Identification of NVP-TNKS656: The use of structure-efficiency relationships to generate a highly potent, selective, and orally active tankyrase inhibitor

Article

pubs.acs.org/jmc

Identiﬁcation of NVP-TNKS656: The Use of Structure−Eﬃciency

Relationships To Generate a Highly Potent, Selective, and Orally

Active Tankyrase Inhibitor

Michael D. Shultz,* Atwood K. Cheung, Christina A. Kirby, Brant Firestone, Jianmei Fan,

Christine Hiu-Tung Chen, Zhouliang Chen, Donovan N. Chin, Lucian DiPietro,^‡Aleem Fazal, Yun Feng,

Pascal D. Fortin, Ty Gould, Bharat Lagu, Huangshu Lei,^§Francois Lenoir, Dyuti Majumdar,

Etienne Ochala,^∥M. G. Palermo, Ly Pham,^⊥Minying Pu, Troy Smith, Travis Stams,

Ronald C. Tomlinson,^#B. Barry Toure, Michael Visser, Run Ming Wang, Nigel J. Waters,^□

́

and Wenlin Shao

Novartis Institutes for Biomedical Research, Inc., 250 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States

S

* Supporting Information

ABSTRACT: Tankyrase 1 and 2 have been shown to be

redundant, druggable nodes in the Wnt pathway. As such,

there has been intense interest in developing agents suitable

for modulating the Wnt pathway in vivo by targeting this

enzyme pair. By utilizing a combination of structure-based

design and LipE-based structure eﬃciency relationships, the

core of XAV939 was optimized into a more stable, more

eﬃcient, but less potent dihydropyran motif 7. This core was

combined with elements of screening hits 2, 19, and 33 and

resulted in highly potent, selective tankyrase inhibitors that are

novel three pocket binders. NVP-TNKS656 (43) was

identiﬁed as an orally active antagonist of Wnt pathway activity in the MMTV-Wnt1 mouse xenograft model. With an

enthalpy-driven thermodynamic signature of binding, highly favorable physicochemical properties, and high lipophilic eﬃciency,

NVP-TNKS656 is a novel tankyrase inhibitor that is well suited for further in vivo validation studies.

biological role of TNKS thus oﬀers an opportunity for its

therapeutic intervention in diseases where TNKS substrates

become dysregulated, such as telomere elongation in cancer,

Wnt pathway activation in cancer, and cardiovascular and

neurodegenerative diseases.⁷Indeed, multiple groups have

reported identiﬁcation of TNKS inhibitors from high

throughput screens and chemical optimization (Figure

1).^4,8−10These inhibitors have been investigated for their

utilities in the treatment of cancer, ﬁbrosis, herpes simplex virus

replication, and remyelination, providing important insight on

therapeutic indications of targeting TNKS.^4,10−15

We recently described the characterization of a novel series

of selective dual inhibitors of TNKS1/2 (2) that are devoid of

activity against PARP1 or PARP2 (Figure 1).⁹These inhibitors

exhibit low nanomolar TNKS inhibition and submicromolar

activity in inhibiting Wnt pathway signaling. Others have

recently published optimization eﬀorts that have resulted in

G007-LK and compound 44.¹⁶⁻¹⁸In this report, we explore a

novel use of lipophilic eﬃciency (LipE = pIC₅₀− log P)¹⁹to

survey a diverse range of chemical structures to more eﬀectively

INTRODUCTION

■

Tankyrase (TNKS) proteins are multifunctional poly(ADP-

ribose) polymerases (PARPs), which are enzymes utilizing

NAD⁺as a substrate to generate ADP-ribose polymers onto

target proteins (PARsylation).¹Both isoforms of TNKS,

TNKS1 and TNKS2, share overlapping functions and similar

structures, including the ankyrin (ANK) repeat domain, the

SAM domain (sterile alpha molecule), and the catalytic PARP

domain.²The ankyrin domain is important for mediating

protein−protein interactions, and the SAM domain mediates

mono- and heteromultimerization of TNKS. Through binding

and modulating substrate proteins, TNKS plays a role in

regulating a variety of cellular processes, including telomere

regulation, mitotic spindle pole function, membrane trans-

location of Golgi-associated vesicles upon insulin stimulation,

and B cell antigen receptor activation.²⁻⁴Previous ﬁndings

from our group have shown that TNKS could regulate the

activity of the Wnt/β-catenin pathway in colon cancer cells

through PARsylation and destabilization of the axin proteins.⁵

In addition to the aforementioned substrates, an array of

proteins representing diverse biological functions have been

predicted to be TNKS targets on the basis of structural and

sequence information of TNKS targeting proteins.⁶The broad

Received: May 30, 2013

A

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 1. Select TNKS inhibitors that have been reported in the literature.

Table 1. A-Ring Modiﬁcations of 1

a

b

c

d

e

LipE = pIC₅₀(TNKS2) − clogP. HEK293 SuperTopFlash reporter gene assay. Rat liver microsome extraction ratio. Measured log D. LipE =

pIC₅₀(TNKS2) − log D.

understand what are the most eﬃcient interactions favored by

TNKS. With this information, we were able to develop TNKS

inhibitors with improved activity, selectivity, and pharmacoki-

netic proﬁles for in vivo testing.

other PARP family members (10−100 fold), low microsomal

stability (rat liver microsomal extraction ration, rLMER = 0.88),

and low solubility (10 μM at pH1, < 5 μM at pH 6.8) which

hampered the use of reasonable formulations for in vivo studies.

For a preclinical proof of concept on the impact of TNKS

inhibition in model systems, a potent, selective, and in vivo

active TNKS inhibitor was required; therefore, the highest

priority for this series was to increase potency, selectivity,

solubility, and metabolic stability. Because compound 1 had a

measured log D of 4.1, it was hypothesized that by reducing the

lipophilicity the oﬀ-target liabilities and poor properties could

be improved.²¹

Following our identiﬁcation of XAV939 (1), our initial hit

ﬁnding eﬀorts discovered several series of TNKS inhibitors that

were evaluated and prioritized for further optimization.^5,9In

parallel, we evaluated 1 to identify potential liabilities and

devise an optimization strategy. There are several attributes of 1

which make it an attractive starting point including low

nanomolar biochemical activity, 78 nM cellular activity, high

ligand eﬃciency²⁰(LE = 0.55), and good lipophilic eﬃciency

(LipE = 4.2 based on measured log D of 4.1). Unfortunately, 1

is not without liabilities which include low selectivity versus

Lipophilic eﬃciency¹⁹(LipE = pIC₅₀− log D) provides a

convenient measure of the diﬀerential between potency and

B

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 1. Synthesis of A-Ring Analogues

a

(a) K₂CO₃, MeOH, 75 °C; (b) DIEA, THF; (c) KOEt, 140 °C, 20 min.

lipophilicity and therefore was ideally aligned with our

optimization goals. It should be noted in Table 1 that early

on we identiﬁed a signiﬁcant disconnect between calculated

logP and measured log P/D for this series (Supporting

Information Figures 1 and 2), and therefore the LipE

calculations throughout this manuscript will be artiﬁcially

high relative to those based on measured log D. There are

several points that should be made at the outset to better

understand how the LipE tool might best be used. First,

because there can be errors in clogP calculations, comparison of

highly dissimilar structural classes should be performed with

caution and preferably utilizing measured log D values. Second,

for structure−eﬃciency relationships (SER), it is the change in

LipE (ΔLipE) between matched pairs that is meaningful in

decision making. If a change between a molecular matched pair

(MMP)²²is unlikely to introduce an ionizable group or aﬀect

the pK_aof an existing one, then ΔclogP can be calculated with

high accuracy and should correlate very well with changes in

measured log P and log D, and therefore may be used with

conﬁdence (Supporting Information Figure 2). Finally,

although it has been reported that there are target LipE values

for optimization,^19,23,24these are highly theoretical and likely

target dependent. Thus, we do not support using an absolute

LipE value to determine if suﬃcient optimization has been

achieved. We were very interested in evaluating the utility of

LipE for enhanced decision making based on SER and describe

herein the results of using LipE for making nonobvious choices

during an iterative optimization process. While we report ligand

eﬃciency for 1, we did not use this parameter during our

optimization eﬀorts and only SER utilizing LipE will be

discussed.

accomplished by the condensation of keto ester derivatives with

an appropriate amidine moiety (Scheme 1). Compound 1, a

nicotinamide mimetic, is structurally similar to many previously

reported PARP antagonists,⁷and we generated a series of

analogues where the sulfur-containing ring was altered (Table

1). For the nonaromatic A-rings analogues (3−5), it became

clear that all changes were not well tolerated as potency and

LipE decreased. Aromatization of ring A (6) did improve the

microsomal stability but was a less eﬃcient inhibitor.

We hypothesized that an oxygen-containing A-ring, such as a

dihydropyran, could improve solubility and metabolic stability;

therefore, we synthesized several of these analogues (e.g., 7 and

8). All analogues were more polar than 1 but also less potent.

Compound 7 is 33-fold less potent in the cellular assay than 1,

a result that would typically remove this modiﬁcation from

further consideration; however, these two compounds have

identical LipE values and therefore 7 was evaluated further.

With a solubility improvement of at least 10-fold and an even

more substantial improvement in microsomal stability, 7 was

able to be formulated for a rat pharmacokinetic (PK) study via

intravenous (iv) administration. After iv administration to

Sprague−Dawley rats, good exposure of compound 7 with low

clearance (8.9 mL/min/kg) and a high volume of distribution

(Vdss = 2.7 L/kg) was observed. The PK proﬁle 2 h postdose

suggests the possibility of enterohepatic recirculation, which is

plausible because of the available hydroxyl for direct

conjugation and biliary elimination. However, this hypothesis

would need to be further evaluated in mechanistic studies.

With the higher selectivity toward PARP 1 and 2 (50−1300

fold,) we decided to preserve the dihydropyran motif and shift

optimization to the other side of the molecule. The synthesis of

C-ring modiﬁcations is outlined in Scheme 2. In an attempt to

identify the minimal pharmacophore required for inhibition, the

phenyl ring was removed or replaced with heterocycles which

led to the complete loss of activity with the exception of 9

For all compounds in Table 1 and throughout this

manuscript (with the exception of compound 4), potency

against TNKS2 was typically 2−10 fold greater than TNKS1.

No signiﬁcant selectivity between these isoforms was observed

for compounds in this manuscript; therefore, this data is not

reported in subsequent tables.

a

Scheme 2. Synthesis of Dihydropyran C-Ring Analogues

RESULTS AND DISCUSSION

■

Computational programs that predict oxidative metabolism

such as Metasite and Stardrop support the concern that we had

at the outset of the optimization process that the A-ring sulfur

was a likely soft spot for oxidative metabolism (Supporting

Information Figure 3), and so replacement of this moiety was

our initial goal.²³⁻²⁶Modiﬁcation of the XAV939 A-ring was

a

K₂CO₃, MeOH, reﬂux overnight.

C

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(Table 2). The improved LipE of 9, relative to the other

analogues, suggests that the phenyl ring was suboptimal. We

The optimization appeared to have entered a structural cul-

de-sac where the vectors oﬀ of the C-ring were not favorable for

further optimization but did allow for modulation of the overall

properties of the molecule. While most of this position was

solvent-exposed, we hypothesized that an amphipathic group

might allow for some hydrophobic TNKS−ligand interactions

while also assisting in overall properties by lowering log D.

Whereas compound 17 has two polar oxygens and one methyl

group, we sought an analogue with a single hydrophobic and

hydrophilic group. The additional methyl group of 18

improved LipE relative to 7 (7.1 based on measured log D)

and apparently eﬃciency peaked at 17 and 18 as other

analogues (not shown) had slightly improved potency but

signiﬁcantly lower LipE. Compound 18 was further evaluated

and found to be extremely stable in liver microsomes (rER =

0.23), over 500-fold more soluble than 1, and on the basis of

measured log D values had signiﬁcantly greater lipophilic

eﬃciency (ΔLipE = +2.9). Suitable formulations for multiple

dose pharmacokinetic (PK)/pharmacodynamic (PD)/eﬃcacy

studies with compound 18 were achieved, and further in vivo

characterization was performed.

Table 2. Structure−Eﬃciency Relationships of C-Ring

Analogues

The pharmacokinetics of compound 18 was evaluated in

both rats and mice. After oral administration in Sprague−

Dawley rats, compound 18 showed good overall exposure and

bioavailability (F = 46%) where an initial rapid phase of

absorption was observed, followed by a second peak at 6 h

postdose. After iv administration, compound 18 again displayed

good exposure and low clearance (12 mL/min/kg), in line with

the low in vitro clearance prediction in rat liver microsomes

(ER = 0.24). Similar to the iv proﬁle of compound 7, the

concentration versus time proﬁle of compound 18 after oral

administration also displayed a second peak 4−6 h postdose,

further supporting the earlier hypothesis for enterohepatic

recirculation of either parent or a direct glucuronidation of the

free hydroxyl. In mice, compound 18 showed dose-propor-

tional exposure after oral dosing and good bioavailability (F =

32%). 18 also showed moderate iv clearance in good agreement

with the in vitro clearance prediction from mouse liver

microsomes. While high oral doses (300 mg/kg) were

tolerated, the low cellular potency only allowed for total drug

concentrations to exceed the cellular IC₅₀values for 4−5 h.

It was becoming increasingly clear that modiﬁcations based

on the ABC rings of 1 would not be able to achieve the low

nanomolar potency we required, and therefore we sought a

more favorable scaﬀold where the C ring was completely

replaced. At this time, we obtained a crystal structure of the

TNKS1 PARP domain in complex with an interesting screening

hit (19) to 2.1 Å resolution. The compound bound to the

NAD⁺donor site, occupying the nicotinamide and diphosphate

pockets. Its interactions with TNKS1 mimic those seen in the

structure of TNKS1 in complex with XAV939²⁴which include

stacking interactions of the pyrimidine ring with Tyr1224 and

hydrogen bonds between the hydroxyl of Ser1221 and Gly1185

with the pyrimidine hydroxyl and pyrimidine nitrogen,

respectively. Additionally, 19 takes advantage of a previously

deﬁned hydrophobic nook⁹where the thiophene moiety makes

van der Waals interactions with Cβ of Ser1186, Pro1187, and

Ile1228. Additionally, the phenyl ring makes van der Waals

interactions with His1184, Ser1186, and Phe1188. Unfortu-

nately, 19 had poor solubility and low microsomal stability,

consistent with the lower LipE (Table 3). By converting the A

ring into the dihydropyran motif (compound 20), the LipE (4.5

based on measured log D) and solubility improved dramatically,

a

b

LipE = pIC₅₀(TNKS2) − clogP. HEK293 SuperTopFlash reporter

c

d

gene assay. Rat liver microsome extraction ratio. Measured log D.

e

LipE = pIC₅₀(TNKS2) − log D.

next explored a series of triﬂuoromethyl replacements (10−17)

to probe this region. Replacement of the triﬂuoromethyl group

with a methyl group did not aﬀect potency but resulted in

improved LipE due to a reduction in lipophilicity, perhaps an

indication that more optimal groups could be identiﬁed. A

broad range of analogues were tested and signiﬁcant improve-

ments in LipE with several compounds (13, 14, 16, and

especially 17) were observed. However, this diverse range of

substitutions all resulted in similar activity (20−200 nM).

Closer examination of the crystal structure revealed that most

of the substituents in this subseries are solvent-exposed and

have very little interaction with the TNKS protein. In contrast,

modiﬁcations at the ortho position of the C-ring obliterates

activity (data not shown). We observed that LipE values can be

increased with polar groups that interact with the bulk solvent,

and while this may improve a compound’s overall proﬁle, in the

context of eﬃcient binding it has little to do with direct

interactions with TNKS.

D

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 3. Structure−Eﬃciency Relationships of Compounds 1, 7, and 19 Tribrid Analogues

a

b

c

d

e

LipE = pIC₅₀(TNKS2) − clogP. HEK293 SuperTopFlash reporter gene assay. Rat liver microsome extraction ratio. Measured log D. LipE =

pIC₅₀(TNKS2) − log D.

a

Scheme 3. Synthesis of Dihydropyran C-Ring Analogues

a

(a) TEA, 2-chloroacetamidine, MeOH; (b) i. NaN₃, H₂O/acetone, ii. Ph₃P, THF; (c) R²-CHO, SiCNBH₃, AcOH, EtOH; RSO₂Cl, TEA, DCM or

RCO₂Cl, TEA, DCM or RCO₂H, EDCI, HOBt, TEA, DCM; (d) i. BocNHR², NaH, THF or DMF; ii. HCl, dioxane, MeOH; (e) R₃NH₂, TEA, 120

°C, 5 min, or R₃NH₂, TEA or DIEA, EtOH, 65 °C; (f) RCO₂Cl, TEA, DCM; (g) RCO₂H, EDCI, HOBt, TEA, DCM.

but microsomal stability remained low. We began by stepwise

deletion of the thiophene and the phenpropanamide moieties

to investigate their importance in binding eﬃciency and

microsomal stability with the synthesis of this series described

in Scheme 3. Truncation to the N-ethyl derivative 21 or to the

secondary amide 22 resulted in lower binding eﬃciency and

cellular activity but improved microsomal stability. Deletion of

the amide moiety (23) also improved microsomal stability but

also increased LipE signiﬁcantly, suggesting this was an

ineﬃcient binding motif. The 3-thiophenyl analogue 24 was

5-fold less active, and activity was lost with other linker

modiﬁcations (data not shown).

and the backbone carbonyl of D1198 were in close proximity to

the phenyl ring (Figure 2). Additionally, a crystallographic

water molecule was bound via the backbone NH of D1198, and

we sought to make a hydrogen bond with this water molecule.

Both compounds 25 and 26 were less potent than 20; however,

25 had signiﬁcantly improved LipE. For a ΔclogP of −1.52,

both analogues would be expected to be about 33-fold less

potent than 20 from a LipE perspective. Compound 25

maintains similar LipE and is 37-fold less potent, and the in

vitro stability improves somewhat, presumably because of the

lower lipophilicity. For analogue 26, however, the 4-fold

decrease in potency is much better than the expected 33-fold

drop and is consistent with the formation of a hydrogen bond

with the bound water. Fluorine derivatives (27−29) were

tested, and compounds 27 and 29 were both more potent and

had improved LipE, however were unstable in microsomes. We

We next kept the thiophene constant and surveyed analogues

of the amide group (Table 4). Examination of the crystal

structure of 19 suggested that there were potential polar

interactions that might be gained, as the hydroxyl of Ser1186

E

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 4. Structure−Eﬃciency Relationships of Tribrid Phenpropyl Amide Analogues

a

b

c

d

e

LipE = pIC₅₀(TNKS2) − clogP. HEK293 SuperTopFlash reporter gene assay. Rat liver microsome extraction ratio. Measured log D. LipE =

pIC₅₀(TNKS2) − log D.

the case of 31 we speculate that the increased LipE may be

diagnostic of the formation of a hydrogen bond with the TNKS

protein, and the strength of this hydrogen bond is of similar

strength to that which was lost upon desolvation. From this

subset of compounds, we concluded that more speciﬁc

interactions with the TNKS protein and improved cellular

activity were possible, but the thiophene remained a key liability

that needed to be addressed. Yet again a more dramatic change

to the structure was needed, and we focused on completely

replacing the phenethyl moiety because of its low contribution

to the overall LipE of the scaﬀold.

Our screening and structural biology eﬀorts identiﬁed a

diverse range of TNKS binders such as 1, which bound to the

nicotinamide pocket,²⁴⁹compounds which bound exclusively to

the adenosine pocket,⁹and also compounds, such as 33,²⁷

which simultaneously occupied the nicotinamide and adenosine

pockets (Figure 3). A crystal structure of TNKS1 in complex

with 33 was determined to 2.2 Å resolution to understand the

binding mode. For this compound, a cyanobenzamide sits in

place of the A and B rings of 1 with the nitrile group forming

polar interactions with the hydroxyl of Ser1221 and the

backbone amide of Gly1185. Compound 33 extends through

the diphosphate binding region and reaches into the adenosine

pocket where the ﬂuorophenyl group forms van der Waals

interactions with the side chains of Phe1188, Ala1191, and

Ile1192 on one side of the groove, and a stacking interaction

with the side chain of His1201 on the opposite side. A

Figure 2. Crystal structure of 19 (pdb 4LI6) suggested potential polar

interactions that might be gained, as (1) the hydroxyl of Ser1186 and

the backbone carbonyl of D1198 were in close proximity to the phenyl

ring and (2) a crystallographic water molecule was bound via the

backbone NH of D1198.

next sought to replace the structural water with hydroxyl

substituents to make direct interactions with Asp1198. Each of

these analogues are more polar than 20 (ΔclogP = −0.7), and

therefore because of the desolvation penalty these analogues

would be expected to be less potent unless a compensating

hydrogen bond was made. Although compound 30 is less active

than 20 but with similar LipE, compounds 31 and 32 have

similar biochemical activity and increased LipE relative to 20. In

F

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

The oral and intravenous pharmacokinetics of 34 and the N-

methylthiophene analogue, 35, were evaluated in mice. After

intravenous administration, both compounds showed low CL

with a moderate volume of distribution. 34 was given orally

over a dose range from 5 to 800 mg/kg as a solution (5 and 25

mg/kg) or a suspension (200−800 mg/kg) dose. 34 exhibited

good oral exposure with a bioavailability of 40−60% over the

dose range. The exposure was slightly overproportional over

the range as evidenced by the dose normalized AUC’s ranging

from 400 to 704 for the 5 and 800 mg/kg doses, respectively.

The oral pharmacokinetics of 35 was evaluated as a solution

dose of 30 and 100 mg/kg. 35 showed signiﬁcantly lower oral

exposure as a result of the lower oral bioavailability of

approximately 12%.

We explored various substitution patterns on the thiophene

(Table 5), none of which improved the microsomal stability or

improved potency. The benzyl derivative (36) had a very slight

improvement in microsomal stability, however lost 18-fold

activity in the cellular assay. The heterocyclic replacements

(e.g., 37) had improvements in LipE, but as was the case for

compounds 13, 14, 16 and 17, these modiﬁcations are solvent-

exposed, modulate compound properties but do not represent

any improvement in binding eﬃciency to TNKS. We then

moved away from aromatic ring systems and then looked at a

series of alkyl substituents (38−41). For these analogues LipE

appeared inversely correlated with the bulkiness of the alkyl

substituent. Another trend appeared that rat liver microsomal

extraction ratio (rLMER) decreased as the compounds became

less lipophilic. Compounds 40 and 41 had a good combination

of LipE, potency, and rLMER. Because 38 was the most potent

of the alkyl derivatives, the N-cyclopropylmethyl analogue 42

was made and found to be equipotent but with improved

rLMER. The oral exposure of compound 42 in rat was 10-fold

greater than 34 (data not shown) and represented a suitable

thiophene replacement to enable further in vivo studies.

Turning once again to our structural data from our previously

reported TNKS inhibitor which binds exclusively to the

adenosine binding pocket, we noted that the ﬂuorophenyl

ring of 34 is in close proximity to the methoxyphenyl ring of 2

(Figure 4) and, despite not being an exact overlay, we

wondered if the methoxy would be a more favorable substituent

because the methoxy group signiﬁcantly improved LipE in the

previous series.⁹Replacing the ﬂuorine of 42 with a methoxy

group led to TNKS656 (compound 43) that was a highly

potent and eﬃcient inhibitor of TNKS1&2 (LipE = 7.0 based

on measured log D), with over 5,000-fold selectivity versus

Figure 3. Superposition of crystal structures 19 (pdb 4LI6, cyan), 33

(pdb 4LI7, green), and 35 (pdb 4LI8, magenta) bound to TNKS1.

Grafting of the nicotinamide/nook binding portion of 19 with the

adenosine binding portion of 33 led to 35, the ﬁrst known three-

pocket tankyrase binder, a compound with excellent potency, highly

favorable physicochemical properties, and metabolic stability.

hydrogen bond is also formed between 33 and the backbone

amide of Asp1198.

Relative to 1 and 20, compound 33 has much lower LipE but

has moderate biochemical and cellular activity. The cyano-

benzamide represented another novel nicotinamide isostere

that we found in our screens, and we hypothesized that if we

retained the high LipE fragment 23 it might be possible to ﬁnd

improvements by grafting on a new substructure, such as the

(4-ﬂuorophenyl)piperidin-4-ylmethanone moiety of 33. The

synthesis of these analogues was accomplished following the

approach outlined in Scheme 4. To our great satisfaction, the

ﬁrst two hybrid molecules based on this design (Table 5, 34

and 35) had excellent potency, substantially improved LipE,

and, in the case of 34 (LipE = 6.6 based on measured log D),

highly favorable physicochemical properties and metabolic

stability. While recent examples exist of compounds that bind in

the nicotinamide and adenosine pockets,¹⁶or to the adenosine

and “nook” pockets,^18,28to our knowledge, this is the ﬁrst

example of a tankyrase inhibitor that binds simultaneously to

the nicotinamide, adenosine, and “nook” pockets. We

benchmarked these compounds in multiple in vitro and in

vivo assays while further optimization continued.

a

Scheme 4.

a

(a) Ethyl 2-bromoacetate, Et₃N, CH₃CN, reﬂux 5 h; (b) NaOH, H₂O, EtOH, 24 h; (c) EDCI, HOBt or HATU, TEA or DIPEA, DCM or DMF.

G

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 5. Incorporation of Compound 33 into the Design of Three-Point Binders

a

b

c

d

e

LipE = pIC₅₀(TNKS2) − clogP. HEK293 SuperTopFlash reporter gene assay. Rat liver microsome extraction ratio. Measured log D. LipE =

pIC₅₀(TNKS2) − log D.

PARPs 1 and 2. With low to moderate microsomal ER values

across species and high solubility, compound 43 was progressed

into advanced in vivo studies.

To assess the potential for in vivo inhibition of Tankyrase

signaling by 43, we ﬁrst examined the pharmacokinetics in mice

(Table 6). After intravenous administration, 43 displayed low

clearance and volume of distribution (V_Dss) (10 mL/min/kg

and 0.6 L/kg, respectively). After oral administration of either a

30 or 100 mg/kg dose, 43 exhibited good exposure and

moderate oral bioavailability of 32% and 53%, respectively.

Some slight overproportional increase in oral exposure was

observed between 30 and 100 mg/kg with the dose normalized

AUC for the 100 mg/kg dose being 2-fold higher than for the

Figure 4. Superposition of the crystal structures of 35 (pdb 4LI8,

magenta) and 2 (pdb 3UDD, yellow).

H

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

select analogues from the optimization process to determine

what eﬀect the optimization had on the thermodynamics of

binding (Supporting Information Table 1, Figure 6). Karlberg

Table 6. Mouse Pharmacokinetic Properties of Compound

43

po

iv, 1 mg/kg

30 mg/kg

33.4

100 mg/kg

183.4

AUC (μM·h)

CL (mL/min/kg)

T1/2 (h)

3.46

9.7

1.3

V_dss(L/kg)

C_max(μM)

T_max(h)

0.6

16.8

0.4

32

79.7

0.8

53

F (%)

30 mg/kg dose. With a cell potency of 3.5 nM and a free

fraction of 16% in mouse from rapid equilibrium dialysis, 43

should have unbound drug concentrations exceeding the IC₅₀

for at least 12 h.

To investigate the in vivo PK/PD activity of 43, we utilized

mammary adenocarcinomas derived from the mouse mammary

tumor virus (MMTV)-Wnt1 transgenic model resulting in Wnt

signaling activation (Figure 5).²⁹Tumor fragments from these

Figure 6. Isothermal calorimetry data for select compounds.

et al. determined that 1 binds with both favorable enthalpic and

entropic contributions, although entropic contributions are

more signiﬁcant.³²Compound 7, despite having TNKS2

aﬃnity lower than that of 1, has a similar LipE, but binding

is enthalpy-driven (ΔH = −10.3 kcal/mol) with a slight

entropic penalty (TΔS = −1.1 kcal/mol). Compound 20 was

the ﬁrst analogue that went beyond the ABC ring system of

XAV939 and was only 3-fold less potent than 7. The improved

thermodynamic proﬁle (ΔH = −18.3 kcal/mol) is likely due in

part to the additional hydrogen bond gained to the amide

carbonyl. 20 has a secondary amide bond in the linker, and

therefore we hypothesize a portion of the large unfavorable

entropic penalty (TΔS = −9.9 kcal/mol) must be paid for the

s-cis amide conformation to bind. Further optimization into the

hybrid 34 beneﬁts from an additional hydrogen bond (ΔH =

−18.7 kcal/mol) while the entropic penalty is only slightly

diminished. The potency gain with 42 is attributed to potency

gained from desolvation of the hydrophobic cyclopropylmethyl

moiety, while the terminal methoxy group of 43 again shifts to

a more enthalpy-driven binding signature. The high enthalpy of

binding is reﬂected in the high selectivity of this class of

compounds against PARPs 1 and 2 as well as a panel of oﬀ-

target activities (data not shown).

Figure 5. PK/PD relationship of 43 in MMTV-Wnt1 tumor bearing

athymic nude mice. Following a single oral dose of 43 at 350 mg/kg,

plasma and tumor samples were collected at various time points (0.5,

1, 2, 4, 8, 16, and 24 h; n = 3) for PK and PD analysis. The data are

represented as 43 concentration in plasma (nM) and %Axin2 mRNA

in tumor relative to untreated controls (mean SEM, n = 3 animals/

time point). p < 0.05 compared to untreated controls using a one-way

ANOVA followed by a posthoc Tukey test.

mice were allografted subcutaneously into recipient nude mice

and grown to approximately 250−300 mm.³To determine if 43

could modulate Wnt/beta-catenin signaling, cohorts of

MMTV-Wnt1 tumor bearing mice (n = 3) were treated with

either vehicle or a single oral dose of 43 at 350 mg/kg over a

24-h time course. Mice treated with 43 showed good plasma

and tumor exposures corresponding to AUC_0−24hof 515 and

325 μM·h, respectively. Consistent with the proposed

mechanism of action of TNKS inhibition, 43-treated tumors

showed a stabilization of Axin1 protein (data not shown) and a

corresponding 70−80% reduction in the Wnt/beta-catenin

target gene Axin2 mRNA level compared to vehicle control

animals. The decrease in Axin2 mRNA expression was detected

as early as 4 h post treatment and persisted through the 24 h

time course, suggesting once daily dosing may be suﬃcient for

sustained target inhibition. Additional studies will be needed to

further model the dose−eﬀect relationship in this model.

The thermodynamics of inhibitor binding, especially

enthalpy of binding, has been identiﬁed as an important factor

in identifying quality compounds.³⁰Using isothermal calorim-

etry (ITC), we measured the thermodynamic signature of

CONCLUSION

■

We previously reported the identiﬁcation of novel inhibitors of

TNKS which is a druggable node in the Wnt pathway.

Beginning with compound 1, we explored the previously

untested approach of utilizing LipE to survey a diverse range of

chemical structures to understand what portions of each

molecule eﬃciently interacted with our target protein. With this

information we undertook a LipE-driven optimization approach

at several junctures. We consciously deprioritized modiﬁcations

to achieve more potent compounds in favor of modiﬁcations to

obtain less potent, but more lipophilic, eﬃcient inhibitors. We

went from compound 1 (cellular IC₅₀of 0.078 μM) to

compound 7 (cellular IC₅₀of 2.7 μM) to compound 20

(cellular IC₅₀of 1.9 μM) before ﬁnally optimizing to 43

(cellular IC₅₀of 0.0035 μM) with demonstrated in vivo activity

for proof of concept studies. The thermodynamic proﬁle of key

analogues generated during the optimization demonstrates that

I

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

TNKS1/PJ34 complex crystals were transferred into a soaking solution

containing 100 mM Bis-Tris pH 5.8, 18% PEG 3350, 320 mM

ammonium sulfate, and 200 μM of compound. Transfer of the crystals

into fresh solution occurred once per hour over a 4 h period. Crystals

were cryoprotected using the same soaking solution with the addition

of 20% glycerol, followed by ﬂash-freezing directly into liquid nitrogen.

Data Collection and Structure Determination. Diﬀraction data

for the TNKS1/19 complex were collected on a Saturn92 CCD

Detector using Cu Kα radiation (λ = 1.5418 Å) from a Rigaku FR-E

rotating anode X-ray generator operating at 45 mV and 45 mA.

Diﬀraction data for the TNKS1/33 complex were collected on a

Quantum315R CCD Detector at the ALS beamline BL5.0.3 at a

wavelength of 1 Å. Diﬀraction data for the TNKS1/35 complex were

collected on a Dectris Pilatus 6 M Detector at the Swiss Light Source

beamline X06SA at a wavelength of 1 Å. The data were measured from

a single crystal, and the reﬂections were indexed, integrated, and scaled

using XDS and XSCALE²⁹or HKL2000.²⁵The space group of the

complex was C2 with two molecules in the asymmetric unit. The

structure was determined with Fourier methods, using the TNKS1/

PJ34 structure (pdb 3UH2) with active site waters and inhibitors

removed as the starting model.

Structure determination was achieved through iterative rounds of

positional and simulated annealing reﬁnement using CNX²⁶or

PHENIX²⁷with model building using COOT²⁸Individual B-factors

were reﬁned, and a bulk solvent correction was applied. The solvent,

sulfate ions, and inhibitor were built into the density in later rounds of

the reﬁnement. Data collection and reﬁnement statistics are shown in

Supporting Information Table 2.

Isothermal Calorimetry. Isothermal calorimetry was used to

determine the thermodynamic properties of ligand binding. Data were

collected at 25 °C on an Auto-iTC₂₀₀(GE Healthcare) using a sample

buﬀer of 25 mM HEPES pH 7.4, 150 mM NaCl, with a matched

DMSO concentration between the titrant and cell. Direct titration of

200 μM compound into 20 μM TNKS1 over 16 injections of 2.5 μL

was performed in the case of 34, 42, and 43. Direct titration of 300

μM compound into 20 μM TNKS1 over 20 injections of 2 μL was

performed in the case of 20. Finally, a reverse titration using 200 μM

TNKS1 into 20 μM 7 over 16 injections of 2.5 μL was performed. The

individual heat values were plotted against the molar ratio, and the

values for the number of binding sites (n), the enthalpy change (ΔH),

and dissociation constant (K_D= 1/K_A) were obtained by nonlinear

regression of the data.

Chemical Synthesis. The purity of all compounds was determined

using high resolution HPLC/MS characterization and found to be

>95% pure except as noted. High resolution LC/ESI-MS data were

recorded using an Agilent 6220 mass spectrometer with electrospray

ionization source and an Agilent 1200 liquid chromatograph with a

gradient from 5% to 95% acetonitrile in water on a C18 reverse phase

column with a diode array detection. The resolution of the MS system

was approximately 11000 (fwhm deﬁnition). HPLC separation was

performed using one of the methods (denoted in the header text)

listed at end of report. Purine and hexakis(1H,1H,3H-

tetraﬂuoropropoxy)phosphazine (protonated molecules m/z

121.05087 and 922.00979, respectively) were used as a reference.

The mass accuracy of the system has been found to be <2 ppm.

General Procedure A: Synthesis of Benzamidines. 4-

(Triﬂuoromethoxy)benzimidamide (15a). Trimethylaluminum (2 M

in toluene, 2.4 mL, 4.8 mmol, 3 equiv) was added to a ca. 5 °C

suspension of ammonium chloride (257 mg, 4.8 mmol, 3 equiv) in 8

mL of toluene. The reaction mixture was then stirred in the cold bath

for 1.5 h and then at room temperature for 0.5 h. 4-

(Triﬂuoromethoxy)benzonitrile (300 mg, 1.6 mmol, 1.0 equiv) was

then added, and the reaction mixture was heated to reﬂux overnight.

The reaction mixture was cooled to room temperature, quenched with

MeOH (30 mL), and cooled in an ice bath, and 2 g of silica gel in

DCM with 1% MeOH was added. The mixture was stirred for 15 min,

and the solid was collected and then washed with excess MeOH. The

ﬁltrate was concentrated in vacuo to give a white solid (327 mg, 1.6

mmol, 100% yield) and used without further puriﬁcation.

our optimization created inhibitors with enthalpy-driven

binding. It is possible that a LipE-driven optimization approach

may be a more general approach for yielding enthalpy-driven

inhibitors, but further studies investigating this are warranted.

For the case of TNKS, LipE was a valuable decision-making

tool for the rapid generation of potent, selective, and in vivo

active compounds for target validation and lead optimization.

MATERIALS AND METHODS

■

Tankyrase AutoPARsylation Assay. PARP catalytic activity was

monitored using the quantitative liquid chromatography/mass

spectrometry (LC-MS) detection of nicotinamide, as previously

described.⁷The autoPARsylation reactions were performed at room

temperature in 384-well Greiner ﬂat-bottom plates. The ﬁnal reaction

mixture contained 2.5% DMSO and inhibitors with concentrations

ranging from 0.0001 to 18.75 μM. GST-TNKS2P, GST-TNKS1P,

PARP1, and PARP2 enzymes were used at ﬁnal concentrations or 5, 5,

5, and 2 nM, respectively. The nicotinamide concentration in the

resulting supernatants was measured by LC-MS. The % inhibition was

calculated as follows: (control − sample)/(control − background) ×

100. “Control” is the average value of eight wells without compound,

and “background” is the average of eight wells mixed with 5×

quenching solution measured prior to initiation of the reaction.

SuperTop Flash (STF) Reporter Gene Assay. Compound

activity in inhibiting Wnt ligand-induced signaling was measured

using a Wnt-responsive SuperTopFlash (STF) luciferase reporter gene

assay in HEK293 cells. The % inhibition was calculated as follows:

(maximum Wnt-induced signaling − sample)/(maximum Wnt-

induced signaling − background) × 100. “Maximum Wnt-induced

signaling” is the STF signal level induced by 20% Wnt3A CM without

compound, and “background” is the STF signal level without the

addition of Wnt3A CM or compound. To demonstrate the speciﬁc

activity of inhibitors regulating Wnt signaling, counter screen was

performed in HEK293T cells expressing a cAMP-response element

(CRE) luciferase reporter gene. Compound activity on the CRE-

reporter was measured in the presence of 10 μM forskolin, which is an

activator of cAMP signaling.

Axin2 Protein ELISA. Compound activity in stabilizing the Axin2

protein was measured by Sandwich enzyme-linked immunosorbent

(ELISA) assay in the colorectal cell line SW480. Cell lysates were

prepared from cells treated with compounds in six-point dilution

starting at 10 μM for 24 h. For the ELISA assay, anti Axin-2 capture

antibody was diluted to a concentration of 1 μg/mL (1:1000) in

carbonate coating buﬀer, pH 9.2 (Sigma, C3041−50CAP). Then 100

μL of the diluted anti Axin-2 capture antibody per well was used to

coat the 96-well plate overnight at 4 °C. Plates were then washed three

times with wash solution, PBST20 (PBS + 0.05% Tween), and blocked

with 1% BSA/PBS for 1.5 h at room temperature while shaking gently.

After blocking, plates were then washed three times with wash

solution. Then 100 μL of prepared SW480 cell lysate was added to

each well and incubated at room temperature for 2 h while shaking

gently. After washing, 100 μL of biotinylated anti-Axin2 antibody was

added to each well and incubated at room temperature for 2 h. Signal

was detected by chemiluminescence (Pierce SuperSignal ELISA Femto

no. 3704) using streptavidin-HRP (R&D Systems, DY998) and

measured on PerkinElmer Wallac 1420 plate reader.

MMTV-Wnt1 Allograft Model. Athymic female nude mice

(Harlan Laboratories) weighing 19−22 g were implanted subcuta-

neously with a 3 × 3 × 3 mm³tumor fragment from an MMTV-Wnt1

tumor-bearing mouse. Tumors were grown to approximately 250−300

mm³. Individual mice were given a single oral dose of vehicle (n = 3)

(4% HCl:10% propylene glycol:20% Solutol HS15:60.5% D5W:0.5%

NaOH) or TNKS656 at 350 mg/kg (n = 18). At 0.5, 1, 2, 4, 8, 16, or

24 h following dosing (n = 3/time point), mice were euthanized, and

blood was collected via cardiac puncture and processed for plasma.

Tumors were excised from mice and frozen at −80 °C for PD analysis.

Protein Puriﬁcation and Crystallization. Human TNKS1

protein was expressed, puriﬁed, and crystallized as previously

described.²⁴To obtain crystals containing either 19, 33, or 35,

J

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

General Procedure B: Synthesis of Pyrimidinones. 2-(4-

(Triﬂuoromethyl)phenyl)-7,8-dihydro-3H-thiopyrano[4,3-d]-

pyrimidin-4(5H)-one (1). K₂CO₃(1.42 g, 10.2 mmol, 2.0 equiv) was

added to a room temperature mixture of methyl 4-oxotetrahydro-2H-

thiopyran-3-carboxylate³¹(876 mg, 5.03 mmol) and 4-

(triﬂuoromethyl)benzimidamide hydrochloride dihydrate (1.42 g,

5.46 mmol, 1.08 equiv) in methanol (25 mL). The reaction mixture

was then reﬂuxed overnight. The reaction mixture was then ﬁltered

with the aid of methanol and then concentrated to ca. 1/4 volume.

The resulting material was dissolved in water and acidiﬁed to ca. pH 3

using 1 N aqueous HCl. The resulting precipitate was ﬁltered and

washed several times with water/MeOH (1:1) followed by several

washes with ether. The resulting white solid was dried in a high

vacuum oven (35 °C) for several hours, aﬀording 1 (1.24 g, 79%) as a

white solid. Exact mass calculated for C₁₄H₁₂F₃N₂OS 313.0622, found

313.0628 (ESI, M + H). 1H NMR (400 MHz, DMSO-d₆) δ ppm

12.94 (br s, 1 H) 8.28 (d, J = 8.08 Hz, 2 H) 7.89 (d, J = 8.59 Hz, 2 H)

3.55 (s, 2 H) 2.91 (s, 4 H)

Ethyl 3-Oxotetrahydro-2H-thiopyran-2-carboxylate (3a). One

drop of MeOH was added to a room temperature suspension of

NaH in Et₂O (2 mL). A solution of ethyl 4-((2-ethoxy-2-oxoethyl)-

thio)butanoate in Et₂O (4 mL) was then added, and the reaction

mixture was heated at reﬂux for 5 h. The reaction mixture was cooled

to room temperature and then quenched with 3 N aq AcOH. After

separation, the aqueous layer was extracted with Et₂O (2 × 80 mL).

The combined organic layers were washed several times with sat.

NaHCO₃solution, dried over MgSO₄, ﬁltered, and concentrated in

vacuo to dryness aﬀording ethyl 3-oxotetrahydro-2H-thiopyran-2-

carboxylate as a brown oil which was used without further puriﬁcation.

Mass calculated for C₁₄H₁₁F₃N₂OS 312.1, found 313.0 (ESI, M + H).

¹H NMR (400 MHz, DMSO-d₆) δ ppm 12.94 (br s, 1H), 8.26 (d, J =

13.13 (s, 1H), 8.68 (d, J = 8.59 Hz, 1H), 8.44 (br s, 1H), 8.13 (d, J =

8.08 Hz, 2H), 7.98 (d, J = 8.59 Hz, 2H), 7.92 (dd, J = 1.26, 7.83 Hz,

1H), 7.82−7.89 (m, 1H), 7.55−7.63 (m, 1H), 7.17−7.25 (m, 1H).

2-(4-(Triﬂuoromethyl)phenyl)quinazolin-4(3H)-one (6). In a 5 mL

microwave vial, N-(2-carbamoylphenyl)-4-(triﬂuoromethyl)benzamide

(226 mg, 0.73 mmol, 1.0 equiv) and KOEt (123 mg, 1.46 mmol, 2.0

equiv) were heated in the microwave at 140 °C for 20 min. The

mixture was poured into 1 N HCl, and the resulting white solid was

collected, washed with heptane, and air-dried to provide the title

compound (202 mg, 0.696 mmol, 95% yield). Mass calculated for

1

C₁₅H₉F₃N₂O 290.1, found 291.2 (ESI, M + H). H NMR (400 MHz,

DMSO-d₆) δ 12.73 (br s, 1H), 8.38 (d, J = 8.08 Hz, 2H), 8.18 (dd, J =

1.52, 8.08 Hz, 1H), 7.93 (d, J = 8.59 Hz, 2H), 7.84−7.90 (m, 1H),

7.77−7.81 (m, 1H), 7.54−7.60 (m, 1H).

2-(4-(Triﬂuoromethyl)phenyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (7). Following general procedure B, the title

compound (190 mg, 35% yield) was prepared from methyl 4-

oxotetrahydro-2H-pyran-3-carboxylate (290 mg, 1.83 mmol) and 4-

(triﬂuoromethyl)benzimidamide hydrochloride dihydrate hydrochlor-

ide dihydrate (526 mg, 2.02 mmol). Mass calculated for

1

C₁₄H₁₁F₃N₂O₂296.08, found 297.0 (ESI, M + H). H NMR (400

MHz, DMSO-d₆) δ ppm 12.93 (br s, 1H), 8.29 (d, J = 8.08 Hz, 2H),

7.89 (d, J = 8.08 Hz, 2H), 4.46 (s, 2H), 3.91 (t, J = 5.56 Hz, 2H), 2.70

(br s, 2H).

2-(4-(Triﬂuoromethyl)phenyl)-5,6-dihydro-3H-pyrano[3,4-d]-

pyrimidin-4(8H)-one (8). Following general procedure B, ethyl 3-

oxotetrahydro-2H-pyran-4-carboxylate (100 mg, 0.436 mmol, 75%

pure) and 4-(triﬂuoromethyl)benzimidamide (117 mg, 0.523 mmol)

were converted to the title compound (82 mg, 0.277 mmol, 64% yield)

as an oﬀ-white solid. MS m/z 297.2 (M + H), Ret time: 1.23 min;

HRMS m/z (M + H) calculated for C₁₄H₁₁F₃N₂O₂: 297.0851, found:

1

8.1 Hz, 2H), 7.87 (d, J = 8.6 Hz, 2H), 2.99 (td, J = 2.9, 5.3 Hz, 2H),

2.76 (t, J = 6.1 Hz, 2H), 2.14−2.01 (m, 2H).

297.0846. H NMR (400 MHz, DMSO-d₆) δ 12.20−13.39 (m, 1H),

8.26 (d, J = 8.03 Hz, 2H), 7.90 (d, J = 8.03 Hz, 2H), 4.47 (s, 2H), 3.87

(t, J = 5.52 Hz, 2H), 2.50 (br s, 2H)

2-(4-(Triﬂuoromethyl)phenyl)-7,8-dihydro-3H-thiopyrano[3,2-d]-

pyrimidin-4(6H)-one (3). Following general procedure B, the title

compound (64 mg, 0.21 mmol, 29% yield) was prepared from ethyl 3-

oxotetrahydro-2H-thiopyran-2-carboxylate (142 mg, 0.72 mmol, 1.0

equiv) and 4-(triﬂuoromethyl)benzimidamide (200 mg, 0.77 mmol,

1.1 equiv). Mass calculated for C₁₄H₁₁F₃N₂OS 312.1, found 313.0

2-(Thiophen-3-yl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-4(5H)-

one (9). Following general procedure B, methyl 4-oxotetrahydro-2H-

pyran-3-carboxylate (100 mg, 0.474 mmol, 75% pure) and thiophene-

3-carboximidamide (93 mg, 0.599 mmol) were converted to the title

compound (73 mg, 0.296 mmol, 62% yield) as a yellow solid. MS m/z

235.2 (M + H), Ret time: 0.78 min; HRMS m/z (M + H) calculated

1

(ESI, M + H). H NMR (400 MHz, DMSO-d₆): δ 12.94 (br s, 1H),

1

8.26 (d, J = 8.1 Hz, 2H), 7.87 (d, J = 8.6 Hz, 2H), 2.99 (td, J = 2.9, 5.3

Hz, 2H), 2.76 (t, J = 6.1 Hz, 2H), 2.14−2.01 (m, 2H).

for C₁₁H₁₀N₂O₂S: 235.0541, found: 235.0549. H NMR (400 MHz,

DMSO-d₆) δ 12.64 (br s 1H), 8.50 (s, 1H), 7.76 (d, J = 4.96 Hz, 1H),

7.66−7.70 (m, 1H), 4.41 (s, 2H), 3.88 (t, J = 5.51 Hz, 2H), 2.60−2.66

(m, 2H)

2-(4-(Triﬂuoromethyl)phenyl)-4a,5,7,7a-tetrahydrothieno[3,4-d]-

pyrimidin-4(3H)-one (4). Following general procedure B, the title

compound (14.7 mg, 0.05 mmol, 9% yield) was prepared from methyl

4-oxotetrahydrothiophene-3-carboxylate (95 mg, 0.56 mmol, 1.0

equiv) and 4-(triﬂuoromethyl)benzimidamide hydrochloride dihydrate

(200 mg, 0.62 mmol, 1.1 equiv). Mass calculated for C₁₃H₉F₃N₂OS

298.04, found 299.0 (ESI, M + H). ¹H NMR (400 MHz, DMSO-d₆) δ

ppm 13.02 (br s, 1H), 8.29 (d, J = 7.6 Hz, 2H), 7.90 (d, J = 8.6 Hz,

2H), 4.23 (d, J = 3.0 Hz, 2H), 4.05 (d, J = 3.0 Hz, 2H).

2-(p-Tolyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-4(5H)-one

(10). Following general procedure B, methyl 4-oxotetrahydro-2H-

pyran-3-carboxylate (100 mg, 0.474 mmol, 75% pure) and 4-

methylbenzimidamide (76 mg, 0.569 mmol) were converted to the

title compound (40 mg, 0.165 mmol, 35% yield) as an oﬀ-white solid.

MS m/z 243.3 (M + H), Ret time: 1.03 min; HRMS m/z (M + H)

calculated for C₁₄H₁₄N₂O₂: 243.1134, found: 243.1138. ¹H NMR (400

MHz, DMSO-d₆) δ 12.65 (br s, 1H), 8.00 (d, J = 8.08 Hz, 2H), 7.33

(d, J = 8.08 Hz, 2H), 4.43 (s, 2H), 3.89 (t, J = 5.56 Hz, 2H), 2.66 (t, J

= 5.31 Hz, 2H), 2.38 (s, 3H).

2-(4-(Triﬂuoromethyl)phenyl)-5,6,7,8-tetrahydroquinazolin-

4(3H)-one (5). Following general procedure B, the title compound (53

mg, 0.18 mmol, 62% yield) was prepared from ethyl 2-oxocyclohex-

anecarboxylate (52 mg, 0.29 mmol, 1.0 equiv) and 4-(triﬂuoromethyl)-

benzimidamide hydrochloride dihydrate (100 mg, 0.38 mmol, 1.3

equiv). Exact mass calculated for C₁₅H₁₃F₃N₂O 294.1, found 295.1

2-(4-Chlorophenyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-

4(5H)-one (11). Following general procedure B, methyl 4-oxotetrahy-

dro-2H-pyran-3-carboxylate (100 mg, 0.474 mmol, 75% pure) and 4-

chlorobenzimidamide (161 mg, 0.569 mmol) were converted to the

title compound (80 mg, 0.283 mmol, 60% yield) as a yellow solid. MS

m/z 263.2 (M + H), Ret time: 0.98 min; HRMS m/z (M + H)

1

(ESI, M + H). H NMR (400 MHz, DMSO-d₆): δ ppm 12.73 (br s,

1H), 8.28 (d, J = 7.6 Hz, 2H), 7.87 (d, J = 8.1 Hz, 2H), 2.70−2.57 (m,

2H), 2.42 (br s, 2H), 1.83−1.63 (m, 4H).

1

N-(2-Carbamoylphenyl)-4-(triﬂuoromethyl)benzamide (6a). In a

20 mL vial, 4-(triﬂuoromethyl)benzoyl chloride (184 mg, 0.88 mmol,

1.2 equiv) was added to 2-aminobenzamide (100 mg, 0.73 mmol, 1.0

equiv) and DIEA (192 uL, 1.1 mmol, 1.5 equiv) in THF, and the

reaction was stirred for 2 h resulting in a white precipitate. The

reaction mixture was diluted with heptane, and the solid was collected,

triturated with water, collected again, and then air-dried to give a white

solid. The title compound (226 mg, 0.733 mmol, 100% yield) was

used without further puriﬁcation. Mass calculated for C₁₅H₁₁F₃N₂O₂

calculated for C₁₃H₁₁ClN₂O₂: 263.0587, found: 263.0587. H NMR

(400 MHz, DMSO-d₆) δ 12.61 (br s 1H), 8.11 (d, J = 8.82 Hz, 2H),

7.58 (d, J = 7.71 Hz, 2H), 4.42 (s, 2H), 3.89 (t, J = 5.51 Hz, 2H),

2.69−2.62 (m, 2H).

2-(4-Bromophenyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-

4(5H)-one (12). Following general procedure B, methyl 4-oxotetrahy-

dro-2H-pyran-3-carboxylate (300 mg, 2.53 mmol, 75% pure) and 4-

bromobenzimidamide (536 mg, 2.28 mmol) were converted to the

title compound (183 mg, 0.554 mmol, 29% yield) as a white solid. MS

m/z 307.1 (M + H), Ret time: 1.11 min; HRMS m/z (M + H)

1

308.1, found 309.1 (ESI, M + H). H NMR (400 MHz, DMSO-d₆) δ

K

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

1

calculated for C₁₃H₁₁BrN₂O₂: 307.0082, found: 307.0070. H NMR

(400 MHz, DMSO-d₆) δ 8.03 (d, J = 8.46 Hz, 2H), 7.74 (d, J = 8.72

Hz, 2H), 7.25 (br s, 1H), 4.43 (s, 2H), 3.90 (t, J = 5.62 Hz, 2H), 2.67

(t, J = 5.31 Hz, 2H).

mL) was cooled to 0 °C under N₂. A solution of 2-chloroacetamidine

(567 g, 4.4 mol) in 890 mL of MeOH was added dropwise over 50

min. The reaction mixture was stirred at 0 °C for 30 min and then at

about 20 °C for 16 h. LCMS at 215 nm and TLC (DCM:MeOH =

10:1) analysis showed that most of 4-oxotetrahydropyran-3-carboxylic

acid methyl ester was consumed. The mixture was then ﬁltered and

concentrated to give a black oil, which was subsequently puriﬁed by

ﬂash column chromatography on silica gel and eluted with DCM to

give a yellow solid/oil mixture, which was further triturated with

MTBE (∼1200 mL) and H₂O:CH₃CN:EA = 1:1:2 (∼600 mL) to give

the title compound as a white solid (318 g, 1.59 mol, 36% yield). MS

m/z 201.2 (M + H). CHN analysis: calculated (results). C 47.89

(47.95), H 4.52 (4.401), N 13.96 (13.76).

2-(4-(Dimethylamino)phenyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (13). Following general procedure B, ethyl 3-

oxotetrahydro-2H-pyran-4-carboxylate (100 mg, 0.290 mmol, 50%

pure) and 4-(dimethylamino)benzimidamide (76 mg, 0.348 mmol)

were converted to the title compound (10 mg, 0.038 mmol, 13% yield)

as a yellow solid. MS m/z 272.4 (M + H), Ret time: 1.05 min; HRMS

m/z (M + H) calculated for C₁₅H₁₇N₃O₂: 272.1399, found: 272.1406.

¹H NMR (400 MHz, DMSO-d₆) δ 12.32 (br s, 1H), 8.00 (d, J = 9.03

Hz, 2H), 6.75 (d, J = 9.03 Hz, 2H), 4.40 (s, 2H), 3.79−3.92 (m, 2H),

2-(Azidomethyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-4(5H)-

one (46). To a solution of 45 (20.6 g, 103 mmol) in acetone (513 mL)

was added an aqueous solution of sodium azide (7.01 g, 108 mmol in

32 mL of water). The reaction was stirred at room temperature for 36

h. Mixture was split into three equal volumes and each was diluted

with dichloromethane (500 mL) and brine was added (100 mL) and

the layers were separated. The aqueous fractions were combined and

extracted with dichloromethane (5 × 100 mL). The combined organic

layers were concentrated in vacuo. The resulting residue was taken up

in dichloromethane (200 mL) and evaporated to dryness three times

3.00 (s, 6H), 2.61 (t, J = 5.27 Hz, 2H).

2-(4-Methoxyphenyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-

4(5H)-one (14). Following general procedure B, methyl 4-oxotetrahy-

dro-2H-pyran-3-carboxylate (100 mg, 0.474 mmol, 75% pure) and 4-

methoxybenzimidamide (85 mg, 0.569 mmol) were converted to the

title compound (28 mg, 0.108 mmol, 23% yield) as an oﬀ-white solid.

MS m/z 259.3 (M + H), Ret time: 0.97 min; HRMS m/z (M + H)

calculated for C₁₄H₁₄N₂O₃: 259.1083, found: 259.1077. ¹H NMR (400

MHz, DMSO-d₆) δ 8.09 (d, J = 8.97 Hz, 2H), 7.0 (d, J = 8.97 Hz, 2H),

4.42 (s, 2H), 3.89 (t, J = 5.56 Hz, 2H), 3.84 (s, 3H), 2.62−2.67 (m,

2H).

2-(4-(Triﬂuoromethoxy)phenyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (15). Following general procedure B, the title

compound (24 mg, 0.075 mmol, 16% yield) was prepared from 15a

(75 mg, 0.47 mmol, 1.0 equiv) and 4-(triﬂuoromethoxy)-

benzimidamide (145 mg, 0.711 mmol, 1.5 equiv). Mass calculated

1

to yield 19.25 g (92 mmol, 90% yield) of a yellow-white solid. H

NMR (400 MHz, DMSO-d₆): δ 12.58 (br s, 1H), 4.36 (s, 2H), 4.28 (s,

2H), 3.85 (t, J = 5.56 Hz, 2H), 2.59 (t, J = 5.56 Hz, 2H).

2-(Aminomethyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-4(5H)-

one (47). To a solution of 2-(azidomethyl)-7,8-dihydro-3H-pyrano-

[4,3-d]pyrimidin-4(5H)-one (19 g, 92 mmol) in THF (945 mL) at 35

°C was added triphenylphosphine (26.5 g, 101 mmol). Gas evolution

was noted immediately and after approximately 40 min a ﬁne

precipitate formed. The reaction was allowed to stir for 1.5 h then

18.17 mL water was added. After an addition 2 h the reaction mixture

was cooled to room temperature, ﬁltered twice, concentrated to

approximately 1/3 of the original volume and ﬁltered once more. The

solid material was washed with THF to yield a total of 15.82 g (87

mmol, 95% yield) of the title compound. ¹H NMR (400 MHz,

MeOD): δ 4.49 (s, 2H), 3.94 (t, J = 5.56 Hz, 2H), 3.76 (s, 2H), 2.66

(t, J = 5.56 Hz, 2H).

1

for C₁₄H₁₁F₃N₂O₃312.2, found 313.1 (ESI, M + H). H NMR (400

MHz, DMSO-d₆) δ 12.82 (br s, 1H), 8.21 (d, J = 7.58 Hz, 2H), 7.51

(d, J = 8.59 Hz, 2H), 4.44 (s, 2H), 3.90 (t, J = 5.56 Hz, 2H), 2.67 (br s,

2H).

Methyl 4-(4-Oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-

2-yl)benzoate (16). Following general procedure B, the title

compound (601 mg, 80% yield) was prepared from methyl 4-

oxotetrahydro-2H-pyran-3-carboxylate (402 mg, 2.63 mmol) and

methyl 4-carbamimidoylbenzoate hydrochloride (0.672 g, 3.13

mmol). Mass calculated for C₁₅H₁₄N₂O₄286.29, found 287.2 (ESI,

M + H). 1H NMR (400 MHz, DMSO-d6) δ ppm 12.86 (br s, 1H),

8.22 (d, J = 8.08 Hz, 2H), 8.07 (d, J = 8.59 Hz, 2H), 4.45 (s, 2H),

3.87−3.93 (m, 5H), 2.69 (br s, 2H).

2-(4-(Methylsulfonyl)phenyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (17). Following general procedure B, methyl 4-

oxotetrahydro-2H-pyran-3-carboxylate (50 mg, 0.316 mmol) and 4-

(methylsulfonyl)benzimidamide (89 mg, 0.379 mmol) were converted

to the title compound (9 mg, 0.029 mmol, 9% yield) as white solid.

MS m/z 307.3 (M + H), Ret time: 0.72 min; HRMS m/z (M + H)

calculated for C₁₄H₁₄N₂O₄S: 307.0753, found: 307.0751. ¹H NMR

(400 MHz, DMSO-d₆) δ 13.01 (br s, 1H), 8.30 (br s, 2H), 8.07 (d, J =

8.53 Hz, 2H), 4.46 (br s, 2H), 3.91 (t, J = 5.27 Hz, 2H), 3.30 (s, 3H),

2.70 (br s, 2H).

2-(4-(2-Hydroxypropan-2-yl)phenyl)-7,8-dihydro-3H-pyrano[4,3-

d]pyrimidin-4(5H)-one (18). MeMgBr (3 M in Et2O, 780 μL, 2.34

mmol) was added to a ca. −78 °C mixture of compound 16 (0.99 mg,

0.346 mmol) in THF (3 mL). After complete addition, the reaction

mixture was removed from the cold bath, stirred, and allowed to warm

to room temperature over 1 h. The reaction mixture was cooled in an

ice bath and slowly quenched with 1 N HCl (ca. 5 mL). The reaction

mixture was then diluted with water and extracted with EtOAc (5 × 10

mL). The combined organic layers were dried over MgSO₄, ﬁltered,

and concentrated in vacuo to dryness, aﬀording the title compound as

an oﬀ-white solid (83 mg, 84% yield). Mass calculated for C₁₆H₁₈N₂O₃

286.33, found 287.2 (ESI, M + H). 1H NMR (400 MHz, DMSO-d₆) δ

ppm 12.63 (br s, 1H), 8.02 (d, J = 8.59 Hz, 2H), 7.59 (d, J = 8.59 Hz,

2H), 5.13 (s, 1H), 4.43 (s, 2H), 3.89 (t, J = 5.56 Hz, 2H), 2.66 (t, J =

5.05 Hz, 2H), 1.45 (s, 6H).

N-((4-Oxo-3,4-dihydroquinazolin-2-yl)methyl)-3-phenyl-N-(thio-

phen-2-ylmethyl)propanamide (19). To a solution of 2-(((thiophen-

2-ylmethyl)amino)methyl)quinazolin-4(3H)-one (100 mg, 0.369

mmol) in methylene chloride (4 mL) were added triethylamine

(113 μL, 0.811 mmol, 2.2 equiv) and 3-phenylpropanoyl chloride (75

mg, 0.442 mmol, 1.2 equiv), and the solution was stirred 4 h. The

reaction mixture was concentrated in vacuo, taken up in EtOAc,

washed with brine, dried over Na₂SO₄, and concentrated in vacuo to

give a white solid. The crude product was absorbed onto silica gel and

added to an Analogix SF15-12g column, and compound was eluted

with 1% to 10% MeOH in DCM over 12 column volumes on a

Biotage SP1. This product was absorbed onto silica gel and added to

an Biotage SNAP-10g column, and the compound was eluted with 5%

to 80% EtOAc in dichloromethane over 12 column volumes on a

Biotage SP1. The concentrated fractions were recrystallized from

diethyl ether, washed with hepatne, and dried under vacuum to give

the title compound (30 mg, 0.074 mmol, 20% yield). Mass calculated

1

for C₂₃H₂₁N₃O₂S 403.1, found 404.3 (ESI, M + H). H NMR (400

MHz, chloroform-d) δ 10.13 (br s, 1H), 8.28 (d, J = 7.58 Hz, 1H),

7.71−7.78 (m, 1H), 7.64 (d, J = 8.08 Hz, 1H), 7.45−7.53 (m, 1H),

7.14−7.25 (m, 6H), 6.88−6.95 (m, 2H), 4.66 (s, 2H), 4.51 (s, 2H),

3.03−3.10 (m, 2H), 2.80−2.87 (m, 2H).

General Procedure C. 2-(((Thiophen-2-ylmethyl)amino)methyl)-

7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-4(5H)-one (23). A mixture

of thiophene-2-carbaldehyde (0.021 mL, 0.22 mmol), 47 (40 mg, 0.22

mmol), and siliabond cyanoborohydride (240 mg, 0.24 mmol, 1

mmol/g) in ethanol (1 mL) was stirred at ambient temperature for 10

min. Acetic acid (0.013 mL, 0.22 mmol) was then added, and the

mixture was stirred at ambient temperature for another 5 min. The

reaction mixture was concentrated in vacuo and puriﬁed via ﬂash

column chromatography (ethyl acetate/hexane, 10:90 to 100:0, then

2-Chloromethyl-3,5,7,8-tetrahydropyrano[4,3-d]pyrimidin-4-one

(45). A mixture of 4-oxo-tetrahydropyran-3-carboxylic acid methyl

ester (1780 g, 11 mol) and NEt₃(830 g, 8.2 mol) in MeOH (3560

L

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

methanol/dichloromethane, 1:99 to 10:90) to provide the title

compound (30 mg, 49% yield). m/z 278.1 (M + H), retention time

= 0.89 min. ¹H NMR (400 MHz, MeOD): δ 7.58 (d, J = 5.05 Hz, 1H),

7.33 (d, J = 3.03 Hz, 1H), 7.07−7.16 (m, 1H), 4.63 (s, 2H), 4.47 (s,

2H), 4.24 (s, 2H), 3.94 (t, J = 5.56 Hz, 2H), 2.73 (t, J = 5.56 Hz, 2H).

General Procedure D. N-((4-Oxo-4,5,7,8-tetrahydro-3H-pyrano-

[4,3-d]pyrimidin-2-yl)methyl)-3-phenyl-N-(thiophen-2-ylmethyl)-

propanamide (20). To a solution of compound 23 (95 mg, 0.34

mmol) in methylene chloride (3 mL) were added triethylamine (143

μL, 1.028 mmol) and 3-phenylpropanoyl chloride (58 mg, 0.34

mmol), and the solution was stirred 4 h. Puriﬁcation was accomplished

on an Isco 12 g column with 1% to 10% methanol in methylene

chloride over 12 column volumes. The resulting partial puriﬁed

product was taken up in DMSO and puriﬁed on a Sunﬁre C18 5 μm

(19 × 100 mm) column. The mobile phase was 10% to 90%

acetonitrile in water (0.1% TFA) to give title compound (39 mg, 0.095

mmol, 28% yield). Mass calculated for C₂₂H₂₃N₃O₃S 409.5, found

410.6 (ESI, M + H); ¹H NMR (400 MHz, MeOD) δ ppm 7.09−7.39

(m, 5H), 6.79−7.00 (m, 2H), 4.77 (s, 1H), 4.35−4.49 (m, 4H), 3.79−

3.98 (m, 2H), 3.35 (s, 1H), 2.81−3.01 (m, 3H), 2.73 (t, J = 7.33 Hz,

1H), 2.49−2.68 (m, 2H).

MeCN in water (0.1% TFA) over 15 min. The fractions were

combined and concentrated in vacuo. The product was dissolved

MeOH, added to a SCX column (1 g), and washed with MeOH (100

mL), and then the product was eluted with 1 N NH₃in MeOH (50

mL) and concentrated to provide the title compound (15 mg, 0.034

mmol, 23% yield). HRMS calculated for C₂₁H₂₂N₄O₃S 411.1491 (M +

H), found 411.1507 (ESI, [M + H]⁺). ¹H NMR (400 MHz, methanol-

d₄) δ 8.73 (d, J = 5.56 Hz, 2H), 7.96−8.12 (m, 2H), 7.20−7.45 (m,

1H), 6.77−7.16 (m, 2H), 4.95−5.14 (m, 2H), 4.56−4.81 (m, 4H),

3.91−4.19 (m, 3H), 3.34 (d, J = 6.06 Hz, 1H), 2.78−3.28 (m, 4H).

N-((4-Oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)-

methyl)-3-(pyridin-3-yl)-N-(thiophen-2-ylmethyl)propanamide (26).

The title compound (10 mg, 0.022 mmol, 15% yield) was prepared

following the same procedure as that for compound 25 from 23 (40

mg, 0.144 mmol, 1.0 equiv) and 3-(pyridin-3-yl)propanoic acid (26

mg, 0.173 mmol, 1.0 equiv). HRMS calculated for C₂₁H₂₂N₄O₃S

411.1491 (M + H), found 411.1492 (ESI, [M + H]⁺). 1H NMR (400

MHz, methanol-d₄) δ 8.54−8.92 (m, 3H), 7.98−8.07 (m, 1H), 7.20−

7.44 (m, 1H), 6.76−7.13 (m, 2H), 4.93−5.11 (m, 1H), 4.49−4.79 (m,

5H), 3.95−4.15 (m, 3H), 2.76−3.28 (m, 6H).

3-(2-Fluorophenyl)-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-

d]pyrimidin-2-yl)methyl)-N-(thiophen-2-ylmethyl)propanamide

(27). To a solution of 47 (45 mg, 0.16 mmol) and 3-(2-

ﬂuorophenyl)propionic acid (27 mg, 0.16 mmol) in dichloromethane

(1 mL) were added 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide

hydrochloride (37 mg, 0.20 mmol), 1-hydroxybenzotriazole (25 mg,

0.16 mmol), and triethylamine (68 μL, 0.49 mmol). The mixture was

stirred at ambient temperature for 24 h. The mixture was added water

and then extracted with dichloromethane two times. The combined

organic layer was concentrated in vacuo and puriﬁed via ﬂash column

chromatography (ethyl acetate:hexane, 10:90 to 100:0) to give a

residue. The residue was further puriﬁed via HPLC to provide the title

compound (12 mg, 17% yield). HRMS calculated for C₂₂H₂₂FN₃O₃S

2-((Ethylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-

4(5H)-one (21a). The title compound (20 mg, 35% yield) was

prepared using general procedure C from acetaldehyde (0.015 mL,

0.28 mmol) and 47 (50 mg, 0.28 mmol). Mass calculated for

1

C₁₀H₁₅N₃O₂209.2, found 210.3 (ESI, M + H). H NMR (400 MHz,

CD₃OD) δ 4.49 (s, 2H) 3.94 (t, J = 5.52 Hz, 2H) 3.79 (s, 2H) 2.85 (q,

J = 7.36 Hz, 2H) 2.65 (t, J = 5.52 Hz, 2H) 1.22 (t, J = 7.28 Hz, 3H).

N-Ethyl-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-

2-yl)methyl)-3-phenylpropanamide (21). The title compound (11

mg, 4% yield) was prepared using general procedure D from 21a (20

mg, 0.096 mmol) and 3-phenylpropanoyl chloride (14 μL, 0.096

mmol). HRMS calculated for C₁₉H₂₃N₃O₃342.1818, found (ESI, [M

1

428.1444, found (ESI, [M + H]⁺), 428.1439. H NMR (400 MHz,

1

+ H]⁺), 342.1811. H NMR (400 MHz, CDCl₃): δ 6.55−6.78 (m,

CDCl₃) δ 10.49 (br s, 1H), 7.15−7.34 (m, 3H), 6.86−7.15 (m, 4H),

4.72 (s, 2H), 4.58 (s, 2H), 4.38−4.48 (m, 2H), 3.95 (t, J = 5.5 Hz,

2H), 3.09 (t, J = 7.8 Hz, 2H), 2.86 (t, J = 7.5 Hz, 2H), 2.57−2.72 (m,

2H).

3-(3-Fluorophenyl)-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-

d]pyrimidin-2-yl)methyl)-N-(thiophen-2-ylmethyl)propanamide

(28). The title compound (8 mg, 11% yield) was prepared according

to same procedure as that for compound 27 from 3-(3-ﬂuorophenyl)-

propanoic acid. HRMS calculated for C₂₂H₂₂FN₃O₃S 428.1444, found

(ESI, [M + H]⁺), 428.1454. ¹H NMR (400 MHz, CDCl₃) δ 10.66 (br

s, 1H), 7.20−7.38 (m, 2H), 6.84−7.07 (m, 5H), 4.72 (s, 2H), 4.57 (s,

2H), 4.44 (s, 2H), 3.95 (t, J = 5.5 Hz, 2H), 3.05 (t, J = 7.5 Hz, 2H),

2.85 (t, J = 7.5 Hz, 2H), 2.54−2.76 (m, 2H).

5H), 4.02 (s, 2H), 3.82 (s, 2H), 3.40 (t, J = 5.3 Hz, 2H), 2.82 (q, J =

6.5 Hz, 2H), 2.37−2.56 (m, 2H), 2.01−2.22 (m, 4H), 0.71 (s, 1H),

0.45−0.64 (m, 3H).

N-((4-Oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)-

methyl)-3-phenylpropanamide (22). To a solution of 47 (24 mg,

0.130 mmol) in dichloromethane (1 mL) were added triethylamine

(0.054 mL, 0.39 mmol, 3.0 equiv) and 3-phenylpropanoyl chloride

(0.02 mL, 0.130 mmol, 1.0 equiv). The mixture was stirred at room

temperature overnight. The solvent was removed, silica gel was added

with methanol, and the mixture was concentrated in vacuo. The

residue was puriﬁed using ﬂash column chromatography (10g Isco

column, 0 to 10% methanol/methylene chloride) to give the title

compound (25 mg, 0.077 mmol, 59% yield). Mass calculated for

3-(4-Fluorophenyl)-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-

d]pyrimidin-2-yl)methyl)-N-(thiophen-2-ylmethyl)propanamide

(29). The title compound (20 mg, 43% yield) was prepared according

to same procedure as that for compound 27 from 2-((thiophen-2-

ylmethylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-

4(5H)-one (30 mg, 0.11 mmol) and 3-(4-ﬂuorophenyl)propanoic acid

(18 mg, 0.11 mmol). HRMS calculated for C₂₂H₂₂FN₃O₃S 428.1444,

1

C₁₇H₁₉N₃O₃313.1, found 314.3 (ESI, M + H). H NMR (400 MHz,

MeOH-d₄) δ 7.12−7.26 (m, 5H), 4.47 (s, 2H), 4.22 (s, 2H), 3.93 (t, J

= 5.56 Hz, 2H), 2.87−2.96 (m, 2H), 2.62−2.67 (m, 2H), 2.54−2.61

(m, 2H).

N-((4-Oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)-

methyl)-3-phenyl-N-(thiophen-3-ylmethyl)propanamide (24). Fol-

lowing the general procedure D, the title compound was prepared.

Mass calculated for C₂₂H₂₃N₃O₃S 409.5, found 410.6 (ESI, M + H).

¹H NMR (MeOD) δ ppm 7.43 (dd, J = 5.05, 3.03 Hz, 1H), 7.11−7.39

1

found (ESI, [M + H]⁺), 428.1454. H NMR (400 MHz, CDCl₃) δ

10.55 (br s 1H), 7.23−7.33 (m, 1H), 7.11−7.22 (m, 2H), 6.92−7.05

(m, 3H), 6.88 (d, J = 3.0 Hz, 1H), 4.70 (s, 2H), 4.58 (s, 2H), 4.42 (s,

2H), 3.95 (t, J = 5.5 Hz, 2H), 3.03 (t, J = 7.3 Hz, 2H), 2.82 (t, J = 7.5

Hz, 2H), 2.65 (t, J = 5.5 Hz, 2H).

(m, 6H), 6.88−7.02 (m, 1H), 4.74 (s, 1H), 4.67 (s, 1H), 4.44 −4.56

(m, 3H), 4.41 (s, 1H), 3.95 (q, J = 5.56 Hz, 2H), 2.95−3.09 (m, 2 H),

2.77−2.95 (m, 2 H), 2.58−2.74 (m, 2 H).

3-(2-Hydroxyphenyl)-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano-

[4,3-d]pyrimidin-2-yl)methyl)-N-(thiophen-2-ylmethyl)-

propanamide (30). The title compound (6 mg, 8% yield) was

prepared according to same procedure as that for compound 27 from

3-(2-hydroxyphenyl)propionic acid. HRMS calculated for

C₂₂H₂₃N₃O₄S 426.1488, found (ESI, [M + H]⁺), 426.1481. ¹H

NMR (400 MHz, CDCl₃) δ 10.91 (br s, 1H), 8.21 (br s, 1H), 7.08−

7.33 (m, 1H), 6.93−7.08 (m, 2H), 6.63−6.93 (m, 4H), 4.54 (s, 2H),

4.48 (s, 2H), 4.37 (s, 2H), 3.86 (t, J = 5.5 Hz, 2H), 2.94−3.07 (m,

2H), 2.78−2.94 (m, 2H), 2.52−2.63 (m, 2H).

N-((4-Oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)-

methyl)-3-(pyridin-4-yl)-N-(thiophen-2-ylmethyl)propanamide (25).

To a solution of 23 (100 mg, 0.369 mmol, 1.0 equiv) and 3-(pyridin-4-

yl)propanoic acid (26 mg, 0.173 mmol, 1.0 equiv) in methylene

chloride (1 mL) were added triethylamine (51 μL, 0.346 mmol, 2.0

equiv) and 2-chloro-1,3-dimethylimidazolidinium chloride (35 mg, 0.2

mmol, 1.2 equiv), and the solution was stirred overnight at room

temperature. The reaction mixture was diluted with dichloromethane,

washed with water (2×), dried over Na₂SO₄, and concentrated in

vacuo. The residue was dissolved in 1 mL of DMSO and puriﬁed on a

Xterra C8 5 μm (19 × 100 mm) column, eluting with 10% to 70%

3-(3-Hydroxyphenyl)-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano-

[4,3-d]pyrimidin-2-yl)methyl)-N-(thiophen-2-ylmethyl)-

M

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

propanamide (31). The title compound (7 mg, 10% yield) was

prepared according to same procedure as that for compound 27 from

3-(3-hydroxyphenyl)propionic acid. HRMS calculated for

C₂₂H₂₃N₃O₄S 426.1488, found (ESI, [M + H]⁺), 426.1494. ¹H

NMR (400 MHz, CDCl₃) δ 10.78 (br s, 1H), 7.23 (d, J = 5.0 Hz, 1H),

7.17 (t, J = 7.8 Hz, 1H), 7.10 (br s, 1H), 6.86−6.98 (m, 1H), 6.74 (t, J

= 6.8 Hz, 4H), 4.56 (d, J = 18.1 Hz, 4H), 4.46 (s, 2H), 3.94 (t, J = 5.5

Hz, 2H), 2.96−3.15 (m, 2H), 2.79−2.91 (m, 2H), 2.63 (t, J = 5.5 Hz,

2H).

3-(4-Hydroxyphenyl)-N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano-

[4,3-d]pyrimidin-2-yl)methyl)-N-(thiophen-2-ylmethyl)-

propanamide (32). The title compound (7 mg, 9% yield) was

prepared according to same procedure as that for compound 27 from

3-(4-hydroxyphenyl)propionic acid. HRMS calculated for

C₂₂H₂₃N₃O₄S 426.1488, found (ESI, [M + H]⁺), 426.1493. ¹H

NMR (400 MHz, CDCl₃): δ 11.17 (br s, 1H), 7.27−7.29 (m, 1H),

7.04 (m, J = 7.5 Hz, 2H), 6.93−7.00 (m, 1H), 6.89 (d, J = 3.0 Hz, 1H),

6.72 (m, J = 8.0 Hz, 2H), 4.71 (br s, 2H), 4.54 (br s, 2H), 4.44 (s, 2H),

3.95 (t, J = 5.3 Hz, 2H), 2.97 (t, J = 7.0 Hz, 2H), 2.83 (t, J = 7.0 Hz,

2H), 2.67 (br s, 2H).

575 Hz, 2H), 7.38 (d, J = 4.4 Hz, 1H), 7.28 (t, J = 8.8 Hz, 2H), 7.00 (s,

1H), 6.95 (s, 1H), 4.85 (s, 2H), 4.47 (s, 2H), 4.38 (s, 2H), 3.85 (t, J =

5.7 Hz, 2H), 3.29 (m, 2H), 2.90 (m, 3H), 2.53 (t, J = 5.4 Hz, 2H),

2.29 (t, J = 10.7 Hz, 2H), 1.76 (m, 2H), 1.63 (m, 2H).

2-((Benzylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (36a). Following general procedure E, the title

compound (30 mg, 0.11 mmol, 11% yield) was prepared from

phenylmethanamine (107 mg, 0.996 mmol). Mass calculated for

C₁₅H₁₇N₃O₂271.3, found 272.4 (ESI, M + H).

N-Benzyl-2-(4-(4-ﬂuorobenzoyl)piperidin-1-yl)-N-((4-oxo-4,5,7,8-

tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)acetamide (36).

Following the procedure for compound 34, the title compound (17.7

mg, 0.034 mmol, 18% yield) was prepared from 36a (46 mg, 0.170

mmol) and 50 (50 mg, 0.19 mmol). Mass calculated for C₂₉H₃₁FN₄O₄

518.6, found 519.4 (ESI, M + H). ¹H NMR (400 MHz,

dichloromethane-d₂) δ 7.98 (br s, 2H) 7.28−7.52 (m, 1H) 7.21 (d,

J = 7.58 Hz, 4H) 4.66 (d, J = 8.08 Hz, 2H) 4.60 (br s, 1H) 4.48 (br s,

2H) 3.85−4.08 (m, 5H) 3.81 (br s, 2H) 3.42 (d, J = 7.07 Hz, 2H) 3.28

(br s, 2H) 2.70 (d, J = 15.66 Hz, 2H) 2.06−2.36 (m, 3H) 1.17−1.41

(m, 2H).

Ethyl 2-(4-(4-Fluorobenzoyl)piperidin-1-yl)acetate (48). To a

solution of (4-ﬂuorophenyl)(piperidin-4-yl)methanone hydrochloride

(51.8 g, 213 mmol) in acetonitrile (400 mL) were added ethyl 2-

bromoacetate (38.2 mL, 345 mmol) and triethylamine (70 mL, 506

mmol). The mixture was heated at reﬂux for 5 h. The solvent was

evaporated under reduced pressure and puriﬁed via ﬂash column

chromatography (ethyl acetate/hexane, 10:90 to 100:0) to provide the

title compound (52.7 g, 85% yield). MS m/z 294.4 (M + 1), retention

time = 1.94 min. ¹H NMR (400 MHz, CDCl₃) δ 7.98 (dd, J = 8.5, 5.5

Hz, 2 H), 7.16 (t, J = 8.5 Hz, 2 H), 4.22 (q, J = 7.2 Hz, 2 H), 3.16−

3.27 (m, 1 H), 3.03 (d, J = 11.5 Hz, 2 H), 2.43 (td, J = 11.2, 3.3 Hz, 2

H), 1.81−2.09 (m, 4 H), 1.51−1.73 (m, 2 H), 1.30 (t, J = 7.3 Hz, 3

H).

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)acetic Acid (50). To a

solution of 48 (52.7 g, 180 mmol) in ethanol (300 mL) was added

sodium hydroxide solution in water (1.8 M, 300 mL). The mixture was

stirred at ambient temperature for 24 h and then neutralized with HCl

(37%) until pH 5−7. The solvent was evaporated under reduced

pressure, dissolved with methanol and methylene chloride, and

ﬁltered. The resulting solution was evaporated to give the title

compound as a white solid (47.7 g, >99% yield). MS m/z 266.3 (M +

H), retention time = 0.87 min. ¹H NMR (400 MHz, CD₃OD) δ 8.11

(dd, J = 8.8, 5.3 Hz, 2H), 7.26 (t, J = 8.8 Hz, 2H), 3.57−3.85 (m, 5H),

3.21 (t, J = 11.0 Hz, 2H), 1.96−2.18 (m, 4H).

2-(((Pyridin-2-ylmethyl)amino)methyl)-7,8-dihydro-3H-pyrano-

[4,3-d]pyrimidin-4(5H)-one (37a). A mixture of 47 (150 mg, 0.83

mmol) and 2-(bromomethyl)pyridine (168 mg, 0.66 mmol) was

heated via microwave in ethanol (1.5 mL) at 120 °C for 5 min. The

reaction mixture was concentrated in vacuo, and the residue was

puriﬁed via ﬂash column chromatography (ethyl acetate/hexane, 20:80

to 100:0 and then methanol/dichloromethane, 1:99 to 30:70) to

provide the title compound (225 mg, 100% yield). MS m/z 273.2 (M

+ 1), retention time = 0.78 min.

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)-N-((4-oxo-4,5,7,8-tetrahy-

dro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)-N-(pyridin-2-ylmethyl)-

acetamide (37). Following the procedure for compound 34, the title

compound (20 mg, 0.038 mmol, 8% yield) was prepared from 37a

(130 mg, 0.477 mmol) and 50 (149 mg, 0.477 mmol). HRMS

calculated for C₂₈H₃₀FN₅O₅520.2360, found (ESI, [M + H]⁺),

520.2367. Retention time = 4.44 min. ¹H NMR (400 MHz, CDCl₃) δ

16.14 (br s, 0.5H), 14.65 (br s, 0.5H), 8.81−8.97 (m, 1H), 7.99−8.08

(m, 2H), 7.82−7.98 (m, 1H), 7.46−7.53 (m, 1H), 7.35−7.43 (m, 1H),

7.19−7.29 (m, 2H), 5.20 (s, 1H), 4.84 (s, 2H), 4.69 (s, 1H), 4.58 (s,

1H), 4.02−4.23 (m, 2H), 3.15−3.28 (m, 2H), 2.86−2.97 (m, 2H),

2.75−2.85 (m, 2H), 2.17−2.40 (m, 2H), 1.85 (d, J = 10.54 Hz, 4H),

1.54−1.73 (m, 2H).

General Procedure E. 2-((Isobutylamino)methyl)-7,8-dihydro-3H-

pyrano[4,3-d]pyrimidin-4(5H)-one (38a). To a solution of 45 (50

mg, 0.25 mmol) in 3 mL of EtOH was added 2-methylpropan-1-amine

(40 mg, 0.54 mmol, 2.2 equiv). The resulting mixture was heated at 65

°C overnight. After 18 h, the reaction mixture was concentrated in

vacuo to give an oil that was dissolved in MeOH and added to a SCX

column. The SCX column was washed with MeOH, and the product

was eluted with 1 N NH₃in MeOH and concentrated in vacuo. The

crude product was absorbed onto silica gel and added to an Biotage

SNAP-10g column, and compound was eluted with 1% to 10% MeOH

in dichloromethane over 12 column volumes to provide the title

compound (13.7 mg, 0.058 mmol, 23% yield). Mass calculated for

C₁₂H₁₉N₃O₂237.30, found 238.4 (ESI, M + H).

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)-N-isobutyl-N-((4-oxo-

4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)-

acetamide (38). To a solution of 50 (50 mg, 0.188 mmol) and

diisopropylethylamine (0.082 mL, 0.47 mmol) in DMF (1 mL) was

added HATU (107 mg, 0.283 mmol, 1.5 equiv), and the mixture was

stirred for 30 min. Then 38a (40 mg, 0.170 mmol) in DMP (1 mL)

was added, and mixture was stirred at ambient temperature overnight.

The crude reaction was puriﬁed directly by prep-HPLC on a Sunﬁre

C18−5 μm (19 × 100 mm) column, eluting with 10% to 90% MeCN

in water (0.1% TFA) over 10 min to provide the title compound (36.4

mg, 0.058 mmol, 31% yield). Exact mass calculated for C₂₆H₃₃FN₄O₄

485.2564, found 485.2588 (ESI, M + H). ¹H NMR (400 MHz,

dichloromethane-d₂) δ ppm 7.98 (dd, J = 8.84, 5.31 Hz, 2H), 7.20 (t, J

= 8.59 Hz, 2H), 4.48−4.67 (m, 4H), 4.21 (br s, 2H), 3.89−4.01 (m,

2H), 3.82 (br s, 3H), 3.23 (br s, 4H), 2.62−2.83 (m, 2H), 2.19 (br s,

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)-N-((4-oxo-4,5,7,8-tetrahy-

dro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)acetamide (34). To a

solution of 47 (1.7 g, 9.6 mmol) and 50 (2.7 g, 10 mmol) in

dichloromethane (50 mL) were added 1-(3-dimethylaminopropyl)-3-

ethylcarbodiimide hydrochloride (2.2 g, 11 mmol), 1-hydroxybenzo-

triazole (1.5 g, 9.6 mmol), and triethylamine (4.0 mL, 29 mmol). The

mixture was stirred at ambient temperature for 24 h. The reaction

mixture was washed with ammonium chloride solution, concentrated

in vacuo, and puriﬁed via ﬂash column chromatography (ethyl acetate/

hexane, 10:90 to 100:0, then methanol/dichloromethane, 1:99 to

10:90) to provide the title compound (2.2 g, 54% yield). HRMS

calculated for C₂₂H₂₅N₄O₄429.1938, found (ESI, [M + H]⁺),

1

429.1925. MS m/z 429.3 (M + 1), retention time = 1.28 min. H

NMR (400 MHz, CDCl₃) δ 11.12 (s, 1H), 8.07 (s, 1H), 7.99 (dd, J =

9.0, 5.5 Hz, 2H), 7.18 (t, J = 8.5 Hz, 2H), 4.58 (s, 2H), 4.41 (d, J = 6.0

Hz, 2H), 3.97 (t, J = 5.5 Hz, 2H), 3.28 (m, 1H), 3.17 (m, 2H), 3.10

(m, 2H), 2.71 (m, 2H), 2.39 (m, 2H), 1.89 (m, 4H).

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)-N-((4-oxo-4,5,7,8-tetrahy-

dro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)-N-(thiophen-2-

ylmethyl)acetamide (35). The title compound (296 mg, 0.56 mmol,

25% yield) was prepared similar to that for compound 34 from 2-

((thiophen-2-ylmethylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (710 mg, 2.3 mmol) and 50 (720 mg, 2.7

mmol). HRMS calculated for C₂₇H₂₉FN₄O₄S 525.1972, found (ESI,

[M + H]⁺), 525.1968. MS m/z 525.2 (M + 1), retention time = 3.08

min. ¹H NMR (400 MHz, DMSO-d₆) δ 11.93 (s, 1H), 8.00 (d, J = 8.8,

N

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

3H), 2.13 (s, 1H), 2.07 (d, J = 6.57 Hz, 1H), 0.98 (d, J = 6.57 Hz, 4H),

0.88 (d, J = 5.56 Hz, 2H).

1H), 3.80−3.88 (m, 3H), 2.63−2.72 (m, 1H), 2.54−2.63 (m, 3H),

0.89−1.05 (m, 1H), 0.42−0.54 (m, 2H), 0.14−0.26 (m, 2H).

2-((Propylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (39a). Following general procedure E, the title

compound (41 mg, 0.173 mmol, 69% yield) was prepared from 45 (50

mg, 0.25 mmol) and propylamine (59 mg, 0.997 mmol). Mass

calculated for C₁₁H₁₇N₃O₂223.2, found 224.3 (ESI, M + H).

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)-N-((4-oxo-4,5,7,8-tetrahy-

dro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)-N-propylacetamide

(39). Following the procedure for compound 38, the title compound

(37.3 mg, 0.064 mmol, 34% yield) was prepared from 39a (38 mg,

0.170 mmol) and 50 (50 mg, 0.188 mmol). Exact mass calculated for

N-(Cyclopropylmethyl)-2-(4-(4-ﬂuorobenzoyl)piperidin-1-yl)-N-

((4-oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)-

acetamide (42). Following the procedure for compound 38, the title

compound (14.4 mg, 0.030 mmol, 16% yield) was prepared from 2-

((cyclopropylmethylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (44 mg, 0.188 mmol) and 50 (50 mg, 0.188

mmol). Exact mass calculated for C₂₆H₃₁FN₄O₄483.2408, found

1

483.2416 (ESI, M + H). H NMR (400 MHz, dichloromethane-d₂) δ

ppm 7.98 (dd, J = 5.56, 8.59 Hz, 2H), 7.19 (t, J = 8.59 Hz, 2H), 4.42−

4.75 (m, 5H), 4.12−4.31 (m, 2H), 3.86−3.98 (m, 2H), 3.36−3.76 (m,

2H), 3.06−3.33 (m, 2H), 2.56−2.76 (m, 2H), 2.22 (br s, 4H), 0.79−

1.04 (m, 1H), 0.61 (d, J = 7.58 Hz, 2H), 0.35−0.54 (m, 1H), 0.12−

0.33 (m, 3H).

1

C₂₅H₃₁FN₄O₄471.2408, found 471.2400 (ESI, M + H). H NMR

(400 MHz, dichloromethane-d₂) δ ppm 7.89 (dd, J = 2.78, 5.81 Hz,

2H), 7.10 (t, J = 8.34 Hz, 2H), 4.37−4.56 (m, 4H), 4.19 (br s, 2H),

3.38−3.91 (m, 6H), 2.97−3.37 (m, 4H), 2.55−2.96 (m, 2H), 1.83−

2.37 (m, 3H), 0.97−1.68 (m, 2H), 0.65−0.93 (m, 3H).

[4-(4-Methoxybenzoyl)piperidin-1-yl]acetic Acid Ethyl Ester (49).

To a solution of (4-methoxyphenyl)(piperidin-4-yl)methanone (20 g,

78 mmol, 1 equiv) in acetonitrile (313 mL) was added triethylamine

(22 mL, 156 mmol, 2 equiv), followed by ethyl 2-bromoacetate (9.5

mL, 86 mmol, 1.1 equiv) dropwise. The resulting mixture was heated

at 85 °C overnight. After 18 h, the reaction mixture was diluted with

saturated NaHCO₃(150 mL) and extracted with ethyl acetate (3 ×

500 mL). The combined organic layers were washed with brine, dried

over anhydrous Na₂SO₄, ﬁltered, and concentrated in vacuo to give an

oil. This oil was puriﬁed by silica gel column chromatography (10−

20% acetone/heptane) to give the title compound as a white solid

(11.7 g, 49% yield). Mass calculated for C₁₇H₂₄N₁O₄306.2. MS (ESI)

m/z 306.4 (M + H)⁺.

[4-(4-Methoxybenzoyl)piperidin-1-yl]acetic Acid (51). To a

solution of 49 (11.7 g, 38 mmol, 1 equiv) in 154 mL of EtOH was

added a solution of NaOH (3.07 g, 77 mmol, 2 equiv) in 154 mL of

H₂O. The resulting mixture was stirred at room temperature

overnight. After 18 h, the reaction mixture was concentrated and

treated with 1 N HCl to pH 5. The resulting mixture was freeze-dried

using a lyophilizer to remove H₂O. The resulting yellow solid was

washed with dichloromethane/methanol, and NaCl was ﬁltered oﬀ.

The ﬁltrate was concentrated to give the title compound as a white

solid (11 g, 38 mmol). Mass calculated for C₁₅H₂₀NO₄278.1. MS

(ESI) m/z 278.8 (M + H)⁺.

N-Ethyl-2-(4-(4-ﬂuorobenzoyl)piperidin-1-yl)-N-((4-oxo-4,5,7,8-

tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)acetamide (40).

To a solution of 45 (100 mg, 0.498 mmol) in EtOH (2 mL) was

added ethylamine (128 mg, 1.99 mmol, 70% in water). The mixture

was then heated at 65 °C for 16 h and concentrated to give 2-

((ethylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]pyrimidin-

4(5H)-one (40a) as a crude yellow solid. This solid (103 mg, 0.493

mmol) was reacted with 50 (131 mg, 0.493 mmol) and O-(7-

azabenzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexaﬂuorophos-

phate (281 mg, 0.740 mmol) in DMF (2.5 mL) using

diisopropylethylamine (159 mg, 1.23 mmol). The resulting mixture

was stirred at ambient temperature for 16 h, and the crude product was

concentrated to dryness. The solid was partitioned between ethyl

acetate and aqueous sodium bicarbonate, and the layers were

separated. The organic layer was washed with brine and aqueous

citric acid, dried (magnesium sulfate), and concentrated to give a crude

oil that was puriﬁed by ﬂash column chromatography (0−10%

methanol in dichloromethane) to give the title compound as a yellow

solid (34 mg, 0.074 mmol, 15% yield). MS m/z 457.4 (M + H), Ret

time: 1.25 min; HRMS m/z (M + H) calculated for C₂₄H₂₉N₄O₄F:

1

457.2248, found: 457.2251. H NMR (400 MHz, CD₃OD) δ 8.00−

8.14 (m, 2H), 7.27 (dt, J = 1.76, 8.66 Hz, 2H), 4.40−4.55 (m, 4H),

4.35 (br s, 1.3 H), 4.27 (br s, 0.7H), 3.87−3.98 (m, 2H), 3.72 (br s,

3H), 3.43−3.55 (m, 3H), 3.15−3.27 (m, 1H), 2.55−2.70 (m, 2H),

1.91−2.21 (m, 4H), 1.24−1.32 (m, 2H), 1.12 (t, J = 7.03 Hz, 1H)

[individual rotamers could not be assigned].

N-(Cyclopropylmethyl)-2-(4-(4-methoxybenzoyl)piperidin-1-yl)-

N-((4-oxo-4,5,7,8-tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)-

methyl)acetamide (43). To a solution of 51 (0.071 g, 0.26 mmol, 1

equiv) and DIEA (0.089 mL, 0.51 mmol, 2 equiv) in 2 mL of DMF

was added HATU (0.107 g, 0.281 mmol, 1.1 equiv). The resulting

mixture was stirred at room temperature for 30 min. 42a (0.06 g, 0.26

mmol, 1 equiv) in 1 mL of DMF was added, and the mixture was

stirred at room temperature overnight. The reaction mixture was

diluted with H₂O (10 mL) and extracted with dichloromethane (3 ×

50 mL). The combined organic layers were concentrated in vacuo and

puriﬁed by preparative HPLC (20−100% CH₃CN/H₂O over 10 min

with 0.1% TFA on a Sunﬁre C18 OBD 50 × 50 mm column with ﬂow

rate of 60 mL/min) to give the title compound as a white solid (0.03 g,

0.06 mmol, 24% yield). HRMS calculated for C₂₇H₃₅N₄O₅495.2607,

found (ESI, [M + H]⁺) 495.2606. MS (ESI) m/z 495.3 (M + H)⁺.

Retention time: 2.89 min (5−95% CH₃CN/H₂O over 7.75 min with

0.1% formic acid, Inertsil ODS3 100 × 3 mm C18 column with ﬂow

2-((Methylamino)methyl)-7,8-dihydro-3H-pyrano[4,3-d]-

pyrimidin-4(5H)-one (41a). Following general procedure E, the title

compound (30.7 mg, 0.157 mmol) was prepared from 45 (50 mg, 0.25

mmol) and 30% methanamine (57 mg, 0.548 mmol) in water. Mass

calculated for C₉H₁₃N₃O₂195.2, found 196.3 (ESI, M + H).

2-(4-(4-Fluorobenzoyl)piperidin-1-yl)-N-methyl-N-((4-oxo-4,5,7,8-

tetrahydro-3H-pyrano[4,3-d]pyrimidin-2-yl)methyl)acetamide (41).

Following the procedure for compound 38, the title compound (22.5

mg, 0.50 mmol, 26% yield) was prepared from 41a (33 mg, 0.169

mmol) and 50 (50 mg, 0.188 mmol). Exact mass calculated for

1

C₂₃H₂₇FN₄O₄443.2095, found 443.2099 (ESI, M + H). H NMR

(400 MHz, dichloromethane-d₂) δ ppm 7.95−8.01 (m, 2H), 7.14−

7.20 (m, 2H), 4.23−4.56 (m, 4H), 3.89−3.95 (m, 2H), 3.05−3.56 (m,

5H), 2.89 (s, 2H), 2.58−2.69 (m, 2H), 2.45−2.57 (m, 2H), 2.14−2.29

(m, 1H), 1.85−2.03 (m, 2H), 1.46 (t, J = 7.58 Hz, 2H).

1

rate of 1.0 mL/min). H NMR (400 MHz, MeOD) δ ppm 12.20−

12.62 (m, 1H), 7.88−7.97 (m, 2H), 7.03 (dd, J = 5.56, 8.59 Hz, 2H),

4.58 (s, 1H), 4.30−4.44 (m, 3H), 3.77−3.86 (m, 5H), 3.44 (d, J = 7.07

Hz, 1H), 3.32−3.39 (m, 1H), 3.13−3.27 (m, 4H), 2.76−3.00 (m, 2H),

2.44 (br s, 1H), 2.04−2.21 (m, 2H), 1.50−1.77 (m, 3H), 1.31−1.46

(m, 1H), 0.86−1.11 (m, 1H), 0.33−0.48 (m, 2H), 0.13−0.27 (m, 2H).

2-((Cyclopropylmethylamino)methyl)-7,8-dihydro-3H-pyrano-

[4,3-d]pyrimidin-4(5H)-one (42a). To a solution of 45 (1 g, 5 mmol,

1 equiv) and DIEA (4.35 mL, 24.9 mmol, 5 equiv) in 7 mL of EtOH

was added cyclopropylmethylamine (1.4 g, 20 mmol, 4 equiv). The

resulting mixture was heated at 65 °C overnight. After 18 h, the

reaction mixture was concentrated in vacuo to give an oil that was

puriﬁed by silica gel column chromatography (100% CH₂Cl₂) to give

the title compound as a white solid (0.23 g, 20%). Mass calculated for

C₁₂H₁₈N₃O₂236.1; MS (ESI) m/z 236.9 (M + H)⁺. Retention time:

0.56 min (5−95% CH₃CN/H₂O over 2 min with 0.1% formic acid,

ASSOCIATED CONTENT

■

S

* Supporting Information

Graphs of calculated versus measured log P as well as

thermodynamic data. This material is available free of charge

via the Internet at http://pubs.acs.org.

1

Inertsil 3 × 3 mm C-8-3 column with ﬂow rate of 2 mL/min). H

NMR (400 MHz, chloroform-d) δ ppm 4.37−4.42 (m, 2H), 3.95 (s,

O

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

inhibition stabilizes axin and antagonizes Wnt signalling. Nature 2009,

461, 614−620.

(6) Guettler, S.; LaRose, J.; Petsalaki, E.; Gish, G.; Scotter, A.;

Pawson, T.; Rottapel, R.; Sicheri, F. Structural basis and sequence rules

for substrate recognition by tankyrase explain the basis for cherubism

disease. Cell 2011, 147, 1340−1354.

(7) Cheung, A. K.; Chin, D. N.; Fan, J.; Shultz, M. D.; Tomlinson, R.

C. 4-Oxo-3,5,7,8-Tetrahydro-4h-Pyrano {4,3-D} Pyrminidinyl Com-

pounds For Use As Tankyrase Inhibitors. WO 2013010092, US 2011-

507313P, 2013.

(8) Gunaydin, H.; Gu, Y.; Huang, X. Novel binding mode of a potent

and selective tankyrase inhibitor. PLoS One 2012, 7, e33740.

(9) Shultz, M. D.; Kirby, C. A.; Stams, T.; Chin, D. N.; Blank, J.;

Charlat, O.; Cheng, H.; Cheung, A.; Cong, F.; Feng, Y.; Fortin, P. D.;

Hood, T.; Tyagi, V.; Xu, M.; Zhang, B.; Shao, W. [1,2,4]Triazol-3-

ylsulfanylmethyl)-3-phenyl-[1,2,4]oxadiazoles: Antagonists of the Wnt

pathway that inhibit tankyrases 1 and 2 via novel adenosine pocket

binding. J. Med. Chem. 2012, 55, 1127−1136.

Accession Codes

PDB codes for the TNKS1 structures in complex with 19, 33,

and 35 are 4LI6, 4LI7, and 4LI8, respectively.

AUTHOR INFORMATION

Corresponding Author

*Phone: 617-871-7551. E-mail: michael.shultz@novartis.com.

■

Present Addresses

^‡Blueprint Medicines, 215 First St., Cambridge, MA 02142.

^§Chongqing Pharmaceutical Research Institute, No. 565,

Tushan Rd., Nanan District, Chongqing, China 400061.

^∥Anti-Infective Group, Medicinal Chemistry, Actelion Pharma-

ceuticals Ltd., Gewerbestrasse 16, CH-4123 Allschwil, Switzer-

land.

^⊥Vertex Pharmaceuticals, Inc., 130 Waverly St., Cambridge, MA

02139.

^#Sanoﬁ-Oncology, 640 Memorial Dr., Cambridge, MA 02139.

^□Epizyme, 400 Technology Square, Cambridge, MA 02139.

(10) Waaler, J.; Machon, O.; Tumova, L.; Dinh, H.; Korinek, V.;

Wilson, S. R.; Paulsen, J. E.; Pedersen, N. M.; Eide, T. J.; Machonova,

O.; Gradl, D.; Voronkov, A.; von Kries, J. P.; Krauss, S. A novel

tankyrase inhibitor decreases canonical Wnt signaling in colon

carcinoma cells and reduces tumor growth in conditional APC mutant

mice. Cancer Res. 2012, 72, 2822−2832.

Notes

The authors declare no competing ﬁnancial interest.

(11) Distler, A.; Deloch, L.; Huang, J.; Dees, C.; Lin, N. Y.; Palumbo-

Zerr, K.; Beyer, C.; Weidemann, A.; Distler, O.; Schett, G.; Distler, J.

H. Inactivation of tankyrases reduces experimental ﬁbrosis by

inhibiting canonical Wnt signallingAnn. Rheum. Dis. 2012, doi:

10.1136/annrheumdis-2012-202275.

(12) Li, Z.; Yamauchi, Y.; Kamakura, M.; Murayama, T.; Goshima, F.;

Kimura, H.; Nishiyama, Y. Herpes simplex virus requires poly(ADP-

ribose) polymerase activity for efficient replication and induces

extracellular signal-related kinase-dependent phosphorylation and

ICP0-dependent nuclear localization of tankyrase 1. J. Virol. 2012,

86, 492−503.

(13) Tenbaum, S. P.; Ordonez-Moran, P.; Puig, I.; Chicote, I.;

Arques, O.; Landolfi, S.; Fernandez, Y.; Herance, J. R.; Gispert, J. D.;

Mendizabal, L.; Aguilar, S.; Cajal, S.; Schwartz, S., Jr.; Vivancos, A.;

Espin, E.; Rojas, S.; Baselga, J.; Tabernero, J.; Munoz, A.; Palmer, H.

G. Beta-catenin confers resistance to PI3K and AKT inhibitors and

subverts FOXO3a to promote metastasis in colon cancer. Nat. Med.

2012, 18, 892−901.

(14) Ulsamer, A.; Wei, Y.; Kim, K. K.; Tan, K.; Wheeler, S.; Xi, Y.;

Thies, R. S.; Chapman, H. A. Axin pathway activity regulates in vivo

pY654-beta-catenin accumulation and pulmonary fibrosis. J. Biol.

Chem. 2012, 287, 5164−5172.

(15) Fancy, S. P. J.; Harrington, E. P.; Yuen, T. J.; Silbereis, J. C.;

Zhao, C.; Baranzini, S. E.; Bruce, C. C.; Otero, J. J.; Huang, E. J.;

Nusse, R.; Franklin, R. J. M.; Rowitch, D. H. Axin2 as regulatory and

therapeutic target in newborn brain injury and remyelination. Nat.

Neurosci. 2011, 14, 1009−1U99.

(16) Bregman, H.; Gunaydin, H.; Gu, Y.; Schneider, S.; Wilson, C.;

DiMauro, E. F.; Huang, X. Discovery of a class of novel tankyrase

inhibitors that bind to both the nicotinamide pocket and the induced

pocket. J. Med. Chem. 2013, 56, 1341−1345.

(17) Lau, T.; Chan, E.; Callow, M.; Waaler, J.; Boggs, J.; Blake, R. A.;

Magnuson, S.; Sambrone, A.; Schutten, M.; Firestein, R.; Machon, O.;

Korinek, V.; Choo, E.; Diaz, D.; Merchant, M.; Polakis, P.; Holsworth,

D. D.; Krauss, S.; Costa, M. A novel tankyrase small-molecule inhibitor

suppresses APC mutation-driven colorectal tumor growth. Cancer Res.

2013, 73, 3132−3144.

(18) Voronkov, A.; Holsworth, D. D.; Waaler, J.; Wilson, S. R.;

Ekblad, B.; Perdreau-Dahl, H.; Dinh, H.; Drewes, G.; Hopf, C.; Morth,

J. P.; Krauss, S. Structural basis and SAR for G007-LK, a lead stage

1,2,4-triazole based specific tankyrase 1/2 inhibitor. J. Med. Chem.

2013, 56, 3012−3023.

ACKNOWLEDGMENTS

■

The authors thank Tim Ramsey for many helpful suggestions

during the preparation of this manuscript. We would like to

acknowledge the work of James Groarke and Guoping Xiao for

protein puriﬁcation of TNKS1. The TNKS1/33 data collection

was performed on the 5.0.3 beamline at the Advanced Light

Source at Berkeley Lab. The Berkeley Center for Structural

Biology is supported in part by the National Institutes of

Health, National Institute of General Medical Sciences, and the

Howard Hughes Medical Institute. The Advanced Light Source

is supported by the Director, Oﬃce of Science, Oﬃce of Basic

Energy Sciences, of the U.S. Department of Energy under

Contract No. DE-AC02-05CH11231. The TNKS1/35 data

collection was performed on the X06SA beamline at the Swiss

Light Source, Paul Scherrer Institut, Villigen, Switzerland.

ABBREVIATIONS USED

■

PARP, poly(ADP-ribose) polymerase; LipE, lipophilic eﬃ-

ciency; MMP, molecular matched pair; SAR, structure−activity

relationship; SER, structure−eﬃciency relationship

REFERENCES

■

(1) Schreiber, V.; Dantzer, F.; Ame, J. C.; de Murcia, G. Poly(ADP-

ribose): nNovel functions for an old molecule. Nat. Rev. Mol. Cell Biol.

2006, 7, 517−528.

(2) Hsiao, S. J.; Smith, S. Tankyrase function at telomeres, spindle

poles, and beyond. Biochimie 2008, 90, 83−92.

(3) Seimiya, H. The telomeric PARP, tankyrases, as targets for cancer

therapy. Br. J. Cancer 2006, 94, 341−345.

(4) Levaot, N.; Voytyuk, O.; Dimitriou, I.; Sircoulomb, F.;

Chandrakumar, A.; Deckert, M.; Krzyzanowski, P. M.; Scotter, A.;

Gu, S.; Janmohamed, S.; Cong, F.; Simoncic, P. D.; Ueki, Y.; La, R. J.;

Rottapel, R. Loss of tankyrase-mediated destruction of 3BP2 is the

underlying pathogenic mechanism of cherubism. Cell 2011, 147,

1324−1339.

(5) Huang, S. M.; Mishina, Y. M.; Liu, S.; Cheung, A.; Stegmeier, F.;

Michaud, G. A.; Charlat, O.; Wiellette, E.; Zhang, Y.; Wiessner, S.;

Hild, M.; Shi, X.; Wilson, C. J.; Mickanin, C.; Myer, V.; Fazal, A.;

Tomlinson, R.; Serluca, F.; Shao, W.; Cheng, H.; Shultz, M.; Rau, C.;

Schirle, M.; Schlegl, J.; Ghidelli, S.; Fawell, S.; Lu, C.; Curtis, D.;

Kirschner, M. W.; Lengauer, C.; Finan, P. M.; Tallarico, J. A.;

Bouwmeester, T.; Porter, J. A.; Bauer, A.; Cong, F. Tankyrase

(19) Leeson, P. D.; Springthorpe, B. The influence of drug-like

concepts on decision-making in medicinal chemistry. Nat. Rev. Drug

Discovery 2007, 6, 881−890.

P

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(20) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand efficiency: a

useful metric for lead selection. Drug Discovery Today 2004, 9, 430−

431.

(21) Arnott, J. A.; Planey, S. L. The influence of lipophilicity in drug

discovery and design. Expert Opin. Drug Discovery 2012, 7, 863−875.

(22) Griffen, E.; Leach, A. G.; Robb, G. R.; Warner, D. J. Matched

molecular pairs as a medicinal chemistry tool. J. Med. Chem. 2011, 54,

7739−7750.

(23) Trunzer, M.; Faller, B.; Zimmerlin, A. Metabolic soft spot

identification and compound optimization in early discovery phases

using MetaSite and LC-MS/MS validation. J. Med. Chem. 2009, 52,

329−335.

(24) Cruciani, G.; Carosati, E.; De, B. B.; Ethirajulu, K.; Mackie, C.;

Howe, T.; Vianello, R. MetaSite: understanding metabolism in human

cytochromes from the perspective of the chemist. J. Med. Chem. 2005,

48, 6970−6979.

(25) T’jollyn, H.; Boussery, K.; Mortishire-Smith, R. J.; Coe, K.; De,

B. B.; Van Bocxlaer, J. F.; Mannens, G. Evaluation of three state-of-the-

art metabolite prediction software packages (Meteor, MetaSite, and

StarDrop) through independent and synergistic use. Drug Metab.

Dispos. 2011, 39, 2066−2075.

(26) Shin, Y. G.; Le, H.; Khojasteh, C.; Hop, C. E. Comparison of

metabolic soft spot predictions of CYP3A4, CYP2C9 and CYP2D6

substrates using MetaSite and StarDrop. Comb. Chem. High

Throughput Screening 2011, 14, 811−823.

(27) Storni, A. Carboxamides. EP 124476, Nov 7, 1984.

(28) Bregman, H.; Chakka, N.; Guzman-Perez, A.; Gunaydin, H.; Gu,

Y.; Huang, X.; Berry, V.; Liu, J.; Teffera, Y.; Huang, L.; Egge, B.;

Mullady, E. L.; Schneider, S.; Andrews, P. S.; Mishra, A.; Newcomb, J.;

Serafino, R.; Strathdee, C. A.; Turci, S. M.; Wilson, C.; DiMauro, E. F.

Discovery of novel, induced-pocket binding oxazolidinones as potent,

selective, and orally bioavailable tankyrase inhibitors. J. Med. Chem.

2013, 56, 4320−4342.

(29) Tsukamoto, A. S.; Grosschedl, R.; Guzman, R. C.; Parslow, T.;

Varmus, H. E. Expression of the int-1 gene in transgenic mice is

associated with mammary gland hyperplasia and adenocarcinomas in

male and female mice. Cell 1988, 55, 619−625.

(30) Freire, E. Do enthalpy and entropy distinguish first in class from

best in class? Drug Discovery Today 2008, 13, 869−874.

(31) Conroy, J. L.; Sanders, T. C.; Seto, C. T. Using the electrostatic

field effect to design a new class of inhibitors for cysteine proteases. J.

Am. Chem. Soc. 1997, 119, 4285−4291.

(32) Karlberg, T.; Markova, N.; Johansson, I.; Hammarstrom, M.;

Schutz, P.; Weigelt, J.; Schuler, H. Structural basis for the interaction

between tankyrase-2 and a potent Wnt-signaling inhibitor. J. Med.

Chem. 2010, 53, 5352−5355.

Q

dx.doi.org/10.1021/jm400807n | J. Med. Chem. XXXX, XXX, XXX−XXX

Article Doi

Identification of NVP-TNKS656: The use of structure-efficiency relationships to generate a highly potent, selective, and orally active tankyrase inhibitor

DOI: 10.1021/jm400807n

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm400807n

Full text of DOI:10.1021/jm400807n

Products guided by the article

R&D Labs maybe for 1419957-25-7

Relevant to this article

Hot Product