Discovery of N-(2-aryl-cyclohexyl) substituted spiropiperidines as a novel class of GlyT1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2005.09.075

Screening of the Roche compound library led to the identification of cis-N-(2-phenyl-cyclohexyl)-spiropiperidine 1 as structurally novel GlyT1 inhibitor. The SAR, which was developed in this series, resulted in the discovery of highly potent compounds displaying excellent selectivity against the GlyT2 isoform.

Full text of DOI:10.1016/j.bmcl.2005.09.075

Bioorganic & Medicinal Chemistry Letters 16 (2006) 349–353
Discovery of N-(2-aryl-cyclohexyl) substituted spiropiperidines as
a novel class of GlyT1 inhibitors
Emmanuel Pinard,* Simona M. Ceccarelli, Henri Stalder and Daniela Alberati
F. Hoffmann-La Roche Ltd, Pharmaceutical Research Basel, Discovery Chemistry, CH-4070 Basel, Switzerland
Received 7 September 2005; revised 26 September 2005; accepted 28 September 2005
Available online 21 October 2005
Abstract—Screening of the Roche compound library led to the identification of cis-N-(2-phenyl-cyclohexyl)-spiropiperidine 1 as
structurally novel GlyT1 inhibitor. The SAR, which was developed in this series, resulted in the discovery of highly potent
compounds displaying excellent selectivity against the GlyT2 isoform.
Ó 2005 Elsevier Ltd. All rights reserved.
NMDA receptor hypofunction is suggested to be in-
volved in the pathophysiology of schizophrenia.1 The
strongest evidence supporting this hypothesis is based
on the observation that schizophrenic like symptoms
can be induced in healthy subjects upon administration
of NMDA blockers such as PCP.2 Thus, therapeutic
intervention aimed at increasing NMDA synaptic tone
is expected to show beneficial effect in schizophrenic
patients. As glycine is an obligatory co-agonist at the
NMDA receptor complex,3 one strategy to enhance
NMDA receptor activity is to elevate extracellular levels
of glycine in the local microenvironment of synaptic
NMDA receptor. Glycine elevation can be achieved by
inhibition of the glycine transporter 1 (GlyT1) which is
co-expressed in the brain with the NMDA receptor
and is responsible for glycine removal from the synaptic
cleft.4,5 Strong support for this approach in the treat-
ment of schizophrenia comes from clinical studies where
glycine6 and D-serine7 (co-agonists at the glycine site of
NMDA receptor) and sarcosine8 (a prototypical weak
GlyT1 inhibitor) improved positive, negative and cogni-
tive symptoms in schizophrenic patients, when added to
conventional therapy. As a result, considerable efforts
have been focused on the development of selective
GlyT1 inhibitors.9 The first examples reported were
glycine or sarcosine derivatives.10 More recently, a wide
variety of nonamino-acid GlyT1 inhibitors have been
disclosed.9b
We report here on our effort to discover and develop
structurally novel, potent and selective GlyT1 inhibi-
tors.11a To achieve this, screening of the Roche com-
pound depository was performed employing a glycine
uptake inhibition assay in cells transfected with human
GlyT1.11a This screening campaign led to the identifica-
tion of the cis-N-(2-phenyl-cyclohexyl)-spiropiperidine 1
as potent inhibitor of GlyT1 (26 nM), displaying more
than 400-fold selectivity against the type 2 isoform.
H
O
N
N
N
cis, rac
1
In this paper, we report on the SAR we have obtained
at the cyclohexyl moiety as well as around the aro-
matic rings. Selectivity against hGlyT2 transporter and
affinity at opioid receptors are also discussed.
cis-N-(2-Aryl-cycloalkyl)-spiropiperidines 1, 5–6, 11–39
were prepared by following two complementary routes
(Scheme 1). The first route, which was followed for
exploring SAR around R2, used the cis-piperidone 40
as a key intermediate. Compound 40 was obtained by
reductive amination of the a-aryl-substituted cyclic
ketone 41 with 1,4-dioxa-8-aza-spiro[4,5]decane fol-
lowed by acidic cleavage of the ketal protective group.
Importantly, this reductive amination provided exclu-
sively the cis-isomer. Conversion of piperidone 40 into
the final spiropiperidine compounds was achieved by
following a well-known 4-step sequence:12a,b Strecker
Keywords: GlyT1; GlyT2; NMDA; Schizophrenia; Transporter;
Glycine; Spiropiperidine.
*
Corresponding author. Tel.: +41 61 688 43 88; fax: +41 61 688 87
14; e-mail: emmanuel.pinard@roche.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.09.075

350
E. Pinard et al. / Bioorg. Med. Chem. Lett. 16 (2006) 349–353
N
O
H
N
O
O
R1
NH2
R1
R1
O
NH
R2
N
(a)
(b)
f
N
NH
R2
e
N
N
NH 2
N
N
(CH2)n
(CH2)n
(CH2)n
cis, rac
40
42
43
trans, rac
trans, rac
46
47
2
c,d
g, h
Scheme 2. Synthesis of trans-N-(2-phenyl-cyclohexyl)-spiropiperidine
2. Reagents and conditions: (a) 1-methyl-1-ethyl-4-oxo-piperidinium
iodide, K2CO3, H2O, EtOH, reflux, 50%; (b) see conditions of (e–h) in
Scheme 1.
H
N
O
R1
H
N
R1
O
N
O
i
R2
N
N
+
R2
N
(CH2)n
H
(CH2)n
41
44
cis, rac
H
O
H
N
N
a, b
e, f, g, h, j
O
1,5-6, 11-39
O
(a)
O
N
N
N
N
R3
n: 0-2
P: Benzyl or
3,5-Bis-trifluoromethyl-benzoyl
H
N
P
(CH2)n
H
N
7-8,10
48
45
O
H
N
O
N
Scheme 1. Synthesis of cis-N-(2-aryl-cycloalkyl)-spiropiperidine 1, 5–
6, 11–39. Reagents and conditions: (a) R1MgBr, CuBrÆMe2S, THF,
0 °C, 40–80%; (b) Dess Martin periodinate, 0.1% H2O, CH2Cl2, rt, 60–
80%; (c) 1,4-dioxa-8-aza-spiro[4,5]decane, cat TsOH, toluene, reflux,
Dean–Stark then NaBH(OAc)3, AcOH, 1,2-dichloroethane, rt, 70%;
(d) 6 N HCl, MeOH, reflux, 70%; (e) R2NH2, KCN, H2O, AcOH, rt;
(f) 90% H2SO4, rt; (g) triethyl orthoformate, AcOH, toluene, reflux,
48 h or 20 min (microwave); (h) NaBH4, MeOH, rt or H2 (1 atm), Pd/
C, AcOH, MeOH, 45 °C, 10–50% overall yield from 40; (i) Ti(OiPr)4,
105 °C then NaBH4, EtOH, rt, 10–60%; (j) performed only when P is
3,5-bis-trifluoromethyl-benzoyl: LiOH, H2O, MeOH, THF, rt, 80–
90%.
(b)
N
(c)
N
N
N
9
Scheme 3. Synthesis of spiropiperidines 7–10. Reagents and condi-
tions: (a) R3CHO, Ti(OiPr)4, 105 °C then NaBH4, EtOH, rt, 60–70%;
(b) acetone, TMSCN, AcOH, 35 °C, 18%; (c) benzylmagnesium
chloride, THF, 70 °C, 24%.
First, the influence of the relative and absolute stereo-
chemistry of the two vicinal chiral centres on GlyT1
activity and selectivity was investigated (Table 1). Inter-
estingly, the cis- and trans-configurated compounds 1
and 2 showed comparably low nanomolar GlyT1 activ-
ity. Moreover, 1 and 2 exhibited an equally low activity
at GlyT2 (P10 lM). Interestingly, in the cis stereomeric
series, the (1R,2R)-enantiomer 3 demonstrated an
extremely high activity at GlyT1 (4 nM) and two orders
of magnitude potency higher than that of the (1S,2S)-
enantiomer 4 (380 nM).16 As a consequence, 3 showed
a remarkably high selectivity against the GlyT2 trans-
porter (>1500-fold).
condensation with an aryl or alkyl amine, acidic hydro-
lysis of the resulting nitrile 42 into the amide 43, ring
closure of 43 in the presence of triethyl orthoformate
followed by reduction of the double bond with sodium
borohydride or hydrogen and Pd/C. Interestingly, the
usually tedious thermal cyclisation step could be
achieved efficiently with shorter reaction time in a
microwave oven. The second route, especially suited
for explorating variation around the left-hand aromatic
group, involved a reductive amination of the already
constructed spiropiperidine building block 44 with
ketone 41. This reaction worked best in our hands in
the presence of titanium(IV) isopropoxide.13 Again, this
reductive amination was stereoselective leading
exclusively to the cis-isomer. Spiropiperidine intermedi-
ate 44 was prepared from the N-protected-piperidone
45 using the same 4-step sequence described above
followed by cleavage of the protective group.
Structural modifications were then directed towards the
cyclohexyl system (Table 2). Replacement of the cis-
cyclohexyl group by smaller (cyclopentyl, 5) or larger
(cycloheptyl, 6) cis-aliphatic ring systems caused more
than a 40-fold loss in GlyT1 activity. This sharp drop
Table 1. In vitro inhibitory activity of 1–4 at GlyT1 and GlyT2a
The trans-isomer 2 was prepared from piperidone 46,
which was obtained from the known trans-2-phenyl-
cyclohexylamine14 using an efficient Hofmann-elimina-
tion–Michael-addition sequence15 (Scheme 2).
b
Compound
EC50 (lM)
Selectivityc
GlyT1
GlyT2
1 (cis, rac)
0.026
0.073
0.004
0.380
12
10
7
461
137
1750
45
2 (trans, rac)
3 (cis, 1R,2R)
4 (cis, 1S,2S)
Derivatives 7–8 and 10 bearing an acyclic R3 group were
prepared by reductive amination of commercially avail-
able spiropiperidine 48 with the required aldehydes
(Scheme 3). The gem-dimethyl derivative 9 was obtained
in 2 steps: Strecker condensation of 48 with acetone fol-
lowed by nucleophilic displacement of the nitrile group
by benzyl magnesium chloride.
17
a EC50 values are the geometric mean of at least two experiments with
<20% variance.
b [3H]Glycine uptake inhibition assay in cells transfected with hGly-
T111a or hGlyT211b cDNAs.
c Ratio of the EC50 (lM) values GlyT2/GlyT1.

E. Pinard et al. / Bioorg. Med. Chem. Lett. 16 (2006) 349–353
351
Table 2. In vitro inhibitory activity of 5–10 at GlyT1a Table 3. In vitro inhibitory activity of 11–19 at GlyT1 and GlyT2a
H
N
H
N
O
O
R1
N
N
N
N
R3
R1
EC50 (lM)
GlyT1 GlyT2
Selectivityc
b
b
Compound
R3
GlyT1 EC50 (lM)
Compound
11
12
13
14
15
16
17
18
19
4-F-Ph
4-Cl-Ph
4-Me-Ph
0.023
0.040
0.085
0.610
0.130
0.067
1.8
16
5
695
125
176
39
5
1.1
15
24
4-MeO-Ph
3-Cl-Ph
3,4-Cl2-Ph
2.6
6.8
nd
20
101
cis, rac
3-CF3,4-Cl-Ph
2-Me-Ph
2-Py
0.510
6.6
13
nd
25
a EC50 values are the geometric mean of at least two experiments with
<20% variance.
6
1.7
b [3H]Glycine uptake inhibition assay in cells transfected with hGly-
T111a or hGlyT211b cDNAs.
cis, rac
c Ratio of the EC50 (lM) values GlyT2/GlyT1.
7
8
>10
>10
9.3
electron-donating groups (14) led to significant drop of
GlyT1 inhibition. In the meta and ortho positions, a de-
crease of activity was also seen upon substitution (15,
17–18). Replacement of the aryl group by the more po-
lar 2-pyridyl group (19) led to a dramatic decrease of
activity. Best selectivity against the GlyT2 transporter
was achieved with the para-fluoro-phenyl-substituted
compound 11 (nearly 700-fold).
9
10
7.3
By keeping the phenyl group fixed at the 2-position on
the cyclohexyl ring, variation around the right-hand
aromatic system was then explored (Table 4).
a EC50 values are the geometric mean of at least two experiments with
<20% variance.
b [3H]Glycine uptake inhibition assay in cells transfected with hGly-
T111a or hGlyT211b cDNAs.
Introduction of electron-withdrawing or -donating
groups at the para position of the aromatic ring was
found to be well tolerated (20–24). Replacement of the
aromatic group by aliphatic residues was then explored.
The methyl derivative 25 was found to be completely
inactive. However, gratifyingly, elongating the linear al-
kyl chain from 1 to 4 carbons resulted in a steep increase
of potency. Excellent activity was indeed reached with
the N-pentyl derivative 28 (34 nM). Addition of an extra
methylene group then led to a slight decrease of activity
(29). Introduction of cycloalkyl residues of various sizes
provided weak compounds (30–33). However, insertion
of 1- or 2-methylene units between the nitrogen atom
and the cycloalkyl residue led to very potent derivatives
as exemplified with the two cyclohexyl analogues 34 and
35 showing and EC50 of 65 and 25 nM, respectively.
Interestingly, this high level of activity was retained by
replacing the cyclohexyl group in 35 by a phenyl group
(37). Introduction of polar functionalities in the alkyl
chain however was found to be detrimental (38–39).
Excellent selectivity against the GlyT2 transporter was
obtained for the most potent derivatives (Table 4).
of activity suggests that the 6-membered ring system has
the adequate size for establishing optimal lipophilic
interaction with the GlyT1 transporter. Replacement
of the cyclohexyl ring by noncyclic linkers was then con-
sidered with the aim of reducing the structural complex-
ity of our compounds. However, these attempts were
met with no success since the benzyl and phenethyl
derivatives as well as the more rigid gem-dimethyl-
substituted phenethyl analogues (7–10) were found
to be much less active. In summary, all modifications
considered around this part of the molecule were not
tolerated and therefore the original cis-cyclohexyl
system remained as the preferred template.
Next, exploration around the left-hand aromatic system
was performed by keeping the N-cyclohexyl-spiropiperi-
dine fixed (Table 3).
In the para position of this aromatic ring, the introduc-
tion of electron-withdrawing groups (11–12) was well
tolerated giving compounds of similar activity to the
original hit compound 1. Substitution with strong
Recently, structurally related N-(substituted-cyclohexyl)
spiropiperidines have been described as Nociceptin/

352
E. Pinard et al. / Bioorg. Med. Chem. Lett. 16 (2006) 349–353
Table 4. In vitro inhibitory activity of 20–39 at GlyT1 and GlyT2a
the NOP receptor. Disappointingly, however, these
compounds showed a consistently higher level of bind-
ing to the l opioid receptor. In particular, worst profile
was achieved with the N-aryl analogues (1, 3 and 21)
which displayed nanomolar activity at this receptor.
Moreover, the trans-diastereoisomer 2 exhibited an
equally high activity at both the NOP and l receptors.
H
O
N
N
R2
N
cis, rac
b
Compound R2
EC50 (lM)
GlyT1 GlyT2
23
2.9
3.4
27
Selectivityc
In summary, starting from hit compound 1, an SAR was
established which led to the identification of cis-N-(2-aryl-
cyclohexyl)-substituted spiropiperidines as a novel class
of highly potent GlyT1 inhibitors displaying excellent
selectivity against the GlyT2 isoform. In this class, affinity
at the l opioid receptor has been identified as a key liabil-
ity which requires optimization. In the next paper, results
of our effort to address this issue will be reported.18
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
4-F-Ph
0.024
0.027
0.065
0.066
0.055
958
107
52
4-Cl-Ph
4-CF3-Ph
4-MeO-Ph
4-Me-Ph
Me
410
164
9
>30
6.6
nd
nd
Et
nPr
0.450 >100
>222
1029
490
nPent
nHex
cPr
0.034
0.049
3.3
35
24
Acknowledgments
nd
We thank Patrick Boissin, Serge Burner, Patrick
Bourdeaux, Sandra Frauchiger, Sylvia Meyer, Marianne
Rueher, and Daniel Zimmerli for their dedicated techni-
cal assistance, and Synese Jolidon, Robert Narquizian,
Matthias Nettekoven, and Andrew Thomas for support
and advice.
cBu
cPent
1.5
>100
>100
11
>67
>159
15
0.63
0.75
0.065
0.025
cHex
CH2-cHex
25
15
385
600
CH2CH2-cHex
CH2-Ph
0.258 >100
0.025
5.3
>388
720
CH2CH2-Ph
CH2CH2-OMe
18
nd
nd
References and notes
39
7.4
CH2-CH2
N
a EC50 values are the geometric mean of at least two experiments with
<20% variance.
1. For a recent review see: Millan, M. J. Psychopharmacology
2005, 179, 30.
2. Jentsch, J. D.; Roth, R. H. Neuropsychopharmacology
1999, 20, 201.
3. Danysz, W.; Parsons, C. G. Pharmacol. Rev. 1998, 50, 597.
4. (a) Eulenburg, V.; Armsen, W.; Betz, H.; Gomeza, J.
Trends Biochem. Sci. 2005, 30, 325; (b) Gomeza, J.; Ohno,
K.; Betz, H. Curr. Opin. Drug Discov. Devel. 2003, 6, 675.
5. Bergeron, R.; Meyer, T. M.; Coyle, J. T.; Greene, R. W.
Proc. Natl. Acad. Sci. U.S.A 1998, 95, 15730.
6. Heresco-Levy, U.; Ermilov, M.; Lichtenberg, P.; Bar, G.;
Javitt, D. C. Biol. Psychiatry 2004, 55, 165.
7. Heresco-Levy, U.; Javitt, D. C.; Ebstein, R.; Vass, A.;
Lichtenberg, P.; Bar, G.; Catinari, S.; Ermilov, M. Biol.
Psychiatry 2005, 57, 577.
b [3H]Glycine uptake inhibition assay in cells transfected with hGly-
T111a or hGlyT211b cDNAs.
c Ratio of the EC50 (lM) values GlyT2/GlyT1.
Orphanin FQ peptide (NOP) receptor ligands.12,17
Therefore, in vitro affinity of some of the most GlyT1
active compounds identified was determined at the
NOP receptor as well as at the related l opioid receptor
to assess their potential side-effect liabilities (Table 5).
Interestingly, the cis-N-(2-aryl-cyclohexyl)-spiropiperi-
dines (1, 3, 21, 28 and 37) displayed low affinity for
8. Tsai, G.; Lane, H.-Y.; Yang, P.; Chong, M.-Y.; Lange, N.
Biol. Psychiatry 2004, 55, 452.
9. For recent reviews, see: (a) Sur, C.; Kinney, G. G. Expert
Opin. Investig. Drugs 2004, 13, 515; (b) Slassi, A.; Egle, I.
Expert Opin. Ther. Patents 2004, 14, 201.
10. (a) Atkinson, B. N.; Bell, S. C.; De Vivo, M.; Kowalski, L.
R.; Lechner, S. M.; Ognyanov, V. I.; Tham, C.-S.; Tsai,
C.; Jia, J.; Ashton, D.; Klitenick, M. A. Mol. Pharmacol.
2001, 60, 1414; (b) Brown, A.; Carlyle, I.; Clark, J.;
Hamilton, W.; Gibson, S.; McGarry, G.; McEachen, S.;
Rae, D.; Thorn, S.; Walker, G. Bioorg. Med. Chem. Lett.
2001, 11, 2007.
11. (a) Ceccarelli, S. M; Pinard, E; Stalder, H. WO Patent
2005040166, 2005; Chem. Abstr. 2005, 142, 447121; (b)
Same conditions as described for the hGlyT1 assay (see
Ref. 11a) with slight modifications: concd [3H]glycine:
200 nM, cold glycine not present.
12. (a) Roever, S.; Adam, G.; Cesura, A. M.; Galley, G.; Jenck,
F.; Monsma, F. J.; Wichmann, J.; Dautzenberg, F. M.
J. Med. Chem. 2000, 43, 1329; (b) Wichmann, J.; Adam,
G.; Roever, S.; Hennig, M.; Scalone, M.; Cesura, A. M.;
Table 5. In vitro binding activities of 1, 2, 4, 21, 28 and 37 at the NOP
and l receptorsa
b
c
d
Compound GlyT1 EC50 (lM) NOP IC50 (lM) l IC50 (lM)
1
2
0.026
0.073
0.004
0.027
0.034
0.025
6
0.15
0.54
0.041
0.23
2.2
0.3
3.7
4.0
3
21
28
37
>10
>10
3.2
a EC50 values are the geometric mean of at least two experiments with
<20% variance.
b [3H]Glycine uptake inhibition assay in cells transfected with hGly-
T111a or hGlyT211b cDNAs.
c Displacement of [3H]NOP in membranes prepared from permanently
transfected HEK293 cells expressing hNOP receptors.12b,c
d Displacement of [3H]naloxone in membranes prepared from BHK
cells transiently expressing hl receptors.12b

E. Pinard et al. / Bioorg. Med. Chem. Lett. 16 (2006) 349–353
353
Dautzenberg, F. M.; Jenck, F. Eur. J. Med. Chem. 2000,
35, 839(c) The hNOP data mentioned in this communica-
tion were generated in the years 2001 and 2002.
16. (a) Enantiomers 3 and 4 were obtained by separation
of racemic compound on chiral-phase HPLC
1
(Chiralpack AD); (b) Absolute configuration of enan-
tiomer 4 has been determined by X-ray crystallogra-
phy analysis of its (1R)-(À)-10-camphorsulfonic acid
salt.
13. Bhattacharyya, S. Tetrahedron Lett. 1994, 35, 2401.
14. Wong, W. C.; Sun, W.; Cui, W.; Chen, Y.; Forray, C.;
Vaysse, P. J.-J.; Branchek, T. A.; Gluchowski, C. J. Med.
Chem. 2000, 43, 1699.
15. Tschaen, D. M.; Abramson, L.; Cai, D.; Desmond, R.;
Dolling, U. H.; Frey, L.; Karady, S.; Shi, Y.-J.; Verho-
even, T. R. J. Org. Chem. 1995, 60, 4324.
17. Roever, S.; Wichmann, J.; Jenck, F.; Adam, G.; Cesura,
A. M. Bioorg. Med. Chem. Lett. 2000, 10, 831.
18. Cecarelli, S. M.; Pinard, E; Stalder, H.; Alberati, D.
Bioorg. Med. Chem. Lett., 2006, 2, 354.