4-Substituted-8-(1-phenyl-cyclohexyl)-2,8-diaza-spiro[4.5]decan-1-one as a novel class of highly selective GlyT1 inhibitors with superior pharmacological and pharmacokinetic parameters

Article abstract of DOI:10.1016/j.bmcl.2006.05.063

A novel class of 4-substituted-8-(1-phenyl-cyclohexyl)-2,8-diaza-spiro[4.5]decan-1-ones have been discovered and developed as potent and selective GlyT1 inhibitors. The molecules are devoid of activity at the GlyT2 isoform and display excellent selectivities against the μ opioid receptor as well as the nociceptin/orphanin FQ peptide (NOP) receptor. These molecules also exhibit superior pharmacological and pharmacokinetic parameters, relative to all GlyT1 inhibitors of the spiropiperidine family, culminating in the identification of 16b with an oral bioavailability of ～60%. In addition, a straightforward two-step procedure for the assembly of the target molecules is also presented.

Full text of DOI:10.1016/j.bmcl.2006.05.063

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4321–4325
4-Substituted-8-(1-phenyl-cyclohexyl)-2,8-diaza-spiro[4.5]decan-1-
one as a novel class of highly selective GlyT1 inhibitors with
superior pharmacological and pharmacokinetic parameters
a
b
c
c
c
`
Daniela Alberati, Dominik Hainzl, Synese Jolidon, Anke Kurt, Emmanuel Pinard,
Andrew W. Thomasc,* and Daniel Zimmerlic
aDiscovery Biology, F. Hoffmann-La Roche Ltd, Pharmaceuticals Division, CH-4070 Basel, Switzerland
bSafety Technical Services, F. Hoffmann-La Roche Ltd, Pharmaceuticals Division, CH-4070 Basel, Switzerland
cDiscovery Chemistry, F. Hoffmann-La Roche Ltd, Pharmaceuticals Division, CH-4070 Basel, Switzerland
Received 7 April 2006; revised 17 May 2006; accepted 17 May 2006
Available online 9 June 2006
Abstract—A novel class of 4-substituted-8-(1-phenyl-cyclohexyl)-2,8-diaza-spiro[4.5]decan-1-ones have been discovered and devel-
oped as potent and selective GlyT1 inhibitors. The molecules are devoid of activity at the GlyT2 isoform and display excellent selec-
tivities against the l opioid receptor as well as the nociceptin/orphanin FQ peptide (NOP) receptor. These molecules also exhibit
superior pharmacological and pharmacokinetic parameters, relative to all GlyT1 inhibitors of the spiropiperidine family, culminat-
ing in the identification of 16b with an oral bioavailability of ꢀ60%. In addition, a straightforward two-step procedure for the
assembly of the target molecules is also presented.
Ó 2006 Elsevier Ltd. All rights reserved.
Compelling evidence suggests that an impairment in
NMDA neurotransmission is involved in the pathophys-
iology of schizophrenia.1 This hypothesis originated
from the observation that PCP, a recreational drug
which has been demonstrated to be a potent and selec-
tive NMDA-R blocker, is able to reproduce, in healthy
subjects, the positive, negative and cognitive symptoms
of schizophrenia.2 Thus, therapeutic intervention aimed
at increasing NMDA synaptic tone is expected to show
beneficial effect in schizophrenic patients. As glycine is
known to act as a positive allosteric modulator of the
NMDA receptor,3 one strategy to enhance NMDA
receptor activity is to elevate extracellular levels of gly-
cine in the local microenvironment of the synaptic
NMDA receptor. Glycine elevation can be achieved by
inhibition of the glycine transporter 1 (GlyT1) which is
co-expressed in the brain together with the NMDA
receptor and is responsible for the maintenance of low
extracellular glycine concentrations.4,5 Strong support
for this novel therapeutic approach comes from clinical
studies where glycine6 and sarcosine7 (a prototypical
weak GlyT1 inhibitor) displayed benefits to schizo-
phrenic patients as an add-on to conventional therapy.
As a result, considerable efforts have been focused on
the development of selective GlyT1 inhibitors.8
We have recently introduced a novel series of N-(2-aryl-
cyclohexyl) substituted spiropiperidines 1a,9 1b,10 211
and 312 as a novel class of GlyT1 inhibitors which
H
H
N
O
O
N
N
R1
N
N
(1'R 2'R or 1'S 2'S)
(1'R 2'R or 1'S 2'S)
1a R1 = H
2
1b R1 = OH
H
H
O
N
O
N
4
HO
N
N
Keywords: GlyT1; GlyT2; NMDA; Schizophrenia; Transporter;
Glycine; Spiropiperidine.
1'
*
Corresponding author. Tel.: +41 61 688 50 48; fax: +41 61 688 87
14; e-mail: andrew.thomas@roche.com
4
(1'R 2'R or 1'S 2'S)
3
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.063

4322
D. Alberati et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4321–4325
display excellent selectivity against the GlyT2 isoform.
These series evolved through the stepwise incorporation
of specific chemical mutations within the structures that
allowed us to successfully address all relevant off-target
liabilities with the accomplishment of excellent selectivi-
ties against the l opioid receptor as well as the nocicep-
tin/orphanin FQ peptide (NOP) receptor achieved for
1b, 2 and 3. The main deficiency within these triaza-
and diazaspiropiperidine series was the relatively low
microsomal stability which translated into non-optimal
in vivo pharmacokinetic properties.
nitriles 6. Encouraging the reaction to proceed to com-
pletion within a few hours, by heating, normally resulted
in isolated yields of the desired product 6 of <40%.
Thereafter, reaction of 6 with either a range of commer-
cially available Grignard reagents at room temperature
(step b) or through reaction with Grignard reagents
formed in situ using i-PrMgCl (step c) normally afforded
high yields of the final products 7–27.
As in part of our previously described work we specifi-
cally used rac-5 in the synthesis strategy and preferred
to separate the enantiomers routinely by chiral-phase
HPLC (Chiralpak ADÒ). Table 1 shows the activity at
GlyT1 for the racemate derivatives and also details the
activity for individual enantiomers in the cases where
we decided to separate them with the fastest eluting
component designated a and the slowest eluting desig-
nated b.
Herein we wish to report on our recent efforts13 to dis-
cover and develop a further structurally novel, potent
and selective chemotype of GlyT1 inhibitors 4 which
have inherently improved pharmacological and pharma-
cokinetic properties when compared to 1–3.
The rationale for the design of this novel chemotype 4 of
GlyT1 inhibitors with a goal to improve upon the phar-
macokinetic properties when compared to 1–3 was de-
rived from some proprietary in-house knowledge that
the poor metabolic stability of this class of compounds
was due to oxidative processes taking place resulting
in cleavage of the N–C10 bond. Therefore, we postulated
that relocating the substituents from C20 to the C10 po-
sition, as shown in 4, would offer some protection to the
molecule from rapid oxidative clearance. More impor-
tantly, to explore this hypothesis, we also significantly
simplified the stereocomplexity of the diazaspiropiperi-
dine series moving from three stereocentres in 2 to only
one stereocentre in 4 where we had already been able to
ascertain the absolute configuration of our key diazaspi-
ropiperidine building block 5.14
For exploration of our preliminary SAR (Table 1), we
initially focused on derivatives where R1 = H and F
since these substituents were found to be optimal in all
of the previous series 1–3. In general the SAR parallels
what we earlier established for 1–3 with even the race-
mates, of nearly all derivatives, prepared showing nano-
molar affinity as GlyT1 inhibitors. Without exception,
strongly electron-withdrawing substituents such as CF3
Table 1. In vitro inhibitory activity at the GlyT1 transporter for
compounds 7–27
H
O
N
R2
N
The synthetic strategy to access the target molecules 4
relied on a straightforward two-step procedure involving
a Bruylants amination reaction15 as shown in Scheme 1.
In our hands, in order to achieve, on a preparatively use-
ful scale, useful yields we found it best to allow the first
step of the sequence (Strecker-type reaction with
TMSCN and AcOH) to proceed slowly (up to 5 days
at rt) followed by isolation of the intermediate amino
R1
Compound R1
R2
Ph
GlyT1 EC50 (lM)
a
rac
a
b
7
8
H
H
H
H
H
H
H
H
H
F
0.077
1.212
0.107
6.969
16.00
4-CF3–Ph
4-CN–Ph
4-CF3O–Ph
4-Cl–Ph
4-Me–Ph
3-F–Ph
9
>30
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
4.563
H
O
N
H
O
0.45
0.81
0.480
0.684
26.231
O
N
N
0.284
0.423
5.61
a)
+
N
HN
3-MeO–Ph
3-Thienyl
4-F–Ph
0.682
0.095
R1
0.056
1.474
0.097
0.419
6
R1
F
4-Me–Ph
3-F–Ph
(rac 5)
R1 = H, F (R or S),
Me, CF3 Ref. 14
F
0.393
R2 Br
H
F
3-CN–Ph
3-CF3O–Ph
3-Me–Ph
2-Me–Ph
2-Thienyl
3-Thienyl
Ph
>30
c)
O
N
i-PrMgCl
F
16.728 21.519
4.000
R2
F
0.563
0.350
0.949
b)
N
F
R2 MgBr
F
0.381
0.213
0.281
0.099
F
R1
7-27 (see Table 1)
Me
Me
0.139
0.073
0.433
4-F–Ph
Scheme 1. Synthesis of compounds 7–27. Reagents and conditions: (a)
TMSCN, AcOH, rt, 3–5 days, 50–62%; (b) THF, rt, on, 73–94% or (c)
THF, 0 °C to rt, 2 h, on, 10–75%.
CF3 4-F–Ph
a Radiometric assay using [3H]-glycine.9

D. Alberati et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4321–4325
4323
and CN were not well tolerated as demonstrated by 8, 9
and 19. As we had observed in an earlier series, m- and
p-CF3O, as exemplified in 10 and 20, was not a favoured
substituent and neither was a m-OMe substituent, for
example, 14. Methyl substituents were well tolerated at
all of the p-, m- and o-positions providing low nanomo-
lar GlyT1 inhibitors for the racemates in 12, 17, 21 and
22.
as those shown in 33–35 and 40 were also shown to be
inactive. However, replacement with an i-Pr group (36
and 41) provided the first examples of potent derivatives
of this type within the diazaspiropiperidine class.
An additional structural modification examined was the
replacement of the cyclohexyl unit of 4 as demonstrated
in the preparation of the derivatives 43–45. Interesting-
ly, the cycloheptane unit was well tolerated with 43
showing a potency of 245 nM as an inhibitor at the
GlyT1 transporter and was free of activity (>30 lM)
at the GlyT2 isoform. However, in stark contrast the
cyclobutane ring was not tolerated since 44 showed an
EC50 > 30 lM. Unfortunately, it was not possible to
prepare the corresponding cyclopentane 45 derivative
following the same protocol as described in Scheme 1.
We also examined some small heterocyclic groups and
pleasingly, the 2- and 3-thiophene substituents 15, 23
and 24 provided very potent compounds. In the previ-
ous series, simple Ph, 4-Cl-phenyl 3-F-phenyl and 4-F-
phenyl were optimal substituents based on their affinity
as GlyT1 inhibitors, and this was found to be the case
also as shown for 7, 11, 13, 16, 18, 26 and 27. In fact,
some of these compounds showed good potency as their
individual enantiomers as tabulated for 7a,b, 11a,b,
13a,b and 16a,b. We also briefly explored changes at
R1 by the introduction of Me (25 and 26) and CF3
(27) groups which were also found to provide potent
derivatives.
F
H
N
F
F
O
H
N
H
N
O
O
N
N
N
43
44
45
F
F
F
In order to further expand the SAR opportunities within
this class, we examined the influence of more diverse
changes to R1 and R2 detailed in Table 2. It was found,
in analogy to our previous work,10 that it was possible
to replace the R1 aryl group with an alkyl group. In this
case the prototypical n-Pr group in combination with
the 4-F-phenyl group provided the most potent deriva-
tive 29 even as a mixture of enantiomers. Separation
of 28 into its enantiomers provided two compounds with
an EC50 of 82 and 62 nM at GlyT1. In the previous ser-
ies, the R2 aryl group was essential for good inhibition
at GlyT1. For this new series cyclic aliphatic units such
Following the strategy outlined below in Scheme 2 it
was possible to prepare the pyrano-derivatives 46 as
an enantiomeric mixture (GlyT1 EC50 = 1.419 lM)
and 47 as a mixture of four diastereoisomers which we
managed to separate (GlyT1 EC50 (lM) = a 3.661, b
7.290, c >30 and d 0.525).
With the difficulties we had previously encountered with
1a it was essential to fully establish excellent levels of
selectivity against the GlyT2 isoform as well as the l
opioid receptor and the nociceptin/orphanin FQ peptide
(NOP) receptor. Table 3 nicely illustrates the levels of
selectivity achieved against each of these receptors for
the main molecules of interest. These results are partic-
ularly exceptional bearing in mind the remarkably high
affinity (15 nM) towards the l opioid receptor and the
NOP receptor for the corresponding triazaspiropiperi-
dine derivative 52 prepared in analogy to Scheme 1
using the commercially available 1-phenyl-1,3,8-triaza-
spiro[4.5]decan-4-one 53.
Table 2. In vitro inhibitory activity at the GlyT1 transporter for
compounds 28–42
H
O
N
R2
R1
N
Compound
R1
R2
GlyT1 EC50
(lM) rac
a
Next, we examined the effect on the metabolic stability
in microsomes within this class since this was a key
parameter which remained to be optimized. For exam-
ple, the triazaspiropiperidine derivative 52 was a high
clearance compound resulting in very low predicted
maximum achievable bioavailability (MAB) of 6% and
16% in mouse and human, respectively. As can be seen
Table 4 confirms our hypothesis since the structural
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
n-Pr
n-Pr
n-Pr
n-Pr
Ph
Ph
4-F–Ph
0.161
0.093
0.229
0.113
0.147
15.237
6.287
>30
2-Thienyl
2-Thienyl-5-Me
2-Thienyl
c-Hexyl
c-Pentyl
c-Pr
Ph
Ph
Ph
Ph
i-Pr
0.970
0.979
0.067
0.099
>30
Ph
Me2C@CH
2-Thienyl
3-Thienyl
c-Pr
4-F–Ph
4-F–Ph
4-F–Ph
4-F–Ph
4-F–Ph
H
N
H
N
O
O
N
N
i-Pr
Et
0.787
1.466
N
HN
Scheme 1
a Radiometric assay using [3H]-glycine.9
52
53

4324
D. Alberati et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4321–4325
Table 4. Microsome stability data for compounds 7, 11, 16, 28, 29, 38,
39 and 52
H
N
H
N
O
O
N
b)
a
Compound CLint (mouse CLint (human
a
MAB (%)
46 X = CH2, Y = O
47 Y = O, X = CH2
N
N
microsomes)
microsomes)
Mouse Human
O
Y
X
7
7a
5
0
9
8
86
100
93
76
46
98
45
51
48
57
28
45
6
63
67
50
12
67
84
60
54
47
53
64
78
16
49
a)
O
b)
7b
2
16
114
8
O
11a
16a
16b
28
10
38
1
48
H
H
O
N
O
N
N
a)
O
3
39
30
34
24
80
39
470
10
13
18
14
9
N
O
HN
28a
28b
29
O
(rac 5)
50
51
F
38
39
Scheme 2. Synthesis of compounds 46 and 47. Reagents and condi-
tions: (a) TMSCN, AcOH, rt, 1 day, 25–30%; (b) PhMgBr, THF,
reflux, 1 h, 30–32%.
4
82
52
a CLint, intrinsic clearance (lL/min/mg protein).
Table 3. In vitro inhibitory activity at the GlyT1 and the GlyT2
transporters and potency in inhibiting the NOP and l receptors for
compounds 7, 11–13, 16–18, 21, 22, 28, 29 and 52
c
Table 5. Selected pharmacokinetic parameters for compound 16b
Compound
F (%)
CLa
T0.5 (h)
Vss (L/kg)
B/P
Compound GlyT1
GlyT2
NOP
l IC50
16bb
60
35
2.8
7.6
3.0
a
a
b
EC50 (lM) EC50 (lM) IC50 (lM) (lM)
a CL, total clearance (mL/min/kg protein).
b Determined in mouse, B/P, brain to plasma ratio; Vss, volume of
distribution.
7a
7b
0.077
0.107
0.45
>30
>30
>30
>30
>30
>30
>30
>30
>30
>30
>30
>30
>30
>30
>30
—
>10
6.15
5.5
7.64
11a
11b
12a
12b
13a
13b
16
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
—
19.8
0.81
0.284
0.48
3.60
9.12
and 16 where we had demonstrated levels of microsomal
stability for both species (human and mouse) in the de-
sired range.
11.9
0.423
0.684
0.095
0.056
0.097
0.393
0.563
0.949
0.350
0.161
0.082
0.062
0.093
0.072
8.04
13.50
7.80
We then examined our most favoured derivative 16b17 in
a single dose in vivo pharmacokinetic study as a repre-
sentative of the 4-aryl-8-(1-phenyl-cyclohexyl)-2,8-dia-
za-spiro[4.5]decan-1-one series (Table 5). As expected
the excellent parameters obtained in vitro resulted in
the highest achieved oral bioavailability (F) of 60% for
all the spiropiperidine classes we have examined which
was also concomitant with excellent levels of brain
penetration.
16a
16b
18
>10
7.40
7.18
21
7.83
21b
22
24.83
3.851
2.76
28
28a
28b
29
—
—
—
—
2.73
1.87
—
>30
—
3.03
0.015
52
0.015
Since the derivatives of type 4 have structural similarity
with PCP we have also demonstrated that they do not
have ‘PCP-like’ behaviour by examining 7, 7a, 7b and
16b in a displacement experiment with H-MK801 in
NMRI mouse forebrain membranes where an
a Radiometric assay using [3H]-glycine.9
b Displacement of [3H]-NOP in membranes prepared from perma-
nently transfected HEK293 cells expressing hNOP receptors.16
c Displacement of [3H]-naloxone in membranes prepared from BHK
cells transiently expressing hl receptors.16
3
IC50 > 30 lM was obtained for each.
changes made in the development of 4 led to compounds
with superior microsomal stability relative to all previ-
ous series. In close examination we can conclude that
Moreover, derivatives from the 4-aryl-8-(1-phenyl-cy-
clohexyl)-2,8-diaza-spiro[4.5]decan-1-one class 4 have
shown no significant inhibition of the hERG K channel
as outlined in Table 6. Notably, we were able to dimin-
ish affinity towards the hERG K channel by incorpora-
tion of a b-oxygen atom as revealed in the structures 46
and 47 which closely follows the relationship12b we gen-
erated in the design of the GlyT-1 inhibitors 3 also free
of hERG K channel activity.
the
4-propyl-8-(1-phenyl-cyclohexyl)-2,8-diaza-spiro-
[4.5]decan-1-one derivatives 28 and 29 displayed similar
MAB values (60% and 47%, respectively, in human
microsomes and 45% and 48%, respectively, in mouse
microsomes), compared to their triazaspiropiperidine
congeners.10 However, the best improvements were ob-
served in the 4-aryl-8-(1-phenyl-cyclohexyl)-2,8-diaza-
spiro[4.5]decan-1-one derivatives with the microsomal
data for nearly all derivatives examined predicting
MAB values generally superior to all our previous series.
In particular, we were encouraged by the derivatives 7
In conclusion, our work in the spiropiperidine class has
culminated in the discovery of a novel, potent and selec-
tive chemotype of GlyT1 inhibitors 4. We stepwise
designed the 4-substituted-8-(1-phenyl-cyclohexyl)-2,8-

D. Alberati et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4321–4325
4325
Table 6. In vitro inhibitory activity at the hERG K channel for
compounds 12c, 27 and 28
2. Jentsch, J. D.; Roth, R. H. Neuropsychopharmacology
1999, 20, 201.
a
b
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 Discovery Dev.
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.
Compound
HERG IC50 (lM)
pKa
16b
46
1.300
>11.000
>14.000
8.75
7.1
47
7.71
a Inhibition of hERG K channel determined by whole-cell patch-clamp
experiments on a transfected CHO cell line.
b Determined by potentiometric titration in a MeOH:water mixture at
rt with the GLpKa instrument from Sirius Analytical Instruments.
7. Tsai, G.; Lane, H.-Y.; Yang, P.; Chong, M.-Y.; Lange, N.
Biol. Psychiatry. 2004, 55, 452.
8. 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.
9. (a) Ceccarelli, S. M.; Pinard, E.; Stalder, H. WO Patent
2005040166, 2005; Chem. Abstr. 2005, 142, 447121; (b)
Pinard, E.; Ceccarelli, S. M.; Pinard, E.; Stalder, H.;
Alberati, D. Bioorg. Med. Chem. Lett. 2006, 16, 349.
10. Ceccarelli, S. M.; Pinard, E.; Stalder, H.; Alberati, D.
Bioorg. Med. Chem. Lett. 2006, 16, 354.
11. Alberati, D.; Ceccarelli, S. M.; Jolidon, S.; Krafft, E. A.;
Kurt, A.; Maier, A.; Pinard, E.; Stalder, H.; Studer, D.;
Thomas, A. W.; Zimmerli, D. Bioorg. Med. Chem. Lett.,
in press, doi:10.1016/j.bmcl.2006.05.064.
12. (a) Alberati, D.; Ceccarelli, S. M.; Jolidon, S.; Pinard, E.;
Thomas, A. W. US Patent 2005154001, 2005; Chem.
Abstr. 2005, 143, 115462.; (b) Alberati, D.; Jolidon, S.;
Krafft, E. A.; Kurt, A.; Maier, A.; Pinard, E.; Thomas, A.
W.; Zimmerli, D. Bioorg. Med. Chem. Lett., in press,
doi:10.1016/j.bmcl.2006.05.058.
13. Jolidon, S.; Pinard, E.; Thomas, A. W. US Patent
2005154000, 2005; Chem. Abstr. 2005, 143, 115461.
14. Krafft, E. A.; Kurt, A.; Maier, A.; Thomas, A. W.;
Zimmerli, D. Synthesis 2005, 3245.
diaza-spiro[4.5]decan-1-one class to display excellent
selectivities against the l opioid receptor as well as the
nociceptin/orphanin FQ peptide (NOP) receptor. In
addition, this work has described in detail our strategy
to improve upon the pharmacological properties and
pharmacokinetic properties of the diazaspiropiperidine
series by specific chemical mutations throughout the
development of the class. Compound 16b has demon-
strated for the first time that GlyT1 inhibitors of the spi-
ropiperidine family can achieve high oral bioavailability.
Acknowledgments
We thank Markus Bender, Serge Burner, Axel Maier,
Eva Krafft, Thierry Meyer and Deborah Studer for
additional technical assistance. Drs. Holger Fischer
and Bjoern Wagner are also thanked for the measure-
ment of pKa values. Dr. Simona Ceccarelli is also
thanked for helpful discussions during the preparation
of the manuscript and Dr. Thomas Woltering is thanked
for useful discussions at the onset of this work.
15. Bruylants, P. Bull. Soc. Chim. Belg. 1924, 33, 467.
16. Wichmann, J.; Adam, G.; Roever, S.; Hennig, M.;
Scalone, M.; Cesura, A. M.; Dautzenberg, F. M. Eur. J.
Med. Chem. 2000, 35, 839.
References and notes
17. Compound 16b was prepared following the two-step
Bruylants amination route established in Scheme 1 start-
ing from (R)-5.15
1. For a recent review, see: Millan, M. J. Psychopharmacol-
ogy 2005, 179, 30.