Discovery of benzoylisoindolines as a novel class of potent, selective and orally active GlyT1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2010.09.124

Benzoylisoindolines were discovered as a novel structural class of GlyT1 inhibitors. SAR studies and subsequent lead optimization efforts focused primarily on addressing hERG liability and on improving in vivo efficacy resulted in the identification of potent GlyT1 inhibitors displaying excellent selectivity and in vivo PD and PK profiles.

Full text of DOI:10.1016/j.bmcl.2010.09.124

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6960–6965
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of benzoylisoindolines as a novel class of potent, selective
and orally active GlyT1 inhibitors
⇑
Emmanuel Pinard , Daniela Alberati, Markus Bender, Edilio Borroni, Virginie Brom, Serge Burner,
Holger Fischer, Dominik Hainzl, Remy Halm, Nicole Hauser, Synèse Jolidon, Judith Lengyel,
Hans-Peter Marty, Thierry Meyer, Jean-Luc Moreau, Roland Mory, Robert Narquizian, Roger D. Norcross,
Philipp Schmid, Roger Wermuth, Daniel Zimmerli
F. Hoffmann-La Roche Ltd, Pharmaceutical Research Basel, CH-4070 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Benzoylisoindolines were discovered as a novel structural class of GlyT1 inhibitors. SAR studies and
subsequent lead optimization efforts focused primarily on addressing hERG liability and on improving
in vivo efficacy resulted in the identification of potent GlyT1 inhibitors displaying excellent selectivity
and in vivo PD and PK profiles.
Received 27 August 2010
Revised 24 September 2010
Accepted 25 September 2010
Available online 7 October 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
GlyT1
Inhibitor
Schizophrenia
Benzoylisoindoline
hERG
Both clinical observations and preclinical studies suggest that
hypofunction of the N-methyl- -aspartate (NMDA) receptor is
identification of RG1678 (7)14 which is the first potent and
selective GlyT1 inhibitor to demonstrate a beneficial effect in
schizophrenic patients in a phase II clinical trial. Following the
identification of RG1678, further medicinal chemistry effort at
F. Hoffmann-La Roche concentrated on studying the structural
diversity allowed at the western arylpiperazine site of RG1678.
Towards this end, keeping the eastern 5-methanesulfonyl-2-((S)-
2,2,2-trifluoro-1-methyl-ethoxy)-benzoyl moiety of RG1678 in
place, a focused screen of cyclic amine systems selected from the
Roche compound depository was performed.
This exercise led to the identification of 5-chloro-substituted-
benzoylisoindoline compound 8 as a fairly potent GlyT1 inhibitor
(135 nM) showing more than 200-fold selectivity against the type
2 isoform (Fig. 2a). Further characterization revealed that 8 pos-
sessed medium lipophilicity, excellent membrane permeability,
medium to low in vitro clearance in mouse and human liver micro-
somes and no significant inhibition of the major drug metabolizing
CYP450 enzymes. In addition, despite its low aqueous solubility in
the lyophylisated solubility assay (LYSA),15 8 proved to be orally
active (ID50 = 20 mg/kg) in the mouse L-687,414-induced hyperlo-
comotion assay, a novel and straightforward functional screening
model recently introduced by us which allows the detection of
the in vivo activity of GlyT1 inhibitors (Fig. 2b).14 The promising
in vitro and in vivo results obtained with benzoylisoindoline com-
pound 8 prompted us to study in detail this novel class of GlyT1
D
implicated in the pathophysiology of schizophrenia.1 Thus, thera-
peutic intervention aiming at restoring NMDA receptor activity
represents a promising novel strategy for the management of
schizophrenia. As glycine is an obligatory co-agonist at the NMDA
receptor complex,2 one strategy to enhance NMDA receptor activ-
ity is to elevate extracellular levels of glycine in the brain through
selective inhibition of glycine uptake mediated by the glycine
transporter-1 (GlyT1) which is co-expressed with the NMDA recep-
tor.3 Strong support for this approach comes from clinical studies
where glycine and D-serine (co-agonists at the glycine site of the
NMDA receptor) and sarcosine (a prototypical weak GlyT1 inhibi-
tor) improved positive, negative and cognitive symptoms in
schizophrenic patients, when added to conventional therapy.4 As
a result, considerable efforts have been focused on the develop-
ment of selective GlyT1 inhibitors.5 The first examples reported
were sarcosine derivatives including 1,6 2,7 and 3.8 More recently,
non amino-acid chemotypes9 like 4,10 5,11 and 6,12 have been
described (Fig. 1). We have recently disclosed our lead
optimisation effort in a benzoylpiperazine class13 that led to the
⇑
Corresponding author. Tel.: +41 61 688 43 88; fax: +41 61 688 87 14.
E-mail address: emmanuel.pinard@roche.com (E. Pinard).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.09.124

E. Pinard et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6960–6965
6961
O
OH
N
N
CF3
O
O
S
O
O
N
N
Cl
MeO
OH
OH
F
3: Lu-AA20465
Lundbeck
2: Org-24598
Organon
1: ALX-5407
Allelix
Cl
O
N
N
H
N
O
F
N
H
H
HN
O
N
Cl
Cl
Cl
S
H
O
CF3
O
N
4: SSR504734
Sanofi
5: DCCCyB
Merck
6: PF-03463275
Pfizer
CF3
O
O
N
N
N
F
SO2Me
F3C
7: RG1678
Hoffmann-La Roche
Figure 1. A selection of published GlyT1 inhibitors.
inhibitors.16 In this Letter, we report on the SAR results and on our
lead optimization effort that led to the identification of the potent,
selective and orally active GlyT1 inhibitors 19, 25 and 27.
Benzoylisoindoline derivatives 8–3416 were synthesized under
standard amide coupling conditions from isoindolines 35 and ben-
zoic acids 3613,14,16 (Scheme 1).
The novel isoindoline building blocks 35a–k were prepared fol-
lowing the synthetic routes depicted in Schemes 2 and 3. Trifluoro-
methyl-substituted dihydro-pyrrolo pyridines 35a–c (Scheme 2)
were obtained upon acidic cleavage of trityl protected precursors
39a–c, themselves prepared from diols 38a–c after a bis-chlorina-
tion or -mesylation followed by cyclization in the presence of tri-
tylamine. Diols 38a–b were obtained by reducing known lactones
37a–b17 and diol 38c was synthesized after reduction of di-methyl
ester 41, itself prepared by oxidation of trifluoromethylquinoline
40 in the presence of ruthenium chloride and periodic acid as de-
scribed by Lee and co-workers.18
hGlyT1 (EC50, µM) = 0.135
hGlyT2 (EC50, µM) = 30
CF3
O
O
N
clogP = 3.4
Pe (Pampa, 10-6 cm/s) = 4.4
Solubility (µg/mL)13 = <1
4
SO2Me
5
Cl
Human CLint (µL/min/mg protein) = 2.4
Mouse CLint (µL/min/mg protein) = 23.3
8
CYP450 3A4, 2D6, 2C9 (IC50, uM) = >38
L-687,414 (50 mg/kg s.c.)
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
##
**
O
R2
O
R2
X
Z
HO
N
Y
(a)
X
R1
NH
3
10
30 30
Y
+
R1
Veh
Compound 8 (mg/ kg p.o.)
Z
SO2Me
SO2Me
Figure 2. (a) Structure of benzoylisoindoline hit 8 and some key in vitro charac-
teristics; (b) Dose-dependent inhibition of L-687,414-induced hyperlocomotion by
8 in mice. The data represent mean horizontal activity counts per group recorded
over a 60-min time period; error bars indicate S.E.M. (n = 8 per group). ##, p <0.01
35
36
8-34
Scheme 1. Synthesis of benzoylisoindolines 8–34. Reagents and conditions:
(a) TBTU, DIPEA, DMF, rt, 60–95%.
**
versus vehicle (Veh) alone; , p <0.01 versus L-687,414 alone.

6962
E. Pinard et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6960–6965
OH
OH
to low in vitro clearances in mouse, rat and human liver micro-
somes (data not shown).
O
X
Z
(a)
X
Z
X
Z
(d)
Y
Y
Y
O
N
Tr
In addition, fully in agreement with their increased GlyT1 affin-
ities compared to the initial hit compound 8, analogues 10 and 11
demonstrated an improved in vivo potency in the L-687,414-in-
duced hyperlocomotion assay with an ID50 reaching 2 mg/kg for
10 and 6 mg/kg for 11 after po administration (Table 1).
In a single dose PK experiment performed in rat, the most
in vivo active compound 10 exhibited low systemic clearance
(1.7 mL/min/kg), medium volume of distribution (2.9 L/kg) and
good brain penetration (brain/plasma = 1). However, 10 showed a
fairly low oral bioavailability of 23%, a result we attributed to its
F3C
F3C
F3C
37a-b
38a-c
39a-c
a: X,Y, Z = N, CH, CH
b: X,Y, Z = CH, N, CH
c: X,Y, Z = CH, CH, N
(c)
(e)
(b)
CO2Me
CO2Me
X
Z
Y
NH
F3C
N
F3C
N
F3C
low aqueous solubility (1 lg/mL) and hence limited absorption.
Further profiling of 10 and 11 in our battery of in vitro safety tests
revealed, in addition, that both compounds had significant inhibi-
tory activities at the hERG potassium channel of 59% and 67%,
35a-c
40
41
Scheme 2. Synthesis of isoindoline intermediates 35a–c. Reagents and conditions:
(a) NaBH4, EtOH, rt, 74–97%; (b) RuCl3ÁH2O, HIO4, H2O, CH3CN, CCl4, rt then Cs2CO3,
MeI, rt, 28%; (c) NaBH4, CaCl2, MeOH, 0 °C to rt, 63%; (d) SOCl2, CH2Cl2, 0 °C to rt (for
38a–b) or MsCl, Et3N, DMAP, CH2Cl2, 0 °C (for 38c) then TrNH2, DIPEA, DMF, 60 °C,
36–51%; (e) TFA, MeOH, CHCl3, rt, 50–90%.
respectively, at a concentration of 10 lM (Table 1).
Therefore, starting from leads 10 and 11, our next step aimed
firstly at improving the oral in vivo potency of our compounds
by increasing the solubility and secondly, at minimizing the hERG
inhibitory activity. We reasoned that both objectives could be
achieved simultaneously by designing compounds having de-
creased lipophilicities. Towards this end, a nitrogen atom insertion
scan was performed on the western aromatic ring of compound 10
and on the 5-phenyl group of 11. This approach delivered the dihy-
dro-pyrrolo-pyridines: 19, 20, 21 and the 2- and 4-pyridines 22
and 23 (Table 1). From this novel set of compounds which exhib-
ited, as expected, a reduced c log P (in the range of 2.2–3.3) and
Isoindolines 35d–k carrying aryl, heteroaryl, cyclic amino or
tetrahydropyranyl moieties in position 5, were prepared from pro-
tected 5-bromo or 5-iodo substituted isoindolines 42 under Suzuki,
Buchwald–Hartwig and Stille conditions (scheme 3).
Our first effort aimed at establishing the SAR at positions 5 and
4 of the western aromatic moiety (Table 1). Starting from 8, dele-
tion of the 5-chloro group leading to compound 9 resulted in a se-
an increased aqueous solubility (25–470 lg/mL), only dihydro-
pyrrolo-pyridine 19 and 4-pyridine 23 showed good affinities at
GlyT1 of 50 and 17 nM, respectively. In the L-687,414 assay, both
compounds, as hoped, exhibited an improved in vivo efficacy over
their respective parent molecules 10 and 11. The improvement was
much more pronounced for 19 for which an ID50 of 0.4 mg/kg after
oral dosing was reached. Pyridine 23 achieved a more moderate
but still favorable potency having an ID50 of 3 mg/kg. Regarding
hERG activity, again, the best results were obtained with com-
pound 19 for which only a 12% block was measured at 10 lM, thus
representing a fivefold reduction in hERG affinity over parent com-
pound 10. Pyridine 23 showed only a slight decrease of hERG po-
tency with a measured 58% block at 10 lM compared to 67% for
vere decrease of GlyT1 affinity (2.54 lM). In contrast, a significant
improvement of activity was observed upon replacing the 5-chloro
substituent with larger electron withdrawing or neutral groups as
seen with the trifluoromethyl (10) and phenyl (11) derivatives,
both showing a GlyT1 activity in the low nanomolar range (21
and 13 nM). Much weaker potencies were however measured with
analogues carrying electron donating groups such as methoxy (12,
180 nM), pyrrolidinyl (13, 180 nM) and morpholinyl (14, 275 nM).
A further reduced activity was observed with compounds bearing a
basic nitrogen containing group as exemplified with the N-methyl-
piperazinyl derivative 15 (2.08 lM). At the position 4 of the isoind-
oline core, a rapid scanning revealed that substituents were always
poorly tolerated. For example, methyl (16), methoxy (17) and ni-
trile (18) analogues showed GlyT1 affinities only in the micromolar
or high nanomolar range. The two highly potent analogues 10 and
11 were selected for further evaluation. Both compounds exhibited
excellent selectivity versus the GlyT2 isoform (like all active ana-
logues in this series), good membrane permeability, and medium
11. Reasoning that the pronounced aromatic character of the wes-
tern part was possibly contributing to the hERG activity of 23,19
subsequent lead optimisation efforts focused on preparing ana-
logues incorporating
a saturated heterocyclic system on the
isoindoline moiety such as the 2- and 4-tetrahydropyranyl (THP)
group (compounds 24 and 25, respectively). This strategy proved
1) conditions a-d
N
R
NH
R1
X
2) conditions e-g
35d-k
42
n0
35d
35e
35f
35g
35h
35i
35j
35k
N
O
N
N
R1
N
N
O
N
O
X
R
I
I
I
Br
Boc
c, f
Br
Boc
c, f
Br
Boc
c, f
Br
Benzyl
d, g
Br
Boc
d, g, e
64%
Boc
a, e
50%
Boc
b, e
69%
Boc
b, e
67%
Conditions
Yield from 42
75%
63%
71%
13%
Scheme 3. Synthesis of isoindoline intermediates 35d–k. Reagents and conditions: (a) PhB(OH)2, K2CO3, cat. Pd(PPh3)4, DMF, 120 °C; (b) 2-or 4-tributylstannylpyridine, cat.
Pd2dba3, cat. AsPh3, CuI, DMF, 90 °C; (c) cyclic amine, cat. Pd2dba3, NaOtBu, 2-(dicyclohexylphosphino)biphenyl, toluene, 120 °C; (d) 2- or 4-dihydropyran-yl-tributylstanne,
cat. Pd2dba3ÁCHCl3, cat. AsPh3, cat. LiCl, DMF, 100 °C; (e) TFA, CH2Cl2, rt; (f) HCl in dioxane (4 N), dioxane, 90 °C; (g) cat Pd/C, ammonium formate, MeOH, 65 °C.

E. Pinard et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6960–6965
6963
Table 1
In vitro and in vivo profiles of 9–25
CF3
O
O
R1
SO2Me
Solubilityc
g/ml)
Pampa Ped
Reversal of L-687,
414-induced
hERG % block at 10 l
Mf
a,b
a,b
Compound
R1
GlyT1 EC50
M)
GlyT2 EC50
M)
c log P
(
l
(
l
(l
hyperlocomotion in
e
mouse ID50 (mg/kg)
N
9
2.54
>30
>30
2.70
3.58
<1
1
0.8
0.7
—
2
—
N
10
0.021
59
F3C
N
11
0.013
12
4.58
4
2.6
6
67
N
12
13
0.180
0.180
>30
2.62
2.98
46
<1
2.7
0.8
—
—
—
—
MeO
N
14.3
N
N
14
15
16
17
18
19
0.275
2.08
>30
>30
>30
>30
>30
>30
2.15
2.60
3.20
2.62
2.13
2.23
355
490
73
4.3
1.5
4.2
4.8
0.8
5
—
—
—
—
—
—
12
N
O
N
—
N
N
N
N
N
3.36
1.94
0.77
—
Me
85
—
OMe
CN
259
280
—
N
N
0.050
0.4
F3C
F3C
A
N
N
N
20
21
0.132
0.250
>30
>30
2.23
2.23
470
429
3.9
5.5
—
—
—
—
F3C
N
N
N
22
23
24
25
0.084
0.017
0.205
0.014
>30
>30
>30
>30
3.30
3.09
3.07
2.91
30
25
<1
76
0.8
4.2
0.9
3.3
—
—
N
3
58
—
N
O
N
—
N
0.5
23
B
O
a
b
c
d
e
f
EC50 values are the average of at least two independent experiments.
[3H]-glycine uptake inhibition assay in cells transfected with hGlyT114 or hGlyT214 cDNAs.
Solubility measured in a lyophilisation solubility assay (LYSA).15
Pe: PAMPA permeation constant (10À6 cm/s) through artificial membranes.22
Compound given orally, n = 8.
Patch clamp assay, values are the average of at least two independent experiments.

6964
E. Pinard et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6960–6965
Table 2
In vitro and in vivo profiles of 26–34
O
R2
N
R1
N
N
R1 =
F3C
O
SO2Me
B
A
a,b
a,b
Compound
R1
R2
GlyT1 EC50
M)
GlyT2 EC50
M)
c log P
Solubilityc
g/ml)
Pampa Ped
Reversal of L-687,414-induced
hyperlocomotion in
mouse ID50 (mg/kg)
hERG% block
(
l
(
l
(
l
at 10 l
Mf
e
26
27
A
B
0.070
>30
1.97
114
4.1
1
10
O
0.028
>30
2.65
191
4.2
0.5
10
O
O
28
29
A
A
0.030
0.111
>30
>30
2.59
1.90
445
513
5.5
4.3
—
3
37
—
CF3
CF3
O
O
O
30
31
B
A
0.137
0.805
>30
>30
2.59
2.90
334
72
4.1
0.8
—
—
—
—
F
32
B
0.016
>30
3.76
15
1.4
0.3
36
F
F
33
34
A
A
0.030
3.0
>30
>30
3.15
0.73
42
6.4
0.9
1
38
—
O
N
495
—
a
b
c
d
e
f
EC50 values are the average of at least two independent experiments.
[3H]-glycine uptake inhibition assay in cells transfected with hGlyT114 or hGlyT214 cDNAs.
Solubility measured in a lyophilisation solubility assay (LYSA).15
Pe: PAMPA permeation constant (10À6 cm/s) through artificial membranes.22
Compound given orally, n = 8.
Patch clamp assay, values are the average of at least two independent experiments.
successful since the 4-THP derivative 25, displaying an excellent
GlyT1 affinity (14 nM) showed more than twofold reduction in
series from which RG1678 emerged,13,14 more extended substitu-
ents such as the 3,3,3-trifluoro-propoxy (29 and 30) or benzyloxy
(31) groups were less well tolerated. The attachment of fluoro-
substituted phenyl groups delivered compounds with excellent
activities as exemplified with compounds 32 (16 nM) and 33
(30 nM). Finally, an abrupt drop of GlyT1 activity was observed
upon introduction of cyclic amine systems like the morpholine
group (34). Among the set of potent GlyT1 compounds thus iden-
tified, the most favorable hERG profiles were obtained with the iso-
propoxy anologues 26 and 27 for which only a 10% block was
hERG activity (23% block at 10
23. In addition, this compound exhibited an improved aqueous sol-
ubility (76 g/mL) and an excellent in vivo potency with an ID50 of
lM) over the parent compound
l
0.5 mg/kg in the L-687,414-induced hyperlocomotion assay. In
contrast to 25, the 2-THP isomer 24 showed a fairly low GlyT1
activity (205 nM) and was therefore not profiled further.
In summary, from the SAR and optimisation studies performed
on the western aromatic side of our benzoylisoindoline class, the
dihydro-pyrrolo-pyridine fragment A and the 4-THP-isoindoline
fragment B (see Table 1) emerged as two excellent scaffolds. In-
deed, their combination with the 5-methanesulfonyl-2-((S)-2,2,2-
trifluoro-1-methyl-ethoxy)-benzoyl moiety on the eastern side re-
sulted in the two potent and selective GlyT1 inhibitors 19 and 25
showing balanced molecular properties, excellent in vitro selectiv-
ity and good in vivo potency. Keeping these two newly identified
scaffolds in place, our next effort was directed at exploring the
SAR at the position 2 of the eastern 5-methanesulfonyl-substi-
tuted-benzoyl moiety. Towards this end, a set of diverse substitu-
ents was evaluated including alkoxy groups of varied chain length
and steric bulk, aromatic groups and amine systems. Results
obtained with selected examples are presented in Table 2. With
isopropoxy (26 and 27) and isobutyloxy (28) analogues, good to
excellent GlyT1 affinities of respectively 70, 28 and 30 nM were
obtained. As already observed in the parent benzoylpiperazine
measured at a concentration of 10 lM. The more extended isobu-
tyloxy derivative 28 and the more lipophilic fluoro-substituted
Table 3
Pharmacokinetics properties of 19, 25 and 27 in rata
Compound
19
25
27
iv dose (mg/kg)
CL (mL/min/kg)
Vss (L/kg)
po dose (mg/kg)
Bioavailability F (%)
T1/2 (h)
2.0
1.7
1.4
2
55
6.0
0.2
2.2
21
3.7
6.8
65
1.7
22
2.0
5.7
40
3.2
0.6
1.6
0.6
Brain/plasma
a
Values are an average of two independent experiments.

E. Pinard et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6960–6965
6965
Table 4
Effect of 19, 25, 32 on the extracellular glycine levels in rats striatum at 10 mg/kg poa and their plasma and brain exposures.
Compound
Max. fold inc. of glycine
Fold inc. of glycine at 3 h
Plasmab (ng/mL)
Brainb (ng/mL)
Brainb/GlyT1 EC50
19
25
27
1.8
2.1
1.9
1.8
2.0
1.5
2500
480
279
500
274
159
20
39
13
a
n = 6.
b
Measured at 3 h post dosing.
phenyl compounds 32 and 33 showed more than threefold higher
hERG activities.20 Gratifyingly, in addition to their favorable hERG
profiles, isopropoxy substituted compounds 26 and 27 exhibited
good to excellent in vivo potencies with ID50 values of respectively
1 and 0.5 mg/kg after oral dosing.
The compounds showing the highest in vivo potencies: the
dihydro-pyrrolo-pyridine 19 and the two THP-substituted deriva-
tives 25 and 27 were selected for further evaluation. Pleasingly,
in a CEREP selectivity screen performed against a panel of 92 tar-
gets including transmembrane and soluble receptors, enzymes,
ion channels and monoamine transporters,21 all three analogues
Acknowledgments
We would like to thank Damian Beck, Mesut Cakmak, Daniel
Hafner and Debora Studer for their dedicated technical assistance.
References and notes
1. For a recent review see: Javitt, D. C. Int. Rev. Neurobiol. 2007, 78, 69.
2. Danysz, W.; Parsons, C. G. Pharmacol. Rev. 1998, 50, 597.
3. (a) Sur, C.; Kinney, G. G. Curr. Drug Targets 2007, 8, 643; (b) Thomsen, C. Drug
Discovery Today 2006, 3, 539.
4. Shim, S. S.; Hammonds, M. D.; Kee, B. S. Eur. Arch. Psychiatry Clin. Neurosci. 2008,
258, 16.
5. For recent reviews see: (a) Hashimoto, K. Open Med. Chem. J. 2010, 4, 10; (b)
Javitt, D. C. Curr. Opin. Drug Discov. Dev. 2009, 12, 468; (c) Harsing, L. G.; Juranyi,
Z.; Gacsalyi, I.; Tapolcsanyi, P.; Czompa, A.; Matyus, P. Curr. Med. Chem. 2006,
13, 1017; (d) Lindsley, C. W.; Wolkenberg, S. E.; Kinney, G. G. Curr. Top. Med.
Chem. 2006, 6, 1883.
6. 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.
7. 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.
8. Smith, G.; Ruhland, T.; Mikkelsen, G.; Andersen, K.; Christoffersen, C. T.;
Alifrangis, L. H.; Mork, A.; Wren, S. P.; Harris, N.; Wyman, B. M.; Brandt, G.
Bioorg. Med. Chem. Lett. 2004, 14, 4027.
9. Wolkenberg, S. E.; Sur, C. Curr. Top. Med. Chem. 2010, 10, 170.
10. Dargazanli, G.; Estenne Bouhtou, G.; Magat, P.; Marabout, B.; Medaisko, F.;
Roger, P.; Sevrin, M.; Veronique, C. FR 2838739, 2003; Chem. Abstr. 2003, 139,
337891.
11. Blackaby, W. P.; Lewis, R. T.; Thomson, J. L.; Jennings, A. S. R.; Goodacre, S. C.;
Street, L. J.; MacLeod, A. M.; Pike, A.; Wood, S.; Thomas, S.; Brown, T. A.; Smith,
A.; Pillai, G.; Almond, S.; Guscott, M. R.; Burns, H. D.; Eng, W.; Ryan, C.; Cook, J.;
Hamill, T. G. ACS Med. Chem. Lett., 2010. ASAP.
demonstrated a highly selective profile (<20% inhibition at 10 lM
measured for all targets). In single dose pharmacokinetic studies
in rats (Table 3), 19 exhibited a low systemic clearance. In contrast,
an intermediate clearance was measured for the THP derivatives
25 and 27. These results correlated well with the intrinsic clear-
ances measured in rat microsomes which were low for 19 (Clint
1.9 L/min/mg protein) and medium for 25 and 27 (13.9 and
18.9 L/min/mg protein, respectively). The three compounds
:
l
l
showed intermediate volumes of distribution and good oral bioav-
ailabilities (40–65%). The brain penetration was good for 25 and 27
(brain/plasma: 0.6) and more moderate for 19 (brain/plasma: 0.2).
The terminal half-lives for all three compounds were in the favor-
able range, although significantly shorter for the more rapidly
cleared THP derivatives 25 and 27 (3.2 and 1.6 h, respectively) than
for the dihydro-pyrrolo-pyridine 19 (6 h).
This favorable data prompted us to evaluate the effect of com-
pounds 19, 25 and 27 on the extracellular level of glycine in rat
striatum in a microdialysis study (Table 4).
We were delighted to observe that, at an oral dose of 10 mg/kg,
the three compounds produced a robust glycine increase in the
range of 1.8 to 2.1-fold over basal levels. After 3 h, at which time
the extracellular fluid sampling was stopped and the animal sacri-
ficed, the PD effect remained strong for compounds 19 and 25 (1.8-
and 2.0-fold, respectively), but was significantly diminished for 27
(1.5-fold), a result fully in agreement with the half-lives measured
in plasma for the three compounds in PK studies (vide supra). The
PD effect at 3 h post dosing correlated well with the brain level cor-
rected by the in vitro GlyT1 potencies of the tested compounds
(Table 4).
In summary, we report here on the discovery of benzoylisoind-
olines, a novel structural class of GlyT1 inhibitors originating from
our previously reported benzoylpiperazine series. From the initial
compound 8, the SAR and the lead optimization effort, focused pri-
marily on addressing the hERG liability and on improving in vivo
efficacy resulted in the identification of potent GlyT1 inhibitors
such as 19, 25 and 27 displaying excellent overall in vitro and
in vivo profiles.
12. Lowe, J. A.; Hou, X.; Schmidt, C.; David, T. F.; McHardy, S.; Kalman, M.;
DeNinno, S.; Sanner, M.; Ward, K.; Lebel, L.; Tunucci, D.; Valentine, J. Bioorg.
Med. Chem. Lett. 2009, 19, 2974.
13. Pinard, E.; Alberati, D.; Borroni, E.; Fischer, H.; Hainzl, D.; Jolidon, S.; Moreau, J.
L.; Narquizian, R.; Nettekoven, M.; Norcross, R. D.; Stalder, H.; Thomas, A. W.
Bioorg. Med. Chem. Lett. 2008, 18, 5134.
14. Pinard, E.; Alanine, A.; Alberati, D.; Bender, M.; Borroni, E.; Bourdeaux, P.;
Brom, V.; Burner, S.; Fischer, H.; Hainzl, D.; Halm, R.; Hauser, N.; Jolidon, S.;
Lengyel, J.; Marty, H. P.; Meyer, T.; Moreau, J. L.; Mory, R.; Narquizian, R.;
Nettekoven, M.; Norcross, R. D.; Puellmann, B.; Schmid, P.; Schmitt, S.; Stalder,
H.; Wermuth, R.; Wettstein, J. G.; Zimmerli, D. J. Med. Chem. 2010, 53, 4603.
15. Solubility was measured in
a lyophilisation solubility assay: Compound
initially in DMSO solution is lyophilized then dissolved in 0.05 M phosphate
buffer (pH 6.5), stirred for 1 h and shaken for 2 h. After one night, the solution
is filtered and the filtrate analyzed by direct UV measurement or by HPLC-UV.
16. Jolidon, S.; Narquizian, R.; Norcross, R. D.; Pinard, E. WO Patent 2006082001;
Chem. Abstr. 2006, 145, 230529.
17. Goble, S. D.; Pasternak, A.; Mills, S. G.; Zhou, C.; Yang, L. WO Patent
2004082616; Chem. Abstr. 2004, 141, 314341.
18. Cho, K.-N.; Park, M.-S.; Shim, Y.-K.; Lee, K.-I. B. Korean Chem. Soc. 2002, 23, 1830.
19. Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G. J. Med. Chem. 2006, 49, 5029.
20. A similar trend was observed in the parent benzoylpiperazine series. See Ref.14
.
21. The panel consisted in targets included in the standard CEREP high throughput
profile screen complemented with enzymes (phosphodiesterases, kinases,
MAO-A, COMT, acetylcholineesterase). http://www.cerep.fr.
22. Kansy, M.; Senner, F.; Gubernator, K. J. Med. Chem. 1998, 41, 1007.