Stereospecific synthesis and structure-activity relationships of unsymmetrical 4,4-diphenylbut-3-enyl derivatives of nipecotic acid as GAT-1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2010.09.025

Two complementary stereospecific synthetic approaches for the preparation of unsymmetrical ortho-substituted N-(4,4-diphenylbut-3-enyl) derivatives of nipecotic acid are described. Determination of the activity of the prepared compounds at the GAT-1 trans

Full text of DOI:10.1016/j.bmcl.2010.09.025

Bioorganic & Medicinal Chemistry Letters 21 (2011) 602–605
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Stereospecific synthesis and structure–activity relationships
of unsymmetrical 4,4-diphenylbut-3-enyl derivatives of nipecotic
acid as GAT-1 inhibitors
Domenica A. Pizzi , Colin P. Leslie , Romano Di Fabio, Catia Seri, Giovanni Bernasconi , Michela Squaglia à,
⇑
⇑
Gennaro Carnevale §, Alessandro Falchi –, Elisabetta Greco ||, Laura Mangiarini, Michele Negri
Neurosciences Centre of Excellence for Drug Discovery, GlaxoSmithKline, Medicines Research Centre, Via A. Fleming 4, 37135, Verona, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 June 2010
Revised 3 September 2010
Accepted 4 September 2010
Available online 15 September 2010
Two complementary stereospecific synthetic approaches for the preparation of unsymmetrical
ortho-substituted N-(4,4-diphenylbut-3-enyl) derivatives of nipecotic acid are described. Determination
of the activity of the prepared compounds at the GAT-1 transporter highlighted differing SAR require-
ments of the E- and Z-phenyl rings, and led to the discovery of a compound with comparable potency
to tiagabine. Some attempts to replace nipecotic acid with alternative novel amino acids are also
described.
Keywords:
GAT-1
Ó 2010 Published by Elsevier Ltd.
GABA transporter inhibitor
The amino acid 4-aminobutyric acid (GABA) is recognised as the
major inhibitory neurotransmitter in the brain.1–3 Considerable
direct and indirect evidence suggests that alterations of the
GABAergic system may to be implicated in several psychiatric
and neurological disorders, including anxiety, pain, Huntington’s
chorea, epilepsy and Parkinson’s disease.4–8 A number of strategies
are available for increasing GABAergic tone including agonism of
GABA receptors, inhibition of GABA catabolism and inhibition
of GABA reuptake. The latter approach requires inhibitors of one
or more of the GABA uptake transporters (GAT 1–3, BGT-1),9 of
which the GAT-1 transporter has been most extensively studied.
The therapeutic potential of the earliest GAT-1 inhibitors, such
as nipecotic acid (1) and guvacine (2), was limited by their inability
to penetrate the blood–brain barrier (BBB). Subsequently, it was
demonstrated that modification of these structures with a lipo-
philic side chain led to highly potent GAT-1 inhibitors exemplified
by SKF-89976A (3), SKF-100300A (4) and tiagabine (5) (Fig. 1),
which are able to cross the blood-brain barrier.10 The pharmacol-
ogy of these compounds has been extensively studied.11,12
In particular, tiagabine has been shown to exhibit anticonvul-
sant activity in clinical trials and is currently approved as an
add-on therapy for the treatment of partial seizures.
A body of preclinical evidence suggests that GAT-1 inhibitors
have the potential to function as effective anxiolytic agents.13 Fur-
thermore, at the time this work was initiated proof of concept
CO2H
CO2H
N
O
N
N
OH
H
H
nipecotic acid
guvacine
SKF-89976A
⇑
Corresponding authors. Tel.: +39 045 8218931; fax: +39 045 8218196 (D.A.P.);
1
2
3
tel.: +39 045 8218546; fax: +39 045 8218196 (C.P.L.).
E-mail addresses: domenica.a.pizzi@gsk.com, pizzidomenica@hotmail.com (D.A.
Pizzi), colin.p.leslie@gsk.com (C.P. Leslie).
S
N
O
Present address: European Chemicals Agency, Annankatu 18-PO Box 400, 00120
Helsinki, Finland.
N
O
OH
à
OH
S
Present address: Global Regulatory Affairs, CMC Pre-approval, GlaxoSmithKline,
Via A. Fleming 4, 371335 Verona, Italy.
SKF-100330A
Tiagabine
§
Present address: Arke’ Organics, Zona Industriale della Botte 4, 56012 Fornacette,
Pisa, Italy.
4
5
–
Present address: Chiesi Farmaceutici, Via Palermo 26/A, 43122 Parma, Italy.
||
Present address: San Raffaele Biomedical Science Park, Via Olgettina 58, 20132,
Figure 1. Some examples of potent GABA uptake inhibitors described in the
Milano, Italy.
literature.
0960-894X/$ - see front matter Ó 2010 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2010.09.025

D. A. Pizzi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 602–605
603
(POC) clinical trials undertaken by Cephalon had shown tiagabine
held promise as a novel approach for the treatment of generalised
anxiety disorder (GAD) and post traumatic stress disorder
(PTSD).14 As part of a research project aimed at the identification
of novel GABA reuptake inhibitors for the treatment of anxiety dis-
orders we were stimulated to revisit the SKF-89976A scaffold to
perform a wider exploration of the structure–activity relationship
(SAR) as remarkably little data are available for the aryl substitu-
tion of this system. Despite reports indicating the importance of
an ‘ortho’ substituent in more than one GAT-1 inhibitor series15,16
and others proving the advantageousness of having unsymmetrical
substitution patterns in the two aryl rings17 only one example of an
unsymmetrical, ortho-substituted derivative of SKF-89976A has
been reported.18 Herein, we describe the synthesis and the biolog-
ical activity of a series of novel unsymmetrical ortho-substituted
N-(4,4-diarylbut-3-enyl) derivatives of nipecotic acid (6) (Fig. 2).
We also communicate a limited exploration of novel alternative
amino acid head groups which were evaluated using the SKF-
89976/tiagabine side chain (7) (Fig. 2).
the first step, the cyclopropylphenylcarbinol intermediate 8 was
prepared by reaction of commercially available cyclopropylphe-
nylketones with an appropriate Grignard reagent, then treatment
with hydrobromic acid in acetic acid afforded E/Z mixtures of
unsymmetric 4-bromo-1,1-diaryl-1-butenes 9. Nucleophilic sub-
stitution with ethyl (R)-nipecotate and subsequent saponification
gave corresponding E/Z mixtures of products 11. Unfortunately,
we were unable to separate these mixtures, or the intermediates
9 and 10, into pure E and Z isomers such that SAR analysis became
impractical. Consequently efforts were made to develop alternative
synthetic routes to allow stereospecific preparation of E and Z iso-
mers. Two complementary stereospecific syntheses (Methods B
and C, Scheme 1) were devised that allowed variation of the E-
or Z-aryl ring at a late stage, facilitating SAR exploration. Method
B exploits the known propensity of sodium bis(2-methoxyethoxy)
aluminium hydride (Red-AlÒ) to reduce 3-butynol derivatives ste-
reoselectively; the high stereoselectivity observed during this
reduction is a consequence of chelation-assisted hydroalumination
giving Z-vinylaluminiums which in turn can be trapped with elec-
trophiles.19,20 Hence, commercially available 3-butyn-1-ol was
transformed under Sonogashira conditions into the corresponding
4-phenyl-3-butyn-1-ol 12. Reduction of this alcohol with Red-AlÒ
in THF followed by addition of iodine at -78 °C afforded isomerical-
ly pure (Z)-4-iodio-4-aryl-3-buten-1-ol 13. Activation of the alco-
hol as its methanesulfonate ester and nucelophilic substitution
with ethyl (R)-nipecotate afforded derivative 14 which was trans-
formed into the corresponding 4,4-diaryl-butenyl derivative 15
through a Suzuki-type coupling with an appropriately substituted
boronic acid. Alkaline saponification then provided the target prod-
ucts 6.
The synthetic routes used for the preparation of compounds 6
are illustrated in Scheme 1.
Initially an established approach for the preparation of
4,4-diarylbut-3-enyl derivatives (Method A) was employed. In
X
Ar
O
N
R
Ar
OH
Y
Alternatively, Method C derives its stereoselectivity from the
syn nature of the hydrostannylation of alkynes.21 In this case
the same starting material, 3-butyn-1-ol, was first converted into
3-butyn-1-yl methanesulfonate and then used to alkylate ethyl
(R)-nipecotate. Alkyne 16 was arylated under Sonogashira
6
7
R = amino acids
H, alkyl, halogen, alkoxy
X≠Y=
Ar = Ph, 3-methyl-thiophen-2-yl
Figure 2. General structures of diaryl butenyl derivatives explored.
Method A
X
X
O
N
X
N
X
X
CO2Et
CO2H
a
b
c
d
Br
Y
Y
HO
Y
Y
8
9
10
11
Method B
OH
I
OH
N
N
X
CO2Et
CO2Et
e
d
Y
f
g
h
6
Y
Y
Y
Y
I
I
12
13
CO2Et
14
15
Method C
OH
CO2Et
N
N
N
X
CO2Et
CO2Et
d
k
g
i
j
l
N
6
15
X
I
Bu3Sn
X
16
17
18
19
Scheme 1. Synthetic routes to butenyl GAT-1 inhibitors. Reagents and conditions: (a) 2-Y-phenylmagnesium halide, THF, rt to reflux; (b) AcOH, HBr; 10 °C; (c) methyl
(R)-nipecotate tartrate, DIPEA, DMF, 60 °C; (d) NaOH, EtOH, rt then HCl, EtOH; (e) CuI, Pd(PPh3)4, TEA, 3-butyn-1-ol; (f) Red-Al, THF, reflux then I2, À78 °C; (g) MsCl, TEA, DCM
then methyl (R)-nipecotate tartrate, K2CO3, KI, 2-butanone; (h) 2-X-phenylboronic acid, Pd(PPh3)4, Na2CO3, DME, 110 °C, microwave irradiation; (i) 1-iodo-2-X-benzene, CuI,
Pd(PPh3)4, TEA; (j) Bu3SnH, PdCl2(PPh3)2, THF; (k) I2, DCM, rt; (l) 2-Y-phenylboronic acid, Pd(PPh3)4, Na2CO3, DME, 110 °C, microwave irradiation.

604
D. A. Pizzi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 602–605
conditions to give 17, which was then subjected to a regio and
stereospecific hydrostannylation to afford the E-vinylstannane
18. The vinylstannane 18 itself is a suitable intermediate for
the preparation of 15 via Stille coupling but in practise it was
most convenient to transform the vinylstannane into vinyliodide
19 via stereospecific iodo-destannylation. The synthetic sequence
was completed, as in Method B, with a Suzuki-type coupling and
alkaline saponification.
The affinity of the compounds prepared for the hGAT-1 trans-
porter was measured in an equilibrium binding assay22 utilising
[3H]-tiagabine as radioligand.23 Data for a selection of the com-
pounds tested is shown in Table 1.
An analysis of the monosubstituted derivatives reveals distinct
SAR differences between the E- and Z-aryl rings. All ortho-substit-
uents examined on the E-phenyl ring led to a loss of activity in
comparison to the parent SKF-89976A (6a–6k, Table 1). Activity
appears to drop with the size of the substituent and with electron
withdrawing groups indicating that both steric and electronic fac-
tors are important. In contrast, many of the same ortho-substitu-
ents on the Z-aryl ring led to increased GAT-1 inhibitory activity
(6l–6n, 6p–6q, 6s–6t, 6v). Bis-ortho substitution of the Z-aryl ring
was therefore investigated (6u, 6w) but the presence of a second
ortho-substituent does not lead to further gains in affinity. A
methyl substituent appears to be optimal although ethyl, chlorine
and bromine also give appreciable increases in affinity. Indeed, the
benefits of an ortho-methyl substituent have previously been noted
by Knutsen15 who reported the symmetrical di-methyl derivative
6ag, however, our results clearly show that the second methyl
group, on the E-aryl group is actually detrimental and lowers affin-
ity compared to the unsymmetrical monosubstituted derivative
(6p vs 6ag). Likewise the majority of the unsymmetrical disubsti-
tuted derivatives prepared (X – H – Y) showed lower activity than
the corresponding monosubstituted analogues (Y = H; X – H)
reflecting the negative influence of substitution on the E-ring
(e.g., see 6ad–6ag vs 6p). Curiously, however, disubstituted deriv-
atives with an ortho-fluorine substituent on the E-ring (Y = F;
X – H) had comparable or slightly higher affinities that the corre-
sponding monosubstituted analogues (Y = H; X – H), despite the
Table 2
In vitro affinity(pKia) of compounds 7
Compound
Ar
R
pKi (hGAT-1)
Tiagabine (5)
SKF-89976A (3)
7.77
7.14
O
O
O
O
S
F
F
OH
OH
OH
OH
7a
7b
7c
7d
6.86
6.08
5.56
<5
N
N
N
Table 1
In vitro affinity (pKia) of compounds 6
Compound
X
Y
pKi (hGAT-1)
Tiagabine (5)
SKF-89976A (3)
6a
6b
6c
6d
6e
6f
6g
6h
6i
6j
6k
6l
6m
6n
6o
6p
6q
6r
6s
—
H
H
H
H
H
H
H
H
H
H
H
H
F
Cl
Br
—
H
F
Cl
Br
CF3
Me
Et
Ph
OMe
OCF3
2,6-di-F
2-Me,4-F
H
H
H
H
H
H
H
H
H
H
H
7.77
7.14
6.43
6.78
6.70
5.72
7.03
6.99
6.17
6.43
6.19
6.16
6.42
7.16
7.39
7.54
7.03
7.69
7.53
5.97
7.25
7.23
6.48
7.47
7.11
7.14
7.05
7.33
6.80
6.45
7.83
7.21
6.42
7.15
7.38
7.42
7.74
7.66
7.28
7.27
7.72
7.41
S
O
N
S
S
N
7e
7f
<5
<5
OH
O
CF3
Me
Et
Ph
OMe
OCF3
2,6-di-F
2-Me,4-F
2,6-di-Me
F
F
F
F
F
Me
Me
Me
Me
Me
Et
N
O
O
N
OH
OH
S
S
S
6t
6u
6v
6w
6x
6y
H
H
7g
7h
<5
<5
H
Me
Et
F
Cl
CF3
F
Cl
CF3
OMe
Me
F
F
F
F
F
F
F
OH
H
O
6z
6aa
6ab
6ac
6ad
6ae
6af
6ag
6ah
6ai
6aj
6ak
6al
6am
6an
H
N
O
OH
H
7i
<5
<5
H
Cl
Br
N
OCF3
OMe
2-Me,4-F
CF3
O
S
OH
H
7j
H
a
[3H]-Tiagabine radioligand binding assay to determine affinity at human
N
recombinant GAT-1 transporters transiently expressed in HEK293 membranes
using BacMam technology. Each determination lies within 0.3 log units of the mean.
a
See Table 1.

D. A. Pizzi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 602–605
Ar
605
Ar
Acknowledgements
a
Br
R
Ar
Ar
The authors thank Dr. Carla Marchiorro and members of the
Analytical Chemistry department, Verona for support in the analyt-
ical characterisation of the compounds described.
7
20 Ar=Ph
21 Ar=3-methyl-thiophen-2-yl
Scheme 2. Synthetic route to afford compounds 7. Reagents and conditions: (a) (i)
amino acid ester, DIPEA, DMF, 65 °C; (ii) NaOH, EtOH, rt then HCl in Et2O.
Reference and notes
1. GABA in Nervous System Function; Roberts, E., Chase, T. N., Tower, D. B., Eds.;
Raven Press: New York, 1976.
2. Amino Acid Neurotransmission; Stephenson, F. A., Turner, A. J., Eds.; Portland
Press: London, 1997.
3. Krogsgaad-Larsen, P.; Frølund, B.; Frydenvany, K. Curr. Pharm. Design 2000, 6,
1193.
4. Nutt, D. J.; Malizia, A. L. Br. J. Psychiatry 2001, 179, 390; Kalueff, A. V.; Nutt, D. J.
Depress. Anxiety 2007, 24, 495.
5. Lydiard, R. B. J. Clin. Psychiatry 2003, 64, 21.
fact that in isolation a fluorine substituent on the E-ring led to a net
fall in affinity (6a).
Evidently the SARs on the two aryl groups are not simply addi-
tive but can synergise in the right combination. The optimal substi-
tution pattern from our studies was found to be ortho-methyl on
the Z-ring and ortho-fluoro on the E-ring giving a 4,4-diphenyl-
but-3-enyl derivative (6ac) of comparable affinity to tiagabine.
Compound 6ac was further profiled in vitro where it was found
to have low CYP450 inhibition potential (IC50 >10 lM at 1A2,
2C9, 2C19 and 3A4 isoforms) and high selectivity versus the other
GABA transporters (pIC50 <4 at BGT-1, GAT-2 and GAT-3).
As part of our exploration we also briefly investigated replace-
ment of the nipecotic acid headgroup. Many amino acids and ami-
no acid isosteres have previously been evaluated for their ability to
inhibit GABA uptake,24 with best results being achieved with con-
formationally restricted cyclic amino acids, most notably nipecotic
acid. We therefore limited our efforts to a small number of nipecot-
6. Chase, T. N.; Wexler, N. S.; Barbeau, A. Huntington’s Disease; Raven Press: New
York, 1988.
7. Nistico, G.; Engel, J.; Fariello, R. G.; Lloyd, K. G.; Morselli, P. L. Neurotrasmitters,
Seizures and Epilepsy III; Raven Press: New York, 1986.
8. Ondo, W. G.; Hunter, C. Mov. Disord. 2003, 18, 683.
9. In this manuscript the nomenclature proposed by the HUGO gene
nomenclature committee has been adopted. For a detailed overview of the
nomenclature of GABA transporters see Madsen, K. K.; Clausen, R. P.; Larsson,
O. M.; Krogsgaard-Larsen, P.; Schousboe, A.; White, H. S. J. Neurochem. 2009,
109, 139.
10. Ali, F. E.; Bondinell, W. E.; Dandridge, P. A.; Frazee, J. S.; Garvey, E.; Girard, G. R.;
Kaiser, C.; Ku, T. W.; Lafferty, J. J.; Moonsammy, G. I.; Oh, H.; Rush, J. A.; Setler,
P. E.; Stringer, O. D.; Venslavsky, J. W.; Volpe, B. W.; Yunger, L. M.; Zirkle, C. L. J.
Med. Chem. 1985, 28, 653.
11. Halonen, T.; Nissinen, J.; Jansen, J. A.; Pitkanen, A. Eur. J. Pharmacol. 1996, 299,
69.
12. Meldrum, B. S. Epilepsia 1996, 37, S4.
ic acid derivatives and novel cyclic b- and
c-amino acids. The
products 7 (Table 2) were prepared uneventfully as outlined in
Scheme 2; all compounds are racemic or meso and stereochemis-
try, where shown, is used to indicate the relative configurations
of the substituents.
13.
Hug, S.; Greenwood, J. R.; Madsen, K. B.; Larsson, O. M.; Frulund, B.;
Schousboe, A.; Krogsgaard-Larsen, P.; Clausen, R. P. Curr. Top. Med. Chem.
2006, 6, 1861.
14. POC 17th Cong. Eur. Coll. Neuropsychopharmacol. (Oct. 9–13, Stockholm), 2004,
14; Abstr. P.3.206 and Abstr. P.3.004.
The requisite amino acid ester starting materials were pur-
chased or prepared according to literature procedures then alkyl-
ated with the appropriate 4-bromo-1,1-diaryl-1-butenes 20 or
2115 and saponified to give products 7.
Disappointingly, all of the structures investigated were consid-
erably less active than the reference compounds 3 and 5, with
GAT-1 binding only being measurable for the modified nipecotic
acid derivatives 7a–c (Table 2). Our results support the strict
spatial requirements proposed for GAT-1 activity16 and further-
more, the 10-fold drop in potency observed upon introduction of
15. Andersen, K. E.; Braestrup, C.; Grønwald, F. C.; Jørgensen, A. S.; Nielsen, E. B.;
Sonnewald, U.; Sørensen, P. O.; Suzdak, P. D.; Knutsen, L. J. S. J. Med. Chem.
1993, 36, 1716.
16. Knutsen, L. J. S.; Andersen, K. E.; Lau, J.; Lundt, B. F.; Henry, R. F.; Morton, H. E.;
Petersen, H.; Stephensen, H.; Suzdak, P. D.; Swedberg, M. D. B.; Thomsen, C.;
Sørensen, P. O. J. Med. Chem. 1999, 42, 3447.
17. Andersen, K. E.; Behrend, J. L.; Lundt, F.; Petersen, H.; Huusfeldt, P. O.; Suzdak,
P. D.; Swedberg, M. D. B. Bioorg. Med. Chem. 2001, 9, 2773.
18. (a) Larsson, O. M.; Falch, E.; Krogsgaard-Larsen, P.; Schousboe, A. J. Neurochem.
1988, 50, 818; (b) Bondinell, W. E.; Lafferty, J. J.; Zirkle, C. L.; EP 66456 A1, 1982.
19. Crousse, B.; Alami, M.; Linstrumelle, G. Synlett 1997, 8, 992.
20. Ma, S.; Liu, F.; Negishi, E. Tetrahedron Lett. 1997, 38, 3829.
21. Liron, F.; Gervais, M.; Peyrat, J. F.; Alami, M.; Brion, J. D. Tetrahedron Lett. 2003,
44, 2789.
an
a-fluorine to nipecotic acid indicate the importance of main-
22. [3H]-Tiagabine displacement experiments were performed at equilibrium by
taining suitable pKa/pKb of the amino acid functionality.
the addition of HEK293 BacMam membranes expressing hGAT-1 (30–45 lg of
In conclusion, two complementary stereospecific syntheses of
unsymmetric N-(4,4-diaryl-3-butenyl)nipecotic derivatives were
developed. Access to isomerically pure products allowed insights
into the SAR on the aryl rings that otherwise would not have been
possible. Indeed, it was established that the previously reported
beneficial effect of an ortho-substituent on the phenyl rings15 is
limited to ortho substitution of the Z-phenyl ring. In contrast ortho
substitution of the E-phenyl ring is detrimental to activity, except
in the specific case of ortho-fluoro substitution of the E-ring in
combination with ortho substitution of the Z-ring; in this instance
a synergistic effect appears to be operating such that GAT-1 affin-
ities are further enhanced. Accordingly, the optimal substitution
pattern was found to be ortho-methyl on the Z-ring and ortho-flu-
oro on the E-ring giving a 4,4-diphenylbut-3-enyl derivative (6ac)
of comparable affinity to tiagabine.
protein per well) to 96 deep-well plates containing test compounds (1% final
DMSO), [3H]-tiagabine (Amersham Custom Synthesis TRQ10938,ꢀ1 nM final
concentration) and 100 lM SKF-89976-A for non specific binding in a total
volume of 400 ll/well of assay buffer (50 mM Tris–HCl, 1000 mM NaCl, 5 mM
KCl, pH 7.7). After 2 h incubation at room temperature the reaction was
stopped by rapid filtration through GF/B unifilter 96 plates (Perkin–Elmer
6005177), using an Omnifilter-96 harvester (Packard). Filters were washed
(6 Â 1 ml) with ice-cold 0.9% saline and counted for radioactivity using a
TopCount liquid scintillation counter.
Binding data were analysed using an iterative non-linear least square curve
fitting programme (GraphPad Prism™). Affinities (pKi values) were calculated
from the IC50 values using the Cheng–Prusoff equation.
23. [3H]-Tiagabine selectively and potently binds to the GAT-1 transporter but is
not itself a substrate of GAT-1; see Braestrup, C.; Nielsen, E. B.; Sonnewald, U.;
Knutsen, L. J. S.; Andersen, K. E.; Jansen, J. A.; Frederiksen, K.; Andersen, P. H.;
Mortensen, A.; Suzdak, P. D. J. Neurochem. 1990, 54, 639.
24. (a) Fulep, G. H.; Hoesl, C. E.; Hofner, G.; Wanner, K. T. Eur. J. Med. Chem. 2006,
41, 809; (b) Kragler, A.; Höfner, G.; Wanner, K. T. Eur. J. Pharmacol. 2005, 519,
43.