Synthesis and SAR of GlyT1 inhibitors derived from a series of N-((4-(morpholine-4-carbonyl)-1-(propylsulfonyl)piperidin-4-yl)methyl)benzamides

Article abstract of DOI:10.1016/j.bmcl.2006.08.131

This Letter describes the synthesis and SAR, developed through an iterative analog library approach, of potent and selective non-sarcosine-derived GlyT1 inhibitors.

Full text of DOI:10.1016/j.bmcl.2006.08.131

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5968–5972
Synthesis and SAR of GlyT1 inhibitors derived from a series of
N-((4-(morpholine-4-carbonyl)-1-(propylsulfonyl)piperidin-4-yl)-
methyl)benzamides
Zhijian Zhao,a,* Julie A. O’Brien,b Wei Lemaire,b David L. Williams, Jr.,b
Marlene A. Jacobson,b Cyrille Sur,b Doug J. Pettibone,b Philip R. Tiller,c
Sheri Smith,c George D. Hartman,a Scott E. Wolkenberga and
Craig W. Lindsleya,d
aDepartment of Medicinal Chemistry, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
bDepartment of Neuroscience, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
cDepartment of Drug Metabolism, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
dVICB Program in Drug Discovery, Department of Pharmacology, Vanderbilt Medical Center, Nashville, TN 37232, USA
Received 6 July 2006; revised 2 August 2006; accepted 31 August 2006
Available online 20 September 2006
Abstract—This Letter describes the synthesis and SAR, developed through an iterative analog library approach, of potent and selec-
tive non-sarcosine-derived GlyT1 inhibitors.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Schizophrenia is a complex psychiatric disorder charac-
terized by a combination of negative, positive, and cogni-
tive symptoms that requires lifelong daily maintenance
therapy.1 Historically, schizophrenia has been managed
clinically by treatment with dopamine D2 antagonists
(i.e., the dopamine hyperfunction hypothesis); however,
this strategy only improves the positive symptoms of the
disease—negative and cognitive symptoms remain.2 Re-
cent data suggest that N-methyl-D-aspartate (NMDA)
receptor hypofunction is involved in the pathophysiology
of schizophrenia, that is, the NMDA hypofunction (or
glutamate dysfunction) hypothesis, as an alternative the-
ory for the underlying causes of the disease. According to
this hypothesis, any agent that can potentiate NMDA
receptor currents has the potential to ameliorate the
symptoms of schizophrenia.3,4
expression in the neocortex mirrors NMDA receptor
expression suggesting that GlyT1 is optimally positioned
to modulate glycine levels near NMDA receptor-
expressing synapses.7 Hence, increasing the activity of
NMDA receptors by inhibition of GlyT1 has emerged
as an attractive target for drug discovery efforts. Early
work in this arena focused on sarcosine-derived GlyT1
inhibitors which validated this approach in animal mod-
els, but suffered from poor pharmacokinetics (PK) and
brain penetration.1–4 Based on these results, numerous
pharmaceutical companies have launched efforts to
identify non-sarcosine-derived GlyT1 inhibitors, and
reports are beginning to accrue.
Recently, we reported on the optimization of HTS lead
1 to afford 2, a potent (GlyT1 IC50 = 2.6 nM) and selec-
tive (versus GlyT2 and TauT) GlyT1 inhibitor based on
In the forebrain, glycine is a required co-agonist for the
NMDA receptor and modulates NMDA-dependent
excitatory neurotransmission; therefore, elevation of
synaptic glycine levels should enhance NMDA receptor
function.5 In the CNS, glycine levels are tightly main-
tained by two transporters: GlyT1 and GlyT2.6 GlyT1
O
OMe
O
NH2
N
N
H
H
Cl
N
S
N
S
O
O
O
O
1 (IC50 = 135 nM)
2 (IC50 = 2.6 nM)
Keywords: Schizophrenia; GlyT1; Inhibitor; Piperidine.
*
Figure 1. 4,4-Disubstituted piperidine GlyT1 inhibitors.
Corresponding author. E-mail: zhijian_zhao@merck.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.131

Z. Zhao et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5968–5972
5969
antipsychotic profile. However, poor aqueous solubility,
high logP (>3.5), and strong time-dependent CYP inhi-
bition (TDI) prevented the further development of 2 and
related analogs.8
O
Cl
CN
CN
BnO
BnO
N
H
d,e
a,b,c
Cl
N
S
N
S
N
O
O
Boc
O
O
In this Letter, we report the results of further lead opti-
mization efforts, through an iterative analog library syn-
thesis approach, of 2 to improve physical properties and
eliminate CYP TDI. From past experience, we surmised
that the 4-phenyl moiety in 2 might be the cause of the
CYP TDI as well as contributing to the undesirable high
logP. Early work in this series indicated that the propyl
sulfonamide was required for maximal GlyT1 activity,
but that a wide range of benzamides were tolerated.8
Therefore, our initial libraries explored alternative, po-
lar moieties in the 4-position while holding the propyl-
sulfonamide and 2,4-dichlorobenzamide constant. For
the scope of this Letter, replacement of the 4-phenyl
moiety with polar 4-carboxamides will be discussed.
3
4
5
O
O
Cl
O
O
Cl
R1
HO
N
N
R2
N
H
f,g
H
h
Cl
Cl
N
S
N
S
O
O
O
O
6
7
Scheme 1. Reagents and conditions: (a) KHMDS, THF, 25 ꢁC, then
BnOCH2Cl, 82%; (b) 4 M HCl/dioxane, 100%; (c) n-PrSO2Cl, i-
Pr2NEt, CH2Cl2, 0–25 ꢁC, 95%; (d) H2, Raney Ni; 2 M NH3 in MeOH,
25 ꢁC, 95%; (e) 2,4-diClPhCOCl, i-Pr2NEt, CH2Cl2, 0–25 ꢁC, 95%; (f)
TMSI, AcCN, 25 ꢁC, 95%; (g) cat. RuCl3/NaIO4, AcCN–CCl4–H2O,
88%; (h) NHR1R2, PS-DCC, HOBt, i-Pr2NEt, CH2Cl2, 25 ꢁC,
72–91%. All compounds purified by mass-guided HPLC.9
As illustrated in Scheme 1, commercially available N-Boc-
4-cyanopiperdine 3 is deprotonated and the resulting
anion trapped with BnOCH2Cl, followed by deprotection
of Boc and sulfonylation to provide analogs 4. Raney
Ni-catalyzed hydrogenation and subsequently acylation
with 2,4-dichlorobenzoyl chloride provided analogs 5.
a 4,4-disubstituted piperidine scaffold (Fig. 1).8 In vivo,
2 selectively increased glycine levels in the prefrontal
cortex (340% of basal) and significantly enhanced pre-
pulse inhibition (PPI) in DBA/2J mice, suggesting an
Table 1. Structures and activities of carboxamide analogs 7
O
O
Cl
R1
N
R2
N
H
Cl
N
O2S
7
a
a
rGlyT1 IC50 (nM)
a
mGlyT1 IC50 (nM)
Compound
NR1R2
hGlyT1 IC50 (nM)
181
7a
ND
ND
ND
ND
ND
ND
ND
3.2
ND
ND
ND
ND
ND
ND
ND
7.6
N
7b
7c
7d
7e
7f
517
544
HN
N
Me2N
472
>5000
69
N
N
N
H
O
7g
7h
109
H2N
N
2.4
O
N
All compounds >30,000 nM versus GlyT2 and TauT.
a Average of at least three measurements11; ND, not determined.

5970
Z. Zhao et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5968–5972
O
O
50 analogs of 7 were synthesized and evaluated in our
GlyT1 inhibition assay. Table 1 highlights selected com-
pounds and data.
O
O
N
N
H
R
N
NH2
a
O
O
N
S
N
S
O
O
Over 80% of the analogs prepared were inactive with only
a small number showing appreciable activity; however,
the morpholino amide analog, 7h, was a clear stand-out
with low nanomolar inhibition of human, rat, and mouse
GlyT1 (IC50s of 2.4, 3.2, and 7.6 nM, respectively) with no
inhibition of GlyT2 or TauT (taurine transporter). Unlike
our previous 4-phenyl series of inhibitors, 1 and 2, 7h dis-
played good aqueous solubility (ꢀ18 mg/mL in saline), no
CYP (3A4, 2C9, 2D6) inhibition, and no CYP 3A4 TDI.
Compound 7h displayed moderate pharmacokinetics
O
9
8
Scheme 2. Reagents and conditions: (a) RCOCl, i-Pr2NEt, CH2Cl2, 0–
25 ꢁC, 95% or (i) RCOOH, PS-DCC, HOBt, DCM (ii) MP-carbonate,
70–88%. All compounds purified by mass-guided HPLC.9
Deprotection of the benzyl group and subsequent oxida-
tion afforded the carboxylic acid 6 which was coupled
with an amine to give amide 7. Following this scheme,
Table 2. Structures and activities of amide analogs 9
O
O
N
N
R
H
O
N
S
O
O
9
a
hGlyT1 IC50 (nM)
a
rGlyT1 IC50 (nM)
a
mGlyT1 IC50 (nM)
Compound
R
9a
>5000
>5000
264
ND
ND
ND
ND
10.1
ND
ND
19.7
22.1
7.4
ND
ND
ND
ND
28.5
ND
ND
42.8
57.3
19.4
ND
95.3
9b
9c
9d
9e
9f
9g
9h
9i
88.4
F3CO
Cl
12.6
42.8
67.8
8.4
F
F
CF3
F3C
F
F
Cl
F
F
12.5
6.3
Cl
Cl
9j
Cl
F
F
9k
9l
131
ND
18.3
F
9.7
Cl
All compounds >30,000 nM versus GlyT2 and TauT.
a Average of at least three measurements11; ND, not determined.

Z. Zhao et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5968–5972
5971
Me
O
Cl
(PK) in rats (Cl = 20.5 mL/min/kg, t1/2 = 0.37 h,
Me
NH2
CN
VD = 0.38). In addition, the logP of 7h was measured to
be 1.4, marking a pronounced increase in polarity by
incorporation of the morpholino amide moiety relative
to 2. As a consequence 7h proved to be a mouse and hu-
man p-glycoprotein (P-gp) substrate (B À A/A À B ratios
of 11.4 and 17.9, respectively); however, excellent passive
permeability was maintained (Papp = 23.6 · 10À6 cm/
s).10 Despite overcoming several limitations of the
4-phenyl series (1 and 2), the P-gp data, predicting poor
CNS exposure for 7h, prevented further in vivo experi-
ments from being conducted.
BnO
a,b
N
BnO
BnO
c
H
(one pot)
Cl
N
S
O
N
S
O
N
S
O
O
O
O
11
4
10
Cl
O
O
Cl
O
O
d,e,f
N
N
N
N
H
H
O
O
Cl
Cl
N
S
O
N
S
O
O
O
12S
12R
Encouraged by 7h, additional libraries were prepared,
according to Schemes 1 and 2, that examined alternative
amide moieties while keeping the 4-morpholino amide
functionality constant. In total, 96 compounds were syn-
thesized and screened in our GlyT1 assay. Table 2 high-
lights selected compounds and data from this effort
which generated a number of potent and selective
(>30,000 nM versus GlyT2 and TauT) GlyT1 inhibitors
with diverse benzamide moieties. In general, aliphatic
amides, such as 9a and 9b, were inactive as were all het-
erocyclic amides (pyridyl, thienyl, etc., data not shown).
Benzamides were uniformly active. For instance, unsub-
stituted phenyl, 9c, had a GlyT1 IC50 of 264 nM which
could be dramatically increased by the addition of a sin-
gle halogen atom as in 9l, possessing a 2-chloro substitu-
ent, with low nanomolar inhibition of human, rat, and
mouse GlyT1 (IC50s of 9.7, 18.3, and 95.3 nM, respec-
tively). A wide-range of halogenation patterns about
the phenyl ring of the benzamide were found to be tol-
erated. The most potent inhibitor from this library, 9j,
possessed a 2,4-dichloro-5-fluorobenzamide moiety and
displayed low nanomolar inhibition of human, rat,
and mouse GlyT1 (IC50s of 6.3, 7.4, and 19.4 nM,
respectively); however, this effort failed to identify a
compound more potent than 7h. Similar to 7h, all ana-
logs 9 displayed good aqueous solubility (10–25 mg/
mL in saline), no CYP (3A4, 2C9, and 2D6) inhibition,
no CYP 3A4 TDI, and displayed moderate PK in rats
(Cl = 18–25 mL/min/kg, t1/2 = 0.30–1.4 h, VD = 0.30–
0.52). Unfortunately, all of the analogs 9 were found
to be mouse and human P-gp substrates (B À A/A À B
ratios of 8.9–19 and 10.6–26.1, respectively) yet
displayed excellent passive permeability (Papp > 22 ·
10À6 cm/s).10 In contrast, the 4-phenyl series, such as 1
and 2, uniformly were not P-gp substrates.8 This was
surprising as analogs 9 did not incorporate additional
hydrogen bond donors relative to the 4-phenyl series;
however, the increase in P-gp susceptibility could be
the result of increased polarity (logP of <1.4 for analogs
9 versus >3.0 for analogs 2). The P-gp results did prove
to be indicative of low CNS penetration for this series,
as 7h displayed low nanomolar total brain levels from
brain:plasma studies in rats.
Scheme 3. Reagents and conditions: (a) (i) MeMgBr, toluene, 25–
40 ꢁC, (ii) anhydrous MeOH; (b) NaBH4, 0–25 ꢁC, 63% (two steps); (c)
2,4-dichlorobenzoyl chloride, i-Pr2NEt, 0–25 ꢁC, 93%; (d) TMSI,
AcCN, 25 ꢁC, 95%; (e) cat. RuCl3/NaIO4, AcCN–CCl4–H2O, 84%; (f)
(i) morpholine, PS-DCC, HOBt, i-Pr2NEt, CH2Cl2, 25 ꢁC, (ii) Chiral
pack AD, 32% for 12R and 37% for 12S.
(S)-enantiomers of 7h were prepared, 12R and 12S,
respectively, and screened in our GlyT1 assay. To our
surprise, both 12R and 12S were inactive (GlyT1
IC50 > 5000 nM); clearly, the SAR for the 4-morphoino
amide series was distinct from the 4-phenyl series.
As the SAR for this series was distinct from our earlier
work,8 we next explored a multi-dimensional library ap-
proach (Fig. 2) wherein we simultaneously evaluated
alternative sulfonamides, diverse benzamides, and
alternative cappings for the eastern amino methyl group
(2ꢁ- and 3ꢁ-amines, ureas, and sulfonamides), 13. This
effort afforded over 100 additional analogs that were
evaluated in our GlyT1 assay. Only benzamide moieties
were tolerated as capping agents for the eastern amino
methyl group—basic amines, ureas, and sulfomamides
possessed no GlyT1 inhibitory activity. Alternative
sulfonamides for the piperidine amine maintained some
GlyT1 inhibition, but at a loss of 90- to 500-fold. Based
on these data and the lack of brain penetration for 7h
and related congeners, the 4-morpholino amide series
was not developed further.
In summary, we have developed a novel series of potent
and selective non-sarcosine-derived GlyT1 inhibitors.
Compounds from this series, such as 7h, addressed
several of the liabilities of our early 4-phenyl series,
depicted by 1 and 2. Specifically, incorporation of the
4-morpholino amide moiety increases polarity (logP
ꢀ1.4), improves aqueous solubility (10–25 mg/mL in
saline), and eliminates the CYP TDI which plagued
the 4-phenyl series. However, while the increased
O
R4
R3 = Me, Et, Cyclopropyl, CF3, CH2CF3
R4 = H, Me; R5 = COAr, SO2R, CONHR, R
R4, R5 = alkyl
N
N
O
R5
In the 4-phenyl series, the introduction of a (S)-methyl
group, as shown in 2, improved intrinsic GlyT1 potency
and, for a number of analogs with borderline mouse P-
gp susceptibility (B À A/A À B ratios of ꢀ5.6) eliminat-
ed the P-gp susceptibility (B À A/A À B ratios of 2.2).8
According to Scheme 3, both the enantiopure (R)- and
N
S
R3
O
O
13
Figure 2. Multi-dimensional library generic 13.

5972
Z. Zhao et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5968–5972
M. In Amphetamine and its analogues, Cho; Segals, Eds.;
Academic Press: San Diego, CA, 1994, pp 387–414.
3. (a) Javitt, D. C.; Zukin, S. R. Am. J. Psychiatry 1991, 148,
1301; (b) Olney, J. W.; Farber, N. B. Arch. Gen. Psychiatry
1995, 52, 998; (c) Javitt, D. C.; Zylberman, I.; Zukin, S.
R.; Heresco-Levy, U.; Lindenmayer, J. P. Am. J. Psychi-
atry 1994, 151, 1234; (d) Heresco-Levy, U.; Ermilov, M.;
Lichtenberg, P.; Bar, G.; Javitt, D. C. Biol. Psychiatry
2004, 55, 165; (e) Olney, J. W.; Newcomer, J. W.; Farber,
N. B. J. Psychiatry Res. 1999, 33, 523.
polarity of this series overcame several liabilities, it also
engendered this series to be P-gp substrates and there-
fore not CNS penetrant. Further refinements to this
general scaffold are in progress and additional SAR
and biological studies will be reported in due course.
Acknowledgment
The authors wish to thank Dr. Charles W. Ross III for
obtaining HRMS data (accurate mass measurements).
4. For
a
review Lindsley, C. W.; Shipe, W. D.;
Wolkenberg, S. E.; Theberge, C. R.; Williams, D.
L., Jr.; Sur, C.; Kinney, G. G. Curr. Topics Med.
Chem. 2006, 8, 771.
References and notes
5. (a) Danysz, W.; Parsons, A. C. Pharmacol. Rev. 1998, 50,
597; (b) Smith, K. E.; Borden, L. A.; Hartig, P. R.;
Branchek, T.; Weinshank, R. L. Neuron 1992, 8, 927.
6. (a) Borowsky, B.; Mezey, E.; Hoffman, B. J. Neuron 1993,
10, 851; (b) Liu, Q. R.; Lopez-Corcuera, B.; Mandiyan, S.;
Nelson, H.; Nelson, N. J. Biol. Chem. 1993, 268, 22802; (c)
Kim, K. M.; Kingsmore, S. F.; Han, H. Mol. Pharmacol.
1994, 45, 608; (d) Morrow, J. A.; Collie, I. T.; Dunbar, D.
R.; Walker, G. B.; Shahid, M.; Hill, D. R. FEBS Lett.
1998, 439, 334.
1. (a) Andreasen, N. C. Brain Res. Rev. 2000, 31, 106; (b)
Karayiorgou, M. Clin. Neurosci. Res. 2001, 1, 158; (c)
Meador-Woodruff, J. H.; Healy, D. J. Brain Res. Rev.
2000, 31, 288.
2. (a) Marino, P. J.; Conn, P. J. Curr. Drug Targets. CNS
Neurol. Disord. 2002, 1, 1; (b) Carlsson, A.; Lindqvist, M.
Acta Pharmacol. Toxicol. 1963, 20, 140; (c) Seeman, P.;
Lee, T.; Chau-Wong, M.; Wong, K. Nature 1976, 261,
717; (d) Seeman, P.; Lee, T. Science 1975, 188, 1217; (e)
Bertolino, A.; Breier, A.; Callicott, J. H.; Adler, C.;
Mattay, V. S.; Shapiro, M.; Frank, J. A.; Pickar, D.;
Weinberger, D. R. Neuropsychopharmacology 1999, 22,
125; (f) Laruelle, M.; Abi-Dargham, A.; van Dyck, C. H.;
Gil, R.; D’Souza, C. D.; Erdos, J.; McCance, E.; Rosen-
blatt, W.; Fingado, C.; Zoghbi, S. S.; Baldwin, R. M.;
Seibyl, J. P.; Krystal, J. H.; Charney, D. S.; Innis, R. B.
Proc. Natl. Acad. Sci. USA 1996, 93, 9235; (g) Laruelle,
M.; D’Souza, C. D.; Baldwin, R. M.; Abi-Dargham, A.;
Kanes, S. J.; Fingado, C. L.; Seibyl, J. P.; Zoghbi, S. S.;
Bowers, M. B.; Jatlow, P.; Charney, D. S.; Innis, R. B.
Neuropsychopharmacology 1997, 17, 162; (h) Angrist, B.
7. Zarfa, F.; Gomeza, J.; Olivares, L.; Aragon, C.; Gimenez,
C. Eur. J. Neurosci. 1995, 7, 1342.
8. Lindsley, C. W.; Zhao, Z.; Leister, W. H.; O’Brien, J. A.;
Lemiare, W.; Williams, D. L., Jr.; Chen, T.-B.; Chang, R.
S. L.; Burno, M.; Jacobson, M. A.; Sur, C.; Kinney, G. G.;
Pettibone, D. J.; Tiller, P. R.; Smith, S.; Tsou, N. N.;
Duggan, M. E.; Conn, P. J.; Hartman, G. D. ChemMed-
Chem 2006, 1, 807.
9. Leister, W. H.; Strauss, K. A.; Wisnoski, D. D.;
Zhao, Z.; Lindsley, C. W. J. Comb. Chem. 2003,
5, 322.
10. Hochman, J. H.; Yamazaki, M.; Ohe, T.; Lin, J. H. Curr.
Drug Methods 2002, 3, 257.