N-Substituted Nipecotic Acids as (S)-SNAP-5114 Analogues with Modified Lipophilic Domains

Article abstract of DOI:10.1002/cmdc.201900719

Potential mGAT4 inhibitors derived from the lead substance (S)-SNAP-5114 have been synthesized and characterized for their inhibitory potency. Variations from the parent compound included the substitution of one of its aromatic 4-methoxy and 4-methoxyphenyl groups, respectively, with a more polar moiety, including a carboxylic acid, alcohol, nitrile, carboxamide, sulfonamide, aldehyde or ketone function, or amino acid partial structures. Furthermore, it was investigated how the substitution of more than one of the aromatic 4-methoxy groups affects the potency and selectivity of the resulting compounds. Among the synthesized test substances (S)-1-{2-[(4-formylphenyl)bis(4-methoxyphenyl)-methoxy]ethyl}piperidine-3-carboxylic acid, that features a carbaldehyde function in place of one of the aromatic 4-methoxy moieties of (S)-SNAP-5114, was found to have a pIC₅₀ value of 5.89±0.07, hence constituting a slightly more potent mGAT4 inhibitor than the parent substance while showing comparable subtype selectivity.

Full text of DOI:10.1002/cmdc.201900719

Accepted Article
Title: N-Substituted Nipecotic Acids as (S)-SNAP-5114 Analogues with
Modified Lipophilic Domains
Authors: Michael C. Böck, Georg Höfner, and Klaus Theodor Wanner
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemMedChem 10.1002/cmdc.201900719
Link to VoR: http://dx.doi.org/10.1002/cmdc.201900719
A Journal of
www.chemmedchem.org

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
N-Substituted Nipecotic Acids as (S)-SNAP-5114 Analogues with Modified
Lipophilic Domains
Michael C. Böck, Dr. Georg Höfner, Prof. Dr. Klaus T. Wanner[a]
Dedicated to Prof. Dr. Theodor Severin with warmest wishes on the occasion of his 90th birthday.
[a]
Department of Pharmacy – Center for Drug Research
Ludwig-Maximilians-Universität München
Butenandtstraße 5-13, 81377 Munich, Germany
klaus.wanner@cup.uni-muenchen.de
Supporting information for this article is given via a link at the end of the document
Abstract: Potential mGAT4 inhibitors derived from the lead substance (S)-
SNAP-5114 have been synthesized and characterized for their inhibitory
potency. Variations from the parent compound included the substitution of one
of its aromatic 4-methoxy and 4-methoxyphenyl groups, respectively, with a
more polar moiety, including a carboxylic acid, alcohol, nitrile, carboxamide,
sulfonamide, aldehyde or ketone function, or amino acid partial structures.
Furthermore, it was investigated how the substitution of more than one of the
aromatic 4-methoxy groups affects the potency and selectivity of the resulting
compounds. Among the synthesized test substances (S)-1-{2-[(4-
formylphenyl)bis(4-methoxyphenyl)-methoxy]ethyl}piperidine-3-carboxylic acid,
that features a carbaldehyde function in place of one of the aromatic 4-
methoxy moieties of (S)-SNAP-5114, was found to have a pIC50 value of 5.89
± 0.07, hence constituting a slightly more potent mGAT4 inhibitor than the
parent substance while showing comparable subtype selectivity.
(S)-SNAP-5114 (S)-2 (table 1, entry 3) can be considered the benchmark
mGAT4 inhibitor. Since its publication,[24] it has been the prototype for the
development of further mGAT4 inhibitors with the goal of increasing potency,
subtype selectivity, and chemical stability. The structural modifications
implemented in the (S)-SNAP-5114 scaffold so far include variations of the
spacer between the nipecotic acid partial structure and the trityl rest, as it is for
example the case with DDPM-1457 3 (table 1, entry 4),[25] resulting in
compounds with enhanced stability due to the labile trityl ether function being
avoided. Variations of the substitution pattern of the trityl structure, e.g. by
introduction of an additional methyl group in the 2-position of one of the three
aryl residues (4, table 1, entry 5), was found to lead to compounds with slightly
improved subtype selectivity. Unfortunately, the moderate potency inherent to
the parent compound (S)-2, which is characterized by a pIC50 value of 5.71 ±
0.07 (table 1, entry 3), remains largely unaffected by these structural
modifications. This underlines the necessity of further research in order to
advance the general understanding of the structure activity relationship (SAR)
of mGAT4 inhibitors.
Introduction
The neuronal signal transduction in the mammalian central nervous system
(CNS) is regulated by a complex equilibrium of various excitatory and
inhibitory neurotransmitters. γ-Aminobutyric acid (GABA) is the most abundant
of the latter,[1] with approximately 40% of synapses estimated to be
GABAergic.[2] Pathologically deficient GABA release into the synaptic cleft
results in attenuation of the inhibitory component, which is associated with a
variety of severe neurological disorders including epilepsy,[3,4] neuropathic
To this end, the present study aims at the synthesis of (S)-SNAP-5114 [(S)-2]
analogues that feature a more polar moiety in place of one of the methoxy
groups present in the trityl rest of the parent compound (S)-2 (Scheme 1, a).
As such polar moieties carboxylic acid, alcohol aldehyde, nitrile, carboxamide,
sulfonamide, aldehyde or ketone functions were taken into consideration, as
well as amino acid partial structures. These modifications might result in
increased polar interactions of the inhibitor with the target, thus possibly
affecting its potency and subtype selectivity. In that context, we also aimed to
clarify how a wider variation of the original (S)-SNAP-5114 [(S)-2] structure,
comprising the replacement of one entire 4-methoxyphenyl moiety of the trityl
residue by a polar group (Scheme 1, d), would influence the biological activity.
A further question addressed with this study is how the inhibitory potency of
(S)-SNAP-5114 [(S)-2] analogues that have two (Scheme 1, b) or all three
(Scheme 1, c) 4-methoxy groups in the trityl moiety of (S)-2 substituted with
other residues compare to inhibitors with only one such alteration. The findings
would allow to draw conclusions about the SAR of mGAT4 inhibitors with
regard to polarity, size, and number of matching substituents in the aromatic
domain, and therefore point out important aspects to consider in future
development of more potent mGAT4-inhibitors.
pain,[5] anxiety disorders,[6] depression,[6,7] and Alzheimer´s disease.[8]
A
promising approach to the treatment of these disorders is the application of
drugs that inhibit the GABA reuptake[8,9,10,11] into the presynaptic neurons and
the surrounding glial cells, respectively, thereby enhancing the GABA
concentration in the synaptic cleft and thus prolonging the effect of the
released GABA. The specific and high affinity membrane based[12] transport
proteins accomplishing the GABA reuptake (GATs) are thus an important
therapeutical target. Belonging to the solute carrier 6 (SLC6) family,[13] they
consist of 12 transmembrane helices, four of which (TM 1, TM3, TM6 and
TM8) form the inner ring that in its center halfway across the cell membrane
holds the central substrate binding site S1. Separated from the S1 binding site
by the extracellular gate a second substrate binding site exists at the bottom of
the extracellular vestibule, termed S2,[12] the occupation of which is assumed
to trigger the conformational changes necessary for the release of the
substrate from the S1 pocket into the cell.[14] Four GAT subtypes have been
identified, termed GAT1, GAT2, GAT3, and BGT1 in accordance with the
nomenclature proposed by the Gene Nomenclature Committee of the Human
Genome Organisation (HUGO),[15] or, when cloned from mouse brain, mGAT1
(≙ GAT1), mGAT2 (≙ BGT1), mGAT3 (≙ GAT2) and mGAT4 (≙GAT3).[16,17]
Since the biological test system developed in our group is based on GABA
transporters cloned from mouse cells, in this paper the corresponding
nomenclature will be used.
The predominant GABA transporter in the mammalian CNS, mGAT1, is
located primarily in pre-synaptic neuronal membranes,[18] with its highest
densities found in neocortex, spinal cord, brainstem, cerebellum, basal
ganglia, and hippocampus.[2,19] As the occurrence of mGAT2 and mGAT3 in
the CNS is restricted to low densities in specific brain structures, these
subtypes play only
a marginal role in the termination of the cerebral
GABAergic neurotransmission.[20,21] mGAT4 is the second most abundant
GABA transporter after mGAT1 and found particularly in olfactory bulb,
brainstem, and diencephalon,[22] where it is most commonly expressed on glia
cells.[18] The selective targeting of mGAT4 may therefore provide the possibility
of treating neurological disorders associated with these brain regions with
minimal impairment of the GABA reuptake in other parts of the CNS.
Compared to mGAT1-selective inhibitors such as Tiagabine (Gabatril®) (1,
table 1, entry 1), adverse effects, including dizziness, somnolence, headache,
memory loss, and tremor,[23] could hence be possibly reduced, resulting in
more tolerable medication and better patient compliance.
Scheme 1. Overview of the structural modifications of (S)-SNAP-5114 conducted in this study
Table 1. Binding affinities (pKi) and inhibitory potencies (pIC50) of reference compounds 1-5 from the
literature.
1
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
Initially for the construction of the target compounds 5a-i in which one or more
of the three methoxy substituents of (S)-SNAP-5114 [(S)-2] are replaced by an
alternative polar moiety (compound type a-c in Scheme 1), the corresponding
tertiary alcohols 8a-i had to be synthesized (table 2, entry 1-9, table 3, entry
9). This was accomplished by transforming aryl halides 9a-h into the
corresponding Grignard or organolithium reagents, which were subsequently
reacted
with
the
appropriate
electrophiles,
i.e.
with
4,4´-
dimethoxybenzophenone (10a, table 2, entry 1-6), 4-cyanobenzoyl chloride
(10b, table 2, entry 7), methyl 4-methoxybenzoate (10c, table 2, entry 8), and
dimethyl carbonate (10d, table 2, entry 9). This led to the differently
substituted trityl alcohols 8a-i in acceptable to good yields of 53% – 99%, most
of which exhibit two of the three 4-methoxyphenyl units present in (S)-SNAP-
5114 [(S)-2] together with a third aryl moiety with a different structure (table 2,
entry 1-7). Trityl alcohol 8j (table 3, entry 9), featuring an amide group and two
methoxy groups, respectively, in the 4-position of the three aromatic moieties,
was obtained in a yield of 96% by hydratisation of the nitrile function of 8g
(table 2, entry 7), which was accomplished by the treatment with potassium
tert-butoxide in tert-butanol, following a general procedure from literature.[29]
4,4´-Dimethoxybenzilic acid methyl ester 8k (table 3, entry 10) as precursor for
the preparation of compounds of type d (Scheme 1) was synthesized by
benzilic acid rearrangement of 4,4’-dimethoxybenzil,[30] followed by
esterification with methyl iodide in analogy to a method published for benzilic
acid.[31]
Construction of target compounds 5a-l from N-(2-hydroxyethyl)nipecotic
acid ethyl ester 6 and tertiary alcohols 8a-l
The synthesis of the target compounds 5a-k should be achieved by
etherification of the hydroxy function of N-(2-hydroxyethyl)nipecotic acid ethyl
ester 6 with the tertiary alcohols 8a-k that had been prepared for this purpose.
To this end, in the first step alcohols 8a-k were transformed into the
corresponding tertiary chlorides 7a-k by reaction with acetyl chloride in the
presence of a catalytic amount of dimethyl formamide (table 3, step a). Under
these conditions, also the acetal function of 8a was completely transformed in
an aldehyde group as wanted. Therefore, the reaction mixture obtained after
treatment of 8a with acetylchloride, resulting in 7a with the deprotected
aldehyde function, could be directly used for the next step, the etherification
reaction with N-(2-hydroxyethyl)nipecotic acid ethyl ester (6). In contrast, the
more stable acetal function of 8b was only partially transformed into the
ketone under the conditions for the halide formation. Hence, 8b was first
transformed into ketone 11 by refluxing in aqueous acid (93% yield), which
was subsequently converted in chloride 7b. Due to their high reactivity and
susceptibility to hydrolysis upon exposure to moisture, the tertiary chlorides
were directly used without prior purification or characterization for the next
step, the alcoholysis with racemic 1-(2-hydroxyethyl)nipecotic acid ethyl ester
(6) (table 3, step b). The desired products 12a-k with the newly created trityl
ether function were thus obtained in yields from 42% to 88% over both
reaction steps (based on 8a-k as starting material).
pKi[a]
pIC50
mGAT2
[b]
Ent.
Compound
mGAT1
mGAT1
mGAT3
64%
mGAT4
73%
Tiagabine
7.43
± 0.11
6.88
± 0.12
1
2
3
4
5
50%
-
[(R)-1][d]
rac-SNAP-5114
[rac-2][c]
5.64
± 0.05
-
4.08
4.96
(S)-SNAP-5114
[(S)-2][c]
4.56
± 0.02
4.07
± 0.09
5.29
± 0.04
5.71
± 0.07
56%
DDPM-1457
4.33
± 0.06
4.40
± 0.05
4.42
± 0.11
5.47
± 0.02
5.87
± 0.08
[(S)-3][d]
DDPM-859
4.19
± 0.07
4.12
± 0.08
4.85
± 0.04
5.78
± 0.03
-
[(S)-4][d]
[a] Results of the MS Binding Assays are given as pKi ± SEM. [b] Results of the [3H]GABA uptake
assays are given as pIC50
±
SEM. Percent values represent remaining [3H]GABA uptake in
presence of 100 µM compound. [c] Reference literature [25]. [d] Reference literature [26]
.
Results and Discussion
Chemistry
For the synthesis of an analogue of rac-SNAP-5114 [(S)-2] that features a
hydroxyl function in place of one of the 4-methoxyphenyl groups, we decided
to use a propyl instead of the ethoxy linker to warrant chemical stability of the
product. The synthesis was accomplished by reacting anisyl lithium with ethyl
4-chlorobutanoate to give the required alcohol 8l with an γ-chloropropyl
residue, which upon reaction with ethyl nipecotinate led to the desired N-
substituted nipecotic acid ester 12l in 66% yield (table 3, step c).
Synthesis of the desired (S)-SNAP-5114 [(S)-2] analogues was performed
according to the reaction sequence shown in scheme 2 that had been
described by Schirrmacher et al.[27] for the preparation of [18F] labelled (S)-
SNAP-5114 analogues. According to this plan, the respective tertiary alcohols
8 exhibiting the desired polar function attached to the aryl moieties, or suitable
precursors thereof, are transformed into the corresponding trityl chlorides,
which can e.g. be accomplished by reaction with acetyl chloride. Subsequent
treatment of the prepared trityl chlorides 7 with N-(2-hydroxyethyl)nipecotinate
6, which may be prepared according to literature,[28] will furnish the fully
assembled target compounds in form of their carboxylic acid esters and finally
the free nipecotic acids 5 upon hydrolysis of the carboxylic acid ester function.
Although the stereochemistry of the nipecotic acid partial structure is known to
play a decisive role in the biological activity of mGAT4 inhibitors,[24] we opted
for the synthesis of racemic compounds for economic reasons. However, for
those racemic compounds showing equal or higher activity than (S)-SNAP-
5114 in the biological testing, additionally the (R)- and (S)-isomers should be
synthesized and evaluated for their biological activity. To achieve the
synthesis of these enantiopure compounds following the depicted synthetic
pathway (scheme 2), only rac-6 has to be replaced by its (R)- and (S)-isomer,
respectively.
In order to obtain the free nipecotic acid derivatives 5a-l, compounds 12a-l
were treated with barium hydroxide octahydrate in methanol/water 4:1 for the
hydrolysis of the ester function, followed by workup with carbon dioxide (table
3, step d). Thereupon, the target compounds 5a-l could be isolated in yields
from 61% to 97%.
Synthesis of target compounds 5m-t with amino acid derived residues
substituting one of the three methoxy groups in (S)-SNAP-5114 [(S)-2]
Next, we aimed at the synthesis rac-SNAP-5114 analogues that feature a
hydroxy methyl moiety or an amino acid subunit as a polar group replacing
one of the three methoxy residues in the lipophilic domain of (S)-SNAP-5114
[(S)-2]. As starting material for the synthesis compound 12a, a nipecotic acid
derivative with one of the three methoxy groups of the trityl moiety having
been replaced by a formyl residue was used. Reduction of the aldehyde
function in 12a was accomplished by treatment with sodium borohydride in
methanol, providing the corresponding alcohol 12m in a yield of 89% (table 4,
entry 1). Reductive amination of the aldehyde function in 12a was performed
employing a set of amino acids and amino acid ester hydrochlorides (table 4,
entry 2-8). Upon application of
a
slightly modified standard method[32]
employing sodium acetoxyborohydride in dichloromethane as reductant
(instead of dichloroethane) to 12a and the respective amino acid ester
hydrochlorides led to the glycine methyl ester 12n, β-alanine ethyl ester 12o,
1-aminocyclopropane-1-carboxylic acid ethyl ester 12p, and 2-amino-2-
methylpropanoic acid ethyl ester derivatives 12q (table 4, entry 2-5) in yields
of 42 to 64%. The γ-aminobutyric acid ethyl ester derivative of 12a could not
Scheme 2. Retrosynthetic analysis for the preparation of (S)-SNAP-5114 analogues
Synthesis of the tertiary alcohols 8a-k
2
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
be obtained by this procedure since the product underwent partial γ-
lactamization during workup, resulting in a mixture of the desired compound
and the corresponding γ-butyrolactame. Attempts to reopen the lactame
function under basic conditions (barium hydroxide or sodium hydroxide at rt or
reflux) (table 3, entry 6) led to a completion of the formation of lactame 5r, or
to cleavage of the ether function when lithium hydroxide was used under reflux
conditions. Hence, the reductive amination of 12a was attempted with the free
γ-aminobutyric acid (GABA). Applying the established reaction conditions did
not lead to the desired product 12s, which is likely to be due to the very low
solubility of GABA in dichloromethane. However, when dichloromethane was
replaced by methanol and sodium triacetoxyborohydride by sodium
cyanoborohydride, 12s (table 4, entry 7) was formed in a yield of 40%. The
same method could also be successfully applied to the synthesis of p-
aminobenzoic acid derivative 12t (table 4, entry 8). The free nipecotic acids
5m-q and 5s-t became finally available upon subjecting compounds 12m-q
and 12s-t to alkaline hydrolysis using the procedure as already described
above [Ba(OH)2 · 8H2O, CO2 workup] for the transformation of compounds
12a-l into 5a-l (table 4, step b), the yields of this transformation amounting to
65-96%.
Table 2. Synthesis of the trityl alcohols 2a-i.
Metallation
reagent
Entry
1
Aryl halide
Electrophile
Product
R1
R2
R3
Yield (%)
99
9a
9b
t-BuLi
10a
10a
8a
-CH(OMe)2
-OMe
-OMe
t-BuLi
t-BuLi
8b
8c
2
3
-OMe
-OMe
-OMe
-OMe
87
53
9c[b]
10a
-SO2NMe2
9d
9e
9f
n-BuLi
10a
10a
10a
8d
8e
8f
4
5
6
-OMe
-OMe
-OMe
-OMe
-OMe
-OMe
56
77
92
i-PrMgCl
-COOEt
-CH2OMe
-CN
Mg
Mg
Mg
Mg
9g
9f
10b
10c
10d
8g
8h
8i
7
8
9
-OMe
-OMe
-OMe
94
70
85
-CH2OMe
-CH2OMe
-CH2OMe
-CH2OMe
9f
-CH2OMe
Reagents and conditions: [a] entry 1-3: t-BuLi (2.0 eq), THF, -78°C, 2h, 4,4´-dimethoxybenzophenone (1.0 eq), THF, -78°C-rt; entry 5: iPrMgCl (1.0 eq), THF, -
20°C, 1.5h, 4,4´-dimethoxybenzophenone (1.0 eq), THF, -20°C-rt; entry 6: n-BuLi (1.0 eq), diethyl ether, -78°C, 0.5h, 4,4´-dimethoxybenzophenone (0.83 eq),
diethyl ether, -78°C-rt; entry 7: magnesium (1.0 eq), THF, rt, 4-cyanobenzoylchloride, THF, 0°C-rt; entry 8: magnesium (1.0 eq), THF, rt, methyl 4-
methoxybenzoate (0.89 eq), THF, reflux; entry 9: magnesium (1.0 eq), THF, rt, dimethyl carbonate (0.33 eq), THF, reflux. [b] Synthesized by N-alkylation of 4-
bromobenzenesulfonamide with dimethyl sulfate (2.0 eq) in presence of potassium carbonate (4.0 eq) and tetrabutylammonium tetrafluoroborate (10 mol%) under
reflux conditions.
3
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
Table 3. Synthesis of the N-substituted nipecotic acid ethyl esters 12a-k and their hydrolysis to the free nipecotic acid derivatives 5b-k.
Starting
material
Product of
step a+b
Yield
(%)[f]
Product
of step d
Yield
(%)
Entry
R1
R2
R3
R1
R1
OMe
OMe
8a
8b
12a
12b
5a
5b
1
2
-OMe
-OMe
-OMe
-OMe
70
50
89
86
O
Me
S N
Me
8c
8d
8e
12c
12d
12e
5c
5d
5e
3
4
5
-OMe
-OMe
-OMe
-OMe
-OMe
-OMe
64
48
54
88
80
61
O
O
OEt
8f
8g
8h
12f
12g
12h
5f
5g
5h
6
7
8
-OMe
-OMe
-OMe
-OMe
-OMe
44
85
46
81
86
94
-CH2OMe
8i
8j
12i
12j
5i
5j
9
-OMe
-OMe
-CH2OMe
-OMe
42
49
90
87
O
O
10
NH2
NH2
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
8k
12k
5k
11
-COOMe
-OMe
-OMe
-COOMe
88
-COOH
97
Reagents and conditions: [a] acetyl chloride, dimethyl formamide (cat.), rt; [b] HCl, THF/H2O, reflux; [c] 1-(2-hydroxyethyl)nipecotic acid ethyl ester (6) (1.1 eq),
potassium carbonate (2.5 eq), acetontrile, rt; [d] barium hydroxide octahydrate (2-4 eq), methanol/water 4:1; carbon dioxide; [e] ethyl nipecotinate (1.1 eq),
potassium carbonate (2.5 eq), potassium iodide (0.1 eq), acetonitrile, microwave, 80°C. [f] Yield over two steps.
Table 4. synthesis of the target compounds 5m-t with amino acid derived residues substituting one of the three methoxy groups in (S)-SNAP-5114.
O
O
O
OH
OEt
OEt
N
N
N
O
O
O
b
O
H
a
CH2R1
MeO
CH2R2
MeO
MeO
OMe
OMe
OMe
12m-q
12s-t
5m-q
5s-t
12a
Entry
Product of step a
R1
Yield (%)
89
Product of step b
R2
Yield (%)
69
12m
5m
1
2
-OH
-OH
12n
12o
12p
12q
5n
5o
5p
5q
5r
57
42
59
64
65
74
H
N
O
3
4
5
6
OEt
82
92
47[c]
12s
12t
5s
5t
7
8
40
25
84
96
Reagents and conditions: [a] entry 1: sodium borohydride (2.5 eq), methanol, rt; entry 2-5: sodium triacetoxyborohydride (1.4 eq), amino acid ethylester
hydrochloride (2.0 eq), dichloromethane, rt; entry 7-8: sodium cyanoborohydride (1.4 eq), free amino acid (2.0 eq), methanol, rt. [b] barium hydroxide octahydrate
(2-4 eq), methanol/water 4:1; carbon dioxide. [c] Yield over two steps.
under physiological conditions (pH = 7.4, electrolyte concentration = 0.154
Biological Evaluation
mmol/l) were calculated (clog D: calculated log D) using MarvinSketch.[35]
The N-substituted nipecotic acids 5a-t, and their ester precursors 12a-q and
Replacement of one of the methoxy substituents in the rac-SNAP-5114 (rac-2)
12s-t were tested for their inhibitory potencies on the four GABA transporter
subtypes mGAT1-4 in
a
[3H]GABA uptake assay that was previously
molecule by the larger methoxymethylene group leads to 5f, which at mGAT4
developed by our group.[33] The tests were performed in a standardized
manner using HEK293 cell lines, each expressing one of the four tested GABA
transporter subtypes. Furthermore, binding affinities towards mGAT1 were
examined employing a standardized MS Binding Assay with NO711 as native
MS marker.[34] For compounds that in prelimary experiments did not reduce
[3H]GABA uptake beyond 50% at a test concentration of 100 µM, which equals
a pIC50 of ≤ 4.0, only the percent values of the remaining [3H]GABA uptake are
listed. Correspondingly, the percentage of remaining marker is given in cases
when the tested compound did not cause a reduction of the MS marker
binding beyond 50%, equating to a pKi of ≤ 4.0. When [3H]GABA uptake or
NO711 binding was reduced below 50%, at a concentration of 100 µM,
inhibitory potencies (pIC50 values) and binding affinities (pKi values) were
determined in full scale [3H]GABA uptake and MS Binding Assays,
respectively, measurements being performed as triplicates. For compounds
with pIC50 ([3H]GABA uptake assay) or pKi (MS Binding Assays) values close
to or above 5.0, these experiments were repeated twice and SEM have been
calculated. The results are summarized in table 5.
exhibits a pIC50 value of 5.42 ± 0.10 (table 5, entry 14). This constitutes only a
minor decrease in inhibitory potency at the transporter compared to the parent
compound rac-2 (pIC50 = 5.64 ± 0.05, table 1, entry 2), suggesting that a slight
increase in the space requirements of the substituent is tolerated relatively
well by mGAT4. If a nitrile function is introduced in place of a methoxy moiety,
the loss of potency at mGAT4 is more pronounced, amounting to a pIC50 value
5.07 ± 0.12 (5g, table 5, entry 16). Since the nitrile function is also a hydrogen
bridge acceptor and conveys very similar polarity compared to the methoxy
moiety it replaces (clog D of 5g 2.33, clog D of rac-2 2.32), this finding is likely
attributable to the linear shape of the nitril function not being ideal for
interaction with the target. For rac-5a, featuring an aldehyde function in one of
the aromatic 4-positions of rac-SNAP-5114 (rac-2), a pIC50 of 5.77 ± 0.04 was
determined (table 5, entry 2), which places the potency of compound rac-5a
nominally above that of rac-SNAP-5114 (rac-2) despite the oxygen atom being
placed one bond further away from the aromatic moiety. 5b, which has one of
the methoxy groups of rac-2 replaced by the larger, but similar polar acetyl
moiety (5b: clog D = 2.00), exhibits only a slightly decreased inhibitory potency
compared to the parent compound, with its pIC50 value amounting to 5.43 ±
0.04 (table 5, entry 6). On the other hand, the introduction of
a
As outlined above, the structure of the prototypic mGAT4 inhibitor rac-SNAP-
5114 (rac-2) was modified by formally replacing one of the three aromatic
methoxy moieties in its lipophilic domain with a variety of different functional
groups. For the purpose of estimating the effect that these modifications exert
on the polarity of the resulting molecule, the log D values of the compounds
hydroxymethylene group, leading to the distinctly more polar compound 5m
(clog D = 1.71), is accompanied by a reduction of the pIC50 value from 5.64 ±
0.05 for rac-SNAP-5114 to 4.90, which equates to a potency loss of ~0.75 log
units (table 5, entry 28). Interestingly, an equivalent pIC50 of 4.86 was
5
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
determined for 5e (table 5, entry 12), which has one of the three aromatic 4-
positions of rac-2 occupied by a carboxylic acid moiety. This finding is
astonishing as the carboxyl group is significantly more polar than the
corresponding alcohol moiety present in 5m, with the respective clog D value
amouting to -1.40 under physiological conditions.
function of 5a with the substrate does not, or at least not crucially, contribute to
the biological effects oberserved for this compound.
Due to the presence of an aldehyde function in 5a, it had to be considered
whether the compound 5a might form covalent bonds with the target. In order
to explore this possibility, we examined 5a in comparison with (S)-SNAP-5114
[(S)-2] for inhibition of GABA uptake at mGAT4 by varying the time period
used for preincubation (i.e. the period in which target and test compound are
in contact, before the substrate is added). We studied preincubation times of
10, 25 min (which is our standard preincubation time in our GABA uptake
assays) as well as 60 min and compared the inhibitory potencies of 5a and
(S)-SNAP-5114 [(S)-2] under these conditions. The determined pIC50-values
revealed a very slight enhancement of inhibitory potencies with increasing
preincubation times almost exactly to the same extent for both compounds
(see supporting information, table 1). As 5a and (S)-SNAP-5114 [(S)-2]
behaved identical in this respect, the basic way of interaction of both
compounds with mGAT4 seems to be similar, arguing against a covalent
interaction of 5a with mGAT4.
By contrast, the inhibitory potency of 5j, featuring a carboxamide moiety in this
position, is strongly reduced (pIC50 < 4.0, remaining [3H]GABA at 100 µM =
65.2 %, table 5, entry 22), irrespective of the fact that this group is of similar
size as the carboxylic acid found in 5e while conveying a less pronounced
increase in polarity (clog D = 1.32). Apparently, no correlation between polarity
and inhibitory potency seems to exist. 5c, which features
a N,N-
dimethylsulfonamide group (clog D = 1.98) in place of one of the methoxy
groups of rac-2, is characterized by an even lower activity; the compound
causes a [3H]GABA uptake reduction to 82.0 % at 100 µM, denoting a pIC50 of
well below 4.0 (table 5, entry 8).
Of the compounds obtained from 12a by reductive amination and subsequent
hydrolysis, 5n, featuring a glycine partial structure that is connected to the 4-
position of one of the aromatic moieties by a methylene spacer, reduces the
[3H]GABA uptake to 72.2 % at 100 µM test compound concentration (table 5,
entry 30); its analogue derived from β-alanine, 5o, causes a slightly more
pronounced reduction to 56.3% (table 5, entry 32). In this context, it is
noticeable that the potency of the free nipecotic acid derivatives 5n and 5o is
even lower than that of the corresponding esters 12n [pIC50 (mGAT4) = 4.18,
Table 5, entry 29] and 12o [pIC50 (mGAT4) = 4.66, Table 5, entry 31]. By
contrast, compounds that contain bulky, sterically hindered amino acid partial
structures in the same position, such as 1-aminocyclopropane-1-carboxylic
acid (5p) and 2-amino-2-methylpropanoic acid (5q), are characterized by a
pIC50 of 4.88 (table 5, entry 34) and 4.34 (table 5, entry 36), respectively.
Since these substances retain comparatively more of the inhibitory activity at
mGAT4 of the parent compound, increased space requirements do evidently
not contribute to the decline in potency in this case. Compound 5s, which
formally results from elongation of the amino acid chain of 5o by one
methylene group, hence exhibiting a γ-aminobutyric acid subunit, shows
slightly higher inhibitory potency with a pIC50 value of 4.13 compared to its
shorter chain analogues (compare table 5, entry 39 to entry 30 and 32).
Compound 5t, whose 4-aminobenzoic acid partial structure resembles to
some extent a more rigid homologue of the γ-aminobutyric acid moiety present
in 5s, is characterized by equal inhibitory potency (remaining [3H]GABA at
100µM = 50.0%, table 5, entry 41). Surprisingly, 5r, featuring a lactame
structure derived from the γ-aminobutyric acid substructures present in 5s,
exhibits a similar activity at mGAT4 (pIC50 = 4.19, table 5, entry 37), despite
not possessing an ionized functional group in the aromatic domain and thus
having a distinctly higher clog D value of 1.87 compared to the value of -0.44
calculated for 5s.
Of the rac-SNAP-5114 (rac-2) analogues that have an entire 4-methoxyphenyl
residue replaced by a polar moiety, 5d, possessing an imidazo[1,2-α]pyridine
subunit instead of the aforementioned 4-methoxyphenyl rest, displays a pIC50
of 4.58 (table 5, entry 10). As compared to the potency of rac-2a at mGAT4
(pIC50 = 5.64 ± 0.05) this equates to a distinct, but still moderate reduction of
inhibitory potency of about one log unit. However, the introduction of a small,
polar functional group in place of one of the 4-methoxyphenyl residues is
associated with
a drastic loss of activity at mGAT4. For example, the
compound featuring a carboxylic acid function in this position, 5k, is devoid of
almost any inhibitory activity at mGAT4, reducing the [3H]GABA uptake to only
96.5 % at 100 µM (table 5, entry 24). By comparison, 5e, which also features a
carboxyl group, albeit located on one of the aromats of the lipophilic domain of
rac-2 supplanting a 4-methoxy group, conserves significantly more of the
inhibitory potency of the parent compound [pIC50(mGAT4) = 4.86, Table 5,
entry 12]. The same phenomenon is observed for 5l, that formally results from
the replacement of one of the 4-methoxyphenyl groups of rac-2 with a hydroxy
moiety, in combination with a simultaneous exchange of the ether oxygen of
the spacer by a methylene group to warrant chemical stability. The compound
reduces the [3H]GABA uptake to 92.0% at 100 µM (table 5, entry 26).
Compound 5m on the other hand, which also contains a hydroxy function, but
retains triaryl pattern of the parent compound by substituting one of the three
methoxy groups with a hydroxymethylene moiety, still possesses a pIC50 of
4.90 (table 5, entry 28). These results suggest that decreasing size and steric
demand of the lipophilic domain is accompanied by a pronounced decline in
biological activity at mGAT4. The three aromatic moieties on the quaternary
carbon atom can therefore be regarded as essential for high inhibitory potency
at mGAT4 of these compounds, and variations, e.g. in order to alter the
polarity of a compound, should only be implemented by modification of the
substituents, but not by replacement of one of these aromatic moieties in
favour of a smaller functional group.
In conclusion, the data imply no clear correlation between the polarity and size
of the compound and the inhibitory potency exerted at mGAT4. Still the fact
remains conspicuous that all test compounds with a free carboxylic acid
function of the nipecotic acid residue possessing a pIC50 ≥ 5.0 contain a
functional group that can act as hydrogen bond acceptor, but not as donor,
and that is of similar size as the 4-methoxy moiety in the lipophilic domain it
replaces. Furthermore, the clog D values calculated for these compounds
consistently lie in the range of 2.0 – 2.4, as is the case with the parent
compound rac-2 (clog D = 2.32) and the most potent inhibitor synthesized
within the scope of this study, 5a (clog D = 2.15). This might turn out to be
beneficial with regard to potential therapeutical applications, since compounds
with log D values between 2 and 5,[36] or even better with values closer to 2,[37]
are thought to possess the highest propensity for blood-brain barrier (BBB)
penetration.
Next, we aimed to study how the number of aryl moieties in the lipophilic
domain, which are modified with regard to their substituents, affects the
inhibitory potency. To that end, an array of rac-SNAP-5514 analogues was
synthesized that had one, two, or all three of the methoxy groups replaced by
methoxymethylene moieties. While the introduction of one methoxymethylene
group caused only a minor decrease in inhibitory potency to a pIC50 of 5.42
(5f, table 5, entry 14) as compared to the reference compound rac-2 (pIC50
=
5.64 ± 0.05, table 1, entry 2), the loss of activity became increasingly more
pronounced with each further methoxymethylene group (5h, mGAT4: pIC50
=
4.77, table 5, entry 18; 5i, mGAT4: remaining [3H]GABA at 100µM = 76.1%,
table 5, entry 20). These findings support the notion that the biological system
can tolerate an inapt substituent on one of the aryl groups relatively well as
long as the other two 4-methoxy moieties of the lipophilic domain remain
unchanged and thus are still available for interactions with the target.
However, substitution of further methoxy moieties will result in an exceeding
decline of inhibitory potency at mGAT4.
Since 5a turned out to be the most potent mGAT4 inhibitor of all racemic
compounds tested for this study, its enantiopure isomers (S)-5a and (R)-5a
were synthesized and evaluated for their biological activities as well. As is the
case with the lead substance SNAP-5114, the (S)-enantiomer of 5a was found
to exhibit a more pronounced inhibitory potency at mGAT4 than the (R)-
enantiomer, with the respective pIC50 values amounting to 5.89 ± 0.07 for (S)-
5a (table 5, entry 3) and 4.86 ± 0.03 for (R)-5a (table 5, entry 4). Hence, (S)-5a
constitutes nominally a more potent mGAT4 inhibitor than the benchmark
compound (S)-2 (pIC50 = 5.71 ± 0.07, table 1, entry 3).
Also the nipecotic acid ester derivatives 12 that have been synthesized in this
study were evaluated for their inhibitory potencies at all four GAT subtypes.
With regard to mGAT4 inhibition, most nipecotic acid esters 12 display
distinctly lower potencies at this transporter subtype than their nipecotic acid
analogues 5, signifying the importance of the free nipecotic acid partial
structure for mGAT4 inhibition. Only in case of the nipecotic acids 5j – 5l
which exert very low inhibitory potencies at mGAT4 (pIC50 < 4.0), the inhibitory
potencies of the free acids were lower than that of the corresponding esters
12j – 12l, the differences being, however, marginal. However, carboxylic acid
ester 12k cannot be directly compared to nipecotic acid 5k, as the latter
possesses a carboxy function instead of a methoxy carbonyl moiety attached
to one of the aryl residues of the lipophilic domain.
Next, we aimed to elucidate whether the biological activity of 5a is solely
attributable to competitive inhibition at the target, or whether a reaction of the
aldehyde function with [3H]GABA serving as substrate in the [3H]GABA uptake
assay might have affected the outcome of the biological study. Reaction of
[3H]GABA with the aldehyde function of the test compounds 5a, (S)-5a and
(R)-5a might lead to depletion of the substrate of the uptake assay, thus
falsifying the outcome of the assay. Alternatively, a thus formed intermediate
might be a potent inhibitor by itself. To this end the uptake experiment was
repeated using an assay recently developed by our group,[38] which has GABA
replaced by the chemically inert imidazolacetic acid. Since 5a and its
enantiopure isomers (R)-5a and (S)-5a display similar inhibitory activity when
tested under these conditions as compared to the assay based on [3H]GABA,
it appears reasonable to assume that possible reactions of the aldehyde
Regarding the effects at other GAT subtypes, inhibition of mGAT1 is the
pharmacologically most relevant due to its high abundance in the mammalian
CNS. Of the compounds presented in this study, none of the carboxylic acid
analogues of rac-SNAP-5114 5 show a reasonable inhibitory potency at
mGAT1, i.e. pIC50 values ≥ 4, which is also true for the ester analogues 12
with two exceptions. Thus, for 12d, that has one of the three 4-methoxyphenyl
6
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
groups (of rac-2) substituted by an imidazo[1,2-a]pyridine-7-yl moiety, and
12o, featuring a β-alanine ethyl ester partial structure which is linked to one of
the three aromatic residues in the 4-position by a methylene spacer, pIC50
values of 4.39 and 4.59 were found (12d, table 5, entry 9, and 12o, table 5,
entry 31). All other compounds lie below that mark, i.e. a pIC50 = 4.0, and can
thus be regarded as less potent mGAT1 inhibitors than rac-2, which has a
pIC50 of 4.08 (2, table 1, entry 2). Hence, contrary to what applies to the
potency at mGAT4, transforming the ester function of 12 into a free acid
moiety exerts little to no influence on inhibitory potency at mGAT1, the latter
usually amounting to pIC50 values of below 4.0 for both compound classes.
(12n, table 5, entry 29, pIC50 = 4.29; 12o, table 5, entry 31, pIC50 = 4.72)
display pIC50 values ≥ 4.00 at this transporter subtype.
Serving as lead structure of this study, rac-SNAP-5114 (rac-2) possesses a
comparatively high pIC50 value at mGAT3 of 4.96 (table 1, entry 2); as a
consequence, its desired mGAT4 selectivity is quite poor with regard to this
GAT subtype. The modifications of the rac-2 structure that were undertaken
for the purpose of this study resulted in compounds with lower activity at
mGAT3, 5a being the only exception (pIC50 = 4.98 ± 0.08, table 5, entry 2).
This is of particular importance for those compounds that show relatively high
inhibitory potency at mGAT4 and are thus of interest as pharmacological tools.
In case of 5b and 5f, which comprise an acetyl group and
a
The compounds synthesized and tested for this study were generally found to
be weak binders at mGAT1, which correlates well with the uniformly low
inhibitory potencies at this GABA transporter subtype. In that context, only the
rac-SNAP-5114 analogues possessing a formyl [(5a), (R)-5a], acetyl (5b), or
nitrile function (5g) in the lipophilic domain or in which a 4-methoxyphenyl
residue has been replaced by an imidazo[1,2-a]pyridine-7-yl moiety (5d) or an
hydroxy function (5l) differ by displaying moderate pKi values > 4.0. In detail,
the values determined for these compounds were 4.50 ± 0.06 (5a, table 5,
entry 2), 4.91 ± 0.09 [(R)-5a, table 5, entry 4], 4.51 ± 0.03 (5b, table 5, entry
6), 4.40 ± 0.01 (5g, table 5, entry 16), 4.03 (5d, table 5, entry 10) and 4.49 (5l,
table 5, entry 26), respectively.
methoxymethylene group in the lipophilic domain, the inhibitory potency at
mGAT4 is moderately reduced to pIC50 values of 5.43 ± 0.04 (5b, table 5,
entry 6) and 5.42 ± 0.10 (5f, table 5, entry 14), respectively. However, the
potency loss at mGAT3 caused by these structural modifications is slightly
more pronounced, with a pIC50 value of 4.24 ± 0.02 determined for 5b (table 5,
entry 6) and 4.51 ± 0.04 for 5f (table 5, entry 14). Accordingly, these
compounds display considerably higher mGAT4 selectivity than rac-2, albeit at
the cost of some inhibitory potency.
As compared to the enantiopure benchmark inhibitor (S)-2 (table 1, entry 3),
the most potent compound synthesized for this study, (S)-5a, displays almost
identical subtype selectivity (table 5, entry 3). In particular, the pIC50 values
amount to 5.16 ± 0.05 at mGAT3 [(S)-2: 5.29 ± 0.04] and 4.14 ± 0.00 at
mGAT1 [(S)-2: 4.07 ± 0.09]. At mGAT2 both compounds exert only weak
inhibitory effects, reducing the [3H]GABA uptake to 72.6% [(S)-5a] and 56%
[(S)-2], respectively. Hence, (S)-5a can be considered a viable alternative to
(S)-2 as a pharmacological tool.
Furthermore, none of the free nipecotic acids 5 reduced [3H]GABA below 50%
at a concentration of 100 µM when tested at mGAT2, which corresponds to
pIC50 values < 4.0, except for 5t (pIC50 = 4.18, table 5, entry 41). Nevertheless,
some of the ester precursors 12 exceeded this mark, the most potent among
these being the imidazo[1,2-a]pyridine derivative 12d with a pIC50 of 4.88
(table 5, entry 9). Likewise, the ester derivatives of the compounds that feature
in the lipophilic domain a formyl (12a, table 5, entry 1, pIC50 = 4.57), a
methoxymethyl (12h, table 5, entry 17, pIC50 = 4.30) or an amino acid residue
Table 5. Binding affinities (pKi) and inhibitory potencies (pIC50) of the N-substituted nipecotic acids 6a,c-u and nipecotic acid ethyl esters 12a,c-r,t-u.
[b]
pIC50
pKi[a]
Entry
Compound
R1
R2
R3
R4
X
mGAT1
55.5%
mGAT2
4.57
mGAT3
4.90
mGAT4
55.9%
12a
5a
1
2
3
OMe
OMe
OEt
O
85.4 %
4.41±
0.11
4.17
± 0.08
4.14
4.98
± 0.08
5.16
5.77
± 0.04
5.89
66.3 %
(S)-5a
OMe
OMe
OH
O
77.4 %
72.6 %
± 0.00
± 0.05
± 0.07
4.91±
0.09
4.45
± 0.08
4.47
± 0.05
4.86
± 0.03
(R)-5a
12b
4
5
6
50.1 %
66.4 %
64.2 %
OEt
OH
84.6 %
72.9 %
64.6 %
55.2 %
84.2 %
OMe
OMe
OMe
OMe
OMe
OMe
O
O
O
4.51±
0.03
4.24
± 0.02
5.43
± 0.04
5b
12c
7
OEt
93.8 %
75.7 %
73.2 %
95.9 %
94.3 %
5c
12d
5d
8
9
OH
OEt
OH
62.1 %
77.2 %
4.03
67.9 %
4.39
78.0 %
4.88
77.1 %
4.71
82.0 %
4.24
10
84.9 %
72.9 %
81.8 %
4.58
12e
5e
11
12
OMe
OMe
OMe
OMe
OEt
OH
O
O
74.8 %
71.4 %
84.6 %
96.5 %
87.3 %
85.2 %
79.0 %
4.24
76.1%
4.86
12f
5f
13
14
15
OEt
OH
OEt
55.3 %
89.4 %
97.6 %
66.6 %
67.1 %
76.6 %
58.8 %
72.7 %
80.7 %
56.9 %
50.8 %
OMe
OMe
OMe
OMe
O
O
4.51
± 0.08
5.42
± 0.10
12g
63.6 %
71.9 %
7
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
4.40±
0.10
4.31
± 0.12
5.07
± 0.12
5g
16
OH
74.8 %
82.9 %
12h
5h
12i
5i
17
18
19
20
OEt
OH
OEt
OH
72.9 %
74.1 %
87.5 %
92.7 %
64.9 %
70.5 %
70.3 %
90.7 %
4.30
50.6 %
52.2 %
4.02
4.05
4.77
OMe
CH2OMe
OMe
55.2 %
65.0 %
72.6 %
CH2OMe
O
51.4 %
76.1 %
68.3 %
12j
5j
21
22
OEt
OH
89.3 %
87.0 %
57.8 %
79.6 %
56.1 %
96.1 %
4.10
60.0 %
65.2 %
OMe
OMe
O
O
82.1 %
12k
5k
23
24
-COOMe
-COOH
OEt
OH
90.8 %
96.9 %
89.1 %
100 %
92.6 %
101 %
64.1 %
94.0 %
78.2 %
96.5 %
OMe
12l
5l
25
26
OEt
OH
104.0%
4.49
79.3 %
69.5 %
70.4 %
62.5 %
65.0 %
80.7 %
73.8 %
92.0 %
-OH
OMe
OMe
OMe
OMe
CH2
O
12m
5m
27
28
OEt
OH
75.3 %
55.0 %
50.4 %
78.1 %
49.3 %
91.6 %
4.45
4.23
4.90
58.7 %
12n
5n
29
30
31
32
33
34
35
36
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OEt
OH
OEt
OH
OEt
OH
OEt
OH
O
O
O
O
O
O
O
91.7 %
86.2 %
72.4 %
82.3 %
95.8 %
58.8 %
90.9 %
77.0 %
51.8 %
81.8 %
4.59
4.29
4.47
4.18
72.2 %
4.66
96.1 %
4.72
90.9 %
4.63
12o
5o
94.4 %
75.5 %
78.2 %
66.9 %
80.7 %
69.5 %
103 %
76.0 %
67.4 %
77.0 %
89.1 %
58.8%
69.1 %
49.3 %
61.1 %
56.3 %
66.1 %
4.88
12p
5p
HN
HO
O
12q
5q
57.4 %
4.34
OMe
OMe
OMe
OMe
O
O
5r
37
38
OH
97.0 %
84.9 %
76.0 %
52.4 %
79.7 %
76.7 %
64.2 %
59.3 %
4.19
12s
OEt
67.6 %
OMe
OMe
OMe
OMe
O
O
5s
39
40
OH
86.5 %
101.6%
48.5 %
52.3 %
51.5 %
73.5 %
4.33
4.13
12t
OEt
49.7 %
63.9 %
5t
41
OH
54.1 %
99,5 %
4,18
96,0 %
50,0 %
[a] Results of the MS Binding Assays are given as pKi ± SEM. For compounds with low pKi values only one measurement was performed, therefore no SEM can
be reported. Percent values represent remaining specific NO711 binding in presence of 100 µM compound. [b] Results of the [3H]GABA uptake assays are given
as pIC50 ± SEM. For compounds with low pIC50 values only one measurement was performed, therefore no SEM can be reported. Percent values represent
remaining [3H]GABA uptake in presence of 100 µM compound.
between the inhibitor and polar regions of binding site, and hence improve
binding affinity, inhibitory potency, and selectivity.
Conclusion
The test compounds were accessible through conversion of tertiary alcohols
featuring the respective functional groups or appropriate precursors into the
corresponding chlorides, followed by etherification with ethyl N-(2-
hydroxyethyl)nipecotinate. The obtained N-substituted nipecotic acid esters
In order to gain insight into the structure activity relationship of mGAT4
inhibitors, analogues of rac-SNAP-5114 were synthesized that differ from the
parent compound by having one or more of the 4-methoxy groups attached to
the lipophilic domain, or a complete 4-methoxyphenyl group, replaced by a
were either directly hydrolized at this stage, or functional group
moiety with higher polarity. These modifications might increase interactions
8
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
interconversions were performed beforehand, including the introduction of
various amino acid partial structures via reductive amination.
(K2CO3, 2.5 eq) were added. After stirring 16 hours at room temperature, the
mixture was filtered and reduced in vacuum.
General procedure for the hydrolysis of the ethyl ester function (GP4): The
ester (1.0 eq) was dissolved in MeOH. Bidest. water and barium hydroxide
octahydrate (2.0-4.0 eq) were added and the mixture was stirred at room
temperature until TLC indicated complete consumption of the ester. Carbon
dioxide was passed through the solution until no further precipitate formed.
The suspension was diluted with MeOH (1:1), filtered through a paper filter
and reduced in vacuum. If necessary, the crude acid was purified by flash
column chromatography. The solid residue was solved in MeOH (1.0 ml),
filtered through a syringe filter (Perfect-Flow®, WICOM Germany GmbH,
PTFE, 0.2µM), diluted with bidest. water (4.0 ml) and lyophilized.
The biological evaluation of the obtained N-substituted nipecotic acids
revealed no direct correlation between the polarity of the newly introduced
group and the biological activity of the resulting compound. However,
compounds with clog D values of 2.0 – 2.4 consistently exert the highest
inhibitory potencies. In this context, we found (S)-1-{2-[(4-formylphenyl)bis(4-
methoxyphenyl)methoxy]-ethyl}piperidine-3-carboxylic acid [(S)-5a], that has
one of the three methoxy groups of the parent substance substituted by a
formyl moiety, to be the most potent test compound synthesized in this study.
Being characterized by a pIC50 value of 5.89 ± 0.07 at mGAT4, it even exhibits
slightly higher inhibitory potency than the benchmark substance (S)-SNAP-
5114, while showing comparable subtype selectivity. Furthermore, the racemic
SNAP-5114 analogues comprising in the lipophilic domain an acetyl and an
methoxymethylene substituent, respectively, were found to be slightly less
potent mGAT4 inhibitors than the parent compound, but somewhat more
subtype selective, especially at mGAT3.
General procedure for reductive amination I (GP5): The appropriate amino
acid ester hydrochloride (2.0 eq) and sodium triacetoxyborohydride (1.4 eq)
were added to a solution of aldehyde 3a (1.0 eq) in CH2Cl2. The reaction
mixture was stirred at room temperature until TLC indicated complete
consumption of the aldehyde, quenched with saturated sodium carbonate
(Na2CO3) solution and extracted thrice with CH2Cl2. The combined organic
phases were dried over MgSO4, filtered and reduced in vacuum.
Also, it has been demonstrated that rac-SNAP-5114 analogues that have one
of the aromatic moieties in the lipophilic domain replaced by a small, non-
aromatic moiety show a distinct deprivation of inhibitory potency. This finding
indicates the essentiality of the triaryl structure in the lipophilic domain for the
biological activity of mGAT4 inhibitors. Finally, we determined how the
substitution of several methoxy groups in the lipophilic domain of rac-SNAP-
5114 impacts the inhibitory potency of the resulting compounds using the
example of the methoxymethylene group. As the results of the biological
evaluation suggest, mGAT4 can tolerate one inapt substituent reasonably well.
However, the replacement of additional methoxy groups will result in an
exceeding decline of inhibitory potency.
General procedure for reductive amination II (GP6): The appropriate amino
acid (2.0 eq) and sodium cyanoborohydride (1.4 eq) were added to a solution
of aldehyde 3a (1.0 eq) in MeOH. The reaction mixture was stirred at room
temperature until TLC indicated complete consumption of the aldehyde and
reduced in vacuum.
rac-1-{2-[(4-Formylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}piper-
idine-3-carboxylic acid (5a): GP4 was followed using 12a (101 mg, 0.190
mmol), barium hydroxide octahydrate (241 mg, 0.764 mmol), MeOH/H2O 4:1
(5.0 ml). The crude compound was purified by flash column chromatography
on silica (eluent MeOH). Amorphous colourless solid (85 mg, 89 %).
These findings are expected to be beneficial for future developments of more
potent mGAT4 inhibitors since they reveal essential structure activity
relationships with regard to size, polarity, and electronic effects of the
substituents in the lipophilic domain.
(S)-1-{2-[(4-Formylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}piper-
idine-3-carboxylic acid [(S)-5a]: GP4 was followed using (S)-12a (38 mg,
0.071 mmol), barium hydroxide octahydrate (90 mg, 0.28 mmol), MeOH/H2O
4:1 (5.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH). Amorphous colourless solid (34 mg,
93 %).
Experimental Section
Chemistry
(R-1-{2-[(4-Formylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}piperidine-
3-carboxylic acid [(R)-5a]: GP4 was followed using (R)-12a (37 mg, 0.070
mmol), barium hydroxide octahydrate (89 mg, 0.28 mmol), MeOH/H2O 4:1 (5.0
ml). The crude compound was purified by flash column chromatography on
silica (eluent MeOH). Amorphous colourless solid (32.5 mg, 92 %).
Moisture-sensitive reactions were carried out in oven-dried glassware under
inert gas atmosphere. Commercially available starting materials were used
without further purification. Dry acetonitrile (MeCN) was purchased from VWR
(HiPerSolv Chromanonorm, water content > 30ppm) and tetrahydrofuran
(THF) was freshly distilled from sodium benzophenone ketyl. All other solvents
were distilled prior to use. Microwave reactions were carried out with CEM
Discover® SP Microwave Synthesizer (model no. 909 155). Flash column
chromatography was performed according to Still et al.[39] using Merck silica
rac-1-{2-[(4-Acetylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}piperidine-
3-carboxylic acid (5b): GP4 was followed using 12b (87 mg, 0.16 mmol),
barium hydroxide octahydrate (197 mg, 0.62 mmol), MeOH/H2O 4:1 (3.5 ml).
The crude compound was purified by flash column chromatography on silica
(eluent MeOH). Amorphous colourless solid (69 mg, 86 %).
gel 60 (mesh 0.040
- 0.063 mm) as stationary phase. Thin-layer
chromatography (TLC) was carried out on Merck silica gel 60 F254 sheets. 1H
and 13C NMR spectra were, unless stated otherwise, recorded at room
temperature with JNMR-GX (JEOL 400 or 500 MHz) or Bruker BioSpin
Avance III HD (400 or 500 MHz) and integrated with the NMR software
MestReNova. IR samples were measured as KBr pellets or film with Perkin-
Elmer FT-IR 1600. HRMS data were obtained with JMS-GCmate II (EI, Jeol)
or Thermo Finnigan LTQ FT Ultra (ESI, Thermo Finnigan).
rac-1-(2-{[4-(N,N-dimethylsulfamoyl)phenyl]bis[4-methoxyphenyl]-
methoxy}ethyl)piperidine-3-carboxylic acid (5c): GP4 was followed using
12c (66 mg, 0.11 mmol), barium hydroxide octahydrate (137 mg, 0.434 mmol),
MeOH/H2O 4:1 (3.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH). Amorphous colourless solid (55 mg,
88 %).
rac-1-(2-{Imidazo[1,2-a]pyridin-6-ylbis(4-methoxyphenyl)methoxy}-
ethyl)piperidine-3-carboxylic acid (5d): GP4 was followed using 12d (78
mg, 0.14 mmol), barium hydroxide octahydrate (170 mg, 0.54 mmol),
MeOH/H2O 4:1 (5.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH/CH2Cl2 1:1). Amorphous colourless
solid (59 mg, 80 %).
General procedure for the synthesis of trityl alcohols by Grignard reaction
(GP1): Magnesium turnings were scraped with a glass rod and suspended in
dry THF. The appropriate aryl halide or a solution thereof in THF was added
portion wise. After completion of the Grignard reagent formation a solution of
the electrophile in THF was added dropwise to the solution. The mixture was
stirred for the prescribed time and temperature, quenched with saturated
ammonium chloride (NH4Cl) solution, diluted with water and extracted thrice
with dichloromethane (CH2Cl2). The combined organic phases were dried over
magnesium sulfate (MgSO4), filtered and reduced in vacuum.
rac-1-{2-[(4-Carboxyphenyl)bis(4-methoxyphenyl)methoxy]ethyl}piper-
idine-3-carboxylic acid (5e): GP4 was followed using 12e (200 mg, 0.350
mmol), barium hydroxide octahydrate (222 mg, 0.704 mmol), MeOH/H2O 4:1
(5.0 ml). The crude compound was purified by flash column chromatography
on silica (eluent MeOH). Amorphous colourless solid (110 mg, 61 %).
General procedure for the synthesis of trityl alcohols by lithiation of an aryl
halide and subsequent reaction with 4,4´-dimethoxybenzophenone (GP2): tert-
Butyllithium solution (1.7M in pentane, 2.0 eq.) was added dropwise to a
solution of the aryl halide (1.0 eq.) in THF at -78°C. After 2h a solution of 4,4´-
dimethoxybenzophenone (1.0 eq) in THF was added. The mixture was
allowed to slowly warm up to room temperature overnight, quenched with
water and extracted thrice with CH2Cl2. The combined organic phases were
dried over MgSO4, filtered and reduced in vacuum.
rac-1-(2-{[4-(Methoxymethyl)phenyl]bis[4-methoxyphenyl]methoxy}-
ethyl)piperidine-3-carboxylic acid (5f): GP4 was followed using 12f (72 mg,
0.13 mmol), barium hydroxide octahydrate (83 mg, 0.26 mmol), MeOH/H2O
4:1 (12.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH). Amorphous colourless solid (56 mg,
81 %).
General procedure for the etherification of tertiary alcohols (GP3): The tertiary
alcohol (1.0 eq) was charged with a catalytic amount of DMF. Acetyl chloride
was added and the reaction mixture was stirred at room temperature for 24
hours, reduced in vacuum and dried under high vacuum. The oily or solid
residue was solved in dry MeCN. Racemic, (R)- or (S)-ethyl 1-
(hydroxyalkyl)nipecotinate (1.1 eq) and oven-dried potassium carbonate
rac-1-{2-[(4-Cyanophenyl)bis(4-methoxyphenyl)methoxy]ethyl}piperidine-
3-carboxylic acid (5g): GP4 was followed using 12g (48 mg, 0.090 mmol),
barium hydroxide octahydrate (57 mg, 0.18 mmol), MeOH/H2O 4:1 (2.0 ml).
The crude compound was purified by flash column chromatography on silica
(eluent MeOH). Amorphous colourless solid (39 mg, 86 %).
9
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
rac-1-(2-{Bis[4-(methoxymethyl)phenyl][4-methoxyphenyl]methoxy}-
ethyl)piperidine-3-carboxylic acid (5h): GP4 was followed using 12h (56
mg, 0.10 mmol), barium hydroxide octahydrate (127 mg, 0.403 mmol),
MeOH/H2O 4:1 (5.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH). Amorphous colourless solid (51 mg,
94 %).
rac-1-{2-[(4-{[(4-Carboxyphenyl)amino]methyl}phenyl)bis(4-methoxy-
phenyl)methoxy]ethyl}piperidine-3-carboxylic acid (5t): 12t (37.2 mg,
0.057 mmol) was dissolved in THF (1.5 ml) and H2O (1.5 ml) and barium
hydroxide octahydrate (38.4 mg, 4.0 eq) were added. The mixture was stirred
at room temperature until TLC indicated complete consumption of 12t. Carbon
dioxide was passed through the solution until no further precipitate formed.
The suspension was filtered through a syringe filter (Perfect-Flow®, WICOM
Germany GmbH, PTFE, 0.2µM) and lyophilized. Amorphous colourless solid
(34.0 mg, 96 %).
rac-1-(2-{Tris[4-(methoxymethyl)phenyl]methoxy}ethyl)piperidine-3-
carboxylic acid (5i): GP4 was followed using 12i (61 mg, 0.11 mmol), barium
hydroxide octahydrate (75 mg, 0.44 mmol), MeOH/H2O 4:1 (5.0 ml). The crude
compound was purified by flash column chromatography on silica (eluent
MeOH). Amorphous colourless solid (52 mg, 90 %).
[4-(Dimethoxymethyl)phenyl]bis(4-methoxyphenyl)methanol (8a): GP2
was followed using tert-butyllithium solution (1.7M in pentane, 1.2 ml, 2.04
mmol), 1-bromo-4-(dimethoxymethyl)benzene 9a (235 mg, 1.02 mmol), THF
(3.0 ml). Addition of 4,4´-dimethoxybenzophenone 10a (245 mg, 1.00 mmol) in
THF (5.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent pentane/Et2O 7:3). Colourless oil (393 mg,
99 %).
rac-1-{2-[(4-Carbamoylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}-
piperidine-3-carboxylic acid (5j): GP4 was followed using 12j (63 mg, 0.12
mmol), barium hydroxide octahydrate (146 mg, 0.46 mmol), MeOH/H2O 4:1
(5.0 ml). The crude compound was purified by flash column chromatography
on silica (eluent MeOH). Amorphous colourless solid (52 mg, 87 %).
Bis(4-methoxyphenyl)[4-(2-methyl-1,3-dioxan-2-yl)phenyl]methanol (8b):
GP2 was followed using tert-butyllithium solution (1.7M in pentane, 3.3 ml, 5.6
mmol), 2-(4-bromophenyl)-2-methyl-1,3-dioxane 9b (713 mg, 2.77 mmol), THF
(12.0 ml). Addition of 4,4´-dimethoxybenzophenone 10a (700 mg, 2.83 mmol)
in THF (12.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent pentane/Et2O 8:2). Colourless semi-solid
(1.02 g, 87 %).
rac-1-{2-[Carboxybis(4-methoxyphenyl)methoxy]ethyl}piperidine-3-
carboxylic acid (5k): GP4 was followed using 12k (370 mg, 0.760 mmol),
barium hydroxide octahydrate (481 mg, 1.52 mmol), MeOH/H2O 4:1 (5.0 ml).
The crude compound was purified by flash column chromatography on silica
(eluent MeOH). Amorphous colourless solid (327 mg, 97 %).
rac-1-[4-Hydroxy-4,4-bis(4-methoxyphenyl)butyl]piperidine-3-carboxylic
acid (5l): GP4 was followed using 12l (103 mg, 0.230 mmol), barium
hydroxide octahydrate (148 mg, 0.469 mmol), MeOH/H2O 4:1 (3.6 ml). The
crude compound was purified by flash column chromatography on silica
(eluent MeOH). Amorphous colourless solid (81 mg, 84 %).
4-[Hydroxybis(4-methoxyphenyl)methyl]-N,N-dimethylbenzenesulfon-
amide (8c): GP2 was followed using tert-butyllithium solution (1.7M in
pentane, 2.7 ml, 4.6 mmol), 9c (561 mg, 2.27 mmol), THF (30.0 ml). Addition
of 4,4´- dimethoxybenzophenone 10a (561 mg, 2.27 mmol) in THF (12.0 ml).
The crude compound was precipitated from isohexane/THF. Amorphous
colourless solid (514 mg, 53 %).
rac-1-(2-{[4-(Hydroxymethyl)phenyl]bis[4-methoxyphenyl]methoxy}-
ethyl)piperidine-3-carboxylic acid (5m): GP4 was followed using 12m (58
mg, 0.11 mmol), barium hydroxide octahydrate (69 mg, 0.22 mmol),
MeOH/H2O 4:1 (3.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH). Amorphous colourless solid (38 mg,
69 %).
Imidazo[1,2-a]pyridin-6-ylbis(4-methoxyphenyl)methanol (8d): N-butyl-
lithium solution (2.4 M in hexane, 1.0 ml, 2.4 mmol) was added dropwise to a
suspension of 6-bromoimidazo[1,2-a]pyridine 9d (483 mmg, 2.40 mmol) in
Et2O (24.0 ml) at -78°C. After 30min
a
suspension of 4,4´-
dimethoxybenzophenone 10a (494 mg, 2.0 mmol) in Et2O was added. The
mixture was stirred 1h at -78°C, warmed to room temperature, quenched with
water (20.0 ml) and extracted thrice with ethyl acetate (20.0 ml). The
combined organic phases were washed with saturated Na2CO3 solution, dried
over MgSO4, filtered and reduced in vacuum. The crude compound was
purified by flash column chromatography on silica (eluent CH2Cl2 + 2%
MeOH). Amorphous slightly yellowish solid (400 mg, 56 %).
rac-1-{2-[(4-{[(Carboxymethyl)amino]methyl}phenyl)bis(4-methoxy-
phenyl)methoxy]ethyl}piperidine-3-carboxylic acid (5n): GP4 was followed
using 12n (35 mg, 0.060 mmol), barium hydroxide octahydrate (37 mg, 0.12
mmol), MeOH/H2O 4:1 (5.0 ml). The crude compound was purified by flash
column chromatography on silica (eluent MeOH). Amorphous colourless solid
(29 mg, 65 %).
rac-1-{2-[(4-{[(2-Carboxyethyl)amino]methyl}phenyl)bis(4-methoxy-
phenyl)methoxy]ethyl}piperidine-3-carboxylic acid (5o): GP4 was followed
using 12o (83 mg, 0.13 mmol), barium hydroxide octahydrate (171 mg, 0.542
mmol), MeOH/H2O 4:1 (5.0 ml). The crude compound was purified by flash
column chromatography on silica (eluent MeOH). Amorphous colourless solid
(57 mg, 74 %).
Ethyl 4-[hydroxybis(4-methoxyphenyl)methyl]benzoate (8e): Isopropyl-
magnesium chloride (2.0M solution in THF, 1.0 ml, 2.0 mmol) was added
dropwise to a solution of ethyl 4-iodobenzoate 9e (547 mg, 2.02 mmol) in THF
(5.0 ml) at -20°C. The mixture was stirred for 1.5h and a solution of 4,4´-
dimethoxybenzophenone 10a (445 mg, 1.80 mmol) in THF (5.0 ml) was slowly
added via syringe. Stirring was continued at 0°C over night. The reaction
mixture was allowed to slowly warm up to room temperature, quenched with
saturated NH4Cl solution and extracted thrice with CH2Cl2. The combined
organic phases were dried over MgSO4, filtered and reduced in vacuum. The
resulting crude compound was purified by flash column chromatography on
silica (eluent pentane/Et2O 8:2). Colourless semi-solid (328 mg, 77 %).
rac-1-{2-[(4-{[(1-Carboxycyclopropyl)amino]methyl}phenyl)bis(4-
methoxyphenyl)methoxy]ethyl}piperidine-3-carboxylic acid (5p): GP4
was followed using 12p (51 mg, 0.079 mmol), barium hydroxide octahydrate
(100 mg, 0.317 mmol), MeOH/H2O 4:1 (6.0 ml). The crude compound was
purified by flash column chromatography on silica (eluent MeOH). Amorphous
colourless solid (50 mg, 82 %).
[4-(Methoxymethyl)phenyl]bis(4-methoxyphenyl)methanol (8f): GP2 was
followed using tert-butyllithium solution (1.7M in pentane, 5.0 ml, 8.50 mmol),
1-bromo-4-(methoxymethyl)benzene 9f (848 mg, 4.22 mmol), THF (24.0 ml).
Addition of 4,4´-dimethoxybenzophenone 10a (1.05 g, 4.25 mmol) in THF
(12.0 ml). The crude compound was purified by flash column chromatography
on silica (eluent pentane/Et2O 7:3). Amorphous colourless solid (1.42 g, 92 %).
rac-1-{2-[(4-{[(2-Carboxypropan-2-yl)amino]methyl}phenyl)bis(4-
methoxyphenyl)methoxy]ethyl}piperidine-3-carboxylic acid (5q): GP4
was followed using 12q (35 mg, 0.054 mmol), barium hydroxide octahydrate
(136 mg, 0.431 mmol), MeOH/H2O 4:1 (5.0 ml). The crude compound was
purified by flash column chromatography on silica (eluent MeOH). Amorphous
colourless solid (29.3 mg, 92 %).
4-[Hydroxybis(4-methoxyphenyl)methyl]benzonitrile (8g): GP1 was
followed using a Grignard reagent prepared from 4-bromoanisol 9g (1.72 g,
9.01 mmol) and magnesium (219 mg, 9.01 mmol) in THF (20.0 ml). A solution
of 4-cyanobenzoyl chloride 10c (736 mg, 4.40 mmol) in THF (8.0 ml) was
added dropwise at 0°C. The reaction mixture was stirred 2h at 0°C and 30min
at 25°C. The crude compound was purified by flash column chromatography
on silica (eluent pentane/Et2O 6:4). Amorphous colourless solid (1.42 g, 94 %).
rac-1-[2-(Bis{4-methoxyphenyl}{4-[(2-oxopyrrolidin-1-yl)methyl]phenyl}-
methoxy)ethyl]piperidine-3-carboxylic acid (5r): GP5 was followed using
12a (173 mg, 0.330 mmol), ethyl 4-aminobutanoate hydrochloride (109 mg,
0.650 mmol), sodium triacetoxyborohydride (155 mg, 0.730 mmol), CH2Cl2
(2.0 ml). The crude compound was purified by flash column chromatography
on silica (eluent ethyl acetate + 5% triethylamine), yielding 107 mg of a
mixture
of
rac-ethyl
1-[2-(bis{4-methoxyphenyl}{4-[(2-oxopyrrolidin-1-
Bis[4-(methoxymethyl)phenyl](4-methoxyphenyl)methanol (8h): GP1 was
yl)methyl]phenyl}-methoxy)ethyl]piperidine-3-carboxylate and rac-ethyl 1-{2-
[(4-{[(4-methoxy-4-oxobutyl)amino]methyl}phenyl)bis(4-
followed using
a
Grignard reagent prepared from 1-bromo-4-
(methoxymethyl)benzene 9f (745 mg, 3.70 mmol) and magnesium (90 mg, 3.7
mmol) in THF (6.0 ml). A solution of methyl 4-methoxybenzoate (560 mg, 3.30
mmol) in THF (6.0 ml) was added dropwise. The reaction mixture was stirred
at room temperature for 16h and refluxed 90min. The crude compound was
purified by flash column chromatography on silica (eluent pentane/Et2O 6:4).
Pale yellowish oil (492 mg, 70 %).
methoxyphenyl)methoxy]ethyl}piperidine-3-carboxylate. 84 mg thereof were
subjected to GP4 using barium hydroxide octahydrate (178 mg, 0.564 mmol),
MeOH/H2O 4:1 (3.5 ml). The crude compound was purified by flash column
chromatography on silica (eluent MeOH). Amorphous colourless solid (69 mg,
47 % over both steps).
rac-1-{2-[(4-{[(3-Carboxypropyl)amino]methyl}phenyl)bis(4-methoxy-
phenyl)methoxy]ethyl}piperidine-3-carboxylic acid (5s): GP4 was followed
using 12s (50 mg, 0.081 mmol), barium hydroxide octahydrate (103 mg, 0.326
mmol), MeOH/H2O 4:1 (5.0 ml). Amorphous colourless solid (40 mg, 84 %).
Tris[4-(methoxymethyl)phenyl]methanol (8i): GP1 was followed using a
Grignard reagent prepared from 1-bromo-4-(methoxymethyl)benzene 9f (1.41
g, 7.00 mmol) and magnesium (170 mg, 6.99 mmol) in THF (10.0 ml).
Dimethyl carbonate (209 mg, 2.30 mmol) was added dropwise at room
temperature. The reaction mixture was refluxed 90 min, then stirred at room
10
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
temperature over night. The crude compound was purified by flash column
chromatography on silica (eluent pentane/Et2O 1:1). Colourless oil (771 mg,
85 %).
rac-Ethyl 1-(2-{imidazo[1,2-a]pyridin-6-ylbis(4-methoxyphenyl)methoxy}-
ethyl)piperidine-3-carboxylate (12d): GP5 was followed using 8d (254 mg,
0.710 mmol), acetylchloride (3.0 ml), rac-ethyl 1-(2-hydroxyethyl)nipecotinate
(157 mg, 0.780 mmol), K2CO3 (488 mg, 3.53 mmol), MeCN (5.0 ml). The
crude compound was purified by flash column chromatography on silica
(eluent Et2O/MeOH 9:1). Yellow semi-solid (160 mg, 48 %).
4-[Hydroxybis(4-methoxyphenyl)methyl]benzamide (8j): A mixture of 4-
[hydroxybis(4-methoxyphenyl)methyl]benzonitrile 8g (171 mg, 0.500 mmol)
and potassium tert-butoxide (340 mg, 3.00 mmol) was stirred in tert-butanol
(4.0 ml) at 50°C for 40h. Water (25 ml) was added ad the mixture was
extracted thrice with CH2Cl2 (30.0 ml). The combined organic phases were
dried over MgSO4 and reduced in vacuum. The crude compound was purified
flash column chromatography on silica (eluent isohexane/ethyl acetate 3:7).
Amorphous colourless solid (173 mg, 96 %).
rac-Ethyl
1-(2-{[4-(ethoxycarbonyl)phenyl]bis[4-methoxyphenyl]-
methoxy}ethyl)piperidine-3-carboxylate (12e): GP3 was followed using 8e
(217 mg, 0.550 mmol), acetylchloride (1.0 ml), rac-ethyl 1-(2-
hydroxyethyl)nipecotinate (133 mg, 0.661 mmol), K2CO3 (191 mg, 1.38 mmol),
MeCN (2.0 ml). The crude compound was purified by two consecutive flash
column chromatographies on silica (eluent pentane/Et2O 4:6; pentane/Et2O
8:2 + 5% dimethylethylamine). Colourless semi-solid (173 mg, 54 %).
Methyl 2-hydroxy-2,2-bis(4-methoxyphenyl)acetate (8k): A mixture of 4,4´-
dimethoxybenzilic acid (2.78 g, 9.64 mmol) and 1,8-diazabicyclo[5.4.0]undec-
7-ene (DBU, 2.65 g, 17.4 mmol) in MeCN (4.5 ml) was cooled to 0°C and
iodomethane (5.02 g, 35.0 mmol) was added. The ice bath was removed and
stirring was continued for 48h at ambient temperature. The reaction mixture
was reduced in vacuum and purified by flash column chromatography on silica
(eluent pentane/Et2O 2:1 to 1:1). Amorphous colourless solid (2.64 g, 91 %).
rac-
Ethyl
1-(2-{[4-(methoxymethyl)phenyl]bis[4-methoxyphenyl]-
methoxy}ethyl)piperidine-3-carboxylate (12f): GP3 was followed using 8f
(310 mg, 0.850 mmol), acetylchloride (5.0 ml), rac-ethyl 1-(2-
hydroxyethyl)nipecotinate (191 mg, 0.949 mmol), K2CO3 (294 mg, 2.13 mmol),
MeCN (5.0 ml). The crude compound was purified by two consecutive flash
column chromatographies on silica (eluent Et2O 100%, pentane/Et2O 7:3 + 5%
triethylamine). Colourless oil (207 mg, 44 %).
4-Chloro-1,1-bis(4-methoxyphenyl)butan-1-ol (8l): GP1 was followed using
a Grignard reagent prepared from 4-bromoanisol (0.94 g, 4.98 mmol) and
magnesium (122 mg, 5.02 mmol) in THF (12.0 ml). A solution of 4-chloro-1-(4-
methoxyphenyl)butan-1-one (1.06 g, 5.00 mmol) in THF (12.0 ml) was added
dropwise at 0°C. The reaction mixture was stirred 1h at ambient temperature.
The crude compound was purified by flash column chromatography on silica
(eluent isohexane/ethyl acetate 9:1). Colourless oil (1.06 g, 66 %).
rac-Ethyl
1-{2-[(4-cyanophenyl)bis(4-methoxyphenyl)methoxy]ethyl}-
piperidine-3-carboxylate (12g): GP3 was followed using 8g (364 mg, 1.00
mmol), acetylchloride (1.0 ml), rac-ethyl 1-(2-hydroxyethyl)nipecotinate (242
mg, 1.20 mmol), K2CO3 (346 mg, 2.50 mmol), MeCN (5.0 ml). The crude
compound was purified by flash column chromatography on silica (eluent
pentane/Et2O/triethyl amine 4:4:2). Yellowish semi-solid (447 mg, 85 %).
4-Bromo-N,N-dimethylbenzenesulfonamide (9d):
A
mixture of 4-
rac-Ethyl
1-(2-{bis[4-(methoxymethyl)phenyl][4-methoxyphenyl]-
bromobenzenesulfonamide (2.40 g, 10.2 mmol), tetrabutylammonium
tetrafluoroborate (329 mg, 1.02 mmol), K2CO3 (5.58 g, 40.4 mmol) and
dimethyl sulfate (2.53 g, 20.0 mmol) in MeCN (15.0 ml) was refluxed for 3h.
After cooling to room temperature concentrated ammonium hydroxide solution
(5.0 ml) was added and stirring was continued for 16h. The reaction mixture
was poured into water (20.0 ml) and extracted thrice with CH2Cl2 (30.0 ml).
The combined organic phases were dried over MgSO4, filtered and reduced in
vacuum. The crude compound was purified by flash column chromatography
on silica (eluent pentane/Et2O 7:3). Amorphous colourless solid (2.66 g, 99 %).
methoxy}ethyl)piperidine-3-carboxylate (12h): GP3 was followed using 8h
(380 mg, 1.00 mmol), acetylchloride (3.0 ml), rac-ethyl 1-(2-
hydroxyethyl)nipecotinate (221 mg, 1.10 mmol), K2CO3 (346 mg, 2.5 mmol),
MeCN (5.0 ml). The crude compound was purified flash column
chromatography on silica (eluent Et2O). Colourless oil (260 mg, 46 %).
rac-Ethyl 1-(2-{tris[4-(methoxymethyl)phenyl]methoxy}ethyl)piperidine-3-
carboxylate (12i): GP3 was followed using 8i (243 mg, 0.620 mmol),
acetylchloride (3.0 ml), rac ethyl 1-(2-hydroxyethyl)nipecotinate (452 mg, 2.25
mmol), K2CO3 (774 mg, 5.60 mmol), MeCN (5.0 mL). The crude compound
was purified by flash column chromatography on silica (eluent isohexane/Et2O
8:1 to Et2O 100%). Colourless oil (151 mg, 42 %).
1-{4-[Hydroxybis(4-methoxyphenyl)methyl]phenyl}ethan-1-one
(11):
Concentrated hydrochlorid acid (3.0 ml) was added to a solution of 8b (430
mg, 1.02 mmol) in THF (12.0 ml). The mixture was refluxed overnight, basified
with 2M NaOH and extracted thrice with CH2Cl2. The combined organic
phases were dried over MgSO4, filtered and reduced in vacuum. The crude
compound was purified by flash column chromatography on silica (eluent
pentane/Et2O 1:1). Amorphous colourless solid (345 mg, 93 %).
rac-Ethyl
1-{2-[(4-carbamoylphenyl)bis(4-methoxyphenyl)methoxy]-
ethyl}piperidine-3-carboxylate (12j): GP3 was followed using 8j (512 mg,
1.41 mmol), acetylchloride (5.0 ml), rac-ethyl 1-(2-hydroxyethyl)nipecotinate
(321 mg, 1.55 mmol), K2CO3 (488 mg, 3.53 mmol), MeCN (5.0 ml). The crude
compound was purified flash column chromatography on silica (eluent ethyl
acetate + 5% MeOH). Amorphous slightly yellowish solid (375 mg, 49 %).
rac-Ethyl
1-{2-[(4-formylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}-
piperidine-3-carboxylate (rac-12a): GP3 was followed using 8a (2.68 g, 6.35
mmol), acetylchloride (10.0 ml), rac-ethyl 1-(2-hydroxyethyl)nipecotinate (1.41
g, 6.99 mmol), K2CO3 (2.20 g, 15.9 mmol), acetontrile (20.0 ml). The crude
compound was purified by two consecutive flash column chromatographies on
silica (eluent pentane/Et2O 1:1 + 5% triethyl amine; Et2O 100%). Colourless oil
(2.35 g, 70 %).
rac-Ethyl 1-{2-[2-methoxy-1,1-bis(4-methoxyphenyl)-2-oxoethoxy]ethyl}-
piperidine-3-carboxylate (12k): GP3 was followed using 8k (680 mg, 2.24
mmol), acetylchloride (5.0 ml), rac-ethyl 1-(2-hydroxyethyl)nipecotinate (452
mg, 2.25 mmol), K2CO3 (774 mg, 5.60 mmol), MeCN (5.0 ml). The crude
compound was purified by flash column chromatography on silica (eluent
pentane/Et2O 1:1 to pentane/ Et2O 1:1 + 20% triethylamine). Colourless semi-
solid (956 mg, 88 %).
(S)-Ethyl
1-{2-[(4-formylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}-
piperidine-3-carboxylate [(S)-12a]: GP3 was followed using 8a (509 mg,
1.21 mmol), acetylchloride (3.0 ml), (S)-ethyl 1-(2-hydroxyethyl)nipecotinate
(318 mg, 1.58 mmol), K2CO3 (498 mg, 3.60 mmol), MeCN (5.0 ml). The crude
compound was purified by two consecutive flash column chromatographies on
silica (eluent pentane/Et2O 1:1 + 5% triethyl amine; Et2O 100%). Colourless oil
(324 mg, 41 %).
rac-Ethyl
1-[4-hydroxy-4,4-bis(4-methoxyphenyl)butyl]piperidine-3-
carboxylate (12l): A mixture of 8l (435 mg, 1.29 mmol), rac-ethyl nipecotinate
(233 mg, 1.42 mmol), K2CO3 (446 mg, 3.23 mmol) and KI (21 mg, 0.13 mmol)
in MeCN (5.0 ml) was irradiated in the microwave at 80°C for 24h, filtered and
reduced in vacuum. The crude compound was purified by flash column
chromatography on silica (eluent pentane/ Et2O 2:1 + 5% triethylamine).
Amorphous slightly yellowish solid (374 mg, 66 %).
(R)-Ethyl
1-{2-[(4-formylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}-
piperidine-3-carboxylate [(S)-12a]: GP3 was followed using 8a (380 mg,
0.900 mmol), acetylchloride (1.0 ml), (R)-ethyl 1-(2-hydroxyethyl)nipecotinate
(201 mg, 1.00 mmol), K2CO3 (311 mg, 2.25 mmol), MeCN (1.0 ml). The crude
compound was purified by two consecutive flash column chromatographies on
silica (eluent pentane/Et2O 1:1 + 5% triethyl amine;Et2O 100%). Colourless oil
(270 mg, 56 %).
rac-Ethyl 1-(2-{[4-(hydroxymethyl)phenyl]bis[4-methoxyphenyl]methoxy}-
ethyl)piperidine-3-carboxylate (12m): Sodium borohydride (46 mg, 1.2
mmol) was added to a solution of 12a (254 mg, 0.480 mmol) in MeOH (7.5 ml)
at 0°C. The ice bath was removed and stirring was continued at ambient
temperature for 3h. The mixture was poured into water (30.0 ml) and extracted
thrice with Et2O (30.0 ml). The combined organic phases were dried over
MgSO4, filtered and reduced in vacuum. Colourless oil (226 mg, 89 %).
rac-Ethyl
1-{2-[(4-acetylphenyl)bis(4-methoxyphenyl)methoxy]ethyl}-
piperidine-3-carboxylate (12b): GP3 was followed using 8b (344 mg, 0.950
mmol), acetylchloride (3.0 ml), rac-ethyl 1-(2-hydroxyethyl)nipecotinate (211
mg, 1.05 mmol), K2CO3 (329 mg, 2.38 mmol), MeCN (3.0 ml). The crude
compound was purified by flash column chromatography on silica (eluent
pentane/Et2O 6:4 + 5% trimethylamine). Colourless oil (261 mg, 50 %).
rac-Ethyl
1-{2-[(4-{[(2-methoxy-2-oxoethyl)amino]methyl}phenyl)bis(4-
methoxyphenyl)methoxy]ethyl}piperidine-3-carboxylate (12n): GP5 was
followed using 12a (449 mg, 0.850 mmol), methyl glycinate hydrochloride (212
mg, 1.69 mmol), sodium triacetoxyborohydride (250 mg, 1.18 mmol), CH2Cl2
(5.0 ml). The crude compound was purified by flash column chromatography
on silica (eluent isohexane/ethyl acetate 7:3 + 5% triethylamine). Colourless
semi-solid (289 mg, 57 %).
rac-Ethyl 1-(2-{[4-(N,N-dimethylsulfamoyl)phenyl]bis[4-methoxyphenyl]-
methoxy}ethyl)piperidine-3-carboxylate (12c): GP3 was followed using 8c
(455 mg, 1.06 mmol), acetylchloride (3.0 ml), rac-ethyl 1-(2-
hydroxyethyl)nipecotinate (213 mg, 1.06 mmol), K2CO3 (366 mg, 2.65 mmol),
MeCN (3.0 ml). The crude compound was purified by flash column
chromatography on silica (eluent pentane/Et2O 1:1 + 5% trimethylamine).
Yellowish semi-solid (411 mg, 64 %).
rac-Ethyl
1-{2-[(4-{[(3-ethoxy-3-oxopropyl)amino]methyl}phenyl)bis(4-
methoxyphenyl)methoxy]ethyl}piperidine-3-carboxylate (12o): GP5 was
followed using 12a (480 mg, 0.900 mmol), β-alanine methyl ester
11
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
hydrochloride (314 mg, 2.25 mmol), sodium triacetoxyborohydride (267 mg,
1.26 mmol), CH2Cl2 (10.0 ml). The crude compound was purified by flash
column chromatography on silica (eluent Et2O + 4% triethylamine). Pale yellow
oil (236 mg, 42 %).
application of vacuum. This step was repeated twice resulting in a total elution
volume of 300 µL. Finally, a volume of 130 µL ammonium formate buffer (10
mM, pH 7.0) was added per well and the generated samples analyzed by LC-
ESI-MS/MS using a API 3200 as described in detail previously[40]
.
rac-Ethyl
1-(2-{[4-({[1-(ethoxycarbonyl)cyclopropyl]amino}methyl)-
phenyl]bis[4-methoxyphenyl]methoxy}ethyl)piperidine-3-carboxylate
(12p): GP5 was followed using 12a (233 mg, 0.44 mmol), ethyl 1-
aminocyclopropanecarboxylate hydrochloride (145 mg, 0.88 mmol), sodium
triacetoxyborohydride (131 mg, 0.620 mmol), CH2Cl2 (2.0 ml). The crude
compound was purified by flash column chromatography on silica (eluent
pentane/Et2O + 4% triethylamine). Colourless semi-solid (168 mg, 59 %).
Conflict of interest
The authors declare no conflict of interest.
Keywords: GABA uptake inhibitors • mGAT4 • Medicinal chemistry •
Neurochemistry • Structure-activity relationships
[1]
[2]
[3]
[4]
[5]
P. Krogsgaard-Larsen, Med. Res. Rev. 1988, 8, 1, 27-56
B.S. Meldrum, A.G. Chapman, Epilepsia 1999, 40, 9, 2-6
D.M. Treiman, Epilepsia 2001, 42, 8-12
S.R. Kleppner, A.J. Tobin, Expert Opin. Ther. Targets 2001, 5, 219-239
M.A. Daemen, G. Hoogland, J.M. Cijntje, G.H. Spincemaille, Neurosci.
Lett. 2008, 444, 112-115
P. Nuss, Neuropsychiatr. Dis. Treat. 2015, 11, 165–175
A.V. Kalueff, D. J. Nutt, Depression Anxiety 2007, 24, 495–517
K.L. Lanctot, N. Herrmann, P. Mazzotta, L.R. Khan, N. Ingber, J., Can.
Psychiatr. Assoc. 2004, 49, 439-453
rac-Ethyl 1-{2-[(4-{[(1-ethoxy-2-methyl-1-oxopropan-2-yl)amino]methyl}-
phenyl)bis(4-methoxyphenyl)methoxy]ethyl}piperidine-3-carboxylate
(12q): GP5 was followed using 12a (152 mg, 0.290 mmol), ethyl 2-amino-2-
methylpropanoate
hydrochloride
(96
mg,
0.57
mmol),
sodium
triacetoxyborohydride (85 mg, 0.40 mmol), CH2Cl2 (2.0 ml). The crude
compound was purified by flash column chromatography on silica (eluent
pentane/Et2O 1:1 + 4% triethylamine). Colourless semi-solid (119 mg, 64 %).
[6]
[7]
[8]
rac-4-{[4-({2-[3-(Ethoxycarbonyl)piperidin-1-yl]ethoxy}bis{4-methoxy-
phenyl}methyl)benzyl]amino}butanoic acid (12s): GP6 was followed 4-
aminobutanoic acid (83 mg, 0.80 mmol), 12a (213 mg, 0.40 mmol), sodium
cyanoborohydride (37 mg, 0.56 mmol), MeOH (5.0 ml). The crude compound
was purified by flash column chromatography on silica (eluent MeOH).
Amorphous colourless solid (100 mg, 40 %).
[9]
Ł. Fijałkowski, K. Sałat, A. Podkowa, P. Zaręba, A. Nowaczyk, Eur. J.
Pharm. Sci. 2017, 96, 362-372
[10] K. Sałat, A. Podkowa, P. Kowalczyk, K. Kulig, A. Dzubina, B. Filipek, T.
Librowski, Pharmacol. Rep. 2015, 67, 465-472
[11] A. Schousboe, K.K. Madsen, M.L. Barker-Haliski, H.S. White,
Neurochem. Res. 2014, 39, 1980-1987
[12] A.S. Kristensen, J. Andersen, T.N. Jørgensen, L. Sørensen, J. Eriksen,
C.J. Loland, K. Strømgaard, U. Gether, Pharmacol. Rev. 2011, 63, 3,
585-640
[13] S. Bröer, U. Gether, Br. J. Pharmacol. 2012, 167, 2, 256-278
[14] L. Shi, M. Quick, Y. Zhao, H. Weinstein, J.A. Javitch, Mol Cell. 2008,
30, 667–677
rac-4-{[4-({2-[3-(Ethoxycarbonyl)piperidin-1-yl]ethoxy}bis{4-methoxy-
phenyl}methyl)benzyl]amino}benzoic acid (12t): GP6 was followed using 4-
aminobutanoic acid (217 mg, 1.58 mmol), 12a (420 mg, 0.790 mmol), sodium
cyanoborohydride (73 mg, 1.1 mmol), MeOH (10.0 ml). The crude compound
was purified by flash column chromatography on silica (eluent MeCN/H2O 9:1).
Amorphous slightly yellowish solid (132 mg, 25 %).
[15] K.K. Madsen, R.P. Clausen, O.M. Larsson, P. Krogsgaard-Larsen, A.
Schousboe, H.S. White, J. Neurochem. 2009, 109, 139-144
[16] N. O. Dalby, Eur. J. Pharmacol. 2003, 479, 127-137
[17] B. Christiansen, A. K. Meinild, A. A. Jensen, H. Bräuner-Osbore, J. Biol.
Chem. 2007, 282, 19331-19341
[18] L.A. Borden, Neurochem. Int. 1996, 29, 4, 335-356
[19] Y. Zhou, N.C. Danbolt, Front Endocrinol 2013, 4, 165
[20] S.A. Kempson SA, Y. Zhou Y, N.C. Danbolt, Front Physiol. 2014, 5,
159
Analytical data
The analytical data of the synthesized compounds 5a-t, 8a-l, 9c, 11, 12a-q
and 12s-t are presented in the Supporting Information.
Biological evaluation
[21] Y. Zhou, S. Holmseth, R. Hua, A. C. Lehre, A. M. Olofsson, I. Poblete-
Naredo, S. A. Kempson, N. C. Danbolt, Am J Physiol Renal Physiol.
2012, 302, 313-328
[3H]GABA uptake assays: The [3H]GABA uptake assays were performed
according to Kragler et al.[33] Cells were grown in 150 cm2 plates to a
confluence of 70 – 90 % up to passage 25. After detachment the cells were
centrifuged 5 min at 500 x g, washed for three times in phosphate buffered
saline and, finally, resuspended in Krebs buffer (2.5 mM CaCl2, 1.2 mM
MgSO4, 1.2 mM KH2PO4, 4.7 mM KCl, 11 mM glucose, 25 mM Tris, 119 mM
NaCl, pH 7.2). The uptake assays were performed with aliquots of the
resulting cell suspension (about 100000-200000 cells per well) in a total
volume of 250 µl in 96 well 2.2 ml deep well plates. The cells were equilibrated
for 25 min in Krebs buffer in the presence of the test compound at 37 °C in a
gently shaking water bath. In the case of poor solubility of the test compounds
all samples contained 1 % DMSO. After addition of 25 µl of a solution
containing [3H]GABA (2.22 TBq/mmol, ARC, St Louis, MO, USA) and
unlabeled GABA in Krebs (final concentration 8 nM [3H]GABA and 32 nM
unlabeled GABA for mGAT1, mGAT3, mGAT4 and 20 nM [3H]GABA and 20
nM unlabeled GABA for mGAT2) the cells were incubated for further 4 min
(mGAT1, mGAT3, mGAT4) or 10 minutes (mGAT2), respectively. The
incubation was stopped by filtration through Sartorius MGC micro glass fiber
filters pre-soaked for 1 h in 0.9 % NaCl by means of a Brandell MWXR-96TI
cell harvester (Brandell, Gaithersburg, MD, USA) under reduced pressure. The
filters were rinsed two times with cold 0.9% NaCl and subsequently transferred
to 96 well sample plates (Perkin Elmer LAS, Boston, MA, USA). After addition
of 200 µl Rotiszint Eco Plus (Roth, Karlsruhe, Germany) per well the
radioactivity was determined in a microbeta liquid scintillation counter (Perkin
Elmer LAS, Boston, MA, USA). Non-specific uptake was determined in
presence of 10 mM GABA for mGAT1, mGAT3 and mGAT4 or 100 mM GABA
for mGAT2, respectively. The results are expressed as means ± S.E.M. of at
least three separate experiments, each carried out in triplicate.
[22] K.K. Madsen, H.S. White, A. Schousboe, Pharmacol Ther. 2010, 125,
394-401
[23] I.E. Leppik, L. Gram, R. Deaton, K.W. Sommerville, Epilepsy Res.
1999, 33, 235–246
[24] T.G.M. Dhar, L.A. Borden, S. Tyagarajan, K.E. Smith, T.A. Branchek,
R.L. Weinshank, C. Gluchowski, J. Med. Chem. 1994, 37, 2334-2342
[25] J. Pabel, M. Faust, C. Prehn, B. Woerlein, L. Allmendinger, G. Höfner,
K.T. Wanner, ChemMedChem 2012, 7, 1245-1255
[26] M. Petrera, T. Wein, L. Allmendinger, M. Sindelar, J. Pabel, G. Höfner,
K.T. Wanner ChemMedChem 2016, 11, 519-538
[27] R. Schirrmacher, W. Hamkens, M. Piel, U. Schmitt, H. Lüddens, C.
Hiemke, F. Rösch, J. Label. Compd. Rad. 2001, 44, 627-642
[28] L.J.S. Knutsen, K.E. Andersen, J. Lau, B.F. Lundt, R.F. Henry, H.E.
Morton, L. Nrum, H. Petersen, H. Stephensen, P.D. Suzdak, J. Med.
Chem. 1999, 42, 18, 3447-3462
[29] G.C. Midya, A. Kapat, S. Maiti, J. Dash, J. Org. Chem. 2015, 80, 8,
4148-4151
[30] T. Ohwada, K. Shudo, J. Am. Chem. Soc. 1988, 110, 6, 1862-70
[31] G. Speck, C. Eickmeier, S. Pestel, S. Germeyer, M.P. Pieper, S.
Breitenfelder, M. Grauert, PCT Int. Appl. 2007, 2003064418
[32] A.F. Abdel-Magid, K.G. Carson, B.D. Harris, C.A. Maryanoff, R.D.
Shah, J. Org. Chem. 1996, 61, 3849-3862
[33] A. Kragler, G. Höfner, K.T. Wanner, Eur. J. Med. Chem. 2008, 43,
2404-2411
[34] C. Zepperitz, G. Höfner, K.T. Wanner, ChemMedChem 2006, 1, 298-
217
[35] MarvinSketch, version 6.1.4; ChemAxon, Budapest (Hungary),
MS Binding Assays: The MS Binding Assays were performed with mGAT1
membrane preparations obtained from a stably mGAT1 expressing HEK293
cell line and NO711 as unlabeled marker in competitive binding experiments
as previously described in detail[34]. In short, test compounds were investigated
at 7 concentrations in presence of 10 nM NO 711 in competition experiments.
Non specific binding was defined as binding remaining in the presence of 100
mM GABA. Incubation was terminated by transfer of 200 µL per well onto a
96-well filter plate (Acroprep, glass fibre, 1.0 µm, 350 µL, Pall, Dreieich,
Germany) with a 12 channel pipette. After rapid vacuum-filtration the filters
were washed with ice cold 0.9 % NaCl (5 x 150 µL) and dried at 50°C for 60
min. Afterwards, the filters were eluted with 100 µL methanol containing
[2H10]NO711 (1.4 nM) as internal standard per well into a deepwell plate by
http://www.chemaxon.com/products/marvin/marvinsketch/
July 30, 2019)
(accessed
[36] S. A. Hitchcock, L. D. Pennington, J. Med. Chem. 2006, 49, 7559–7583
[37] J. Mensch, J. Oyarzabal, C. Mackie, P. Augustijns, J. Pharm. Sci. 2009,
98, 4429–4468
[38] publication is to follow
[39] W.C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 14, 2923-2925
[40] G. Höfner, K.T. Wanner, J. Chromatogr. B 2010, 878, 1356-1364
12
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900719
ChemMedChem
FULL PAPER
Entry for the Table of Contents
Searching for novel GABA uptake inhibitors: Synthesis and biological evaluation of a
series of N-substituted nipecotic acids as potential inhibitors of the GABA transporter mGAT4.
Derived from the lead substance (S)-SNAP-5114 the compounds feature polar moieties in the
lipophilic domain in order to potentially enable increased polar interactions with the target.
13
This article is protected by copyright. All rights reserved.