Synthesis of novel GABA uptake inhibitors. 3. Diaryloxime and diarylvinyl ether derivatives of nipecotic acid and guvacine as anticonvulsant agents

Article abstract of DOI:10.1021/jm981027k

(3R)-1-[4,4-bis(3-methyl-2-thienyl)-3-butenyl]-3-piperidinecarboxylic acid 1 (tiagabine, Gabitril) is a potent and selective γ-aminobutyric acid (GABA) uptake inhibitor with proven anticonvulsant efficacy in humans. This drug, which has a unique mechanism

Full text of DOI:10.1021/jm981027k

J . Med. Chem. 1999, 42, 3447-3462
3447
Articles
Syn th esis of Novel GABA Up ta k e In h ibitor s. 3. Dia r yloxim e a n d Dia r ylvin yl
Eth er Der iva tives of Nip ecotic Acid a n d Gu va cin e a s An ticon vu lsa n t Agen ts1
Lars J . S. Knutsen,* Knud Erik Andersen, J esper Lau, Behrend F. Lundt, Rodger F. Henry,‡
Howard E. Morton,‡ Lars Nærum, Hans Petersen,§ Henrik Stephensen, Peter D. Suzdak,|
Michael D. B. Swedberg, Christian Thomsen, and Per O. Sørensen∇
Health Care Discovery and Development, Novo Nordisk A/ S, Novo Nordisk Park, DK-2760 Måløv, Denmark
Received April 21, 1998
(3R)-1-[4,4-bis(3-methyl-2-thienyl)-3-butenyl]-3-piperidinecarboxylic acid 1 (tiagabine, Gabitril)
is a potent and selective γ-aminobutyric acid (GABA) uptake inhibitor with proven anticon-
vulsant efficacy in humans. This drug, which has a unique mechanism of action among
marketed anticonvulsant agents, has been launched for add-on treatment of partial seizures
with or without secondary generalization in patients >12 years of age. Using this new agent
as a benchmark, we have designed two series of novel GABA uptake inhibitors of remarkable
potency, using a putative new model of ligand interaction at the GABA transporter type 1
(GAT-1) uptake site. This model involves the postulated interaction of an electronegative region
in the GABA uptake inhibitor with a positively charged domain in the protein structure of the
GAT-1 site. These two novel series of anticonvulsant agents contain diaryloxime or diarylvinyl
ether functionalities linked to cyclic amino acid moieties and were derived utilizing the new
model, via a series of design steps from the known 4,4-diarylbutenyl GABA uptake inhibitors.
The new compounds are potent inhibitors of [3H]-GABA uptake in rat brain synaptosomes in
vitro, and their antiepileptic potential was demonstrated in vivo by their ability to protect
against seizures induced by the benzodiazepine receptor inverse agonist methyl 4-ethyl-6,7-
dimethoxy-â-carboline-3-carboxylate (DMCM) in mice. From structure-activity studies of these
new GABA uptake inhibitors, we have shown that insertion of an ether oxygen in conjugation
with the double bond in tiagabine (Ki ) 67 nM) improves in vitro potency by 5-fold to 14 nM.
Ch a r t 1. Structures of the Main Reference GABA
Uptake Inhibitors
In tr od u ction
γ-Aminobutyric acid (GABA) is recognized as the
principal brain inhibitory neurotransmitter,2 and this
amino acid is intrinsically involved in the mechanism
of action of a number of marketed and development
stage anticonvulsant drugs.3-9 Two cyclic amino acid
derivatives which act via inhibition of the uptake of
GABA in the central nervous system (CNS) have been
investigated in human clinical trials. These are (3R)-
1-[4,4-bis(3-methyl-2-thienyl)-3-butenyl]-3-piperidine-
carboxylic acid10,11 1 (tiagabine, NNC 05-0328) and 1-[2-
[bis[4-(trifluoromethyl)phenyl]methoxy]ethyl]-1,2,5,6-
tetrahydro-3-pyridinecarboxylic acid (CI-966)12-16
2
(Chart 1). In preclinical rodent models of epilepsy, both
of these relatively lipophilic amino acid derivatives
exhibit a promising seizure protection profile, with ED50
values of 2.6 and 3.0 mg/kg, respectively, following
intraperitoneal (i.p.) dosing in amygdala kindled rats.17
Tiagabine has now been more extensively investi-
gated.18-24
Evidence is mounting that tiagabine, now a marketed
drug, is acting centrally as an inhibitor of GABA uptake
modulating the removal of GABA from the synaptic
cleft,24,25 and clinical findings indicating effectiveness
as an add-on treatment of partial seizures with or
without secondary generalization26-28 have recently
been summarized.29 Tiagabine appears to show some
* Current address for correspondence: Lars J . S. Knutsen, Ph.D.,
Cerebrus Limited, Oakdene Court, Winnersh, WOKINGHAM, Berk-
shire, RG41 5UA, U.K. Tel: +44-(0)-118-989-9303. Fax: +44-(0)-118-
989-9300. E-mail: L.Knutsen@Cerebrus.ltd.uk.
‡ Abbott Laboratories, 1401 Sheridan Road, North Chicago, IL
60064-4000.
§ Current address: H. Lundbeck A/S, Ottiliavej 9, DK-2500 Copen-
hagen - Valby, Denmark.
| Current address: Guilford Pharmaceuticals, 6611 Tributary Street,
Baltimore, MD 21224.
Current address: Astra Pain Control, Department of Pharmacol-
ogy, S-15185 Sodertalje, Sweden.
∇ P.O.S. has now changed name to Per O. Huusfeldt.
10.1021/jm981027k CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/05/1999

3448 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Ch a r t 2. Structures of Other GABA Uptake Inhibitors
Knutsen et al.
clinical advantages over current epilepsy therapy, par-
ticularly in terms of safety and side-effect profile.30-34
The observations24 from the preclinical animal models
have therefore been borne out.
F igu r e 1. Features of diaryloxime and diarylvinyl ether
GABA uptake inhibitors.
vulsant efficacy, compared to 1-3. Two series of novel
GABA uptake inhibitors with general structures 8, 9,
and 10 have been derived using a new proposed model
of the GABA uptake inhibitor binding mode. This new
model postulates the presence of a positively charged
domain in the protein structure of the site controlling
the GABA transporter type 1 (GAT-1).48-51 We suggest
that this domain interacts with an electronegative
moiety which is part of the linker (Figure 1) in the
potent CNS-active compounds comprising cyclic amino
acids N-substituted via this linker to two aromatic
functions. The lipophilic N-substituents have previously
typically contained 1,1-diarylethylene and diarylmethyl
ether functions as the electronegative moieties (Chart
1).
The increasing electronegative character of the linker
(Figure 1) is coupled to the increase in potency of the
compounds as inhibitors of [3H]-GABA uptake in vitro.39
Five-atom linker molecules 11, 12, and 13 show a
corresponding increase in linker electronegativity, il-
lustrated by the red surface shading which is apparent
in the electrostatic potential diagrams for 12 and 13 in
Figure 2.52
Another conclusion from the above structural analysis
is that extension of the electronegative region further
into the middle of the linker leads to greater in vitro
potency when compared to butenyl and diarylmethyl
ether derivatives. These new insights into the under-
standing of the structure of GABA uptake inhibitors
have led to the logical design of the new agents
represented by the general structures 9 and 10 (Chart
3). These diaryloximes and diarylvinyl ethers with
improved potency show promise as a potential treat-
ment for partial clonic/tonic seizures in humans.
At the outset of our investigations in the GABA
uptake and/or re-uptake field only one structural type
of CNS-active GABA uptake inhibitor was known,
illustrated by 1-(4,4-diphenyl-3-butenyl)-3-piperidine-
carboxylic acid35-38 3 (SKF 89976A, Chart 1). This
diphenylbutenyl derivative formed the starting point for
the design of 1, but simple isosteric replacement of
phenyl with thiophene is not the whole explan-
ation behind the improved activity of 1 when compared
to 3. We found subsequently that the presence of the
two “ortho” methyl groups in 1 is very significant in
progressing to drugs of higher potency than 3. Cor-
respondingly, the introduction of “ortho” methyl groups
in the diaryl moiety of 3, giving the novel diarylbutene
4 (Chart 2), increases in vitro potency 5-fold compared
to 3, when compared in Fjalland’s assay39 measuring
inhibition of [3H]-GABA uptake in vitro.
The 4,4-diarylbutenyl structural class was followed
chronologically by diarylmethoxyethyl derivatives10,13-15,40
of which 2 is a representative. More recently, N’Goka
et al.41 have prepared some structural analogues of 3.
The most potent of these, 5 (Chart 2), has a diarylpropyl
system bonded directly to a carbon of the cyclic amino
acid and exhibits in vitro potency similar to that of 3.
Compound 642 (Chart 2), an analogue of 3 containing a
hydroxyisoxazole as an acid isostere, exhibited a certain
degree of seizure suppression in vivo. The triphenyl-
methyl derivative 7 (Chart 2), with some structural
similarity to 2, has been shown to have micromolar
affinity for the GABA transporter type 4 (GAT-4).43
We report here the synthesis and pharmacological
profiles of two new series of GABA uptake inhibitors,
conceptually related to each other and which incorporate
novel structural elements. These structural features
have resulted in a leap in potency in these new agents
when compared to currently available compounds.
These new anticonvulsants are comprised of diaryl/
heteroaryl-oxime44 and -vinyl ether45 moieties joined by
an aliphatic “linker” to a cyclic amino acid (Figure 1).
Syn th etic Ch em istr y
Two main synthetic routes were used for the prepara-
tion of diaryl/heteroaryl oximes44 of general structure
9 (Chart 3, Schemes 1 and 2). The same overall logic,
with detail changes, was followed for preparation of the
diaryl/heteroaryl vinyl ether derivatives45 of general
structure 10 (Chart 3, Schemes 3 and 4), and one new
synthesis unique for vinyl ether GABA uptake inhibitors
is illustrated (Scheme 5). We refer to one of the methods
New P r op osa l of GAT-1 Bin d in g Mod e
The objectives of this study were to provide structur-
ally novel GABA uptake inhibitors with increased in
vitro potency, and correspondingly improved anticon-

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3449
O-alkylation of a diaryl oxime or a diarylacetaldehyde
(in its enolic anion form) with either a 1,2-dihaloethane
or a N-2-(haloethyl)nipecotic acid ester.
Meth od A (Oxim es). This method is illustrated
(Scheme 1) by the synthesis of 1,2,5,6-tetrahydro-1-[2-
[[(diphenylmethylene)aminooxy]ethyl]-3-pyridinecar-
boxylic acid 1444,47,50,53,54 (NNC 05-0711; also referred
to as NNC-711 or NO-711). Benzophenone oxime 15
could be alkylated under mildly basic conditions using
either 1,2-dibromoethane or 1-bromo-2-chloroethane to
provide either 16 or 17 as oils. These halides could both
be used to alkylate the nitrogen of guvacine ethyl ester
18 to give 19, with the bromide 16 providing a higher
yield as would be predicted. The ester 19 was character-
ized as its crystalline hydrochloride salt. NNC 05-0711
hydrochloride 1447 was obtained following saponification
and hydrochloride salt formation.
The synthesis of compounds 20-23 (Table 3) was
carried out using a modification (Scheme 2) of method
A. To prepare 4-methyl-2-thienyl derivatives it was
found necessary to block the 5-position with a suitable
group, and we opted for a trimethylsilyl group. The
aldehyde 24 was prepared from the Grignard reagent
36 derived from 2-bromo-3-methylthiophene followed by
formylation. Carbinol 25 was obtained by reaction with
2-methylphenylmagnesium bromide, but this reaction
was plagued by the formation of an unwanted ether by-
product, presumably owing to the stability of the car-
bonium ion derived from dehydration of protonated 25.
Some recovery of carbinol was obtained during acid
hydrolysis of 25 and 27, the main function of which was
to effect desilylation prior to pyridinium chlorochromate
oxidation of 26 to the ketone 28. Oxime formation
provided the E- and Z-geometric isomers 29 and 30 in
a ratio of 3:2, and these were separated by column
chromatography and converted via a bromide such as
31 into the target oximes 20-23, using the same
procedure as described in method A.
F igu r e 2. Electrostatic potential calculations52 for molecules
11, 12, and 13. The most electronegative surface is represented
by the red shading (the linker is indicated by the red arrows),
graduating toward the electropositive via yellow and green to
blue as the most electropositive. As proposed, the oxime 12
has a less electronegative region in the linker than the vinyl
ether 13; both are significantly different from the pentenyl
analogue 11 of tiagabine. This is reflected in their activities
as inhibitors of [3H]-GABA uptake in vitro, which are 335, 41,
and 14 nM, respectively.
Ch a r t 3. General Structures of Diaryloxime and
Diarylvinyl Ether GABA Uptake Inhibitors
All of the single enantiomer R- and S-nipecotic acid
derivatives described herein were derived by N-alkyl-
ation of the corresponding R- and S-nipecotic acid ethyl
esters. These esters were prepared in high enantiomeric
purity by resolution of commercially available ethyl
nipecotate by procedures involving recrystallization of
the respective diastereomeric dibenzoyl tartaric acid
salts, as utilized in the synthesis of tiagabine 1.10,55
Sch em e 1
Meth od B (Oxim es). This method (Scheme 3) in-
volves the pivotal R-N-(2-bromoethyl)nipecotic acid
ethyl ester 32, derived from the corresponding alcohol
33 which has been described previously.10 Preparation
of alcohol 33 relies on the availability of resolved
nipecotic acid, ethyl ester.55 The bromide 32 could
conveniently be utilized in the O-alkylation of the oxime
34; the oxime precursor, ketone 35, was formed in two
steps by reaction of o-tolualdehyde and the Grignard
reagent 36 derived from 2-bromo-3-methylthiophene,10,56
followed by oxidation of the resultant alcohol. In some
cases we observed an N-alkylation which competed with
the desired O-alkylation, providing what was apparently
the N-oxide or nitrone 37.57 However, this relatively
polar byproduct was readily removed during column
chromatography. The desired ester 38 was characterized
as a crystalline hydrochloride salt, and saponification
afforded 39. This preparation illustrated the possibilities
A-E, used to prepare each of the target compounds, in
the structure-activity relationship (SAR) analysis (Tables
2-4). The underlying principle behind each of the first
four syntheses (methods A-D) is the base-catalyzed

3450 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Knutsen et al.
Sch em e 2
Sch em e 3
for E- and Z-geometric isomers in the final product 39
or its opposite geometric isomer 40 (Table 2). Fractional
crystallization of the oxime 34 from cyclohexane pro-
vided a single E- or Z-isomer, and either of these could
be carried through to the appropriate isomer of final
product, 39 or 40. Unambiguous spectroscopic geo-
metric isomer identification was not straightforward
despite previous investigations in this field58 (see Re-
sults and Discussion section). Observations recorded in
Table 2 indicate that structures 39 and 40, related as
geometric isomers, possess very similar biological activi-
ties.
tion (for example, Scheme 3) and the diaryl ketones
could also be used as starting points for diaryl oximes.
Second, the intermediate glycidic acid, e.g., 41, obtained
from saponification of 42, was convenient for practical
reasons because a base extraction prior to the decar-
boxylation step helped to improve the purity of the
diarylacetaldehyde 43. Under basic conditions, in this
case using phase transfer catalysis, the stabilized enol
form of acetaldehyde 43 could be O-alkylated to give the
bromoethylvinyl ether 44 (Scheme 4). This ether was
reacted further to afford an intermediate ester and
subsequently the amino acid 45 by the method described
for compound 14 (Scheme 1).
Meth od C (Vin yl Eth er s). A number of different
methods exist for the preparation of diarylacetalde-
hydes.59 However, we opted for the classical Darzen’s
glycidic ester condensation60 (Scheme 4) for a number
of reasons. First, the synthesis utilized ubiquitous diaryl
ketones as starting points, as there are a range of
convenient methods available for diaryl ketone prepara
Meth od D (Vin yl Eth er s). In this very direct
synthetic approach, the R-N-(2-bromoethyl)nipecotic
acid ester hydrobromide 32 was used in alkylation of
the enol anion of diphenylacetaldehyde to provide the
ester 46 and subsequently the acid 47 (Scheme 5).
Meth od E (Vin yl Eth er s). This method (Scheme 6)

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3451
Ta ble 1. Pharmacological Data for Reference GABA Uptake Inhibitors
Sch em e 4
Sch em e 5
Resu lts a n d Discu ssion
As inhibitors of rat synaptosome [3H]-GABA uptake
in vitro, the new compounds described herein exhibit
remarkable potency, with Ki values as low as 12 nM in
some cases (see Tables 2-4), in the assay run essentially
as described by Fjalland.39 This represents a 5-fold
potency improvement over tiagabine 1, the most potent
of the currently available reference compounds (Table
1). The compounds, besides being investigated for their
in vitro profiles, have had their in vivo effects evaluated
in a mouse convulsion model, in which seizures were
induced by the inverse benzodiazepine receptor agonist
methyl 4-ethyl-6,7-dimethoxy-â-carboline-3-carboxylate
(DMCM).46,47 Ataxia (or neurological toxicity) in rodents
was assessed for selected compounds using a mouse
rotarod apparatus.47
involves as its crucial step the reaction of a Grignard
reagent 36 with the ethyl ester 48 of the trityloxy-
ethoxyacetic acid 49. The resultant alcohol 50 was
converted into the tosylate ester 51 and was reacted as
described previously for 44 and 16 to provide the ester
52. The synthesis illustrated is that of 13 (Figure 2 and
Chart 4), one of the most potent GABA uptake inhibitors
in this vinyl ether series, which in turn comprises the
most potent published examples of such agents.
The two new series of related, structurally novel
inhibitors of GABA uptake are illustrated by the
representative structures 13 (Figure 2, Table 4) and 14
(Scheme 1, Table 2), which have Ki values39 of 14 ( 1
and 47 ( 3 nM, respectively (Chart 4). The latter
compound is commercially available,53 and its pharma-
cological profile has been described separately.47,54
Before the discovery of these oxime and vinyl ether

3452 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Knutsen et al.
Ta ble 2. Pharmacological and Physical Data for Oxime GABA Uptake Inhibitors

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3453
Ta ble 2 (Continued)
structures, it was assumed that the optimal length of
the linker (Figure 1) joining the diaryl moiety and the
cyclic amino acid in lipophilic GABA uptake inhibitors
was four atoms. This was certainly the case for the three
most prominent agents 1, 2, and 3 (Chart 1). Indeed,
early preparation of the 5,5-diphenylpentenyl derivative
5331 (Chart 5) provided only a relatively weak (Ki > 1
µM) in vitro inhibitor of [3H]-GABA uptake. The same
trend was observed when 1 was extended from a 4,4-
dithienylbutenyl linker to the 5,5-dithienylpentenyl
derivative 11 (Figure 2 and Chart 5). The in vitro value
for inhibition of [3H]-GABA uptake was likewise shown

3454 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Knutsen et al.
Ta ble 3. Data for 4-Methyl-2-thienyl/2-Methylphenyl Oxime GABA Uptake Inhibitors
to be considerably higher (Ki ) 335 nM) for the homo-
logue 11; the corresponding Ki value for 1 is 67 nM.
oxime derived from the corresponding ketone, allowed
the synthesis of a relatively large range of compounds
for the examination of SAR in this series44 (Tables 2
and 3). The conclusions of this study are that in terms
of in vitro potency the oximes are generally more potent
than diarylbutenes with a corresponding diaryl/hetero-
aryl substitution. Although the linker now comprises
five atoms, the bis(3-methyl-2-thienyl) oxime derivative
12 with R-nipecotic acid as amino acid, corresponding
to 1, is among the most potent in vitro from this oxime
series, with a Ki value of 41 nM for inhibition of [3H]-
GABA uptake. The rank order of potency among the
different cyclic amino acid derivatives (for example, in
Table 2, 60, 61, 14, and 62, 63, 6447) is that the
R-nipecotic acid derivatives are substantially more
potent than the corresponding S-nipecotic acid deriva-
tives,55 and that the guvacines10,69 have generally
similar in vitro potency to the R-nipecotic acid deriva-
tives. The “ortho” effect of aryl/heteroaryl methyl groups,
which is apparent in 1,10 is not as prominent in this
oxime series compared to the butenyl series, illustrated
by the observation that the ditoluyl derivative 6447 (Ki
) 55 ( 4 nM) is not more potent than 1447 (Ki ) 47 (
3 nM). Although exceptions do exist, this may reflect
the longer linker, which presumably affects the binding
of the diaryl moiety at the uptake site. Furthermore,
the increased electronegativity of the oxime linker
compared to a butenyl linker may also have an effect
on binding conformation.
It was therefore surprising to us when five-atom
linker compounds such as 13 and 14 (Chart 4) were
found to possess greater in vitro potency than the known
butenyl derivatives 1, 2, and 3. The step to a five-atom
linker was initially taken in the above oximes and vinyl
ethers because we found that synthesis of the shorter
(i.e. four-atom linker) oxime analogues was not achiev-
able. We further demonstrated that five was the opti-
mum linker length in the oxime and vinyl ether series
by preparing six-atom linker compounds 54 and 55.
These novel structures were found to be only very weak
inhibitors of [3H]-GABA uptake in vitro, with Ki values
in the micromolar range (Chart 5).
The oxime ether functionality has been used previ-
ously in drug design to provide novel biologically active
molecules. Some selected examples include Ridogrel
56,61 a thromboxane synthetase inhibitor, the antineo-
plastic agent Amidox 57,62 the phosphodiesterase in-
hibitor 58,63 muscarinic agonists such as 5964 (Chart 6),
as well as a range of oxime-based cephalosporin anti-
biotics.
The novel anticonvulsants described herein44,45 have
been examined for their in vitro effects as inhibitors of
[3H]-GABA uptake, expressed as Ki values in nM.39
Representative compounds from each series, ordered by
their substitution pattern, are shown in Tables 2-4,
with selected reference GABA uptake inhibitors fea-
tured in Table 1. The tables also contain ED50 values
for protection against DMCM-induced seizures,46,47
utilizing a 30 min pretreatment time in mice. Since new
anticonvulsant agents to be considered as back-ups to
1 should ideally have minimized sedative effects, neu-
rological toxicity in mice assessed in a standard rotarod
test47 has also been included for selected compounds in
Tables 2-4. The ED50 values in this latter paradigm
are included in parentheses below the ED50 values for
DMCM-induced seizures.
Significantly, the R-nipecotic acid derivatives 39 and
40, which are geometric isomers, have very similar
biological activity, which implies that thiophenes and
phenyl rings are isosterically interchangeable in this
oxime series. Pyridine/phenyl isosterism is however not
permitted, illustrated by the poor in vitro actvity of 65
(Ki of >2 µM) compared to 60 (Ki of 81 nM). One
â-homoproline derivative,65 66, is featured in Table 2;
this compound has an unremarkable in vitro effect, but
performs comparatively well against the seizure in vivo
measure.
The relatively facile synthesis of diaryloxime ether-
based GABA uptake inhibitors by O-alkylation of an
The vinyl ether series (Table 4) has provided the most

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3455
Ta ble 4. Pharmacological and Physical Data for Vinyl Ether GABA Uptake Inhibitorsa

3456 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Ta ble 4 (Continued)
Knutsen et al.
a
Crystallization solvents: a, acetone; b, 2-propanol; c, Et2O; d, CH2Cl2; e, H2O; f, PhCH3; g, EtOH; h, CH3OH; i, EtOAC; j, n-heptane.
Amino acids (X): NIP represents nipecotic acid (see 1); Guv, guvacine (see 2); Hom, â-homoproline; all are alkylated on nitrogen. Figures
in parentheses below ED50 values for DMCM-induced seizures relate to mouse rotarod ED50 values in mg/kg.47
potent in vitro inhibitors of GAT-1 so far discovered.
Again, the special activity of the vinyl ether containing
the bis(3-methyl-2-thienyl) heteroaromatic moieties is
borne out, with compound 13 exhibiting an Ki value of
14 nM for inhibition of [3H]-GABA uptake. Several other
structures in Table 4, particularly those containing
2-chlorophenyl derivatives also exhibit high in vitro
potency. Generally, we conclude that the vinyl ether
derivatives, of a particular aromatic/heteroaromatic
substitution have a greater in vitro potency than the
corresponding oxime structures. The rank order of
potency among the different amino acid derivatives is
as for the oxime series, the R-nipecotic acid derivatives
being generally more potent in vitro than the corre-
sponding guvacines, although exceptions exist.
Clearly, when one is moving from an in vitro assay
for inhibition of [3H]-GABA uptake to an in vivo
paradigm where compounds are investigated for seizure
control, many more variables come into play. Issues
such as penetration across the blood-brain barrier,
plasma half-life, distribution, and bioavailability become
important. Despite these potential variables, the com-

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3457
Sch em e 6
Ch a r t 4. Structures of NNC 05-1010 13 and NNC
05-0711 14
Ch a r t 6. Examples of Biologically Active Molecules
Containing the Oxime Functionality
Ch a r t 5. GABA Uptake Inhibitors with 5- and 6-Atom
Linkers
in Table 2 which have been examined in mouse rotarod,
those with improved ratios compared to the marketed
drug are 14, 74, and 75, with ratios of between 4 and 5.
In Table 4, vinyl ethers such as 76-79 also exhibited
improved therapeutic ratios compared to 1. However,
during our investigations, we identified a series of
4-methyl-2-thienyl/2-methylphenyloxime derivatives, 20-
23 (Table 3, Scheme 2), which became the subject of a
special study. We wished to evaluate these ligands in
detail in order to compare the pharmacological activity
of the E- and Z-isomers and both the R- and S-
enantiomers.
The oximes 20-23 are not outstandingly potent as
inhibitors of [3H]-GABA uptake in vitro, as the R-
nipecotic acid derivatives 20 (Z) and 22 (E) had Ki values
of 783 and 261 nM. However, in view of this in vitro
profile, 20 and 23 are remarkably potent in terms of
their in vivo anticonvulsant activity, with ED50 values
of 4.5 and 5.5 mg/kg, respectively. Rather like the other
E- and Z-isomers such as 39 and 40, there was little
observed difference in anticonvulsant potency between
20 and 22, but these two examples exhibited an im-
pounds featured in Tables 2-4 have a remarkably
consistent effect in protection against seizures. A range
of these novel anticonvulsants have ED50 values of <2
mg/kg in this paradigm. Three oxime derivatives, 64,
67, and 68 (Table 2) have ED50 values below 1 mg/kg,
as do six compounds, 13, 45, and 69-73, in the vinyl
ether series (Table 4).
Tiagabine has been extensively investigated in human
subjects after oral dosing as add-on therapy for the
control of partial seizures. The usual maintenance dose
in human is 30-50 mg/day, with doses of up to 70 mg/
day being well tolerated. The ED50 value for tiagabine
against DMCM-induced seizures in mice is 1.2 mg/kg
intraperitoneally (i.p.), so the total daily dose in humans
is of a similar order to the rodent anticonvulsant ED50
in this paradigm, which gives some prediction of poten-
tial human doses. To provide a superior drug candidate
to 1, it is important to improve upon the therapeutic
ratio of ED50(rotarod)/ED50(DMCM seizures), which is
3.75 for 1 (4.5/1.2) (Table 1). Of the oxime derivatives

3458 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Knutsen et al.
5-fold. These highly potent GABA uptake inhibitors may
have potential in the treatment of epilepsy in humans,
and further structural modifications in this series of
anticonvulsant agents are planned.
Future challenges in this field include the design of
new ligands which are specific for other subtypes of
GABA transporters.48-51 These transporters have a
different brain distribution to GAT-1, and inhibitors
with subtype selectivity, which are now emerging,66,67
may show a different therapeutic and side-effect profile.
Exp er im en ta l Section
Ch em istr y. Melting points were determined in open capil-
lary tubes on a Bu¨chi 535 melting point apparatus and are
uncorrected. The structures of all compounds are consistent
with spectroscopic data, and satisfactory elemental analyses
(for C, H, N with
a Perkin-Elmer model 240 elemental
analyzer; S and Cl were determined by the Scho¨niger combus-
tion method) were obtained within (0.4% of theoretical values
where given. 1H NMR spectra were recorded on a Bruker
WM400 spectrometer with TMS as standard, with selected
representative chemical shifts quoted in ppm (δ) in the solvents
indicated. Compounds used as starting materials are either
known compounds or compounds which can be prepared by
methods known per se. Column chromatography was carried
out using the technique described by W. C. Still et al.,68 on
Merck silica gel 60 (Art 9385) using thick-walled glass
columns. HPLC was carried out on a Waters model 510
chromatograph interfaced via a system module to a Waters
490 multiwavelength detector to a reversed phase C18 column
(250 × 4 mm, 5 mm, 100 Å; eluent flow rate 1 mL/min at 35
°C). Retention times are given in minutes.
1-[2-[[(Dip h en yl)im in o]oxy]et h yl]-3-(1,2,5,6-t et r a h y-
d r op yr id in -1-yl)ca r boxylic a cid (14) (Meth od A, Sch em e
1). Dip h en ylm et h a n on e, O-(2-ch lor oet h yl)oxim e (17).
Benzophenone oxime 15 (3.94 g, 20 mmol), 1-bromo-2-chloro-
ethane (28.7 g, 200 mmol), and dried, powdered K2CO3 (5.53
g, 80 mmol) were heated at reflux in acetone (60 mL) for 72 h.
The reaction mixture was cooled and filtered and the filtrate
was evaporated to an oily residue. Flash chromatography (95:5
n-heptane/EtOAc) provided diphenylmethanone, O-(2-chloro-
ethyl)oxime 17 (3.82 g, 73%) as an oil: Rf ) 0.36 (SiO2, 90:10
n-heptane/EtOAc); 1H NMR (CDCl3) 3.79 (t, 2H), 4.38 (t, 2H),
7.29-7.50 (m, 10H).
F igu r e 3. X-ray structure of 2-methylphenyl, 4-methyl-2-
thienylmethanone oxime, 30.
proved therapeutic ratio of 11 when compared to other
GABA uptake inhibitors. The compounds featured in
Table 3 are all oximes derived from 2-methylphenyl(4-
methyl-2-thienyl)-methanone, and because of this prom-
ising therapeutic ratio, all of the four combinations of
isomers, i.e., the E- and Z-oxime isomers as well as the
R- and S-nipecotic acid derivatives, were prepared and
tested. The precursor to compounds 20 and 21, 2-meth-
ylphenyl, 4-methyl-2-thienylmethanone oxime, 30, was
subjected to X-ray crystallographic analysis, confirming
that it was a Z-isomer (Figure 3).
It is difficult to assign the cis or trans geometry across
the oxime double bond in these four O-alkylated oxime
GABA uptake inhibitors by NMR even though all four
E/Z and R/S combinations 20-23 were isolated; this is
a simpler matter when monoaryl oxime ethers are
involved since the vicinal proton NMR coupling can be
readily analyzed.58 Hence we opted for an X-ray analysis
of 30, allowing assignment of E- and Z-isomer identity
to the pharmacologically promising series, 20-23.
Eth yl 1-[2-[[(Dip h en yl)im in o]oxy]eth yl]-3-(1,2,5,6-tet-
r a h yd r op yr id in -1-yl)ca r boxyla te (19). To a solution of 17
(13.09 g, 50 mmol) in PhCH3 (500 mL) were introduced
guvacine ethyl ester69 18 (7.76 g, 50 mmol) and powdered,
dried K2CO3 (13.82 g, 100 mmol). The reaction mixture was
heated at reflux for 90 h and cooled. Filtration and evaporation
provided an oil which was purified by flash chromatography
(9:1 cyclohexane/EtOAc initially, later increasing polarity to
1:1) to provide ethyl 1-[2-[[(diphenyl)imino]oxy]ethyl]-3-(1,2,5,6-
tetrahydropyridin-1-yl) carboxylate 19 (5.63 g, 30%) as a
gum: Rf ) 0.26 (SiO2, 1:1 cyclohexane/EtOAc); 1H NMR
(CDCl3) 2.26-3.34 (m, 2H), 2.57 (t, 2H), 2.87 (t, 2H), 3.26 (q,
2H), 3.72 (s, 3H), 4.38 (t, 2H), 6.99 (m, 1H), 7.29-7.52 (m,
10H). Starting material 17 (8.87 g, 67%) was also isolated. The
ethyl ester 19 was converted into a hydrochloride salt by
dissolving the gum in PhCH3 at 75 °C and precpitating by
treatment with chlorotrimethylsilane and CH3OH (1 equiv),
a method we have described previously.10 The white solid
hydrochloride had a melting point of 110-116 °C. Anal.
(C23H26N2O3‚HCl) C, H, N, Cl.
Con clu sion
We have demonstrated that improved anticonvulsant
efficacy in mice can be achieved by rational design of
novel uptake inhibitors at the GABA transporter type
1 (GAT-1) site with the goal of providing new clinically
effective drugs for treatment of partial seizures.
Given the background of 1 as a benchmark GABA
uptake inhibitor and a clinically effective anticonvulsant
drug, we have prepared novel compounds of increased
potency, using a new putative model of ligand interac-
tion at the GABA uptake site GAT-1 (Figure 1). This
model involves the postulated interaction of an elec-
tronegative region in the small molecule GABA uptake
inhibitor with a positive domain in the protein structure
of the site controlling GABA uptake. The new agents
synthesized on the basis of this new model have potent
anticonvulsant properties in rodent seizure models, and
some have an improved therapeutic ratio.
1-[2-[[(Dip h en yl)im in o]oxy]et h yl]-3-(1,2,5,6-t et r a h y-
d r op yr id in -1-yl)ca r boxylic Acid (14). To a solution of 19
hydrochloride salt (16.3 g, 43 mmol) in EtOH (100 mL) was
added 10 N aqueous NaOH solution (43 mL). The mixture was
stirred for 2.5 h at room temperature and evaporated in vacuo.
The creamy residue was dissolved in water (250 mL) and
cooled to ca. 5 °C. The pH was adjusted to ca. 2 with 2 N
aqueous HCl, and the mixture was extracted with CH2Cl2 (4
By a series of design steps, 12 (Table 2) was derived
from 1, and in turn led to the discovery of 13 (Table 2),
which is one of the most potent known inhibitors of [3H]-
GABA uptake in vitro. An insertion of an ether oxygen
(Chart 3) in 1 thereby improves in vitro potency by

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3459
39 as a white solid hydrochloride salt (3.25 g, 71%): mp 204-
× 200 mL). The combined organic extracts were dried (MgSO4).
The residue after evaporation was crystallized first from
CH2Cl2/cyclohexane, followed by recrystallization from 2-pro-
panol to provide 14 as a white solid (9.4 g, 56%): mp 214.5-
216.5 °C; 1H NMR (CDCl3) 2.46-2.56 (m, 2H), 2.57 (t, 2H),
2.87 (t, 2H), 3.26 (q, 2H), 3.72 (s, 3H), 4.38 (t, 2H), 6.99 (m,
1H), 7.29-7.52 (m, 10H). Anal. (C23H26N2O3‚HCl) C, H, N, Cl.
205 °C; [R]20 -6.8° (c ) 1.0, CH3OH); 1H NMR [(CD3)2SO]
D
1.76 (s, 3H), 2.09 (s, 3H), 2.96 (t, 2H), 4.39 (t, 2H), 6.92 (d,
1H), 7.20-7.38 (m, 4H), 7.68 (d, 1H). Anal. (C23H26N2O3‚HCl)
C, H, N.
(3R)-1-[2-[[2,2-Bis(2-m eth ylp h en yl)eth en yl]oxy]eth yl]-
3-p ip er id in eca r boxylic Acid (45) (Meth od C, Sch em e 4).
Bis(2-m et h ylp h en yl)a cet a ld eh yd e (43). Hexamethyldi-
silazane (1.55 g, 96 mmol, 1.99 mL) in dry THF (150 mL) was
cooled to -70 °C. n-BuLi (2.5 M in hexanes) (40 mL, 100 mmol)
was added dropwise while keeping the temperature below -65
°C, and the solution was stirred for 1 h at -70 °C. Ethyl
bromoacetate (16.0 g, 96 mmol, 10.65 mL) was introduced
dropwise over a 10 min period, followed by bis(2-methyl-
phenyl)methanone10 (16.8 g, 80 mmol) in dry THF (200 mL),
again keeping the temperature below -65 °C. After being
stirred for 20 min at this temperature, the reaction mixture
was allowed to reach 0 °C, and it was poured into a vigorously
stirred mixture of H2O (500 mL) and Et2O (200 mL). The
phases were separated, the aqueous phase was extracted with
Et2O (200 mL), and the combined ether extracts were dried
(MgSO4). The residue on evaporation, which contained ethyl
3,3-bis(2-methylphenyl)-2-oxiranecarboxylate 42, was dis-
solved in EtOH (100 mL) and saponified by stirring with 10
N aqueous NaOH (40 mL) at room temperature for 0.5 h. The
reaction mixture was acidified to pH 2 with aqueous HCl and
was extracted with Et2O (400 mL). The Et2O phase was
extracted twice with a mixture of 10 N aqueous NaOH (8 mL)
and H2O (300 mL). The combined aqueous extracts were
acidified with 0.5 M aqueous citric acid, and the mixture was
extracted with Et2O (2 × 200 mL). The combined Et2O extracts
were dried (MgSO4) and evaporated to afford the residual solid
3,3-bis(2-methylphenyl)-2-oxiranecarboxylic acid 41 which was
dissolved in 1,4-dioxan (50 mL). Decarboxylation was per-
formed by heating this solution at 130 °C for 2 h, and 43 was
obtained as an oil (10.1 g, 56%) following evaporation: 1H NMR
(CDCl3) 2.25 (s, 6H), 5.19 (s, 1H), 6.95 (d, 2H), 7.15-7.28 (m,
6H), 10.0 (1H, s).
(3R)1-[2-[[2-(2-Met h ylp h en yl)-2-(3-m et h yl-2-t h ien yl)-
im in o]oxy]eth yl]-3-p ip er id in eca r boxylic a cid (39) (Meth -
od B, Sch em e 3). (3R)-N-(2-Br om oeth yl)n ip ecotic Acid
Eth yl Ester (32). To ethyl R-3-piperidine carboxylate (ethyl
nipecotate)55 (100 g, 640 mmol) in anhydrous acetone (400 mL)
were added 2-bromoethanol (95.4 g, 760 mmol), dried, pow-
dered K2CO3 (175.8 g, 1270 mmol), and KI (21 g, 127 mmol).
The reaction mixture was stirred at room temperature for 18
h. The reaction mixture was filtered, and the filtrate was
concentrated in vacuo to a residue which was fractionated to
provide R-N-(2-hydroxyethyl)nipecotic acid ethyl ester 33 (95
g, 74%): bp 110-115 °C/0.1 mmHg; 1H NMR (CDCl3) 1.28 (t,
3H); 1.6 (m, 2H); 1.75 (m, 1H); 1.9 (m, 1H); 2.2 (m, 1H); 2.4
(m, 1H); 2.55 (m, 3H); 2.75 (m, 1H); 2.9 (m, 1H); 3.10 (br s,
1H); 3.63 (m, 2H); 4.16 (q, 2H). This alcohol 33 (40.3 g, 200
mmol) was dissolved in PhCH3 (300 mL). A solution of SOBr2
(41.6 g, 200 mol) in PhCH3 (50 mL) was added dropwise, and
the reaction mixture was stirred at room temperature for 2 h.
Et2O (300 mL) was added, and the solid product (65 g) was
collected by filtration. The bromide 32 (46.6 g, 68%) was
thereby obtained as a white solid hydrobromide salt, mp
187.5-194.5 °C, by recrystallization from 2-propanol: 1H NMR
(CDCl3) 1.27 (t, 3H); 1.58 (q, 1H); 2.02 (d, 1H); 2.30-2.50 (m,
2H); 2.84-3.04 (m, 2H); 3.42-3.58 (m, 3H); 3.70 (d, 1H); 3.80
(d, 1H); 3.95 (m, 2H); 4.17 (q, 2H); 11.6 (br s, 1H).
2-Meth ylp h en yl(3-m eth yl-2-th ien yl)m eth a n on e Oxim e
(34). 2-Methylphenyl(3-methyl-2-thienyl)methanone 35 was
prepared from 2-bromo-3-methylthiophene56 and 2-methyl-
benzaldehyde by the standard method described previously for
bis(3-methyl-2-thienyl)methanone.10 Compound 35 (2.28 g,
10.5 mmol) was dissolved in pyridine (70 mL), NH2OH‚HCl
(1.50 g, 21.5 mmol) was introduced, and the reaction mixture
was heated at reflux for 66 h before being concentrated in
vacuo to a residue. EtOAc (100 mL) and water (100 mL) were
introduced, and the EtOAc phase was separated before being
extracted with water (100 mL). The organic phase was dried
(MgSO4) and evaporated to an oil which was purified by flash
chromatography (19:1 n-heptane/EtOAc) to provide a mixture
of geometric isomers which was recrystallized from cyclohex-
ane to provide 34 (0.7 g, 29%) as a single geometric isomer:
mp 99-100 °C; 1H NMR [(CD3)2SO] 1.73 (s, 3H), 2.07 (s, 3H),
6.88 (d, 1H), 7.17-7.33 (m, 4H), 7.61 (d, 1H), 11.81 (s, 1H).
By crystallization of the mother liquors, the other geometric
2,2-Bis(2-m et h ylp h en yl)et h en yl 2-Br om oet h yl E t h er
(44). 1,2-Dibromoethane (16.9 g, 90 mmol, 7.75 mL), tetra-
butylammonium bromide (1.45 g, 4.5 mmol) and 12 N aqueous
NaOH solution (25 mL) were added to a solution of 43 (10.0 g,
45 mmol) in CH2Cl2 (50 mL). The two phases were stirred
vigorously for 20 h, and the mixture was acidified with 4 N
aqueous HCl (75 mL). H2O (200 mL) was added, and the
organic phase was separated before being washed with H2O
(200 mL) and dried (MgSO4). Evaporation (coevaporation with
PhCH3 and PhCH2CH3) provided 44 as an oil (12.8 g, 86%):
1H NMR (CDCl3) 2.20 (s, 6H), 3.37 (2H, t), 4.08 (2H, t) 6.28 (s,
1H), 6.90-7.28 (m, 8H).
(3R)-1-[2-[[2,2-Bis(2-m eth ylp h en yl)eth en yl]oxy]eth yl]-
3-p ip er id in eca r boxylic Acid (45). Ethyl R-3-piperidine
carboxylate hydrochloride55 (2.25 g, 11.6 mmol) and powdered,
dried K2CO3 (6.4 g, 46.3 mmol) were added to a solution of 44
(12.8 g, 38.6 mmol) in acetone (100 mL). The reaction mixture
was heated at reflux for 18 h and cooled. Filtration and
evaporation provided an oil which was purified by flash
chromatography on silica gel (9:1 cyclohexane/EtOAc initially,
increasing polarity to 1:1) to provide ethyl (3R)-1-[2-[[2,2-bis-
(2-methylphenyl)ethenyl]oxy]ethyl]-3-piperidine carboxylate
(2.23 g, 47%) as a gum. This ethyl ester was hydrolyzed by
dissolution in EtOH (20 mL) and treatment with 10 N aqueous
NaOH solution (7 mL) at ambient temperature for 3 h. Water
(300 mL) was added, and the mixture was washed with Et2O
(3 × 50 mL) to remove unsaponified material. The aqueous
solution was adjusted to pH 5 with 2 N aqueous HCl and then
extracted with CH2Cl2 (5 × 50 mL), dried (MgSO4), and
evaporated to a gum. The residual 45 (1.9 g) was converted
into a crystalline hydrochloride salt by the method described
for compound 19. The salt was collected by filtration and
recrystallized from H2O to provide analytically pure 45 as the
hydrochloride salt (1.4 g, 62%): mp 217-222 °C; [R]20D -23.5°
1
isomer of 34 was obtained (0.5 g, 21%): mp 136-138 °C; H
NMR [(CD3)2SO] 1.87 (s, 3H), 2.15 (s, 3H), 6.87 (d, 1H), 7.17-
7.32 (m, 4H), 7.39 (d, 1H), 11.19 (s, 1H).
Eth yl (3R)-1-[2-[[2-(2-Meth ylp h en yl)-2-(3-m eth yl-2-th i-
en yl)im in o]oxy]eth yl]-3-p ip er id in eca r boxyla te (38). To
a solution of the lower-melting oxime 34 (9.75 g, 42 mmol) in
acetone (100 mL) was added 32 (22.5 g, 65.2 mmol), PhCH3
(200 mL), and dried, powdered K2CO3 (23.22 g, 168 mmol).
The reaction mixture was heated at reflux for 2 h, cooled, and
filtered. The filtrate was evaporated to an oil, which was
purified by flash chromatography (9:1 n-heptane/EtOAc). The
ester 38 (12.1 g, 69%) was converted into the corresponding
hydrochloride salt, using the method ((CH3)3SiCl/CH3OH)
described for the ester 19. The resultant solid was recrystal-
lized from PhCH3/cyclohexane to afford white crystals: mp
124-125.5 °C; 1H NMR (CDCl3) 1.22 (t, 3H), 1.42 (q, 1H), 1.72
(s, 3H), 2.16 (s, 3H), 3.46 (t, 2H), 4.12 (q, 2H), 4.79 (t, 2H),
6.81 (d, 1H), 7.20-7.38 (m, 4H), 7.41 (d, 1H). Anal. (C23H30N2-
O3S‚HCl) C, H, N, Cl, S.
(3R)-1-[2-[[2-(2-Meth ylp h en yl)-2-(3-m eth yl-2-th ien yl)-
im in o]oxy]eth yl]-3-p ip er id in eca r boxylic Acid (39). The
hydrochloride salt of ester 38 (4.87 g, 10.8 mmol) was
hydrolyzed as described for compound 19. The residue after
evaporation was recrystallized from H2O/CH3OH to provide
1
(c ) 0.02, CH3OH); H NMR [(CD3)2SO] 2.06 (s, 3H), 2.12 (s,

3460 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Knutsen et al.
3H), 4.34 (br t, 2H), 6.54 (s, 1H), 7.02-7.20 (m, 8H). Anal.
(C22H29NO3‚HCl) C, H, N, Cl.
1-(3-methyl-2-thienyl)ethenyl]-3-methylthiophene 50 (0.54 g,
50%) as a gum. Compound 50 (0.53 g, 19 mmol) was dissolved
in dry toluene (20 mL), and the solution was cooled to 0 °C. A
solution of n-BuLi (2.5 M in hexanes) (0.9 mL, 23 mmol) was
introduced, the reaction mixture was allowed to stand at 0 °C
for 1 h, and a solution of p-toluenesulfonyl chloride (0.47 g, 25
mmol) in PhCH3 (10 mL) was added. The mixture was stirred
at room temperature for 20 h, and to the resultant solution of
tosylate 51 were added ethyl R-3-piperidine carboxylate
hydrochloride (0.59 g, 3.8 mmol) and powdered, dried K2CO3
(1.04 g, 7.5 mmol). The temperature was increased to 80 °C,
and stirring was maintained for 50 h. The reaction mixture
was cooled, and water (50 mL) was added. The PhCH3 phase
was separated, and the water phase was extracted with EtOAc
(50 mL). The combined organic extracts were dried (MgSO4)
and evaporated to give an oil, which was purified by flash
chromatography (19:1 cyclohexane/EtOAc initially, later in-
creasing polarity to 5:1) to provide 52 (0.25 g, 31%) as a gum:
1H NMR (CDCl3) 1.20 (t, 3H), 1.93 (s, 3H), 1.96 (s, 3H), 2.06
(t, 1H), 2.24 (t, 1H), 2.66 (t, 2H), 4.01 (t, 2H), 4.09 (q, 2H),
6.50 (s, 1H), 6.75 (d, 1H), 6.77 (d, 1H), 7.02 (d, 1H), 7.14 (d,
1H).
(3R)-1-[2-[(2,2-Diph en yleth en yl)oxy]eth yl]-3-piper idin e-
ca r boxylic Acid (47) (Meth od D, Sch em e 5). Eth yl (3R)-
1-[2-[(2,2-Dip h en ylet h en yl)oxy]et h yl]-3-p ip er id in eca r -
boxyla te (46). Diphenylacetaldehyde (4.9 g, 25 mmol) was
added dropwise to a mixture of NaH (1.5 g, 5 mmol, 80% oil
dispersion) and dry PhCH3 (25 mL) at 0 °C. This mixture was
stirred at room temperature for 0.5 h, heated to 50 °C, and
allowed to cool again to room temperature. R-1-(2-Bromoethyl)-
3-ethoxycarbonylpiperidine hydrobromide 32 (8.6 g, 25 mmol)
was introduced portionwise while the temperature was kept
below 30 °C with an ice-water bath. After being stirred for 1
h, the reaction mixture was filtered, and the filtrate was
evaporated to a residue. Flash chromatography (4:1 heptane/
THF) provided 46 (6.6 g, 69%) as a gum: 1H NMR (CDCl3)
1.22 (t, 3H), 1.43 (q, 1H), 1.55 (dq, 1H), 2.11 (dt, 1H), 2.30 (t,
3H), 2.69 (t, 2H), 2.99 (br d, 1H), 4.03 (t, 2H), 4.10 (q, 2H),
6.51 (s, 1H), 7.18-7.32 (m, 8H), 7.41 (d, 2H).
(3R)-1-[2-[[2,2-Diph en yleth en yl)oxy]eth yl]-3-piper idin e-
ca r boxylic Acid (47). Ethyl (3R)-1-[2-[(2,2-Diphenylethenyl)-
oxy]ethyl]-3-piperidine carboxylate (3.0 g, 79 mmol) was
hydrolyzed as described for compound 14. The crystalline
residue after evaporation was triturated with Et2O to provide
47 as a white solid (2.65 g, 86%): mp 217-218 °C; [R]20D -7.6°
(c ) 0.02, CH3OH); 1H NMR [(CD3)2SO] 2.82-3.05 (br m, 2H),
(3R)-1-[2-[[2,2-Bis(3-m eth yl-2-th ien yl)eth en yl]oxy]eth yl]-
3-p ip er id in eca r boxylic Acid (13). Compound 52 (420 mg,
1 mmol) was dissolved in ethanol (20 mL) and hydrolyzed as
described for compound 19. The residue after evaporation was
recrystallized from H2O to provide 13 as a white solid
hydrochloride salt: mp 55-70 °C (dec); 1H NMR (CDCl3) 1.26
(q, 1H), 1.40 (q, 1H), 1.91 (s, 3H), 1.93 (s, 3H), 1.97 (t, 1H),
2.09 (t, 1H), 2.33 (br t, 1H), 2.24 (t, 1H), 2.54 (t, 2H), 2.70 (br
d, 1H), 2.90 (br d, 1H), 3.41 (t, 1H), 4.02 (t, 2H), 4.09 (q, 2H),
6.67 (s, 1H), 6.79 (d, 1H), 6.81 (d, 1H), 7.23 (d, 1H), 7.32 (d,
1H). Anal. (C20H25NO3S2‚0.75H2O) C, H, N, S, Cl.
Electr osta tic P oten tia l Ca lcu la tion s. Geometry optimi-
zation was performed by using the semiempirical method AM1
(Figure 2). Electrostatic potentials were based upon the
Hartree-Fock 6-31G** ab initio procedure52 implemented in
the program Spartan.
P h a r m a cologica l Meth od s. Syn a p tosom a l [3H]-GABA
Up ta k e. Uptake of [3H]-GABA into synaptosomal preparations
was assayed by a filtration assay.39 Rat forebrain was rapidly
excised and homogenized in 20 mL of ice-cold 0.32 M sucrose
with a hand-driven Teflon/glass Potter-Elvehjem homogenizer.
The homogenate was centrifuged for 10 min at 600g at 4 °C.
The pellet was resuspended in 50 volumes of ice-cold buffer
(120 mM NaCl, 0.18 mM KCl, 2.30 mM CaCl2, 4.0 mM MgSO4,
12.66 mM Na2HPO4, 2.97 mM NaH2PO4, and 10.0 mM glucose,
pH 7.4) at 4 °C. Fifty microliters of this synaptosomal
suspension (0.1 mg protein), diluted into 300 mL of phosphate
buffer and 100 mL of test substance solutions in water, was
preincubated for 8 min at 30 °C. Then 50 mL of [3H]-GABA
(final concentration 0.9 nM) and unlabeled GABA (final
concentration 0.9 nM) was added before continuing incubation
for another 8 min. Synaptosomes were then recovered by rapid
filtration through Whatman GF/F glass fiber filters under
vacuum. Filters were washed twice, each time with 10 mL of
ice-cold isotonic saline, and the tritium trapped on the filters
was assessed by conventional scintillation counting in 4 mL
of FilterCount (Packard). Noncarrier-mediated uptake was
determined in the presence of nipecotic acid (500 mM) and was
subtracted from total binding to give carrier-mediated [3H]-
GABA uptake. IC50 values were calculated from dose-response
curves (3 points minimum), and inhibitory constants (Ki) were
calculated using the equation: Ki ) IC50/(1 + [L]/Km), where
[L] is the concentration of [3H]-GABA (25 nM) and Km the
affinity constant for GABA (3700 nM).71 The Ki value (with
standard error) obtained for each compound is shown in Tables
1-4.
4.42 (dt, 2H), 6.86 (s, 1H), 7.18-7.38 (m, 10H). Anal. (C22H25
ClNO3‚HCl) C, H, N, Cl.
-
(3R)-1-[2-[[2,2-Bis(3-m eth yl-2-th ien yl)eth en yl]oxy]eth yl]-
3-p ip er id in eca r boxylic Acid (13) (Meth od E, Sch em e 6).
Eth yl 2-(Tr ityloxy)eth oxya ceta te (48). 2-(Trityloxy)ethanol
(3.98 g, 13 mmol) was dissolved in dry THF (50 mL), and a
2.5 M solution of n-BuLi in hexanes (5.5 mL, 13.7 mmol) was
added at 5 °C. In a separate vessel, a solution of BrCH2CO2H
(1.81 g, 13 mmol) was treated with a 2.5 M solution of n-BuLi
in hexanes (5.5 mL, 13.7 mmol) at 5 °C before the two solutions
were mixed. This reaction mixture was heated at reflux for
68 h, cooled, and water (200 mL) was added. The reaction
mixture was washed with EtOAc (100 mL), and the aqueous
phase was acidified with 0.5 M citric acid solution (50 mL).
Extraction with EtOAc (2 × 100 mL) and drying of the
combined extracts (MgSO4) provided crude 2-(trityloxy)ethoxy-
acetic acid 49 (2.78 g, 58%). This acid was dissolved in CH2Cl2
(30 mL), and dicyclohexyl carbodiimide (1.72 g, 8.3 mmol) was
introduced, followed by 4-pyrrolidinopyridine70 (0.11 g, 0.74
mmol) and ethanol (0.89 mL, 2 equiv). The reaction mixture
was stirred for 16 h at room temperature and filtered to
remove dicyclohexyl urea. The filtrate was evaporated, and
the residue was purified by flash chromatography (100:1
cyclohexane/EtOAc initially, then 30:1) to provide 48 (1.5 g,
50%) as an oil: 1H NMR (CDCl3) 1.24 (t, 3H), 3.24 (t, 2H),
3.66 (t, 2H), 4.06 (s, 2H), 4.08 (q, 2H), 7.12-7.58 (m, 15H).
Eth yl (3R)-1-[2-[[2,2-Bis(3-m eth yl-2-th ien yl)eth en yl]-
oxy]eth yl]-3-p ip er id in eca r boxyla te (52). 2-Bromo-3-meth-
ylthiophene56 (1.5 g, 8.5 mmol) and Mg turnings (0.22 g) were
heated gently in anhydrous THF (30 mL), and the reaction
rapidly became exothermic. After the resultant Grignard
reagent 36 was stirred for 0.2 h under a nitrogen atmosphere,
the reaction mixture was heated at reflux for 0.5 h, and ester
48 (1.5 g, 3.8 mmol) was introduced as a solution in anhydrous
THF (20 mL). The mixture was again heated at reflux for 0.5
h and then cooled, and saturated aqueous NH4Cl (100 mL)
was carefully added. Stirring for 0.5 h at room temperature
was followed by extraction with EtOAc (3 × 70 mL). The
combined extracts were dried (MgSO4) and evaporated. The
resultant residue was dissolved in a mixture of 2 N aqueous
HCl (50 mL), THF (50 mL), and EtOH (50 mL), and the
solution was heated at 50 °C for 1 h (to ensure deprotection
and alkene formation), before basification to pH 9.5 with
aqueous NaOH. The organic solvents were evaporated, and
the aqueous residue was extracted with EtOAc (3 × 75 mL).
Drying of the combined organic extracts (MgSO4) and evapora-
tion provided an oil, which was purified by flash chromatog-
raphy (9:1 cyclohexane/EtOAc) to afford 2-[2-(2-hydroxyethoxy)-
DMCM-In d u ced Seizu r es. These tests were conducted as
described previously.46,47 Briefly, male or female NMRI mice
(Bomholdtgaard, Ry, Denmark) weighing 20-25 g were dosed
with 15 mg/kg DMCM i.p. 30 min after i.p. administration of
test compound. Eight mice were tested per dose. The number

Synthesis of Novel GABA Uptake Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18 3461
(13) Bjorge, S.; Black, A.; Bockbrader, H.; Chang, T.; Gregor, V. E.;
Lobbestael, S. J .; Nugiel, D.; Pavia, M. R.; Radulovic, L.; Woolf,
T. Synthesis and Metabolic Profile of CI966: A Potent, Orally-
Active Inhibitor of GABA Uptake. Drug Dev. Res. 1990, 21,
189193.
(14) Pavia, M. R.; Lobbestael, S. J .; Nugiel, D.; Mayhugh, D. R.;
Gregor, V. E.; Taylor, C. P.; Schwarz, R. D.; Brahce, L.;
Vartanian, M. G. Structure-activity studies on benzhydrol-
containing nipecotic acid and guvacine derivatives as potent,
orally active inhibitors of GABA uptake. J . Med. Chem. 1992,
35, 4238-4248.
(15) Taylor, C. P.; Vartanian, M. G.; Schwarz, R. D.; Rock, D. M.;
Callahan, M. J .; Davis, M. D. Pharmacology of CI966: A Potent
GABA Uptake Inhibitor, In Vitro and in Experimental Animals.
Drug Dev. Res. 1990, 21, 195-215.
(16) Sedman, A. J .; Gilmet, G. P.; Sayed, A. J .; Posvar, E. L. Initial
Human Safety and Tolerance Study of a GABA Uptake Inhibitor,
CI966: Potential Role of GABA as a Mediator in the Pathogen-
esis of Schizophrenia and Mania. Drug Dev. Res. 1990, 21, 235-
242.
of mice failing to exhibit clonic seizures within 15 min were
considered to be protected.
Motor Im p a ir m en t. These tests were conducted as de-
scribed previously.47 Briefly, NMRI mice were trained for 2
min to stay on top of a rotating rod (6 rev/min) as the rod
rotated toward the animal. After a 2 min training period mice
(n ) 8/dose) were administered the test compound, as de-
scribed in the DMCM seizure tests above, and were then
placed on the rotating rod. Each mouse was observed for 2
min, and a mouse that fell off the rod twice or more was
considered ataxic.
Dr u gs. Compounds were suspended in duphasol-x or dis-
solved in distilled water and were administered in a volume
of 10 mL/kg. DMCM (a generous gift from Schering AG, Berlin,
Germany, courtesy of Dr. Sprzagala) was dissolved in 0.02 N
saline and was delivered in a volume of 15 mL/kg.
Da ta An a lysis. ED50 values were the doses (mg/kg) that
protected 50% of the mice against seizures or the dose
producing falls from the rotating rod in 50% of the mice,
respectively.
(17) Suzdak, P. Lipophilic GABA Uptake Inhibitors: Biochemistry,
Pharmacology and Therapeutic Potential. Drugs Fut. 1993, 18,
1129-1136.
(18) Braestrup, C.; Nielsen, E. B.; Wolffbrandt, K. H.; Andersen, K.
E.; Knutsen, L. J . S.; Sonnewald, U. Modulation of GABA
Receptor Interaction with GABA Uptake Inhibitors. Int. Congr.
Ser. Excerpta Med. 1987, 750 (Pharmacology), 125-128.
(19) Nielsen, E. B.; Suzdak, P. D.; Andersen, K. E.; Knutsen, L. J .
S.; Sonnewald, U.; Braestrup, C. Characterization of Tiagabine
(NO-328), a New Potent and Selective GABA Uptake Inhibitor.
Eur. J . Pharmacol. 1991, 196, 257266.
(20) Suzdak, P. D.; Swedberg, M. D. B.; Andersen, K. E.; Knutsen,
L. J . S.; Braestrup, C. In vivo Labeling of the Central GABA
Uptake Carrier with 3H-Tiagabine. Life Sci. 1992, 51, 1857-
1868.
(21) Smith, S. E.; Parvez, N. S.; Chapman, A. G.; Meldrum, B. S.
The γ-Aminobutyric Acid Uptake Inhibitor, Tiagabine, is Anti-
convulsant in Two Animal Models of Reflex Epilepsy. Eur. J .
Pharmacol. 1995, 273, 259-265.
(22) Halonen, T.; Nissinen, J .; J ansen, J . A.; Pitkanen, A. Tiagabine
prevents seizures, neuronal damage and memory impairment
in experimental status epilepticus. Eur. J . Pharmacol. 1996, 299,
69-81.
Ack n ow led gm en t. We acknowledge the contribu-
tion of Claus Bruun J ensen, J an Sørensen, Freddy
Pedersen, Paw Bloch, Charlotte Gade, and Karina
Madsen for synthesis of novel compounds, Dorthe
Andersen for determination of Ki values, and Hanne
Nielsen and Wei Liu for in vivo pharmacology as-
sistance. The encouragement and helpful comments of
Prof. Claus Braestrup are acknowledged.
Su p p or tin g In for m a tion Ava ila ble: X-ray structural
data for 2-methylphenyl, 4-methyl-2-thienylmethanone oxime,
30. This material is available free of charge via the Internet
at http://pubs.acs.org.
(23) Dalby, N. O.; Nielsen, E. B. Tiagabine exerts an epileptogenic
effect in amygdala kindling epileptogenesis in the rat. Neurosci.
Lett. 1997, 229, 135-137. Morimoto, K.; Sato, H.; Yamamoto,
Y.; Watanabe, T.; Suwaki, H. Antiepileptic effects of tiagabine,
a selective GABA uptake inhibitor, in the rat kindling model of
temporal lobe epilepsy. Epilepsia 1997, 38, 966-974.
Refer en ces
(1) This work was initially presented in part as posters at the 33rd.
National Organic Chemistry Symposium in Bozeman, Montana,
J une 1993, and at the Gordon Conference, Medicinal Chemistry,
New London, New Hampshire, August 1997. For part 2, see:
Andersen, K. E.; Begtrup, M.; Chorgade, M. S.; Lee, E. C.; Lau,
J .; Lundt, B. F.; Petersen, H.; Sørensen, P. O.; Thøgersen, H.
The Synthesis of Novel GABA uptake inhibitors. Part 2. Syn-
thesis of 5-hydroxytiagabine, a Human Metabolite of the GABA
Reuptake Inhibitor Tiagabine. Tetrahedron 1994, 50, 8699-
8710.
(2) Costa, E. Building a Bridge Between Neurobiology and Mental
Illness. J . Psychiat. Res. 1992, 26, 449-460.
(3) DeLorenzo, R. Mechanisms of Action of Anticonvulsant Drugs.
Epilepsia 1988, 29, S35-S47.
(4) Natsch, S.; Hekster, Y. A.; Keyser, A.; Deckers, C. L. P.;
Meinardi, H.; Renier, W. O. Newer anticonvulsant drugs: role
of pharmacology, drug interactions and adverse reactions in drug
choice. Drug Saf. 1997, 17, 228-240.
(5) Walker, M. C.; Patsalos, P. N. Update on novel antiepileptic
drugs. Emerging Drugs 1997, 2, 381-393.
(6) Gram, L. Tiagabine: A Novel Drug with a GABAergic Mecha-
nism of Action. Epilepsia 1994, 35, S85-S87.
(7) Rogawski, M. A.; Porter, R. J . Antiepileptic Drugs: Pharmaco-
logical Mechanisms and Clinical Efficacy with Consideration of
Promising Development Stage Compounds. Pharmacol. Rev.
1990, 42, 223-286.
(8) Upton, N. Mechanisms of Action of New Antiepileptic Drugs:
Rational Design and Serendipitous Findings. TIPS 1994, 15,
456-462
(9) Chadwick, D. W. An overview of the efficacy and tolerability of
new antiepileptic drugs. Epilepsia 1997, 38, S59-S62.
(10) Andersen, K. E.; Braestrup, C.; Grønwald, F. C.; J ørgensen, A.
S.; Nielsen, E. B.; Sonnewald, U.; Sørensen, P. O.; Suzdak, P.
D.; Knutsen, L. J . S. The Synthesis of Novel GABA uptake
inhibitors, 1. Elucidation of the Structure-Activity Studies
leading to the choice of R1[4,4-bis(3-Methyl-2-thienyl)-3-piperi-
dine carboxylic acid (Tiagabine) as an Anticonvulsant Drug
Candidate. J . Med. Chem. 1993, 36, 1716 1725.
(11) Mengel, H. Tiagabine. Epilepsia 1994, 29, S81-S84.
(12) Taylor, C. P. GABA Receptors and GABAergic Synapses as
Targets for Drug Development. Drug Dev. Res. 1990, 21, 151-
160.
(24) Suzdak, P. D.; J ansen, J . A.
A review of the preclinical
pharmacology of tiagabine - a potent and selective anticonvul-
sant GABA uptake inhibitor. Epilepsia 1995, 36, 612-626.
(25) Schousboe, A.; Larsson, O. M.; Krogsgaard-Larsen, P. GABA
Uptake Inhibitors as Anticonvulsants. In GABA Mechanisms in
Epilepsy; Tunnicliff, G., Raess, B. U., Eds.; Wiley-Liss: New
York, 1991; p 165-187. Update on the mechanism of action of
antiepileptic drugs.
(26) Bauer, J .; Stawowy, B.; Lenders, T.; Bettig, U.; Elger, C. E.
Efficacy and tolerability of tiagabine - results of an add-on study
in patients with refractory partial seizures. J . Epilepsy 1995, 8,
83-86.
(27) Meldrum, B. S. Update on the mechanism of action of anti-
epileptic drugs. Epilepsia 1996, 37, S4-S11.
(28) Benmenachem, E. International experience with tiagabine add-
on therapy. Epilepsia 1995, 36, S14-S21.
(29) Tiagabine, Gabitril, Tiabex. Drugs Fut. 1997, 22, 1396-1397.
(30) Sachdeo, R. C.; Biton, V.; Boellner, S. W.; Schachter, S. C.; Alto,
G. H.; Kaply, C. W.; Phillips, H. B.; Kardatzke, D.; Pixton, G.;
Sommerville, K. W. Long-term safety of tiagabine‚HCl. Epilepsia
1995, 36, 283-283.
(31) Uthman, B.; Rowan, A. J .; Ahmann, P.; Wannamaker, B.;
Schachter, S.; Rask, C. Safety and efficacy of 3 dose levels of
tiagabine‚HCl versus placebo as adjunctive treatment for com-
plex partial seizures. Ann. Neurol. 1993, 34, 272.
(32) Dodrill, C. B.; Arnett, J . L.; Sommerville, K. W.; Shu, V.
Cognitive and quality-of-life effects of differing dosages of
tiagabine in epilepsy. Neurology 1997, 48, 1025-1031.
(33) Mengel, H. B.; Pierce, M. W.; Mant, T.; Christensen, M. S.;
Gustafson, L. Tiagabine Safety and Tolerance During 2-weeks
Multiple Dosing to Healthy Volunteers. Epilepsia 1991, 32
(Suppl. 1).
(34) Chadwick, D.; Richens, A.; Duncan, A.; Dam, M.; Gram, L.;
Morrow, J .; Mengel, H.; Shu, V.; Mc. Kelvy, J . F.; Pierce, M. W.
Tiagabine Hydrochloride. Safety and Efficacy as Adjunctive
Treatment for Complex Partial Seizures. Epilepsia 1991, 32
(Suppl. 3).

3462 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 18
Knutsen et al.
(35) Ali, F. E.; Bondinell, W. E.; Danbridge, P. A.; Frazee, J . S.;
Garvey, E.; Girard, G. R.; Kaiser, C.; Ku, T. W.; Lafferty, J . J .;
Moonsammy, G. I.; Oh, H. J .; Rush, J . A.; Setler, P. E.; Stringer,
O. D.; Venslavsky, J . W.; Volpe, B. W.; Yunger, L. M.; Zirkle, C.
L. Orally Active and Potent Inhibitors of Gamma Aminobutyric
Acid uptake. J . Med. Chem. 1985, 28, 653-660.
(36) Yunger, L. M.; Fowler P. J .; Zarevics, P.; Setler, P. E. Novel
Inhibitors of γ-Aminobutyric Acid (GABA) Uptake: Anticonvul-
sant Actions in Rats and Mice. J . Pharm. Exp. Ther. 1984, 228,
109-115.
(55) Akkerman, A. M.; De J ongh, D. K.; Veldstra, H. Synthetic
Oxytocics. I. 3-(Piperidyl-(N)-Methyl)indoles and Related Com-
pounds. Rec. Trav. Chim. 1951, 70, 899-916. Bettoni, G.;
Duranti, E.; Tortorella, V. Absolute Configuration and Optical
Purity of 3-Substituted Piperidines. Gazz. Chim. Ital. 1972, 102,
189195.
(56) Kellogg, R. M.; Schaap, A. P.; Harper, E. T.; Wynberg, H. Acid-
Catalysed Brominations, Deuterations, Rearrangements of
Thiophenes under Mild Conditions. J . Org. Chem. 1968, 33,
2902-2909.
(37) Krogsgaard-Larsen, P.; Hjeds, H.; Falch, E.; J oergensen, F. S.;
Nielsen, L. Recent Advances in GABA Agonists, Antagonists and
Uptake Inhibitors: Structure-Activity Relationships and Thera-
peutic Potential. Adv. Drug. Res. 1988, 17, 381-456.
(57) Buehler, E. Alkylation of syn- and anti-Benzaldoximes. J . Org.
Chem. 1967, 32, 261-264.
(58) See, for example: Laforest, J .; Thuiller, G. Etude de la config-
uration d’oximes a activite cardiotrope. J . Heterocycl. Chem.
1977, 14, 793-796; Haney, W. G.; Brown, R. G.; Isaacson, E. I.;
Delgado, J . N. Synthesis and Structure Activity Relationships
of Selected Isomeric Oxime O-Ethers as anticholinergic agents.
J . Pharm. Sci. 1977, 66, 1602-1606. Conde, S.; Corral, C.;
Lissavetzky, J . E- and Z-isomerism of 2-acetylthiophene oximes.
J . Heterocycl. Chem. 1985, 22, 301-304.
(59) (a) Ashwood, M. S.; Bell, L. A.; Houghton, P. G.; Wright, S. H.
B. Synthesis of 1,1-Diaryl-2,-2-dimethoxyethanes. Synthesis
1988, 379-381. (b) Borch, R. F. A New Procedure for the
Darzen’s Synthesis of Glycidic Esters. Tetrahedron Lett. 1972,
36, 3761-3763. (c) Martin, S. F. Synthesis of Aldehydes, Ketones
and Carboxylic Acids from Lower Carbonyl Compounds by C-C
Coupling Reactions. Synthesis 1979, 633-665. (d) Maruoka, K.;
Nagahara S.; Ooi, T.; Yamamoto, H. An Efficient, Catalytic
Procedure for Epoxide Rearrangement. Tetrahedron Lett. 1989,
30, 5607-5610. (e) Meyers, A. I.; J agdmann, G. E. Enamidines.
Versatile Vehicles for Homologation of Carbonyl Compounds. J .
Am. Chem. Soc. 1982, 104, 877-879. (f) Mateson, D. S.; Moody,
R. J . Homologation of Carbonyl Compounds to Aldehydes with
Lithium Bis(ethylenedioxyboryl)methide. J . Org. Chem. 1980,
45, 1091-1095.
(38) Krogsgaard-Larsen, P.; Falch, E.; Larsson, O. M.; Schousboe,
A. GABA Uptake Inhibitors; Relevance to Antiepileptic Drug
Research. Epilepsy Res. 1987, 1, 7793.
(39) Fjalland, B. Inhibition by Neuroleptics of Uptake of [3H]-GABA
into Rat Brain Synaptosomes. Acta Pharmacol. Toxicol. 1978,
42, 7376.
(40) Falch, E.; Krogsgaard-Larsen, P. GABA Uptake Inhibitors.
Syntheses and Structureactivity Studies on GABA Analogues
Containing Diarybutenyl and Diarylmethoxyalkyl N-substitu-
ents. Eur. J . Med. Chem. 1991, 26, 6978.
(41) N’Goka, V.; Schlewer, G.; Linget, J .-M.; Chambon, J .-P.; Wer-
muth, C.-G. GABA-Uptake Inhibitors: Construction of a General
Pharmacophore Model and Successful Prediction of a New
Representative. J . Med. Chem. 1991, 34, 2547-2557.
(42) Falch, E.; Larsson, O. M.; Schousboe, A.; Krogsgaard-Larsen,
P. GABA-A agonists and GABA Uptake Inhibitors: Structure-
Activity Relationships. Drug Dev. Res. 1990, 21, 169-188.
(43) Dhar, T. G. M.; Borden, L. A.; Tyagarajan, S.; Smith, K. E.;
Branchek, T. A.; Weinshank, R. L.; Gluchowski, C. Design,
Synthesis and Evaluation of Substituted Triarylnipecotic Acid
Derivatives as GABA Uptake Inhibitors: Identification of a
Ligand with Moderate Affinity and Selectivity for the Cloned
Human GABA Transporter GAT-3. J . Med. Chem. 1994, 37,
2334. See also refs 48-51 and Dhar, T. G. M.; Nakanishi, H.;
Borden, L. A.; Gluchowski, C. On the bioactive conformation of
the GABA uptake inhibitor SK&F 89976A. BioMed. Chem. Lett.
1996, 1535-1540.
(44) Knutsen, L. J . S.; Andersen, K. E.; J ørgensen, A. S.; Sonnewald,
U. Azacyclic Carboxylic Acid Derivatives, Their Preparation and
Use. Novo Nordisk A/S, U.S. Patent 5,039,685, August 13, 1991.
(45) Knutsen, L. J . S.; J ørgensen, A. S.; Andersen, K. E.; Sonnewald,
U. N-Substituted Azaheterocyclic Carboxylic Acids and Phar-
maceutical Uses. Novo Nordisk A/S, U.S. Patent 5,071,859,
December 10, 1991.
(60) Blicke, F. F.; Faust, J . A. Antispasmodics. XV. â-Diethylamino-
ethyl Esters of â,â-Diphenylglycidic, â,â-Diphenyllactic and â,â-
Diphenylglyceric Acids. J . Am. Chem. Soc. 1954, 76, 3156-3159.
(61) Ridogrel (R-68070). Drugs Fut. 1991, 16, 488-489.
(62) vant Reit, B.; Elford, H. L. Amidox. Drugs Fut. 1991, 16, 990-
991.
(63) De Chaffoy de Courcelles, D.; De Loore, K.; Freyne, E.; J anssen,
P. A. J . Inhibition of human cardiac cyclic AMP-phosphodi-
esterases by R 80122, a new selective cyclic AMP-phosphodi-
(46) Klitgaard, H.; Knutsen, L. J . S.; Thomsen, C. Contrasting effects
of adenosine A1 and A2 receptor ligands in different chemo-
convulsive models. Eur. J . Pharmacol. 1993, 224, 221-228.
(47) Suzdak, P. D.; Frederiksen, K.; Andersen, K. E.; Sørensen, P.
O.; Knutsen, L. J . S.; Nielsen, E. B. NNC-711, a Novel Potent
and Selective GABA Uptake Inhibitor: Pharmacological Char-
acterization. Eur. J . Pharmacol. 1992, 223, 189-198. For DMCM
method descriptions, see also ref 19 and Swedberg, M. D. B.;
J acobsen, P.; Honore´, T. Anticonvulsant, anxiolytic and dis-
criminative effects of the AMPA antagonist 2,3-dihydroxy-6-
nitro-7-sulfamoyl-benzo[f]quinoxaline (NBQX). J . Pharmacol.
Exp. Ther. 1995, 274, 1113-1121.
(48) Borden, L. A. GABA transporter heterogeneity: pharmacology
and cellular localization. Neurochem. Int. 1996, 29, 335-356.
Note that this article and ref 43 utilize different terminology
for the GAT transporters from that in ref 49.
(49) Liu, Q. R.; Lo´pez-Corcurera, B.; Mandiyan, S.; Nelson, H.;
Nelson, N. Molecular Characterization of Four Pharmacologi-
cally Distinct γ-Aminobutyric Acid Transporters in Mouse Brain.
J . Biol. Chem. 1993, 268, 2106-2112.
esterase III inhibitor:
a comparison with other cardiotonic
compounds. J . Pharmacol. Exp. Ther. 1992, 263, 6-14.
(64) Schwarz, R. D.; Davis, R. E.; J aen, J . C.; Spencer, C. J .; Tecle,
H.; Thomas, A. J . Characterization of muscarinic agonists in
recombinant cell lines. Life Sci. 1993, 52, 465-72.
(65) Nielsen, L.; Brehm, L.; Krogsgaard-Larsen, P. GABA agonists
and uptake inhibitors. Synthesis, absolute stereochemistry, and
enantioselectivity of (R)-(-)- and (S)-(+)-homo-â-proline J . Med.
Chem. 1990, 33, 71-7.
(66) Dalby, N. O.; Thomsen, C.; Fink J ensen, A.; Lundbeck, J .;
Sokilde, B.; Man, C. M.; Sorensen, P. O.; Meldrum, B. Anticon-
vulsant properties of two GABA uptake inhibitors NNC 05-2045
and NNC 05-2090, not acting preferentially on GAT-1. Epilepsy
Res. 1997, 28, 51-61.
(67) Thomsen, C.; Sorensen, P. O.; Egebjerg, J . 1-(3-(9H-Carbazol-
9-yl)-1-propyl)-4-(2-methoxyphenyl)-4-piperidinol, a novel subtype-
selective inhibitor of the mouse type-II GABA transporter. Br.
J . Pharmacol. 1997, 120, 983-985.
(68) Still, W. C.; Kahn, M.; Mitra, A. Rapid Chromatographic
Technique for Preparative Separations with Moderate Resolu-
tion. J . Org. Chem. 1978, 43, 29235.
(69) Morlacchi, F.; Cardellini, M.; Liberatore, F. Preparation of
unsaturated heterocyclic compounds, II. Synthesis of guvacine.
Ann. Chim. 1967, 57, 1456-1461.
(70) Hassner, A.; Alexanian, V. Direct room temperature esterifica-
tion of carboxylic acids. Tetrahedron Lett. 1978, 46, 4475-4478.
(71) Braestrup, C.; Nielsen, E. B.; Sonnewald, U.; Knutsen, L. J . S.;
Andersen, K. E.; J ansen, J . A.; Frederiksen, K.; Andersen, P.
H.; Mortensen, A.; Suzdak, P. D. (R)-N-(4,4-Bis-(3-Methyl-2-
Thienyl)-but-3-en-1-yl)nipecotic acid binds with high affinity to
the brain γ-aminobutyric acid uptake carrier. J . Neurochem.
1990, 54, 639-647.
(50) Borden, L. A.; Dhar, T. G. M.; Smith, K. E.; Weinshank, R. L.;
Branchek, T. A.; Gluchowski, C. Tiagabine, SK&F 89976A, CI-
966, and NNC-711 are selective for the cloned GABA transporter
GAT-1. Eur. J . Pharmacol. 1994, 269, 219-224.
(51) Bennett, E. R.; Kanner, B. I. The membrane topology of GAT-1,
a (Na+Cl-) -coupled gamma-aminobutyric acid transporter from
rat brain. J . Biol Chem. 1997, 272, 1203-1210.
(52) Program Spartan Version 4.0, available from Wave Function,
Inc., 18401 Von Karman, Suite 370, Irvine, CA 92715. Ab initio
level used for the calculations was HF 6-31G**.
(53) NNC 05-071147 is marketed as
a pharmacological tool by
Research Biochemicals, Inc., One Strathmore Road, Natick, MA
017602418.
(54) Richards, D. A.; Bowery, N. G. Comparative effects of the GABA
uptake inhibitors, tiagabine and NNC-711, on extracellular
GABA levels in the rat ventrolateral thalamus. Neurochem. Res.
1996, 21, 135-140.
J M981027K