Discovery of a novel series of pyrazolo[1,5-a]pyrimidine-based phosphodiesterase 2A inhibitors structurally different from N-((1S)-1-(3-fluoro-4-(trifluoromethoxy)phenyl)-2-methoxyethyl)-7-methoxy-2-oxo-2,3-dihydropyrido[2,3-b]pyrazine-4(1H)-carboxamide (TAK-915), for the treatment of cognitive disorders

Article abstract of DOI:10.1248/cpb.c17-00564

It has been hypothesized that selective inhibition of phosphodiesterase (PDE) 2A could potentially be a novel approach to treat cognitive impairment in neuropsychiatric and neurodegenerative disorders through augmentation of cyclic nucleotide signaling pathways in brain regions associated with learning and memory. Following our earlier work, this article describes a drug design strategy for a new series of lead compounds structurally distinct from our clinical candidate 2 (TAK-915), and subsequent medicinal chemistry efforts to optimize potency, selectivity over other PDE families, and other preclinical properties including in vitro phototoxicity and in vivo rat plasma clearance. These efforts resulted in the discovery of N-((1S)-2-hydroxy-2-methyl-1-(4-(trifluoromethoxy)phenyl)propyl)-6-methyl-5-(3-methyl-1H-1,2,4triazol-1-yl)pyrazolo[1,5-a]pyrimidine-3-carboxamide (20), which robustly increased 3′,5′-cyclic guanosine monophosphate (cGMP) levels in the rat brain following an oral dose, and moreover, attenuated MK-801-induced episodic memory deficits in a passive avoidance task in rats. These data provide further support to the potential therapeutic utility of PDE2A inhibitors in enhancing cognitive performance.

Full text of DOI:10.1248/cpb.c17-00564

1
058
Chem. Pharm. Bull. 65, 1058–1077 (2017)
Vol. 65, No. 11
Regular Article
Highlighted Paper selected by Editor-in-Chief
Discovery of a Novel Series of Pyrazolo[1,5-a]pyrimidine-Based
Phosphodiesterase 2A Inhibitors Structurally Different from N-((1S)-1-
(
2
3-Fluoro-4-(trifluoromethoxy)phenyl)-2-methoxyethyl)-7-methoxy-2-oxo-
,3-dihydropyrido[2,3-b]pyrazine-4(1H)-carboxamide (TAK-915), for the
Treatment of Cognitive Disorders
,
a
a
a
a
Satoshi Mikami,* Masanori Kawasaki, Shuhei Ikeda, Nobuyuki Negoro,
a
a
a
a
Shinji Nakamura, Izumi Nomura, Tomoko Ashizawa, Hironori Kokubo,
b
b
a
a
a,†
Isaac Dylan Hoffman, Hua Zou, Hideyuki Oki, Noriko Uchiyama, Yuuto Hiura,
a
a
a
a
a
Maki Miyamoto, Yuuki Itou, Masato Nakashima, Hiroki Iwashita, and Takahiko Taniguchi
a
Pharmaceutical Research Division, Takeda Pharmaceutical Company, Limited; 2–26–1 Muraoka-Higashi, Fujisawa,
b
Kanagawa 251–8555, Japan: and Takeda California, Inc.; 10410 Science Center Drive, San Diego, California 92121,
United States.
Received July 12, 2017; accepted August 14, 2017
It has been hypothesized that selective inhibition of phosphodiesterase (PDE) 2A could potentially
be a novel approach to treat cognitive impairment in neuropsychiatric and neurodegenerative disorders
through augmentation of cyclic nucleotide signaling pathways in brain regions associated with learning
and memory. Following our earlier work, this article describes a drug design strategy for a new series of
lead compounds structurally distinct from our clinical candidate 2 (TAK-915), and subsequent medicinal
chemistry efforts to optimize potency, selectivity over other PDE families, and other preclinical proper-
ties including in vitro phototoxicity and in vivo rat plasma clearance. These efforts resulted in the dis-
covery of N-((1S)-2-hydroxy-2-methyl-1-(4-(trifluoromethoxy)phenyl)propyl)-6-methyl-5-(3-methyl-1H-1,2,4-
triazol-1-yl)pyrazolo[1,5-a]pyrimidine-3-carboxamide (20), which robustly increased 3′,5′-cyclic guanosine
monophosphate (cGMP) levels in the rat brain following an oral dose, and moreover, attenuated MK-801-in-
duced episodic memory deficits in a passive avoidance task in rats. These data provide further support to the
potential therapeutic utility of PDE2A inhibitors in enhancing cognitive performance.
Key words phospodiesterase 2A; schizophrenia; pyrazolo[1,5-a]pyrimidine; structure-based drug design;
phototoxicity; intramolecular hydrogen bond
Schizophrenia is a chronic, severe, and disabling mental novel antipsychotic with improved efficacy and superior safety
disorder affecting approximately 1% of the general population profile for the treatment of schizophrenia, especially against
1
–4)
worldwide,
resulting in profound disruption in emotion and negative and cognitive symptoms of the disease.
Owing to a growing body of preclinical research, phospho-
cognition, which has a major impact on patients’ and caregiv-
ers’ lives. The symptoms can typically be divided into three diesterase (PDE) 2A has been recently receiving significant
domains: positive (delusions and hallucinations), negative (lack attention for its therapeutic potential to treat central nervous
of motivation and social withdrawal), and cognitive (memory, system (CNS) disorders, including schizophrenia and Al-
10–41)
attention, and executive function) symptoms. The most re- zheimer’s disease.
PDE2A is one of the 11 known PDE
cent medications commonly prescribed for schizophrenia families and is highly expressed in brain regions such as the
include a new generation of antipsychotics called atypical or striatum, hippocampus, and frontal cortex, which are closely
1
3–18)
second-generation antipsychotics (SGAs), which function via linked to emotion, learning, and memory.
inhibition of dopamine D2 and serotonin 5-HT2A receptors.
In addition,
5
)
PDE2A has been shown to hydrolyze the intracellular second
It has been reported that these agents are reasonably effective messengers, cAMP and 3ꢀ,5ꢀ-cyclic guanosine monophosphate
19)
at treating the positive symptoms of schizophrenia, but show (cGMP), molecules that play important roles in regulating
little or no efficacy against negative and cognitive symptoms cyclic nucleotide signaling implicated in neuronal plasticity
6
)
20–24)
of the disease. Furthermore, although generally more effec- and memory.
In light of these features, it is hypothesized
tive and better tolerated than typical or first-generation anti- that inhibition of PDE2A could exert procognitive activity
psychotics (FGAs), atypical antipsychotics also cause a variety through augmentation of cyclic nucleotide signaling in the
of drug-related side effects such as sedation, weight gain, an aforementioned brain areas, thus leading to the treatment of
increased risk of type II diabetes and high cholesterol, extra- cognitive dysfunction in a range of neuropsychiatric and neu-
7
–9)
pyramidal symptoms, hypotension, and prolactin elevation.
rodegenerative disorders.
Thus, there is a clear and high unmet medical need for a
On the basis of this biological hypothesis, we initiated a
drug discovery program to develop PDE2A inhibitors. A se-
†
Present address: University of Fukui School of Medical Sciences; 23–3 ries of recent publications from our research group described
42,43)
Matsuokashimoaizuki, Eiheiji-cho, Yoshida, Fukui 910–1193, Japan.
medicinal chemistry efforts
leading to the identification
*
To whom correspondence should be addressed. e-mail: satoshi.mikami@takeda.com
©
2017 The Pharmaceutical Society of Japan

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1059
Fig. 1. Transition from High-Throughput Screening Hit 1 to Clinical Candidate 2 (TAK-915)
Fig. 2. Origin of a Novel PDE2A Lead Compound 5
35,43)
of clinical candidate 2 (TAK-915
) originating from high- structural features from both chemotypes to deliver a new
throughput screening hit 1 (Fig. 1). Over the course of the dis- class of lead with no MDR1 liability. Despite the fact that
covery of 2, we identified another novel right-hand side (RHS) compounds 2 and 3 possess a nearly identical topological
benzylamine moiety structurally distinct from 2, as exempli- polar surface area (TPSA), the former has a superior MDR1
fied by compound 3 (Fig. 2). Although compound 3 exhibited efflux ratio, indicating that the hydrogen bond donor (HBD)
a good balance of potency and PDE selectivity, it was found count (two HBDs in 2 vs. three HBDs in 3) may play a crucial
4
4,45)
to be a P-glycoprotein (P-gp) substrate as evidenced by high role in defining each compound as a P-gp substrate.
We
multidrug resistance protein 1 (MDR1) efflux ratio, resulting therefore reasoned that replacing the 3,4-dihydropyrido[2,3-
in lower brain exposure. On the other hand, the previously b]pyrazin-2(1H)-one core in 3 with the pyrazolo[1,5-a]-
described pyrazolo[1,5-a]pyrimidine series, as represented pyrimidine core derived from 4 would serve to improve brain
by compound 4, demonstrated favorable MDR1 efflux ratios, penetration by reducing the overall HBD count. Indeed, hy-
but had suboptimal PDE2A inhibitory activity due to an un- brid compound 5 showed reasonable inhibition of PDE2A with
42)
optimized core and RHS benzylamine moieties. With these an IC50 of 66nM and no indication of being a P-gp substrate.
attributes of the two compound classes in mind, we combined These promising attributes of 5 as a lead compound encour-

1
060
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
Fig. 3. Effect of a 6-Methyl Substitution on PDE2A Inhibitory Activity
Fig. 4. X-Ray Crystal Structure of 7 Bound in the PDE2A Catalytic Site (PDB 5XKM)
aged us to focus on further optimization, aiming mainly to
improve inhibitory activity.
Our previous structure–activity relationship (SAR) study
of methyl substitutions on the pyrazolo[1,5-a]pyrimidine core
of 1 demonstrated that the 6-methyl substitution provided
an approximately 10-fold increase in the PDE2A inhibitory
activity (see Fig. 1 for numbering of the substitution posi-
tions). The 5-methyl substitution was well tolerated, while the
introduction of the methyl group into the 2- or 7-position was
42)
detrimental for potency. In good agreement with the previ-
ous SAR, introduction of the methyl group at the 6-position of
compound 5 to give compound 6 (Fig. 3) resulted in a 13-fold
Fig. 5. Drug Design Strategies to Further Increase Potency
boost in potency. An X-ray crystal structure of compound 7 and brain-penetrating PDE2A inhibitors, as represent-
bound to PDE2A revealed an unoccupied and relatively large ed by N-((1S)-2-hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
hydrophobic space around the 5-position of the core (Fig. 4); phenyl)propyl)-6-methyl-5-(3-methyl-1H-1,2,4-triazol-1-yl)-
this finding supported the tolerance for the 5-methyl substitu- pyrazolo[1,5-a]pyrimidine-3-carboxamide (20). This new
tion. These findings provided an avenue for further elabora- series successfully overcame in vitro phototoxicity issues
1
tion at the 5-position (R of the structure in Fig. 5) that might identified by the introduction of substituents at the 5-position
lead to potency enhancement. Overall, with the aim of further of the core. In addition, we report an X-ray crystal structure
increasing the inhibitory activity to the levels of our leading of 20 bound in the PDE2A active site to better understand the
clinical candidate 2 (IC =0.61 nM), our medicinal chemistry origin of high potency and PDE selectivity, as well as in vivo
5
0
effort was focused on variations of substituents at the 5 and data linking target engagement and pharmacodynamic (PD)
-positions of the pyrazolo[1,5-a]pyrimidine core. Taking into effects to relevant in vivo efficacy in a preclinical model of
account synthetic tractability, we initially chose to prepare cognitive impairment.
-substituted pyrazolo[1,5-a]pyrimidine analogs with no meth-
6
5
2
yl group at the 6-position (R =H of the structure in Fig. 5).
Results and Discussion
Table 1 summarizes the effects of carbon-linked substitu-
Herein, we describe the SAR of 5- and/or 6-substituted
pyrazolo[1,5-a]pyrimidine derivatives, culminating in the tions at the 5-position of the pyrazolo[1,5-a]pyrimidine core.
discovery of highly potent, selective, orally bioavailable, The corresponding data for 5 are also included in the table for

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1061
Table 1. Profiles of Pyrazolo[1,5-a]pyrimidine Derivatives with Substitutions at the 5-Position
a) IC50 values (95% confidence intervals given in parentheses) were calculated from percent inhibition data (dupli-
cate, n=1). All values are rounded off to two significant digits. b) Minimum selectivity (rounded off to two significant
digits) over other PDEs. c) MDR1 efflux ratios in P-glycoprotein (P-gp)-overexpressing cells. d) Plasma clearance
calculated following 0.1mg/kg, i.v. cassette dosing in rats (non-fasted). e) Values (%) represent the difference between
viability of cells exposed to 50µM of the test compound in the absence of UV irradiation and that in the presence of
UV irradiation, where cell viability was measured by intracellular ATP content (%) at 50µM relative to vehicle control.
A larger value is interpreted as a higher risk of phototoxicity. f ) LogD values at pH 7.4. g) Precipitates observed.
comparison. Introducing an isopropyl group at the 5-position improved activity. However, compound 9 displayed an unfa-
8) had little impact on potency relative to the non-substituted vorable MDR1 efflux ratio and high in vivo clearance, likely
(
analog 5, while the introduction of a bigger and less lipophilic being driven by increased polarity due to the tetrahydro-2H-
tetrahydro-2H-pyran group (9) tended to show moderately pyran group, and increased number of methylene moieties

1
062
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
Table 2. Profiles of Di-substituted Pyrazolo[1,5-a]pyrimidine Derivatives
a) IC50 values (95% confidence intervals given in parentheses) were calculated from percent inhibition data (duplicate,
n=1). All values are rounded off to two significant digits. b) Minimum selectivity (rounded off to two significant digits)
over other PDEs. c) MDR1 efflux ratios in P-glycoprotein (P-gp)-overexpressing cells. d) Plasma clearance calculated
following 0.1mg/kg, i.v. cassette dosing in rats (non-fasted). e) Values (%) represent the difference between viability of
cells exposed to 50µM of the test compound in the absence of UV irradiation and that in the presence of UV irradiation,
where cell viability was measured by intracellular ATP content (%) at 50µM relative to vehicle control. A larger value is
interpreted as a higher risk of phototoxicity. f ) LogD values at pH 7.4. g) Precipitates observed.
a)
Table 3. Inhibitory Activities of 19 and 20 against Human PDEs
Compd.
PDE1A
PDE2A
PDE3A
PDE4D2
PDE5A1
PDE6AB
PDE7B
PDE8A1
PDE9A2 PDE10A2 PDE11A4
1
9
0
3300
2700
0.31
0.51
>10000
>10000
3300
9400
8800
9100
1800
8000
>10000
>10000
>10000
>10000
>10000
>10000
1700
3100
3000
6100
2
a) IC50 values (nM) were calculated from percent inhibition data (duplicate, n=1). All values are rounded off to two significant digits.
susceptible to oxidative metabolism, respectively. To address icity in animals and humans. A phototoxicity liability could
such liabilities of the aliphatic heterocycle, we next sought impose significant limitations in a drug clinical use, including
to add heteroaryl groups. The 2-pyridyl substitution in com- the requirement to avoid sun exposure by means of sunscreen,
47,48)
pound 10 provided not only dramatically improved potency protective clothing, and eyewear.
We therefore sought to
and excellent selectivity over other PDEs, but also a favorable minimize the in vitro phototoxicity liability observed in this
MDR1 efflux ratio and significantly reduced in vivo clear- compound series. With the hope of altering the electronic
ance. In addition, 3- and 4-pyridine derivatives 11 and 12, nature of the core, thus lowering the risk of in vitro phototox-
respectively, retained desirable attributes of the 2-pyridine icity, nitrogen-linked heteroaryl groups were explored. Instal-
derivative 10, including inhibitory activity, PDE selectivity lation of a pyrazole at the 5-position of the core to obtain 14,
and MDR1 efflux ratio; however, they showed increased in a regioisomer of 13, resulted in moderately decreased potency
vivo clearance probably due to oxidation of the methyl groups and PDE selectivity, but a favorable MDR1 efflux ratio and in
adjacent to the pyridine nitrogen atoms and/or N-oxidation of vivo clearance. Of particular note was the fact that 14 showed
the pyridines. To overcome the potential metabolic liability of greatly improved phototoxic liability. This encouraging result
the methylpyridine derivatives, the difluoromethyl pyrazole prompted us to synthesize the structurally related imidazole
analog 13 was prepared, and it exhibited favorable in vivo analog 15. The compound exhibited improved potency and
clearance coupled with excellent potency and PDE selectivity, PDE selectivity with no appreciable in vitro phototoxicity
and a favorable MDR1 efflux ratio. Unfortunately, however, potential, but was found to be susceptible to P-gp mediated
the series of carbon-linked heteroaryl derivatives 10–13 were efflux, most likely due to the polar or basic nitrogen atom of
4
4,45)
all found to possess significant in vitro phototoxicity potential, the imidazole ring.
In order to attenuate the basicity of the
as evidenced by the large reduction in the viability of cells imidazole ring in 15, a two-pronged approach was implement-
exposed to the compounds in the presence of UV radiation, ed: first, the methyl group on the imidazole ring was replaced
4
6)
relative to that in the absence of UV. Although the incidence with an electron-withdrawing trifluoromethyl group to give
of in vivo phototoxicity depends on drug distribution to light- 16, and second, an additional nitrogen atom was incorporated
exposed tissues like the skin and eye, it is generally recog- into the imidazole ring to afford two isomeric triazoles 17
nized that reducing the in vitro phototoxicity potential of drug and 18. As anticipated, all modifications improved the P-gp
candidates during the lead-optimization process is a prudent efflux ratio, and maintained favorable clearance and in vitro
strategy to minimize the potential risk for in vivo phototox- phototoxicity profile, although the result of the phototoxicity

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1063
assay of compound 17 was inconclusive due to its precipita- lacks the substituent at the 5-position. Metabolite identifica-
tion at the concentration tested, 50µM. Despite these benefits, tion studies on 6 with human and rat hepatocytes revealed a
both compounds showed a concomitant 5- to 10-fold loss in single metabolite common to both species, and also indicated
potency, in addition to a decrease in PDE selectivity. Overall, the 6-methyl group of the core to be the most likely site of ox-
substitutions at the 5-position of the pyrazolo[1,5-a]pyrimidine idative metabolism. Based on these results, it was speculated
core did not result in analogs with aligned potency, PDE selec- that the triazole rings in 19 and 20 may contribute to attenu-
tivity, and other preclinical properties.
ation of oxidative metabolism of the methyl groups through
In an attempt to rescue insufficient potency and PDE selec- steric shielding effects. In the meantime, further profiling of
tivity of compounds 17 and 18, which were otherwise close to physicochemical properties of both compounds demonstrated
meeting our goals, the methyl substitution at the 6-position of that 19 had extremely poor aqueous solubility that could lead
the pyrazolo[1,5-a]pyrimidine core was examined (19, 20 in to non-linear oral absorption, while 20 was approximately
Table 2), since the 6-methyl substitution was previously shown 10-fold more soluble with high degree of crystallinity similar
42)
to significantly enhance both potency and PDE selectivity.
to 19 (Table 4). These results are consistent with the solubil-
In good agreement with the previous findings, introducing ity issue identified in the in vitro phototoxicity assay. Conse-
the methyl group at the 6-position of the pyrazolo[1,5-a]- quently, compound 20 was advanced to further characteriza-
pyrimidine core of 17 and 18 to provide 19 and 20, respec- tion by an X-ray co-crystallization analysis as well as a series
tively, demonstrated a substantial improvement of potency of in vivo studies.
and PDE selectivity (see Table 3 for inhibitory activity profile
The X-ray co-crystal structure of 20 bound to the catalytic
against each PDE family), without compromising MDR1 ef- domain of the PDE2A enzyme is presented in Fig. 6A and
flux ratios and in vitro phototoxicity risks. The phototoxic that of 7 in Fig. 6B. Compound 20 retains a number of key
potential of compound 19 was not accurately assessed due to interactions observed in the case of the first generation lead
42)
its low aqueous solubility, with precipitation observed at 50µM 7, including (i) a C–H hydrogen bonding interaction of the
as in the case of 17. It appears certain, however, that its pho- pyrazolo[1,5-a]pyrimidine core with the carbonyl oxygen atom
totoxic potential was considerably lower than those of carbon- of Gln859, (ii) π–π and CH–π interactions of the core with the
linked heteroaryl derivatives such as 10–13. This is because Phe862 and Ile826 located above and below the core, respec-
compound 19 showed no signs of phototoxicity liability at tively, (iii) van der Waals interactions of the 6-methyl group
25µM where no precipitate occurred, while carbon-linked de- of the core with a hydrophobic cavity surrounded by Tyr827,
rivatives 10–13 all displayed significant levels of UV-induced Leu858, and Met847, (iv) interactions of the amide carbonyl
cytotoxicity at the same concentration (data not shown). It is group with two water molecules, one of which serves as a
also worth noting that plasma clearance of 19 and 20 in rat bridging water molecule to interact with the OH of Tyr655
was significantly improved compared with that of 6, which (these interactions are not shown in the figure for simplicity),
and (v) hydrophobic interactions of the RHS trifluorometh-
Table 4. Thermodynamic Solubility of Compounds 19 and 20
oxyphenyl group with the hydrophobic pocket newly formed
due to ligand binding. In addition to the above conserved in-
teractions, new ones characteristic of 20 that may greatly con-
tribute to the potency enhancement are observed. The triazole
ring incorporated into the 5-position of the core effectively
fills the space sandwiched between Ile866 on the top and
Met847 on the bottom, thus making favorable van der Waals
interactions. It appears that a hydrogen bond between the tri-
azole nitrogen atom with a bridging water molecule and the
Thermodynamic solubility (µg/mL)
Compd.
Crystallinity (%)
a)
pH 1.2
pH 6.8
pH 6.8+GCDC
1
2
9
0
0.45
7
0.32
4.1
16
89
83
140
a) Solubility in a pH 6.8 solution containing sodium glycochenodeoxycholate
(GCDC). This is considered as simulated intestinal fluid (pH 6.8) in the presence of
bile acid.
Fig. 6. (A) X-Ray Crystal Structure of 20 Bound in the PDE2A Catalytic Site (PDB 5VP1) and (B) X-Ray Crystal Structure of 7 Bound in the
PDE2A Catalytic Site (PDB 5XKM)
The intramolecular hydrogen bonds are indicated by light dotted lines.

1
064
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
13–18)
backbone carbonyl group of Leu858 may have some positive where the PDE2A enzyme is highly expressed,
were mea-
impacts on the potency. Furthermore, the branched tertiary sured to assess the PD effects of compound 20 (Fig. 8). Con-
11,29,42,43,50)
alcohol moiety would serve to provide greater conformational sistent with the previous observations,
compound 20
rigidity to the molecule through not only an intramolecular dose-dependently increased cGMP levels in all selected brain
hydrogen bond between the oxygen atom of the tertiary alco- regions at doses as low as 3mg/kg, corresponding to ca. 37%
hol and the amide NH, but also restriction of the bond rotation occupancy levels, while not affecting cAMP levels. These re-
imparted by the gem-dimethyl group, and thereby offer a ben- sults suggest that oral administration of 20 at doses of 3mg/kg
efit with reduced entropic cost of binding by biasing the RHS or above would lead to significant pharmacological effects
benzylamine moiety toward the bioactive conformation.
in preclinical models of cognition. We therefore performed a
The preclinical pharmacokinetic (PK) profiles of compound passive avoidance test in MK-801-treated rats to evaluate the
0 in rat and mouse are summarized in Table 5. Compound effects of 20 at an oral dose of 3mg/kg on deficits in episodic
0 exhibited favorable PK parameters in both rat and mouse, memory induced by N-methyl-D-aspartate (NMDA) recep-
2
2
51,52)
characterized by reduced plasma clearance and moderate tor hypofunction
(Fig. 9). Subcutaneous administration
oral bioavailability. In addition, the compound demonstrated of MK-801 at 0.1mg/kg significantly decreased the avoid-
reasonable brain uptake with a brain to plasma ratio of 0.32 ance time in the retention test, compared to saline treatment
2
in rat, although it had a high TPSA value (114Å ) outside (p≤0.01). Pretreatment with 3mg/kg of 20 significantly atten-
4
4,45,49)
the traditional range of CNS drugs.
These results sug- uated the decrease in the avoidance time induced by MK-801
gested that the intramolecular hydrogen bonds observed in 20 (p≤0.01). The ability of 20 to reverse deficits in episodic
effectively shield the polarity of the molecule, thus improving memory produced by MK-801, suggests its potential for the
blood-brain barrier permeability.
In order to assess target engagement, the dose–occupancy disorders with impaired glutamatergic neurotransmission.
relationship of compound 20 was characterized by measur- Chemistry The RHS benzylamine moiety 28 was pre-
treatment of cognitive deficits seen in a range of psychiatric
ing PDE2A occupancy 2h after oral administration of five pared as outlined in Chart 1. Benzaldehyde 21 was initially
escalating doses: 0.3, 1, 3, 10, and 30mg/kg. As shown in converted to the corresponding hydantoin 22 using the Bu-
53)
Fig. 7, compound 20 dose-dependently occupied PDE2A in cherer–Bergs reaction. Hydantoin 22 was then hydrolyzed
rat striatum with a 50% target occupancy oral dose (ED )
5
0
of 3.9mg/kg. Based on these results, cyclic nucleotide levels
in the frontal cortex, hippocampus, and striatum of rat brain,
Table 5. PK Parameters and Kp Values of Compound 20 in Rats and
a)
Mice
Rat
Mouse
b)
CL
(mL/h/kg)
1169.0
2329.0
58.1
768.0
1337.0
181.0
0.7
total
c)
Vdss (mL/kg)
d)
Cmax (ng/mL)
e)
Tmax (h)
2.0
f )
AUC0
(ng·h/mL)
272.8
31.2
628.6
48.2
–8h
g)
F
(%)
h)
Kp
0.32
—
a) Cassette dosing at 0.1mg/kg, i.v. and 1mg/kg, p.o. (non-fasted), in an average
of 3 rats or mice. b) Total clearance. c) Volume of distribution at steady state. d)
Maximum plasma concentration. e) Time of maximum concentration. f ) Area under
the plasma concentration vs. time curve (0–8h). g) Oral bioavailability. h) Brain-to-
plasma ratio at 2h after oral administration of 20 at a dose of 10mg/kg.
Fig. 8. Effects of Compound 20 on Cyclic Nucleotides in Rat Brain
cGMP (A) and cAMP (B) levels 2h after oral administration of compound
2
(
0 (1, 3, 10mg/kg) were measured in the rat frontal cortex (Fcx), hippocampus
Hipp), and striatum (Str). Data are presented as mean± S.E.M. (n=9). *p<0.025,
**p<0.005, ***p<0.0005, vs. vehicle by Shirley–Williams’ test.
Fig. 7. In Vivo PDE2A Target Occupancy in the Striatum Measured 2h
after Oral Administration of 20 in Rats

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1065
under basic conditions to afford amino acid 23. The protec- desired enantiomer 27a was removed under acidic conditions
tion of the amino group of 23 with a Boc group using Schot- to give RHS benzylamine 28 as a hydrochloride salt.
ten–Baumann conditions and subsequent methylation of the
A series of 5-substituted pyrazolo[1,5-a]pyrimidine deriva-
carboxylic acid 24 gave methyl ester 25. Treatment of 25 tives 8–18 were prepared following the procedures depicted
with methyl Grignard reagent, followed by chiral separation in Chart 2. Starting from elaboration of the pyrazolopyrimi-
provided enantiomers 27a and b. Finally, the Boc group of the dine core 30 by the condensation of 5-amino-1H-pyrazole 29
with a mixture of cis- and trans-ethyl 3-ethoxyacrylate under
basic conditions, the ethyl ester of 30 was hydrolyzed to af-
ford carboxylic acid 31. Bischlorination of 31 with phosphoryl
chloride (POCl ) gave 32, which was subsequently reacted
3
with RHS benzylamine 28, to furnish 5-chloropyrazolo[1,5-
a]pyrimidine 33. Attachment of the substituent at the 5-posi-
tion of the pyrazolo[1,5-a]pyrimidine core was finally accom-
plished by the following three approaches: i) Suzuki–Miyaura
cross-coupling of the corresponding boronic acids or esters to
deliver 11–13; ii) Suzuki–Miyaura cross-coupling followed by
catalytic hydrogenation of the olefin moieties to deliver 8 and
9; and iii) nucleophilic displacement with the corresponding
azoles to deliver 14–18. However, Suzuki–Miyaura cross-
coupling employing 33 to obtain 2-pyridyl derivative 10 was
unsuccessful, probably due to the relatively unstable nature of
2
-pyridylboron species as well as the presence of polar func-
tionalities such as the amide and the tertially alcohol moieties
in 33 that could potentially coordinate with the palladium cat-
alyst. Therefore, attempts were made to introduce the 2-pyri-
dyl group at the early step prior to the amide coupling with
Fig. 9. Effects of Compound 20 on MK-801-Induced Episodic Memory
Deficits Assessed by the Passive Avoidance Task in Rats
Vehicle or compound 20 (3mg/kg) was orally administered 90min before the s.c.
administration of saline or MK-801. Thirty minutes after the s.c. administration of the RHS benzylamine moiety. After chlorination of 30 with
saline or MK-801 (0.1mg/kg), the rat was placed on the illuminated chamber. The
POCl , 5-chloropyrazolo[1,5-a]pyrimidine 36 was successfully
3
latency to cross over into the dark chamber was recorded for up to a maximum of
3
00s. The retention test was carried out 24h after the acquisition trial. Data (la- coupled with lithium 2-pyridyltriolborate salt 37 via Suzuki–
tency to cross over into the dark compartment on the retention trial) are expressed
Miyaura cross-coupling catalyzed by palladium(II) acetate
and triphenylphosphine (Ph P) in the presence of copper(I)
#
#
as mean± S.E.M., n=20. **p≤0.01 (versus control by Wilcoxon’s test), p≤0.01
(
versus vehicle+MK-801 by Wilcoxon’s test).
3
Reagents and conditions: (a) KCN, (NH ) CO , EtOH, H O, 50 to 60°C, 3h, (taken on crude); (b) KOH, water, 90°C, 3d, (taken on crude); (c) Boc O, 2M NaOH aq.,
4
2
3
2
2
THF, r.t., overnight, (taken on crude); (d) MeI, K CO , DMF, r.t., 2h, 46% (4 steps from 21); (e) MeMgBr, THF, 0°C, 1h, 76%; (f) Chiralpak AD, hexane–EtOH=19:1; (g)
2
3
HCl, EtOAc, r.t., 1.5h, 84%.
Chart 1. Synthesis of RHS Benzylamine Moiety 28

1
066
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
Reagents and conditions: (a) ethyl 3-ethoxyacrylate (cis- and trans-mixture), Cs CO , DMF, 100°C, 2h, 91%; (b) NaOH aq., THF, EtOH, 50–60°C, 2h–overnight,
2
3
7
5–98%; (c) POCl , DIEA, 130°C, 4h, 74%; (d) RHS benzylamine 28, DIEA, CH CN, 0°C to r.t., 16h, 77%; (e) alkyl or arylboronic acid pinacolester, (Amphos) PdCl ,
3
3
2
2
K CO , toluene, H O, 120–150°C (microwave), 25min–2.5h, 63–78%; (f) azole, K CO , DMF, 80–90°C, 30min–2h, 7–74%; (g) H , 5% Pd/C–ethylenediamine complex,
2
3
2
2
3
2
MeOH, r.t., 16h, 83–98%; (h) POCl , 100°C, 16h, 61%; (i) lithium 4-methyl-1-pyridin-2-yl-2,6,7-trioxa-1-borabicyclo[2.2.2]octan-1-uide 37, Pd(OAc) , Ph P, CuI, DMF,
3
2
3
8
0°C, 2h, 87%; (j) RHS benzylamine 28, HATU, DIEA, DMF, r.t., 12h, 84%.
Chart 2. Synthesis of 5-Substituted Pyrazolo[1,5-a]pyrimidine Derivatives 8–18
5
4)
iodide, providing coupled product 38. Ester hydrolysis of 38 to give the corresponding carboxylic acids 52a, b, which were
and subsequent amide formation with RHS benzylamine 28 finally coupled with benzylamine 28 to afford amide products
furnished the final compound 10.
19 and 20.
The synthetic route to access compounds 19 and 20 is
highlighted in Chart 3. Aminopyrazole 43 was initially Conclusion
constructed in a two-step sequence of reactions, involving
In summary, a new class of PDE2A inhibitors, structurally
the condensation of tert-butyl 2-cyanoacetate (40) and tert- different from our clinical candidate TAK-915 were identified
butoxy bis(dimethylamino)methane (41), followed by reaction by incorporating features from our previously reported PDE2A
with hydrazine. The aminopyrazole 43 was then condensed inhibitor series such as 3 and 4. With an initial goal of enhanc-
with 1,3-dicarbonyl related electrophile 46 prepared by bro- ing the potency of new lead compound 5, substitutions on the
mination of methyl methacrylate (44), prior to treatment pyrazolo[1,5-a]pyrimidine core, especially at the 5- and 6-po-
with a methanol solution of sodium methoxide to provide sitions, were explored on the basis of previous SAR knowledge
pyrazolo[1,5-a]pyrimidine core 47. Regioselective installation and crystal structure information of pyrazolo[1,5-a]pyrimidine
of triazole rings at the 5-position of the core was carried out analog 7. Our optimization efforts resulted in the discov-
by two different approaches using conditions mild enough to ery of N-((1S)-2-hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
be compatible with the acid-labile tert-butyl ester moiety. One phenyl)propyl)-6-methyl-5-(3-methyl-1H-1,2,4-triazol-1-yl)-
approach involved diphenylphosphoryl azide (DPPA)-mediated pyrazolo[1,5-a]pyrimidine-3-carboxamide (20),
a
potent
azidation prior to the regioselective 1,3-dipolar cycloaddition brain-penetrating PDE2A inhibitor highly selective over other
of the resulting azido group of 48 to the C=C bond of the PDE families. Introduction of a nitrogen-linked heteroaryl
betainic form of α-formylethylidenetriphenylphosphorane (49) substituent at the 5-position as in 20 was key to alleviating
with subsequent spontaneous elimination of triphenylphos- in vitro phototoxicity risk. In addition, despite its high TPSA
5
5,56)
2
phine oxide,
to deliver 51a. The other involved Appel-type value (114Å ) which is beyond the optimal range for CNS
chlorination with carbon tetrachloride (CCl ) and PPh , fol- drugs, compound 20 exhibited acceptable brain exposure
4
3
lowed by S Ar displacement with 3-methyl-1H-1,2,4-triazole probably due to cloaking of HBD and HBA groups via intra-
N
to afford 51b in a regioselective manner. The resulting tert- molecular hydrogen bonding. This was also supported by an
butyl esters 51a, b were hydrolyzed under acidic conditions X-ray crystal structure of 20 bound to PDE2A. In the subse-

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1067
Reagents and conditions: (a) neat, r.t., 30min (taken on crude); (b) NH NH ·H O, MeOH, 70°C, overnight, 64%; (c) Br , EtOAc, r.t., 16h, (taken on crude); (d) NaOMe,
2
2
2
2
MeOH, 70°C, 3h, (taken on crude); (e) Cs CO , DMF, 100°C, 16h, 84%; (f) DPPA, DBU, THF, 60°C, 5h, 74%; (g) 49, toluene, 80°C, 3h, then MgCl , 60°C, 2h, 97%; (h)
2
3
2
CCl , PPh , 1,2-dichloroethane, 75–85°C, 4.5h, 88%; (i) 3-methyl-1H-1,2,4-triazole, K CO , DMF, 60°C, 2h, 69%; (j) MsOH, CH CN, r.t.–60°C, 5.5h–overnight, 85–91%;
4
3
2
3
3
(
k) RHS benzylamine 28, EDCI·HCl, HOBt·H O, Et N, DMF, r.t., 2h–overnight, 75–82%.
2 3
Chart 3. Synthesis of 6-Methyl-5-Substituted Pyrazolo[1,5-a]pyrimidine Derivatives 19 and 20
quent in vivo studies on 20, we confirmed dose-dependent tar- (TFA) or 5mM ammonium acetate and detected at 220nm.
get engagement in rat brain after oral dosing. Correlated with Column chromatography was carried out on silica gel ((Merck
these results, compound 20 exhibited dose-dependent increas- Kieselgel 60, 70–230 mesh, Merck) or (Chromatorex NH-DM
es in cGMP levels in the brain and, moreover, improved cog- 1020, 100–200 mesh, Fuji Silysia Chemical, Ltd.)), or on pre-
nitive function in a rat model of schizophrenia following oral packed Purif-Pack columns (SI or NH, particle size: 60µm,
administration at the dose where PD effects were noted. Thus, Fuji Silysia Chemical, Ltd.). Analytical HPLC was performed
potent and selective PDE2A inhibitors, such as compound 20, using a Corona Charged Aerosol Detector or photo diode
are hypothesized to provide a novel therapeutic approach to array detector with a Capcell Pak C18AQ (3.0mm ID×50mm
treat cognitive dysfunction in a range of neuropsychiatric and L, Shiseido, Japan) or L-column2 ODS (2.0mm ID×30mm L,
neurodegenerative disorders, including schizophrenia and Alz- CERI, Japan) column at a temperature of 50°C and a flow rate
heimer’s disease.
of 0.5mL/min. Mobile phases A and B under neutral condi-
tions were a mixture of 50mmol/L ammonium acetate, water,
and acetonitrile (1:8:1, v/v/v) and a mixture of 50mmol/L
Experimental
General Chemistry Information All solvents and re- ammonium acetate and acetonitrile (1:9, v/v), respectively.
agents were obtained from commercial sources and were used The ratio of mobile phase B was increased linearly from 5
as received. Microwave-assisted reactions were carried out in to 95% over 3min, and then maintained at 95% over the next
a single-mode reactor, Biotage Initiator 2.0 or 2.5 microwave 1min. Mobile phases A and B under acidic conditions were
synthesizer. Yields were not optimized. All reactions were a mixture of 0.2% formic acid in 10mmol/L ammonium for-
monitored by TLC analysis on Merck Kieselgel 60 F254 plates mate and 0.2% formic acid in acetonitrile, respectively. The
or Fuji Silysia NH plates, or LC-MS analysis. LC-MS analysis ratio of mobile phase B was increased linearly from 14 to 86%
was performed on a Shimadzu LC-MS system operating in over 3min, and then maintained at 86% over the next 1min.
atmospheric pressure chemical ionization (APCI) (+ or −) or All final test compounds were purified to >95% chemical
electrospray ionization (ESI) (+ or −) ionization mode. Ana- purity as measured by analytical HPLC. Elemental analyses
lytes were eluted using a linear gradient with a mobile phase were carried out by Takeda Analytical Laboratories, and all
of water/acetonitrile containing 0.05% trifluoroacetic acid results were within ± 0.4% of the theoretical values. Melting

1
068
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
points were determined on a BÜCHI B-545 melting point ap- 7.44–7.56 (3H, m), 7.88–7.97 (1H, m), 8.26 (1H, d, J=7.4Hz),
paratus or a DSC1 system (Mettler-Toledo International Inc., 8.64 (1H, s), 8.75–8.88 (3H, m), 9.25 (1H, d, J=8.3Hz). MS
+
Greifensee, Switzerland). Proton nuclear magnetic resonance (ESI/APCI) m/z 472.1 [M+H] . HPLC purity: 97.3%. Anal.
1
(
H-NMR) spectra were recorded on a Varian Mercury-300 Calcd for C H F N O : C, 58.60; H, 4.28; N, 14.86. Found: C,
23 20 3 5 3
(
300MHz) or Bruker DPX300 (300MHz) instrument. All 58.61; H, 4.51; N, 14.86.
1
H-NMR spectra were consistent with the proposed struc-
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
tures. All proton shifts are given in parts per million (ppm) phenyl)propyl)-5-(6-methylpyridin-3-yl)pyrazolo[1,5-a]-
downfield from tetramethysilane (δ) as the internal standard pyrimidine-3-carboxamide (11) The title compound was
in deuterated solvent, and coupling constants (J) are in hertz prepared as a pale yellow solid after trituration with hex-
(
Hz). NMR data are reported as follows: chemical shift, inte- ane–ethyl acetate (5:1) in 63% yield from 33 and 2-methyl-5-
gration, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; (4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridine using the
quint, quintet; m, multiplet; dd, doublet of doublets; td, triplet procedure analogous to that described for the synthesis of
of doublets; ddd, doublet of doublet of doublets; and brs, broad 35, except that basic silica gel was employed in a column
1
singlet), and coupling constants. Very broad peaks for protons chromatography purification in place of silica gel. H-NMR
of, for example, hydroxyl and amino groups are not always (300MHz, DMSO-d ) δ: 1.03 (3H, s), 1.33 (3H, s), 2.60 (3H,
6
indicated.
s), 4.93 (1H, d, J=8.3Hz), 5.19 (1H, s), 7.23–7.33 (2H, m),
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- 7.48–7.59 (3H, m), 7.98 (1H, d, J=7.5Hz), 8.50 (1H, s), 8.81
phenyl)propyl)-5-isopropylpyrazolo[1,5-a]pyrimidine-3- (1H, dd, J=8.3, 2.3Hz), 9.24 (1H, d, J=8.7Hz), 9.41 (1H, d,
carboxamide (8) A mixture of 34 (38.6mg, 0.09mmol) and J=7.5Hz), 9.54 (1H, d, J=2.3Hz). MS (ESI/APCI) m/z 486.1
+
5
% Pd/C–ethylenediamine complex (10mg) in MeOH (5mL) [M+H] . HPLC purity: 100%. mp 237°C.
was hydrogenated under balloon pressure at room temperature
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
(
r.t.) for 16h. The catalyst was removed by filtration and the phenyl)propyl)-5-(2-methylpyridin-4-yl)pyrazolo[1,5-a]-
filtrate was concentrated in vacuo. The residue was purified pyrimidine-3-carboxamide (12) A mixture of 33 (130mg,
by column chromatography (silica gel, hexane–ethyl acetate, 0.303mmol), 2-methyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-
1
9:1 to 0:100) to afford 8 (32.1mg, 0.074mmol, 83%) as a 2-yl)pyridine (100mg, 0.455mmol), (Amphos) PdCl (21.5mg,
2 2
1
white amorphous solid. H-NMR (300MHz, DMSO-d ) δ: 0.0303mmol), K CO3 (62.9mg, 0.455mmol), toluene (2mL)
6
2
0
.99–1.08 (3H, m), 1.27 (3H, s), 1.41 (3H, d, J=4.5Hz), 1.43 and water (0.2mL) was heated at 120°C under microwave ir-
3H, d, J=4.5Hz), 3.20–3.34 (1H, m), 4.81–5.00 (2H, m), radiation for 2.5h. After being cooled to r.t., the reaction mix-
.22–7.34 (3H, m), 7.48–7.61 (2H, m), 8.44 (1H, s), 9.07 (1H, ture was directly purified by column chromatography (basic
d, J=8.7Hz), 9.19 (1H, d, J=7.2Hz). MS (ESI/APCI) m/z 437.1 silica gel, ethyl acetate–methanol, 100:0 to 19:1) to give 12
(
7
+
[
M+H] . HPLC purity: 98.9%.
(115mg, 0.237mmol, 78%) as an off-white solid after recrys-
1
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- tallization from heptane–ethyl acetate. H-NMR (300MHz,
phenyl)propyl)-5-(tetrahydro-2H-pyran-4-yl)pyrazolo[1,5-a]- CDCl ) δ: 1.21 (3H, s), 1.50 (3H, s), 1.73 (1H, s), 2.72 (3H,
3
pyrimidine-3-carboxamide (9) A mixture of 35 (84.4mg, s), 5.13 (1H, d, J=8.5Hz), 7.13–7.20 (2H, m), 7.46–7.54 (3H,
0.177mmol) and Pd/C (en) (15mg) in MeOH (10mL) and m), 7.91 (1H, dd, J=5.1, 1.3Hz), 8.09 (1H, s), 8.67 (1H, s),
tetrahydrofuran (THF) (dry) (2mL) was hydrogenated under 8.75 (1H, d, J=5.3Hz), 8.88 (1H, d, J=7.4Hz), 9.19 (1H, d,
+
balloon pressure at r.t. for 16h. The catalyst was removed J=8.3Hz). MS (ESI/APCI) m/z 486.1 [M+H] . HPLC purity:
by filtration and the filtrate was concentrated in vacuo. The 100%. mp 192°C. Anal. Calcd for C H F N O : C, 59.38; H,
2
4
22
3
5
3
residue was purified by column chromatography (silica gel, 4.57; N, 14.43. Found: C, 59.03; H, 4.60; N, 14.24.
hexane–ethyl acetate, 19:1 to 0:100) to afford 9 (82.9mg,
5-(1-(Difluoromethyl)-1H-pyrazol-4-yl)-N-((1S)-2-hy-
.173mmol, 98%) as a white amorphous solid. H-NMR droxy-2-methyl-1-(4-(trifluoromethoxy)phenyl)propyl)-
300MHz, DMSO-d ) δ: 1.04 (3H, s), 1.28 (3H, s), 1.86–2.11 pyrazolo[1,5-a]pyrimidine-3-carboxamide (13) The title
1
0
(
6
(4H, m), 3.13–3.28 (1H, m), 3.45–3.59 (2H, m), 3.97–4.10 (2H, compound was prepared as a white amorphous solid in 77%
m), 4.88–4.99 (2H, m), 7.24–7.34 (3H, m), 7.50–7.61 (2H, m), yield from 33 and 1-(difluoromethyl)-4-(4,4,5,5-tetramethyl-
8
.45 (1H, s), 9.03 (1H, d, J=8.7Hz), 9.21 (1H, d, J=7.2Hz). 1,3,2-dioxaborolan-2-yl)-1H-pyrazole using the procedure
+
MS (ESI/APCI) m/z 479.1 [M+H] . HPLC purity: 99.6%.
analogous to that described for the synthesis of 35, except
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- that the crude material was purified by basic silica gel column
phenyl)propyl)-5-(pyridin-2-yl)pyrazolo[1,5-a]pyrimidine- chromatography followed by silica gel column chromatog-
1
3
0
0
0
-carboxamide (10)
A
mixture of crude 39 (79.0mg, raphy. H-NMR (300MHz, DMSO-d ) δ: 1.04 (3H, s), 1.36
6
.329mmol), 28 (94.0mg, 0.329mmol), HATU (163mg, (3H, s), 4.89 (1H, d, J=8.3Hz), 5.28 (1H, s), 7.23–7.35 (2H,
.428mmol), N,N-diisopropylethylamine (DIEA) (0.172mL, m), 7.49–7.59 (2H, m), 7.73 (1H, d, J=7.5Hz), 7.97 (1H, t,
.987mmol), and N,N-dimethylformamide (DMF) (3mL) was J=59.0Hz), 8.45 (1H, s), 8.81 (1H, s), 9.29 (1H, s), 9.32–9.42
+
stirred at r.t. for 12h. The mixture was diluted with water (2H, m). MS (ESI/APCI) m/z 511.1 [M+H] . HPLC purity:
and extracted with EtOAc. The organic layer was separated, 100%.
washed with saturated aqueous NaCl, dried over anhydrous
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
Na SO and concentrated in vacuo. The residue was purified phenyl)propyl)-5-(1H-pyrazol-1-yl)pyrazolo[1,5-a]pyrimi-
2
4
by column chromatography (basic silica gel, hexane–ethyl dine-3-carboxamide (14) To a mixture of 33 (126.3mg,
acetate, 100:0 to 0:100) to give 10 (130mg, 0.276mmol, 84% 0.295mmol) and 1H-pyrazole (24.1mg, 0.353mmol) in DMF
(
63% yield in 2 steps from 39)) as a pale yellow amorphous (4mL) was added K CO (48.8mg, 0.353mmol). The mixture
2 3
1
solid. H-NMR (300MHz, CDCl ) δ: 1.21 (3H, s), 1.49 (3H, was stirred at 90°C for 30min and then poured into water.
3
s), 1.96 (1H, s), 5.14 (1H, d, J=8.3Hz), 7.14–7.22 (2H, m), The mixture was extracted with EtOAc, washed with water

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1069
and saturated aqueous NaCl, dried over anhydrous Na SO4 8.55 (1H, s), 8.88 (1H, d, J=0.8Hz), 9.07 (1H, d, J=8.3Hz),
2
+
and concentrated in vacuo. The residue was purified by 9.50 (1H, d, J=7.5Hz). MS (ESI/APCI) m/z 476.2 [M+H] .
column chromatography (silica gel, hexane–ethyl acetate, HPLC purity: 99.4%. mp 260°C.
19:1 to 0:100) to afford 14 (102mg, 0.222mmol, 75%) as a
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
white solid after trituration with hexane–ethyl acetate (5:1). phenyl)propyl)-5-(3-methyl-1H-1,2,4-triazol-1-yl)pyrazolo-
1
H-NMR (300MHz, DMSO-d ) δ: 1.02 (3H, s), 1.36 (3H, [1,5-a]pyrimidine-3-carboxamide (18) The title compound
6
s), 4.86 (1H, d, J=8.3Hz), 5.34 (1H, s), 6.75–6.81 (1H, m), was prepared as a white solid after trituration with diisopropyl
7
.23–7.33 (2H, m), 7.48–7.58 (2H, m), 7.76 (1H, d, J=7.5Hz), ether in 35% yield from 33 and 3-methyl-1H-1,2,4-triazole
8
.05 (1H, d, J=1.5Hz), 8.46 (1H, s), 9.02–9.14 (2H, m), 9.37 using the procedure analogous to that described for the syn-
+
(1H, d, J=7.5Hz). MS (ESI/APCI) m/z 461.1 [M+H] . HPLC thesis of 14, except that basic silica gel was employed in a
purity: 100%. mp 241°C.
column chromatography purification in place of silica gel.
1
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
H-NMR (300MHz, DMSO-d ) δ: 1.02 (3H, brs), 1.36 (3H, s),
6
phenyl)propyl)-5-(4-methyl-1H-imidazol-1-yl)pyrazolo[1,5- 2.47 (3H, brs), 4.86 (1H, d, J=8.1Hz), 5.43 (1H, s), 7.21–7.35
a]pyrimidine-3-carboxamide (15) The title compound was (2H, m), 7.47–7.59 (2H, m), 7.63 (1H, d, J=7.5Hz), 8.52 (1H,
prepared as a pale yellow solid after trituration with hexane– s), 9.06 (1H, d, J=8.1Hz), 9.46 (1H, d, J=7.5Hz), 9.55 (1H, s).
+
ethyl acetate (5:1) in 70% yield from 33 and 4-methyl-1H- MS (ESI/APCI) m/z 476.1 [M+H] . HPLC purity: 100%.
imidazole using the procedure analogous to that described for
(S)-N-(2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
the synthesis of 14, except that a preparative HPLC (column: phenyl)propyl)-6-methyl-5-(4-methyl-1H-1,2,3-triazol-1-yl)-
L-Column2 ODS 20mm ID×150mm L; mobile phase A: pyrazolo[1,5-a]pyrimidine-3-carboxamide (19) To a mix-
0
.1% TFA in water; mobile phase B: 0.1% TFA in acetonitrile; ture of 52a (2.81g, 10.9mmol) and 28 (3.26g, 11.4mmol) in
flow rate: 20mL/min) was employed in place of silica gel DMF (250mL) were added HOBt·H O (2.06g, 13.1mmol),
2
1
column chromatography. H-NMR (300MHz, DMSO-d ) δ: EDCI·HCl (2.55g, 13.1mmol) and triethylammonium acetate
6
1
.02 (3H, s), 1.35 (3H, s), 2.24 (3H, s), 4.87 (1H, d, J=8.3Hz), (TEA) (7.65mL, 54.4mmol). The mixture was stirred at r.t.
.31 (1H, s), 7.24–7.33 (2H, m), 7.47–7.57 (2H, m), 7.73 (1H, d, overnight and then quenched with water at r.t. The mixture
J=7.9Hz), 8.07 (1H, s), 8.45 (1H, s), 8.85 (1H, d, J=0.8Hz), was extracted with EtOAc, washed with saturated aqueous
5
9
.00 (1H, d, J=8.3Hz), 9.46 (1H, d, J=7.5Hz). MS (ESI/APCI) NaHCO , water, and then saturated aqueous NaCl, dried over
3
+
m/z 475.1 [M+H] . HPLC purity: 99.2%. mp 130°C.
anhydrous Na SO , and concentrated in vacuo. The residue
2
4
(
S)-N-(2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- was purified by column chromatography (silica gel, hexane–
phenyl)propyl)-5-(4-(trifluoromethyl)-1H-imidazol-1-yl)- ethyl acetate, 100:0 to 3:17) to give 19 (3.86g, 7.89mmol,
1
pyrazolo[1,5-a]pyrimidine-3-carboxamide (16) The title 73%) as a white solid. H-NMR (300MHz, CDCl ) δ: 1.17
3
compound was prepared as a pale yellow solid after tritura- (3H, s), 1.48 (3H, s), 1.74 (1H, s), 2.50 (3H, d, J=0.8Hz),
tion with hexane–ethyl acetate (5:1) in 70% yield from 33 2.84 (3H, d, J=0.8Hz), 5.05(1H, d, J=8.3Hz), 7.17 (2H, d,
and 4-(trifluoromethyl)-1H-imidazole using the procedure J=7.9Hz), 7.38–7.48 (2H, m), 8.60 (1H, s), 8.69–8.82 (3H,
1
analogous to that described for the synthesis of 14. H-NMR m). The obtained product (3.86g, 7.89mmol) was dissolved
(
300MHz, DMSO-d ) δ: 1.02 (3H, s), 1.34 (3H, s), 4.86 (1H, in EtOAc (277mL) at 50–60°C. To this solution was added
6
d, J=8.3Hz), 5.36 (1H, s), 7.23–7.33 (2H, m), 7.47–7.58 (2H, hexane (150mL) at the same temperature and a white solid
m), 7.89 (1H, d, J=7.5Hz), 8.53 (1H, s), 8.92–9.02 (2H, m), precipitated. After 2h of stirring at r.t., further hexane (88mL)
9
.15 (1H, s), 9.62 (1H, d, J=7.5Hz). MS (ESI/APCI) m/z 529.1 was added to the mixture, which was stirred at r.t. for 10min.
+
[
M+H] . HPLC purity: 100%. mp 220°C.
The white precipitate was collected by filtration, washed with
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- hexane–ethyl acetate, and dried to give 19 (3.17g, 6.48mmol,
1
phenyl)propyl)-5-(4-methyl-1H-1,2,3-triazol-1-yl)pyrazolo- 82%) as a white solid. H-NMR (300MHz, CDCl ) δ: 1.17
3
[
(
1
1,5-a]pyrimidine-3-carboxamide (17) A mixture of 33 (3H, s), 1.48 (3H, s), 1.74 (1H, s), 2.50 (3H, d, J=0.8Hz),
430mg, 1.00mmol), 4-methyl-1H-1,2,3-triazole (100mg, 2.84 (3H, d, J=0.8Hz), 5.05 (1H, d, J=8.3Hz), 7.17 (2H, d,
.20mmol) and K CO (208mg, 1.50mmol) in DMF (4mL) J=8.3Hz), 7.39–7.48 (2H, m), 8.60 (1H, s), 8.69–8.83 (3H, m).
2
3
+
was stirred at 80°C for 2h. The mixture was quenched with MS (ESI/APCI) m/z 490.2 [M+H] . HPLC purity: 100%. mp
water at r.t. and extracted with EtOAc. The organic layer was 238–239°C. Anal. Calcd for C H F N O : C, 53.99; H, 4.53;
2
2
22
3
7
3
separated, washed with water and saturated aqueous NaCl, F, 11.64; N, 20.03. Found: C, 53.82; H, 4.42; F, 11.64; N, 19.81.
dried over anhydrous MgSO4 and concentrated in vacuo (S)-N-(2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
508.7mg). The residue was purified by using preparative phenyl)propyl)-6-methyl-5-(3-methyl-1H-1,2,4-triazol-1-yl)-
(
HPLC (column: L-Column2 ODS 20mm ID×150mm L; pyrazolo[1,5-a]pyrimidine-3-carboxamide (20) To a sus-
mobile phase A: 0.1% TFA in water; mobile phase B: 0.1% pension of 52b (606mg, 2.35mmol) in DMF (12mL)
TFA in acetonitrile; flow rate: 20mL/min). The first eluting were added 28 (805mg, 2.82mmol), EDCI·HCl (540mg,
factions (tR1) were concentrated to dryness, and washed with 2.82mmol), HOBt·H O (431mg, 2.82mmol), and TEA
2
saturated aqueous NaHCO , extracted with EtOAc. The or- (0.392mL, 2.82mmol) at r.t. The mixture was stirred at r.t.
3
ganic layer was separated, washed with water and saturated for 2h. The mixture was poured into water and extracted
aqueous NaCl, dried over anhydrous MgSO and concentrated with EtOAc. The organic layer was separated, washed with
4
in vacuo. The residue (42.7mg) was crystallized from hex- saturated aqueous NaCl, dried over anhydrous MgSO4 and
ane–ethyl acetate to give 17 (34.4mg, 0.072mmol, 7.2%) as concentrated in vacuo. The residue was purified by column
1
a white solid. H-NMR (300MHz, DMSO-d ) δ: 1.04 (3H, s), chromatography (basic silica gel, hexane–ethyl acetate, 4:1 to
6
1
.37 (3H, s), 2.44 (3H, s), 4.87 (1H, d, J=8.3Hz), 5.39 (1H, s), 1:9) to give 20 (858mg, 1.75mmol, 75%) as a white solid after
1
7.23–7.34 (2H, m), 7.48–7.58 (2H, m), 7.93 (1H, d, J=7.5Hz), recrystallization from diisopropyl ether/ethyl acetate. H-NMR

1
070
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
1
(
300MHz, DMSO-d ) δ: 1.00 (3H, s), 1.31 (3H, s), 2.46 (3H, 51.2mmol, 76%) as a white solid. H-NMR (300MHz, CDCl )
6
3
s), 2.64 (3H, d, J=0.8Hz), 4.87 (1H, d, J=8.0Hz), 5.32 (1H, δ: 1.05 (3H, s), 1.30–1.47 (12H, m), 4.50 (1H, d, J=6.4Hz),
s), 7.27 (2H, d, J=8.0Hz), 7.51 (2H, d, J=8.7Hz), 8.49 (1H, s), 5.53 (1H, d, J=8.7Hz), 7.18 (2H, d, J=7.9Hz), 7.33 (2H, d,
−
8
.85 (1H, d, J=8.3Hz), 9.46–9.53 (2H, m). MS (ESI/APCI) m/z J=8.7Hz). MS (ESI/APCI) m/z 348.0 [M−H] .
+
490.3 [M+H] . HPLC purity: 100%. mp 212°C. Anal. Calcd
tert-Butyl ((1S)-2-Hydroxy-2-methyl-1-(4-(trifluorometh-
for C H F N O : C, 53.99; H, 4.53; N, 20.03. Found: C, 54.11; oxy)phenyl)propyl)carbamate (27a) Resolution of the en-
2
2
22
3
7
3
H, 4.50; N, 20.04.
-Amino-2-(4-(trifluoromethoxy)phenyl)acetic Acid (23) Chiralpak AD 20mm ID×250mm L column (hexane–ethanol,
The solution of potassium cyanide (19.4g, 299mmol) in water 950:50) at 80mL/min. Resolution of 26 (22.3g, 63.8mmol)
170mL) at 50°C was added dropwise to a solution of 4-tri- provided 27a as a white solid (10.2g, 29.1mmol, 46%, 91%
antiomers of 26 was carried out chromatographically using a
2
(
fluoromethoxy benzaldehyde (21) (34.1mL, 239mmol) and theoretical) as the first eluting enantiomer. Analytical HPLC
ammonium carbonate (62.0g, 645mmol) in EtOH (273mL) analysis carried out on a 4.6mm ID×250mm L Chiralpak
and water (109mL). The mixture was stirred at 60°C for 3h. AD column with the same eluent as above at a flow rate of
1
After cooling to r.t., EtOH was removed under reduced pres- 1.0mL/min indicated that 27a was of >99.9% ee. H-NMR
sure. To the mixture was acidified to pH ca. 1 with concd (300MHz, CDCl ) δ: 1.05 (3H, s), 1.36 (3H, s), 1.40 (9H, brs),
3
HCl at 0°C. The solid was filtered off and washed with water. 1.51 (1H, brs), 4.50 (1H, d, J=7.5Hz), 5.53 (1H, d, J=8.3Hz),
To a solution of potassium hydroxide (66.2g, 1000mmol) in 7.13–7.23 (2H, m), 7.29–7.39 (2H, m).
water (250mL) was added the product at r.t. The mixture was
stirred at 90°C overweekend. After cooling to r.t., the mix- yl)propan-2-ol Hydrochloride (28)
(1S)-1-Amino-2-methyl-1-(4-(trifluoromethoxy)phen-
mixture of 27a
A
ture was neutralized with concd HCl. The resulting solid was (10.2g, 29.3mmol) and 4M HCl solution in EtOAc (70.0mL,
filtered off, washed with water, and dried to give 23 (56.0g, 280mmol) was stirred at r.t. for 1.5h. The mixture was con-
2
38mmol, 100%) as a pale yellow solid. This was used in the centrated in vacuo and the resulting solid was triturated with
next reaction without further purification. MS (ESI/APCI) m/z hexane–diisopropyl ether, collected by filtration, rinsed with
−
2
34.0 [M−H] .
hexane–diisopropyl ether, and dried to afford 28 (7.00g,
1
(
(tert-Butoxycarbonyl)amino)(4-(trifluoromethoxy)- 24.5mmol, 84%) as a pale red solid. H-NMR (300MHz,
phenyl)acetic Acid (24) To
a
solution of 23 (56.0g, DMSO-d ) δ: 0.98 (3H, s), 1.23 (3H, s), 4.23 (1H, brs), 5.39
6
2
38mmol) in THF (476mL) was added Boc O (82.9mL, (1H, s), 7.37–7.48 (2H, m), 7.63 (2H, d, J=8.7Hz), 8.46 (3H,
2
357mmol) and 2M NaOH aqueous solution (357mL, br s).
714mmol) at r.t. The mixture was stirred at r.t. The mixture Ethyl 5-Hydroxypyrazolo[1,5-a]pyrimidine-3-carboxyl-
was poured into water at r.t. and extracted with Et O. The ate (30) To a mixture of ethyl 3-amino-1H-pyrazole-4-car-
2
aqueous layer was acidified to pH ca. 3 with 1M HCl aqueous boxylate (29) (30.0g, 193mmol) and ethyl 3-ethoxy-2-prope-
solution at 0°C and then extracted with EtOAc. The organic noate (cis- and trans-mixture, 41.9mL, 290mmol) in DMF
layer was washed with saturated aqueous NaCl, dried over (387mL) was added cesium carbonate (113g, 348mmol). The
anhydrous MgSO4 and concentrated in vacuo to afford 24 mixture was stirred at 100°C for 2h, diluted with water and
(63.3g, 189mmol) as a pale yellow solid. This was subjected then acidified to pH ca. 5 with AcOH. The resulting solid was
to the next reaction without further purification. MS (ESI/ filtered by filtration, washed with water and dried to afford 30
−
1
APCI) m/z 333.9 [M−H] .
(36.4g, 176mmol, 91%) as a beige solid. H-NMR (300MHz,
tert-Butyl (2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- DMSO-d ) δ: 1.28 (3H, t, J=7.1Hz), 4.28 (2H, q, J=7.1Hz),
6
phenyl)propyl)carbamate (25) To a solution of 24 (63.3g, 6.15 (1H, d, J=7.9Hz), 8.13 (1H, s), 8.57 (1H, d, J=7.9Hz),
189mmol) in DMF (378mL) were added MeI (14.2mL, 11.73 (1H, brs).
2
27mmol) and K CO (31.3g, 227mmol) at r.t. The mixture 5-Hydroxypyrazolo[1,5-a]pyrimidine-3-carboxylic Acid
2
3
was stirred at r.t. for 2h. The mixture was poured into water (31) To a solution of 30 (20.8g, 100mmol) in THF (223mL)
at r.t. and extracted with EtOAc. The organic layer was sepa- and EtOH (111mL) was added 2M NaOH aqueous solution
rated, washed with saturated aqueous NaCl, dried over an- (201mL, 401mmol). The mixture was stirred at 50°C over-
hydrous MgSO and concentrated in vacuo. The residue was night. The mixture was evaporated under reduced pressure to
4
purified by column chromatography (silica gel, hexane–ethyl remove solvents and then neutralized with 2M HCl aqueous
acetate, 100:0 to 4:1) to give 25 (38.7g, 111mmol, 46% solution (201mL, 401mmol). The resulting solid was collected
1
from 21) as an off-white solid. H-NMR (300MHz, CDCl ) δ: by filtration, rinsed with water/EtOH and dried to give 31
1
3
1
.26–1.53 (9H, m), 3.73 (3H, s), 5.27–5.41 (1H, m), 5.54–5.75 (17.7g, 99mmol, 98%) as a beige solid. H-NMR (300MHz,
(
1H, m), 7.20 (2H, d, J=8.3Hz), 7.35–7.44 (2H, m). MS (ESI/ DMSO-d ) δ: 6.14 (1H, d, J=7.9Hz), 8.09 (1H, s), 8.56 (1H, d,
6
−
APCI) m/z 348.1 [M−H] .
J=7.9 Hz).
tert-Butyl (2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)-
5-Chloropyrazolo[1,5-a]pyrimidine-3-carbonyl Chloride
phenyl)propyl)carbamate (26) To a solution of 25 (23.5g, (32) To cooled (0°C) POCl (71.0mL, 762mmol) were added
3
67.3mmol) in THF (336mL) was added dropwise MeMgBr 31 (5.00g, 27.9mmol) and DIEA (16.1mL, 92.1mmol). The
(269mL, 269mmol) at 0°C. The mixture was stirred at 0°C mixture was stirred at 130°C for 4h and then concentrated in
under Ar for 1h. The mixture was quenched with saturated vacuo. The residue was diluted with toluene and water at 0°C,
aqueous NH Cl at 0°C and extracted with EtOAc. The organic and then extracted with EtOAc. The organic layer was sepa-
4
layer was separated, washed with saturated aqueous NaCl, rated, dried over anhydrous Na SO , filtered through a cake
2
4
dried over anhydrous MgSO4 and concentrated in vacuo. of silica gel pad (eluted with ethyl acetate) and concentrated
The residue was purified by column chromatography (silica in vacuo. The resulting solid was triturated with heptane, col-
gel, hexane–ethyl acetate, 19:1 to 2:3) to give 26 (17.9g, lected by filtration, washed with heptane and dried to give 32

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1071
1
(
4.46g, 20.7mmol, 74%) as a beige solid. H-NMR (300MHz, extracted with EtOAc. The combined organic phases were
CDCl ) δ: 7.16 (1H, d, J=7.2Hz), 8.65 (1H, s), 8.70 (1H, d, washed with water and saturated aqueous NaCl, dried over
3
J=7.2 Hz).
anhydrous Na SO4 and concentrated in vacuo. The residue
2
-Chloro-N-((1S)-2-hydroxy-2-methyl-1-(4-(trifluo- was purified by column chromatography (silica gel, hexane–
5
romethoxy)phenyl)propyl)pyrazolo[1,5-a]pyrimidine-3-car- ethyl acetate, 19:1 to 1:1) to afford 36 (3.69g, 16.3mmol,
+
boxamide (33) To a cooled (0°C) mixture of 32 (1.48g, 61%) as a white solid. MS (ESI/APCI) m/z 226.1 [M+H] .
6
.85mmol) and CH CN (30mL) were added 28 (1.96g,
Ethyl
5-(Pyridin-2-yl)pyrazolo[1,5-a]pyrimidine-3-car-
3
6.85mmol) and DIEA (3.58mL, 20.6mmol), and the mixture boxylate (38) A mixture of ethyl 36 (300mg, 1.33mmol), lithi-
was stirred at 0°C to r.t. for 16h. The mixture was poured into um 4-methyl-1-(pyridin-2-yl)-2,6,7-trioxa-1-borabicyclo[2.2.2]-
water and extracted with EtOAc. The organic layer was sepa- octan-1-uide (37) (566mg, 2.66mmol), Ph3P (139mg,
rated, washed with saturated aqueous NaCl, dried over anhy- 0.532mmol), copper(I) iodide (127mg, 0.665mmol) and
drous Na SO , and concentrated in vacuo. The residue was Pd(OAc)2 (29.9mg, 0.133mmol) in DMF (5mL) was stirred
2
4
purified by column chromatography (silica gel, hexane–ethyl at 80°C for 2h under Ar. The mixture was filtered through
acetate, 100:0 to 0:100) to give 33 (2.27g, 5.29mmol, 77%) a cake of basic silica gel pad (eluted with ethyl acetate), fol-
1
as white amorphous solids. H-NMR (300MHz, CDCl ) δ: lowed by a cake of silica gel pad (eluted with ethyl acetate).
3
1
.19 (3H, s), 1.40 (3H, s), 1.88 (1H, s), 5.12 (1H, d, J=8.7Hz), The appropriate fractions were concentrated in vacuo. The
6
.99 (1H, d, J=7.2Hz), 7.17–7.23 (2H, m), 7.45–7.53 (2H, m), resulting solid was triturated with hot ethyl acetate and in-
8.56 (1H, d, J=8.7Hz), 8.62 (1H, s), 8.67 (1H, d, J=7.2Hz). soluble materials were filtered off. After concentration of the
+
MS (ESI/APCI) m/z 429.1 [M+H] .
filtrate, the residue was purified by column chromatography
N-((1S)-2-Hydroxy-2-methyl-1-(4-(trifluoromethoxy)- (basic silica gel, hexane–ethyl acetate, 100:0 to 3:7) to give
phenyl)propyl)-5-(prop-1-en-2-yl)pyrazolo[1,5-a]pyrimi- 38 (309mg, 1.15mmol, 87%) as a pale yellow solid after tritu-
1
dine-3-carboxamide (34) Into a microwave vial equipped ration with diisopropyl ether. H-NMR (300MHz, CDCl ) δ:
3
with a magnetic stirrer were added 33 (76.1mg, 0.177mmol), 1.47 (3H, t, J=7.1Hz), 4.46 (2H, q, J=7.1Hz), 7.40–7.47 (1H,
4
,4,5,5-tetramethyl-2-(prop-1-en-2-yl)-1,3,2-dioxaborolane m), 7.86–7.95 (1H, m), 8.25 (1H, d, J=7.4Hz), 8.59 (1H, s),
(
46.6mg, 0.266mmol), K CO3 (36.8mg, 0.266mmol), tolu- 8.71–8.76 (2H, m), 8.80 (1H, d, J=7.4Hz). MS (ESI/APCI) m/z
2
+
ene (4mL) and water (0.4mL), followed by (Amphos) PdCl2 269.1 [M+H] .
2
(
12.6mg, 0.0177mmol). The reaction vial was flushed with
nitrogen, sealed and heated by microwave irradiation at Acid (39) A mixture of 38 (117mg, 0.436mmol), 4M NaOH
50°C for 25min. The mixture was purified by column chro- aqueous solution (0.545mL, 2.18mmol), THF (3.5mL) and
matography (basic silica gel, hexane–ethyl acetate, 19:1 to EtOH (3.5mL) was stirred at 60°C for 2h. After being cooled
:4), followed by a second column purification (silica gel, to 0°C, the mixture was acidified with 6M HCl aqueous solu-
hexane–ethyl acetate, 19:1 to 1:4) to afford 34 (48.8mg, tion (0.363mL) and concentrated in vacuo to give crude 39
5-(Pyridin-2-yl)pyrazolo[1,5-a]pyrimidine-3-carboxylic
1
1
1
0
.112mmol, 63%) as a yellow amorphous solid. H-NMR (79.0mg, 0.329mmol, 75%). This was used in the next reaction
(
300MHz, DMSO-d ) δ: 1.00 (3H, s), 1.27 (3H, s), 2.37 (3H, without further purification.
6
s), 4.88–4.98 (2H, m), 5.80 (1H, s), 6.36 (1H, s), 7.24–7.34 (2H,
tert-Butyl 2-Cyano-3-(dimethylamino)acrylate (42) A
m), 7.49–7.58 (2H, m), 7.65 (1H, d, J=7.5Hz), 8.45 (1H, s), mixture of tert-butyl 2-cyanoacetate (40) (24.3g, 172mmol)
9
.02 (1H, d, J=8.7Hz), 9.22 (1H, d, J=7.5Hz). MS (ESI/APCI) and tert-butoxy bis(dimethylamino)methane (41) (35.5mL,
+
m/z 435.2 [M+H] .
172mmol) was stirred at r.t. for 30min. The mixture was
5
-(3,6-Dihydro-2H-pyran-4-yl)-N-((1S)-2-hydroxy- evaporated to give 42 as a pale yellow solid. This was used
1
2
-methyl-1-(4-(trifluoromethoxy)phenyl)propyl)pyrazolo- in the next reaction without further purification. H-NMR
[
1,5-a]pyrimidine-3-carboxamide (35) Into a microwave (300MHz, CDCl ) δ: 1.50 (9H, s), 3.17 (3H, s), 3.36 (3H, s),
3
vial equipped with a magnetic stirrer were added 33 (142mg, 7.62 (1H, s).
.330mmol), 2-(3,6-dihydro-2H-pyran-4-yl)-4,4,5,5-tetrameth- tert-Butyl 3-Amino-1H-pyrazole-4-carboxylate (43)
yl-1,3,2-dioxaborolane (104mg, 0.495mmol), K CO (68.5mg, mixture of 42 (33.8g, 172mmol) and hydrazine monohydrate
0
A
2
3
0.495mmol), toluene (4mL) and water (0.4mL), followed by (8.39mL, 172mmol) in MeOH (344mL) was stirred at 70°C
(
Amphos) PdCl (23.4mg, 0.0330mmol). The reaction vial overnight. After cooling to r.t., the mixture was poured into
2
2
was flushed with nitrogen, sealed, and heated by microwave water at r.t. and extracted with EtOAc. The organic layer was
irradiation at 150°C for 25min. The mixture was purified separated, washed with saturated aqueous NaCl, dried over
by column chromatography (silica gel, hexane–ethyl acetate, anhydrous MgSO4 and concentrated in vacuo. The residue
1
9:1 to 1:4) to afford 35 (110mg, 0.230mmol, 70%) as a yel- was crystallized from diisopropyl ether to give 43 (20.0g,
1
1
low solid. H-NMR (300MHz, DMSO-d ) δ: 1.00 (3H, s), 1.28 109mmol, 64%) as a pale orange solid. H-NMR (300MHz,
6
(
3H, s), 2.67–2.94 (2H, m), 3.84–3.97 (2H, m), 4.35–4.43 (2H, DMSO-d ) δ: 1.47 (9H, s), 4.86–6.14 (2H, m), 7.07–7.92 (1H,
6
+
m), 4.86–5.00 (2H, m), 7.24–7.34 (3H, m), 7.49–7.56 (2H, m), m), 11.44–12.34 (1H, m). MS (ESI/APCI) m/z 184.1 [M+H] .
.59 (1H, d, J=7.5Hz), 8.42 (1H, s), 9.06 (1H, d, J=8.7Hz), Methyl 2,3-Dibromo-2-methylpropanoate (45) To
solution of methyl methacrylate (60.1g, 600mmol) in EtOAc
Ethyl 5-Chloropyrazolo[1,5-a]pyrimidine-3-carboxylate (353mL) at 0°C was added a solution of bromine (30.7mL,
36) POCl3 (30mL, 322mmol) was added to 30 (5.54g, 600mmol) in EtOAc (30mL). The mixture was stirred at r.t.
6.7mmol) and the mixture was stirred at 100°C for 16h. for 16h and then saturated aqueous Na S O was added. The
7
9
a
+
.20 (1H, d, J=7.5Hz). MS (ESI/APCI) m/z 477.2 [M+H] .
(
2
2
2
3
After POCl was removed under reduced pressure, the residue mixture was diluted with water and extracted with EtOAc.
3
was partitioned between EtOAc and NaHCO aqueous solu- The combined organic phases were washed with saturated
3
tion. The phases were separated and the aqueous phase was aqueous NaCl, dried over anhydrous Na SO and concentrated
2
4

1
072
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
in vacuo to afford 45 (152g, 584mmol, 97%) as a colorless oil. added methanesulfonic acid (2.78mL, 42.0mmol) at 0°C. The
1
H-NMR (300MHz, DMSO-d ) δ: 1.95 (3H, s), 3.76 (3H, s), mixture was stirred at r.t. overnight. The starting material re-
6
4
.08–4.14 (1H, m), 4.16–4.22 (1H, m).
Methyl 3,3-Dimethoxy-2-methylpropanoate (46) A sus- 8.40mmol) was added. The mixture was stirred at r.t. for 1h
pension of sodium methoxide (28% MeOH solution, 226g, and then neutralized with 1M NaOH aqueous solution at 0°C.
.17mol) in MeOH (234mL) was heated to 70°C and then a The organic solvent was evaporated under reduced pressure
mained on TLC and further methanesulfonic acid (0.555mL,
1
solution of 45 (152g, 584mmol) in MeOH (20mL) was added and then diluted with water. The resulting solid was collected
rapidly. The mixture was stirred at 70°C for 3h. After cool- by filtration, washed with water and then hexane, and dried to
1
ing to r.t., the mixture was filtered and washed with MeOH, give 52a (1.98g, 7.67mmol, 91%) as a white solid. H-NMR
and then the filtrate was concentrated in vacuo. The residue (300MHz, DMSO-d ) δ: 2.40 (3H, s), 2.52 (3H, d, J=0.8Hz),
6
was partitioned between Et O and water. The organic layer 8.50 (1H, d, J=1.1Hz), 8.62 (1H, s), 9.50 (1H, d, J=1.1Hz),
2
+
was washed with water and saturated aqueous NaCl, dried 12.52 (1H, brs). MS (ESI/APCI) m/z 259.2 [M+H] .
over anhydrous MgSO , filtered, and concentrated in vacuo
tert-Butyl 5-Chloro-6-methylpyrazolo[1,5-a]pyrimidine-
4
to give 46 (54.5g, 336mmol, 58%) as a pale yellow oil. 3-carboxylate (50) Carbon tetrachloride (9.73mL, 100mmol)
This was used in the next reaction without further purifica- was added to a solution of PPh (27.1g, 100mmol) in 1,2-di-
3
1
tion. H-NMR (300MHz, CDCl ) δ: 1.17 (3H, d, J=7.2Hz), chloroethane (223mL) at r.t. The mixture was stirred at the
.71–2.85 (1H, m), 3.35 (3H, s), 3.38 (3H, s), 3.70 (3H, s), 4.50 same temperature under N for 30min. To this mixture was
3
2
2
(
1H, d, J=7.6 Hz).
added a suspension of 47 (5.0g, 20.1mmol) in 1,2-dichloroeth-
tert-Butyl 5-Hydroxy-6-methylpyrazolo[1,5-a]pyrimidine- ane (111mL) at r.t. and the resulting mixture was stirred under
3
-carboxylate (47) To a mixture of 43 (20.0g, 109mmol) N at 75–85°C for 4.5h, and then concentrated in vacuo. The
2
and 46 (26.6g, 164mmol) in DMF (219mL) was added ce- residue was partitioned between EtOAc and water. The phases
sium carbonate (64.2g, 197mmol). The mixture was stirred were separated and the aqueous phase was extracted with
at 100°C for 16h, diluted with water and then acidified to pH EtOAc. The combined organic phases were washed with satu-
ca. 4 with AcOH. The resulting solid was filtered by filtration, rated aqueous NaCl, dried over anhydrous Na SO , and con-
2
4
washed with water and dried to afford 47 (22.6g, 91.8mmol, centrated in vacuo. The residue was purified by column chro-
1
8
1
4%) as a white solid. H-NMR (300MHz, DMSO-d ) δ: matography (silica gel, hexane–ethyl acetate, 100:0 to 7:3)
6
1
.53 (9H, s), 1.96 (3H, d, J=1.1Hz), 7.97 (1H, s), 8.53 (1H, d, to give 50 (4.75g, 17.7mmol, 88%) as a white solid. H-NMR
J=1.1Hz), 11.22 (1H, brs). MS (ESI/APCI) m/z 250.1 [M+H] . (300MHz, CDCl ) δ: 1.62 (9H, s), 2.42 (3H, d, J=0.9Hz), 8.40
+
3
tert-Butyl 5-Azido-6-methylpyrazolo[1,5-a]pyrimidine- (1H, s), 8.52 (1H, d, J=0.9 Hz).
3
-carboxylate (48) To a suspension of 47 (4.99g, 20.0mmol)
in THF (100mL) was added 1,8-diazabicyclo[5.4.0]undec-7-ene pyrazolo[1,5-a]pyrimidine-3-carboxylate (51b) To a solu-
DBU) (4.52mL, 30.0mmol) and DPPA (5.17mL, 24.0mmol) tion of 50 (1.08g, 4.03mmol) in DMF (20mL) was added
at r.t. The mixture was stirred at 60°C under Ar for 3h. DBU 3-methyl-1H-1,2,4-triazole (0.469g, 5.65mmol) and K CO
tert-Butyl 6-Methyl-5-(3-methyl-1H-1,2,4-triazol-1-yl)-
(
2
3
(1.51mL, 10.0mmol) and DPPA (1.72mL, 8.00mmol) were (0.781g, 5.65mmol) at r.t. The mixture was stirred at r.t.
added thereto. The mixture was stirred at 60°C under Ar for overnight and warmed to 60°C for 2h. After cooling to r.t.,
2
h. The mixture was poured into water at r.t. and extracted water was added to the mixture. The solid was collected by
with a 1:1 mixture of hexane and EtOAc. The organic layer filtration, washed with water, and dried to give 51b (0.870g,
1
was separated, washed with saturated aqueous NaCl, dried 2.77mmol, 69%) as a tan solid. H-NMR (300MHz, DMSO-
over anhydrous MgSO , filtered through a short pad of silica d6) δ: 1.56 (9H, s), 2.43 (3H, s), 2.54 (3H, d, J=1.1Hz), 8.56
4
gel, eluted with a 1:1 mixture of hexane and EtOAc, and (1H, s), 9.10 (1H, s), 9.43 (1H, d, J=1.1Hz). MS (ESI/APCI)
+
concentrated in vacuo. The resulting solid was triturated with m/z 315.2 [M+H] .
hexane to give 48 (4.03g, 14.7mmol, 74%) as a pale green
6-Methyl-5-(3-methyl-1H-1,2,4-triazol-1-yl)pyrazolo[1,5-
1
solid. H-NMR (300MHz, DMSO-d ) δ: 1.51–1.65 (9H, m), a]pyrimidine-3-carboxylic Acid (52b) To a suspension of
6
2
.12–2.55 (3H, m), 8.38–8.49 (1H, m), 8.85–9.11 (1H, m). MS 51b (0.87g, 2.77mmol) in CH CN (14mL) was added MsOH
(1.26mL, 19.4mmol) at 0°C. After being stirred at r.t. for
tert-Butyl 6-Methyl-5-(4-methyl-1H-1,2,3-triazol-1-yl)- 5h, the mixture was warmed to 60°C for 30min. One molar
3
+
(
ESI/APCI) m/z 275.1 [M+H] .
pyrazolo[1,5-a]pyrimidine-3-carboxylate (51a) A mixture NaOH aqueous solution (19.4mL, 19.4mmol) was added there-
of 48 (4.36g, 15.9mmol) and 49 (6.07g, 19.1mmol) in toluene to at 0°C. CH CN was evaporated. The solid was collected
3
(
79mL) was stirred at 80°C for 3h. To the mixture was added by filtration, washed with water, dried to give 52b (0.606g,
1
MgCl (4.54g, 47.7mmol). After 15min, to the mixture was 2.35mmol, 85%) as a tan solid. H-NMR (300MHz, DMSO-
2
added MgCl2 (3.03g, 31.8mmol). The mixture was stirred d6) δ: 2.42 (3H, s), 2.48–2.53 (3H, m), 8.60 (1H, s), 9.09 (1H,
at 60°C under Ar for 2h. The solid was filtered off, washed s), 9.44 (1H, d, J=1.1Hz), 12.49 (1H, brs). MS (ESI/APCI) m/z
+
with hot toluene. The filtrate was evaporated. The residue was 259.1 [M+H] .
purified by column chromatography (silica gel, hexane–ethyl
Enzyme Assay Protocol
acetate, 19:1 to 9:11) to give 51a (4.83g, 15.4mmol, 97%) as
Preparation of Human PDE
Human PDE1A, 3A, 4D2, 5A1, 7B, 8A1, 9A2, and 11A4
1
a white solid. H-NMR (300MHz, CDCl ) δ: 1.64 (9H, s), 2.46
3
(
8
3H, d, J=0.8Hz), 2.78 (3H, d, J=1.1Hz), 8.46–8.50 (2H, m), enzymes were purchased from BPS Bioscience. Human
+
.71–8.73 (1H, m). MS (ESI/APCI) m/z 315.2 [M+H] .
PDE6AB enzyme was purchased from SB Drug Discovery.
6
-Methyl-5-(4-methyl-1H-1,2,3-triazol-1-yl)pyrazolo[1,5- Human PDE2A3 full-length gene was transduced into Sf9
a]pyrimidine-3-carboxylic Acid (52a) To a stirred sus- cells, and the enzyme was purified by His-tag affinity col-
pension of 51a (2.64g, 8.40mmol) in CH CN (84mL) was umn and gel filtration. Human PDE10A2 was generated from
3

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1073
2
COS-7 cells that had been transfected with the full-length (pore size 0.4µm, 0.143cm surface area) with polyethylene
gene. The enzymes were stored at −70°C until use.
PDE2A3 Enzyme Inhibitory Assay
terephthalate membrane (Corning Life Sciences, Lowell, MA,
U.S.A.) at a density of 1.125×105 cells/well. The cells were
PDE activity was measured using an SPA (Scintillation preincubated with M199 at 37°C for 30min. Subsequently,
Proximity Assay) (GE Healthcare). To evaluate the inhibitory transcellular transport was initiated by the addition of M199
activity of a compound, 10µL of serially diluted compounds either to apical compartments (75µL) or to basolateral com-
were incubated with 20µL of PDE enzyme (final concentra- partments (250µL) containing 10µM digoxin, 200µM lucifer
tion 0.023nM) in assay buffer (50mM N-(2-hydroxyethyl)- yellow (as a marker for the tightness of the monolayer), and
piperazine-Nꢀ-2-ethanesulfonic acid (HEPES)–NaOH, 8.3mM 10 µM test compounds. The assay was terminated by the re-
MgCl2, 1.7mM ethylene glycol bis(2-aminoethyl ether)- moval of each assay plate after 2h. Aliquots (25µL) from the
N,N,Nꢀ,Nꢀ-tetraacetic acid (EGTA), and 0.1% bovine serum opposite compartments were mixed with CH CN contain-
3
albumin (BSA) (pH 7.4)) for 30min at r.t. Final concentration ing alprenolol and diclofenac as internal standards, and then
of dimethyl sulfoxide (DMSO) in the reaction solution was centrifuged. The compound concentrations in the supernatant
1%. Compounds were tested in duplicate in 96-well half-area were measured by LC-MS/MS. The apparent permeability
plates (Corning) or 384-well OptiPlates (PerkinElmer, Inc.). (Papp) of test compounds in the receiver wells was determined
We used an 8 concentration serial dilution dose response rang- and the efflux ratio (ER) for the MDR1 membrane permeabil-
ing from 100µM to 10pM compound concentrations. To start ity test was calculated using the following equation:
3
the reaction, 10µL of substrate [ H] cGMP (final concentra-
tion 77nM, PerkinElmer, Inc.) were added to a total volume of
ER = Papp, B to A / Papp, A to B
4
0µL. After 60min at r.t., 20µL of 20mg/mL yttrium silicate where Papp, AtoB is the apical-to-basal passive permeability-sur-
SPA beads containing zinc sulfate were added to terminate face area product and Papp, BtoA is the basal-to-apical passive
the PDE reaction. After resting undisturbed for an additional permeability-surface area product.
6
0min, the assay plates were counted in a scintillation counter
Phototoxicity Test Phototoxicity assay was carried out
4
6
(
PerkinElmer, Inc.) to allow calculation of the inhibition rate. as described in the OECD guideline No. 432 with some
Inhibition rate was calculated based on 0% control wells with modifications for a high-throughput screening. BALB/c 3T3
enzyme and DMSO, and 100% control wells without enzyme. cells were cultured at 37°C under 5% CO2 in Dulbecco’s
All IC50 values were obtained by fitting the results to the fol- modified Eagle’s medium (DMEM) supplemented with 10%
lowing 4 Parameter Logistic Equation:
fetal bovine serum, 50IU/mL penicillin and 50mg/mL strep-
tomycin. Cells were seeded at 2.5×10 cells/well in 384-well
white plates, and cultured in DMEM supplemented with 10%
3
(
C−x)×D
y= A+(B − A) / (1+(10
))
where A is the minimum y value, B is the maximum y value, fetal bovine serum, 2mM L-glutamine, 1mM sodium pyruvate,
C is Log(EC ) value, and D is the slope factor.
Human PDE Enzyme Assay
50IU/mL penicillin, and 50µg/mL streptomycin for 1d. Two
384-well plates per test compound (50µM) in Earle’s Balanced
5
0
PDE activities were measured using an SPA (GE Health- Salt Solution (EBSS) supplemented with 1mM HEPES were
care). To evaluate the inhibitory activity, 10µL of serially di- preincubated for 1h. One of the two plates was irradiated
2
2
luted compounds were incubated with 20µL of PDE enzymes (+UV) for 60min with 1.4–1.7mW/cm (5–6 J/cm ), whereas
(
8
except for PDE1A) in assay buffer (50mM HEPES–NaOH, the other plate was kept in the dark (−UV). In both plates the
.3mM MgCl , 1.7mM EGTA, and 0.1% BSA (pH 7.4)) for treatment medium was replaced with the culture medium and
2
30min at r.t. The PDE1A enzyme assay was performed in a after another 24h of culture, the cell viability was determined
different assay buffer (50mM Tris–HCl, 8.3mM MgCl , 0.2mM by measuring the cellular ATP content. The cellular ATP con-
2
CaCl , 0.1% BSA, and 30nM Calmodulin (pH 7.5)). The final tent was measured using the Celltiter-Glo™ assay kit (Prome-
2
concentration of DMSO in the assay was 1%. Compounds ga) following the manufacture’s instruction. ATP content was
were tested in duplicate in 96-well half-area plates (Corning). calculated as follows. ATP content (% of control)=(relative
We used an 4 concentration serial dilution dose response rang- light unit (RLU) of test compound/RLU of 1% DMSO)×100.
ing from 10µM to 10nM compound concentrations. To start
Estimation of LogD at pH 7.4 LogD , which is the
7.4
3
the reaction, 10µL of substrate ([ H] cGMP (final concentra- partition coefficient of the compounds between 1-octanol and
tion 77nM, PerkinElmer, Inc.) for PDE1A, 5A1, 6AB, 9A2, aqueous buffer at pH 7.4, was measured using a chromato-
3
59)
1
0A2, and 11A4 or [ H] cAMP (final concentration 14.7nM, graphic procedure based on a published method. The instru-
PerkinElmer, Inc.) for PDE3A, 4D2, 7B, and 8A1) were added ments utilized were a Waters Alliance 2795 HPLC system and
for a final assay volume of 40µL. After 60min incubation at a 2996 UV-Vis detector (Milford, MA, U.S.A.).
r.t., 20µL of 20mg/mL yttrium silicate SPA beads containing
Thermodynamic Solubility Measurement Using the
ZnSO were added to terminate the PDE reaction. After rest- Shake-Flask Method The measurement of thermodynamic
4
59)
ing undisturbed for more than 120min, the assay plate was solubility was carried out as previously described. Briefly,
counted in a scintillation counter (PerkinElmer, Inc.) to allow the drug substances were weighed into Thomson filter vials
calculation of the inhibition rate.
(Chrom Tech, Inc., Minnesota, U.S.A.). JP1 (pH 1.2), JP2
Transcellular Transport Study Using a Transporter- (pH 6.8), and JP2-containing 20mM GCDC were added to
Expression System Human MDR1-expressing LLC-PK1 the vials. The vials were incubated at 37°C for 18h and the
cells were cultured with minor modifications to the method resulting suspensions were filtered by compressing the vials.
57)
reported previously. The transcellular transport study was The drug concentration of the filtrates was determined using
58)
performed as reported previously. In brief, the cells were a UHPLC system.
grown for 7d in HTS Transwell 96-well permeable support Powder X-Ray Diffraction (PXRD) and Crystallinity

1
074
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
Calculation PXRD patterns were collected using a RINT coordinates and structure factors have been deposited in the
UltimaIV powder X-ray diffractometer (Rigaku Corp., Tokyo, Protein Data Bank with accession code 5VP1.
59)
Japan) according to previously described conditions. The
Animal Experiments The care and use of animals and
peak intensities of the crystalline (Ic) and non-crystalline (Ia) the experimental protocols were approved by the Experimen-
fractions were integrated from the baseline collection accord- tal Animal Care and Use Committee of Takeda Pharmaceuti-
6
0,61)
ing to Herman’s method.
The crystallinity was calculated cal Company Limited.
using the following equation with an autocrystallinity calcula-
Pharmacokinetic Analysis in Rat or Mouse Cassette
Dosing Compound 20 was administered intravenously
0.1mg/kg) or orally (1mg/kg) by cassette dosing to nonfasted
male Crl:CD(SD)(IGS) rats (8W, n=3) or male ICR mice (8W,
61,62)
tion software (Rigaku).
Crystallinity= Ic×100 / (Ic+Ia)×100
(
Protein Expression and Purification The PDE2A cata- n=3). The combination for a cassette dosing was determined
lytic domain (578–919) was cloned into a pFastBac vector, to avoid combinations of compounds with the same molecu-
for expression in Sf9 cells, utilizing an N-terminal 6×poly- lar weight. The solution of compounds in dimethylacetamide
histidine tag containing a TEV cleavage site. Large scale containing 50% (v/v) 1,3 butanediol at 0.1mg/mL/kg was
production of recombinant protein was carried out in Sf9 administered intravenously to isoflurane-anesthetized mice via
cells. The pellet from 10L of baculovirus infected Sf9 cells femoral vein. The suspension of compounds in 0.5% methyl
was resuspended in 600mL lysis buffer containing 25mM Tris cellulose with water was used for vehicle (1mg/kg) and was
pH 7.6, 1M NaCl, 20mM imidazole, 5% glycerol, and 3 Roche administered orally by gavage. After administration, blood
cOmplete Protease Inhibitor tablets. The cell suspension was samples were collected via tail vein by syringes with heparin
homogenized with the Polytron PT-3100, centrifuged for 1h at 5, 10, 15, 30min, 1, 2, 4, and 8h (intravenously (i.v.)) and
at 13000rpm (JA-14 rotor), and the clarified supernatant was 15, 30min, 1, 2, 4, and 8h (per os (p.o.)), and centrifuged to
brought to 800mL with lysis buffer before batch binding with obtain the plasma fraction. The plasma samples were depro-
1
0mL of Probond Ni resin (Invitrogen) for 2h at 4°C, roll- teinized by mixing with acetonitrile followed by centrifuga-
ing. The beads were collected by low speed centrifugation tion. The compound concentrations in the supernatant were
3500rpm with JS-4.2 rotor), loaded into a gravity column, measured by LC-MS/MS with a standard curve. Pharmaco-
(
and washed slowly overnight with 2L of wash buffer con- kinetic parameters were calculated by the non-compartmental
taining 25mM Tris pH 7.6, 1M NaCl, 20mM imidazole, 5% analysis. The area under the concentration–time curve (AUC)
glycerol. The following day the protein was eluted with buffer and the area under the first moment curve (AUMC) were
containing 25mM Tris pH 7.9, 50mM NaCl, 250mM imidazole, calculated using the linear trapezoidal method. The mean
10% glycerol. The 1.5mL sample eluted from the Nickel cap- residence time (MRT) was calculated as AUMC/AUC. The
ture step was brought to 9mL with Mono Q buffer A contain- total clearance (CLtotal) was calculated as doseiv/AUC . The
i.v.
ing 25mM Tris pH 7.9, and 10% glycerol. After the full sample volume of distribution (Vd ) was calculated as CLtotal×MRTi.v..
ss
volume was bound to the Mono Q column, a salt gradient was Oral bioavailability (F) was calculated as (AUCp.o./dosep.o.)/
applied from 0M NaCl to ca. 800mM NaCl in 40mL. Frac- (AUC /dose )×100.
i.v.
i.v.
tions corresponding to the unphosphorylated protein (identi-
Brain and Plasma Concentration in Rats Compound
fied by MS with MW=40178Da) were pooled for further 20 was administered orally to Long–Evans rats (male, non-
purification by size-exclusion chromatography on a Superdex fasted, 7-week old) at 10mg/kg. Blood and whole brain
2
00 column equilibrated in 1×TBS pH 7.4, 0.5mM DTT, 1mM samples were collected 2h after oral administration. The
EDTA, 10% glycerol. Peak SEC fractions were collected and blood samples were centrifuged to obtain the plasma fraction.
concentrated to 12mg/mL for crystallization. The brain samples were homogenized in saline to obtain the
Crystallization and Structure Determination Crystals brain homogenate. Compound concentrations were measured
suitable for data collection were first grown using the vapor in aliquots of rat plasma and brain, which were mixed well
diffusion method in hanging drops at r.t. by adding 0.5µL pro- with acetonitrile containing an internal standard and then
tein solution with 1mM IBMX (1-methyl-3-(2-methylpropyl)- centrifuged. The supernatants were diluted with solvents for
2
,3,6,7-tetrahydro-1H-purine-2,6-dione) and 0.5µL reservoir LC-MS/MS analysis (mobile phase A: 10mM ammonium
solution (30% PEG 3350, 0.1M Tris pH 7.5, and 0.2M MgCl2). formate–formic acid (100:0.2, v/v), mobile phase B: aceto-
PDE2A IBMX crystals were soaked in a drop containing nitrile–formic acid (100:0.2, v/v)). The diluted solutions were
5
0
mM compound 20, 31% PEG 3350, 0.1M Tris pH 7.5, and injected into an LC-MS/MS (API5000, AB Sciex, Foster City,
.2M MgCl for 3d. Crystals were transferred through a fresh CA, U.S.A.) equipped with a Shimadzu Shim-pack XR-ODS
2
cryo-protected soak drop immediately before being harvested column (2.2µm packing particle size, 2.0mm ID×30mm L)
and flash frozen in liquid nitrogen. X-Ray diffraction data was maintained at 50°C. The chromatographic separation was per-
collected at ALS beamline 5.0.2 using a Pilatus3 6M (Dectris) formed using gradient elution at a flow rate of 0.7mL/min. The
detector from a single cryogenically protected crystal (100K) LC time program was as follows: Mobile phase B was held at
at a wavelength of 1Å. The crystals belong to the space 5% for 0.1min, and increased linearly to 95% in 0.1min. After
group C121 and contain three enzyme molecules per asym- maintaining B at 95% for another 0.8min, it was decreased
metric unit. X-Ray diffraction data was reduced using the to 5% in 0.01min, followed by re-equilibration for 0.59min.
63)
HKL2000 software package. The structure was determined The total cycle time for one injection was 1.6min. Compound
by molecular replacement with PHASER within the CCP4 20 was detected using multiple reaction monitoring mode and
6
4)
program suite and refined with REFMAC. Several cycles of the transition m/z 490.01→240.96. Analyst™ software (version
65)
model building using MIfit and refinement using REFMAC 1.4.2) was used for data acquisition and processing.
were performed for improving the quality of the model. The In Vivo Occupancy Study In vivo target occupancy study

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
1075
of compound 20 was conducted using LC-MS/MS. Compound instructions. Values were expressed as pmol/mg tissue weight.
2
0 was suspended in 0.5% (w/v) methylcellulose in distilled
Step-through Passive Avoidance Task This task was
6
6)
67)
water, and PF-05270430
was dissolved in dimethylacet- performed as previously described
with some modifica-
amide and 1,3-butanediol (1:1). Sprague-Dawley (SD) rats tions. This experiment was conducted in 7–8-week-old male
were pretreated with vehicle (p.o., n=6) or compound 20 Long–Evans rats. The apparatus (Brainscience idea, Osaka,
(0.3, 1, 3, 10, 30mg/kg, p.o., n=4 in each group) 2h before Japan) consisted of an illuminated chamber (25×10×25cm)
sampling. PF-05270430 (0.1mg/mL/kg) was administered by connected to a dark chamber (30×30×30cm) by a sliding door
bolus intravenous injection via lateral tail vein 30min before (8×8cm). On the training day, each animal was subjected to a
sampling. Rats were decapitated, and blood and brain samples single pre-training trial 4–6h before the acquisition trial. The
(striatum as target tissue and cerebellum as reference tissue) rat was placed in the light chamber, and the sliding door was
were collected. Brain samples were weighed and saline was opened 30s later. As soon as the rat entered the dark chamber
added (20% (w/v)), followed by homogenization with Lysing with all four paws, the door was closed. The rat was then al-
Matrix I beads (MP Biomedicals). Homogenized samples were lowed to remain in the dark chamber for 30s before being
stored at –30°C until quantification of tracer (PF-05270430) returned to its home cage. In the acquisition trial, the rat was
using LC-MS/MS. The supernatants were diluted with sol- placed in the light chamber, and the sliding door was opened.
vents for LC-MS/MS analysis (mobile phase A: 10mM am- The time required for the rat to enter the dark chamber was
monium formate–formic acid (100:0.2, v/v), mobile phase then recorded. As soon as the rat entered the dark chamber,
B: acetonitrile–formic acid (100:0.2, v/v)). The diluted solu- the door was closed, and an electric shock (0.5mA, 3s) was
tions were injected into an LC-MS/MS (API5000, AB Sciex) delivered from the floor grid. The rat was then returned to its
equipped with a Shimadzu Shim-pack XR-ODS (2.2µm pack- home cage. The retention test was conducted 24h later. The
ing particle size, 2.0mm ID×30mm L) maintained at 50°C. rat was again placed in the light chamber with the sliding
The chromatographic separation was performed using gradient door. After 30s, the door was opened and the latency for the
elution at a flow rate of 0.7mL/min. The LC time program rat to cross over into the dark compartment was recorded. If
was as follows: Mobile phase B was held at 5% for 0.1min, the animal did not enter the dark chamber within 300s, the
and increased linearly to 95% in 0.1min. After maintain- retention test was terminated, and the animal was given a
ing B at 95% for another 0.8min, it was decreased to 5% in ceiling score of 300s. Vehicle or compound 20 (3mg/kg) was
0.01min, followed by re-equilibration for 0.59min. The total administered p.o. 2h prior, and saline or MK-801 (0.1mg/kg)
cycle time for one injection was 1.6min. PF-05270430 was was administered s.c. 30min prior to the acquisition trial. The
detected using multiple reaction monitoring mode and the statistical significance of differences between group latency
transition m/z 432→386. Analyst™ software (version 1.4.2) scores was determined by Wilcoxon’s test with significance
was used for data acquisition and processing. Specific tracer set at p≤0.05.
binding (B ) in the striatum was represented as the difference
SP
between the tracer concentration in striatum and that in cer-
Acknowledgments We acknowledge Mr. Kouta Matsu-
ebellum. PDE2A occupancy was calculated using the follow- miya for metabolite identification studies on 6, Mr. Mitsuyoshi
ing equation: Occupancy (%)=(BSP, base−BSP, drug)/BSP, base×100, Nishitani for obtaining the single crystal X-ray structure of
where BSP, base and BSP, drug are the concentrations at baseline S2, Ms. Miki Hara for chiral separations, Ms. Shoko Takeuchi
(vehicle treatment) and at drug treatment, respectively. Curve for thermodynamic solubility and crystallinity measurements
fitting of the saturation curve was carried out by nonlinear on 19 and 20, Dr. Gyorgy Snell and Mr. Scott Lane (Takeda
regression using GraphPad Prism 5.02 (GraphPad Software, California, Inc.) for crystallographic data collection and pro-
Inc., San Diego, CA, U.S.A.).
cessing of 20, and Dr. Toshihiro Imaeda for supplying large
Measurement of Cyclic Nucleotide Contents in Rat Brain amounts of key intermediate 28.
Animals
Five-week-old male Long–Evans rats were purchased from
Conflict of Interest The authors declare no conflict of
Japan SLC, Inc. (Japan). The rats were housed in groups of interest.
3
/cage in a light-controlled room (12h light/dark cycles with
lights on at 07:00). Food and water were provided ad libitum.
After a one week acclimation period, the six-week-old rats ticle contains supplementary materials. Synthesis of (+)-di-(p-
Supplementary Materials The online version of this ar-
were used for experiments.
Measurements
toluoyl)-D-tartaric acid salt (S2) of free base of 28 suitable for
X-ray crystallography. Synthesis of pyrazolo[1,5-a]pyrimidine
Compound 20 was suspended in 0.5% (w/v) methylcellu- derivatives 5 and 6. Full details of the X-ray structure analysis
lose in distilled water, and was administered in a volume of of S2. Molecular formula strings.
2
mL/kg body weight for rats. Rats were administered orally
with either vehicle or compound 20 (1, 3, 10mg/kg) after >1h References
of habituation. A microwave fixation system (Muromachi
Kikai, Tokyo, Japan) was used to sacrifice unanesthetized rats
by exposure of the head to the microwave beam at 2h after
administration of 20. Brain tissues were isolated and then
homogenized in 0.5mol/L HCl, followed by centrifugation.
The concentration of cyclic nucleotides in the supernatant was
measured using a cyclic AMP EIA kit or cyclic GMP EIA
kit (Cayman Chemical, U.S.A.) following the manufacturer’s
1) van Os J., Kapur S., Lancet, 374, 635–645 (2009).
Tandon R., Keshavan M. S., Nasrallah H. A., Schizophr. Res., 100,
–19 (2008).
Tandon R., Keshavan M. S., Nasrallah H. A., Schizophr. Res., 102,
–18 (2008).
McGrath J., Saha S., Chant D., Welham J., Epidemiol. Rev., 30,
7–76 (2008).
Miyamoto S., Duncan G. E., Marx C. E., Lieberman J. A., Mol.
2
)
4
3)
1
4)
6
5
)

1
076
Chem. Pharm. Bull.
Vol. 65, No. 11 (2017)
Psychiatry, 10, 79–104 (2005).
Murphy B. P., Chung Y.-C., Park T.-W., McGorry P. D., Schizophr.
Res., 88, 5–25 (2006).
Lieberman J. A., Stroup T. S., McEvoy J. P., Swartz M. S., Rosen-
heck R. A., Perkins D. O., Keefe R. S., Davis S. M., Davis C.
E., Lebowitz B. D., Severe J., Hsiao J. K., N. Engl. J. Med., 353,
33) Chappie T. A., Humphrey J. M., Verhoest P. R., Yang E., Helal C. J.,
WO 2012114222, August 30, 2012.
34) Helal C. J., Chappie T. A., Humphrey J. M., WO 2012168817, De-
cember 13, 2012.
35) Mikami S., Nakamura S., Ashizawa T., Sasaki S., Taniguchi T., No-
mura I., Kawasaki M., WO 2013161913, October 31, 2013.
36) Nakamura S., Mikami S., Kawasaki M., Nomura I., Ashizawa T.,
Taniguchi T., WO 2014010732, January 16, 2014.
37) Kawasaki M., Mikami S., Nakamura S., Negoro N., Ikeda S.,
Nomura I., Ashizawa T., Imaeda T., Seto M., Sasaki S., Marui S.,
Taniguchi T., WO 2015012328, January 29, 2015.
6
)
7
)
1209–1223 (2005).
8)
Gothelf D., Apter A., Reidman J., Brand-Gothelf A., Bloch Y.,
Gal G., Kikinzon L., Tyano S., Weizman R., Ratzoni G., J. Neural
Transm., 110, 545–560 (2003).
9)
Ayano G., J. Schizophr. Res., 3, 1027.1–1027.5 (2016).
1
0) Niewoehner U., Schauss D., Hendrix M., Koenig G., Boess F. G.,
van der Staay F. J., Schreiber R., Schlemmer K. H., Grosser R., WO
002050078, June 27, 2002.
1) Boess F. G., Hendrix M., van der Staay F.-J., Erb C., Schreiber R.,
38) Seto M., Banno Y., Imaeda T., Kajita Y., Ashizawa T., Kawasaki
M., Nakamura S., Mikami S., Nomura I., Taniguchi T., Marui S.,
WO 2015060368, April 30, 2015.
39) Shen D.-M., Egbertson M., Berger R., Qian X., Qian Y., Harper B.,
Yang M., Zack Z. Q., Rada V. L., Wang D., Cernak T. A., Sinz C. J.,
Wang M., Wilson J. E., Xu S., WO 2015096651, July 2, 2015.
40) Trabanco A. A., Buijnsters P., Rombouts F. J. R., Expert Opin. Ther.
Pat., 26, 933–946 (2016).
41) Gomez L., Massari M. E., Vickers T., Freestone G., Vernier W., Ly
K., Xu R., McCarrick M., Marrone T., Metz M., Yan Y. G., Yoder
Z. W., Lemus R., Broadbent N. J., Barido R., Warren N., Schmelzer
K., Neul D., Lee D., Andersen C. B., Sebring K., Aertgeerts K.,
Zhou X., Tabatabaei A., Peters M., Breitenbucher J. G., J. Med.
Chem., 60, 2037–2051 (2017).
2
1
van Staveren W., de Vente J., Prickaerts J., Blokland A., Koenig G.,
Neuropharmacology, 47, 1081–1092 (2004).
1
2) Rutten K., Van Donkelaar E. L., Ferrington L., Blokland A., Bollen
E., Steinbusch H. W. M., Kelly P. A. T., Prickaerts J. H. H. J., Neu-
ropsychopharmacology, 34, 1914–1925 (2009).
13) Bolger G. B., Rodgers L., Riggs M., Gene, 149, 237–244 (1994).
14) Van Staveren W. C., Steinbusch H. W., Markerink-Van Ittersum M.,
Repaske D. R., Goy M. F., Kotera J., Omori K., Beavo J. A., De
Vente J., J. Comp. Neurol., 467, 566–580 (2003).
15) Reyes-Irisarri E., Markerink-Van Ittersum M., Mengod G., de Vente
J., Eur. J. Neurosci., 25, 3332–3338 (2007).
42) Mikami S., Sasaki S., Asano Y., Ujikawa O., Fukumoto S., Na-
kashima K., Oki H., Kamiguchi N., Imada H., Iwashita H., Tanigu-
chi T., J. Med. Chem., 60, 7658–7676 (2017).
43) Mikami S., Nakamura S., Ashizawa T., Nomura I., Kawasaki M.,
Sasaki S., Oki H., Kokubo H., Hoffman I. D., Zou H., Uchiyama N.,
Nakashima K., Kamiguchi N., Imada H., Suzuki N., Iwashita H.,
Taniguchi T., J. Med. Chem., 60, 7677–7702 (2017).
1
6) Stephenson D. T., Coskran T. M., Wilhelms M. B., Adamowicz W.
O., O’Donnell M. M., Muravnick K. B., Menniti F. S., Kleiman R.
J., Morton D., J. Histochem. Cytochem., 57, 933–949 (2009).
1
7) Lakics V., Karran E. H., Boess F. G., Neuropharmacology, 59,
67–374 (2010).
8) Stephenson D. T., Coskran T. M., Kelly M. P., Kleiman R. J., Mor-
3
1
ton D., O’Neill S. M., Schmidt C. J., Weinberg R. J., Menniti F. S.,
Neuroscience, 226, 145–155 (2012).
44) Hitchcock S. A., Pennington L. D., J. Med. Chem., 49, 7559–7583
(2006).
1
2
9) Bender A. T., Beavo J. A., Pharmacol. Rev., 58, 488–520 (2006).
0) Frey U., Huang Y. Y., Kandel E. R., Science, 260, 1661–1664 (1993).
45) Hitchcock S. A., J. Med. Chem., 55, 4877–4895 (2012).
46) OECD, “OECD Guidelines for the Testing of Chemicals, Section
4: Health Effects (Test No. 432: in vitro 3T3 NRU Phototoxic-
ity Test). 2004”: ‹http://www.oecd-ilibrary.org/environment/test-no-
432-in-vitro-3t3-nru-phototoxicity-test_9789264071162-en;jsessionid=
3qtfu6logg1hf.x-oecd-live-02›, accessed August 9, 2017.
47) Stein K. R., Scheinfeld N. S., Expert Opin. Drug Saf., 6, 431–443
(2007).
21) Son H., Lu Y. F., Zhuo M., Arancio O., Kandel E. R., Hawkins R.
D., Learn. Mem., 5, 231–245 (1998).
2
2) Lu Y., Kandel E. R., Hawkins R. D., J. Neurosci., 19, 10250–10261
1999).
(
2
3) Prickaerts J., de Vente J., Honig W., Steinbusch H. W. M., Blokland
A., Eur. J. Pharmacol., 436, 83–87 (2002).
2
2
4) Sanderson T. M., Sher E., Neuropharmacology, 74, 86–95 (2013).
5) Abaarghaz M., Biondi S., Duranton J., Limanton E., Mondadori C.,
Wagner P., WO 2005063723, July 14, 2005.
6) Schmidt B., Weinbrenner S.,Flockerzi D., Kuelzer R., Tenor H.,
Kley H.-P., WO 2006024640, March 9, 2006.
48) Peukert S., Nunez J., He F., Dai M., Yusuff N., DiPesa A., Miller-
Moslin K., Karki R., Lagu B., Harwell C., Zhang Y., Bauer D.,
Kelleher J. F., Egan W., Med. Chem. Commun., 2, 973–976 (2011).
49) Wager T. T., Hou X., Verhoest P. R., Villalobos A., ACS Chem. Neu-
rosci., 1, 435–449 (2010).
50) Podzuweit T., Nennstiel P., Müller A., Cell. Signal., 7, 733–738
(1995).
51) Neill J. C., Barnes S., Cook S., Grayson B., Idris N. F., McLean S.
L., Snigdha S., Rajagopal L., Harte M. K., Pharmacol. Ther., 128,
419–432 (2010).
2
2
2
2
7) Schmidt B., Weinbrenner S., Flockerzi D., Külzer R., Tenor H.,
Kley H.-P., WO 2006072612, July 13, 2006.
8) Schmidt B., Weinbrenner S., Flockerzi D., Külzer R., Tenor H.,
Kley H.-P., WO 2006072615, July 13, 2006.
9) Buijnsters P., De Angelis M., Langlois X., Rombouts F. J. R.,
Sanderson W., Tresadern G., Ritchie A., Trabanco A. A., VanHoof
G., Van Roosbroeck Y., Andrés J.-I., ACS Med. Chem. Lett., 5,
52) van der Staay F. J., Rutten K., Erb C., Blokland A., Behav. Brain
Res., 220, 215–229 (2011).
1049–1053 (2014).
53) Ware E., Chem. Rev., 46, 403–470 (1950).
3
0) Rombouts F. J. R., Tresadern G., Buijnsters P., Langlois X., Tovar
F., Steinbrecher T. B., Vanhoof G., Somers M., Andrés J.-I., Tra-
banco A. A., ACS Med. Chem. Lett., 6, 282–286 (2015).
54) Yamamoto Y., Takizawa M., Yu X.-Q., Miyaura N., Heterocycles,
80, 359–368 (2010).
55) Harvey G. R., J. Org. Chem., 31, 1587–1590 (1966).
56) L’abbé G., Angew. Chem. Int. Ed. Engl., 14, 775–782 (1975).
57) Sugimoto H., Hirabayashi H., Kimura Y., Furuta A., Amano N.,
Moriwaki T., Drug Metab. Dispos., 39, 8–14 (2011).
31) Redrobe J. P., Jørgensen M., Christoffersen C. T., Montezinho L. P.,
Bastlund J. F., Carnerup M., Bundgaard C., Lerdrup L., Plath N.,
Psychopharmacology (Berl.), 231, 3151–3167 (2014).
3
2) Helal C. J., Identification of a Brain Penetrant, Highly Selective
Phosphodiesterase 2A Inhibitor for the Treatment of Cognitive
Impairment Associated with Schizophrenia (CIAS). 244th National
Meeting & Exposition, Philadelphia, PA, United States, August
58) Takeuchi T., Yoshitomi S., Higuchi T., Ikemoto K., Niwa S., Ebi-
hara T., Katoh M., Yokoi T., Asahi S., Pharm. Res., 23, 1460–1472
(2006).
59) Yamamoto K., Ikeda Y., J. Drug Deliv. Sci. Tec., 33, 13–18 (2016).
60) Hermans P. H., Weidinger A., J. Appl. Phys., 19, 491–506 (1948).
19–23, 2012.

Vol. 65, No. 11 (2017)
Chem. Pharm. Bull.
mifit›, June 12, 2010.
1077
61) Ikeda Y., Ban J., Ishikawa T., Hashiguchi S., Urayama S., Horibe
H., Chem. Pharm. Bull., 56, 1406–1411 (2008).
66) Zhang L., Villalobos A., Beck E. M., Bocan T., Chappie T. A., Chen
L., Grimwood S., Heck S. D., Helal C. J., Hou X., Humphrey J. M.,
Lu J., Skaddan M. B., McCarthy T. J., Verhoest P. R., Wager T. T.,
Zasadny K., J. Med. Chem., 56, 4568–4579 (2013).
67) Miyamoto M., Takahashi H., Kato K., Hirai K., Ishihara Y., Goto
G., J. Pharmacol. Exp. Ther., 277, 1292–1304 (1996).
6
6
6
2) Yamamura S., Momose Y., Int. J. Pharm., 212, 203–212 (2001).
3) Otwinowski Z., Minor W., Methods Enzymol., 276, 307–326 (1997).
4) Collaborative Computational Project, Acta Crystallogr. D Biol.
Crystallogr., 50, 760–763 (1994).
6
5) Smith B., Badger J., “MIFit, 2010. 10”; ‹https://github.com/mifit/