Scholarly article on tert-butyl 5-(4,4,5,5-tetraMethyl-1,3,2-dioxaborolan-2-yl)-3,4-dihydropyridine-1(2H)-carboxylate 1121057-77-9 from Journal of Medicinal Chemistry p. 12887

pubs.acs.org/jmc

Article

[1,2,4]Triazolo[1,5‑a]pyrimidine Phosphodiesterase 2A Inhibitors:

Structure and Free-Energy Perturbation-Guided Exploration

́

Gary Tresadern,* Ingrid Velter, Andres A. Trabanco, Frans Van den Keybus, Gregor J. Macdonald,

Marijke V. F. Somers, Greet Vanhoof, Philip M. Leonard, Marieke B. A. C. Lamers,

Yves E. M. Van Roosbroeck, and Peter J. J. A. Buijnsters *

Cite This: https://dx.doi.org/10.1021/acs.jmedchem.0c01272

Read Online

sı

*

Metrics & More

Article Recommendations

Supporting Information

ACCESS

ABSTRACT: We describe the hit-to-lead exploration of a

[1,2,4]triazolo[1,5-a]pyrimidine phosphodiesterase 2A (PDE2A)

inhibitor arising from high-throughput screening. X-ray crystallog-

raphy enabled structure-guided design, leading to the identiﬁcation

of preferred substructural components. Further rounds of

optimization used relative binding free-energy calculations to

prioritize diﬀerent substituents from the large accessible chemical

space. The free-energy perturbation (FEP) calculations were

performed for 265 putative PDE2A inhibitors, and 100 compounds

were synthesized representing a relatively large prospective

application providing unexpectedly active molecules with IC₅₀′s

from 2340 to 0.89 nM. Lead compound 46 originating from the

FEP calculations showed PDE2A inhibition IC₅₀of 1.3 0.39 nM,

∼100-fold selectivity versus other PDE enzymes, clean cytochrome P450 proﬁle, in vivo target occupancy, and promise for further

lead optimization.

nucleotide levels that play a key role in processes of

neuroplasticity such as long-term potentiation (LTP), which is

considered as a neurophysiological correlate of memory and

learning.¹⁰The exact underlying mechanism of memory

enhancement following PDE2A inhibition is not fully under-

stood, but it is likely occurring via the nitric oxide synthase

(NOS)/cGMP/protein kinase-G pathway rather than the

cAMP/protein kinase A pathway.^11,12

Cognitive deﬁcits are prominent in neurodegenerative

disorders such as Alzheimer’s disease (AD). The earliest

cognitive symptoms of AD include a decline in memory

function,^13,14manifest as problems with planning and task

execution (executive function),¹⁵described as an inability to

successfully form new memory traces for integration into

existing memory networks.^16,17While the impairment of

episodic memory is frequently the earliest detectable neuro-

cognitive sign, the earliest symptoms often cooccur with

neuropsychiatric behavioral symptoms.¹⁸Upon progression of

the disease, more cognitive function is lost¹⁹and patients

INTRODUCTION

■

The ubiquitous nucleotides 3′,5′-cyclic adenosine mono-

phosphate (cAMP) and 3′,5′-cyclic guanosine monophosphate

(cGMP) are second messengers that play a critical role in many

signal transduction pathways. Both cyclic nucleotides are

hydrolyzed by phosphodiesterase (PDE) enzymes to the

corresponding biologically inactive adenosine and guanosine

monophosphates, AMP and GMP, respectively. PDE enzymes

are encoded by 21 genes and are grouped into 12 families, PDE1

to PDE12.¹Their expression is tissue- and cell-speciﬁc;

moreover, the diﬀerent isozymes and splice variants are

expressed in speciﬁc cell compartments and targeted to

particular protein complexes, resulting in a spatiotemporal

integration of multiple signaling pathways.²

PDE2A is a homodimer of two monomers approximately 105

kDa each. The enzyme has two regulatory units for each

monomer, a GAF-A and a GAF-B domain that have distinct

roles in dimerization and activation, respectively. Binding of

cGMP to the GAF-B domain allosterically activates the enzyme

resulting in a 6- to 30-fold increase of cAMP hydrolysis.³⁻⁵Once

the enzyme is activated, it hydrolyzes both nucleotides with

similar K_Mvalues restoring the cAMP and cGMP concentrations

to basal levels. PDE2A is expressed in the brain and is most

prevalent in cortex, hippocampus, and striatum,⁶⁻⁸supporting a

role in cognitive processes, neuronal plasticity, and memory.⁹

Inhibition of PDE2A will result in elevation of intracellular cyclic

Received: July 21, 2020

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

A

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 1. Previous tricyclic PDE2A inhibitor leads 1 and 2, along with triazolopyrimidine HTS hit 3 which is the subject of this report.

develop concurrent neuropsychiatric symptoms. Disease-

modifying treatments in AD are yet to be established, and

currently available symptomatic treatments are only moderately

eﬀective or limited in their use because of their side eﬀects.

Therefore, improved treatments to alleviate or even reverse the

symptoms of reduced cognitive function represent a signiﬁcant

unmet medical need. Thus, PDE2A inhibition may be

procognitive and have a positive impact in the treatment of

cognitive dysfunction in early stage AD,^20,21and PDE2

inhibition has been the subject of recent medicinal chemistry

reports.²²⁻²⁶

for PDE2A and PDE10A. We began our experimental follow-up

by solving the X-ray crystallographic structure of 3 with the

PDE2A catalytic domain, reﬁned to 2.35 Å resolution (R_work

=

21.5% and R_free= 29.1%). There were two PDE2A molecules in

the asymmetric unit, but each active site overlaid well and

showed minimal variation in ligand binding. Molecule 3 sits in

the hydrophobic clamp between Phe830 and Phe862 (Figure 2).

We previously described the structure-based optimization of

tricyclic PDE2A inhibitors such as 1 and 2.^27,28Although highly

potent and selective, poor solubility and high plasma clearance

could not be overcome, limiting their developability. Here, we

report a series of [1,2,4]triazolo[1,5-a]pyrimidine PDE2A

inhibitors arising from high throughput screening (HTS). X-

ray crystallography of the initial hit 3 allowed structure-guided

design, leading to improved target potency. An X-ray structure

of a second example conﬁrmed the binding mode, and we

employed computational relative binding free-energy (RBFE)

calculations to guide the molecular design. The accurate

prediction of protein−ligand interaction energies has been,

and remains, an important but very challenging goal for

computational structure-based design.²⁹We applied the free-

energy perturbation (FEP) approach that calculates the binding

energy diﬀerence between structurally similar ligands, making it

well suited for lead optimization (LO). The method is

undergoing a resurgence of interest^30,31and we have seen a

good performance in drug discovery programs³²⁻³⁴including

PDE2A^28,35and for other large protein−ligand data sets.³⁶⁻³⁸

Here, FEP calculations on 250 triazolopyrimidines were

followed by the synthesis of 100 examples, giving a valuable

demonstration of FEP-guided LO. Overall, the resulting lead

compounds showed excellent PDE2A inhibition potency (IC₅₀

< 1 nM) and selectivity of up to 100- to 1000-fold versus other

PDE enzymes. Brain penetration was conﬁrmed for selected

leads and the most promising examples, 42 and 46, showed in

vivo target engagement in a PDE2A occupancy assay.

Figure 2. X-ray structure of HTS hit 3 with PDE2A viewed from the

entrance to the active site (A) and top down (B), PDB accession code

6ZND. Important amino acids are highlighted in green, and catalytic

Zn²⁺and Mg²⁺ions are shown, along with active site water molecules

(red spheres); hydrogen bonds between Gln812 and Gln859 and the

triazolopyrimidine scaﬀold are indicated with dashed yellow lines in

panel B.

Most active site interactions with the ligand are formed with the

triazolopyrimidine core. Gln812 donates a 2.3 Å hydrogen bond

from its NH side chain to the pyrimidine sp²nitrogen acceptor

of the bicycle. Another hydrogen bond with a distance of 2.5 Å is

donated from the NH side chain of Gln859 to the triazole sp²

nitrogen and Phe862 forms an oﬀ-center π-stack above the

scaﬀold. The linking amido piperidinyl group has an end-on π-

interaction with Phe830 and is largely solvent exposed along

with the pendant trimethoxyphenyl residue that is bound on the

solvent exposed surface of helix 15. The para methoxy group

forms a hydrogen bond with a water molecule. The X-ray

structure settled the uncertainty of whether the alkoxyphenyl in

3 may interact with the “switch” Gln859 as seen for well-known

PDE inhibitors such as Papaverine and Rolipram.^39,40Further

examination of the binding mode suggested that there was space

to substitute on the free positions of the pyrimido ring along

with ample opportunity for amide replacements.

The chemistry exploration of the initial hit began with the

preparation of the right-hand side (RHS, Chart 1) amides 3−18

(Table 1). The compounds were tested in a full-length human

PDE2A inhibition assay measuring changes in levels of cAMP by

a scintillation proximity approach. Consistent with the binding

mode, a variety of benzamides were tolerated as substituents on

the piperidine. Initial hit 3 was racemic and a 66 nM PDE2A

RESULTS AND DISCUSSION

■

HTS hit 3 (Figure 1) had a PDE2A inhibition IC₅₀of 66 3.2

nM and approximately eightfold selectivity versus PDE10A (the

closest structural PDE family member to PDE2A). It was

inactive (IC₅₀> 10,000 nM) in all cases when tested for

inhibition activity in a larger panel of PDE enzymes (PDE1B,

PDE3A, PDE4D, PDE5A, PDE7A, PDE9A, and PDE11A).

Compound 3 originated from commercial sources and was part

of our internal screening library. It was tested in over 100

internal screens and was inactive (IC₅₀> 10,000 nM) other than

B

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a b

,

Chart 1. Medicinal Chemistry Strategy for Exploration of

HTS Hit 3

Table 1. RHS Exploration of HTS Hit 3

inhibitor. The dimethoxy analogue 4 showed similar PDE2A

inhibition activity (IC₅₀= 47 16 nM). Homologation via a

methylene spacer between the amide carbonyl and the distal

aromatic resulted in a ∼10-fold drop in activity for 5 compared

to 4 (5, IC₅₀= 479 16 nM vs 4, IC₅₀= 47 16 nM). The X-ray

structure of 3 (Figure 2) suggests that this was due to a loss of

the optimal interactions with Met847 and Phe862. Simpliﬁca-

tion to the unsubstituted phenyl 6 (IC₅₀= 339 31 nM) or

replacement of the distal aromatic by small aliphatic groups, 7

(IC₅₀= 246 12 nM) and 8 (IC₅₀= 1580 180 nM), were

detrimental for activity, again because of the loss of preferred

aromatic interactions with Met847 and Phe862.

The PDE2A activity was dependent on the position and type

of substitution on the aromatic ring. For instance, meta and para

substituents were preferred, exempliﬁed with ortho methoxy 9

(IC₅₀= 912 200 nM) that was >12-fold less active than meta

10 (IC₅₀= 74 27 nM) or para 11 (IC₅₀= 50 22 nM). It was

expected that the ortho substituent caused the rotation of the

distal aromatic ring and disrupted its interaction with Phe862.

Meanwhile, it was clear that electron donating groups were

beneﬁcial for activity compared to withdrawing, for instance, the

comparison of para methoxy 11 (IC₅₀= 50 22 nM) to para

cyano 12 (IC₅₀= 589 280 nM) and triﬂuoromethyl 13 (IC₅₀=

324 96 nM) substituents. The electron-rich aromatic rings

were preferred as they form an energetically more favorable T-

shaped π-interaction with Phe862, Figure 2. Conformational

restriction of the methoxy groups into the benzodioxane

derivative 14 (IC₅₀= 49 21 nM) maintained a similar range

of activity compared to 4 (IC₅₀= 47 16 nM), although 4 was

racemic. Fully aromatic bicyclic examples such as benzofuran 15

improved the inhibition activity to 24 8.5 nM. This prompted

us to expand the exploration around these bicyclic hetero-

aromatics, resulting in analogues such as 16 and 17 that reached

PDE2A potencies of 2.1 0.9 and 2.6 1.1 nM, respectively.

The tetrahydrobenzothiophene group attached via the ﬁve-

membered ring in 18 was a 10 5.2 nM PDE2A inhibitor and

was selected as the template to explore central core and LHS

variations. Finally, the observation of the S stereoisomer binding

in the X-ray structure was formally conﬁrmed by chiral

separation of 17. Indeed, the S stereoisomer retains most of

the PDE2A inhibitory activity with 20 being 60-fold more active

than its enantiomer 19 (20, IC₅₀= 1.5 0.1 nM vs 19, IC₅₀= 87

27 nM).

a

Primary inhibition activity versus PDE2A, and activity versus close

b

structural family members PDE3A/B and PDE10A. PDE2A

bioactivity measurements are the mean of at least two independent

repeats and provided with standard deviations except 6, 8, and 12,

which are provided with the standard error on dose response for n =

1. All data were generated with PDE3A except compounds 8, 9, and

15 that were tested versus PDE3B.

The identiﬁcation of a selective inhibitor was a key aspect of

our strategy to later validate the role of PDE2A inhibition for

cognition. The main concern was selectivity versus PDE10A

given its close structural similarity to PDE2A (34% sequence

identity, 50% similarity for the catalytic domain) and its

implication in other central nervous system disorders.⁴¹Indeed,

the selectivity data available for all the compounds in this article

conﬁrms only weak and sparse activity (range 3160−10,000 nM

c

for 12 out of 140 IC₅₀measurements) versus multiple other PDE

family members (PDE1B, PDE4D, PDE5A, PDE7A, PDE9A,

C

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

and PDE11A), Table S1. The HTS hit 3 showed a modest 8-fold

selectivity, but this was substantially improved to approximately

30- to 50-fold with bicyclic amides such as 16−18. This could be

rationalized because of the interaction with Leu858 (Figure 2),

which forms the boundary at the bottom of the benzamide

pocket and interacts directly with the meta and para substituents

of the distal aromatic ring, or in the case of 16−18, the fused 5-

membered ring. A glycine is found at the same position in

PDE10A (Gly725), and because of the absence of the side chain,

the pocket extends further, meaning that this region of the

binding site, traditionally referred to as the Q2 pocket,⁴⁰can be

exploited for PDE selectivity.⁴²

constant presence of the tetrahydrobenzothiophene amide. In

contrast, 18 and 21 displayed only 1.4- to 4-fold selectivity

versus PDE3A. The sequence similarity between PDE2A and

PDE3A (20% sequence identity, 33% similarity for the catalytic

domain) was lower than that for PDE10A, and the loss of

selectivity versus PDE3A was initially a surprise. Nevertheless,

the complete lack of PDE3A inhibition activity for some

benzamide replacements (3, 7, and 19 Table 1) oﬀered promise

that PDE3A selectivity could be modulated in the same Q2

pocket that was targeted for PDE10A selectivity.⁴⁰Comparison

of PDE2A and PDE3B⁴³crystal structures revealed diﬀerences,

for instance, Met847 (PDE2A, Figure 1) was instead Phe976

(PDE3B) that also adopts a diﬀerent conformation and changes

the shape of the pocket” to “for instance, Met847 in PDE2A

(Figure 1) was instead Phe976 in PDE3B. The Phe976 also

adopts a diﬀerent conformation compared to Met847 and

changes the shape of the pocket.

Turning to the core modiﬁcations (Table 2), replacement of

the piperidine in 18 by a morpholine was detrimental for activity

a b

,

Table 2. Central Core Modiﬁcations

To complete the initial structure−activity relationship (SAR)

exploration around the triazolopyrimidine hit series, the LHS

group (Chart 1) was modiﬁed while maintaining the piperidine

central core and RHS residue present in 18. The introduction of

a methyl substituent on the pyrimidine ring brought immediate

improvements in potency and compound 26 showed a 3.5 2.3

nM PDE2A inhibitor IC₅₀activity (Table 3). According to the

crystal structure binding mode of 3, the methyl group in 26

would occupy a small pocket deep in the binding site. The larger

isopropyl group in 27 would not ﬁt and hence the compound

was much less active (27, IC₅₀= 6031 1943 nM). The smaller

methoxy group in the same position recovered the activity (28,

IC₅₀= 195 85 nM). Triazolopyrimidine compounds bearing

electron-withdrawing cyano (29) and triﬂuoromethyl (30)

substituents showed good PDE2A inhibition values (10 5.9

and 50 19 nM, respectively), although they were less potent

than the methyl analogue 26. Interestingly, the CHF₂group in

31 delivered an equipotent PDE2A inhibitor (31, 3.4 2.3 nM

vs 26, 3.5 2.3 nM). Modiﬁcation of the heterobicyclic ring

(32−34) resulted in all cases in a substantial loss of activity. The

IC₅₀of pyrazolopyrimidine 32, imidazolopyrimidine 33, and

azaindole 34 were 30 22, 407 66, and 6020 1980 nM,

respectively. Finally, the introduction of substituents in the

triazolo ring in 26 was clearly detrimental for activity with

a

Primary inhibition activity versus PDE2A, and activity versus close

b

structural family members PDE3A/B and PDE10A. PDE2A

bioactivity measurements are the mean of at least two independent

repeats and provided with standard deviations except 22 and 25

which are provided with the standard error on dose response for n =

1. All data were generated with PDE3A except compound 23 that

was tested versus PDE3B.

methyl- (35, IC₅₀= 562

65 nM) or triﬂuoromethyl-

substituted (36, IC₅₀= 2951 478 nM) analogues 160- and

842-fold less active than the PDE2A inhibitor 26. In general,

selectivity improved by the introduction of the small substituent

on the pyrimidine ring, for instance 31 was nearly 200-fold and

16-fold selective versus PDE10A and PDE3A, respectively.

Compound 31 was tested in a larger PDE inhibition panel

including PDE1B, PDE4D, PDE5A, PDE7A, and PDE11A,

where it showed >10,000 nM activity in all assays except for

PDE4D, where a moderate activity (IC₅₀= 6460 nM) was

measured.

From the initial SAR studies, it was clear that although the

RHS substituent tolerated a diverse set of aromatic amides, the

substitution pattern and electronic nature of the substituents

were important for achieving increased PDE2A inhibition. This

was due to forming optimal nearby interactions, particularly with

Met847 and Phe862. All modiﬁcations of the central scaﬀold

were less active than the parent piperidine example. Meanwhile,

the LHS group that hydrogen bonds with the Gln812 switch and

Gln859, permitted a CH₃and CHF₂substituent entering deep

into the binding pocket. Therefore, using the optimal LHS along

with the piperidine central ring, benzylic heteroaromatic groups

c

with 21 being ninefold less active than 18 (21, IC₅₀91 29 vs

18, 10 5.2 nM). Although the unsubstituted piperazine was

substantially less active (22, IC₅₀= 1412 196 nM), the N-

methyl analogue 23 showed a double-digit nanomolar PDE2A

inhibition activity (IC₅₀= 87

41 nM). Acylation of the

piperazine nitrogen in 24 was inactive up to the 10 μM

concentration limit. Ring enlargement was also detrimental for

activity with the homopiperidine 25 being 26-fold less active

than the corresponding piperidine matched pair (25, IC₅₀= 263

24 nM vs 18, IC₅₀= 10 5.2 nM). Hence, the piperidine was

optimal for the activity, something that could be understood

from the X-ray structure because of its close lipophilic

interaction with the side chains of Tyr655 and Ile826. The

selectivity of the most active examples in this exploration, 21 and

23, was similar to the parent piperidine central core, 18. Indeed,

all three examples maintained the same 28-fold selectivity versus

PDE10A, demonstrating the origin of selectivity from the

D

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a b

,

Table 3. LHS Exploration of HTS Hit 3

a

b

Primary inhibition activity versus PDE2A, and activity versus close structural family members PDE3A/B and PDE10A. PDE2A bioactivity

measurements are the mean of at least two independent repeats and provided with standard deviations except for 33, 35, and 36, which are

c

provided with the standard error on dose response for n = 1. All data were generated with PDE3A except for compounds 28, 29, 32, and 33 that

were tested versus PDE3B.

E

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

were studied as RHS benzamide replacements (Table 4). The

SAR obtained did not correlate well with the phenyl amide

a b

,

Table 4. Second Round RHS Modiﬁcations

Figure 3. X-ray structure of lead 39 with PDE2A viewed from the

entrance to the active site (A) and top down (B), PDB accession code

6ZQZ. Important amino acids are highlighted in green, and catalytic

Zn²⁺and Mg²⁺ions are shown, along with active site water molecules

(red spheres); hydrogen bonds are indicated with dashed yellow lines in

panel B.

with the phenol oxygen of the neighboring Tyr655. The

piperidine core is largely solvent exposed along with the pendant

pyridine ring. Positive electron density peaks in the ﬁnal

diﬀerence density maps indicated a second alternative

conformation of the pyridine ring related by a 180° rotation of

the pyridine, positioning the chlorine substituent toward the

solvent space. B-Factor reﬁnement indicated less than 20%

occupancy of this second conformation in each of the four

subunits. The alternative orientation was therefore not included

in the ﬁnal model. Nevertheless, the second possible orientation

was consistent with conformational eﬀects contributing to lower

inhibition activity of benzyl- compared to benzamide-

substituents (Table 4). The pyridyl ring was again sandwiched

between Met847 and Phe862, and the nitrogen formed a

hydrogen bond to a crystallographic water at the bottom of the

Q2 pocket.

Racemization of the Central Scaﬀold. Up to this point,

we had several promising leads, including 15−18, 26, and 31

that showed PDE2A inhibition <10 nM and in most cases >10-

fold selectivity versus PDE3A and PDE10A. Unfortunately, it

was noticed that after chiral separation, in some cases,

enantiopure compounds in solution turned into mixtures after

several days. Racemization presented a challenge for the

progression of the series. Interestingly, the problem was

overcome by introducing a methyl substituent into the 4-

position of the piperidine ring. Several examples were

synthesized and shown not only to maintain but even improve

potency (Figure 4). This was likely because of the preference for

hydrophobic interactions with the side chains of Tyr655 and

Ile826, as seen for the central core modiﬁcations (Table 2).

a

Primary inhibition activity versus PDE2A, and activity versus close

b

structural family members PDE3B and PDE10A. PDE2A bioactivity

measurements are the mean of at least two independent repeats and

provided with standard deviations.

subseries. First, both the dimethoxyphenyl- (37) and the

benzofuran-containing (38) triazolopyrimidine analogues were

less active than the equivalent amides 4 and 15 (37, IC₅₀= 741

345 nM vs 4, IC₅₀= 47 16 nM and 38, IC₅₀= 1148 207 nM

vs 15, IC₅₀= 24

8.5 nM, respectively). Furthermore, the

relative order of activity was reversed. Interestingly, the benzyl-

like pyridine derivatives 39 and 40 were 24 6.3 and 69 39

nM PDE2A inhibitors and recovered selectivity, being >120-fold

selective for both PDE3A and PDE10A. This conﬁrmed the

comparison between PDE2A and PDE3B X-ray structures that

revealed that the sequence and structural diﬀerences in the Q2

pocket could lead to PDE3A selectivity. It was expected that

these electron-rich heteroaromatics would make favorable

interactions with Phe862 as mentioned above, but the

diﬀerences in activity suggested that diﬀerent ligand con-

formations were possible compared to equivalent amides.

Therefore, intrigued to understand if this would translate into

structural diﬀerences when bound to the enzyme, we attempted

and solved an X-ray crystal structure of 39 with the PDE2A

catalytic domain.

Compound 41 showed high PDE2A inhibition activity (IC₅₀

1.1 0.4 nM) and >90-fold selectivity versus PDE3A (IC₅₀

=

The structure of PDE2A in complex with lead 39 was solved at

1.9 Å resolution (R_work= 19.6% and R_free= 24.4%), and the

binding site is shown in Figure 3. There were four PDE2A

monomers per asymmetric unit and each had a single 39

molecule bound in its active site. Structural overlay indicated

little conformational diﬀerence between the four subunits; the

electron density was best deﬁned for subunit D. The

triazolopyrimidine core of 39 was again anchored by two

hydrogen bonds to Gln859 and Gln812 while being held in a

hydrophobic clamp by Ile826, Phe830 and Phe862. The diﬂuoro

methyl group sits in a small, well-buried pocket, as expected from

the X-ray structure of 3. The proton of the diﬂuoro methyl group

is electropositive and makes a favorable electrostatic interaction

2240 nM) and PDE10A (IC₅₀= 97 nM). It was also tested for

inhibition against a larger panel of PDE enzymes and was

inactive (IC₅₀> 10,000 nM) in all cases (PDE1B, PDE4D,

PDE7A, and PDE11A) except for PDE5A, where a weak activity

was detected (IC₅₀= 6760 nM). The chiral example 42 showed

sub-nanomolar PDE2A inhibition (IC₅₀= 0.6 0.3 nM) and

was 2200- and 90-fold selective versus PDE3A and PDE10A,

respectively.

RBFE Calculations. Up to this point, we had mostly used

amide substitution on the piperidine nitrogen to explore the hit

series. However, molecules with benzylic functionality such as

39 and 40 demonstrated that a large chemically accessible space

may be possible. Hence, we turned to computational

F

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 4. 4-Methyl-substituted piperidines 41−43, introduction of the methyl group at the 4-position of the piperidine central scaﬀold to prevent

racemization.

prioritization using RBFE calculations with FEP methodology.

The method requires that perturbed molecules share an

overlapping binding mode during the computational simula-

tions. This was expected here as the two crystallographic binding

orientations of 3 and 39 were highly similar, suggesting that

diﬀerent chemical decoration can be probed with FEP

calculations. Others have used binding free-energy calculations

for PDE10⁴⁴and for other PDE enzymes.⁴⁵We have described

accurate RBFE predictions for ligand modiﬁcations in a diﬀerent

subpocket of the PDE2A binding site.³⁵Here, the ligand changes

are relatively solvent exposed and that was expected to help the

performance of the calculations by avoiding issues with

kinetically trapped solvent.

We began by testing the FEP performance with a small

retrospective validation study using 20 existing molecules

involving a mixture of structurally similar amides and benzyls

covering a 4 log unit range of activity. A good correlation, R²

0.69, and mean unsigned error (MUE) of 0.89 0.35 kcal/mol

was seen when comparing experimental and calculated binding

aﬃnities, Figure S1 and Table S2. This prompted us to continue

and perform successive rounds of FEP to help prioritize on-

going chemistry explorations on a roughly monthly basis. The

calculations were applied to diﬀerent virtual libraries containing

varied substituents on the piperidine nitrogen. The input for the

FEP calculations was guided using docking that performed well

at reproducing the expected binding modes, Figure S2. The

reagents for each virtual library were selected based on their

similarity to the known actives, that is, aromatic amides, and

their chemical feasibility and availability. These included amides,

benzylic groups, and heteroaromatic halide ((het)aryl) reagents

(Table 5). This study was part of a larger in-house

evaluation^28,32,33,35of FEP, and therefore, we chose compounds

from across the range of predicted aﬃnity to avoid issues of

selection bias and deliver a fair evaluation of the FEP

performance.⁴⁶

Table 5. Summary of the Rounds of FEP Calculations That

Were Applied to Prospective Molecular Design

no

MUE

(kcal/mol)

a

round R-substituents

calculated synthesized

R²

1

amides & (het)

aryls

59

5

0.77

1.02 0.90

2

3

4

5

(het)aryls

benzyls

amides

57

58

39

52

29

10

27

0.51

0.28

0.41

0.58

0.78 0.27

1.01 0.33

0.47 0.32

0.62 0.29

pyridyls

a

99% conﬁdence interval. (Het)arylsheteroaromatic halide

reagents delivering compounds with a heteroaromatic ring directly

bonded to the piperidine nitrogen.

In total, 100 molecules were synthesized from the diﬀerent

libraries. The experimental and predicted binding free energies

(δG, kcal/mol) are compared in Figure 5. The most active

molecules are those in the bottom left of the plot with the largest

negative favorable binding free energy. The correlation of R²

0.57 and overall MUE of 0.79 0.16 kcal/mol were consistent

with the retrospective validation. A key reason for exploring

quantitative RBFE predictions is the poor activity prediction for

congeneric series by methods such as docking. This was also the

case here because the docking, although able to reproduce the

binding pose, was unable to predict binding aﬃnity for the same

set of compounds, R²0.01 and MUE of 2.27 0.33 kcal/mol,

Figure S2.

Figure 5. Correlation of experimental and calculated binding activity

(∂G) for the compounds synthesized in each round of the FEP-guided

design. The compound data points are colored and shaped according to

the rounds of FEP and synthesis, see Table 5. The diagonal lines of

unity and 1 or 2 kcal/mol errors are shown. The trendline (black,

partly obscured by the line of unity), equation, and R²were calculated

based on all the data combined, showing almost linear slope and very

little oﬀset. Selected compounds are labeled.

amides as highly active. An analogue of 42 where the benzofuran

was replaced by indole (changing only the oxygen atom for NH)

was a high aﬃnity 0.9 0.4 nM PDE2A inhibitor but structurally

There were various expected and unexpected examples

amongst the FEP results. The calculations predicted multiple

G

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 6. Comparative examples of 4-substituted pyridine RHS substituents arising from the FEP calculations.

very close to previous examples. Therefore, although this was an

interesting example of successful activity prediction, that is itself

a diﬃcult achievement, it is a relatively obvious result. However,

amongst a series of 4-substituted pyridyls, 44−46, example 46

with the 4-phenyl substituted pyridine was a surprising 1.3 0.4

nM PDE2A inhibitor that showed ∼100-fold increase in activity

compared to analogues 44 and 45 (Figure 6). The phenyl ring in

the 4-pyridyl position would have been disregarded based on the

relatively low activity of 44 that presents an amide with an

aromatic character in this position and 45 with a lipophilic t-

butyl group. Curiously, the FEP calculation for this molecule was

performed twice in two separate libraries of compounds, as seen

by the almost overlapping star and square labels highlighted in

Figure 5. It was part of a general library of heteroaromatic halide

reagents (round 2), and again in the more focused halo-pyridyl

reagent library (round 5). In both cases, it was predicted to be a

potent example that was conﬁrmed after synthesis and testing

and without the calculations this example would have been

overlooked.

The diﬀerence in activity between 45 and 46 was correctly

predicted by the FEP calculations, predicted to be 4.0 kcal/mol

in favor of 46 compared to 2.9 kcal/mol experimentally. In other

words, the calculations were capturing the multiple SAR eﬀects

that include the aromatic interaction with Phe862, the CH−pi

interactions between the 4-phenyl substituent of the ligand and

Ile866, and displacement of the crystallographic water at the

bottom of the Q2 pocket from a relatively hydrophobic and

exposed region of the protein surface (Figures 2 and 3 for

binding site details). This is manifest in the mostly good

correlation seen for the diﬀerent FEP rounds. Overall, relatively

substantial improvements in potency were seen for a pocket that

while solvent exposed contains multiple hydrophobic side

chains, Met847, Leu858, Phe862, and Ile866. This makes it an

ideal pocket for improving binding aﬃnity and also avoids

diﬃculties for FEP calculations, such as trapped solvent in

buried pockets or needing to introduce polar interactions with

optimal distance and angles to compete with the solvent.

Altogether these factors may explain the good performance of

FEP in this case.

due to a change in the ionization state of the piperidine nitrogen,

which was treated as neutral in all cases but may vary based on

the benzylic substituent. The correlation of the amide library in

round 4 was low, but this was due to a smaller range of the

resulting experimental activity (1.3 log units), making it diﬃcult

to achieve high correlation but the MUE remained good.

Overall, the improvement with each round was attributed to

the closer structural similarity between the compounds

(beneﬁcial for the convergence of the calculations) because

each library became more focused on analogues of previously

identiﬁed actives. We did not include additional experimental

reference molecules in the subsequent rounds, although this can

be used as a strategy to improve results. In our experience, this

application of FEP has been particularly successful for various

reasons. On the one hand, the SAR is relatively easy to capture

with important hydrophobic interactions in a solvent accessible

region, where water displacement can be modeled. The site of

perturbation is exposed, does not require protein movement,

and is far away from the active site metal ions which can cause

diﬃculties, all in contrast to our previous FEP studies on

PDE2.³⁵Practical issues also make this system amenable, the

high-resolution crystal structures were convergent on a common

binding mode for diﬀerent substituents, the substituents

themselves are distal and often unlikely to disturb the

conformation of the remaining molecule, there is no expected

change in formal charge of the ligands, and all perturbations took

place in only one region of the molecule. Thus, many reasons

combine to make this an ideal system for studying with RBFE

calculations.⁴⁷

Finally for the calculations, it is worthwhile to consider the

value and impact that accurate binding energy predictions can

have on drug discovery. An analysis of 50,000 medicinal

chemistry transformations from 30 projects revealed only 8.5%

having a potency gain of >1 log unit (∼1.4 kcal/mol).⁴⁸For a

small set of 10 compounds, a prediction with 1 kcal/mol of error

increases the chance a log unit improvement in activity to 36%, a

∼4-fold gain.⁴⁷As more compounds are assessed, the impact

improves. This simple model is informative but has various

caveats, for instance not all transformations are intended to

improve potency. More recent analysis for a prototypical drug

discovery program has suggested a 2- to 4-fold increase in the

number of potent molecules when assessing at least ﬁve ideas for

each synthesized compound.⁴⁹In this study, we synthesized 100

molecules from across a range of activity prediction. FEP

correctly identiﬁed 90% (9 out of 10) of the 10 most active

amongst its best ranked 10, compared to 20% (2 out of 10) from

docking. Although this represents an unexpectedly good

performance, it highlights the beneﬁts for increased computa-

tional assistance in molecular design and the value of RBFE

predictions compared to classical approaches.

The error between experimental and predicted binding

aﬃnity of the successive rounds of FEP improved, for instance,

rounds 4 and 5 showed MUE of 0.47 0.32 and 0.62 0.29

kcal/mol, respectively, compared to the ﬁrst rounds with MUE

∼1 kcal/mol. The ﬁrst round included molecules with two

diﬀerent types of substitution on the piperidine nitrogen, all the

subsequent rounds only used one type of chemical substituent.

The benzyl library, round 3, performed the worst, R²0.28 and

MUE 1.01 0.33 kcal/mol.

The lower correlation could be due to alternative binding

conformations, although the low populated conformation in the

X-ray structure of 39 was seen in the simulations, or it could be

H

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Table 6. Emerging Leads from the Initial Exploration of HTS Triazolopyrimidine Hit 3

d

PDE2A IC₅₀

(nM)

solubility

(μM pH 7.4)

rat Cl_int

hum Cl_int

rat t_1/2

hum t_1/2

(min)

b

c

compd

selectivity

CYP450

(μL/min/mg)

(min)

15

17

18

31

39

42

43

46

24 8.5

2.6 1.1

10 5.2

3.4 2.3

24 6.3

0.6 0.3

1.7 2.0

1.3 0.4

21

30

4

11

234

89

n.d.

105

102

92

n.d.

>100

4

n.d.

0, -, 2, 9, 0, 0

69, -, 44, 16, 0, 51

14

35

>347

74

11

26

157

19

32

>347

49

14

38

98

40

<4

19

129

55

9

72

43

<4

28

98

99

37

28, -, 31, 7, 43, 41

n.d.

112

99

>10, >10, 9, >10, >10,

>10

a

b

c

n.d. not determined. Minimum fold-selectivity calculated with respect to either PDE3A or PDE10A. Cytochrome P450 inhibition, all

compounds measured as % inhibition at 10 μM compound concentration, except 46, which corresponds to IC₅₀in μM. All measurements follow

d

the same order: 3A4, 2C8, 2C9, 2C19, 2D6, 1A2. All compounds incubated at 1 μM for 60 min in human and rat liver microsomes to determine

the metabolic turnover (half-life, Cl_int).

The initial systematic exploration of the HTS hit 3, followed

by the computationally guided library exploration, led to a

collection of early lead candidates with PDE2A inhibition

activity in the range of 0.6−24 nM (Table 6). Selectivity was

improved during the optimization with later examples such as

39, 42, 43, and 46 showing approximately 100-fold selectivity

versus either PDE3A or PDE10A. This set of promising

compounds was further proﬁled in vitro for its ADMET

properties. In general, solubility was good and no relevant

cytochrome P450 (CYP450) inhibition was seen for most

compounds up to 10 μM. The metabolic stability was an issue

for the leads that contained the tetrahydrobenzothiophene

amide such as 18 and 31 with both showing high turnover in

human and rat liver microsomal stability assays. Meanwhile,

examples such as the benzofuran-containing lead 42 suggested

the high metabolic turnover could be overcome. This compared

favorably with the previous tricycle 1 that was highly

metabolized (compound was undetectable after 15 min of

incubation in human and rat liver microsomes) in similar

assays.²⁷

shown in Table 7. On the basis of these results, compound 46

showed promise with the best brain to plasma ratio, but further

optimization would be needed to resolve metabolic clearance

and increase absolute levels.

An in vivo target occupancy assay was used to evaluate PDE2

target engagement of compounds 42 and 46 in rat brain slices. A

tritiated version of compound 1^27,28was used as a radioactive

tracer (10 μCi i.v.). In this experiment, MP-10⁴²(2.5 mg/kg

s.c.), a potent and selective PDE10 inhibitor, was predosed

because it was found to greatly increase the signal-to-noise ratio

of the tracer.²⁷This is understood to be due to PDE10

inhibition, leading to increased intracellular cGMP that is known

to activate PDE2 through its GAF domain. Using this protocol,

42 showed 92% occupancy of PDE2 after p.o. administration

(10 mg/kg suspension in 20% HP-β-CD solution). Compound

46 showed 90% PDE2 occupancy after s.c. administration (10

mg/kg suspension in 20% HP-β-CD solution).

Chemistry. A general retrosynthesis toward targeted amide

compounds⁵⁰is outlined in Scheme 1. The ﬁnal compounds 47

were synthesized via amide coupling between the cyclic amines

48 and readily accessible or commercially available (hetero)aryl

carboxylic acids 49. In turn, triazolopyrimidine-amines 48 were

obtained from the acetylamines 50 or from the piperidin-3-one

51 via known procedures.

We investigated the ability of a representative set of

compounds to cross the blood−brain barrier (BBB). Plasma

and brain levels measured 1 h after dosing at 10 mg/kg are listed

in Table 7. The tetrahydrobenzothiophene amides (18 and 31)

5-Unsubstituted triazolo[1,5-a]pyrimidine amines 48a−d

were prepared, as shown in Scheme 2. Thus, condensation of

the Boc-protected acetylamines 50a−d with N,N-dimethylfor-

mamide dimethylacetal (DMF/DMA) provided the intermedi-

ate 3-dimethylaminobuten-2-ones 52a−d, which were then

cyclized with 1H-1,2,4-triazol-3-amine (53) in acetic acid at 100

°C, to provide triazolopyrimidines 54a−d. Cleavage of the Boc-

protecting group in 54a−d gave the desired intermediates 48a−

d. Optionally, intermediates 48a−b were also separated in their

enantiomers via supercritical ﬂuid chromatography (SFC)

puriﬁcation to provide intermediates 48aa (R)−48ab (S) and

48ba (R)−48bb (S), respectively.⁵¹

The synthesis of 5-substituted triazolopyrimidine amines

48e−f commenced with the condensation of 3-acetylpiperidine

50a with carbon disulﬁde, followed by S-methylation using

methyl iodide and sodium hydride, which gave the bismethyl-

sulfanyl butenone intermediate 55 (Scheme 3). Cyclization of

55 with the aminotriazole 53 provided the 5-thiomethyl-

triazolopirimidine 56. Oxidation of the thiomethyl substituent

in 56 with mCPBA gave the corresponding sulfone 57, which

was then converted via aromatic nucleophilic substitution with

sodium methoxide and sodium cyanide to the respective

Table 7. Brain and Plasma Levels for Compounds after 1 h of

a Single Dose at 10 mg/kg in Rats

plasma levels

a

brain levels

a

B/P

ratio

b

compd

(ng/mL)

c log P TPSA

c

18

679 28

831 450

1640 57

338 68

240

346 260

788

596 75

1

0.4

0.5

1.8

1.9

2.1

1.8

3.4

92

77

59

c

31

d

42

4

c

46

a

Data are expressed as the mean average of at least two runs and

b

provided with standard deviations. c log P calculated with Biobyte

software. Compound dosed s.c. at 1 mg/mL in 20% HP-β-CD

solution in Sprague Dawley rat. Compound dosed p.o. at 1 mg/mL

c

d

in a 20% HP-β-CD solution pH 3.6−3.7 in the Long−Evans rat.

and the chiral benzofuran (42) were less brain penetrant (B/P =

0.4−0.5) than the 4-phenyl substituted pyridine (46) (B/P =

1.8). The better in vitro clearance seen for 42 permitted p.o.

dosing with higher absolute levels in plasma and brain.

Meanwhile, the better B/P ratio of 46 may be explained by its

lower total polar surface area (TPSA) and higher lipophilicity, as

I

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Scheme 1. General Retrosynthetic Route

a

Scheme 2. Synthesis of Intermediates 48a−d

a

Reagents and conditions: (a) DMF/DMA, 120 °C, 3 h; (b) AcOH, 100 °C, 1 h, 28−62% over 2 steps; (c) HCl, 1,4-dioxane, RT, 1 h, 70−93%.

a

Scheme 3. Synthesis of Intermediates 48e−f

Scheme 4. Synthesis of Intermediates 48g−i

a

Reagents and conditions: (a) NaH, CS₂, CH₃I; THF, 0 °CRT, 16

h, 21%; (b) 150 °C, 3 h, 12%; (c) mCPBA, CHCl₃, 60 °C, 2 h, 99%;

(d) R = OMe: NaOMe, MeOH, RT, 30 min; (e) R = CN: NaCN,

DMSO, RT, 30 min, 43%; (f) TFA, DCM, RT, 2 h; R = OMe: 99%, R

= CN: 68%.

a

Reagents and conditions: (a) 120 °C, 3 h, 95%; (b) AcOH, 120 °C,

30 min, for 53: 80%, for 2-aminoimidazole: 55%; (c) C₆H₆, KOtBu, 0

°CRT, 5 h, 98%; (d) (i) 53, DMF, 60 °C, 28 h; (ii) Et₃N, di-tert-

butyl dicarbonate, 0 °CRT, 1 h, 86%; (e) HCl, iPrOH, MeOH, 1−

24 h, RT (48g, 55%; 48h, 95%; 48i, 61%).

methoxy (48e) and cyano (48f) triazolopyrimidine intermedi-

ate amines after Boc deprotection.

The synthetic sequence for the preparation of intermediates

48g−h is shown in Scheme 4 and was similar to Scheme 1 for the

synthesis of intermediates 48a−d. Condensation of Boc-

protected 3-acetylpiperidine 50a with N,N-dimethylacetamide

dimethyl acetal aﬀorded the enaminone 58, which was then

reacted with either 1H-imidazol-2-amine or with 53 to provide

the corresponding 5- or 7-methyl-substituted [1,2,4]triazolo-

[1,5-a]pyrimidine (48g) or imidazo[1,2-a]pyrimidine (48h)

intermediate amines. The 5-(diﬂuoromethyl)-7-(3-piperidyl)-

[1,2,4]triazolo[1,5-a]pyrimidine amine derivative 48i was

obtained in three steps starting from 50a. First, the condensation

of 50a with ethyl diﬂuoroacetate aﬀorded the 1,3-dicarbonyl

intermediate 59 in excellent yield. Then, reaction of 59 with 53

led to intermediate 48i after Boc cleavage.

51 was treated with LDA in the presence of N-phenyl-

bis(triﬂuoromethanesulfonimide), aﬀording a mixture of triﬂate

enol ethers 60 and 61 that could be separated by column

chromatography on a silica gel. Both compounds were

converted to the corresponding boronate esters 62 and 63 in

moderate yields via palladium-catalyzed reaction with bis-

(pinacolato)diboron.

Boronic ester 62 was used in Suzuki cross-coupling reactions

with chloro-triazolopyrimidines 64 and tetrakis-

(triphenylphosphine)palladium(0) to give tetrahydropyridines

65a−d in modest yields because of the instability of the chloro-

heteroaromatic derivatives 64, which hydrolyzed in the reaction

conditions. The double bond in derivatives 65a−d was

satisfactorily reduced under catalytic hydrogenation conditions,

leading to the corresponding N-Boc-protected piperidines using

For the preparation of the piperidine intermediates 48j−o,

the boronic ester intermediates 62 and 63 were prepared from

Boc-protected piperidine3-one 51, as shown in Scheme 5. First,

J

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Scheme 5. Synthesis of 3,4- and 3,6-Dihydropyridyl Pinacol Boronic Ester Intermediates

a

Reagents and conditions: (a) LDA, THF, −78 °CRT, 16 h; (b) AcOK, Pin₂B₂, Pd(OAc)₂, PCy₃, 2-MeTHF, 80 °C, 16 h.

a

Scheme 6. Synthesis of Intermediates 48j−48o

a

Reagents and conditions: (a) Pd(PPh₃)₄, Na₂CO₃, 1,4-dioxane, EtOH, water, (9−27%). (b) H₂, Pd/C, MeOH, (R¹= iPr, R²= H, 99%) or Pt/C

(47%−88%); (c) HCl, iPrOH. MeOH, RT, 16 h (94−100%); (d) (SIPr)Pd(allyl)Cl, Na₂CO₃, 1,4-dioxane, EtOH, water (18%); (e) H₂, Pt/C,

MeOH (99%); (f) HCl, iPrOH. MeOH, RT, 16 h,91%; (h) (SIPr)Pd(allyl)Cl, Na₂CO₃, 1,4-dioxane, EtOH, water,63%; (i) H₂, Pd/C, MeOH, (R¹

= iPr, R²= H, 86%); (j) HCl, iPrOH. MeOH, RT, 16 h, 94%.

a

either Pd/C or Pt/C as catalysts. Boc removal under acidic

conditions provided piperidines 48j−m in nearly quantitative

yields. In a similar fashion, the boronic ester 63 was coupled to 4-

chloroazaindole (66) and chloro pyrazolopyrimidine (68) using

the N-heterocyclic carbene complex allyl[1,3-bis(2,6-

diisopropylphenyl)imidazol-2-ylidene]chloropalladium(II) to

provide tetrahydropyridines 67 and 69, respectively. Catalytic

reduction of the double bond in 67 and 69, followed by Boc

cleavage aﬀorded the piperidines 48n and 48o, respectively

(Scheme 6).

The piperidine intermediates 48a−o were transformed into

the ﬁnal targets via amide coupling reaction conditions with the

corresponding carboxylic acids of acid chlorides (Scheme 7).

Piperidines bearing benzyl-like substituents at the nitrogen

atom 37, 38, and 40 were prepared by reductive amination with

sodium triacetoxyborohydride of diﬂuoromethyltriazolopyrimi-

dine 48i with readily available aldehydes 70 (Scheme 8).

Alternatively, substitution-alkylation of 48i (Scheme 7) with the

derivative 71 provided the benzylamine 39 in 37% yield.

Scheme 7. Synthesis of Compounds 3, 7−18, 21−36

a

Reagents and conditions: for compounds 3−7, 10−12, 14−18, and

21−36 (a) (Het)ArCOOH, HBTU, DIPEA, DCM, 1 h, 15−95%; for

compounds 8, 9, and 13 (b) (Het)ArCOCl, DIPEA, DCM, RT, 1 h,

11−80%.

Finally, the target compounds containing a trisubstituted

piperidine were prepared following the synthetic route outlined

in Scheme 9. Thus, catalytic hydrogenation of commercially

available methyl 4-methylpyridine-3-carboxylate 72 using PtO₂

in acetic acid (Scheme 9), followed by Boc-protection of the

K

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

able to rapidly identify highly potent PDE2A inhibitors with

greater than 100-fold selectivity over PDE3A and PDE10A.

Furthermore, several examples (39, 42, and 46) demonstrated

that selectivity versus other members of the PDE enzyme family

was resolved by targeting the Q2 pocket. Racemization issues

were overcome by the introduction of a methyl substituent in

the central core, leading to an optimal trisubstituted piperidine

scaﬀold. Four lead compounds were evaluated to test BBB

penetration with 42 and 46 showing most promise, and in vivo

PDE2 target engagement was conﬁrmed for these two

compounds via p.o. and s.c. administration, respectively. The

FEP calculations were useful for identifying unexpectedly active

compounds from amongst the virtual libraries that were proﬁled.

We consider the approach of FEP calculations on increasingly

larger virtual libraries and chemical spaces to oﬀer signiﬁcant

potential for future LO eﬀorts.⁵²It provides a way to identify

stand-out or unexpectedly active compounds, thereby having a

larger impact on the LO process. Finally, despite the good results

here, it is important to note that FEP can be problematic in real-

world prospective applications as described in the recent large

report of Schindler et al.⁵³We will follow up with a detailed

report on the optimization of leads from this series.

Scheme 8. Synthesis of Compounds 37−40

a

Reagents and conditions: (a) NaBH(OAc)₃, DIPEA, DCM, RT, 4 h

(36%−86%); (b) DIPEA, DMF, 60 °C, 4 h (37%).

resulting piperidine, provided methyl 4-methylpiperidine-3-

carboxylateas a cis/trans mixture of diastereoisomers, with trans

isomer 73 being the major component in the mixture. The

desired diastereoisomer 73 could be separated by column

chromatography. Saponiﬁcation of the methyl ester in 73,

followed by CDI-mediated coupling with methoxymethylamine,

provided Weinreb amide 74. Addition of methylmagnesium

bromide to amide 74 led to the methylketone 75, which was

then condensed with diﬂuoroethylacetate to aﬀord the 1,3-

dicarbonylic compound 76. In turn, 76 was reacted with 2-

aminotriazole 53 to provide the triazolopyrimidine intermediate

77. The Boc N-protecting group was partially cleaved in the

cyclization reaction conditions; therefore, the crude reaction

mixture (RM) was treated with Boc₂O to reprotect the

deprotected intermediate in order to enable a better separation

of 77 from the reaction mixture. Deprotection of the Boc N-

protecting group under acidic conditions provided piperidine

78. SFC puriﬁcation at this stage allowed isolation of the

eutomer of 78, which was then submitted to amide coupling or

to nucleophilic aromatic substitution conditions to provide the

target molecules 41−43 and 44−46, respectively.⁵⁰

EXPERIMENTAL SECTION

■

X-ray Crystallography. Cocrystals of C-His tagged human 579-

921 PDE2A with HTS hit 3 and lead 39 were grown in 0.2 M MgCl₂, 0.1

M Tris-HCl pH 8.0 and 20% PEG 3350 and 0.2 M MgCl₂, 0.1 M Tris-

HCl pH 8.0 and 24% PEG 3350, respectively, both in the presence of 1

mM compound. Crystals were ﬂash-frozen directly from the hanging

drop without additional cryo-protectant by plunging into liquid

nitrogen. The HTS hit 3 data set was collected in house on a Rigaku

Saturn 944 CCD detector. The lead 39 data set was collected at the

Diamond Light Source at beamline 102 on an ADSC Q315 CCD

detector. Data sets were indexed, integrated, and scaled using

MOSFLM and SCALA (CCP4). The PDE2A model with PDB ID

1Z1L was used for molecular replacement with the data sets using

PHASER (CCP4). The resulting models were then given 10 cycles of

atomic reﬁnement with tight geometric weights using REFMAC5. In

each case, the diﬀerence electron density map calculated after molecular

replacement and initial reﬁnement was examined for the presence of

possible ligands. Once the ligand electron density was located, a

molecular structure ﬁle and reﬁnement library ﬁle were produced using

SKETCHER (CCP4). The ligands were ﬁtted into the electron density

using COOT and reﬁned using REFMAC5 (CCP4). Water molecules

were then added using the water placement option in COOT and

reﬁned using REFMAC5 (CCP4). Metal ions were also added and

CONCLUSIONS

■

The initial exploration of a triazolopyrimidine PDE2A inhibitor

HTS hit (3) has been described. Guided by the X-ray structure

and later with FEP relative binding energy calculations, we were

a

Scheme 9. Synthesis of Compounds 41−43

a

Reagents and conditions: (a) H₂, PtO₂, AcOH, 50 °C, 5 d, 97%; (b) Boc₂O, DCM, DIPEA, DMAP, RT, 1 h, 75%; (c) NaOH, MeOH, RT, 14 h,

100%; (d) CDI, MeNHOMe, Et₃N, THF, ACN, RT80 °C, 7 d,100%; (e) MeMgBr, THF, −15 °CRT, 1 h, 90%; (f) KOtBu, toluene, 0 °C

RT, 2 h, 100%; (g) (i) 53, DMF, 80 °C, 24 h; (ii) Boc₂O, Et₃N, RT, 30 min, 30%; (h) HCl, iPrOH, MeOH, 2 h, RT (72%); (i) (Het)ArCOOH,

HBTU, DIPEA, DCM, 1 h, 15−95%; (j) (Het)Cl or (Het)Br, DIPEA, nBuOH, 160 °C, 4 h (6−41%).

L

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

reﬁned. The structural geometry of both the protein and ligand were

ﬁnally checked using Maestro and MOE.

metabolic half-life (t_1/2) is calculated using the slope of the log−linear

regression from the percentage parent compound remaining versus

time relationship. The in vitro intrinsic clearance (Cl_int) (mL/min/mg

microsomal protein) is derived from the half-life taking into account the

incubation volume and the weight of microsomal protein in the

incubation.

Protocols for Measuring Inhibition of PDE’s In Vitro. PDEs

1B1 and 11A4 were expressed in HEK cells from full-length human

recombinant clones. Human recombinant PDEs 2A, 4D3, 5A3, 7A1,

9A1, 10A2, and rat PDE10A2 were expressed in Sf9 cells using a

recombinant baculovirus construct containing the full-length sequence

containing a a 6xHis sequence, following the start Met to allow metal

aﬃnity puriﬁcation of the recombinant protein. Cells were harvested,

and the PDE protein was puriﬁed by metal chelate chromatography on

Ni-sepharose 6FF. PDE6A, PDE3A, and PDE8A were purchased as

partially puriﬁed Sf9 cell lysates (Scottish Biomedical, UK). All enzymes

were diluted in 50 mM Tris pH 7.8, 1.7 mM EGTA, and 8.3 mM MgCl₂

except for PDE9A, which was diluted in 50 mM Tris pH 7.8 and 5 mM

MnCl₂, and PDE1B was diluted in 50 mM Tris pH 7.8, 8.3 mM MgCl₂

complemented with 624 U/mL calmodulin and 800 μM CaCl₂. The

aﬃnity of the compounds for PDEs was measured by a scintillation

proximity assay (SPA). PDE yttrium silicate SPA beads allow the PDE

activity to be measured by direct binding of the primary phosphate

groups of noncyclic AMP or GMP to the beads via an iron complex

chelation mechanism. The amount of bound tritiated product ([³H]-

AMP or [³H]-GMP) was measured by liquid scintillation counting in a

TopCount (Packard).

In Vivo PDE2 Occupancy Method. Male Wistar rats were treated

by systemic administration (s.c. or p.o.) of vehicle or single dose of

selective PDE2A inhibitors (n = 3 per dose). After 15 min, every rat

received a s.c. injection of the PDE10 inhibitor MP-10 at the dose of 2.5

mg/kg, followed 55 min later by an i.v. injection of the selective PDE2A

radioligand [³H]-1.^27,28Dosing rats with MP-10,⁴²a potent and

selective PDE10 inhibitor, prior to the radioligand injection allowed to

increase the signal to noise ratio of [³H]-1 because its binding is

increased when intracellular cGMP concentration increased.^27,28Rats

were sacriﬁced 5 min after the tracer injection, and their brains were

dissected and frozen. Then, 20 μm of thick coronal sections were cut

using a cryostat, collected on glass slides, and dried. Brain sections were

loaded in a β-imager for 12 h. The speciﬁc binding was determined as

the diﬀerence between [³H]-1 binding quantiﬁed in the striatum (a

brain area showing a high density of PDE2A) and in the cerebellum (a

brain area where PDE2A is virtually absent). Occupancy was calculated

as the inhibition of speciﬁc [³H]-1 binding in drug-treated animals

relative to vehicle-treated animals. The percentage of PDE2A

occupancy was measured at one dosage.

Compounds were dissolved and diluted in 100% dimethyl sulfoxide

(DMSO) in polystyrene plates to a concentration of 100-fold the ﬁnal

concentration in the assay. Rat or human PDE10A or human PDE2A or

human PDE3A or PDE3B enzyme solution (10 μL) was added to 20 μL

of incubation buﬀer (50 mM Tris pH 7.8, 8.3 mM MgCl₂, 1.7 mM

EGTA), 10 μL of substrate solution consisting of a mixture of

nontritiated and tritiated substrate (60 nM cAMP, 0.008 μCi ³H-cAMP

Computational Docking Procedure. Docking used the Glide

̈

software from Schrodinger Inc. The corresponding crystal structure

solved with 39 was used. The protein was prepared using the protein

preparation wizard⁵⁴in Maestro.⁵⁵Hydrogen-bond constraints were

enforced with Gln812 and Gln859 to ensure accurate placement of the

triazolopyrimidine scaﬀold. All molecules were prepared for docking

using the LigPrep tool. All default settings were used, and ionization

states were manually checked. The ligands were parameterized for use

with the OPLSv2.1 force ﬁeld up front using the tools available in

Maestro. The Glide XP scoring function was used with expanded

sampling, all other docking parameters were set to the defaults. Results

were visually inspected to ensure that each molecule adopted the

desired binding mode consistent with the crystal structure.

3

for human and rat PDE10A, 10 μM cGMP, 0.01 μCi H-cGMP for

PDE2A, 0.1 μM cAMP, 0.024 μCi ³H-cAMP for PDE3A or PDE3B),

and 0.4 μL of compound in 100% DMSO in a 384-well plate and

incubated for 60 min at room temperature for human and rat PDE10A

and PDE3A/B or for 40 min at room temperature for PDE2A. After

incubation, the reaction was stopped with 20 μL of stop solution,

consisting of PDE SPA beads (17.8 mg beads/mL in 18 mM zinc sulfate

for human and rat PDE10A or in 200 mM zinc sulfate for PDE2A and

PDE3A, and in 400 mM zinc sulfate for PDE3B). After sedimentation

of the beads for 30 min, the radioactivity was measured in a

PerkinElmer TopCount scintillation counter and results were expressed

as counts per minute (cpm). To measure the low control, no enzyme

was added to the reaction mixture. The same assay principle was applied

for the measurement of the inhibition of other members of the PDE

family, with appropriate modiﬁcations of enzyme concentration,

incubation buﬀer, substrate solution, incubation time, and stop

solution. Data were calculated as the percentage of inhibition of total

activity measured in the absence of test compound (% control). A best-

ﬁt curve was ﬁtted by a minimum sum of squares method to the plot of

% control versus compound concentration, from which an IC₅₀value

(inhibitory concentration causing 50% inhibition of hydrolysis) was

obtained. Compounds 1−40 were tested on rat PDE10A, and 41−46

were tested on human PDE10A.

Computational FEP Procedure. All FEP calculations were

conducted using the Schrodinger molecular modeling suite version

2015-1, the implementation is often referred to as FEP+. Calculations

were set up using the mapper technology to deﬁne the perturbations.

Default protocols were used with a 5 ns simulation length for ligands

both in complex and in solution. As mentioned above, the FEP

calculations were performed with the molecules in the chosen docked

pose. Molecules were treated in a neutral form and missing OPLS force-

ﬁeld parameters were calculated and ﬁtted up front using the force-ﬁeld

builder tool. The results of the simulations with 5 ns simulation time

and 12 λ windows were used to deﬁne the molecules recommended for

synthesis in the prospective application. The experimental pIC₅₀were

converted to expected binding free energies at 298 K using the relation

RT ln(IC₅₀); any oﬀset with respect to binding free energy cancels when

comparing relative diﬀerences. The calculations deliver the RBFE

diﬀerences between compounds, and the resulting δδG’s were ﬁtted to

the mean of the experimental δG to provide the calculated δG with

reduced oﬀset. MUEs are reported by comparing experimental and

calculated δG (kcal/mol) and shown with the corresponding 99%

conﬁdence interval. The force ﬁeld available at the time was OPLSv2.1.

To compare with newer versions, the predicted ΔG for the

retrospective dataset was recalculated using Schrodinger suite 2019−

3 and OPLSv3e. A high degree of correlation was seen with OPLSv2.1,

R 0.9. Input structures for FEP calculations are provided in the

Supporting Information.

CYP450 Inhibition Assay. The potential to reversibly inhibit the

major human P450 isoforms (CYP450s 1A2, 2C9, 2C19, 2D6, and

3A4) was determined in the CYP450 inhibition cocktail assay. Test

compounds are incubated across a concentration range (ﬁve

concentrations) with human liver microsomes and probe substrates

for each of the six isoforms (CYP450 1A2, 2C8, 2C9, 2C19, and 2D6 in

cocktail format and 3A4 in the single probe format) to estimate the IC₅₀

for inhibition of the probe substrate by the test compound. After the

deﬁned incubation time, the reaction is quenched and samples are

centrifuged and prepared for analysis by liquid chromatography/mass

spectrometry (LC/MS). The percentage inhibition of the probe

substrate metabolite formation is plotted against inhibitor concen-

tration and an IC₅₀calculated by curve ﬁtting.

Experimental Procedures and Compound Characterization.

All reactions were carried out by employing standard chemical

techniques under an inert atmosphere. Solvents used for extraction,

washing, and chromatography were of high-performance liquid

chromatography (HPLC) grade. Unless otherwise noted, all solvents

and chemicals used were of reagent grade. Anhydrous solvents

tetrahydrofuran (THF), DCM, and DMF were purchased from

Microsomal Metabolic Stability Assay. The metabolic stability

of a test compound was tested at Cyprotex by using liver microsomes

(0.5 mg/mL protein) from human and preclinical species incubated up

to 60 min at 37 °C with 1 μM of the test compound. The in vitro

M

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Sigma-Aldrich or Acros. All ﬁnal compounds were characterized by ¹H

NMR and LC/MS. ¹H nuclear magnetic resonance spectra were

recorded on Bruker spectrometers: DPX-360 MHz, DPX-400 MHz,

Bruker Avance Neo spectrometer operating at 400 MHz, or on a Bruker

Avance III-HD spectrometer operating at 600 MHz using chloroform-d

(deuterated chloroform, CDCl₃) or DMSO-d₆(deuterated DMSO,

dimethyl-d₆sulfoxide) as solvents. Chemical shifts (d) are reported in

parts per million relative to tetramethylsilane, which was used as the

internal standard.

dimethyl acetal (58 mL) was heated at 100 °C for 16 h. The reaction

mixture was concentrated in vacuo to give the title compound (30 g),

which was used in the next step without further puriﬁcation.

Intermediate 52c, tert-Butyl 3-[3-(Dimethylamino)prop-2-enoyl]-

azepane-1-carboxylate. Prepared with the same procedure for 52b,

starting from tert-butyl 3-acetylazepane-1-carboxylate [1782629-29-1]

(18.2 g) to give to give the title compound, (22 g), which was used in

the next step without further puriﬁcation.

Intermediate 52d, Di-tert-butyl 2-Acetylpiperazine-1,4-dicarbox-

ylate, Prepared in Three Steps. Step 1, intermediate 79, di-tert-butyl 2-

[methoxy(methyl)carbamoyl]piperazine-1,4-dicarboxylate. To a sus-

pension of 1,4-bis[(t-butoxy)carbonyl]piperazine-2-carboxylic acid

[173774-48-6] (60 g, 0.18 mol) in THF (285 mL) was added di-1H-

imidazol-1-yl-methanone (CDI, 40 g, 0.25 mol), and the mixture was

stirred at RT for 2 h. In another ﬂask, to a suspension of N-methoxy-

methanamine hydrochloride (24.45 g, 0.25 mol) in ACN (190 mL) was

added trimethylamine (36.35 mL, 0.26 mol). Both mixtures were

combined and stirred at RT for 16 h. The solvents were evaporated. The

residue was dissolved in DCM and washed with water, 20% AcOH

solution, and ﬁnally an aqueous NaOH (1 N) solution. The organic

layer was dried over MgSO₄, ﬁltered, and evaporated. The product was

puriﬁed by column chromatography (silica gel, eluent: a gradient DCM

100 to 3% MeOH in DCM). The pure fractions were evaporated,

Elemental analyses were performed with a Carlo-Erba EA1110

analyzer. Thin-layer chromatography (TLC) was carried out on silica

gel 60 F254 plates (Merck) using reagent grade solvents. Open column

chromatography was performed on a silica gel, mesh 230−400 particle

size and 60 Å pore size (Merck) under standard techniques. Automated

ﬂash column chromatography was performed using ready-to-connect

cartridges from Merck, on an irregular silica gel, particle size 15−40 pm

(normal phase disposable ﬂash columns) on a SPOT or LAFLASH

system from Armen Instrument. All ﬁnal compounds were conﬁrmed to

be >95% pure via HPLC methods. All the LC/MS analyses were

performed using an Agilent G1956A LC/MS quadrupole coupled to an

Agilent 1100 series LC system consisting of a binary pump with a

degasser, autosampler, thermostated column compartment, and diode

array detector. The mass spectrometer was operated with an

atmospheric pressure electrospray ionization source in the positive

ion mode. The capillary voltage was set to 3000 V, and the fragmentor

voltage was set to 70 V, and the quadrupole temperature was

maintained at 100 °C. The drying gas ﬂow and temperature values were

12.0 L/min and 350 °C, respectively. Nitrogen was used as the

nebulizer gas at a pressure of 35 psig. Data acquisition was performed

with Agilent ChemStation software. Analyses were carried out on a

YMC pack ODS-AQ C18 column (50 mm long × 4.6 mm I.D.; 3 μm

particle size) at 35 °C, with a ﬂow rate of 2.6 mL/min. A gradient

elution was performed from 95% (water + 0.1% formic acid)/5%

acetonitrile to 5% (water + 0.1% formic acid)/95% acetonitrile in 4.8

min; the resulting composition was held for 1.0 min; from 5% (water +

0.1% formic acid)/95% acetonitrile to 95% (water + 0.1% formic acid)/

5% acetonitrile in 0.2 min. The standard injection volume was 2 μL.

Acquisition ranges were set to 190−400 nm for the UV-PDA detector

and 100−1400 m/z for the MS detector. Optical rotations measure-

ments were carried out on a 341 PerkinElmer polarimeter in the

indicated solvents. Melting points were determined with a DSC823e

(Mettler Toledo) and were measured with a temperature gradient of 30

°C/min. The reported values are peak values. Purities of all new

compounds were determined by analytical reverse phase (RP)-HPLC

using the area percentage method on the UV trace recorded at a

wavelength of 254 nm, and compounds were found to have ≥95%

purity unless otherwise speciﬁed.

The absolute stereochemical conﬁguration for some of the

compounds was determined using vibrational circular dichroism

(VCD). They were measured on a Bruker Equinox 55 equipped with

a PMA 37 in a KBr liquid cell using CD₂CI₂as the solvent (PEM:

1350cm^−l, LIA: l mV, resolution: 4 cm⁻¹). A description on the use of

VCD for the WO 2018/083101 PCT/EP2017/077918-29-5 10 15 20

25 30 determination of absolute conﬁguration can be found in Dyatkin

A.B. et al., Chirality, 14, 215−219 (2002). Ab initio calculations: A

thorough conformational search was performed at molecular mechanics

level using a Macromodel to do a mixed torsional/low-mode sampling

with the OPLS-2005 force ﬁeld. The located minima were optimized

using Jaguar at the B3LYP/6-31G** level with a Poisson−Boltzmann

continuum solvation model to mimic a dichloromethane solvent. All

conformations within 10 kJ/mol interval were used to simulate VCD

and IR spectra. Dipole and rotational strengths were calculated at the

same B3LYP/6-31G** level using Jaguar. The calculated VCD spectra,

generated after scaling the frequencies with a factor of 0.97, converting

to a Lorentzian bandshape, and summing up the contribution of each

conformer assuming a Boltzmann ensemble, were visually compared

with the experimental spectra for assigning the correct stereo chemistry.

Intermediate 52b, tert-Butyl 2-[3-(Dimethylamino)prop-2-enoyl]-

morpholine-4-carboxylate. A stirred mixture of t-butyl 2-acetylmor-

pholine-4 carboxylate (25 g, 0.11 mol) in N,N-dimethylacetamide

1

yielding intermediate 79 (56.8 g, 83.7% yield). H NMR (400 MHz,

DMSO-d₆): δ ppm 1.37 (d, J = 6.06 Hz, 18H), 2.86−2.96 (m, 1H),

3.06−3.12 (m, 3H), 3.22−3.34 (m, 1H), 3.39−3.55 (m, 1H), 3.59−

3.63 (m, 1H), 3.73 (s, 3H), 3.76−3.84 (m, 1H), 4.11−4.19 (m, 1H),

4.70 (br d, J = 2.83 Hz, 1H). Step 2, intermediate 80, di-tert-butyl 2-

acetylpiperazine-1,4-dicarboxylate. Intermediate 77 (56.8 g, 0.15 mol)

in THF (200 mL) was charged in a ﬂask under N₂and cooled to 0 °C.

Methylmagnesium bromide (265 mL, 1.4 M in toluene/THF 75/25)

was added dropwise, with the temperature not exceeding 15 °C. After

addition, the reaction mixture was stirred at RT for 1 h. Then, the

reaction mixture was poured on ice with 100 mL of AcOH. The product

was extracted with Et₂O, and the organic layer was washed with a 5%

NaHCO₃solution. The organic layer was dried over MgSO₄, ﬁltered,

and evaporated to give intermediate 80 (36.6 g, 73%), used as such in

the next step. Gas chromatography−MS m/z 328 (M). Step 3,

intermediate 52d, di-tert-butyl 2-acetylpiperazine-1,4-dicarboxylate.

That was prepared with the same procedure as 52b, from di-tert-

butyl 2-acetylpiperazine-1,4-dicarboxylate (intermediate 78) (10 g) to

give the title compound (11 g, 94%). MS (ESI) m/z: 384 [M + H]⁺.

Representative Procedure for the Preparation of Intermedi-

ates 54a−54d. Intermediate 54a, tert-Butyl 3-([1,2,4]Triazolo[1,5-

a]pyrimidin-7-yl)piperidine-1-carboxylate. Intermediate 52a (33.0 g,

0.12 mol), 1H-1,2,4-triazol-5-amine hydrochloride (1:1) (12.3 g, 0.14

mol), and AcOH (75 mL) were stirred at 130 °C for 1 h. The reaction

mixture was cooled on an ice bath and left under stirring overnight at

RT. The reaction mixture was concentrated in vacuo, co-evaporated

with toluene, diluted with Et₂O (0.7 L), and then poured onto ice (0.5

L). The layers were separated, and the aqueous layer was back extracted

with Et₂O (3 × 0.3 L). The combined organic layers were treated with

brine, dried over MgSO₄, ﬁltered, and evaporated to give the title

compound (racemic) as a brown oil (25 g, 70.5%). MS (ESI) m/z: 304

[M + H]⁺. This material could be used as such. Alternatively, 54a was

also separated into enantiomers 54aa and 54ab by puriﬁcation by Prep

SFC (stationary phase: CHIRALPAK Daicel OJ 20 × 250 mm; mobile

phase: CO₂, MeOH) yielding intermediate 54aa (R!, 10 g , 28%, R_t=

5.59 min) and intermediate 54ab (S!, 10 g, 28%, R_t= 7.4 min).

Intermediate 54b, tert-Butyl 2-([1,2,4]Triazolo[1,5-a]pyrimidin-7-

yl)morpholine-4-carboxylate. Prepared with the same procedure for

54a, starting from intermediate 52b (30 g, 0.11 mol) to give the title

compound 54b (racemic mixture, 19 g, 62%) after puriﬁcation on a

silica gel (eluent: DCM, 1% MeOH in DCM, 2%, 4%). MS (ESI) m/z:

305 [M + H]⁺.

Intermediate 54c, tert-Butyl 3-([1,2,4]Triazolo[1,5-a]pyrimidin-7-

yl)azepane-1-carboxylate. Prepared with the same procedure for 54a,

starting from intermediate 52c (5.0 g, 16.87 mmol) to give 1.69 g of the

title compound 54c (racemic mixture, 31.56%) after puriﬁcation on a

N

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

silica gel (eluent: DCM, 1% MeOH in DCM, 2%). MS (ESI) m/z: 318

[M + H]⁺.

24.1 mmol) were stirred in melt at 150 °C for 3 h. The reaction mixture

was cooled to RT and dissolved in DCM. The organic layer was washed

with water, dried over MgSO₄, ﬁltered, and evaporated under reduced

pressure. The resulting residue was puriﬁed by column chromatography

on a silica gel using a gradient (DCM, 0.25% MeOH in DCM, 0.5, 1%)

as the eluent to provide the title compound (1 g, 12%) MS (ESI) m/z:

350 [M + H]⁺

Intermediate 57, tert-Butyl 3-(5-Methylsulfonyl-[1,2,4]triazolo-

[1,5-a]pyrimidin-7-yl)piperidine-1-carboxylate. A solution of inter-

mediate 56 (1.39 g, 3.98 mmol) and mCPBA (2.75 g, 15.9 mmol) in

chloroform (CHCl₃, 50 mL) was stirred at reﬂux for 2 h. After this time,

the reaction mixture was cooled to RT, then washed with NaOH 1 N,

dried over MgSO₄, ﬁltered, and evaporated in vacuo to yield the title

compound (1.50 g, 99%), which was used as such in the next step. MS

(ESI) m/z: 380 [M − H]⁺.

Intermediate 81e, tert-Butyl 3-(5-Methoxy-[1,2,4]triazolo[1,5-

a]pyrimidin-7-yl)piperidine-1-carboxylate. Sodium methoxide

(NaOMe, 234 mg, 4.33 mmol) was added at RT to a solution of

intermediate 57 (1.5 g, 3.93 mmol) in methanol (MeOH, 50 mL), and

the resulting mixture was stirred at this temperature for 12 h. After this

time, the solvent was evaporated in vacuo and the resulting residue was

partitioned between water and DCM. The organic layer was separated,

and then the aqueous layer was back extracted with DCM (2×). The

combined organic layer was dried over MgSO₄, ﬁltered, and evaporated

under reduced pressure. The resulting residue was puriﬁed by column

chromatography on a silica gel using a gradient (DCM, 1% MeOH in

DCM, 2%) as the eluent to provide the title compound (560 mg, 43%).

MS (ESI) m/z: 334 [M + H]⁺.

Intermediate 81f, tert-Butyl 3-(5-Cyano-[1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)piperidine-1-carboxylate. Intermediate 57 (950 mg,

2.50 mmol) and sodium cyanide (249 mg, 4.98 mmol) were stirred in

DMSO (5 mL) at RT for 30 min. The reaction mixture was poured in

water and extracted with EtOAc. The organic layer was washed with

brine, dried on MgSO₄, ﬁltered, and evaporated under reduced

pressure. The resulting residue was puriﬁed via ﬂash column

chromatography on a silica gel using a gradient (DCM, 1% MeOH in

DCM, 2%) as the eluent to give intermediate 79f (820 mg, 100%). MS

(ESI) m/z: 329 [M + H]⁺.

Intermediate 48e, 5-Methoxy-7-(3-piperidyl)-[1,2,4]triazolo[1,5-

a]pyrimidine. TFA (0.626 mL, 8.18 mmol) was added at RT to a

solution of intermediate 81e (560 mg, 1.68 mmol) in DCM (50 mL),

and the resulting mixture was stirred for 2 h at RT. Saturated aqueous

NaHCO₃solution was then added to the reaction mixture and stirred

for 10 min. The organic layer was separated, dried on MgSO₄, ﬁltered,

and evaporated under reduced pressure to give intermediate 48e (390

mg, 99.5%), which was used in the next step without further

puriﬁcation.

Intermediate 48f, 7-(3-Piperidyl)-[1,2,4]triazolo[1,5-a]-

pyrimidine-5-carbonitrile. Prepared in a similar way to procedure

3b, starting from intermediate 81f (820 mg, 2.5 mmol) to give the title

compound 48f (390 mg, 68%), which was used in the next step without

further puriﬁcation.

Intermediate 81g, tert-Butyl 3-(5-Methyl-[1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)piperidine-1-carboxylate. 81g is prepared in two steps.

Step 1, intermediate 58, tert-butyl 3-[(Z)-3-(dimethylamino)but-2-

enoyl]piperidine-1-carboxylate. Intermediate 50a (2.0 g, 8.8 mmol)

was added to N,N-dimethylacetamide dimethyl acetal (3.52 g), and the

mixture was reﬂuxed for 3 h. The reaction mixture was evaporated to

give the crude product, which was used as such in the next step (2.5 g,

95%). Step 2, intermediate 81g, tert-butyl 3-(5-methyl-[1,2,4]triazolo-

[1,5-a]pyrimidin-7-yl)piperidine-1-carboxylate. A solution of inter-

mediate 58 (2.5 g, 8.43 mmol) and 1H-1,2,4-triazol-5-amine hydro-

chloride (0.71 g, 8.43 mmol) in acetic acid (50 mL) was stirred at reﬂux

during 30 min. Saturated aqueous Na₂CO₃solution was added, and the

product was extracted with DCM. The organic layer was dried on

MgSO₄, ﬁltered, and evaporated. The resulting residue was puriﬁed by

ﬂash column chromatography on a silica gel using a gradient (DCM, 1%

MeOH in DCM, 2%) as the eluent to give intermediate 81g (2.14 g,

80%). MS (ESI) m/z: 318 [M + H]⁺.

Intermediate 54d, Di-tert-butyl 2-([1,2,4]Triazolo[1,5-a]-

pyrimidin-7-yl)piperazine-1,4-dicarboxylate. Prepared with the

same procedure for 54a, starting from intermediate 52d (5 g, 13.04

mmol) to give 2.42 g of the title compound 54d (racemic mixture, 40%)

after puriﬁcation on a silica gel (eluent: DCM, 1% MeOH in DCM,

2%). MS (ESI) m/z: 405 [M + H]⁺.

Intermediate 48a, 7-(3-Piperidyl)-[1,2,4]triazolo[1,5-a]-

pyrimidine. To a solution of intermediate 54a (53.8 g, 0.18 mol) in

1,4 dioxane (600 mL) was added HCl-1,4 dioxane (200 mL, 4 M), and

the reaction mixture was stirred for 2 h at RT. The product crystallized

out as a hydrochloric salt, and it was ﬁltered and washed with ether to

give very hygroscopic crystals. These were then triturated in ether,

ﬁltered, and dried to give the title compound 48a as a HCl salt (racemic,

39.7 g, 93%). MS (ESI) 204 [M + H]⁺. ¹H NMR (360 MHz, DMSO-

d₆): δ ppm 1.74−2.00 (m, 3H), 2.13−2.23 (m, 1H), 2.76−3.01 (m,

1H), 3.22−3.40 (m, 2H), 3.58−3.66 (m, 1H), 3.94−4.04 (m, 1H), 7.34

(d, J = 4.39 Hz, 1H), 8.75 (s, 1H), 8.90 (d, J = 4.76 Hz, 1H), 9.27−9.51

(m, 1H), 9.65 (br d, J = 9.15 Hz, 1H). Anal. Calcd (C₂₅H₂₃N₅O): C, H,

N. Alternatively, 48a (4 g) was also separated into enantiomers by

puriﬁcation by Prep SFC (Stationary phase: CHIRALPAK Daicel OJ

20 × 250 mm; mobile phase: CO₂, EtOH with 0.5% iPr NH₂) yielding

intermediates: 48aa (R!, 2g), R_t= 6.43 min, OR = +56.59° (589 nm, 20

°C, 0.42 w/v %; MeOH) and intermediate 48ab (S!, 3 g), R_t= 9.15 min,

OR = −53.9° (589 nm, 20 °C, 0.43 w/v %; MeOH).

Intermediate 48b, 2-([1,2,4]Triazolo[1,5-a]pyrimidin-7-yl)-

morpholine. To a solution of intermediate 54b (19.0 g, 62.2 mmol)

in MeOH (100 mL) was added HCl−Et₂O (200 mL, 1 M), and the

reaction mixture was stirred for 16 h at RT. The solvent was evaporated

under reduced pressure, and the residue was placed in DCM and

treated with a saturated aq sol of NaHCO₃. The biphasic layer was

separated, and then the aqueous layer was extracted 2× DCM. The

combined OL was dried over MgSO₄, ﬁltered, and evaporated to

1

provide 48b (racemic, 9 g, 71%). MS (ESI) m/z: 206 [M + H]⁺. H

NMR (360 MHz, DMSO-d₆): δ ppm 2.94−3.25 (m, 3H) 3.28−3.41

(m, 1H) 3.85 (br d, J = 12.64 Hz, 1H) 4.08−4.20 (m, 1H) 4.21−4.31

(m, 1H) 5.65 (dd, J = 11.03, 2.45 Hz, 1H) 7.40 (d, J = 4.53 Hz, 1H) 8.78

(s, 1H) 8.96 (d, J = 4.53 Hz, 1H) 10.03 (d, J = 9.70 Hz, 1H) 10.24 (br d,

J = 10.69 Hz, 1H). This material was also separated into enantiomers by

puriﬁcation by Prep SFC (stationary phase: CHIRALPAK Daicel OJ 20

× 250 mm; mobile phase: CO₂, MeOH) yielding intermediates: 54ba

(R!, 1.52 g, 8%), R_t= 8.19 min and 54bb (S!, 1.77 g), R_t= 11.6 min.

Intermediate 48c, 7-(Azepan-3-yl)-[1,2,4]triazolo[1,5-a]-

pyrimidine. Prepared as in the procedure for 48b, starting from

intermediate 54c (600 mg, 1.89 mmol) to give the title compound 48c

(400 mg, 73%), and used in the next step without further puriﬁcation.

MS (ESI) m/z: 218 [M + H]⁺.

Intermediate 48d, 7-Piperazin-2-yl-[1,2,4]triazolo[1,5-a]-

pyrimidine. Prepared as in the procedure for 48b, starting from

intermediate 54d (2.42 g, 5.98 mmol) to give the title compound 48d

(1.85 g, 98%), and used in the next step without further puriﬁcation.

Intermediate 55, tert-Butyl 3-[3,3-Bis(methylsulfanyl)prop-2-

enoyl]piperidine-1-carboxylate. NaH (60% dispersion in mineral

oil, 8.77 g, 0.22 mol) was added to a solution of tert-butyl 3-

acetylpiperidine-1-carboxylate (50a, 45.3 g, 0.20 mol) in anhydrous

THF (300 mL) at 0 °C under an atmosphere of N₂, and the resulting

mixture was stirred at this temperature for 10 min. After this time,

carbon disulﬁde (16.7 g, 0.22 mol) and then methyl iodide (59.4 g, 0.42

mol) were added dropwise. The reaction mixture was allowed to reach

RT and stirred for 4 h. Water was added, and the resulting mixture was

extracted with EtOAc. The combined organic layer was washed with

brine, dried over MgSO₄, ﬁltered, and evaporated under reduced

pressure. The resulting residue was puriﬁed via ﬂash column

chromatography on a silica gel using a gradient (DCM, 0.25%

MeOH in DCM, 0.5, 1%) as the eluent to give the title compound

(13.88 g, 21%). MS (ESI) m/z: 332 [M + H]⁺.

Intermediate 56, tert-Butyl 3-(5-Methylsulfanyl-[1,2,4]triazolo-

[1,5-a]pyrimidin-7-yl)piperidine-1-carboxylate. Intermediate 55 (8 g,

24.1 mmol) and 1H-1,2,4-triazol-5-amine hydrochloride (1:1) (2.03 g,

O

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Intermediate 48g, 5-Methyl-7-(3-piperidyl)-[1,2,4]triazolo[1,5-a]-

pyrimidine and 54a. Prepared according to procedure 3, starting from

intermediate 81g (2.14 g, 6.74 mmol), to give the title compound as a

mixture with 54a (1.07 g, 54.6%). ¹H NMR (360 MHz, DMSO-d₆): δ

ppm 1.76−2.01 (m, 3H) 2.12−2.23 (m, 1H) 2.64 (s, 3H) 2.87−2.98

(m, 1H) 3.22−3.41 (m, 2H) 3.59 (br d, J = 11.71 Hz, 1H) 3.89−4.00

(m, 1H) 7.31 (s, 1H) 8.69 (s, 1H) 9.55 (br d, J = 10.25 Hz, 1H) 9.80 (br

d, J = 10.25 Hz, 1H). MS (ESI) m/z: 218 [M + H]⁺.

Intermediate 82, tert-Butyl 3-(7-Methylimidazo[1,2-a]pyrimidin-

5-yl)piperidine-1-carboxylate. Prepared in a similar way to inter-

mediate 81g, starting from intermediate 58 (2.5 g, 8.43 mmol) and 1H-

imidazol-2-amine (0.71 g, 8.43 mmol) to give the title compound 82

(0.86 g, 32%). MS (ESI) m/z: 317 [M + H]⁺.

as the eluent (DCM, 1% CH₃CN in DCM, 5%) to provide the

intermediate 60 (42 g, 50%) and intermediate 61 (24.4 g, 29%).

Intermediate 60, ¹H NMR (400 MHz, DMSO-d₆): δ ppm 1.45 (s, 9H),

1.88 (quin, J = 6.06 Hz, 2H), 2.36−2.44 (m, 2H), 3.43−3.50 (m, 2H),

7.11 (s, 1H).

Intermediate 62, tert-Butyl 5-(4,4,5,5-Tetramethyl-1,3,2-dioxa-

borolan-2-yl)-3,4-dihydro-2H-pyridine-1-carboxylate. A degassed

solution of intermediate 60 (24.4 g, 73.65 mmol) in 1,4-dioxane

(700 mL) was added to a previously degassed mixture of PdCl₂(dppf)

(1.61 g, 2.19 mmol), dppf (1.22 g, 2.2 mmol), KOAc (21.552 g, 219.6

mmol), and bis(pinacolato)diboron (20.53 g, 80.84 mmol) at RT and

under a nitrogen atmosphere. The reaction mixture was stirred during

16 h at 80 °C, after which time it was ﬁltered over Celite. The ﬁltrate

was evaporated under reduced pressure, and the resulting residue was

puriﬁed via ﬂash column chromatography on a silica gel using a gradient

(DCM, 2% CH3CN in DCM, 5%) as the eluent to give the title

compound 62 as a white solid after recrystallization from heptane (10.8

g, 47%). MS (ESI) m/z: 209 [M-Boc + H]⁺.

Intermediate 48h, 7-Methyl-5-(3-piperidyl)imidazo[1,2-a]-

pyrimidine. Prepared according to the procedure for 48a, starting

from intermediate 82 (0.86 g, 2.72 mmol), to give the title compound

48h (0.75 g, 95%). ¹H NMR (400 MHz, DMSO-d₆): δ ppm 1.96−2.05

(m, 3H), 2.05−2.19 (m, 1H), 2.71 (s, 3H), 2.92−3.13 (m, 1H), 3.35−

3.54 (m, 1H), 3.99−4.2 (m, 1H), 7.53 (s, 1H), 8.27 (d, J = 2.42 Hz,

1H), 8.69 (d, J = 2.83 Hz, 1H), 9.78 (br s, 1H), 9.95 (br s, 1H).

Intermediate 59, tert-Butyl 3-(4,4-Diﬂuoro-3-oxo-butanoyl)-

piperidine-1-carboxylate. Intermediate 50a (17.4 g, 0.077 mol) was

stirred in benzene (200 mL) at 0−5 °C under N₂. Potassium tert-

butoxide (10.5 g, 93.6 mol) was added at 0−5 °C and 2,2-diﬂuoro-

acetic acid ethyl ester (11.43 g, 0.92 mol) was added dropwise at 0−5

°C. The reaction mixture was stirred at RT for 5 h, then it was brought

to pH = 7 with the aid of an aqueous H₂SO₄solution (10%, 10 mL).

The reaction mixture was extracted with EtOAc (2 × 50 mL), then the

combined organic layer was washed with brine, dried over MgSO₄,

ﬁltered and evaporated under reduced pressure, to give intermediate 59

(23 g, 98%), which was pure enough to be used in the next step. MS

(ESI) m/z: 304 [M − H]⁺.

Intermediate 81i, tert-Butyl 3-[5-(Diﬂuoromethyl)-[1,2, 4]-

triazolo[1,5-a]pyrimidin-7-yl]piperidine-1-carboxylate. Intermediate

59 (23 g, 75.3 mmol), 1H-1,2,4-triazol-5-amine hydrochloride (1:1)

(18.16 g, 150.7 mmol), and DMF (270 mL) were stirred at 60 °C for 28

h, then at 50 °C for 20 h. The reaction mixture was cooled to 0 °C, then

Et₃N (52 mL, 376 mmol) and ditert-butyl dicarbonate (16 g, 75.3

mmol) were added. The mixture was stirred for 1 h at RT, then

concentrated in vacuo. The resulting residue was partitioned between

DCM and an aqueous NH₄OH solution (20%) and stirred for 30 min.

The organic layer was separated, then dried over MgSO₄, ﬁltered and

concentrated under reduced pressure. The resulting residue was

puriﬁed by ﬂash column chromatography using a gradient heptane/

EtOAc, 100/0 to 60/40, as the eluent to give the title compound (23 g,

86%). MS (ESI) m/z: 354 [M + H]⁺.

Intermediate 48i, 5-(Diﬂuoromethyl)-7-(3-piperidyl)-[1,2,4]-

triazolo[1,5-a]pyrimidine. Prepared according to the procedure for

48a, starting from intermediate 81i (23 g, 65.09 mmol), to give the title

compound as an HCl salt (13 g, 61%). ¹H NMR (360 MHz, DMSO-

d₆): δ ppm 1.86−2.04 (m, 3H), 2.18 (br d, J = 7.48 Hz, 1H), 2.94 (br d,

J = 10.79 Hz, 1H), 3.29−3.49 (m, 2H), 3.59 (br d, J = 12.14 Hz, 1H),

3.98−4.14 (m, 2H), 7.15 (t, J = 54.52 Hz, 1H), 7.62 (s, 1H), 8.89 (s,

1H), 9.46 (br d, J = 11.57 Hz, 1H), 9.74 (br d, J = 10.70 Hz, 1H). MS

(ESI) m/z: 254 [M + H]⁺.

Intermediates 60, tert-Butyl 5-(Triﬂuoromethylsulfonyloxy)-3,4-

dihydro-2H-pyridine-1-carboxylate, and 61, tert-Butyl 5-(Triﬂuor-

omethylsulfonyloxy)-3,6-dihydro-2H-pyridine-1-carboxylate. A sol-

ution of commercially available 1-Boc-3-piperidinone (50 g, 250.9

mmol) in THF (125 mL) was added dropwise at −78 °C to a stirred

solution of LDA in a mixture of THF/heptane/ethylbenzene (179 mL,

2 M, 358 mmol) in THF (625 mL) under a nitrogen atmosphere. After

15 min, a solution of N,N-bis(triﬂuoromethylsulfonyl)aniline (107.5 g,

300.9 mmol) in THF (250 mL) was added dropwise to the reaction

mixture at −78 °C, and the resulting mixture was stirred for another 15

min. The reaction mixture was allowed to warm to RT overnight, and

then it was cooled to 0 °C and quenched with a saturated aqueous

NH₄Cl solution (190 mL), diluted with water, and extracted with

DCM. The organic layer was dried over MgSO₄, ﬁltered, and

concentrated under reduced pressure. The resulting residue was

puriﬁed by ﬂash column chromatography on a silica gel using a gradient

Intermediate 63, tert-Butyl 5-(4,4,5,5-Tetramethyl-1,3,2-dioxa-

borolan-2-yl)-3,6-dihydro-2H-pyridine-1-carboxylate. Prepared in a

similar way to intermediate 62, starting from intermediate 61 (24.4 g,

73.6 mmol), to give the title compound 63 (10.8 g, 35% yield). MS

(ESI) m/z: 209 [M-Boc + H]⁺.

General Procedure for the Suzuki Couplings of Intermedi-

ates 62 and 63 with the Corresponding Bicyclic Heteroar-

omatic Chlorides. Intermediate 65a, tert-Butyl 5-(5-Isopropyl-

[1,2,4]triazolo[1,5-a]pyrimidin-7-yl)-3,4-dihydro-2H-pyridine-1-car-

boxylate. A mixture of intermediate 62 (2.04 g, 6.61 mmol),

commercially available 7-chloro-5-(propan-2-yl)-[1,2,4]triazolo[1,5-

a]pyrimidine (1 g, 5.08 mmol), allylchloro[1,3-bis(2,6-di-i-propyl-

phenyl)-4,5-dihydroimidazol-2-ylidene]palladium(II) (584 mg, 1.02

mmol), Na₂CO₃(1.08 g, 10.17 mmol), 1,4-dioxane (5 mL), EtOH (5

mL), and distilled water (1 mL) was stirred under microwave

irradiation at 120 °C 2 h. The volatiles were removed under reduced

pressure, and the resulting residue was partitioned between water and

EtOAc. The organic layer was washed with brine, dried over MgSO₄,

ﬁltered, and concentrated in vacuo. The resulting residue was puriﬁed

by Prep HPLC using as the stationary phase a column Vydac Denali

C1810 μm, 250 g, 5 cm, and as the mobile phase a mixture of 0.25%

NH₄HCO₃solution in water and MeOH to give the title compound

65a (180 mg, 10%). MS (ESI) m/z: 344 [M + H]⁺.

Intermediate 65b, tert-Butyl 5-[5-(Triﬂuoromethyl)-[1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl]-3,4-dihydro-2H-pyridine-1-carboxy-

late. Prepared in a similar way to intermediate 65a, starting from

intermediate 62 (1.00 g, 3.23 mmol) and commercially available 7-

chloro-5-(triﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidine (0.72 g,

3.23 mmol) and employing Pd(PPh₃)₄(0.75 g, 0.65 mmol) as the

catalyst to give the title compound 65b (0.32 g, 26.8%). MS (ESI) m/z:

370 [M + H]⁺.

Intermediate 65c, tert-Butyl 5-(2,5-Dimethyl-[1,2,4]triazolo[1,5-

a]pyrimidin-7-yl)-3,4-dihydro-2H-pyridine-1-carboxylate. Prepared

in a similar way to intermediate 65a, starting from intermediate 62

(2.2 g, 7.12 mmol) and commercially available 7-chloro-2,5-dimethyl-

[1,2,4]triazolo[1,5-a]pyrimidine (1.0 g, 5.48 mmol) to give the title

compound 65c (0.17 g, 9.4%). MS (ESI) m/z: 330 [M + H]⁺.

Intermediate 65d, tert-Butyl 5-[5-Methyl-2-(triﬂuoromethyl)-

[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]-3,4-dihydro-2H-pyridine-1-car-

boxylate. Prepared in a similar way to intermediate 65a, starting from

intermediate 62 (1.31 g, 4.23 mmol) and commercially available, 7-

chloro-5-methyl-2-(triﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidine

(1.00 g, 4.23 mmol) and employing Pd(PPh₃)₄(0.97 g, 0.85 mmol) as

the catalyst to give the title compound 65d (0.39 g, 9.26%). MS (ESI)

m/z: 384 [M + H]⁺.

Intermediate 83a, tert-Butyl 3-(5-Isopropyl-[1,2,4]triazolo[1,5-

a]pyrimidin-7-yl)piperidine-1-carboxylate. Intermediate 65a (180

mg, 0.52 mmol) and a solution of thiophenol (0.1 mL, 4%, 0.039 mmol)

was added to a suspension of Pd/C (10%) (100 mg, 0.094 mmol) in

methanol (40 mL) under N₂ﬂow. The reaction mixture was stirred at

RT under H₂atmosphere until 1 equiv hydrogen was absorbed. The

catalyst was removed by ﬁltration over Celite and the ﬁltrate was

P

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

evaporated to give the title compound 83a (180 mg, 99.4%), which was

used in the next step without further puriﬁcation.

a silica gel using a mixture heptane/EtOAc, 80/20, as the eluent gave

the title compound 69 (1.3 g) MS (ESI) m/z: 351 [M + H]⁺.

Intermediate 85, tert-Butyl 3-[5-(Diﬂuoromethyl)pyrazolo[1,5-

a]pyrimidin-7-yl]piperidine-1-carboxylate. Prepared in a similar way

to 83a, starting from intermediate 69 (1.3 g, 3.71 mmol) to give the title

compound 85 (820 mg, 63%), MS (ESI) m/z: 351 [M + H]⁺. ¹H NMR

(400 MHz, DMSO-d₆): δ ppm 1.36 (br s, 9H), 1.49−1.63 (m, 1H),

1.74 (d, J = 11.3 Hz, 1H), 1.93−2.07 (m, 1H), 3.10 (ddd, J = 13.2, 9.7,

3.1 Hz, 1H), 3.40 (br s, 1H), 3.66−3.74 (m, 1H), 3.79 (d, J = 13.1 Hz,

1H), 4.15 (d, J = 13.2 Hz, 1H), 7.00 (t, J = 54.4 Hz, 1H), 6.94 (J = 2.5

Hz, 1H), 7.21 (s, 1H), 8.40 (d, J = 2.4 Hz, 1H).

Intermediate 48o, 5-(Diﬂuoromethyl)-7-(3-piperidyl)pyrazolo-

[1,5-a]pyrimidine. Prepared in a similar way to 48a, starting from

intermediate 85 (820 mg, 2.33 mmol) to give the title compound 48o as

an HCl salt (520 mg, 88%), which used as such in the next step.

Final Compound 4, (3,4-Dimethoxyphenyl)-[3-([1,2,4]triazolo-

[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. 3,4-Dimethoxyben-

zoic acid (760 mg, 4.17 mmol) was stirred in THF (40 mL). DIPEA

(2.16 g, 0.75 g/mL, 16.7 mmol) and HBTU (1.74 g, 4.59 mmol) were

added, stirring was continued for 20 min at RT. A solution of

intermediate 48a (1.38 g, 5.00 mmol) in DMF (10 mL) was added, and

stirring was continued for 2 h at RT. The solvents were evaporated

under reduced pressure, and the resulting residue was diluted with ice

and water and then extracted with 2× DCM. The combined organic

layers were washed with an aqueous solution of NaOH (1 N) and brine,

separated, dried over MgSO₄, ﬁltered, and then concentrated in vacuo.

This material was puriﬁed by recrystallization from EtOAc, yielding

compound 4 (1.04 g, 67.8% yield). MS (ESI) m/z: 368 [M + H]⁺. mp

159.2 C; ¹H NMR (400 MHz, DMSO-d₆): δ ppm 1.62−1.74 (m, 1H)

1.84 (dt, J = 13.62, 3.89 Hz, 1H) 1.96−2.06 (m, 1H) 2.20−2.27 (m,

1H) 3.08−3.18 (m, 1H) 3.39 (dd, J = 12.92, 10.09 Hz, 1H) 3.68−3.74

(m, 1H) 3.79 (s, 6H) 3.96−4.08 (m, 1H) 4.38 (br d, J = 12.51 Hz, 1H)

6.93−7.00 (m, 3H) 7.26 (d, J = 4.84 Hz, 1H) 8.56 (s, 1H) 8.81 (d, J =

4.44 Hz, 1H).

Final Compound 5, 2-(3,4-Dimethoxyphenyl)-1-[3-([1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]ethenone. 3,4-Dimethoxy-

phenylacetic acid (0.047 g, 0.24 mmol) was stirred in DCM (5 mL).

DIPEA (0.14 mL, 0.86 mmol) and HBTU (90 mg, 0.24 mmol) were

added, and stirring was continued for 30 min at RT. Intermediate 48a

(0.060 g, 0.22 mmol) was added, and stirring was continued for 2 h at

RT. The RM was quenched with water (1 mL), and then it was ﬁltered

on an Extrelut column. The OL was evaporated in vacuo. This material

was puriﬁed by Prep HPLC (stationary phase: RP Vydac Denali C18

10 μm, 250 × 5 cm, mobile phase: 0.25% NH₄HCO₃solution in water,

MeOH), to give compound 5 (57 mg, 69 % yield). MS (ESI) m/z: 382

[M + H]⁺. ¹H NMR (600 MHz, DMSO-d_6,100 °C): δ ppm 1.46−1.56

(m, 1H); 1.77 (m, 1H); 1.93 (dtd, J = 13.00, 11.21, 11.21, 4.05 Hz,

1H); 2.15−2.21 (m, 1H); 3.26−3.34 (m, 1H); 3.51 (br t, J = 9.91 Hz,

1H); 3.67 (m, 2H); 3.71 (s, 3H); 3.73 (m, 4H); 4.07−4.26 (m, 1H);

4.38−4.54 (m, 1H); 6.75 (br d, J = 7.60 Hz, 1H); 6.79−6.89 (m, 2H);

7.17 (d, J = 4.62 Hz, 1H); 8.59 (s, 1H); 8.78 (d, J = 4.62 Hz, 1H).

Final Compound 6, Phenyl-[3-([1,2,4]triazolo[1,5-a]pyrimidin-7-

yl)-1-piperidyl]methanone. Benzoyl chloride (0.03 g, 0.22 mmol) was

added at RT to a solution of intermediate 48a (0.06 g, 0.22 mmol) and

DIPEA (0.084 g, 0.75 g/mL, 0.65 mmol) in DCM (2 mL), and the

resulting mixture was stirred at RT for 1 h. The RM was quenched with

water (1 mL), and then it was ﬁltered on an Extrelut column. The OL

was evaporated in vacuo. This material was puriﬁed by ﬂash column

chromatography on a silica gel, eluting with 0−2% MeOH in DCM, to

give compound 6 (47 mg, 70% yield). MS (ESI) m/z: 308 [M + H]⁺. ¹H

NMR (600 MHz, DMSO-d_6.100 °C): δ ppm 1.64−1.72 (m, 1H);

1.79−1.88 (m, 1H) 2.03 (dtd, J = 13.11, 10.95, 10.95, 3.96 Hz, 1H);

2.21−2.28 (m, 1H); 3.13−3.19 (m, 1H); 3.42 (dd, J = 12.96, 9.99 Hz,

1H); 3.65−3.76 (m, 1H); 3.91−4.03 (m, 1H); 4.31−4.42 (m, 1H);

7.24 (d, J = 4.62 Hz, 1H); 7.34−7.44 (m, 5H); 8.53 (s, 1H); 8.80 (d, J =

4.62 Hz, 1H).

Intermediate 83b, tert-Butyl 3-[5-(Triﬂuoromethyl)-[1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl]piperidine-1-carboxylate. Intermediate

62b (320 mg, 0.87 mmol) and a solution of thiophenol (0.1 mL, 4%,

0.039 mmol) was added to a suspension of Pt/C (5%) (100 mg) in

methanol (40 mL) under N₂ﬂow. The reaction mixture was stirred at

50 °C under a H₂atmosphere until 1 equiv hydrogen was absorbed. The

catalyst was removed by ﬁltration over Celite, and the ﬁltrate was

evaporated. The resulting residue was puriﬁed via ﬂash column

chromatography on a silica gel using a gradient (DCM, 2% MeOH in

DCM) as the eluent to give the title compound 83b (150 mg, 46.6%).

MS (ESI) m/z: 372 [M + H]⁺.

Intermediate 83c, tert-Butyl 3-(2,5-Dimethyl-[1,2,4]triazolo[1,5-

a]pyrimidin-7-yl)piperidine-1-carboxylate. Intermediate 65c (170

mg, 0.52 mmol) and a solution of thiophenol (0.1 mL, 4%, 0.039 mmol)

were added to a suspension of Pt/C (5%) (100 mg) in methanol (40

mL) under N₂ﬂow. The reaction mixture was stirred at 50 °C under a

H₂atmosphere until 1 equiv hydrogen was absorbed. The catalyst was

removed by ﬁltration over Celite, and the ﬁltrate was evaporated. The

resulting residue was dissolved in DCM (10 mL), and then it was

treated with MnO₂(45 mg, 0.52 mmol) and stirred at RT for 1 h. The

reaction mixture was ﬁltered over Celite, and the solvent was

evaporated under reduced pressure to provide the title compound

83c (150 mg, 87.7%), which was used in the next step without further

puriﬁcation.

Intermediate 83d, tert-Butyl 3-[5-Methyl-2-(triﬂuoromethyl)-

[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]piperidine-1-carboxylate. Pre-

pared in a similar way to intermediate 83c, starting from intermediate

65d (150 mg, 0.39 mmol) to give the title compound 83d (100 mg,

66%). MS (ESI) m/z: 386 [M + H]⁺.

Intermediate 48j, 5-Isopropyl-7-(3-piperidyl)-[1,2,4]triazolo[1,5-

a]pyrimidine. To a solution of intermediate 83a (180 mg, 0.52 mmol)

in MeOH (50 mL) was added HCl−iPrOH (50 mL), and the reaction

mixture was stirred for 16 h at RT. The reaction mixture was evaporated

to give the title compound as an HCl salt (160 mg, 96%) MS (ESI) m/z:

246 [M + H]⁺.

Intermediate 48k, 7-(3-Piperidyl)-5-(triﬂuoromethyl)-[1,2,4]-

triazolo[1,5-a]pyrimidine. Prepared in a similar way to 48j, starting

from intermediate 83b (150 mg, 0.40 mmol) to give the title compound

48k as an HCl salt (140 mg, quantitative) MS (ESI) m/z: 272 [M +

H]⁺.

Intermediate 48l, 2,5-Dimethyl-7-(3-piperidyl)-[1,2,4]triazolo-

[1,5-a]pyrimidine. Prepared in a similar way to 48j, starting from

intermediate 83c (150 mg, 0.45 mmol) to give the title compound 48l

as an HCl salt (130 mg, 94%). MS (ESI) m/z: 232 [M + H]⁺.

Intermediate 48m, 5-Methyl-7-(3-piperidyl)-2-(triﬂuoromethyl)-

[1,2,4]triazolo[1,5-a]pyrimidine. Prepared in a similar way to 48j,

starting from intermediate 83d (100 mg, 0.45 mmol) to give the title

compound 48m as an HCl salt (90 mg, 97%). MS (ESI) m/z: 286 [M +

H]⁺.

Intermediate 67, tert-Butyl 5-(1H-Pyrrolo[2,3-b]pyridin-4-yl)-3,4-

dihydro-2H-pyridine-1-carboxylate. Prepared in a similar way to 65a,

starting from intermediate 63 (1.32 g, 4.26 mmol) and commercially

available 4-chloro-7-azaindole (0.5 g, 3.28 mmol). Puriﬁcation via ﬂash

column chromatography on a silica gel using a gradient (DCM, 1%

MeOH in DCM, 2%) as the eluent to give the title compound 67 (180

mg, 18%). MS (ESI) m/z: 300 [M + H]⁺.

Intermediate 84, tert-Butyl 3-(1H-Pyrrolo[2,3-b]pyridin-4-yl)-

piperidine-1-carboxylate. Prepared in a similar way to 83b, starting

from intermediate 67 (180 mg, 0.60 mmol) to give the title compound

84 (180 mg, 99%), which used as such in the next step.

Intermediate 48n, 4-(3-Piperidyl)-1H-pyrrolo[2,3-b]pyridine. Pre-

pared in a similar way to 48a, starting from intermediate 84 (180 mg,

0.59 mmol) to give the title compound 48n as an HCl salt (150 mg,

91.6%), which was used directly in the next step.

Intermediate 69, tert-Butyl 5-[5-(Diﬂuoromethyl)pyrazolo[1,5-

a]pyrimidin-7-yl]-3,4-dihydro-2H-pyridine-1-carboxylate. Prepared

in a similar way to 65a, starting from intermediate 62 (0.59 g, 1.92

mmol) and 7-chloro-5-(diﬂuoromethyl)pyrazolo[1,5-a]pyrimidine

(0.30 g, 1.5 mmol). Puriﬁcation via ﬂash column chromatography on

Final Compound 7, Cyclopropyl-[3-([1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in a similar manner

as compound 5, starting from cyclopropanecarboxylic acid (0.02 g, 0.24

mmol) and intermediate 48a (0.06 g, 0.22 mmol) to give compound 7

Q

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

1

(49 mg, 84% yield). MS (ESI) m/z: 272 [M + H]⁺. H NMR (600

MHz, DMSO-d₆): δ ppm 0.62−0.69 (m, 1H) 0.69−0.75 (m, 2H)

0.76−0.81 (m, 1H) 1.56−1.67 (m, 1H) 1.84 (dt, J = 13.67, 3.90 Hz,

1H) 1.90−1.95 (m, 1H) 2.00 (dtd, J = 13.07, 11.05, 11.05, 4.13 Hz, 1H)

2.22 (dq, J = 12.96, 4.16 Hz, 1H) 3.10−3.15 (m, 1H) 3.38 (br t, J =

11.31 Hz, 2H) 3.55−3.65 (m, 1H) 4.14−4.28 (m, 1H) 4.54 (ddt, J =

13.03, 3.61, 1.59, 1.59 Hz, 1H) 7.24 (d, J = 4.62 Hz, 1H) 8.58 (s, 1H)

8.81 (d, J = 4.62 Hz, 1H).

3.67−3.76 (m, 1H); 3.86−4.06 (m, 1H); 4.18−4.42 (m, 1H); 7.23 (d, J

= 4.46 Hz, 1H); 7.56−7.58 (m, 2H); 7.83 (d, J = 8.26 Hz, 2H); 8.54 (s,

1H); 8.80 (d, J = 4.62 Hz, 1H).

Final Compound 13, [(3S)-([1,2,4]Triazolo[1,5-a]pyrimidin-7-yl)-

1-piperidyl]-[4-(triﬂuoromethyl)phenyl]methanone. Prepared in a

similar manner as compound 8, starting from intermediate 48ab (S!,

0.06 g, 0.22 mmol) and 4-(triﬂuoromethyl)benzoyl chloride (0.045 g,

0.22 mmol). Puriﬁcation by ﬂash chromatography on a silica gel, eluting

with 0−2% MeOH in DCM, aﬀorded compound 9 (65 mg, 80%). MS

(ESI) m/z: 376 [M + H]⁺. SFC [CHIRALCEL Daicel OD 4.6 × 500

mm; mobile phase: CO₂, MeOH with 0.2% 2-propylamine] R_t= 12.68

min. ¹H NMR (600 MHz, DMSO-d_6,100 °C): δ ppm 1.62−1.78 (m,

2H) 1.81−1.91 (m, 1H) 2.04 (dtd, J = 13.13, 10.98, 10.98, 4.05 Hz, 2H)

2.20−2.29 (m, 1H) 3.13−3.23 (m, 1H) 3.42 (dd, J = 12.96, 9.99 Hz,

1H) 3.66−3.76 (m, 1H) 3.85−4.08 (m, 1H) 4.23−4.47 (m, 1H) 7.24

(d, J = 4.46 Hz, 1H) 7.60 (d, J = 7.93 Hz, 2H) 7.74 (d, J = 7.93 Hz, 2H)

8.50 (s, 1H) 8.80 (d, J = 4.62 Hz, 1H).

Final compound 14, 2,3-Dihydro-1,4-benzodioxin-6-yl-[(3S)-

([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. Pre-

pared in a similar manner to compound 11, starting from 2,3-

dihydro-1,4-benzodioxine-6-carboxylic acid (0.100 g, 0.56 mmol) and

intermediate 48ab (S!, 0.184 g, 0.67 mmol). Puriﬁcation via Prep

HPLC (stationary phase: RP Vydac Denali C1810 μm, 250 × 5 cm,

mobile phase: 0.25% NH₄HCO₃solution in water, MeOH) provided

compound 14 (114 mg, 56%). MS (ESI) m/z: 366 [M + H]⁺. SFC

[CHIRALCEL Daicel OJ 4.6 × 500 mm; mobile phase: CO₂, MeOH

with 0.2% 2-propylamine] R_t= 13.5 min. ¹H NMR (360 MHz, DMSO-

d₆): δ ppm 1.67 (br s, 1H) 1.82 (br s, 1H) 1.99 (q, J = 10.49 Hz, 1H)

2.21 (br dd, J = 12.99, 3.48 Hz, 1H) 3.12 (br s, 1H) 3.26−3.32 (m, 1H)

3.62−3.74 (m, 1H) 4.27 (s, 5H) 4.36−4.98 (m, 1H) 6.91 (br d, J = 9.51

Hz, 3H) 7.32 (br s, 1H) 8.68 (br s, 1H) 8.86 (br d, J = 4.39 Hz, 1H).

Final Compound 15, Benzofuran-5-yl-[3-([1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in a similar manner

to compound 11, starting from 1-benzofuran-5-carboxylic acid (0.150

g, 0.93 mmol) and intermediate 48a (0.207 g, 1.02 mmol). Puriﬁcation

by ﬂash chromatography on a silica gel, eluting with 0−2% MeOH in

DCM, followed by crystallization from diethyl ether, aﬀorded

compound 15 (241 mg, 75%). MS (ESI) m/z: 348 [M + H]⁺. mp

216.2 C. ¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.71 (br s, 1H) 2.01

(br d, J = 11.34 Hz, 2H) 2.16−2.28 (m, 1H) 3.17−3.34 (m, 2H) 3.60−

3.82 (m, 1H) 3.87−4.27 (m, 1H) 4.29−5.00 (m, 1H) 7.04 (dd, J = 2.20,

0.73 Hz, 1H) 7.38 (br d, J = 8.42 Hz, 2H) 7.64 (br s, 1H) 7.77 (s, 1H)

8.08 (s, 1H) 8.48−8.77 (m, 1H) 8.85 (br s, 1H).

Final Compound 16. Step 1, intermediate 16a, 1H-indol-5-yl-[(3S)-

([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. Inter-

mediate 16a was prepared in a similar manner to compound 11,

starting from indole-5-carboxylic acid (0.4 g, 2.48 mmol) and

intermediate 48ab (S!, 0.823 g, 2.98 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−2% MeOH in DCM,

aﬀorded intermediate 16a (800 mg, 93%). MS (ESI) m/z: 347 [M +

H]⁺. SFC [CHIRALCEL Daicel OJ 4.6 × 500 mm; mobile phase: CO₂,

MeOH with 0.2% 2-propylamine] R_t= 17.37 min. ¹H NMR (360 MHz,

DMSO-d₆): δ ppm 1.68 (br d, J = 12.44 Hz, 1H) 1.82 (br s, 1H) 1.88−

2.07 (m, 1H) 2.23 (br d, J = 9.51 Hz, 1H) 3.13 (br s, 1H) 3.25−3.36 (m,

2H) 3.63−3.79 (m, 1H) 4.02 (br s, 1H) 4.28−4.75 (m, 1H) 6.52 (br s,

1H) 7.16 (br d, J = 8.05 Hz, 1H) 7.28−7.37 (m, 1H) 7.37−7.47 (m,

2H) 7.69 (s, 1H) 8.67 (br s, 1H) 8.85 (br d, J = 4.02 Hz, 1H) 11.29 (br

s, 1H).

Final Compound 16, 1-[5-[(3S)-([1,2,4]Triazolo[1,5-a]pyrimidin-

7-yl)piperidine-1-carbonyl]indol-1-yl]ethenone. NaH (0.012 mg,

0.29 mmol, 60% dispersion in mineral oil) was added at RT to a

solution of intermediate 16a (100 mg, 0.29 mmol) in DMF (5 mL), and

the reaction mixture was stirred at RT for 30 min. After this time, acetyl

chloride (0.023 mg, 0.29 mmol) was added to the reaction mixture,

which was stirred at RT for another 30 min before being quenched with

water. The mixture was extracted with DCM, and then the organic layer

was dried over MgSO₄, ﬁltered, and evaporated under reduced pressure.

The resulting residue was puriﬁed via Prep HPLC (stationary phase: RP

Vydac Denali C1810 μm, 250 × 5 cm, mobile phase: 0.25%

NH₄HCO₃solution in water, MeOH), which provided compound 16

Final Compound 8, Cyclohexyl-[3-([1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]methanone. Cyclohexanecarbonyl chlor-

ide (0.116 mL, 0.87 mmol) was added at RT to a solution of

intermediate 48a (250 mg, 0.87 mmol) and DIPEA (0.574 mL, 0.75 g/

mL, 3.47 mmol), and the resulting mixture was stirred at RT for 1 h.

The reaction mixture was quenched with water (1 mL) and saturated

aqueous solution of Na₂CO₃(1 mL), and then the mixture was ﬁltered

through an Extrelut ﬁlter. The organic layer was evaporated under

reduced pressure, and the resulting residue was puriﬁed via Prep HPLC

(stationary phase: RP Vydac Denali C1810 μm, 250 × 5 cm, mobile

phase: 0.25% NH₄HCO₃solution in water, MeOH) to aﬀord

compound 8 (30 mg, 11%). MS (ESI) m/z: 312 [M − H]⁺. H

1

NMR (360 MHz, DMSO-d₆): δ ppm 1.15 (br s, 1H) 1.28−1.39 (m,

2H) 1.47 (br s, 1H) 1.58−1.73 (m, 4H) 1.75−1.85 (m, 1H) 1.87−2.02

(m, 2H) 2.17 (br d, J = 12.81 Hz, 1H) 2.52−2.83 (m, 2H) 3.04−3.32

(m, 3H) 3.51 (br d, J = 11.34 Hz, 1H) 3.72−4.01 (m, 1H) 4.39 (br d, J =

7.68 Hz, 1H) 4.47−4.66 (m, 1H) 7.31 (dd, J = 16.10, 4.39 Hz, 1H) 8.72

(d, J = 13.54 Hz, 1H) 8.86 (dd, J = 10.25, 4.39 Hz, 1H).

Final Compound 9, (2-Methoxyphenyl)-[3(S)-([1,2,4]triazolo[1,5-

a]pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in a similar

manner as compound 8, starting from intermediate 48ab (S!, 0.20 g,

0.72 mmol) and 2-methoxybenzoyl chloride (0.281 g, 2.17 mmol).

Puriﬁcation by ﬂash chromatography on a silica gel, eluting with 0−2%

MeOH in DCM, aﬀorded compound 9 (167 mg, 68%). MS (ESI) m/z:

338 [M + H]⁺. (DMSO-d₆): δ ppm 1.49 (br d, J = 9.51 Hz, 1H) 1.56−

1.73 (m, 2H) 1.87−2.05 (m, 3H) 2.16 (br s, 2H) 2.96−3.21 (m, 2H)

3.25−3.33 (m, 2H) 3.36−3.46 (m, 1H) 3.50−3.75 (m, 5H) 3.81 (s,

1H) 3.85 (s, 3H) 4.37−4.49 (m, 1H) 4.54 (br d, J = 13.17 Hz, 1H) 4.74

(br d, J = 11.34 Hz, 1H) 6.77−6.83 (m, 1H) 6.87−7.18 (m, 5H) 7.24−

7.33 (m, 3H) 7.33−7.44 (m, 2H) 8.49 (s, 1H) 8.74 (d, J = 1.83 Hz, 1H)

8.80 (d, J = 4.76 Hz, 1H) 8.90 (d, J = 4.39 Hz, 1H).

Final Compound 10. Prepared in a similar manner as compound 8,

starting from intermediate 48ab (S!, 1 equiv) and 3-methoxybenzoyl

chloride (1 equiv). MS (ESI) m/z: 338 [M + H]⁺; ¹H NMR (600 MHz,

DMSO-d_6,100 °C): δ ppm 1.64−1.73 (m, 1H) 1.79−1.87 (m, 1H)

1.96−2.08 (m, 1H) 2.18−2.28 (m, 1H) 3.11−3.21 (m, 1H) 3.41 (dd, J

= 12.96, 9.99 Hz, 1H) 3.71 (tt, J = 10.28, 3.92 Hz, 1H) 3.79 (s, 3H)

3.93−4.04 (m, 1H) 4.30−4.42 (m, 1H) 6.90−6.93 (m, 2H) 6.95−6.98

(m, 1H) 7.24 (d, J = 4.46 Hz, 1H) 7.27−7.33 (m, 1H) 8.52 (s, 1H) 8.80

(d, J = 4.62 Hz, 1H).

Final Compound 11, (4-Methoxyphenyl)-[3(S)-([1,2,4]triazolo-

[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in a similar

manner to compound 4, starting from 4-methoxybenzoic acid (0.100 g,

0.66 mmol) and intermediate 48ab (S!, 0.218 g, 0.79 mmol) and using

DCM as the solvent. Puriﬁcation via Prep HPLC (stationary phase: RP

Vydac Denali C1810 μm, 250 × 5 cm, mobile phase: 0.25%

NH₄HCO₃solution in water, MeOH) provided compound 11 (180

mg, 81%). MS (ESI) m/z: 338 [M + H]⁺. ¹H NMR (600 MHz, DMSO-

d₆): δ ppm 1.56−1.90 (m, 2H); 1.99 (m, 1H); 2.18−2.25 (m, 1H);

2.93−3.27 (m, 2H); 3.64−3.74 (m, 1H); 3.79 (s, 3H); 4.03−4.83 (m,

2H); 6.97 (br s, 2H); 7.32 (br s, 1H); 7.39 (br d, J = 7.93 Hz, 2H); 8.68

(br s, 1H); 8.85 (br s, 1H).

Final Compound 12, 4-[3(S)-([1,2,4]Triazolo[1,5-a]pyrimidin-7-

yl)piperidine-1-carbonyl]benzonitrile. Prepared in a similar manner to

compound 11, starting from 4-cyanobenzoic acid (0.100 g, 0.68 mmol)

and intermediate 48ab (S!, 0.225 g, 0.82 mmol). Puriﬁcation via Prep

HPLC (stationary phase: RP Vydac Denali C1810 μm, 250 × 5 cm,

mobile phase: 0.25% NH₄HCO₃solution in water, MeOH) provided

compound 12 (35 mg, 15%). MS (ESI) m/z: 333 [M + H]⁺. ¹H NMR

(600 MHz, DMSO-d_6,100 °C): δ ppm 1.65−1.75 (m, 1H); 1.81−1.90

(m, 1H); 2.04 (dtd, J = 13.11, 11.03, 11.03, 3.96 Hz, 1H); 2.21−2.27

(m, 1H); 3.13−3.20 (m, 1H); 3.41 (dd, J = 13.05, 10.07 Hz, 1H);

R

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(82 mg, 73%). MS (ESI) m/z: 389 [M + H]⁺. SFC [CHIRALCEL

Daicel OJ 4.6 × 500 mm; mobile phase: CO₂, EtOH with 0.2% 2-

propylamine] R_t= 18.02 min and R_t= 18.4 min. ¹H NMR (360 MHz,

DMSO-d₆): δ ppm 1.73 (m, 1H) 1.94−2.07 (m, 1H) 2.23 (br d, J = 9.15

Hz, 1H) 2.67 (s, 3H) 3.19 (br s, 1H) 3.24−3.33 (m, 1H) 3.56−3.82 (m,

2H) 3.91−4.22 (m, 1H) 4.36−4.95 (m, 1H) 6.82 (d, J = 3.66 Hz, 1H)

7.37 (br s, 2H) 7.71−7.75 (m, 1H) 7.94 (br d, J = 3.29 Hz, 1H) 8.35 (br

s, 1H) 8.72 (br s, 1H) 8.85 (br s, 1H).

Final Compound 17, (1-Methylsulfonylindol-5-yl)-[3-([1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in

a similar manner to compound 11, starting from 1-methanesulfonyl-

2,3-dihydro-1H-indole-5-carboxylic acid (0.500 g, 2.07 mmol) and

intermediate 48a (0.687 g, 2.49 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−2% MeOH in DCM,

followed by crystallization from diethyl ether, aﬀorded compound 17

(511 mg, 58%). MS (ESI) m/z: 427 [M + H]⁺. mp 238.2 C. SFC

[CHIRALCEL Daicel OJ 4.6 × 500 mm; mobile phase: CO₂, MeOH

with 0.2% 2-propylamine] R_t= 17.87 min and R_t= 20.28 min. ¹H NMR

(360 MHz, DMSO-d₆): δ ppm 1.68 (br s, 1H) 1.76−1.92 (m, 1H) 2.02

(br d, J = 9.51 Hz, 1H) 2.22 (br dd, J = 12.99, 3.48 Hz, 1H) 3.05 (s, 3H)

3.14 (br t, J = 8.05 Hz, 3H) 3.37−3.52 (m, 1H) 3.62−3.75 (m, 1H) 3.98

(br t, J = 8.60 Hz, 2H) 4.18 (br s, 1H) 4.67 (br s, 1H) 7.04−7.31 (m,

3H) 7.33 (br s, 1H) 8.64 (br d, J = 11.34 Hz, 1H) 8.86 (br d, J = 4.03

Hz, 1H).

Final Compound 18, 4,5,6,7-Tetrahydrobenzothiophen-2-yl-

[(3S)-([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone.

Prepared in a similar manner to compound 11, starting from 4,5,6,7-

tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.100 g, 0.55 mmol)

and intermediate 48ab (0.687 g, 2.49 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−2% MeOH in DCM,

followed by crystallization from diethyl ether, aﬀorded compound 17

(511 mg, 58%). MS (ESI) m/z: 427 [M + H]⁺. mp 238.2 C. SFC

[CHIRALCEL Daicel OJ 4.6 × 500 mm; mobile phase: CO₂, EtOH

with 0.2% 2-propylamine] R_t= 17.87 min and R_t= 20.28 min. ¹H NMR

(360 MHz, DMSO-d₆): δ ppm 1.68 (br s, 1H) 1.76−1.92 (m, 1H) 2.02

(br d, J = 9.51 Hz, 1H) 2.22 (br dd, J = 12.99, 3.48 Hz, 1H) 3.05 (s, 3H)

3.14 (br t, J = 8.05 Hz, 3H) 3.37−3.52 (m, 1H) 3.62−3.75 (m, 1H) 3.98

(br t, J = 8.60 Hz, 2H) 4.18 (br s, 1H) 4.67 (br s, 1H) 7.04−7.31 (m,

3H) 7.33 (br s, 1H) 8.64 (br d, J = 11.34 Hz, 1H) 8.86 (br d, J = 4.03

Hz, 1H).

Final Compound 21, 4,5,6,7-Tetrahydrobenzothiophen-2-yl-

[(2S)-([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)morpholin-4-yl]-

methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.175 g,

0.96 mmol) and intermediate 48bb (S! 0.32 g, 1.15 mmol). Puriﬁcation

by ﬂash chromatography on a silica gel, eluting with 0−2% MeOH in

DCM, followed by SFC puriﬁcation [CHIRALCEL Daicel OJ 20 × 250

mm; mobile phase: CO₂, MeOH with 0.2% 2-propylamine] provided

compound 21 (205 mg, 58%). R_t= 8.06 min. (ESI) m/z: 370 [M + H]⁺.

¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.78 (br s, 4H) 2.63 (br s, 2H)

2.74 (br s, 2H) 3.08 (br s, 1H) 3.13−3.21 (m, 1H) 3.88 (br t, J = 11.16

Hz, 1H) 4.05−4.22 (m, 1H) 4.27 (br d, J = 13.17 Hz, 1H) 4.99 (br d, J =

12.44 Hz, 1H) 5.35 (br d, J = 9.51 Hz, 1H) 7.40 (br d, J = 4.03 Hz, 2H)

8.78 (s, 1H) 8.94 (br d, J = 4.02 Hz, 1H).

Final Compound 22, 4,5,6,7-Tetrahydrobenzothiophen-2-yl-[3-

([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)piperazin-1-yl]methanone.

Prepared in a similar manner to compound 11, starting from 4,5,6,7-

tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.484 g, 2.66 mmol)

and intermediate 48d (1.00 g, 3.19 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−6% MeOH in DCM,

followed by crystallization from EtOAc, provided compound 22 (707

1

mg, 72%). (ESI) m/z: 369 [M + H]⁺. mp 193.45 °C. H NMR (400

MHz, DMSO-d₆): δ ppm 1.72−1.83 (m, 4H) 2.57 (br t, J = 6.06 Hz,

2H) 2.71 (br t, J = 5.65 Hz, 3H) 2.89−2.98 (m, 1H) 3.13 (dt, J = 12.41,

3.68 Hz, 1H) 3.32 (ddd, J = 13.02, 9.79, 3.43 Hz, 1H) 3.39 (dd, J =

12.92, 8.48 Hz, 1H) 4.02−4.10 (m, 1H) 4.57 (ddd, J = 12.92, 3.63, 1.21

Hz, 1H) 4.67 (dd, J = 8.48, 3.63 Hz, 1H) 7.06 (s, 1H) 7.38 (d, J = 4.44

Hz, 1H) 8.52 (s, 1H) 8.82 (d, J = 4.44 Hz, 1H).

Final Compound 23, [4-Methyl-3-([1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)piperazin-1-yl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. A solution of compound 22 (0.2 g, 0.54 mmol),

paraformaldehyde (0.020 g, 0.66 mmol), and NaBH(OAc)₃(0.69 g,

3.25 mmol) in DCM (20 mL) was stirred at RT for 32 h. The mixture

was then ﬁltered through Celite, and the ﬁltrate was washed with water.

The organic layer was ﬁltered over Extrelut, and the ﬁltrate was

evaporated under reduced pressure. The resulting residue was puriﬁed

by recrystallization from DIPE to provide compound 23 (192 mg,

92%). (ESI) m/z: 383 [M + H]⁺. mp 174.6 °C. ¹H NMR (360 MHz,

DMSO-d₆): δ ppm 1.68−1.81 (m, 4H) 2.19 (s, 3H) 2.35−2.46 (m, 1H)

2.56−2.62 (m, 2H) 2.71 (br s, 2H) 3.10 (br d, J = 11.71 Hz, 2H) 3.21−

3.32 (m, 1H) 4.19 (dd, J = 9.88, 2.93 Hz, 1H) 4.28 (br d, J = 12.81 Hz,

1H) 4.51 (br d, J = 13.17 Hz, 1H) 7.28 (s, 1H) 7.41 (d, J = 4.39 Hz, 1H)

8.76 (s, 1H) 8.91 (d, J = 4.39 Hz, 1H).

Final Compound 24, 1-[4-(4,5,6,7-Tetrahydrobenzothiophene-2-

carbonyl)-2-([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)piperazin-1-yl]-

ethenone. Acetyl chloride (0.07 g, 0.89 mmol) was added at 0 °C to a

solution of compound 22 (0.30 g, 0.81 mmol) and DIPEA (0.127 g,

0.98 mmol, 0.75 g/mL) in DCM (50 mL). The reaction mixture was

allowed to reach RT and stirred until the starting material was

consumed, as indicated by LC/MS and TLC. The reaction mixture was

quenched with a saturated aqueous NaHCO₃solution, and then the

organic layer was separated, dried over MgSO₄, ﬁltered, and

concentrated under reduced pressure. The resulting residue was

puriﬁed by ﬂash chromatography on a silica gel, eluting with 0−2%

MeOH in DCM, and the resulting material triturated in diethyl ether to

provide compound 24 (293 mg, 87.6%). (ESI) m/z: 411 [M + H]⁺. mp

186 °C. ¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.61−1.78 (m, 5H)

1.98 (s, 1H) 2.20 (s, 3H) 2.37−2.47 (m, 2H) 2.52−2.66 (m, 2H) 3.37−

3.47 (m, 1H) 3.85−3.99 (m, 1H) 3.99−4.09 (m, 2H) 4.12−4.22 (m,

1H) 4.80 (br s, 1H) 5.01 (br s, 1H) 5.66−5.75 (m, 1H) 5.81 (br s, 1H)

6.21−6.98 (m, 1H) 7.25 (br d, J = 4.03 Hz, 1H) 7.35 (d, J = 4.39 Hz,

1H) 8.71 (br s, 1H) 8.81−8.88 (m, 1H) (80/20 mixture of rotamers).

Final Compound 25, 4,5,6,7-Tetrahydrobenzothiophen-2-yl-[3-

([1,2,4]triazolo[1,5-a]pyrimidin-7-yl)azepan-1-yl]methanone. Pre-

pared in a similar manner to compound 5, starting from 4,5,6,7-

tetrahydrobenzo[b]thiophene-2-carboxylic acid [40133-07-1] (1.1

equiv) and intermediate 48c (1 equiv). (ESI) m/z: 382 [M + H]⁺.

¹H NMR (600 MHz, DMSO-d_6,100 °C, mixture of rotamers): δ ppm

Final Compound 19, (1-Methylsulfonylindol-5-yl)-[(3R)-3-([1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in a

similar manner to compound 18, starting from 1-methanesulfonyl-2,3-

dihydro-1H-indole-5-carboxylic acid (0.145 g, 0.60 mmol) and

intermediate 48aa (R!, 0.2 g, 0.72 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−4% MeOH in DCM

aﬀorded compound 19 (120 mg, 46.6%). MS (ESI). SFC

[CHIRALCEL Daicel OJ 4.6 × 500 mm; mobile phase: CO₂, MeOH

with 0.2% 2-propylamine] R_t= 20.21 min. OR = −10.5° (589 nm, 20

1

°C, 0.438 w/v %; MeOH). H NMR (360 MHz, DMSO-d₆): δ ppm

1.68 (br s, 1H) 1.76−1.92 (m, 1H) 2.02 (br d, J = 9.51 Hz, 1H) 2.22 (br

dd, J = 12.99, 3.48 Hz, 1H) 3.05 (s, 3H) 3.14 (br t, J = 8.05 Hz, 3H)

3.37−3.52 (m, 1H) 3.62−3.75 (m, 1H) 3.98 (br t, J = 8.60 Hz, 2H) 4.18

(br s, 1H) 4.67 (br s, 1H) 7.04−7.31 (m, 3H) 7.33 (br s, 1H) 8.64 (br d,

J = 11.34 Hz, 1H) 8.86 (br d, J = 4.03 Hz, 1H).

Final Compound 20, (1-Methylsulfonylindol-5-yl)-[(3S)-3-([1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl)-1-piperidyl]methanone. Prepared in a

similar manner to compound 18, starting from 1-methanesulfonyl-2,3-

dihydro-1H-indole-5-carboxylic acid (0.145 g, 0.60 mmol) and

intermediate 48ab (S!, 0.2 g, 0.72 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−4% MeOH in DCM,

aﬀorded compound 20 (128 mg, 46.6%). MS (ESI). SFC

[CHIRALCEL Daicel OJ 4.6 × 500 mm; mobile phase: CO₂, MeOH

with 0.2% 2-propylamine] R_t= 17.88 min. OR = + 9.36° (589 nm, 20

1

°C, 0.417 w/v %; MeOH). H NMR (360 MHz, DMSO-d₆): δ ppm

1.68 (br s, 1H) 1.76−1.92 (m, 1H) 2.02 (br d, J = 9.51 Hz, 1H) 2.22 (br

dd, J = 12.99, 3.48 Hz, 1H) 3.05 (s, 3H) 3.14 (br t, J = 8.05 Hz, 3H)

3.37−3.52 (m, 1H) 3.62−3.75 (m, 1H) 3.98 (br t, J = 8.60 Hz, 2H) 4.18

(br s, 1H) 4.67 (br s, 1H) 7.04−7.31 (m, 3H) 7.33 (br s, 1H) 8.64 (br d,

J = 11.34 Hz, 1H) 8.86 (br d, J = 4.03 Hz, 1H).

1.69−1.87 (m, 6H) 1.90−2.00 (m, 3H) 2.06−2.14 (m, 1H) 2.66−2.78

(m, 4H) 3.68−3.74 (m, 1H) 3.80−3.86 (m, 1H) 3.89−3.99 (m, 2H)

S

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

4.17−4.23 (m, 1H) 7.06 (s, 1H) 7.22−7.26 (m, 1H) 8.56−8.58 (m,

1H) 8.76−8.85 (m, 1H).

prep HPLC on a RP Vydac Denali column (C1810 μm, 250 g, 5 cm)

and using as the mobile phase a mixture of a 0.25% NH₄HCO₃solution

in water and MeOH aﬀorded compound 31 (130 mg, 95%). ESI m/z:

418 [M + H]⁺. ¹H NMR (400 MHz, DMSO-d₆): δ ppm 1.68−1.82 (m,

5H) 1.88 (dt, J = 13.52, 3.73 Hz, 1H) 2.06−2.17 (m, 1H) 2.21−2.28

(m, 1H) 2.56 (t, J = 5.85 Hz, 2H) 2.71 (t, J = 5.85 Hz, 2H) 3.19−3.28

(m, 1H) 3.54 (dd, J = 13.12, 9.89 Hz, 1H) 3.75−3.82 (m, 1H) 4.23 (br

d, J = 13.32 Hz, 1H) 4.56 (dd, J = 13.32, 3.63 Hz, 1H) 7.03 (t, J = 54.50

Hz, 1H) 7.13 (s, 1H) 7.51 (s, 1H) 8.75 (s, 1H).

Final Compound 32, 3-[5-(Diﬂuoromethyl)pyrazolo[1,5-a]-

pyrimidin-7-yl]-1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.108 g,

0.59 mmol) and intermediate 48o (0.150 g, 0.59 mmol). Puriﬁcation by

ﬂash chromatography on a silica gel, eluting with 0−30% EtOAc in

heptane, followed by crystallization from diisopropyl ether, aﬀorded

compound 32 (190 mg, 77%). ESI m/z 417 [M + H]⁺. mp 192.8 °C. ¹H

NMR (600 MHz, DMSO-d₆): δ ppm 1.64−1.81 (m, 5H) 1.84−1.91

(m, 1H) 2.05−2.12 (m, 1H) 2.21−2.26 (m, 1H) 2.54−2.61 (m, 2H)

2.67−2.75 (m, 2H) 3.17−3.23 (m, 1H) 3.46 (dd, J = 12.76, 10.12 Hz,

1H) 3.83−3.88 (m, 1H) 4.26 (br d, J = 13.21 Hz, 1H) 4.57 (dd, J =

12.76, 3.67 Hz, 1H) 6.90 (t, J = 54.87 Hz, 1H) 6.87−6.90 (m, 1H) 7.19

(d, J = 8.80 Hz, 2H) 8.34 (dd, J = 2.35, 0.88 Hz, 1H).

Final Compound 33, [3-(7-Methylimidazo[1,2-a]pyrimidin-5-yl)-

1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-yl)methanone.

Prepared in a similar manner to compound 11, starting from 4,5,6,7-

tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.131 g, 0.72 mmol)

and intermediate 48h (0.250 g, 0.86 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−3% MeOH in DCM,

followed by crystallization from diethyl ether, aﬀorded compound 33

(168 mg, 61%). ESI m/z: 381 [M + H]⁺. mp 192.7 °C. ¹H NMR (360

MHz, DMSO-d₆): δ ppm 1.63−1.91 (m, 8H); 2.17 (br d, J = 12.44 Hz,

1H); 2.52 (s, 3H); 2.53−2.59 (m, 2H); 2.64−2.81 (m, 2H); 3.22 (br d,

J = 16.10 Hz, 2H); 3.27−3.31 (m, 1H); 4.32 (br d, J = 12.81 Hz, 1H);

4.51 (br d, J = 11.34 Hz, 1H); 6.90 (s, 1H); 7.14 (s, 1H); 7.67 (s, 1H);

8.01 (br s, 1H).

Final Compound 26, [3-(5-Methyl-[1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.100 g,

0.55 mmol) and intermediate 48g (0.687 g, 2.49 mmol). Puriﬁcation by

ﬂash chromatography on a silica gel, eluting with 0−2% MeOH in

DCM, followed by crystallization from diethyl ether, aﬀorded

compound 26 (511 mg, 58%). (ESI) m/z: 382 [M + H]⁺. mp 201.2

°C. ¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.60−1.81 (m, 6H) 1.88

(br d, J = 13.54 Hz, 1H) 1.99 (br d, J = 12.08 Hz, 1H) 2.20 (br d, J =

13.17 Hz, 1H) 2.58 (br t, J = 5.67 Hz, 2H) 2.62 (s, 3H) 2.70 (br d, J =

5.12 Hz, 2H) 3.11 (br s, 1H) 3.58−3.67 (m, 1H) 4.29 (br d, J = 12.08

Hz, 1H) 4.60 (br d, J = 12.81 Hz, 1H) 7.26 (s, 2H) 8.60 (s, 1H).

Final Compound 27, [3-(5-Isopropyl-[1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.076 g,

0.42 mmol) and intermediate 48j (0.160 g, 0.5 mmol). Puriﬁcation by

ﬂash chromatography on a silica gel, eluting with 0.5−1% MeOH in

DCM, followed by crystallization from diisopropyl ether, aﬀorded

compound 27 (54 mg, 31%). ESI m/z: 410 [M + H]⁺. mp 158 °C. ¹H

NMR (360 MHz, DMSO-d₆): δ ppm 1.30 (d, J = 6.95 Hz, 6H) 1.74 (br

s, 5H) 1.96−2.31 (m, 3H) 2.54−2.59 (m, 2H) 2.70 (br d, J = 4.76 Hz,

2H) 3.11−3.24 (m, 2H) 4.11−4.36 (m, 1H) 4.45−4.63 (m, 1H) 7.24

(s, 1H) 7.28 (s, 1H) 8.62 (s, 1H).

Final Compound 28, [3-(5-Methoxy-[1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.101 g,

0.56 mmol) and intermediate 48e (0.130 g, 0.56 mmol). Puriﬁcation by

ﬂash chromatography on a silica gel, eluting with 0.5−1% MeOH in

DCM, followed by crystallization from diethyl ether, aﬀorded

compound 28 (86 mg, 39%). ESI m/z: 398 [M + H]⁺. mp 165.5 °C.

¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.58−1.88 (m, 7H) 1.91−2.06

(m, 1H) 2.16 (br d, J = 12.81 Hz, 1H) 2.53−2.61 (m, 2H) 2.65−2.76

(m, 2H) 3.10 (br s, 1H) 3.20−3.30 (m, 1H) 3.51−3.60 (m, 1H) 3.99 (s,

3H) 4.26 (br d, J = 12.08 Hz, 1H) 4.55 (br d, J = 13.17 Hz, 1H) 6.80 (s,

1H) 7.25 (br s, 1H) 8.46 (s, 1H).

Final Compound 34, [3-(1H-Pyrrolo[2,3-b]pyridin-4-yl)-1-piper-

idyl]-(4,5,6,7-tetrahydrobenzothiophen-2-yl)methanone. Prepared

in a similar manner to compound 11, starting from 4,5,6,7-

tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.083 g, 0.46 mmol)

and intermediate 48n (0.150 g, 0.55 mmol). Puriﬁcation by

crystallization from methanol aﬀorded compound 34 (114 mg, 68%).

ESI m/z: 366 [M + H]⁺. mp 275.7 °C. ¹H NMR (360 MHz, DMSO-

d₆): δ ppm 1.59−1.87 (m, 6H); 1.89−2.07 (m, 2H); 2.52−2.58 (m,

2H); 2.67−2.75 (m, 2H); 3.02−3.32 (m, 3H); 4.37 (br d, J = 11.71 Hz,

2H); 6.57 (br s, 1H); 6.97 (d, J = 5.12 Hz, 1H); 7.10 (s, 1H); 7.44 (t, J =

3.11 Hz, 1H); 8.15 (d, J = 4.76 Hz, 1H); 11.65 (br s, 1H).

Final Compound 35, [3-(2,5-Dimethyl-[1,2,4]triazolo[1,5-a]-

pyrimidin-7-yl)-1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.065 g,

0.36 mmol) and intermediate 48l (0.130 g, 0.43 mmol). Puriﬁcation by

ﬂash chromatography on a silica gel, eluting with 0.5−1% MeOH in

DCM, followed by crystallization from diisopropyl ether, aﬀorded

compound 35 (12 mg, 39%). ESI m/z: 396 [M + H]⁺. ¹H NMR (600

MHz, DMSO-d₆): δ ppm 1.62−1.70 (m, 1H) 1.81 (br s, 6H) 1.86−1.91

(m, 1H) 1.96−2.04 (m, 1H) 2.17−2.24 (m, 1H) 2.46 (s, 3H) 2.54−

2.56 (m, 2H) 2.57 (s, 3H) 2.71 (br t, J = 6.19 Hz, 2H) 3.19 (ddd, J =

13.33, 11.11, 3.30 Hz, 1H) 3.43 (dd, J = 13.05, 10.07 Hz, 1H) 3.59 (tt, J

= 10.30, 3.90 Hz, 1H) 4.21 (dt, J = 13.29, 4.00 Hz, 1H) 4.53 (ddt, J =

12.98, 3.65, 1.59, 1.59 Hz, 1H) 7.04 (s, 1H) 7.05 (s, 1H).

Final Compound 36, [3-[5-Methyl-2-(triﬂuoromethyl)-[1,2,4]-

triazolo[1,5-a]pyrimidin-7-yl]-1-piperidyl]-(4,5,6,7-tetrahydroben-

zothiophen-2-yl)methanone. Prepared in a similar manner to

compound 11, starting from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-

carboxylic acid (0.038 g, 0.21 mmol) and intermediate 48m (0.09 g,

0.25 mmol). Puriﬁcation by ﬂash chromatography on a silica gel, eluting

with 0.5−1% MeOH in DCM, aﬀorded compound 36 (75 mg, 78%).

ESI m/z: 450 [M + H]⁺. ¹H NMR (400 MHz, DMSO-d₆): δ ppm 1.63−

1.81 (m, 5H); 1.85−2.06 (m, 2H); 2.19−2.26 (m, 1H); 2.50−2.55 (m,

2H); 2.64−2.72 (m, 5H); 3.16−3.26 (m, 1H); 3.44 (dd, J = 12.9, 10.1

Final Compound 29, 7-[1-(4,5,6,7-Tetrahydrobenzothiophene-2-

carbonyl)-3-piperidyl]-[1,2,4]triazolo[1,5-a]pyrimidine-5-carboni-

trile. Prepared in a similar manner to compound 11, starting from

4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.157 g, 0.86

mmol) and intermediate 48f (0.135 g, 1.04 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−1.5% MeOH in DCM,

followed by crystallization from diethyl ether, aﬀorded compound 29

(73 mg, 21%). ESI m/z: 393 [M + H]⁺. ¹H NMR (360 MHz, DMSO-

d₆): δ ppm 1.58−1.82 (m, 5H) 1.89 (br d, J = 13.17 Hz, 1H) 1.97−2.14

(m, 1H) 2.20 (br d, J = 13.17 Hz, 1H) 2.52−2.61 (m, 2H) 2.70 (br d, J =

5.85 Hz, 2H) 3.11 (br s, 1H) 3.25−3.32 (m, 1H) 3.69−3.79 (m, 1H)

4.31 (br d, J = 12.44 Hz, 1H) 4.61 (br d, J = 12.81 Hz, 1H) 7.23 (s, 1H)

8.00 (s, 1H) 8.99 (s, 1H).

Final Compound 30, 4,5,6,7-Tetrahydrobenzothiophen-2-yl-[3-

[5-(triﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]-1-

piperidyl]methanone. Prepared in a similar manner to compound 11,

starting from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid

(0.062 g, 0.34 mmol) and intermediate 48k (0.140 g, 0.41 mmol).

Puriﬁcation by ﬂash chromatography on a silica gel, eluting with 0−

0.5% MeOH in DCM, aﬀorded compound 30 (110 mg, 74%). ESI m/z:

436 [M + H]⁺. ¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.04 (d, J =

6.22 Hz, 1H) 1.61−1.80 (m, 5H) 1.87 (br d, J = 12.81 Hz, 1H) 2.10−

2.25 (m, 2H) 2.52−2.61 (m, 3H) 2.65−2.76 (m, 2H) 3.02−3.24 (m,

1H) 3.36−3.53 (m, 1H) 3.75−3.84 (m, 1H) 4.30 (br d, J = 13.90 Hz,

1H) 4.59 (br d, J = 10.25 Hz, 1H) 7.24 (s, 1H) 7.80 (s, 1H) 8.96 (s,

1H).

Final Compound 31, [3-[5-(Diﬂuoromethyl)-[1,2,4]triazolo[1,5-

a]pyrimidin-7-yl]-1-piperidyl]-(4,5,6,7-tetrahydrobenzothiophen-2-

yl)methanone. Prepared in a similar manner to compound 11, starting

from 4,5,6,7-tetrahydrobenzo[b]thiophene-2-carboxylic acid (0.060 g,

0.33 mmol) and intermediate 48i (0.100 g, 0.39 mmol). Puriﬁcation by

T

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Hz, 1H); 3.57−3.67 (m, 1H); 4.18−4.25 (m, 1H); 4.52−4.57 (m, 1H);

7.02 (s, 1H); 7.38 (s, 1H).

DMF), mp 153.3 °C, ¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.62−

1.73 (m, 2H); 1.74−1.88 (m, 1H); 2.04 (br dd, J = 12.62, 4.21 Hz, 1H);

2.37 (br s, 1H); 2.44 (s, 3H); 2.53−2.71 (m, 2H); 2.99 (br d, J = 10.98

Hz, 1H); 3.54−3.62 (m, 2H); 3.82 (dt, J = 8.51, 4.35 Hz, 1H); 7.04 (t, J

= 55.30 Hz, 1H); 7.26 (d, J = 6.59 Hz, 2H); 7.70 (s, 1H); 8.82 (s, 1H)

and 42 (S*) (348 mg, 33%). R_t= 4.66 min, OR = −48.46° (589 nm, 20

°C, 0.43 w/v %; DMF), mp 154.4 °C.

Final Compound 37, 5-(Diﬂuoromethyl)-7-[1-[(3,4-

dimethoxyphenyl)methyl]-3-piperidyl]-[1,2,4]triazolo[1,5-a]-

pyrimidine. DIPEA (0.52 mL, 3.02 mmol) and 3,4-dimethoxybenzal-

dehyde [120-14-9] (0.076 g, 0.46 mmol) were added to a solution of

intermediate 48i (0.15 g, 0.46 mmol) in DCM (30 mL). The reaction

mixture was stirred at RT for 1 h, and then sodiumtriacetoxyborohy-

dride [56553-60-7] (0.15 g, 1.84) was added. The reaction mixture was

stirred at RT for 2 h, and then it was washed with sat. aq NaHCO₃

solution, dried over MgSO₄, ﬁltered, and concentrated by evaporation.

The resulting residue was puriﬁed by ﬂash column chromatography on

a silica gel using a mixture of EtOAc and heptane, 80/20, as the eluent

to provide compound 37 (160 mg, 86%). ESI m/z: 404 [M + H]⁺. ¹H

NMR (400 MHz, DMSO-d₆, 80 °C): δ ppm 1.60−1.71 (m, 2H); 1.79−

1.89 (m, 1H); 2.00−2.08 (m, 1H); 2.35−2.43 (m, 1H); 2.59−2.68 (m,

2H); 2.96−3.01 (m, 1H); 3.50 (s, 2H); 3.73 (s, 3H); 3.76 (s, 3H);

3.78−3.88 (m, 1H); 6.83 (dd, J = 8.1, 1.61 Hz, 1H); 6.85−6.89 (d, J =

8.5 Hz, 1H); 7.09 (t, J = 54.5 Hz, 1H); 6.93 (d, J = 1.84 Hz, 1H); 7.66

(s, 1H) 8.72 (s, 1H).

Final Compound 41, Benzofuran-5-yl-[3-[5-(diﬂuoromethyl)-

[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]-4-methyl-1-piperidyl]-

methanone. Prepared in a similar manner to compound 42, starting

from intermediate 78 (rac, 220 mg, 0.82 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−1% MeOH in DCM,

1

aﬀorded compound 41 %. ESI m/z: 412 [M + H]⁺. H NMR (400

MHz, DMSO-d₆): δ ppm 0.82 (d, J = 6.46 Hz, 3H) 1.37−1.50 (m, 1H)

1.84−1.92 (m, 1H) 2.41−2.46 (m, 1H) 2.50−2.53 (m, 1H) 3.15 (td, J

= 12.92, 2.83 Hz, 1H) 3.34−3.40 (m, 1H) 3.56−3.64 (m, 1H) 4.17 (br

d, J = 12.92 Hz, 1H) 4.32 (br d, J = 12.51 Hz, 1H) 6.84−7.14 (m, 1H)

6.96 (dd, J = 2.42, 0.81 Hz, 1H) 7.38 (dd, J = 8.48, 1.61 Hz, 1H) 7.55−

7.56 (m, 1H) 7.58 (d, J = 8.48 Hz, 1H) 7.75 (d, J = 1.61 Hz, 1H) 7.94

(d, J = 2.02 Hz, 1H) 8.70 (s, 1H).

Final Compound 42, Benzofuran-5-yl-[(3S,4R)-3-[5-(diﬂuoro-

methyl)-[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]-4-methyl-1-piperidyl]-

methanone. 1-Benzofuran-5-carboxylic acid (214 mg, 1.32 mmol) was

stirred in DCM (10 mL). DIPEA (851 mg, 0.75 g/mL, 6.58 mmol) and

HBTU (499 mg, 1.32 mmol) were added, and stirring was continued

for 20 min at RT. Intermediate 78 (RS!, 400 mg, 1.32 mmol; WO 2018/

083098) was added, and stirring was continued for 1 h at RT. The RM

was quenched with 1 N aqueous NaOH (1 mL) solution and stirred for

5 min, then the OL was separated and dried over MgSO₄, ﬁltered, and

concentrated in vacuo. This material was puriﬁed by silica gel ﬂash

chromatography eluting with 0−3% MeOH in DCM to aﬀord

compound 42 (120 mg, yield 52.931%). ESI m/z: 412 [M + H]⁺. ¹H

NMR (400 MHz, DMSO-d₆): δ ppm 0.82 (d, J = 6.46 Hz, 3H) 1.37−

1.50 (m, 1H) 1.84−1.92 (m, 1H) 2.41−2.46 (m, 1H) 2.50−2.53 (m,

1H) 3.15 (td, J = 12.92, 2.83 Hz, 1H) 3.34−3.40 (m, 1H) 3.56−3.64

(m, 1H) 4.17 (br d, J = 12.92 Hz, 1H) 4.32 (br d, J = 12.51 Hz, 1H)

6.84−7.14 (m, 1H) 6.96 (dd, J = 2.42, 0.81 Hz, 1H) 7.38 (dd, J = 8.48,

1.61 Hz, 1H) 7.55−7.56 (m, 1H) 7.58 (d, J = 8.48 Hz, 1H) 7.75 (d, J =

1.61 Hz, 1H) 7.94 (d, J = 2.02 Hz, 1H) 8.70 (s, 1H).

Final Compound 38, 7-[1-(Benzofuran-5-ylmethyl)-3-piperidyl]-

5-(diﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidine. Prepared in a

similar manner to compound 37, starting from benzo[b]furan-5-

carboxaldehyde (0.10 g, 0.69 mmol) and intermediate 48i (0.15 g, 0.46

mmol). Puriﬁcation by Prep HPLC on a RP Vydac Denali column

(C1810 μm, 250 g, 5 cm) and using as mobile phase a mixture of a

0.25% NH₄HCO₃solution in water and ACN aﬀorded the title

compound as a free base, which was then dissolved in DIPE and treated

with HCl to aﬀord compound 38 as a HCl salt (180 mg, 86%). ESI m/z:

384 [M + H]⁺. ¹H NMR (360 MHz, DMSO-d₆): δ ppm 1.80−2.17 (m,

4H); 3.03 (br d, J = 11.71 Hz, 1H); 3.43−3.55 (m, 3H); 4.24 (br t, J =

12.08 Hz, 1H); 4.38−4.52 (m, 2H); 7.14 (t, J = 54.52 Hz, 1H); 7.01−

7.05 (m, 1H); 7.55 (s, 1H); 7.57 (d, J = 9.88 Hz, 1H); 7.69 (d, J = 8.42

Hz, 1H); 7.91 (s, 1H); 8.08 (d, J = 2.20 Hz, 1H); 8.85 (s, 1H); 11.06 (br

s, 1H). Anal. Calcd (C₂₀H₁₉F₂N₅O): C, H, N.

Final Compound 39, 7-[1-[(2-Chloro-6-methyl-4-pyridyl)methyl]-

3-piperidyl]-5-(diﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidine. A

solution of 2-chloro-4-chloromethyl-6-methylpyridine [162046-59-5]

(0.81 g, 4.6 mmol), intermediate 48i (1.00 g, 3.07 mmol) and DIPEA

(2.13 mL, 12.4 mmol) in DMF (100 mL) was stirred and heated at 40

°C for 4 h. The reaction mixture was cooled to RT, and then it was

evaporated under reduced pressure. The resulting residue was dissolved

in DCM, washed with a satd aq NaHCO₃soln (1×), dried over MgSO₄,

ﬁltered, and evaporated in vacuo. The resulting residue was puriﬁed by

prep HPLC on a RP Vydac Denali column (C1810 μm, 250 g, 5 cm)

and using as a mobile phase a mixture of a 0.25% NH₄HCO₃solution in

water and ACN aﬀorded the title compound as a racemic mixture (1.15

g, 95%). ESI m/z: 393 [M + H]⁺. Separation in enantiomers by

preparative SFC [CHIRALPAK Daicel AD 20 × 250 mm; mobile

phase: CO₂, EtOH with 0.2% 2-propylamine] aﬀorded compounds 39

(R*) (430 mg, 37%) R_t= 3.75 min, OR = +39.71° (589 nm, 20 °C,

0.345 w/v %; DMF), mp 164.56 °C, ¹H NMR (360 MHz, DMSO-d₆):

δ ppm 1.62−1.73 (m, 2H); 1.74−1.88 (m, 1H); 2.04 (br dd, J = 12.62,

4.21 Hz, 1H); 2.37 (br s, 1H); 2.44 (s, 3H); 2.53−2.71 (m, 2H); 2.99

(br d, J = 10.98 Hz, 1H); 3.54−3.62 (m, 2H); 3.82 (dt, J = 8.51, 4.35

Hz, 1H); 7.04 (t, J = 55.30 Hz, 1H); 7.26 (d, J = 6.59 Hz, 2H); 7.70 (s,

1H); 8.82 (s, 1H) and 41 (S*) (407 mg, 35%). R_t= 4.67 min, OR =

−42.56° (589 nm, 20 °C, 0.43 w/v %; DMF), mp 165.83 °C.

Final Compound 43, (3,4-Dimethoxyphenyl)-[(3S,4R)-3-[5-(di-

ﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]-4-methyl-1-

piperidyl]methanone. Prepared in a similar manner to compound 42,

starting from 3,4-dimethoxybenzoic acid (120 mg, 0.66 mmol) and

intermediate 78 (SR!, 200 mg, 0.66 mmol). Puriﬁcation by ﬂash

chromatography on a silica gel, eluting with 0−2% MeOH in DCM,

1

aﬀorded compound 41 %. ESI m/z: 432 [M + H]⁺. mp 227 °C. H

NMR (360 MHz, DMSO-d₆): δ ppm 0.00 (s, 1H) 0.77 (d, J = 6.59 Hz,

3H); 1.35−1.47 (m, 1H); 1.79−1.95 (m, 1H); 2.40−2.47 (m, 1H);

3.55−3.63 (m, 1H); 3.77 (br s, 3H); 3.79 (s, 4H); 3.93−4.65 (m, 1H);

6.95 (br s, 1H); 7.01 (br s, 3H); 7.10 (s, 1H); 7.68 (br s, 1H); 8.83 (s,

1H).

Final Compound 44, [3-[2-(5-Bromo-2-methyl-pyrazol-3-yl)-

ethyl]-6-ﬂuoro-1-naphthyl]oxy-tert-butyl-diphenyl-silane; 2-[3-[5-

(Diﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidin-7-yl]-4-methyl-1-

piperidyl]-N-methyl-pyridine-4-carboxamide; Methane. DIPEA

[7087-68-5] (372 mg, 2.88 mmol) was added at RT to a solution of

intermediate 78 (RS!, 175 mg, 0.58 mmol) and 2-bromo-n-methyl-4-

pyridinecarboxamide [337536-01-3] (372 mg, 1.73 mmol) in nBuOH

[71-36-3] (3 mL). The resulting mixture was then stirred for 4 h at 180

°C under microwave irradiation, then it was cooled to RT, and the

solvent was evaporated under reduced pressure. The resulting residue

was dissolved in DCM (5 mL) and washed with a saturated aqueous

NaHCO₃solution (1 mL). The biphasic mixture was passed through an

Extrelut ﬁlter, which was afterward washed with DCM (×2). The

combined OL was evaporated under reduced pressure. The resulting

residue was puriﬁed by Prep HPLC using as the stationary phase a

Uptisphere C18 ODB10 μm, 200 g, 5 cm column and as the mobile

phase a mixture of 0.25% NH₄HCO₃solution in water and CH₃CN, to

provide the title compound. This was recrystallized from ether, to give

compound 44 (95 mg, 41%). ESI m/z: 402 [M + H]⁺. mp 180.87 °C.

¹H NMR (400 MHz, DMSO-d₆): δ ppm 0.78 (d, J = 6.53 Hz, 3H) 1.42

Final Compound 40, 5-(Diﬂuoromethyl)-7-[1-[(2,6-dimethyl-4-

pyridyl)methyl]-3-piperidyl]-[1,2,4]triazolo[1,5-a]pyrimidine. Pre-

pared in a similar manner to compound 37, starting from 2,6-

dimethylisonicotinaldehyde [18206-06-9] (0.99 g, 7.36 mmol) and

intermediate 48i (1.20 g, 3.68 mmol). Puriﬁcation by prep HPLC on a

Uptisphere column (C1810 μm, 200 g, 5 cm) and using as the mobile

phase a mixture of a 0.25% NH₄HCO₃solution in water and MeOH

aﬀorded the title compound as a racemic mixture (1.04 g, 76%). ESI m/

z: 373 [M + H]⁺. Separation in enantiomers by preparative SFC

[CHIRALPAK Daicel AD 30 × 250 mm; mobile phase: CO₂, EtOH

with 0.2% 2-propylamine] aﬀorded compounds 40 (R*) (373 mg,

36%) R_t= 3.53 min, OR = +48.93° (589 nm, 20 °C, 0.515 w/v %;

U

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(br dd, J = 12.42, 3.89 Hz, 1H) 1.91 (br dd, J = 13.55, 3.51 Hz, 1H) 2.44

(br s, 2H) 2.78 (d, J = 4.52 Hz, 3H) 3.11 (td, J = 12.80, 3.01 Hz, 1H)

3.36 (dd, J = 12.80, 11.04 Hz, 1H) 3.51−3.58 (m, 1H) 4.45 (br d, J =

13.30 Hz, 1H) 4.53−4.59 (m, 1H) 6.95 (dd, J = 5.14, 1.13 Hz, 1H)

7.16−7.19 (m, 1H) 7.61 (s, 1H) 8.16 (d, J = 5.02 Hz, 2H) 8.74 (s, 1H).

Final Compound 45, 7-[1-(4-tert-Butyl-2-pyridyl)-4-methyl-3-

piperidyl]-5-(diﬂuoromethyl)-[1,2,4]triazolo[1,5-a]pyrimidine. Pre-

pared in a similar manner to compound 44, starting from 4-tert-butyl-

2-chloropyridine (112 mg, 0.66 mmol) and intermediate 78 (SR!, 100

mg, 0.33 mmol) and using NMP (3 mL) as the solvent. Puriﬁcation by

ﬂash chromatography on a silica gel, eluting with 0−2% MeOH in

DCM, followed by crystallization from diisopropyl ether, aﬀorded

compound 45 (37.6 mg, 29%). ESI m/z: 401 [M + H]⁺. ¹H NMR (360

MHz, chloroform-d): δ ppm 0.80−0.98 (m, 5H); 1.21−1.35 (m, 11H);

1.41−1.56 (m, 1H); 1.61−1.79 (m, 1H); 1.91−1.99 (m, 1H); 2.36 (br

s, 1H); 3.03−3.23 (m, 2H); 3.66 (br s, 1H); 4.55 (br dd, J = 12.26, 2.74

Hz, 2H); 6.53−6.85 (m, 2H); 6.86 (s, 1H); 7.33 (s, 1H); 8.10 (d, J =

5.12 Hz, 1H); 8.60 (s, 1H).

Final Compound 46, 5-(Diﬂuoromethyl)-7-[4-methyl-1-(4-phe-

nyl-2-pyridyl)-3-piperidyl]-[1,2,4]triazolo[1,5-a]pyrimidine. Pre-

pared in a similar manner to compound 44, starting from 2-chloro-4-

phenylpyridine [42260-39-9] (139 mg, 0.74 mmol) and intermediate

78 (SR!, 112 mg, 0.37 mmol) and using NMP (3.5 mL) as the solvent.

Puriﬁcation by ﬂash chromatography on a silica gel, eluting with 0−2%

MeOH in DCM, followed by crystallization from diisopropyl ether,

aﬀorded compound 46 (8.6 mg, 6%). ESI m/z: 421 [M + H]⁺. mp

127.58 °C. ¹H NMR (360 MHz, chloroform-d): δ ppm 0.76−0.96 (m,

4H) 1.40−1.61 (m, 1H) 1.93−2.01 (m, 1H) 2.41 (br s, 1H) 3.09−3.30

(m, 2H) 3.61−3.77 (m, 1H) 4.58−4.67 (m, 2H) 6.52−6.87 (m, 1H)

6.91 (dd, J = 5.31, 1.28 Hz, 1H) 7.12 (s, 1H) 7.34 (s, 1H) 7.41−7.53

(m, 3H) 7.64−7.72 (m, 2H) 8.25 (d, J = 5.49 Hz, 1H) 8.62 (s, 1H).

Frans Van den Keybus − Medicinal Chemistry, Janssen

Pharmaceutica N. V., B-2340 Beerse, Belgium

Gregor J. Macdonald − Medicinal Chemistry, Janssen

Pharmaceutica N. V., B-2340 Beerse, Belgium

Marijke V. F. Somers − Discovery Sciences, Janssen Research &

Development, Janssen Pharmaceutica N. V., B-2340 Beerse,

Belgium

Greet Vanhoof − Discovery Sciences, Janssen Research &

Development, Janssen Pharmaceutica N. V., B-2340 Beerse,

Belgium

Philip M. Leonard − Structural Biology, Charles River Discovery

(Previously BioFocus), Chesterford Research Park, CB10 1XL

Essex, U.K.

Marieke B. A. C. Lamers − Structural Biology, Charles River

Discovery (Previously BioFocus), Chesterford Research Park,

CB10 1XL Essex, U.K.

Yves E. M. Van Roosbroeck − Medicinal Chemistry, Janssen

Pharmaceutica N. V., B-2340 Beerse, Belgium

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.0c01272

Author Contributions

The manuscript was written through contributions of all

authors. All authors have given approval to the ﬁnal version of

the manuscript.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

ASSOCIATED CONTENT

* Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c01272.

■

sı

We thank Thomas Steinbrecher and Amr Ragab at Schrodinger

for assistance with the free-energy perturbation calculations. We

also thank Ilse Lenaerts for performing the occupancy

experiments. We thank the members of analysis and puriﬁcation

department, in particular Michel Carpentier and Alex de Groot

for help with data analysis.

Selectivity data for all compounds; data for retrospective

validation of FEP approach; comparison of binding

aﬃnity predictions from docking with experiment; and

additional chemistry details (PDF)

Molecular formula strings for 3−46 (CSV)

Input structures for FEP calculations (ZIP)

ABBREVIATIONS

■

AD, Alzheimer’s disease; cAMP, 3′,5′-cyclic adenosine mono-

phosphate; cGMP, 3′,5′-cyclic guanosine monophosphate;

CYP450, cytochrome P450; HTS, high-throughput screening;

FEP, free-energy perturbation; HP-β-CD, (2-hydroxypropyl)-

beta-cyclodextrin; LHS, left-hand side; LO, lead optimization;

LTP, long-term potentiation; MUE, mean unsigned error; NOS,

nitric oxide synthase; PDB, protein data bank; PDE,

phosphodiesterase; RBFE, relative binding free energy; RHS,

right-hand side; SAR, structure−activity relationship; SD,

standard deviation; SFC, supercritical ﬂuid chromatography

Accession Codes

Atomic coordinates and structure factors for compounds 3 and

39 have been deposited in the RCSB Protein Data Bank under

the accession code 6ZND and 6ZQZ respectively. Authors will

release the atomic coordinates and experimental data upon

article publication.

AUTHOR INFORMATION

Corresponding Authors

■

REFERENCES

Gary Tresadern − Computational Chemistry, Janssen

Pharmaceutica N. V., B-2340 Beerse, Belgium; orcid.org/

0000-0002-4801-1644; Phone: +32 1464 1569;

Email: gtresade@its.jnj.com

Peter J. J. A. Buijnsters − Medicinal Chemistry, Janssen

Pharmaceutica N. V., B-2340 Beerse, Belgium; Phone: +32 1460

7439; Email: pbuijnst@its.jnj.com

■

(1) Maurice, D. H.; Ke, H.; Ahmad, F.; Wang, Y.; Chung, J.;

Manganiello, V. C. Advances in targeting cyclic nucleotide

phosphodiesterases. Nat. Rev. Drug Discovery 2014, 13, 290−314.

(2) Francis, S. H.; Blount, M. A.; Corbin, J. D. Mammalian cyclic

nucleotide phosphodiesterases: molecular mechanisms and physio-

logical functions. Physiol. Rev. 2011, 91, 651−690.

(3) Pandit, J.; Forman, M. D.; Fennell, K. F.; Dillman, K. S.; Menniti,

F. S. Mechanism for the allosteric regulation of phosphodiesterase 2A

deduced from the X-ray structure of a near full-length construct. Proc.

Natl. Acad. Sci. U.S.A. 2009, 106, 18225−18230.

Authors

Ingrid Velter − Medicinal Chemistry, Janssen Pharmaceutica N.

V., B-2340 Beerse, Belgium

(4) Wu, A. Y.; Tang, X.-B.; Martinez, S. E.; Ikeda, K.; Beavo, J. A.

Molecular determinants for cyclic nucleotide binding to the regulatory

domains of phosphodiesterase 2A. J. Biol. Chem. 2004, 279, 37928−

37938.

́

Andres A. Trabanco − Medicinal Chemistry, Janssen Research &

Development, Janssen-Cilag S. A., 45007 Toledo, Spain;

orcid.org/0000-0002-4225-758X

V

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(5) Erneux, C.; Couchie, D.; Dumont, J. E.; Baraniak, J.; Stec, W. J.;

Abbad, E. G.; Petridis, G.; Jastorff, B. Specificity of cyclic GMP

activation of a multi-substrate cyclic nucleotide phosphodiesterase from

rat liver. Eur. J. Biochem. 1981, 115, 503−510.

Taniguchi, T. Discovery of clinical candidate 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): a

highly potent, selective, and brain-penetrating phosphodiesterase 2A

inhibitor for the treatment of cognitive disorders. J. Med. Chem. 2017,

60, 7677−7702.

(24) Helal, C. J.; Arnold, E. P.; Boyden, T. L.; Chang, C.; Chappie, T.

A.; Fennell, K. F.; Forman, M. D.; Hajos, M.; Harms, J. F.; Hoffman, W.

E.; Humphrey, J. M.; Kang, Z.; Kleiman, R. J.; Kormos, B. L.; Lee, C.-

W.; Lu, J.; Maklad, N.; McDowell, L.; Mente, S.; O’Connor, R. E.;

Pandit, J.; Piotrowski, M.; Schmidt, A. W.; Schmidt, C. J.; Ueno, H.;

Verhoest, P. R.; Yang, E. X. Application of structure-based design and

parallel chemistry to identify a potent, selective, and brain penetrant

Phosphodiesterase 2A inhibitor. J. Med. Chem. 2017, 60, 5673−5698.

(25) Stachel, S. J.; Berger, R.; Nomland, A. B.; Ginnetti, A. T.; Paone,

D. V.; Wang, D.; Puri, V.; Lange, H.; Drott, J.; Lu, J.; Marcus, J.; Dwyer,

M. P.; Suon, S.; Uslaner, J. M.; Smith, S. M. Structure-Guided design

and procognitive assessment of a potent and selective phosphodiester-

ase 2A inhibitor. ACS Med. Chem. Lett. 2018, 9, 815−820.

(6) Lakics, V.; Karran, E. H.; Boess, F. G. Quantitative comparison of

phosphodiesterase mRNA distribution in human brain and peripheral

tissues. Neuropharmacology 2010, 59, 367−374.

(7) 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. Immunohistochemical localization of phosphodiesterase

2A in multiple mammalian species. J. Histochem. Cytochem. 2009, 57,

933−949.

(8) Stephenson, D. T.; Coskran, T. M.; Kelly, M. P.; Kleiman, R. J.;

Morton, D.; O’Neill, S. M.; Schmidt, C. J.; Weinberg, R. J.; Menniti, F.

S. The distribution of phosphodiesterase 2A in the rat brain.

Neuroscience 2012, 226, 145−155.

(9) Blokland, A.; Menniti, F. S.; Prickaerts, J. PDE Inhibition and

cognition enhancement. Expert Opin. Ther. Pat. 2012, 22, 349−354.

(10) Bliss, T. V. P.; Collingridge, G. L. A synaptic model of memory:

long-term potentiation in the hippocampus. Nature 1993, 361, 31−39.

(11) Bollen, E.; Puzzo, D.; Rutten, K.; Privitera, L.; De Vry, J.;

Vanmierlo, T.; Kenis, G.; Palmeri, A.; D’Hooge, R.; Balschun, D.;

Steinbusch, H. M.; Blokland, A.; Prickaerts, J. Improved long-term

memory via enhancing cGMP-PKG signaling requires cAMP-PKA

signaling. Neuropsychopharmacology 2014, 39, 2497−2505.

(12) Lueptow, L. M.; Zhan, C.-G.; O’Donnell, J. M. Cylic cGMP-

mediated memory enhancement in the object recognition test by

inhibitors of phosphodiesterase-2 in mice. Psycopharmacology 2016,

233, 447−456.

(26) Helal, C. J.; Arnold, E.; Boyden, T.; Chang, C.; Chappie, T. A.;

Fisher, E.; Hajos, M.; Harms, J. F.; Hoffman, W. E.; Humphrey, J. M.;

Pandit, J.; Kang, Z.; Kleiman, R. J.; Kormos, B. L.; Lee, C.-W.; Lu, J.;

Maklad, N.; McDowell, L.; McGinnis, D.; O’Connor, R. E.; O’Donnell,

C. J.; Ogden, A.; Piotrowski, M.; Schmidt, C. J.; Seymour, P. A.; Ueno,

H.; Vansell, N.; Verhoest, P. R.; Yang, E. X. Identification of a potent,

highly selective, and brain penetrant phosphodiesterase 2A inhibitor

clinical candidate. J. Med. Chem. 2018, 61, 1001−1018.

(27) Buijnsters, P.; De Angelis, M.; Langlois, X.; Rombouts, F. J. R.;

Sanderson, W.; Tresadern, G.; Ritchie, A.; Trabanco, A. A.; Vanhoof,

(13) Elias, M. F.; Beiser, A.; Wolf, P. A.; Au, R.; White, R. F.;

D’Agostino, R. B. The preclinical phase of Alzheimer disease: A 22-year

prospective study of the Framingham Cohort. Arch. Neurol. 2000, 57,

808−813.

́

G.; Roosbroeck, Y. V.; Andres, J.-I. Structure-based design of a potent,

selective, and brain penetrating PDE2A inhibitor with demonstrated

target engagement. ACS Med. Chem. Lett. 2014, 5, 1049−1053.

(28) Rombouts, F. J. R.; Tresadern, G.; Buijnsters, P.; Langlois, X.;

(14) Newman, S. K.; Warrington, E. K.; Kennedy, A. M.; Rossor, M.

N. The earliest cognitive change in a person with familial Alzheimer ’s

disease: presymptomatic neuropsychological features in a pedigree with

familial Alzheimer ’s disease confirmed at necropsy. J. Neurol.,

Neurosurg. Psychiatry 1994, 57, 967−972.

́

Tovar, F.; Steinbrecher, T. B.; Vanhoof, G.; Somers, M.; Andres, J.-I.;

Trabanco, A. A.-I.; Trabanco, A. A. Pyrido[4,3-e][1,2,4]triazolo[4,3-

a]pyrazines as selective, brain penetrant phosphodiesterase 2 (PDE2A)

inhibitors. ACS Med. Chem. Lett. 2015, 6, 282−286.

(29) Jorgensen, W. L. Efficient drug lead discovery and optimization.

Acc. Chem. Res. 2009, 42, 724−733.

(15) Mayeux, R. Early Alzheimer ’s Disease. N. Engl. J. Med. 2010, 362,

2194−2201.

(30) Abel, R.; Wang, L.; Harder, E. D.; Berne, B. J.; Friesner, R. A.

Advancing drug discovery through enhanced free energy calculations.

Acc. Chem. Res. 2017, 50, 1625−1632.

(16) Gabrieli, J. D. E. Memory systems analyses of mnemonic

disorders in aging and age-related diseases. Proc. Natl. Acad. Sci. U.S.A.

1996, 93, 13534−13540.

(31) Cournia, Z.; Allen, B.; Sherman, W. Relative binding free energy

calculations in drug discovery: Recent advances and practical

considerations. J. Chem. Inf. Model. 2017, 57, 2911−2937.

́

(17) Belleville, S.; Sylvain-Roy, S.; de Boysson, C.; Menard, M.-C.

Chapter 23 Characterizing the memory changes in persons with mild

cognitive impairment. Prog. Brain Res. 2008, 169, 365−375.

(18) Lyketsos, C. G.; Lopez, O.; Jones, B.; Fitzpatrick, A. L.; Breitner,

J.; DeKosky, S. Prevalence of neuropsychiatric symptoms in dementia

and mild cognitive impairment: results from the cardiovascular health

study. JAMA 2002, 288, 1475−1483.

́

(32) Ciordia, M.; Perez-Benito, L.; Delgado, F.; Trabanco, A. A.;

Tresadern, G. Application of free energy perturbation for the design of

BACE1 inhibitors. J. Chem. Inf. Model. 2016, 56, 1856−1871.

̈

́

(33) Keranen, H.; Perez-Benito, L.; Ciordia, M.; Delgado, F.;

Steinbrecher, T. B.; Oehlrich, D.; van Vlijmen, H. W. T.; Trabanco,

A. A.; Tresadern, G. Acylguanidine beta secretase 1 inhibitors: A

combined experimental and free energy perturbation study. J. Chem.

Theory Comput. 2017, 13, 1439−1453.

(19) Blennow, K.; de Leon, M. J.; Zetterberg, H. Alzheimer ’s disease.

Lancet 2006, 368, 387−403.

(20) Gomez, L.; Breitenbucher, J. G. PDE2A inhibition: potential for

the treatment of cognitive disorders. Bioorg. Med. Chem. Lett. 2013, 23,

6522−6527.

́

(34) Deflorian, F.; Perez-Benito, L.; Lenselink, E. B.; Congreve, M.;

Van Vlijmen, H.; Mason, J. S.; de Graaf, C.; Tresadern, G. Accurate

prediction of GPCR ligand binding affinity with free energy

perturbation. J. Chem. Inf. Model. 2020, 60, 4311−4325,

DOI: 10.1021/acs.jcim.0c00449.

(21) Wu, Y.; Li, Z.; Huang, Y.-Y.; Wu, D.; Luo, H.-B. Novel

phosphodiesterase inhibitors for cognitive improvement in Alzheimer ’s

disease. J. Med. Chem. 2018, 61, 5467−5483.

(22) 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.

Design and synthesis of novel and selective phosphodiesterase 2

(PDE2a) inhibitors for the treatment of memory disorders. J. Med.

Chem. 2017, 60, 2037−2051.

́

̈

(35) Perez-Benito, L.; Keranen, H.; van Vlijmen, H. W. T.; Tresadern,

G. Predicting binding free energies of PDE2A inhibitors. The

difficulties of protein conformation. Sci. Rep. 2018, 8, 4883.

́

(36) Perez-Benito, L.; Casajuana-Martin, N.; Jimenez-Roses, M.; van

Vlijmen, H.; Tresadern, G. Predicting activity cliffs with free-energy

perturbation. J. Chem. Theory Comput. 2019, 15, 1884−1895.

́

(37) Gapsys, V.; Perez-Benito, L.; Aldeghi, M.; Seeliger, D.; van

Vlijmen, H.; Tresadern, G.; de Groot, B. L. Large scale relative protein

ligand binding affinities using non-equilibrium alchemy. Chem. Sci.

2020, 11, 1140−1152.

(23) 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.;

W

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

́

̈

(38) Wan, S.; Tresadern, G.; Perez-Benito, L.; Vlijmen, H.; Coveney,

(53) Schindler, C.; Baumann, H.; Blum, A.; Bose, D.; Buchstaller, H.-

P.; Burgdorf, L.; Cappel, D.; Chekler, E.; Czodrowski, P.; Dorsch, D.;

P. V. Accuracy and precision of alchemical relative free energy

predictions with and without replica-exchange. Adv. Theory Simul.

2019, 3, 1900195.

̈

Eguida, M. K. I.; Follows, B.; Fuchß, T.; Gradler, U.; Gunera, J.;

Johnson, T.; Jorand Lebrun, C.; Karra, S.; Klein, M.; Knehans, T.;

Kotzner, L.; Krier, M.; Leiendecker, M.; Leuthner, B.; Li, L.; Mochalkin,

̈

(39) Zhang, K. Y. J.; Card, G. L.; Suzuki, Y.; Artis, D. R.; Fong, D.;

Gillette, S.; Hsieh, D.; Neiman, J.; West, B. L.; Zhang, C.; Milburn, M.

V.; Kim, S.-H.; Schlessinger, J.; Bollag, G. Glutamine switch mechanism

for nucleotide selectivity by phosphodiesterases. Mol. Cell 2004, 15,

279−286.

(40) Jansen, C.; Kooistra, A. J.; Kanev, G. K.; Leurs, R.; de Esch, I. J. P.;

de Graaf, C. PDEStrIAn: A Phosphodiesterase structure and ligand

interaction annotated database as a tool for structure-based drug design.

J. Med. Chem. 2016, 59, 7029−7065.

(41) Wilson, L. S.; Brandon, N. J. Emerging biology of PDE10A. Curr.

Pharm. Des. 2015, 21, 378−388.

(42) Verhoest, P. R.; Chapin, D. S.; Corman, M.; Fonseca, K.; Harms,

J. F.; Hou, X.; Marr, E. S.; Menniti, F. S.; Nelson, F.; O’Connor, R.;

Pandit, J.; Proulx-LaFrance, C.; Schmidt, A. W.; Schmidt, C. J.; Suiciak,

J. A.; Liras, S. Discovery of a novel class of phosphodiesterase 10A

inhibitors and identification of clinical candidate 2-[4-(1-Methyl-4-

pyridin-4-yl-1H-pyrazol-3-yl)-phenoxymethyl]-quinoline (PF-

2545920) for the treatment of schizophrenia. J. Med. Chem. 2009, 52,

5188−5196.

I.; Musil, D.; Neagu, C.; Rippmann, F.; Schiemann, K.; Schulz, R.;

Steinbrecher, T. B.; Tanzer, E.-M.; Unzue Lopez, A.; Viacava Follis, A.;

Wegener, A.; Kuhn, D. Large-Scale assessment of binding free energy

calculations in active drug discovery projects. J. Chem. Inf. Model. 2020,

DOI: 10.1021/acs.jcim.0c00900.

(54) Madhavi Sastry, G.; Adzhigirey, M.; Day, T.; Annabhimoju, R.;

Sherman, W. Protein and ligand preparation: Parameters, protocols,

and influence on virtual screening enrichments. J. Comput.-Aided Mol.

Des. 2013, 27, 221−234.

(55) Schrodinger LLC. 120 West 45th Street 17th Floor, Tower 45

New York, NY 10036-4041, USA at http://www.schrodinger.com/

Maestro/, 2014. (accessed Sept 18, 2020).

(43) Scapin, G.; Patel, S. B.; Chung, C.; Varnerin, J. P.; Edmondson, S.

D.; Mastracchio, A.; Parmee, E. R.; Singh, S. B.; Becker, J. W.; Van der

Ploeg, L. H. T.; Tota, M. R. Crystal structure of human

phosphodiesterase 3B: Atomic basis for substrate and inhibitor

specificity. Biochemistry 2004, 43, 6091−6100.

(44) Li, Z.; Huang, Y.; Wu, Y.; Chen, J.; Wu, D.; Zhan, C.-G.; Luo, H.-

B. Absolute binding free energy calculation and design of a

subnanomolar inhibitor of phosphodiesterase-10. J. Med. Chem. 2019,

62, 2099−2111.

(45) Moraca, F.; Negri, A.; de Oliveira, C.; Abel, R. Application of free

energy perturbation (FEP+) to understanding ligand selectivity: A case

study to assess selectivity between pairs of phosphodiesterases (PDE ’s).

J. Chem. Inf. Model. 2019, 59, 2729−2740.

(46) Abel, R.; Wang, L.; Mobley, D. L.; Friesner, R. A. A critical review

of validation, blind testing, and real-world use of alchemical protein-

ligand binding free energy calculations. Curr. Top. Med. Chem. 2017, 17,

2577−2585.

(47) Mey, A. S.; Allen, B.; Macdonald, H. E. B.; Chodera, J. D.; Kuhn,

M.; Michel, J.; Mobley, D. L.; Naden, L. N.; Prasad, S.; Rizzi, A.; Scheen,

J.; Shirts, M. R.; Tresadern, G.; Xu, H.. Best practice for free energy

calculations. 2020, arXiv:2008.03067. Published Online: August 21,

2020, https://arxiv.org/abs/2008.03067 (accessed Sept 22, 2020).

(48) Shirts, M. R.; Mobley, D. L.; Brown, S. P. Free-Energy

Calculations in Structure-Based Drug Design. In Drug Design. Structure-

and Ligand-Based Approaches; Merz, K. M., Jr., Ringe, D., Reynolds, C.

H., Eds.; Cambridge University Press: New York, 2010; pp 61−86.

(49) Abel, R.; Manas, E. S.; Friesner, R. A.; Farid, R. S.; Wang, L.

Modeling the value of predictive affinity scoring in preclinical drug

discovery. Curr. Opin. Struct. Biol. 2018, 52, 103−110.

(50) (a) Van Roosbroeck, Y. E. M.; Van den Keybus, F. A. M.;

Tresadern, G. J.; Buijnsters, P. J. J. A.; Velter, A. I.; Jacoby, E.;

Macdonald, G. J.; Gijsen, H. J. M.; Ahnaou, A.; Drinkenburg, W. H. I.

M. [1,2,4]Triazolo[1,5-a]pyrimidine Derivatives as PDE2 Inhibitors.

WO 2018083098 A1, May 11, 2018. (b) Van Roosbroeck, Y. E. M.;

Buijnsters, P. J. J. A.; Tresadern, G. J.; Jacoby, E.; Oehlrich, D.; Gijsen,

H. J. M. [1,2,4]Triazolo[1,5-a]pyrimidine Compounds as PDE2

Inhibitors. WO 2018083103 A1, May 11, 2018.

(51) The absolute conﬁguration was determined using vibrational

VCD. See Supporting Information.

(52) Ghanakota, P.; Bos, P. H.; Konze, K.; Staker, J.; Marques, G.;

Marshall, K.; Leswing, K.; Abel, R.; Bhat, S. Combining cloud-based

free energy calculations, synthetically aware enumerations and goal-

directed generative machine learning for rapid large scale chemical

exploration and optimization. J. Chem. Inf. Model. 2020, DOI: 10.1021/

acs.jcim.0c00120.

X

https://dx.doi.org/10.1021/acs.jmedchem.0c01272

J. Med. Chem. XXXX, XXX, XXX−XXX

Article Doi

Triazolo[1,5-a]pyrimidine Phosphodiesterase 2A Inhibitors: Structure and Free-Energy Perturbation-Guided Exploration

DOI: 10.1021/acs.jmedchem.0c01272

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.0c01272

Full text of DOI:10.1021/acs.jmedchem.0c01272

Products guided by the article

R&D Labs maybe for 1121057-77-9

Relevant to this article

Redox reactions in palladium catalysis: On the accelerating and/or inhibiting effects of copper and silver salt additives in cross-coupling chemistry involving electron-rich phosphine ligands

Copper-catalyzed decarboxylative cyclization: Via tandem C-P and C-N bond formation: Access to 2-phosphorylmethyl indoles

Synthesis of dibutyl-trimethylsilanylmethyl-amine and its application towards SO₂absorption with phase change behaviors

Solid-Phase Synthesis, Separation, and Stereochemical Aspects of P-Chiral Methane- and 4,4'-Dimethoxytriphenylmethanephosphonate Analogues of Oligodeoxyribonucleotides

Short Communication. A potential chiral derivatizing agent for 1,2-diglycerides

Article Doi

Triazolo[1,5-a]pyrimidine Phosphodiesterase 2A Inhibitors: Structure and Free-Energy Perturbation-Guided Exploration

DOI: 10.1021/acs.jmedchem.0c01272

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.0c01272

Full text of DOI:10.1021/acs.jmedchem.0c01272

Products guided by the article

R&D Labs maybe for 1121057-77-9

Relevant to this article

Redox reactions in palladium catalysis: On the accelerating and/or inhibiting effects of copper and silver salt additives in cross-coupling chemistry involving electron-rich phosphine ligands

Copper-catalyzed decarboxylative cyclization: Via tandem C-P and C-N bond formation: Access to 2-phosphorylmethyl indoles

Synthesis of dibutyl-trimethylsilanylmethyl-amine and its application towards SO2absorption with phase change behaviors

Solid-Phase Synthesis, Separation, and Stereochemical Aspects of P-Chiral Methane- and 4,4'-Dimethoxytriphenylmethanephosphonate Analogues of Oligodeoxyribonucleotides

Short Communication. A potential chiral derivatizing agent for 1,2-diglycerides

Synthesis of dibutyl-trimethylsilanylmethyl-amine and its application towards SO₂absorption with phase change behaviors