonyl-6-aminohexane (13).
The latter compound was reacted with trityl chloride re-
sin to give 14 and subsequently treated with TBAF to
provide the resin-bound 1,6-diaminohexane (5).
3
approach is to introduce a radiolabel, such as H; alter-
nately, 13C- and 15N-labels can be introduced for biolog-
ical NMR studies. In both cases, the labels can be
introduced without changing the chemical structure of
PhTX-56 (3). The introduction of fluorescent labels is
also of major interest and has been performed (L. S. Jen-
sen and K. Strømgaard, unpublished), but this obvi-
ously changes the parent structure and requires
verification of the biological activity of the fluorescent
analogs.
A tritiated version of PhTX-56 was obtained by treat-
ment of a diiodo-precursor (4) with tritium gas, yielding
[3H]PhTX-56 (10) in high radiochemical purity (>99.5%)
and, most importantly, with high specific radioactivity
(>50 Ci/mmol). [3H]PhTX-56 (10) will be an important
tool in binding studies with AMPA receptors, as this
is the first radioligand to bind to the pore-forming re-
gion of AMPA receptors. In addition, [3H]PhTX-56
(10) can provide information on the pharmacokinetic
properties of this class of compounds, and most impor-
tantly, whether these compounds penetrate the blood–
brain barrier (BBB) or not. This is of utmost importance
to evaluate the therapeutic properties of these com-
pounds, and previous studies do not seem to have clar-
ified this point. Finally, [3H]PhTX-56 (10) can be used in
ex vivo autoradiographic studies, which will provide
information on regional binding, as well as binding
kinetics (Kd and Bmax), and when compared with similar
experiments using [3H]AMPA, a measurement of the
In this manner, the three starting resins, 15, 16, and 17
labeled with 13C2, 15N2, and 13C2,15N2, respectively, were
prepared (Scheme 4). The synthesis of [13C2]PhTX-56
(18), [15N2]PhTX-56 (19), and [13C2,15N2]PhTX-56 (20)
was then carried out using a methodology which is similar
to that described for the synthesis of diiodo-PhTX-56 (4).
The crude products were purified by preparative HPLC-
MS, and the target molecules were achieved in yields of
32%, 29%, and 35%, respectively for the eight-step
sequences.
Scheme 4. Synthesis of [13C2]PhTX-56 (18), [15N2]PhTX-56 (19), and [13C2,15N2]PhTX-56 (20). Resin-bound 1,6-diaminohexanes labeled with either
13C2], [15N2] or [13C2,15N2] were used as starting materials and the target molecules were synthesized according to the procedure shown in Scheme 1.
[

T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
5109
occurrences of Ca2+-permeable AMPA receptors might
be feasible.
cation, except for DMF, which was stored over 3 A
molecular sieves, and THF, which was distilled under
N2 from Na/benzophenone. TLC was carried out on
Merck Kieselgel 60 F254 aluminum sheets. 1H NMR
and 13C NMR spectra were recorded on a Varian Mer-
cury spectrometer at 300 MHz or on a Varian Gemini
200 BB at 300 MHz, using CDCl3 or CD3OD as sol-
vents. Chemical shifts are reported in ppm (d). Coupling
constants (J) are given in Hz. Multiplicities of 1H NMR
signals are given as follows: s, singlet; br s, broad singlet;
d, doublet; t, triplet; dt, doublet of triplets; dm, doublet
of multiplets; q, quartet; p, pentet; m, multiplet. Analyt-
ical and preparative high-performance liquid chro-
matography (HPLC-MS) was performed on a Sciex
API150ex instrument equipped with Atmospheric Pres-
sure Chemical Ionization (APCI) ion source. The HPLC
system consisted of two Shimadzu LC10ADvp pumps.
UV trace was obtained with a Gilson UV/VIS 155 UV
detector operating at 254 nm. Evaporative light scatter-
ing (ELS) trace was obtained with Sedere Sedex 55 Light
Scattering Detector and was used for estimation of the
purity of the final products. Analytical HPLC-MS was
performed on a 50 · 4.6 mm YMC RP18 column, using
2 mL/min of a mixture of A: water with 0.05% TFA and
B: acetonitrile with 5% water and 0.035% TFA. Prepara-
tive HPLC-MS (split-flow MS detection) was run with
500 lL injections (20 mg samples in 1.0 mL DMSO) to
a Phenomenex Synergi Hydro 4 lm, 21.20 · 50 mm
column eluted with the same solvent gradient at
22.7 mL/min. Accurate mass determination was per-
formed on a JEOL JMS-HX110/100A HF mass spec-
trometer using a 3-nitrobenzyl alcohol (NBA) matrix
and Xe ionizing gas, all being within 5 ppm of theoreti-
cal values.
˚
Three different PhTX-56 analogs, which incorporated
13C, 15N, or both into the polyamine moiety, were de-
signed. Increasing evidence shows that the polyamine
moiety determines both selectivity and, to some extent,
the potency, and previous studies show an almost non-se-
lective interaction of the aromatic head-group. Therefore,
it is of major interest to study the interaction of the poly-
amine portion of PhTX-56 with the pore-forming region
of the AMPA receptor. Labeled diamine precursors,
which were either commercially available or prepared in
two steps, were bound to a solid phase and using previ-
ously developed solid phase synthetic methods, the three
compounds [13C2]PhTX-56 (18), [15N2]PhTX-56 (19),
and [13C2,15N2]PhTX-56 (20) were synthesized in good
yield.
Recent advances in biological NMR spectroscopic tech-
niques are allowing for molecular level studies of ligand–
receptor interactions.25 Although transmembrane pro-
teins, such as AMPA receptors, are particularly difficult
to study owing to their lipid-embedded domains, there
are a few examples of how these obstacles could be over-
come, for example by studying fragments of the recep-
tor.25c The 13C and 15N ligands prepared in this study
can be applied in a similar fashion, thereby providing
detailed information of ligand–receptor interaction.
4. Conclusion
The polyamine toxin derivative, PhTX-56 (3), is an
important tool for the study of the pore-forming region
of AMPA receptors. Here, we have prepared a range of
labeled analogs of PhTX-56 that allow for molecular
level studies of ligand–receptor interactions, as well as a
range of other studies pertaining to AMPA receptor
structure and function.
5.1. 1,4-Dicyanobutane (11)
To a solution of 1,4-dibromobutane (2.16 g, 10 mmol)
and 18-crown-6 (0.22 g, 0.83 mmol) in dry acetonitrile
(5 mL) was added dry KCN (2.61 g, 40 mmol). The mix-
ture was refluxed for 15 h with vigorous magnetic stir-
ring. After the addition of 15 mL water, the mixture
was extracted three times with CH2Cl2 (5 mL). The com-
bined extracts were dried (MgSO4) and concentrated in
vacuo to give a colorless oil, which was purified by silica
gel column chromatography (EtOAC/Pet. ether
[3H]PhTX-56 (10) was prepared in high radiochemical
purity and high specific activity, and results from studies
employing 10 will be reported in due course. Three ana-
logs of PhTX-56, where 13C- and 15N-labels were incor-
porated into the polyamine moiety, as in compounds 18,
19, and 20, were also prepared; these compounds are
useful probes to study ligand–receptor interactions by
means of biological NMR spectroscopy.
1
1:5 ! 1:2) to give a colorless oil (1.06 g, 96%); H and
13C NMR as previously reported.26
5.1.1. [13C2]1,4-Dicyanobutane (11a). Yield: 392 mg, 95%;
1H NMR (CDCl3) d 1.81–1.90 (m, 4H), 2.42–2.52 (m,
4H). 13C NMR (CDCl3): d 15.6, 16.3, 23.7 (2C), 118.6
(2C).
5. Experimental
Resin-bound diamines (trityl chloride resin, 1% divinyl-
benzene, 200–400 mesh) and (S)-Fmoc-O-(tert-butyl)ty-
rosine were obtained from Novabiochem (La¨ufelingen,
Switzerland). (S)-Fmoc-tyrosine(3,5-I2)-OH was ob-
tained from Advanced ChemTech (Louisville, KY,
USA). All other starting materials were obtained com-
mercially from Aldrich or Fluka. All reactions on solid
phase were carried out using Bohdan MiniBlocks from
Mettler-Toledo (Columbus, OH, USA). All starting
materials and solvents were used without further purifi-
5.1.2. [13C2,15N2]1,4-Dicyanobutane (11b). Yield: 343 mg,
1
82%; H NMR (CDCl3) d 1.75–1.98 (m, 4H), 2.41–2.62
(m, 4H). 13C NMR (CDCl3): d 16.6, 17.4, 24.8 (2C),
119.6 (2C).
5.2. 1,6-Diaminohexane (12)
To a suspension of lithium aluminum hydride (0.19 g,
5 mmol) in dry diethyl ether (10 mL) under nitrogen at

5110
T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
0 ꢂC was added a solution of 1,4-dicyanobutane (11,
0.22 g, 2 mmol) in dry diethyl ether (7 mL). The mixture
was heated under reflux for 1 h and then stirred at room
temperature for 24 h. Water (1 mL) was then added to
hydrolyze the remaining lithium aluminum hydride.
The mixture was filtered and the granular solid was
washed several times with diethyl ether and CH2Cl2.
The organic layers were evaporated to give yellow crys-
tals (163 mg, 70%), which were used without further
purification. 1H and 13C NMR as previously reported.27
DMF/CH2Cl2 (1:1) (4 mL) was added a solution of 1-
N-(trimethylsilyl)ethoxycarbonyl-6-aminohexane (13,
104 mg, 0.4 mmol) in DMF/CH2Cl2 (1:1) (4 mL). The
reaction mixture was stirred at room temperature for
2 h. The resin was quenched with MeOH (4 mL), drained,
washed with DMF (3 · 5 mL), CH2Cl2 (3 · 5 mL),
MeOH (3 · 5 mL), and CH2Cl2 (3 · 5 mL), and dried in
vacuo to give resin-bound 1-N-(trimethylsilyl)ethoxycar-
bonyl-6-aminohexane (14). The above resin was suspend-
ed in dry THF (5 mL) under nitrogen at 50 ꢂC. A solution
of TBAF (1.0 mL, 1 M in THF) was added slowly and the
mixture was stirred at 50 ꢂC for 30 min. The resin was
drained, washed with DMF (3 · 2.5 mL), CH2Cl2
(3 · 2.5 mL), MeOH (3 · 2.5 mL), and CH2Cl2
(3 · 2.5 mL), and dried in vacuo to give 5.
5.2.1. [13C2]1,6-Diaminohexane (12a). Yield: 67 mg, 67%;
1H NMR (CDCl3) d 1.11–1.85 (m, 8H), 2.67 (dt,
J = 132.9/6.8 Hz, 4H).
5.2.2. [13C2,15N2]1,6-Diaminohexane (12b). Yield: 133 mg,
1
36%; H NMR (CDCl3) d 1.05–1.79 (m, 8H), 2.68 (dt,
J = 133.0/6.6 Hz, 4H).
Similar reactions were carried out treating trityl chloride
resin with [13C2]1-N-(trimethylsilyl)ethoxycarbonyl-
6-aminohexane (13a), [15N2]1-N-(trimethylsilyl)ethoxy-
carbonyl-6-aminohexane (13b), and [13C2,15N2]1-
N-(trimethylsilyl) ethoxycarbonyl-6-aminohexane (13c),
respectively, to give resin-bound [13C2]1,6-diaminohex-
ane (5a), [15N2]1,6-diaminohexane (5b), and [13C2,
15N2]1,6-diaminohexane (5c), respectively, which were
reacted according to the procedure described below.
5.3. 1-N-(Trimethylsilyl)ethoxycarbonyl-6-aminohexane
(13)
To a solution of 1,6-diaminohexane (12, 0.511 g,
5 mmol) in CH2Cl2 (8 mL) were added triethylamine
(1.4 mL, 10 mmol) and a solution of 2-(trimethylsilyl)-
ethyl 4-nitrophenyl carbonate (1.56 g, 5.5 mmol) in
CH2Cl2 (3 mL). The reaction mixture was stirred over-
night at room temperature. The solvent was evaporated
in vacuo and the resulting yellow oil was redissolved in
CH2Cl2 and washed several times with saturated NaCl,
saturated NaHCO3, and 2 M NaOH until the organic
phase was only weakly yellow. The organic phase was
dried (MgSO4) and concentrated in vacuo to give yellow
crystals (1.06 g, 86%), which were used without further
purification. 1H CDCl3: d 0.02 [s, Si(CH3)3], 0.91 (t,
J = 8.5 Hz, SiCH2), 1.24–1.32 (m, 4H), 1.49–1.62 (m,
4H), 2.60 (t, J = 6.9 Hz, 2H), 3.14 (t, J = 7.1 Hz, 2H),
4.11 (t, J = 7.9 Hz, OCO–CH2), 4.31 (br s, NH).
5.5. Synthesis of PhTX-56 (3) and derivatives.
General procedure
Resin-bound 1,6-diaminohexane (0.16 mmol) was sus-
pended in CH2Cl2 (4 mL). Diisopropylethylamine
(DIEA) (0.96 mmol) and o-nitrobenzenesulfonyl chlo-
ride (0.64 mmol) were added successively and the reac-
tion mixture was stirred under nitrogen at room
temperature for 3 h. The resin was drained, washed with
DMF (3 · 2.5 mL), CH2Cl2 (3 · 2.5 mL), MeOH
(3 · 2.5 mL), and CH2Cl2 (3 · 2.5 mL), and dried in vac-
uo. The above resin was suspended in dry THF/CH2Cl2
(1:1) (2 mL) under nitrogen. A solution of 3-[N-[2-(trim-
ethylsilyl)ethoxycarbonyl]amino]pentanol, prepared as
previously described,12d (0.8 mmol) in dry THF/CH2Cl2
(1:1) (1 mL), tributylphosphine (0.8 mmol), and a solu-
tion of 1,10-(azadicarbonyl)-dipiperidine (ADDP) in
dry THF/CH2Cl2 (1:1) (1 mL) were added successively.
The reaction mixture was stirred at room temperature
under nitrogen for 3 h. The resin was drained, washed
with DMF (3 · 2.5 mL), CH2Cl2 (3 · 2.5 mL), MeOH
(3 · 2.5 mL), and CH2Cl2 (3 · 2.5 mL), and dried in vac-
uo. The procedure was repeated two more times, and the
resulting resin was dried in vacuo. The above resin was
suspended in dry THF (5 mL) under nitrogen at 50 ꢂC.
A solution of TBAF (1 M in THF, 0.64 mmol) was add-
ed slowly, and the reaction mixture was stirred at 50 ꢂC
for 30 min. The resin was drained, washed with DMF
(3 · 2.5 mL), CH2Cl2 (3 · 2.5 mL), MeOH (3 ·
2.5 mL), and CH2Cl2 (3 · 2.5 mL), and dried in vacuo.
A solution of an N-Fmoc-protected (S)-tyrosine deriva-
tive (0.48 mmol) and HATU (0.48 mmol) in DMF
(1 mL), followed by a solution of collidine (0.72 mmol)
in DMF (0.5 mL), was added to the resin. The reaction
mixture was stirred at room temperature for 2 h, and the
resin was subsequently washed with DMF (3 · 2.5 mL),
CH2Cl2 (3 · 2.5 mL), MeOH (3 · 2.5 mL), and CH2Cl2
5.3.1. [13C2]1-N-(Trimethylsilyl)ethoxycarbonyl-6-amino-
1
hexane (13a). Yield: 130 mg, 87%; H NMR (CDCl3) d
0.04 [s, Si(CH3)3], 0.97 (t, J = 8.5 Hz, SiCH2), 1.27–
1.39 (m, 4H), 1.43–1.67 (m, 4H), 2.69 (qd, J = 5.8/
132.1 Hz, 2H), 3.14 (qd, J = 6.1/136.7 Hz, 2H), 4.13 (t,
J = 8.2 Hz, OCO–CH2), 4.59 (br s, NH).
5.3.2. [15N2]1-N-(Trimethylsilyl)ethoxycarbonyl-6-amino-
1
hexane (13b). Yield: 215 mg, 64%; H NMR (CDCl3) d
0.04 [s, Si(CH3)3], 0.97 (t, J = 8.5 Hz, SiCH2), 1.28–1.55
(m, 8H), 2.67 (t, J = 6.8 Hz, 2H), 3.05–3.21 (m, 2H),
4.13 (t, J = 8.2 Hz, OCO–CH2), 4.61 (dm, J = 90 Hz).
5.3.3. [13C2,15N2]1-N-(Trimethylsilyl)ethoxycarbonyl-6-
aminohexane (13c). Yield: 144 mg, 27%; 1H NMR
(CDCl3) d 0.01 [s, Si(CH3)3], 0.94 (t, J = 8.4 Hz, SiCH2),
1.17–1.56 (m, 8H), 2.64 (d, J 135.1 Hz, 2H), 3.20 (dm,
J = 137.0 Hz, 2H), 4.10 (t, J = 8.3 Hz, OCO–CH2),
4.71 (dm, J = 90 Hz).
5.4. Synthesis of resin-bound 1,6-diaminohexane (5)
To a suspension of trityl chloride resin (0.2 mmol, 125 mg
loading 1.6 mmol/g) and Et3N (139 lL, 0.8 mmol) in

T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
5111
(3 · 2.5 mL). The product was treated with 20% piperi-
5.7. Synthesis of [3H]PhTX-56 (10)24
dine in DMF (v/v, 2 mL) and the reaction mixture was
agitated for 3 min at room temperature. The resulting
resin was washed with DMF (3 · 2.5 mL), treated again
with 20% piperidine in DMF (v/v, 2 mL) for 20 min, and
then washed with DMF (3 · 2.5 mL), CH2Cl2
(3 · 2.5 mL), MeOH (3 · 2.5 mL), and CH2Cl2
(3 · 2.5 mL), and dried in vacuo. A solution of butyric
acid (0.48 mmol) and HATU (0.48 mmol) in DMF
(1 mL), followed by a solution of collidine (0.72 mmol)
in DMF (0.5 mL), was added to the resin. The mixture
was agitated at room temperature for 2 h. The resin
was drained, washed with DMF (3 · 2.5 mL), CH2Cl2
(3 · 2.5 mL), MeOH (3 · 2.5 mL), and CH2Cl2
(3 · 2.5 mL), and dried in vacuo. The resin was
treated with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU,
0.6 mmol) in DMF (1 mL) and mercaptoethanol
(0.6 mmol) in DMF (1 mL) for 30 min. The resin was
drained and washed with DMF (5 · 2.5 mL). The proce-
dure was repeated 3 more times. The resin was washed
with DMF (3 · 2.5 mL), CH2Cl2 (3 · 2.5 mL), MeOH
(3 · 2.5 mL), and CH2Cl2 (3 · 2.5 mL), and then treated
with a solution of CH2Cl2/TFA/triisopropylsilane/H2O
(47.5:47.5:2.5:2.5 v/v, 2 mL) for 2 h. The resin was
drained and washed with MeOH (2 mL) and CH2Cl2
(2 mL). The solutions of the cleaved product and the
washings were combined and evaporated in vacuo to
give a sticky solid, which was purified by preparative
HPLC to give the final product as a yellowish oil.
The reaction mixture containing 4 (10 mg, 14.6 lmol) in
anhydrous DMF (0.8 mL), DIPA (10 lL, 57 lmol), and
10% palladium on charcoal (20 mg) was frozen at
ꢀ80 ꢂC and evacuated three times. Tritium gas was intro-
duced at atmospheric pressure and the reaction mixture
was stirred at room temperature overnight. After the
removal of un-reacted tritium gas, the reaction mixture
was filtered. The crude product was purified by prepara-
tive HPLC to give [3H]PhTX-56. The HPLC conditions
were as follows: Zorbax SB-C18 (4.5 · 250 mm) column,
using a gradient, with A: water and 0.1% TFA, and B: ace-
tonitrile, gradient: 5–50% in 20 min. Flow rate was 1 mL/
min, with a radiodetector (3Hb-RAM) and UV detection
at 235 nm. The radiochemical purity was 99.52% and the
specific activity as determined by UV detection was found
to be 50.2 Ci/mmol.
5.8. Synthesis of [13C2]PhTX-56 (18), [15N2]PhTX-56
(19), and [13C2,15N2]PhTX-56 (20)
The compounds were synthesized according to the gen-
eral procedure using resin-bound 1,6-diaminohexane as
starting materials. Resin-bound 1,6-diaminohexanes
15, 16, and 17 were used for the synthesis of 18, 19,
and 20, respectively (Scheme 4).
5.8.1. (S)-N-[5-[(6-Aminohexyl)amino]pentyl]-4-hydroxy-
a-[(1-oxobutyl)amino]benzene-propaneamide bis(trifluo-
1
5.5.1. (S)-N-[5-[(6-Aminohexyl)amino]pentyl]-4-hydroxy-
a-[(1-oxobutyl)amino]benzene-propaneamide bis(trifluo-
roacetate) (diiodo-PhTX-56, 4). Yield: 35%; H NMR
roacetate) ([13C2]PhTX-56, 18). Yield: 32%; H NMR
(CD3OD) d 0.86 (t, J = 7.4 Hz, 3H), 1.18–1.31 (m,
2H), 1.37–1.77 (m, 14H), 2.17 (t, J = 7.1 Hz, 2H),
2.68–2.84 (m, 3H), 2.89–3.00 (m, 3H), 3.01–3.27 (m,
4H), 4.42 (t, J = 7.8 Hz, 1H), 6.69 (d, J = 8.5 Hz, 2H),
7.04 (d, J = 8.5 Hz, 2H); HRMS calcd for C22
13C2H43N4O3 [M+1] 437.6006; found: 437.6019.
1
(CD3OD) d 0.86 (t, J = 7.4 Hz, 3H), 1.22–1.69 (m,
16H), 2.18 (t, J = 7.3 Hz, 2H), 2.78 (dd, JAB = 13.6 Hz,
JAX = 8.0 Hz, 1H), 2.84–3.23 (m, 9H), 4.42 (t,
J = 7.6 Hz, 1H), 7.61 (s, 2H); HRMS calcd for
C24H41I2N4O3 [M+1] 687.1268; found: 687.1249.
5.8.2. (S)-N-[5-[(6-Aminohexyl)amino]pentyl]-4-hydroxy-
a-[(1-oxobutyl)amino]benzene-propaneamide bis(trifluoro-
acetate) ([15N2]PhTX-56, 19). Yield: 29%; 1H NMR
(CD3OD) d 0.86 (t, J = 7.3 Hz, 3H), 1.18–1.28 (m, 2H),
1.38–1.75 (m, 14H), 2.17 (t, J = 7.6 Hz, 2H), 2.79 (dd,
J = 13.6/8.1, 1H), 2.80–3.22 (m, 9H), 4.43 (t, J = 7.9 Hz,
1H), 6.69 (d, J = 8.5 Hz, 2H), 7.04 (d, J = 8.5 Hz, 2H);
HRMS calcd for C24H43N215N2O3 [M+1] 437.6021;
found: 437.6029.
5.6. Synthesis of PhTX-56 (3) and D2-PhTX-56 (9)
A dried, hydrogen- or deuterium-flushed 25 mL two-
necked flask equipped with a balloon filled with either
hydrogen or deuterium was charged with a suspension
of Pd/C 10% (5 mg) in DMF (0.5 mL) and a solution
of diiodo-PhTX-56 (4, 30 mg, 0.04 mmol) in DMF
(1 mL). The mixture was stirred vigorously for 20 h.
The reaction mixture was filtered over a patch of Celite
and washed with EtOAc. The solvents were removed in
vacuo and the resulting crude product was purified by
preparative HPLC-MS.
5.8.3. (S)-N-[5-[(6-Aminohexyl)amino]pentyl]-4-hydroxy-
a-[(1-oxobutyl)amino]benzene-propaneamide bis(trifluo-
roacetate) ([13C2,15N2]PhTX-56, 20). Yield: 35%; 1H
NMR d 0.84 (t, J = 7.2 Hz, 3H), 1.14–1.28 (m, 2H),
1.31–1.80 (m, 14H), 2.21 (t, J = 6.9 Hz, 2H), 2.58–2.79
(m, 3H), 2.79–2.97 (m, 3H), 3.04–3.24 (m, 4H), 4.47 (t,
J = 8.1 Hz, 1H), 6.63 (d, J = 8.1 Hz, 2H), 7.09 (d,
J = 8.2 Hz, 2H); HRMS calcd for C22 13C2H43
N215N2O3 [M+1] 439.3343; found: 439.3329.
5.6.1. (S)-N-[5-[(6-Aminohexyl)amino]pentyl]-4-hydroxy-
a-[(1-oxobutyl)amino]benzene-propaneamide bis(trifluo-
roacetate) (PhTX-56, 3). Yield: 35%; 1H and 13C
NMR as previously reported.19d
5.6.2. (S)-N-[5-[(6-Aminohexyl)amino]pentyl]-4-hydroxy-
a-[(1-oxobutyl)amino]benzene(3,5-D2)-propaneamide bis-
(trifluoroacetate) (D2-PhTX-56, 9). Yield: 33%; 1H NMR
(CD3OD) d 0.87 (t, J = 7 Hz, 3H), 1.35–1.60 (m, 16H),
2.17 (t, J = 7 Hz, 2H), 2.74–2.82 (m, 1H), 2.84–3.23
(m, 9H), 4.45 (t, J = 7 Hz, 1H), 7.04 (s, 2H).
Acknowledgments
Technical assistance of Ms. Dorthe Kroghave Toftdahl
˚
and Mr. Erik Damgard, Department of Spectroscopy,

5112
T. F. Andersen et al. / Bioorg. Med. Chem. 13 (2005) 5104–5112
16. (a) http://www.ebixa.com; (b) Ferris, S. H. Expert Opin.
Pharmacother. 2003, 4, 2305.
Medicinal Chemistry Research, H. Lundbeck A/S
(Valby, Denmark), is appreciated.
17. Eldefrawi, A. T.; Eldefrawi, M. E.; Konno, K.; Mansour,
N. A.; Nakanishi, K.; Oltz, E.; Usherwood, P. N. R. Proc.
Natl. Acad. Sci. USA 1988, 85, 4910.
References and notes
18. (a) Blaschke, M.; Keller, B. U.; Rivosecchi, R.; Hollmann,
M.; Heinemann, S.; Konnerth, A. Proc. Natl. Acad. Sci.
USA 1993, 90, 6528; (b) Herlitze, S.; Raditsch, M.;
Ruppersberg, J. P.; Jahn, W.; Monyer, H.; Schoepfer,
R.; Witzemann, V. Neuron 1993, 10, 1131; (c) Washburn,
M. S.; Dingledine, R. J. Pharmacol. Exp. Ther. 1996, 278,
669; (d) Brackley, P. T. H.; Bell, D. R.; Choi, S. K.;
Nakanishi, K.; Usherwood, P. N. R. J. Pharmacol. Exp.
Ther. 1997, 266, 1573.
1. (a) Gerner, E. W.; Meyskens, F. L. Nat. Rev. Cancer 2004,
4, 781; (b) Thomas, T.; Thomas, T. J. J. Cell. Mol. Med.
2003, 7, 113; (c) Igarashi, K.; Kashiwagi, K. Biochem.
Biophys. Res. Commun. 2000, 271, 559; (d) Casero, R. A.;
Woster, P. M. J. Med. Chem. 2001, 44, 1.
2. (a) Williams, K. Cell. Signal. 1997, 9, 1; (b) Johnson, T. D.
Trends Pharmacol. Sci. 1996, 17, 22.
3. (a) Bra¨uner-Osborne, H.; Egebjerg, J.; Nielsen, E. Ø.;
Madsen, U.; Krogsgaard-Larsen, P. J. Med. Chem. 2000,
43, 2609; (b) Ozawa, S.; Kamiya, H.; Tsuzuki, K. Prog.
Neurobiol. 1998, 54, 581; (c) Dingledine, R. K. B.; Bowie,
D.; Traynelis, S. F. Pharmacol. Rev. 1999, 51, 7.
4. Parsons, C. G.; Danysz, W.; Quack, G. Drug News
Perspect. 1998, 11, 523.
5. (a) Bliss, T. V. P.; Collingridge, G. L. A. Nature 1993, 361,
31; (b) Malenka, R. C.; Nicoll, R. A. Science 1999, 285,
1870.
19. (a) Strømgaard, K.; Brierley, M. J.; Andersen, K.; Sløk, F.
A.; Mellor, I. R.; Usherwood, P. N. R.; Krogsgaard-
Larsen, P.; Jaroszewski, J. W. J. Med. Chem. 1999, 42,
5224; (b) Strømgaard, K.; Brier, T. J.; Andersen, K.;
Mellor, I. R.; Saghyn, A.; Tikhonov, D.; Usherwood, P.
N. R.; Krogsgaard-Larsen, P.; Jaroszewski, J. W. J. Med.
Chem. 2000, 43, 4526; (c) Mellor, I. R.; Brier, T. J.;
Pluteanu, F.; Strømgaard, K.; Saghyan, A.; Eldursi, N.;
Brierley, M. J.; Andersen, K.; Jaroszewski, J. W.; Kro-
gsgaard-Larsen, P.; Usherwood, P. N. R. Neuropharma-
cology 2003, 44, 70; (d) Kromann, H.; Krikstolaityte, S.;
Andersen, A. J.; Andersen, K.; Krogsgaard-Larsen, P.;
Jaroszewski, J. W.; Egebjerg, J.; Strømgaard, K. J. Med.
Chem. 2002, 45, 5745.
6. Kemp, N.; Bashir, Z. I. Prog. Neurobiol. 2001, 65, 339.
7. (a) Carroll, R. C.; Beattie, E. C.; von Zastrow, M.;
Malenka, R. C. Nat. Rev. Neurosci. 2001, 2, 315; (b)
Luscher, C.; Frerking, M. Trends Neurosci. 2001, 24, 665;
¨
(c) Sheng, M.; Lee, S. H. Cell 2001, 105, 825; (d) Malinow,
R.; Malenka, R. C. Annu. Rev. Neurosci. 2002, 25, 103.
8. Lees, G. J. Drugs 2000, 59, 33.
9. (a) Gill, R.; Lodge, D. Int. Rev. Neurobiol. 1997, 40, 197; (b)
Pellegrini-Giampietro, D. E.; Gorter, J. A.; Bennet, M. V.
L.; Zukin, R. Z. T. Trends Neurosci. 1997, 20, 464; (c)
Bleakman, D.; Lodge, D. Neuropharmacology 1998, 37,
1187; (d) Weiss, J. H.; Sensi, S. L. Trends Neurosci. 2000, 23,
365.
10. (a) Armstrong, N.; Sun, Y.; Chen, G.-Q.; Gouaux, E.
Nature 1998, 395, 913; (b) Gouaux, E. J. Physiol. 2004,
554, 249.
11. Sun, Y.; Olson, R.; Horning, M.; Armstrong, N.; Mayer,
M. L.; Gouaux, E. Nature 2002, 417, 245.
20. Manov, N.; Tzouros, M.; Bigler, L.; Bienz, S. Tetrahedron
2004, 60, 2387.
21. (a) Choi, S.-K.; Kalivretenos, A. G.; Usherwood, P. N.
R.; Nakanishi, K. Chem. Biol. 1995, 2, 23; (b) Bixel,
M. G.; Krauss, M.; Liu, Y.; Bolognesi, M. L.; Rosini,
M.; Mellor, I. S.; Usherwood, P. N. R.; Melchiorre, C.;
Nakanishi, K.; Hucho, F. Eur. J. Biochem. 2000, 267,
110; (c) Bixel, M. G.; Weise, C.; Bolognesi, M. L.;
Rosini, M.; Brierly, M. J.; Mellor, I. R.; Usherwood,
P. N. R.; Melchiorre, C.; Hucho, F. J. Biol. Chem.
2001, 276, 6151.
22. Strømgaard, K.; Andersen, K.; Ruhland, T.; Krogsgaard-
Larsen, P.; Jaroszewski, J. W. Synthesis 2001, 877.
23. Andersen, T. F.; Strømgaard, K. Tetrahedron Lett. 2004,
45, 7929.
12. Jin, R. S.; Banke, T. G.; Mayer, M. L.; Traynelis, S. F.;
Gouaux, E. Nat. Neurosci. 2003, 6, 803.
13. (a) Madden, D. R. Curr. Opin. Drug Discov. Dev. 2002, 5,
741; (b) Wollmuth, L. P.; Sobolevsky, A. I. Trends
Neurosci. 2004, 27, 321.
24. The tritiation reaction was carried out by J. Selvasekaran
and S. Gupta, American Radiolabeled Chemicals, Inc. (St.
Loius, MO, USA). Contact in Europe: ANAWA Trading
SA (Wangen, Switzerland).
14. (a) Chen, G. Q.; Cui, C.; Mayer, M. L.; Gouaux, E.
Nature 1999, 402, 817; (b) Mayer, M. L.; Olson, R.;
Gouaux, E. J. Mol. Biol. 2001, 311, 815.
15. (a) Strømgaard, K.; Andersen, K.; Krogsgaard-Larsen, P.;
Jaroszewski, J. W. Mini-Rev. Med. Chem. 2001, 1, 317; (b)
Strømgaard, K.; Mellor, I. R. Med. Res. Rev. 2004, 24, 589;
(c) Mellor, I. R.; Usherwood, P. N. R. Toxicon 2004, 43,
493.
25. (a) Meyer, B.; Peters, T. Angew. Chem. Int. Ed. 2003, 42,
864; (b) Villar, H. O.; Yan, J. L.; Hansen, M. R. Curr.
Opin. Chem. Biol. 2004, 8, 387; (c) Opella, S. J.; Marassi,
F. M. Chem. Rev. 2004, 104, 3587.
26. Park, H.-K.; Kim, I. B.; Chin, C. S. J. Organomet. Chem.
1987, 326, 431.
27. Balaban, A. T.; Dinculescu, A.; Elguero, J.; Faure, R.
Magn. Reson. Chem. 1985, 23, 553.