Discovery of Potent and Selective Inhibitors of Phosphodiesterase 1 for the Treatment of Cognitive Impairment Associated with Neurodegenerative and Neuropsychiatric Diseases

Article abstract of DOI:10.1021/acs.jmedchem.5b01751

A diverse set of 3-aminopyrazolo[3,4-d]pyrimidinones was designed and synthesized. The structure-activity relationships of these polycyclic compounds as phosphodiesterase 1 (PDE1) inhibitors were studied along with their physicochemical and pharmacokinetic properties. Systematic optimizations of this novel scaffold culminated in the identification of a clinical candidate, (6aR,9aS)-2-(4-(6-fluoropyridin-2-yl)benzyl)-5-methyl-3-(phenylamino)-5,6a,7,8,9,9a-hexahydrocyclopenta[4,5]imidazo[1,2-a]pyrazolo[4,3-e]pyrimidin-4-(2H)-one phosphate (ITI-214), which exhibited picomolar inhibitory potency for PDE1, demonstrated excellent selectivity against all other PDE families and showed good efficacy in vivo. Currently, this investigational new drug is in Phase I clinical development and being considered for the treatment of several indications including cognitive deficits associated with schizophrenia and Alzheimer's disease, movement disorders, attention deficit and hyperactivity disorders, and other central nervous system (CNS) and non-CNS disorders.

Full text of DOI:10.1021/acs.jmedchem.5b01751

Subscriber access provided by UNIV OSNABRUECK
Article
Discovery of Potent and Selective Inhibitors of Phosphodiesterase
for the Treatment of Cognitive Impairment Associated
1
with Neurodegenerative and Neuropsychiatric Diseases
Peng Li, Hailin Zheng, Jun Zhao, Lei Zhang, Wei Yao, Hongwen Zhu, J. David Beard, Koh
Ida, Weston Lane, Gyorgy Snell, Satoshi Sogabe, Charles J. Heyser, Gretchen L. Snyder,
Joseph P. Hendrick, Kimberly E. Vanover, Robert E. Davis, and Lawrence P Wennogle
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01751 • Publication Date (Web): 20 Jan 2016
Downloaded from http://pubs.acs.org on January 28, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 51
Journal of Medicinal Chemistry
1
2
3
4
5
6
Discovery of Potent and Selective Inhibitors of Phosphodiesterase 1 for the Treatment of
Cognitive Impairment Associated with Neurodegenerative and Neuropsychiatric Diseases
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
*,†
†
†, €
†
†
†, £
†
Peng Li, Hailin Zheng, Jun Zhao, Lei Zhang, Wei Yao, Hongwen Zhu, J. David Beard, Koh
¶
‡
‡
¶
§
†
Ida, Weston Lane, Gyorgy Snell, Satoshi Sogabe, Charles J. Heyser, Gretchen L. Snyder, Joseph P.
†
†
†
†
Hendrick, Kimberly E. Vanover, Robert E. Davis and Lawrence P. Wennogle
†
th
IntraꢀCellular Therapies, Inc., 430 East 29 Street, Suite 900, New York, New York, 10016, United
States
‡
Department of Structural Biology, Takeda California, Inc., 10410 Science Center Dr., San Diego, Caliꢀ
fornia 92121, United States
¶
Pharmaceutical Research Division, Takeda Pharmaceutical Company, Ltd., 26ꢀ1, MuraokaꢀHigashi 2ꢀ
chome, Fujisawa, Kanagawa 251ꢀ8555, Japan
§
Department of Neurosciences, University of California, San Diego, 9500 Gilman Dr. #0608, La Jolla,
California, 92093, United States
──────────────────────────────────────────────
∗
Author to whom correspondence should be addressed
Peng Li, Ph.D.
Vice President, Medicinal Chemistry
IntraꢀCellular Therapies, Inc.
th
4
30 East 29 Street, Suite 900
New York, NY 10016
Phone: 646ꢀ440ꢀ9388
Email: pli@intracellulartherapies.com
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 51
ABSTRACT
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A diverse set of 3ꢀaminopyrazolo[3,4ꢀd]pyrimidinones was designed and synthesized. The structureꢀ
activity relationships of these polycyclic compounds as phosphodiesterase 1 (PDE1) inhibitors were
studied along with their physicochemical and pharmacokinetic properties. Systematic optimizations of
this novel scaffold culminated in the identification of a clinical candidate, (6aR,9aS)ꢀ2ꢀ(4ꢀ(6ꢀ
fluoropyridinꢀ2ꢀyl)benzyl)ꢀ5ꢀmethylꢀ3ꢀ(phenylamino)ꢀ5,6a,7,8,9,9aꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hexahydrocyclopenta[4,5]imidazo[1,2ꢀa]pyrazolo[4,3ꢀe]pyrimidinꢀ4ꢀ(2H)ꢀone phosphate (ITIꢀ214),
which exhibited picomolar inhibitory potency for PDE1, demonstrated excellent selectivity against all
other PDE families, and showed good efficacy in vivo. Currently, this investigational new drug is in
Phase I clinical development and being considered for the treatment of several indications including
cognitive deficits associated with schizophrenia and Alzheimer’s disease, movement disorders, attention
deficit and hyperactivity disorders, and other CNS and nonꢀCNS disorders.
INTRODUCTION
3
',5'ꢀCyclic adenosine monophosphate (cAMP) and 3',5'ꢀcyclic guanosine monophosphate (cGMP, 1)
1
play important roles in signal transduction and synaptic transmission. Cyclic nucleotide phosꢀ
phodiesterases (PDEs) degrade cAMP and cGMP by hydrolysis of 3'ꢀphosphodiester bonds, and conseꢀ
2
quently regulate intracellular levels of these ubiquitous second messengers. To date, 11 different PDE
3
families (PDE1 to PDE11) have been identified, encoded by 21 genes. Unique among the PDEs, the
4
,5
enzymes of the PDE1 family are tightly regulated by intracellular calcium and calmodulin levels. Risꢀ
es in intracellular calcium lead to increased PDE1 activity and result in increased cyclic nucleotide degꢀ
radation. PDE1 enzymes are dual cyclic nucleotide phosphodiesterases that can accommodate and hyꢀ
6
drolyze both cAMP and cGMP. The ability of PDE1 to hydrolyze both cAMP and cGMP allows these
enzymes to dynamically influence signaling in multiple excitatory cell systems.
ACS Paragon Plus Environment

Page 3 of 51
Journal of Medicinal Chemistry
PDE1 exists as three isoforms (PDE1A, 1B and 1C), derived from three distinct genes. Each of these
1
2
5
isoforms has multiple splice variants. These isoforms/splice variants appear to be distributed in a cell
typeꢀspecific manner in the brain and other tissues. For example, in the brain the predominant expresꢀ
sion of PDE1A is in the cortex and neostriatum, PDE1B is in the neostriatum, prefrontal cortex, hippoꢀ
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
campus and olfactory tubercle and PDE1C is more ubiquitously expressed. The cell type specific exꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pression as well as regulation by intracellular calcium and calmodulin, and regionally specific localizaꢀ
tions of the PDE1 isoforms place these enzymes in a position to influence phasic, but not tonic signaling
from diverse neurotransmitter systems. These systems include those utilizing dopamine, acetylcholine,
7
glutamate and certain peptidergic systems. As one example, PDE1B is coꢀlocalized with dopamine D1
receptors (D1R), which initiate signal transduction via production of cAMP. Both PDE1B and D1R are
7–9
abundantly expressed in the prefrontal cortex, a brain region that governs working memory function.
Modulation of dopamine function through D1 receptors has been shown to alter certain mnemonic proꢀ
1
0
cesses. Insufficient D1Rꢀmediated signaling in the prefrontal cortex has been shown to impair cogniꢀ
11
tive function in both preclinical and clinical studies. As D1 receptors signal through modulation of adꢀ
enylate cyclase and production of cAMP, inhibition of PDE1 will amplify dopaminergic neurotransmisꢀ
sion downstream of these receptors. Increased D1Rꢀmediated signaling would be expected to ameliorate
1
2,13
cognitive deficits associated with schizophrenia,
as well as other central nervous system (CNS) and
nonꢀCNS disorders. However, the foregoing does not preclude the possibility that PDE1 inhibition may
also modulate other neurotransmitter systems that signal through either cAMP or cGMP systems.
Changes in these neurotransmitter systems are known to influence cognitive function and include doꢀ
paminergic, muscarinic, glutamatergic and nitric oxide systems. Thus, by uniquely preventing degradaꢀ
tion of both cAMP and cGMP signaling, PDE1 inhibition is likely to improve cognitive function in a
variety of neurodegenerative and neuropsychiatric diseases, including schizophrenia, Parkinson's disꢀ
1
4
ease and dementia.
The role of PDE1 in excitatory cellular systems compelled us to design PDE1 specific inhibitors for
the treatment of cognitive dysfunction. Since the catalytic domains of all PDE enzymes share a high deꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 51
gree of sequence homology, this was an ambitious endeavor. The key interactions of substrates at the
catalytic sites of these PDE enzymes include an important hydrogen bond with a conserved activeꢀsite
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
glutamine and a hydrophobic clamp that sandwiches the purineꢀlike core of the cAMP or cGMP, as conꢀ
15–17
firmed by Xꢀray crystallography.
Various PDE1 inhibitors with different heterocyclic systems mimꢀ
1
8,19
icking the structure of guanine have been documented.
However, most of the known PDE1 inhibiꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
tors in the public domain show inhibitory activity to other PDE enzymes in addition to PDE1. Achieving
high selectivity between enzymes of the PDE superfamily can be a challenging task, yet necessary in
order to develop PDE drug candidates with good safety profiles. In particular, drug candidates need to
2
0
avoid PDE6 activity to minimize effects on vision.
In an effort to develop potent and selective PDE1 inhibitors for the treatment of symptoms associated
with neuropsychiatric and neurodegenerative diseases, we analyzed the published PDE inhibitors with
different chemical scaffolds and found the pyrazolo[3,4ꢀd]pyrimidinones reported by Xia and coꢀ
19
workers could be a good starting point for design of new PDE1 inhibitors. These compounds were deꢀ
veloped as PDE1 and PDE5 dual inhibitors with nanomolar to micromolar inhibitory activities. Accordꢀ
ing to the published Xꢀray structures of PDE enzymes complexed with cGMP and various PDE inhibiꢀ
1
5–17
tors mimicking guanine,
the Nꢀ7 nitrogen in the purine ring of guanine usually forms a hydrogen
bond with an adjacent amino acid residue located in the catalytic site. For the pyrazolo[3,4ꢀ
d]pyrimidinone series of compounds, the shift of the nitrogen from position 3 to position 2 of the pyraꢀ
zole ring could disrupt the hydrogen bond network. To reꢀestablish the hydrogen bond mentioned above,
we designed the polycyclic 3ꢀaminopyrazolo[3,4ꢀd]pyrimidinone scaffold 2 with the intention to form
the important hydrogen bond through the amino group substitution at carbonꢀ3 of the pyrazole ring
(Figure 1). By the combined use of ligandꢀbased and structureꢀbased drug design processes, we have
designed and developed multiple novel scaffolds as PDE1 inhibitors. In this paper, we will focus our
discussion on the development and structureꢀactivity relationships (SAR) of the polycyclic 3ꢀ
aminopyrazolo[3,4ꢀd]pyrimidinone scaffold 2 as novel PDE1 inhibitors. This work led to the discovery
of
(6aR,9aS)ꢀ2ꢀ(4ꢀ(6ꢀfluoropyridinꢀ2ꢀyl)benzyl)ꢀ5ꢀmethylꢀ3ꢀ(phenylamino)ꢀ5,6a,7,8,9,9aꢀ
ACS Paragon Plus Environment

Page 5 of 51
Journal of Medicinal Chemistry
hexahydrocyclopenta[4,5]ꢀimidazo[1,2ꢀa]pyrazolo[4,3ꢀe]pyrimidinꢀ4ꢀ(2H)ꢀone phosphate (3, ITIꢀ214),
which is a picomolar PDE1 inhibitor with excellent selectivity against other PDE family members and
against a panel of enzymes, receptors, transporters and ion channels. Currently, compound 3 is in human
clinical trials.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structures of cGMP (1), 3-amino-pyrazolo[3,4-d]pyrimidine scaffold (2) and com-
pound 3.
CHEMISTRY
To efficiently synthesize a diverse set of polycyclic pyrazolo[3,4ꢀd]pyrimidinone compounds for SAR
studies, several synthetic methods have been developed as illustrated in Schemes 1–3. The synthetic
route exemplified in Scheme 1 was employed to conveniently explore the SAR of the left lower moiety
of this series of compounds with variations of substituents (R –R ), ring size (n) and stereochemistry.
2
4
Commercially available 6ꢀchloroꢀ3ꢀmethyluracil was first protected with a paraꢀmethoxybenzyl (PMB)
group, followed by chlorine displacement with hydrazine hydrate to give compound 5. Treatment of 5
with phenyl isothiocyanate using a procedure developed based on a reported method allowed us to conꢀ
struct the pyrazolo[3,4ꢀd]pyrimidinone ring system and introduce a phenylamino group at position 3 in
2
1
one step. The obtained compound 6 was then reacted with a paraꢀsubstituted benzyl halide to give N2ꢀ
substituted 7 as the major product along with a small amount of N1ꢀsubstituted isomer, which can be
readily separated by column chromatography. After the removal of the PMB protective group with triꢀ
fluoromethanesulfonic acid (TfOH) and trifluoroacetic acid (TFA), the resulting compound 8 was chloꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 51
rinated with phosphorus oxychloride to furnish the key intermediate 9. Reaction of 9 with various amino
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
alcohols HOCHR CR R (CH ) NH followed by ringꢀclosure with thionyl chloride afforded the final
4
2
3
2 n
2
products 19–23, 25–29, 38 and 66, as listed in Tables 1 and 4. Compound 24 was synthesized by debenꢀ
zylation of 27 with boron trichloride.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To expedite the SAR development of compounds with variations of substituents at position N1 or N2
of the pyrazole ring, another synthetic route was developed, as illustrated in Scheme 2. Intermediate 6
was deprotected with TfOH and TFA, followed by chlorination with phosphorus oxychloride to afford
1
0. Amination of 10 with (1R,2R)ꢀ2ꢀaminocyclopentanol upon heating produced 11, which was treated
with thionyl chloride to afford the tetracyclic intermediate 12. For rapidly generating analogs, the
sidechain R was introduced in the last step of the synthesis by reacting 12 with a halide RX. Thus, N1ꢀ
substituted compounds 30–36, 39, 44 and 46 were obtained as the major products. N2ꢀsubstituted isoꢀ
mers 37, 40–43, 45, 47 and 67–69 can also be prepared using this method.
The third synthetic route was developed to rapidly explore the SAR of the substituents at position 3 of
the pyrazole ring, as illustrated in Scheme 3. Intermediate 5 was treated with POCl and DMF to afford
3
13. Benzylation of 13 with 2ꢀ(4ꢀ(chloromethyl)phenyl)ꢀ6ꢀfluoropyridine gives exclusively the N2ꢀ
substituted compound 14, which upon treatment with TfOH and TFA furnished 15. Compound 16 was
22
synthesized from 15 by BOPꢀmediated direct amination. The subsequent stereoselective ringꢀclosure
was achieved by treating 16 with N,Nꢀdimethylchloroꢀsulfitemethaniminium chloride, in situ generated
23
from thionyl chloride and DMF, to afford tetracyclic compound 17. Chlorination of 17 with Lithium
bis(trimethylsilyl)amide (LiHMDS) and hexachloroethane produced the key intermediate 18, which enꢀ
ables us to efficiently introduce various substituents into position 3 in the final step of the synthesis. For
instance, 3ꢀalkylamino and 3ꢀbenzylamino substituted compounds 49–57 and 59 can be conveniently
synthesized by reacting 18 with the corresponding amines at elevated temperature. Reaction of 18 with
4
ꢀmethoxybenzyl amine followed by PMB deprotection with TfOH and TFA yielded compound 48. 3ꢀ
Phenylamino substituted compound 3 was prepared from 18 in good yields by a palladiumꢀcatalyzed
ACS Paragon Plus Environment

Page 7 of 51
Journal of Medicinal Chemistry
crossꢀcoupling reaction using tris(dibenzylideneacetone)dipalladium (Pd (dba) ) as a catalyst and
2
3
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Xantphos as a ligand. Compounds 58 and 71 were synthesized by reacting 18 with pyridinꢀ4ꢀamine and
4ꢀfluorobenzenamine, respectively, using LiHMDS as a strong base. Aryloxyꢀsubstituted 72 and 73
were produced by treating 18 with the corresponding aryl alcohols and potassium carbonate at 150 °C.
Methylation of 3 with methyl iodide and cesium carbonate furnished 60. Compound 74 with a phenylꢀ
thio group at position 3 was prepared by treating 17 with LiHMDS and phenyl disulfide. Synthesis of 3ꢀ
benzyl substituted 75 and 76 were accomplished by Pdꢀcatalyzed Suzuki coupling of 18 with borate inꢀ
termediates, which were produced in situ from Bꢀmethoxyꢀ9ꢀborabicycloꢀ[3.3.1]nonane (BꢀOMeꢀ9ꢀ
BBN) and the corresponding arylmethylzinc bromides. Reduction of 38 and 47 with diisobutylaluminiꢀ
um hydride (DIBALH) afforded 61 and 65, respectively. Treatment of 38 with phosphorus pentasulfide
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
yielded a thiocarbonyl analog, which was subsequently reacted with HgCl and ammonia to give 62.
2
Compound 63 was prepared using a similar method by heating 38 with phosphorus pentasulfide and
ammonia in methanol at 150 °C in a microwave reactor.
a
Scheme 1
O
O
O
NHPh
NH
NHPh
N
O
O
NHPh
N
N
N
N
N
a,b
c
d
e
N
NH
2
N
N
O
N
N
O
N
O
N
N
O
N
H
Cl
H
O
N
H
PMB
PMB
PMB
R
1
4
5
6
7
8
R
1
O
N
O
NHPh
NHPh
N
N
N
f
g, h
A
N
N
B
N
N
N
Cl
C
n
R
4
R
2
R
3
R
1
R
1
9
19 − 23, 25 − 29, 38, 66
a
Reagents and conditions: (a) pꢀmethoxybenzyl chloride, K CO , DMF, 60 °C; (b) hydrazine hydrate,
2
3
EtOH, MeOH, reflux; (c) PhNCS, DMF, 120 °C; (d) R C H CH X, K CO , DMF; (e) TFA, TfOH, 25
1
6
4
2
2
3
°
C; (f) POCl , reflux; (g) HOCHR CR R (CH ) NH , DMF; (h) SOCl , CH Cl .
3 4 2 3 2 n 2 2 2 2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 51
a
Scheme 2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
O
O
N
O
N
NHPh
N
NHPh
N
NHPh
N
NHPh
N
N
N
N
O
N
N R
NHPh
N
N
N
N
N
HN
N
N
N
e
N
N
a,b
c
H
d
H
6
N
R
OH
N
Cl
H
1
0
11
12
30−36, 39, 44, 46
37, 40−43, 45, 47, 67−69
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (a) TFA, TfOH, CH Cl , 25 °C; (b) POCl , reflux; (c) (1R,2R)ꢀ2ꢀ
2
2
3
aminocyclopentanol hydrochloride, DIPEA, DMF, 120 °C; (d) SOCl , CH Cl ; (e) RX, K CO or
2
2
2
2
3
Cs CO .
2
3
a
Scheme 3
O
O
O
O
N
N
N
N
N
N
a
b
c
d
N
N
N
5
N
O
N
O
N
H
HN
N
NH
PMB
N
OH
O
N
PMB
N
N
N
F
g
F
F
1
3
14
15
16
e
O
O
O
R
Cl
N
N
N
N
N
N
f
N
N
N
N
N
N
N
N
N
N
N
N
3
, 48-57, 71
F
18
F
17
F
j
h
i
R
2
R
2
O
O
O
N
S
O
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
72 (R
3 (R
2
2
= H)
= F)
F
7
5 (R
= H)
= F)
F
74
F
2
2
7
76 (R
a
Reagents and conditions: (a) POCl , DMF; (b) 2ꢀ(4ꢀ(chloromethyl)phenyl)ꢀ6ꢀfluoropyridine, K CO ,
3
2
3
DMF, 25 °C; (c) TFA, TfOH, CH Cl , 25 °C; (d) BOP, DBU, (1R,2R)ꢀ2ꢀaminocyclopentanol hydroꢀ
2
2
chloride, THF, 25 °C; (e) SOCl , DMF, 25 °C; (f) LiHMDS, C Cl , THF, 25 °C; (g) alkyl amine, heatꢀ
2
2
6
ing; or aryl amine, Pd (dba) , Xantphos, K CO , tꢀamyl alcohol, reflux; or (hetero)aryl amine,
2
3
2
3
ACS Paragon Plus Environment

Page 9 of 51
Journal of Medicinal Chemistry
LiHMDS; (h) ArOH, K CO , microwave, 150 °C; (i) ArCH ZnX, BꢀOMeꢀ9ꢀBBN, PdOAc, SꢀPhos,
2
3
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DMF, 100 °C; (j) (PhS) , LiHMDS, THF, 25 °C.
2
RESULTS AND DISCUSSION
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
SAR development and X-ray crystallography. As a primary assay for exploring SAR, all of the synꢀ
thesized compounds were evaluated for their PDE1 inhibitory activities. These compounds were found
to be selective for inhibition of PDE1 versus other PDE family members. However, most of the comꢀ
pounds in this series also exhibited weak interactions with PDE4 as an offꢀtarget activity. Because of
this potential offꢀtarget PDE inhibitory activity, a human recombinant PDE4A assay was developed as a
counter screen to evaluate compound selectivity for these two enzyme classes. To delineate the SAR
around the investigational new drug 3, four representative sets of compounds were selected as summaꢀ
rized in Tables 1–4.
The SAR of compounds with different ring size and substituents on the Cꢀring is summarized in Table
1
. Compound 19 with a 6ꢀmembered Cꢀring exhibited a PDE1 inhibitory activity of 65 nM. Shrinking
the Cꢀring into a 5ꢀmembered ring slightly improved compound inhibitory activity (21 vs. 19). To estabꢀ
lish additional interactions with PDE1, various substituents were introduced on the Cꢀring. A gemꢀ
dimethyl substitution on the Cꢀring of 19 led to 13ꢀfold increase in PDE1 inhibitory potency and the reꢀ
sulting compound 20 demonstrated over 2000ꢀfold selectivity against PDE4A. Similarly, introduction of
a methyl or isopropyl group at the C ꢀposition of the 5ꢀmembered Cꢀring resulted in 9ꢀ and 13ꢀfold inꢀ
a
crease in PDE1 inhibition activity, respectively (22 and 23 vs. 21). Hydroxymethyl substituted 24 and
tetrahydrofuranꢀ3ꢀyl substituted 25 exhibited 2–5 fold better potency relative to 21 at the expense of apꢀ
proximately 2ꢀfold decrease of selectivity over PDE4A. It appeared that a small hydrophobic alkyl
group was preferred at the C ꢀposition, while a bulky substituent at this position had a detrimental effect
a
on compound binding to PDE1. For instance, C ꢀbenzyl substituted analog 26 was 13 times less potent
a
in PDE1 inhibition relative to 21. Conversely, such a structural modification enhanced PDE4A inhibitoꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 51
ry potency, with compound 26 showing only 33ꢀfold selectivity over PDE4A. Compound 27, containing
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a bulky substituent at the C ꢀposition, exhibited lesser PDE1 inhibitory potency than 21 as well. We also
a
attempted to introduce a small alkyl group on the C ꢀposition and found C ꢀmethylated compound 28
b
b
exerted more potent PDE1 inhibitory activity and better selectivity than 21. To take advantage of the
preferred alkyl substitutions at both C ꢀ and C ꢀpositions, 29 with a cyclopentyl group fused onto Cꢀring
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
b
was synthesized. This compound showed a PDE1 inhibitory activity of 0.85 nM, which was 34ꢀfold
more potent than 21. Among the different sizes of Cꢀring and substituents, compounds with a cyclopenꢀ
tyl fused 5ꢀmembered Cꢀring demonstrated increased PDE1 inhibitory potency and good selectivity, and
therefore compounds with this ring system were selected for further optimizations.
Table 1. SAR of Pyrazolo[3,4-d]pyrimidine Derivatives with Variations of Substituents on C-Ring
O
NHPh
N
B
N
A N
N
N
C
b
n
a
R
4
1
R
2
R
3
R
1
R
= Ph
a
b
Compd
b
Ki (
ꢀM)
Ratio
a
n
R
4
R
2
R
3
PDE1
PDE4A
> 100
1
9
0.065
> 1500
20
21
22
0.0048
0.029
9.9
44
2063
1517
1030
0.0033
3.4
2
2
3
4
0.0023
0.0060
1.5
4.1
652
683
OH
2
2
5
6
0.014
12
857
O
0.39
0.25
13
33
Ph
27
> 100
> 400
OBn
ACS Paragon Plus Environment

Page 11 of 51
Journal of Medicinal Chemistry
0.0080
30
3750
1529
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
28
2
9
0.00085
1.3
a
b
K values are the means of at least two determinations. Ratio of K
i
i
(PDE4A)/K (PDE1).
i
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To mimic the ribose moiety of cGMP and create additional interactions with the amino acid residues
located in the catalytic site of PDE1, compounds 30–47 with various substituents on the pyrazole ring
were synthesized, as listed in Table 2. If these compounds and cGMP were to bind to PDE1 in a similar
binding pose, substitution at Nꢀ1 position would correspond to the anchoring position of the ribose
group of cGMP. However, compared to the unsubstituted compound 12, no apparent improvement on
PDE1 binding affinity was observed for compounds 30 and 31, both containing N1ꢀsubstituted alkyl
groups. Unexpectedly, the N1ꢀ(6ꢀfluoropyridinꢀ2ꢀyl)methyl substituted analog 32 exhibited 25ꢀfold inꢀ
crease in PDE4A inhibitory potency while its PDE1 inhibitory potency was only improved 2ꢀfold (32 vs.
1
2). Compounds 33–35 were designed with an extended linker between the phenyl group and the pyraꢀ
zolopyrimidinone core, yet this kind of structural modifications had little effect on PDE1 binding affiniꢀ
ty. 34 tended to inhibit PDE4A more potently, while the trifluoromethyl analog 35 was a relatively more
potent PDE1 inhibitor and demonstrated 33ꢀfold selectivity over PDE4A. An interesting SAR was obꢀ
served when we compared N1ꢀ versus N2ꢀsubstituted isomers. Compound 36 was 22 times more potent
than 12 in binding to PDE1. Remarkably, its N2ꢀsubstituted isomer 37 exerted over 730ꢀfold increase in
PDE1 binding affinity relative to 12 and exhibited over 630ꢀfold selectivity against PDE4A. To follow
up on such an interesting finding, a number of compounds with N2ꢀsubstituted analogs were syntheꢀ
sized. Compound 38, containing a pꢀtrifluoromethyl benzyl sidechain at N2ꢀposition, demonstrated subꢀ
nanomolar PDE1 binding affinity and excellent selectivity over PDE4A. By replacing the trifluoromeꢀ
thyl group with a hydrophilic methylsulfonyl group, the resulting compound 40 maintained subꢀ
nanomolar PDE1 inhibitory potency, yet its selectivity over PDE4A decreased 8ꢀfold. Its N1ꢀisomer 39
showed similar PDE4A selectivity, but its PDE1 inhibitory activity was weakened 61ꢀfold. Sulfonamide
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 51
4
1 and pyrrolidine derivatives 42 exhibited similar potency and selectivity. Nꢀmethylation of the pyrrolꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
idine ring had little effect on compound binding profile (43 vs. 42). By introducing a paraꢀfluoro biꢀ
phenylmethyl sidechain, the binding affinity of compound 45 was increased over 3400ꢀfold relative to
1
2. This subꢀnanomolar inhibitor was further optimized by replacing the terminal benzene ring with a 2ꢀ
pyridyl group to adjust its physicochemical properties. The resulting analog 47 is a picomolar PDE1 inꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hibitor with a K of 85 pM and with over 2800ꢀfold selectivity against PDE4A. The SAR of these comꢀ
i
pounds clearly demonstrates that N2ꢀsubstitution of the polycyclic compounds is preferred. For instance,
45 and 47 are hundreds of times more potent than their N1ꢀisomers, 44 and 46. In addition, paraꢀ
substituted benzyl sidechains generally provide compounds with excellent potency and selectivity.
Table 2. SAR of Pyrazolo[3,4-d]pyrimidine Derivatives with Variations of Substituents on the A-
Ring
O
O
NHPh
NHPh
3
3
N
2
N
2
B
N
A N
B
N
A N R
N
N
1
N
N
C
D
C
D
1
R
N1-Substitution
N2-Substitution
a
b
Compd
Isomer
R
H
Ki (
ꢀM)
Ratio
PDE1
PDE4
9.8
1
3
2
0
–
1.4
7.0
2.6
N1
0.77
2.0
8.4
3
3
3
1
2
3
N1
N1
O
F
5.0
1.7
0.67
0.39
0.58
N
N1
N1
N1
N1
N2
1.1
2.0
4.1
0.86
18
3.7
0.43
33.3
11
34
O
3
3
3
5
6
7
CF3
0.54
CF
CF
3
3
0.065
0.0019
0.70
1.2
N
N
632
ACS Paragon Plus Environment

Page 13 of 51
Journal of Medicinal Chemistry
3
3
4
4
8
9
0
1
N2
N1
N2
N2
N2
N2
CF
3
0.00088
0.030
0.93
3.0
1057
100
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
S
CH
CH
3
3
O
O
S
0.00049
0.0014
0.0016
0.0027
0.065
1.2
133
O
O
S
NH2
857
O
H
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
42
1.2
750
N
4
3
3.1
1148
4
4
4
4
5
6
N1
N2
N1
F
F
0.24
0.00041
0.042
> 100
0.46
4.7
> 410
1122
112
N
4
7
N2
0.000085
0.24
2824
N
a
b
K values are the means of at least two determinations. Ratio of K (PDE4A)/ K
i
i
i
(PDE1).
According to the available Xꢀray structures of PDE enzymes and their binding patterns with cGMP
1
5,16
and PDE inhibitors mimicking guanine,
the Nꢀ7 nitrogen in the purine ring usually forms a hydrogen
bond with an adjacent amino acid residue. Changing the purine ring system into a pyrazolo[3,4ꢀ
d]pyrimidinone scaffold would likely disrupt the hydrogen bond network, especially for the key hydroꢀ
gen bond with the conserved activeꢀsite glutamine, so an amino group was introduced at position 3 of
the designed scaffold in an attempt to reꢀestablish a preferred hydrogen bond pattern. To understand the
SAR and further improve compound binding profiles, a list of 3ꢀamino substituted compounds were
synthesized as shown in Table 3. Compound 48 inhibited PDE1 with a K of 5.8 nM. Introducing a meꢀ
i
thyl group on the NH caused 5ꢀfold decrease of PDE1 binding affinity (49 vs. 48). Capping the nitroꢀ
2
gen with two methyl groups resulted in 21ꢀfold drop in binding potency (50 vs. 48). Significant lessenꢀ
ing of PDE1 binding affinity was also observed for pyrrolidinyl substituted analog 57, which suggested
that the 3ꢀamino group might be involved in the formation of a hydrogen bond and probably served as a
hydrogen donor. When a larger alkyl group was substituted on the nitrogen, the PDE1 binding affinity of
the resulting compounds, 51–54 was improved as the size of the alkyl substituents increased. Similarly,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 51
cyclohexyl substituted 56 showed higher potency than cyclopentyl substituted 55. Remarkably, comꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pound 3 with a phenylamino group at position 3 exhibited potent PDE1 inhibitory activity (K = 58 pM),
i
which was 100ꢀfold more potent than the 3ꢀamino substituted analog 48. Replacing the phenyl group
with a 4ꢀpyridyl group caused over 500ꢀfold decrease in PDE1 binding potency without significant
change in the PDE4A binding affinity (58 vs. 3). Insertion of a methylene group between the NH and
benzene ring led to 84ꢀfold decline in PDE1 binding potency (59 vs. 3). NꢀMethylation of the PhNH
group resulted in an over 300ꢀfold drop in PDE1 binding affinity while PDE4A binding potency was
largely maintained (60 vs. 47). It was apparent that an arylamino substituent was preferred at position 3
compared to alkylamino, cycloalkylamino and heteroarylamino groups. An uncapped NꢀH at this posiꢀ
tion was essential for binding to PDE1. N,Nꢀdisubstituted analogs gave significantly weaker PDE1 bindꢀ
ing affinities.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. SAR of Pyrazolo[3,4-d]pyrimidine Derivatives with Variations of Substituents at Position
3
O
R
3
N
2
B
N
A N
N
N
C
D
1
N
R
1
b
a
Compd
R
R1
Ki (ꢀM)
Ratio
PDE1
0.0058
0.028
PDE4
2.4
4
4
5
8
9
0
NH
2
F
F
F
F
F
F
F
F
F
F
F
414
154
9.2
MeNH
4.3
Me
N
0.12
1.1
2
n
51
PrNH
PrNH
BuNH
0.015
6.2
413
458
851
2000
229
571
29
i
i
5
2
0.012
5.5
53
0.0074
0.0013
0.024
6.3
5
4
2ꢀmethylbutylꢀNH
cyclopentylꢀNH
cyclohexylꢀNH
Nꢀpyrrolidinyl
PhNH
2.6
55
5.5
5
6
0.0028
0.16
1.6
57
3
4.6
0.000058
0.16
2759
ACS Paragon Plus Environment

Page 15 of 51
Journal of Medicinal Chemistry
5
8
4ꢀpyridylꢀNH
PhCH NH
PhNMe
F
F
H
0.031
0.37
4.0
12
816
24
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
59
2
0.0049
6
a
0
0.031
b
0.74
K values are the means of at least two determinations. Ratio of K (PDE4A)/ K (PDE1).
i
i
i
We determined the Xꢀray crystal structure of the catalytic domain of the human recombinant PDE1B
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
complexed with compound 3 to identify interactions between PDE1 and this inhibitor and to further exꢀ
plore SAR around this compound. As shown in Figure 2 and S1A, the pyrazolo[3,4ꢀd]pyrimidinone core
is held tightly by a "hydrophobic clamp" formed primarily by F424, L388 and F392 located above and
below the aromatic core of 3. The well positioned π–π stacking and hydrophobic interactions between
these contact residues and the planar core contribute significantly to the picomolar bind affinity of comꢀ
15,16
pound 3. Compound 3 appears to adopt a binding pose different from that of cGMP.
In compound 3,
the exocyclic carbonyl group of pyrimidone and the amino group at position 3 form bidentate hydrogen
bonds with the ꢀcarboxamide of the conserved Q421 (Figure 3 and S1B). A waterꢀmediated hydrogen
γ
bond is observed between the guanidyl nitrogen on the Bꢀring with the bound metal zinc. The involveꢀ
ment of the NꢀH group at position 3 of the Aꢀring in hydrogen bond formation explains our observations
that N,Nꢀdisubstituted compounds, such as 50, 57 and 60, were significantly weaker inhibitors than
monoꢀsubstituted analogs. In addition, replacing this NꢀH at position 3 with O, S, or CH caused 35–90
2
fold decrease in PDE1 binding affinity (72, 74–75 vs. 3), as shown in Table 4. When the key carbonyl
group of the Bꢀring in 38 and 47 was reduced into a methylene group, the PDE1 binding affinities of the
resulting analogs 61 and 65 were weakened 50ꢀ and 294ꢀfold, respectively. The oxygen of this carbonyl
group serves as a hydrogen acceptor in the hydrogen bonding formation with Q421. Converting the carꢀ
bonyl group into an imine, a hydrogen donor, would prevent the formation of this key hydrogen bond
and possibly disrupt other interactions. There was an over 3000ꢀfold decrease in PDE1 binding affinity
caused by this structural modification (62 vs. 38).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Crystal structure of the catalytic domain of human recombinant PDE1B enzyme com-
plexed with compound 3. Hydrogen bonds between 3 and the invariant Q421 of PDE1B were indicated
in orange dashed lines. The pyrazolopyrimidinone core of 3 was sandwiched between residues F424,
L388 and F392, forming π–π stacking with F424, π–edge interactions with F392 and hydrophobic interꢀ
action with L388.
Figure 3. Ligplot presentation of PDE1B complexed with compound 3. Key hydrogen bonds and
contact residues located in the catalytic site of PDE1B are highlighted.
ACS Paragon Plus Environment

Page 17 of 51
Journal of Medicinal Chemistry
The phenylamino group at position 3 of Aꢀring is buried deeply in a hydrophobic subꢀpocket formed
primarily by F392, L388, M389, V417, S420, T385 and L409 (Figure S1C). The π–π edge interactions
with F392 and additional hydrophobic interactions with these residues provide 3 with greater binding
affinity than alkylamino substituted analogs (Table 3 and Figure 2). The cyclopentyl group fused on the
Cꢀring with a stereochemistry of (6aR, 9aS) bends down and fits nicely into the catalytic pocket and has
van der Waals interactions with M336 and several other residues. Compounds 3 and its close analog 47
exhibited 62ꢀ and 93ꢀfold higher affinity than the corresponding enantiomers 70 and 66 with the ring
fusion carbons adopting (6aS, 9aR) configuration. These enantiomers may not fit in the binding pocket
well since the cyclopentyl moiety of these molecules may collide with the amino acid residues in the
adjoining catalytic pocket primarily formed by M336, H223 and I371, and consequently forcing the
compounds to adopt less favorable binding poses. The Nꢀmethyl group at position 5 inserts into a small
cavity, surrounded by I371, H373, F424 and Y222 (Figure 3). The additional interactions of the Nꢀ
methyl group with these residues supports the observed SAR that Nꢀmethylated compound 38 was 330
times more potent in PDE1 inhibitory activity than its desmethyl analog 63. Interestingly, Nꢀ
demethylation did not cause significant change in the PDE4 inhibitory potency. Compound 63 became a
PDE1 and PDE4A dual inhibitor by way of losing PDE1 inhibitory activity, as shown in Table 4. The Nꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
24
methyl group will prevent potential tautomerization of the pyrimidone Bꢀring into pyrimidinol, which
would disrupt the important hydrogen bonding of the exocyclic carbonyl group with Q421. The pheꢀ
nylene and pyridyl groups of the 4ꢀ(6ꢀfluoropyridinꢀ2ꢀyl)benzyl sidechain of 3 each had π–π interacꢀ
tions with F424 and F427, respectively, in addition to various van der Waals interactions with other resꢀ
idues such as G423, S420, F340 and I428 (Figure 2, 3 and S1D). These additional interactions created
by the N2ꢀsubstituted sidechain provide 3 with over 24000 times more potent binding affinity to PDE1
relative to unsubstituted analog 12. It was evident that the observed SAR of these polycyclic pyraꢀ
zolopyrimidinones and the key interactions between PDE1B and the bound ligand 3 revealed by the
crystal structure described above are well correlated with each other. The planar pyrazolopyrimidinone
core, 3ꢀphenylamino group, paraꢀsubstituted benzyl sidechain at N2ꢀposition and the fused cyclopentyl
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 51
group with an absolute configuration of (6aR, 9aS), all provide major contributions to the subꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
nanomolar PDE1 binding potency of this series of compounds.
Table 4. SAR of Tetracyclic Pyrazolo[3,4-d]pyrimidine Derivatives.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
W
3
R
2
5
N
R
N
Z
2
B
N
A N
N
C
1
6
a
9a
D
R
1
a
b
Compd
Z
W
R
Stereochemistry
6a 9a
R
1
R
2
K
i
(
ꢀM)
Ratio
HLM Clint
ml/min/kg)
(
PDE1
PDE4A
6
6
3
6
6
4
6
6
1
2
8
3
4
7
5
6
CH
2
NH
Me
Me
Me
H
R
R
R
R
R
R
R
S
S
S
S
S
S
S
S
R
S
S
S
S
R
S
S
S
S
S
S
CF
3
3
3
3
3
H
0.044
2.8
37
18
841
6.4
64
32
C=NH
C=O
C=O
C=O
C=O
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
O
CF
CF
CF
CF
H
H
H
F
0.00088
0.29
0.93
0.55
0.42
0.24
18
1057
2
139
22
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
0.00063
0.000085
0.025
667
2824
720
1585
2740
5323
6757
2759
1556
413
83
73
pyridinꢀ2ꢀyl
pyridinꢀ2ꢀyl
pyridinꢀ2ꢀyl
H
H
H
H
H
H
H
H
F
189
59
CH
2
C=O
C=O
C=O
C=O
C=O
C=O
C=O
C=O
C=O
C=O
C=O
C=O
0.0053
0.000073
0.000062
0.000074
0.000058
0.0054
0.00015
0.0030
0.0066
0.0052
0.0022
0.0023
8.4
72
67
68
69
3
R
R
R
R
S
3ꢀfluoropyridinꢀ2ꢀyl
4ꢀfluoropyridinꢀ2ꢀyl
5ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
6ꢀfluoropyridinꢀ2ꢀyl
0.20
0.33
0.50
0.16
8.4
276
218
125
78
7
7
7
7
7
7
0
1
2
3
4
5
6
K
51
R
R
R
R
R
R
0.062
0.25
0.22
0.93
0.77
0.78
107
214
266
208
165
81
H
F
O
33
S
H
H
F
179
350
339
CH
2
7
a
CH
2
b
i
i i
values are the means of at least two determinations. Ratio of K (PDE4A)/ K (PDE1).
Table 5. PDE Family Selectivity of Selected PDE1 Inhibitors.
a
Compd
i
K (ꢀM)
PDE1
PDE2A
37
PDE3B
1.5
PDE4A
1.3
PDE5A
1.6
PDE6
1.1
PDE7B
0.91
0.30
1.2
PDE8A
33
PDE9A
>100
6.1
PDE10A
7.9
PDE11A
7.1
2
4
3
3
9
7
8
7
0.00085
0.000085
0.00088
0.0019
24
1.2
0.24
0.93
1.2
0.60
0.59
1.2
0.43
1.6
8.2
2.7
2.6
39
30
54
50
5.4
28
> 100
18
42
5.5
6.1
> 100
3.0
> 100
17
19
72
3
0.000058
3.1
0.16
0.63
0.32
0.36
1.8
1.3
a
K
i
values are the means of at least two determinations.
Metabolic stability, profiling and pharmacokinetics. Metabolic stability studies using human liver
microsomes (HLM) showed that the subꢀnanomolar PDE1 inhibitors 38 and 47 process relatively high
ACS Paragon Plus Environment

Page 19 of 51
Journal of Medicinal Chemistry
microsomal clearance, as shown in Table 4. Their analogs, 62, 63 and 65, exhibited lower HLM clearꢀ
ance, but with a significant loss in PDE1 binding potency. To improve compound metabolic stability
while maintaining potency, the paraꢀposition of the potentially metabolically labile phenylamino group
at position 3 of the Aꢀring was capped with a fluoro group. The fluoroꢀcapped analog 64 showed similar
potency with 38, and demonstrated slightly lower HLM clearance. A similar trend was observed for the
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
ꢀbenzyl substituted pair of compounds 75 and 76. Clearance of paraꢀfluoro analog 73, however, was
not improved over 72. Fluorination of the phenylamino group on the Aꢀring of compound 3 caused a 7ꢀ
fold drop in PDE4A selectivity and a slight decrease in PDE1 binding potency, without a significant efꢀ
fect on microsomal stability (71 vs. 3). Another attempt to improve metabolic stability was made by a
fluorine scan of the 4ꢀ(pyridinꢀ2ꢀyl)benzyl group, which was the preferred sidechain at the N2 position,
enabling 47 to bind to PDE1 with an K of 0.000085 ꢀM. Compounds 67–69 and 3 with a fluoro group
i
substituted at different positions of the pyridine ring were evaluated in the HLM stability assay. Comꢀ
pound 3 demonstrated relatively lower clearance than 47 and other fluorinated analogs while maintainꢀ
ing great potency and selectivity, as shown in Tables 4 and 5. Compounds 3 and 47 both exhibited exꢀ
cellent selectivity over other PDE family members. These compounds did not show meaningful PDE1
subꢀtype selectivity. For instance, the K values of compound 3 against recombinant fullꢀlength human
i
PDE1A, PDE1B and PDE1C are 0.000033
ꢀM, 0.00038 ꢀM and 0.000035 ꢀM, respectively. Comꢀ
pound 3 was profiled in a panel of enzymes, receptors, transporters and ion channels from Caliper at 10
ꢀ
M, which was over 170,000 times higher than its K for PDE1, and demonstrated good selectivity.
i
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 51
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 6. Physicochemical and Pharmacokinetic Properties of Selected PDE1 Inhibitors.
Compd
2
9
47
41
38
64
74
72
73
75
76
3
LogD at pH 7.4
LogD at pH 1.7
Solubility in PBS (
>5
1.92
2
3.54
1.96
4.47
1.16
35
4.57
1.11
31
> 5
2.49
<1
4.61
1.05
8
4.77
1.21
7
4.96
1.36
<1
>5
4.41
1.27
22
ꢀ0.58
>100
ꢀ0.96
>100
1.42
<1
ꢀM)
a
Mouse PK parameters
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Plasma Cmax(obs) (ng/mL) 130
Brain Cmax(obs) (ng/mL) 474
Plasma AUClast (ngꢀh/mL) 286
48
28
65
21
0.3
650
16
60
21
93
165
170
210
192
0.9
88
86
44
490
387
826
588
0.7
285
122
548
4.5
133
24
71
64
101
134
154
1.2
81
785
15
108
81
119
102
0.9
69
Brain AUClast (ngꢀh/mL) 542
166
6.9
109
1.6
b
Brain/Plasma ratio
2.9
0.02
0.8
c
Rat PK parameters
iv
t
1/2 (h)
0.72
2.48
40
1.4
7.44
62
1.5
7.48
57
1.4
Vd (L/kg)
6.83
55
CL (mL/min/kg)
AUC
∞
(ngꢀh/mL)
1/2 (h) 3.1
max (ng/mL) 102
Tmax (obs) (h) 4.0
420
1.5
271
1.5
75
292
2.2
42
300
1.5
po
t
C
158
0.50
636
16
252
0.75
802
28
0.25
214
8
0.50
164
6
AUC
∞
(ngꢀh/mL) 816
F (%)
a
Screening mode PK studies in male C57BL/6 mice. Compounds in 0.05M pH 4.0 citrate buffer containing 5% Labrasol and 5% TPGS were administered to
b
c
mice at 10 mg/kg via oral gavage. Ratio of brain AUC0ꢀ2h/plasma AUC0ꢀ2h. Male SpragueꢀDawley rats were administered with test compounds in 0.05M pH 3.1
citrateꢀphosphate buffer at 1 mg/kg by intravenous injection or in 0.02 N HCl at 9.75 mg/kg by oral gavage.
The pharmacokinetic properties of selected compounds were evaluated in mice after oral administraꢀ
tion, as summarized in Table 6. All of the tested compounds were orally bioavailable. The subꢀ
nanomolar PDE1 inhibitor 29 exhibited good brain exposure with a brain to plasma ratio of 2.9, but this
hydrophobic compound has poor physicochemical properties, as shown in Table 6, and was not suitable
for further development. When a nitrogen was introduced into the terminal phenyl ring on the sidechain
of 29, the resulting compound 47 showed 10 times more potent PDE1 binding affinity. Moreover, 47
exhibited much better solubility in PBS buffer and had appropriate physicochemical properties for a
CNS drug candidate, as shown in Table 6. The plasma and brain exposure of this compound is lower
than that of 29 giving it a brain to plasma ratio of 0.3. Compound 47 has two basic pKa′s (7.0 and 4.3).
In contrast, compound 3 has a single basic pKa (7.0), as fluorination on the αꢀcarbon of the pyridine ring
prevents ionization of the sidechain. The fineꢀtuning of physicochemical and ionization properties proꢀ
vided 3 with improved brain exposure and better metabolic stability in human liver microsomes, as well
ACS Paragon Plus Environment

Page 21 of 51
Journal of Medicinal Chemistry
as a preferred brain to plasma ratio of 0.7 in mice after oral dosing. Oral bioavailability of compounds in
Table 6 was determined in a screening mode using one formulation. In general, formulations can be opꢀ
timized further on a compoundꢀspecific basis. Compound 3 exhibited higher bioavailability than 47 in
rats, as shown in Table 6. Compound 41 is a more hydrophilic compound with a LogD of 1.96 at pH7.4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
and a large topological polar surface area (TPSA, 134 Å ) due to the polar sulfonamide group on the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
sidechain. These properties result in this compound having very good plasma exposure, but limited
brain penetration (brain/plasma ratio = 0.02). Such a compound is suitable for nonꢀCNS indications. The
hydrophobic analog 38 with a pꢀtrifluorobenzyl sidechain exhibited a brain to plasma ratio of 4.5 in
mice, but this compound has high clearance rates both in vivo and in human liver microsomes. Comꢀ
pound 38 showed low oral bioavailability in rats (Table 6). Capping the paraꢀposition of the potentially
metabolically labile phenylamino group with a fluoro group resulted in relatively lower oral exposure
(64 vs. 38). Replacing the key phenylamino group of 3 with an arylthio, aryloxy or benzyl group not onꢀ
ly led to decreased PDE1 inhibitory potency, but also resulted in poor physicochemical properties and
low plasma and brain exposure after oral administration in mice (72–76 vs. 3). Among all of these tested
compounds, compound 3 possesses a good overall profile with balanced physicochemical properties,
excellent potency and selectivity and good pharmacokinetics.
In vivo Studies. Compound 3 was studied in various animal models of motor and cognitive functions
to evaluate its in vivo efficacy before being nominated for clinical development. As one of the proof of
concept studies, compound 3 was evaluated for effects on memory performance using the rat novel obꢀ
ject recognition (NOR) test. Compound 3 was found to significantly enhance memory performance in
the test with a minimum effective dose of 3 mg/kg (Figure 4). As is customary in this assay, an inverted
2
5
Uꢀshaped dose response curve was observed. An extensive pharmacological profiling of compound 3
using animal models of cognitive dysfunction and other CNS disorders has been conducted and comꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 51
pound 3 exhibited good efficacy in these models after oral administration. These results will be reported
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
elsewhere.
4
0
5
F amiliar
N o vel
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
*
30
*
25
20
15
10
5
0
Vehicle
1
3
6
10
Dosage (mg/kg, IP)
Figure 4. Effect of compound 3 on cognitive performance of male Sprague-Dawley rats in the nov-
el object recognition task. Data are expressed as mean (with SEM). * indicates a statistical differ-
ence between the familiar and novel object (p < 0.05).
CONCLUSION
A novel series of 3ꢀaminopyrazolo[3,4ꢀd]pyrimidinones has been designed and synthesized as PDE1
inhibitors. Systematic optimization of this scaffold led to the discovery of compound 3, a picomolar
PDE1 inhibitor with excellent selectivity against the members of other PDE families, PDE2 – 11. The
interactions between compound 3 and the contact residues located in the catalytic pocket of PDE1 were
elucidated based upon the observed SAR and the high resolution crystal structure of PDE1B complexed
with compound 3. Compound 3 possesses good physicochemical and pharmacokinetic properties for a
CNSꢀactive drug candidate, and showed efficacy in various animal models of motor and cognitive funcꢀ
tions. In multiple human Phase 1 clinical trials compound 3 has demonstrated a favorable safety profile
and has been generally wellꢀtolerated across a broad range of doses both in healthy volunteers and in
ACS Paragon Plus Environment

Page 23 of 51
Journal of Medicinal Chemistry
patients with schizophrenia. The pharmacokinetic profile of compound 3 supports once daily dosing,
being rapidly absorbed with a mean plasma halfꢀlife of approximately 14 to 18 hours. More detailed
preclinical and clinical results of 3 will be reported in other forums.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL SECTION
Chemistry. Unless otherwise noted, all materials were obtained from commercial suppliers and used
without further purification. All reactions involving airꢀ or moistureꢀsensitive reagents were performed
under an argon atmosphere. All microwave assisted reactions were conducted with an Initiator Eight
EXP Microwave System from Biotage, Charlotte, NC, USA. Silica gel or alumina column chromatogꢀ
raphy was performed with a CombiFlash Companion purification system from Teledyne ISCO, Lincoln,
NE. All compounds synthesized, unless otherwise noted, were purified by column chromatography
and/or preparative reverse–phase HPLC. A Waters preparative HPLC system equipped with a Delta
6
00EF pump, 2996 PDA detector and WFCIII fraction collector was used for compound purification
using the following general HPLC method. Column: Gemini, AXIA packed, 10 m C18 110Å, 250 ×
1.2 mm; mobile phase A, 0.1% formic acid in water; mobile phase B, 0.1% formic acid in acetonitrile;
ꢀ
2
gradient was adjusted and optimized based on compound polarity; run time 22 min; flow rate 24
1
mL/min; detection at 210–350 nm. HꢀNMR spectra were determined in the cited solvent on a Bruker
1
3
DRX 300 or Avance III (400 MHz or 500 MHz) spectrometer. C NMR spectra were recorded at 75 or
126 MHz. Chemical shifts are reported in delta ( ) units, parts per million (ppm) downfield from tetraꢀ
δ
methylsilane. Coupling constants are reported in hertz (Hz). Splitting patterns are designated as follows:
s, singlet; br, broad singlet; d, doublet; t, triplet; q, quartet; m, multiplet. Purity of all final products was
>
95% as determined by reverse–phase HPLC on a Waters Alliance 2695 system with a 2996 PDA deꢀ
tector. HPLC method: Column, XBridge Shield RP18 column, 150 mm × 4.6 mm, 5 m, 30 °C; mobile
ꢀ
phase A, 0.02 mol/L phosphate buffer (pH 6.8)/acetonitrile (13/7, v/v); mobile phase B, acetonitrile/0.02
mol/L phosphate buffer (pH 6.8) (7/3, v/v); gradient, 0.0–30 min, 0–100% B; 30–30.1 min, 100–0% B;
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 51
3
0.1–40 min, 0% B; flow rate 1.0 mL/min; detection at 225 nm. Mass spectral (MS) data were deterꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mined on a Micromass Quattro micro API Mass, or LCT Premier XE spectrometer from Waters. The
chemical yields reported below are not optimized and correspond as specific examples of a single prepaꢀ
ration.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Representative synthetic procedures of polycyclic pyrazolopyrimidine derivatives shown in
Scheme 1. 2-(Bi-phenyl-4-ylmethyl)-7,8,9-trihydro-5-methyl-3-(phenylamino)-[2H]-pyrazolo[4,3-
e]pyrimido-[1,2-a]pyrimidin-4(5H)-one
methylpyrimidineꢀ2,4(1H,3H)ꢀdione. A mixture of 6ꢀchloroꢀ3ꢀmethylpyrimidineꢀ2,4(1H,3H)ꢀdione (4)
8.46 g, 52.7 mmol), pꢀmethoxybenzyl chloride (7.9 mL, 58.2 mmol) and potassium carbonate (7.3 g,
(19).
Step
a.
6ꢀChloroꢀ1ꢀ(4ꢀmethoxybenzyl)ꢀ3ꢀ
(
58.9 mmol) in anhydrous DMF (100 mL) was heated at 60 °C for an hour. After hot filtration, the filꢀ
trate was evaporated to dryness under reduced pressure. The residue was crystallized from hexane to
give crude 6ꢀchloroꢀ1ꢀ(4ꢀmethoxybenzyl)ꢀ3ꢀmethylpyrimidineꢀ2,4(1H,3H)ꢀdione as a yellowish crystalꢀ
line solid (13.8 g, 93% yield), which was used directly in the next step without further purification. A
small amount of the crude material was further purified by silicaꢀgel column chromatography using a
1
gradient of 0 – 50% ethyl acetate in hexane as eluent to afford the pure product as a crystalline solid. H
NMR (500 MHz, CDCl3)
δ
7.30 (d, J = 8.6 Hz, 2H), 6.87 (d, J = 8.6 Hz, 2H), 5.93 (s, 1H), 5.21 (s, 2H),
+
3
.80 (s, 3H), 3.35 (s, 3H). MS (ESI) m/z 281.1 [M+H] .
Step b. 6ꢀHydrazinylꢀ1ꢀ(4ꢀmethoxybenzyl)ꢀ3ꢀmethylpyrimidineꢀ2,4(1H,3H)ꢀdione (5). To a solution of
the crude 6ꢀchloroꢀ1ꢀ(4ꢀmethoxybenzyl)ꢀ3ꢀmethylpyrimidineꢀ2,4(1H,3H)ꢀdione (2.4 g, 8.6 mmol) in
EtOH (25 mL) and MeOH (50 mL) was added anhydrous hydrazine (1.2 mL). The reaction mixture was
refluxed for 3 h and then cooled to room temperature. A large amount of ether was added into the reacꢀ
tion mixture and then filtered. The filter cake was dried under vacuum to give the title compound as a
1
crystalline solid (2.0 g, 84% yield). H NMR (500 MHz, DMSOꢀd )
δ 8.07 (s, 1H), 7.15 (d, J = 8.5 Hz,
6
2
H), 6.88 (d, J = 8.5 Hz, 2H), 5.14 (s, 1H), 5.02 (s, 2H), 4.40 (br, 2H), 3.72 (s, 3H), 3.12 (s, 3H). MS
+
(ESI) m/z 277.1 [M+H] .
ACS Paragon Plus Environment

Page 25 of 51
Journal of Medicinal Chemistry
Step c. 7ꢀ(4ꢀMethoxybenzyl)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀ
1
2
dione (6). To a suspension of 6ꢀhydrazinylꢀ1ꢀ(4ꢀmethoxybenzyl)ꢀ3ꢀmethylpyrimidineꢀ2,4(1H,3H)ꢀdione
(10.1 g, 36.6 mmol) in DMF (100 mL) was added phenyl isothiocyanate (19.8 g, 14.7 mmol). The reacꢀ
tion mixture was heated at 115°C for 3 days and then evaporated to dryness under reduced pressure. The
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
residue was washed with hexane (2 × 200 mL) and then crystallized from methanol to give the title
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
compound as a crystalline solid (9.9 g, yield, 72%). H NMR (500 MHz, CDCl )
δ
7.53 (s, 1H), 7.43 (d,
3
J = 8.7 Hz, 2H), 7.40 – 7.34 (m, 2H), 7.20 (d, J = 8.4 Hz, 2H), 7.18 – 7.12 (m, 1H), 6.83 (d, J = 8.8 Hz,
+
2H), 5.05 (s, 2H), 3.77 (s, 3H), 3.35 (s, 3H). MS (ESI) m/z 378.3 [M+H] .
Step d. 2ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ7ꢀ(4ꢀmethoxybenzyl)ꢀ5ꢀmethylꢀ3ꢀ(phenylꢀamino)ꢀ2Hꢀpyrazolo[3,4ꢀ
d]pyrimidineꢀ4,6(5H,7H)ꢀdione (7; R = Ph). To a solution of 7ꢀ(4ꢀmethoxybenzyl)ꢀ5ꢀmethylꢀ3ꢀ
1
phenylaminoꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀdione (24 mg, 0.064 mmol) and 4ꢀ
(bromomethyl)biphenyl (17 mg, 0.070 mmol) in acetone (3.0 mL) was added potassium carbonate (8.8
mg, 0.064 mmol). The reaction mixture was stirred at room temperature overnight. After the solvent
was removed under reduced pressure, the residue was purified by silicaꢀgel column chromatography usꢀ
ing a gradient of 0 – 75% ethyl acetate in hexane as eluent to afford the title compound as a white solid
1
(24 mg, 69% yield). H NMR (300 MHz, CDCl )
δ
7.56 (d, J = 7.4 Hz, 2H), 7.52 – 7.39 (m, 6H), 7.39 –
3
7
.26 (m, 3H), 7.15 – 7.02 (m, 3H), 6.94 (d, J = 7.8 Hz, 2H), 6.85 – 6.70 (m, 3H), 5.10 (s, 2H), 4.99 (s,
+
2H), 3.72 (s, 3H), 3.32 (s, 3H). MS (ESI) m/z 544.3 [M+H] .
Step e. 2ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀ
dione (8; R = Ph). To a solution of 2ꢀ(biphenylꢀ4ꢀylmethyl)ꢀ7ꢀ(4ꢀmethoxybenzyl)ꢀ5ꢀmethylꢀ3ꢀ(phenylꢀ
1
amino)ꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀdione (3.4 g, 6.25 mmol) in trifluoroacetic acid (28
mL) was added trifluoromethanesulfonic acid (1.11 mL, 12.5 mmol). The reaction mixture was stirred at
room temperature overnight. After TFA was removed under reduced pressure, the residue was treated
with water (100 mL), followed by 28% ammonium hydroxide (10 mL). The resulting precipitate was
filtered, washed with water (4 × 50 mL), and then dried under vacuum. The obtained crude product was
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 51
further purified by silicaꢀgel column chromatography using a gradient of 0 – 100% ethyl acetate in hexꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
ane as eluent to afford the title compound as a white solid (2.6 g, 96% yield). H NMR (500 MHz,
CDCl3)
δ 9.22 (s, 1H), 7.52 (d, J = 7.7 Hz, 2H), 7.46 (d, J = 8.1 Hz, 2H), 7.44 – 7.38 (m, 2H), 7.38 –
7
.28 (m, 3H), 7.16 – 7.11 (m, 1H), 7.08 (d, J = 8.1 Hz, 2H), 6.97 (d, J = 8.0 Hz, 2H), 6.83 (s, 1H), 5.01
+
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(s, 2H), 3.33 (s, 3H). MS (ESI) m/z 424.2 [M+H] .
Step f. 2ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ6ꢀchloroꢀ5ꢀmethylꢀ3ꢀ(phenylamino)ꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ
4(5H)ꢀone (9; R1
=
Ph). 2ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ2Hꢀpyrazolo[3,4ꢀ
d]pyrimidineꢀ4,6(5H,7H)ꢀdione (2.4 g, 5.67 mmol) was suspended in POCl (80 mL). The mixture was
3
heated at reflux overnight and then evaporated under reduced pressure. The residue was treated with waꢀ
ter and ethyl acetate and then adjusted to pH = 9 with 28% ammonium hydroxide. The aqueous phase
was extracted with ethyl acetate (3 × 100 mL), and the combined organic phase was evaporated to dryꢀ
ness. The residue was purified by silicaꢀgel column chromatography using a gradient of 0 – 75% ethyl
1
acetate in hexane as eluent to afford the title compound as a white solid (1.4 g, 56% yield). H NMR
(500 MHz, CDCl3)
δ
7.53 (d, J = 7.6 Hz, 2H), 7.50 – 7.39 (m, 4H), 7.37 – 7.32 (m, 1H), 7.31 – 7.26 (m,
2
3
H), 7.14 (d, J = 8.1 Hz, 2H), 7.11 – 7.04 (m, 1H), 6.84 (d, J = 7.9 Hz, 2H), 6.43 (s, 1H), 5.17 (s, 2H),
+
.61 (s, 3H). MS (ESI) m/z 442.2 [M+H] .
Step g. 2ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ6ꢀ(3ꢀhydroxypropylamino)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ2Hꢀpyrazolo[3,4ꢀ
d]pyrimidinꢀ4(5H)ꢀone. A mixture of 2ꢀ(biphenylꢀ4ꢀylmethyl)ꢀ6ꢀchloroꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ2Hꢀ
pyrazolo[3,4ꢀd]pyrimidinꢀ4(5H)ꢀone (100 mg, 0.23 mmol), 3ꢀaminopropanꢀ1ꢀol (85 L, 1.1 mmol) and
N,Nꢀdiisopropylethylamine (200 L, 1.1 mmol) in DMF (3.0 mL) was heated at 110 °C overnight. After
ꢀ
ꢀ
the solvent was removed under reduced pressure, the residue was purified with a Waters semiꢀ
preparative HPLC system using a gradient of 0 – 60% acetonitrile in water containing 0.1% formic acid
1
over 16 min to afford the title compound as a white solid (80 mg, 73% yield). H NMR (500 MHz,
CDCl3)
δ 7.52 (d, J = 7.2 Hz, 2H), 7.47 – 7.38 (m, 4H), 7.36 – 7.29 (m, 1H), 7.28 – 7.24 (m, 2H), 7.15
(d, J = 8.1 Hz, 2H), 7.06 – 6.98 (m, 1H), 6.84 (d, J = 7.8 Hz, 2H), 6.48 (s, 1H), 5.85 (br, 1H), 5.09 (s,
ACS Paragon Plus Environment

Page 27 of 51
Journal of Medicinal Chemistry
2
4
H), 3.79 (t, J = 5.4 Hz, 2H), 3.75 – 3.68 (m, 2H), 3.36 (s, 3H), 1.94 – 1.81 (m, 2H). MS (ESI) m/z
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
81.2 [M+H] .
Step
h.
2ꢀ(Biphenylꢀ4ꢀylmethyl)ꢀ7,8,9ꢀtrihydroꢀ5ꢀmethylꢀ3ꢀ(phenylamino)ꢀ[2H]ꢀpyrazolo[4,3ꢀ
e]pyrimidoꢀ[1,2ꢀa]pyrimidinꢀ4(5H)ꢀone (19; R = Ph, R = R = R = H, n = 1). To a solution of thionyl
1
2
3
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
chloride (9.0 ꢀL, 0.10 mmol) in methylene chloride (3.0 mL) was added 2ꢀ(biphenylꢀ4ꢀylmethyl)ꢀ6ꢀ(3ꢀ
hydroxy propyl amino)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4(5H)ꢀone (25 mg,
0.052 mmol) at 0 °C. The reaction mixture was warmed to room temperature and stirred overnight. Afꢀ
ter the solvent was removed under reduced pressure, the residue was purified with a Waters semiꢀ
preparative HPLC system using a gradient of 0 – 35% acetonitrile in water containing 0.1% formic acid
1
over 16 min to afford the final compound as an offꢀwhite solid (13 mg, 54% yield). H NMR (500 MHz,
CDCl3)
δ 7.52 (d, J = 7.4 Hz, 2H), 7.46 (d, J = 8.2 Hz, 2H), 7.44 – 7.38 (m, 2H), 7.37 – 7.31 (m, 1H),
7.30 – 7.23 (m, 2H), 7.11 (d, J = 8.2 Hz, 2H), 7.08 – 7.02 (m, 1H), 6.91 (d, J = 7.8 Hz, 2H), 5.11 (s,
2
H), 3.96 (t, J = 6.0 Hz, 2H), 3.69 (t, J = 5.7 Hz, 2H), 3.26 (s, 3H), 2.07 – 1.93 (m, 2H). MS (ESI) m/z
+
+
4
63.2 [M+H] . HRMS (ESI) m/z calcd for C H N O [M+H] : 463.2246; found: 463.2260. HPLC puriꢀ
28 27 6
ty: 98.7%; retention time: 13.7 min.
Representative synthetic procedures of pyrazolo-pyrimidinones shown in Scheme 2. (6aR,9aS)-
5
,6a,7,8,9,9a-hexahydro-1-(cyclopentylmethyl)-5-methyl-3-phenylamino-
cyclopent[4,5]imidazo[1,2-a]pyrazolo[4,3-e]pyrimidin-4(1H)-one (30). Step a and b. 6ꢀChloroꢀ5ꢀ
methylꢀ3ꢀphenylaminoꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4(5H)ꢀone (10) To solution of 7ꢀ(4ꢀ
a
methoxybenzyl)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ1Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀdione (6.6 g, 17.5
mmol) in CH Cl (200 mL) was added TFA (30 mL) slowly at room temperature, followed by trifluoꢀ
2
2
romethanesulfonic acid (10 mL). After the addition was completed, the reaction mixture was stirred at
room temperature for 2 h. The solvent was removed under reduced pressure, and then 200 mL of 1.0 N
NaOH was added at 0 °C. The mixture was extracted with ethyl acetate five times (5 × 100 mL). The
combined organic phase was dried over Na SO , and then filtered. The filtrate was evaporated to dryꢀ
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 51
ness to give 5ꢀmethylꢀ3ꢀ(phenylamino)ꢀ2Hꢀpyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀdione as a white solꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
id (4.3 g, yield 96%). MS (ESI) m/z 258.1 [M+H] . The crude 5ꢀmethylꢀ3ꢀ(phenylamino)ꢀ2Hꢀ
pyrazolo[3,4ꢀd]pyrimidineꢀ4,6(5H,7H)ꢀdione (4.3 g, 16.7 mmol) was suspended in POCl (120 mL), and
3
then heated at reflux for 2 days. After POCl was removed under reduced pressure, the residue was susꢀ
3
pended in 100 mL of water and carefully adjusted to pH = 1 – 2 with 7.0 % NH OH at 0 °C. The mixꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
ture was then extracted with CH Cl /MeOH (10:1, v/v) five times (5 × 100 mL). The combined organic
2
2
phase was washed with water (2 × 50 mL), dried over Na SO , and then evaporated under reduced vacꢀ
2
4
uum to give crude 6ꢀchloroꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ1Hꢀpyrazolo [3,4ꢀd]pyrimidinꢀ4(5H)ꢀone (3.3 g,
yield: 72%). A small amount of the crude material was further purified by semiꢀpreparative reverse–
phase HPLC using a gradient of 0 – 55% of acetonitrile in water with 0.1% formic acid as eluent to afꢀ
1
ford the pure product as a white solid. H NMR (500 MHz, DMSOꢀd )
δ 12.98 (s, 1H), 7.96 (s, 1H),
6
7
.67 (d, J = 8.1 Hz, 2H), 7.33 – 7.18 (m, 2H), 6.91 – 6.81 (m, 1H), 3.57 (s, 3H). MS (ESI) m/z 276.1
+
[M+H] .
Step
c.
6ꢀ((1R,2R)ꢀ2ꢀHydroxycyclopentylamino)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ1Hꢀpyrazolo[3,4ꢀ
d]pyrimidinꢀ4(5H)ꢀone (11). A solution of 6ꢀchloroꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ1Hꢀpyrazolo[3,4ꢀd] pyꢀ
rimidinꢀ4(5H)ꢀone (140 mg, 0.51 mmol), (1R,2R)ꢀ2ꢀamino cyclopentanol hydrochloride (280 mg, 2.0
mmol) and N,Nꢀdiisopropylethylamine (0.44 mL, 2.5 mmol) in DMF (3.0 mL) was heated at 120 °C
overnight. After the solvent was removed under reduced pressure, the residue was purified by silicaꢀgel
column chromatography using a gradient of 0 – 25% methylene chloride/methanol (5/1, v/v) in hexane
1
as eluent to afford the title compound as an offꢀwhite solid (157 mg, 91% yield). H NMR (500 MHz,
DMSOꢀd6)
δ 11.97 (s, 1H), 7.73 – 7.57 (m, 3H), 7.30 – 7.18 (m, 2H), 6.86 – 6.79 (m, 1H), 6.76 (d, J =
6
.4 Hz, 1H), 4.81 (d, J = 3.8 Hz, 1H), 4.18 – 3.93 (m, 2H), 3.34 (s, 3H), 2.16 – 2.01 (m, 1H), 1.99 –
+
1.81 (m, 1H), 1.74 – 1.60 (m, 2H), 1.58 – 1.38 (m, 2H). MS (ESI) m/z 341.2 [M+H] .
Step d. (6aR,9aS)ꢀ5,6a,7,8,9,9aꢀHexahydroꢀ5ꢀmethylꢀ3ꢀ(phenylamino)ꢀcyclopent[4,5]imidazo[1,2ꢀ
a]pyrazolo[4,3ꢀe]pyrimidinꢀ4(1H)ꢀone (12). 2.0 M thionyl chloride solution in CH Cl (130 ꢀL, 0.26
2
2
ACS Paragon Plus Environment

Page 29 of 51
Journal of Medicinal Chemistry
mmol) was added to a solution of 6ꢀ((1R,2R)ꢀ2ꢀhydroxycyclopentylamino)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ
Hꢀpyrazolo[3,4ꢀd]pyrimidinꢀ4(5H)ꢀone (30 mg, 0.088 mmol) in a mixture of CH Cl (1.0 mL) and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
2
THF (1.5 mL) at room temperature. After the completion of the addition, the reaction mixture was
stirred at room temperature overnight. The mixture was treated with 28% ammonium hydroxide (50 L)
ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and then evaporated to dryness. The obtained residue was purified by silicaꢀgel column chromatography
using a gradient of 0 – 100% ethyl acetate in hexane as eluent to afford the title compound as an offꢀ
1
white solid (26 mg, 91% yield). H NMR (500 MHz, DMSOꢀd )
δ 7.55 (s, 1H), 7.40 (dd, J = 8.5, 7.4
6
Hz, 2H), 7.22 (d, J = 7.3 Hz, 2H), 7.18 – 7.13 (m, 1H), 4.76 – 4.62 (m, 2H), 3.35 (s, 3H), 2.23 – 2.16
(m, 1H), 1.99 – 1.91 (m, 1H), 1.87 – 1.78 (m, 1H), 1.75 – 1.66 (m, 2H), 1.59 – 1.51 (m, 1H). MS (ESI)
+
m/z 323.2 [M+H] .
Step
e.
(6aR,9aS)ꢀ5,6a,7,8,9,9aꢀHexahydroꢀ1ꢀ(cyclopentylmethyl)ꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀ
cyclopent[4,5]imidazo[1,2ꢀa]pyrazolo[4,3ꢀe]pyrimidinꢀ4(1H)ꢀone (30). A suspension of (iodomeꢀ
thyl)cyclopentane (65 mg, 0.31 mmol), cesium carbonate (101 mg, 0.31 mmol) and (6aR,9aS)ꢀ
5
,6a,7,8,9,9aꢀhexahydroꢀ5ꢀmethylꢀ3ꢀ(phenylamino)ꢀcyclopent[4,5]imidazo[1,2ꢀa]pyrazolo[4,3ꢀ
e]pyrimidinꢀ4(1H)ꢀone (50 mg, 0.016 mmol) in butanꢀ2ꢀone (1.2 mL) in a sealed vial was heated in a
microwave reactor at 140 °C for 0.5 h. After the solvent was removed under reduced pressure, the resiꢀ
due was purified with a semiꢀpreparative HPLC system using a gradient of 0 – 40% acetonitrile in water
containing 0.1% formic acid over 16 min to afford the title compound as a white solid (8.0 mg, 13%
1
yield). H NMR (500 MHz, CDCl )
δ 7.55 (d, J = 7.7 Hz, 2H), 7.52 (s, 1H), 7.31 (dd, J = 8.6, 7.3 Hz,
3
1H), 6.99 – 6.90 (m, 1H), 4.93 – 4.78 (m, 2H), 4.07 (dd, J = 13.8, 7.0 Hz, 1H), 3.87 (dd, J = 13.8, 7.8
+
Hz, 1H), 3.43 (s, 3H), 2.59 – 1.21 (m, 15H). MS (ESI) m/z 405.2 [M+H] . HRMS (ESI) m/z calcd for
+
C H N O [M+H] : 405.2403; found: 405.2396. HPLC purity: 96.0%; retention time: 18.5 min.
23 29
6
(6aR,9aS)-5,6a,7,8,9,9a-Hexahydro-5-methyl-3-phenylamino-2-(4-(pyridin-2-yl)benzyl)-
cyclopent[4,5]imidazo[1,2-a]pyrazolo[4,3-e]pyrimidin-4(2H)-one (47). To a solution of (6aR,9aS)ꢀ
5,6a,7,8,9,9aꢀhexahydroꢀ5ꢀmethylꢀ3ꢀphenylaminoꢀcyclopent[4,5]imidazo[1,2ꢀa]pyrazolo[4,3ꢀ
ACS Paragon Plus Environment