Discovery of ARD-69 as a Highly Potent Proteolysis Targeting Chimera (PROTAC) Degrader of Androgen Receptor (AR) for the Treatment of Prostate Cancer

Article

pubs.acs.org/jmc

Cite This: J. Med. Chem. XXXX, XXX, XXX −XXX

Discovery of ARD-69 as a Highly Potent Proteolysis Targeting

Chimera (PROTAC) Degrader of Androgen Receptor (AR) for the

Treatment of Prostate Cancer

Xin Han,^†,‡,∇Chao Wang,^{†,‡,∇,▲}Chong Qin,^†,‡,∇Weiguo Xiang,^†,‡,∇Ester Fernandez-Salas,^†,∥

Chao-Yie Yang,^†,‡Mi Wang,^†,‡Lijie Zhao,^†,‡Tianfeng Xu,^†,‡Krishnapriya Chinnaswamy,^#

James Delproposto,^#Jeanne Stuckey,^#and Shaomeng Wang*^{,†,‡,§,⊥}

‡

§

∥

⊥

#

^†The Rogel Cancer Center, Departments of Internal Medicine, Pharmacology, Pathology, Medicinal Chemistry, and Life

Sciences Institute, University of Michigan, Ann Arbor, Michigan 48109, United States

S

* Supporting Information

ABSTRACT: We report herein the discovery of highly potent

PROTAC degraders of androgen receptor (AR), as exempliﬁed

by compound 34 (ARD-69). ARD-69 induces degradation of

AR protein in AR-positive prostate cancer cell lines in a dose-

and time-dependent manner. ARD-69 achieves DC₅₀values of

0.86, 0.76, and 10.4 nM in LNCaP, VCaP, and 22Rv1 AR+

prostate cancer cell lines, respectively. ARD-69 is capable of

reducing the AR protein level by >95% in these prostate cancer

cell lines and eﬀectively suppressing AR-regulated gene

expression. ARD-69 potently inhibits cell growth in these AR-

positive prostate cancer cell lines and is >100 times more potent

than AR antagonists. A single dose of ARD-69 eﬀectively

reduces the level of AR protein in xenograft tumor tissue in mice. Further optimization of ARD-69 may ultimately lead to a new

therapy for AR+, castration-resistant prostate cancer.

In the last few years, the proteolysis targeting chimera

(PROTAC) strategy has gained momentum with its promise in

the discovery and development of completely new types of

small-molecule therapeutics by inducing targeted protein

degradation.¹⁹⁻²⁵A PROTAC molecule is a heterobifunctional

small molecule containing one ligand, which binds to the target

protein of interest, and a second ligand for an E3 ligase system,

tethered together by a chemical linker.²⁶Because AR protein

plays a key role in CRPC, AR degraders designed based upon the

PROTAC concept could be potentially very eﬀective for the

treatment of CRPC when the disease becomes resistant to AR

antagonists or to androgen synthesis inhibitors.²⁷⁻³⁰Naito et al.

have recently reported AR degraders designed based on the

PROTAC concept, which were named speciﬁc and nongenetic

inhibitors of apoptosis protein (IAP)-dependent protein erasers

(SNIPERs).³¹An AR SNIPER molecule, e.g., compound 1 was

designed using an AR antagonist and a ligand for the cellular

inhibitor of apoptosis protein 1 (cIAP1) as the E3 ligase. While

SNIPER AR degraders are eﬀective in inducing partial

degradation of the AR protein in cells, they induce the

autoubiquitylation and proteasomal degradation of the cIAP1

protein, the E3 ligase needed for induced degradation of AR

protein, thus limiting their AR degradation eﬃciency and

INTRODUCTION

■

Despite improvements in medical treatments over the past three

decades, prostate cancer (PCa) remains a signiﬁcant cause of

cancer-related death, and is second only to lung cancer among

men in developed countries.^1,2In addition to surgery and

radiotherapy, androgen deprivation therapies (ADTs) are

frontline treatments for prostate cancer patients with high-risk

localized disease, and second-generation antiandrogens, such as

abiraterone and enzalutamide, have been shown to beneﬁt

patients with advanced prostate cancer.^3,4Nevertheless, patients

who progress to metastatic castration-resistant prostate cancer

(mCRPC), a hormone-refractory form of the disease, face a high

mortality rate and no cure is currently available.^5,6

The androgen receptor (AR) and its downstream signaling

play a critical role in the development and progression of both

localized and metastatic prostate cancer.⁷Previous strategies

that successfully target AR signaling have focused on blocking

androgen synthesis by drugs such as abiraterone and inhibition

of AR function by AR antagonists such as enzalutamide and

apalutamide (ARN-509).⁸⁻¹⁴However, such agents become

ineﬀective in advanced prostate cancer with AR gene

ampliﬁcation, mutation, and alternate splicing.^15,16It is very

clear, however, that in most patients with CRPC, the AR protein

continues to be expressed and tumors are still dependent on AR

signaling. Consequently, AR is an attractive therapeutic target

for mCRPC.^17,18

Received: October 21, 2018

A

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 1. Chemical structures of two previously reported AR degraders.

Figure 2. Chemical structures of representative AR antagonists.

therapeutic eﬃcacy. Compound 2 (ARCC-4) has been recently

reported as another PROTAC degrader, which was designed

using enzalutamide as the AR antagonist and a von Hippel−

Lindau (VHL) ligand.^27,32ARCC-4 was shown to be more

potent and eﬀective than enzalutamide at inducing apoptosis

and inhibiting proliferation of AR-ampliﬁed prostate cancer cells

(Figure 1).^27,32

In the present study, we report our design, synthesis, and

evaluation of PROTAC AR degraders prepared using several

diﬀerent classes of AR antagonists (Figure 2). Because the

cereblon/cullin 4A and VHL/cullin 2 neddylation degradation

systems have been successfully employed for the design of

PROTAC degraders for diﬀerent proteins, we have evaluated

both these types of E3 ligases for the design of AR PROTAC

degraders. The linker in a PROTAC molecule plays a key role for

its degradation potency and eﬃcacy and accordingly, we have

performed extensive optimization of the linker. This study has

resulted in the discovery of highly potent PROTAC AR

degraders (hereafter called AR degraders), exempliﬁed by

compound 34 (ARD-69). This compound (ARD-69) achieves

DC₅₀< 1 nM and D_max> 95% in LNCaP and VCaP AR+

prostate cancer cell lines. Our present study lays the foundation

for the development of a completely new class of therapeutic

agents for the treatment of CRPC.

However, the aryl group in diﬀerent AR antagonists (shown in

blue in Figure 2) has been extensively modiﬁed and a large

number of substituents on this aryl group have been shown to be

well tolerated. The extensive available structure−activity

relationship (SAR) data suggest that this aryl group is exposed

to solvent, making it a potentially suitable site for tethering in the

design of PROTAC AR degraders. Indeed, in both compounds 1

and 2, this aryl group was employed as the tethering site.

Therefore, we have employed the corresponding aryl group in

diﬀerent AR antagonists for our design and synthesis of

PROTAC AR degraders.

We designed and synthesized a series of potential PROTAC

AR degraders (8−25) using as enzalutamide (4), a Food and

Drug Administration-approved AR antagonist, and a potent,

previously reported VHL ligand,³⁴and diﬀerent linkers to tether

the AR antagonist and VHL ligand together (Table 1). We ﬁrst

evaluated these compounds in the AR-positive (AR+) LNCaP

cells for their ability to induce degradation of AR protein at

diﬀerent concentrations by Western blotting and obtained the

data summarized in Table 1.

Our Western blotting data showed that compounds 8−11,

which contain a linker with two to ﬁve methylene groups,

eﬀectively reduce AR protein level at 0.1, 1, and 10 μM

concentrations. However, the maximum reduction of AR

protein achieved by compounds 8−11 at 0.1−10 μM is only

∼30%. Compound 12 containing a linker with six methylene

groups eﬀectively reduces the AR protein by 54% at 0.1 μM and

by 84% at 1 μM and is therefore a more potent AR degrader.

Interestingly, at 10 μM, compound 12 reduces AR protein by

64%, less than that at 1 μM, probably due to the “hook” eﬀect

that has been observed previously for PROTAC molecules.²¹

Accordingly, we evaluated several highly potent AR degraders at

a concentration range of 1−100 nM to better determine their

potency. Our data showed that compound 13, containing a

RESULTS AND DISCUSSION

■

In the design of PROTAC AR degraders, it is critical to identify a

suitable site for tethering in both an AR antagonist and an E3

ligase ligand. Although co-crystal structures of AR agonists

complexed with AR and co-crystal structures of AR antagonists

bound to mutated AR proteins in an agonist conformation are

available,³³no crystal structure of AR in an antagonist

conformation complexed with an AR antagonist is available.

B

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

than compounds 15 and 16. These data indicate that in these

PROTAC AR degraders, there is an optimal linker length of 11−

12 atoms for achieving the most eﬀective AR degradation.

We next investigated the chemical composition of the linker,

by keeping the total linker length approximately the same as that

in compound 15 or 16. Compound 18 with a poly(ethylene

glycol) linker fails to induce AR degradation at 0.1, 1, and 10

μM, indicating that both the linker length and the linker

chemical composition are important for a PROTAC AR

degrader to eﬀectively induce AR degradation.

Table 1. Optimization of Linker Length, Composition, and

VHL Moiety

a

Compounds 8−18 contain an amide group tethering the

phenyl group in enzalutamide with the linker. We replaced the

amide bond in the linker with an ethynyl group, which we had

used successfully in our design of extremely potent bromodo-

main and extra-terminal (BET) degraders.²³To increase the

solubility of the compounds, we introduced a pyridine group

into the linker, directly connected to the ethynyl group. This led

to the design and synthesis of compound 19, which eﬀectively

reduces AR protein level by 30, 63, and 96% at 10 nM, 100 nM,

and 1 μM, respectively, and is therefore a very potent and

eﬀective AR degrader. To further improve its solubility, we

incorporated a piperazinyl group into the linker in compound

19, yielding compound 20. Compound 20 eﬀectively reduces

AR protein level by 50, 80, and 80% at 100 nM, 1 μM, and 10

μM, respectively.

In the design of VHL ligands, it has been shown that

introduction of an (S)-methyl group (R²) signiﬁcantly improves

the binding aﬃnity to VHL.¹⁹To conﬁrm this, we developed a

ﬂuorescence polarization (FP)-based binding assay for VHL

protein (Supporting Information) and tested the binding

aﬃnities of two VHL ligands, VHL-a and VHL-b. Our binding

data showed that VHL-a and VHL-b bind to VHL protein with

IC₅₀values of 458 and 164 nM (Figure 3). Hence, VHL-b is 3

times more potent than VHL-a. Accordingly, we introduced an

(S)-methyl group (R²) into compound 20, yielding compound

21. Compound 21 eﬀectively reduces AR protein by 37, 61, and

92% at 10 nM, 100 nM, and 1 μM, respectively, and is more

potent than 20.

Encouraged by the high potency of compound 21 in

induction of AR degradation, we further explored the linker

length in 21 by synthesis of compounds 22−25, which have

either a shorter or longer linker. Compound 22 with one more

methylene group and compound 23 with one less methylene

group than the linker in compound 21 achieve very similar

potencies compared to 21. However, compound 24 with two

less methylene groups than compound 21 induces AR

degradation by 3, 31, and 28% at 100 nM, 1 μM, and 10 μM,

respectively, and is thus much less potent and eﬀective than 21.

Compound 25 with three less methylene groups than

compound 21 is completely ineﬀective in inducing AR

degradation at 0.1−10 μM. Taken together, our data

demonstrate that both the length and composition of the linker

in a PROTAC AR degrader play a major role in induction of AR

degradation.

In our design of PROTAC AR degraders (8−25), we tethered

enzalutamide (4) to the terminal amide group in the VHL

ligand. We also investigated other possible tethering positions in

the VHL ligand for the design of AR degraders.

a

All of the data were average of three independent experiments with a

treatment time of 6 h.

Based upon a number of co-crystal structures of VHL ligands

in complex with VHL protein (e.g., PDB: 4W9H and

5LLI),³³⁻³⁶the (S)-methyl group in VHL-b is exposed to the

solvent environment and can be used as a possible tethering

point for the design of AR degraders. To facilitate the synthesis

linker with 7 methylene groups appears to be less potent than 12,

but compounds 14−16, containing a linker with 8−10

methylene groups, have potencies similar to 12. Compound

17, containing a linker with 11 methylene groups, is less potent

C

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 3. Chemical structures of VHL ligands and their binding aﬃnities to VHL protein as determined in our FP-based binding assay (Supporting

Information).

of new AR degraders, we appended an amide group to the (S)-

methyl group and synthesized three VHL ligands (Figure 3)

with diﬀerent “tail” groups based upon previous published SAR

studies on VHL ligands.^34,35These three ligands, VHL-c, VHL-d

and VHL-e all bind to VHL with a high aﬃnity, and VHL-d is the

most potent with an IC₅₀value of 26.8 nM to VHL (Figure 3).

We next designed and synthesized two potential AR degraders

(26 and 27) using enzalutamide, VHL-d and the linkers used in

the two potent AR degraders 15 and 21. Western blotting

analysis showed that both 26 and 27 eﬀectively induce AR

degradation in a dose-dependent manner with >50% of AR

protein being reduced at 0.1 μM and >80% at 1 μM in the

LNCaP cell line.

Compound 29 induces 81% of AR degradation at 100 nM and

97% at 1 μM in the LNCaP cell line and is therefore a potent and

eﬀective AR degrader (Table 2).

Next, we designed and synthesized a number of potential AR

degraders using compound 29 as the template AR degrader and

other four AR antagonists shown in Figure 2. The results are

summarized in Table 3. Compound 30 was synthesized by

replacing the enzalutamide “core” structure with that of ARN-

509 (apalutamide). Compound 30 eﬀectively induces AR

degradation in the LNCaP cell line and has a potency similar

to that of compound 29. Compound 31 was synthesized by

replacing the enzalutamide core structure with that of

bicalutamide. Compound 31 is much less potent than

compounds 29 and 30 in reducing AR protein level in the

LNCaP cell line. Compound 32 was synthesized by replacing

the enzalutamide core structure in 29 with a potent AR

antagonist reported by Guo et al. from Pﬁzer.¹⁰Compound 32

reduces AR protein by 76, 98, and 99% at 10 nM, 100 nM, and 1

μM, respectively, in the LNCaP cell line. In direct comparison,

compound 32 is more potent than compounds 29 and 30.

On the basis of the promising AR degradation data for 27, we

synthesized compound 28 with a shorter and more rigid linker

than in 27. Compared to compound 27, compound 28 is

similarly potent and eﬀective in reducing the AR protein level in

the LNCaP cell line, at 0.01−1 μM.

We next designed and synthesized compound 29 also with a

rigid linker, but its length is similar to that in compound 28.

D

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Takeda Corporation.¹⁴Compound 33 reduces AR protein by

65, 93, and 78% at 100 nM, 1 μM, and 10 μM, respectively, in the

LNCaP cell line. Our data thus showed that compound 32

containing the AR antagonist reported by Guo et al. from

Pﬁzer¹⁰is the most potent AR degrader from this series of AR

degraders designed using the same linker and the same VHL

ligand but diﬀerent AR antagonists.

Table 2. Investigation of the Middle Linking Site from the

VHL ligand

a

Because the VHL ligand plays a key role in our designed

PROTAC AR degraders induction of AR degradation, we

synthesized two new VHL ligands (VHL-e and VHL-f) on the

basis of a recent study.³⁵Our binding data showed that both

VHL-e and VHL-f bind to VHL protein with a high aﬃnity with

IC₅₀values of 190 and 28.6 nM, respectively. To investigate the

stereospeciﬁcity of these compounds, we also synthesized VHL-

g and VHL-h by changing a chiral center in VHL-e. VHL-g and

VHL-h bind to VHL protein with IC₅₀values of 179 and 78 μM,

respectively, and are >100 times less potent than VHL-e,

highlighting the importance of stereochemistry for both chiral

centers in the binding to VHL protein.

Employing VHL-e and VHL-f as the VHL ligands and using

compound 32 as the template AR degrader molecule, we

synthesized compounds 34 and 35 (Table 4). For the purpose of

comparison, we also synthesized compound 36 by employing

VHL-c as the VHL ligand and compound 32 as the template

molecule.

a

All of the data were average of three independent experiments with a

Western blotting showed that compounds 34 and 35 are

highly potent and eﬀective in reducing AR protein in the LNCaP

cell line and both compounds reduce the AR protein level by

>75% at 10 nM and >95% at both 100 nM and 1 μM

concentrations. Compound 36 is also very eﬀective and reduces

the AR protein level by 39, 77, and 99% at 10 nM, 100 nM, and 1

μM, respectively. Hence, compounds 34 and 35 represent two

extremely potent AR degraders.

treatment time of 6 h.

Table 3. Investigation of the Eﬀect of Diﬀerent AR

a

Antagonists

We synthesized two control degrader compounds (37 and

38) by employing VHL-g and VHL-h as the VHL ligands and

compound 34 as the template degrader molecule. Our Western

blotting showed that both 37 and 38 fail to reduce AR protein

level at 0.1−10 μM in the LNCaP cell line (Table 4). This

suggests that strong binding of our designed PROTAC AR

degraders to VHL protein is necessary for eﬀective induction of

AR degradation.

Cereblon ligands have been used successfully in the design of

highly potent degraders of BET protein^20,23and other

proteins.³⁷⁻³⁹We sought to determine if cereblon ligands can

be used for the design of eﬀective AR degraders. We synthesized

compounds 40 and 41 using two cereblon ligands and the same

linkers as those in two potent AR degraders 32 and 39 designed

using a VHL ligand. Although 32 and 39 eﬀectively induce AR

degradation at 10 nM, 100 nM, and 1 μM concentrations in the

LNCaP cell line, compounds 40 and 41 have only a modest

eﬀect in reducing the level of AR protein. Taken together, these

data show that VHL ligands are very suitable for the design of

highly potent AR degraders, but more studies are needed to

determine if the cereblon/cullin 4A E3 ligase system can be used

for the design of highly potent and eﬀective AR degraders (Table

5).

We further evaluated three potent AR degraders, compounds

32, 34, and 35, for their dose-dependent AR degradation in the

LNCaP cell line, with compound 6 included as the AR

antagonist control. Our Western blotting data showed that

each of these three AR degraders induces AR degradation in a

dose-dependent manner, whereas the AR antagonist (6) is

completely ineﬀective, even at 10 μM. While compounds 32, 34,

a

All of the data were average of three independent experiments with a

treatment time of 6 h.

Compound 33 was synthesized by replacing the enzalutamide

core structure in 29 with an AR antagonist reported by the

E

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Table 4. Investigation of VHL Ligands

Table 5. Investigation of Diﬀerent E3 Ligands

a

All of the data were average of three independent experiments with a

treatment time of 6 h.

50% protein degradation) values of 0.86 and 0.76 nM in the

LNCaP and VCaP cell lines, respectively, and >95% AR

degradation at 10 nM in both cell lines. ARD-69 achieves a DC₅₀

of 10.4 nM and near-complete degradation at 1 μM in 22Rv1

cells.

We next investigated the ability of ARD-69 to suppress AR-

regulated gene expression in LNCaP and VCaP cell lines, with a

potent AR antagonist (6) included as the control (Figures 6 and

7). Our data showed that ARD-69 eﬀectively suppresses the

expression of PSA, TMPRSS2, and FKBP5 genes in both LNCaP

and VCaP cell lines in a dose-dependent manner and is capable

of reducing the mRNA level of both PSA and TMPRSS2 genes

by >50% at 10 nM. In addition, ARD-69 is also very eﬀective in

suppressing the ERG gene in VCaP cell lines in a dose-

dependent manner. In direct comparison, the AR degrader

ARD-69 is >100 times more potent than the potent AR

antagonist 6 in suppressing the AR-regulated gene transcription

in both LNCaP and VCaP cell lines.

Because AR signaling drives cell growth for AR-positive

prostate cancer cells, we tested the ability of ARD-69 to inhibit

cell growth, with enzalutamide (4) and compound 6 included as

the AR antagonist controls. Our data showed that ARD-69 is

highly potent in inhibition of cell growth in both LNCaP and

VCaP cell lines and achieves IC₅₀values of 0.25 and 0.34 nM in

the LNCaP and VCaP cell lines, respectively (Figure 8). In the

22Rv1 cell line, compound 34 (ARD-69) achieves an IC₅₀value

of 183 nM and AR antagonists 4 and 6 have IC₅₀values of >10

μM. Therefore, ARD-69 is >100 times more potent than

enzalutamide (4) and compound 6 in LNCaP, VCaP, and

22Rv1 AR+ prostate cancer cell lines.

We investigated the mechanism of AR degradation induced

by ARD-69 in LNCaP and VCaP cells. Our data showed that AR

degradation induced by ARD-69 can be eﬀectively blocked by

pretreatment with an AR antagonist (6), a VHL ligand (VHL-

d), a NEDD8 activating E1 enzyme inhibitor (MLN4924), or a

proteasome inhibitor (MG132) in both LNCaP (Figure 9) and

VCaP (Figure 10) cell lines. These mechanistic data clearly

demonstrate that ARD-69 is a bona ﬁde PROTAC AR degrader.

We examined the pharmacodynamics (PD) of ARD-69 in the

VCaP xenograft tumor tissue in mice (Figure 11). Our PD data

showed that a single administration of ARD-69 at 50 mg/kg via

a

All of the data were average of three independent experiments with a

treatment time of 6 h.

and 35 induce partial AR degradation at 10 nM, they all induce

essentially complete AR degradation at both 100 nM and 1 μM.

Based on the data obtained at 10 nM, compound 34 appears to

be the most potent AR degrader among 32, 34, and 35, and

consequently, we pursued an extensive investigation of

compound 34 (ARD-69).

We evaluated the kinetics of ARD-69 in induction of AR

degradation at 100 nM in the LNCaP and VCaP AR+ cell lines.

Our data showed that ARD-69 eﬀectively reduces the AR

protein level within 2 h and achieves near-complete AR

depletion with a 4 h treatment (Figure 4). Our kinetic data

thus showed that induced AR degradation by ARD-69 in

LNCaP and VCaP cells is rapid.

We further evaluated ARD-69 in LNCaP, VCaP, and 22Rv1

cell lines for its potency in inducing AR degradation with a 24 h

treatment time and obtained the data summarized in Figure 5.

ARD-69 achieves DC₅₀(the drug concentration that results in

F

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 4. Western blotting analysis of AR protein in LNCaP cells treated with AR degrader 34 (ARD-69), with glyceraldehyde 3-phosphate

dehydrogenase (GAPDH) used as the loading control. Cells were treated with 100 nM of ARD-69 for indicated time points.

Figure 5. Examination of dose-dependent AR degradation by 34 (ARD-69) in LNCaP, VCaP, and 22Rv1 prostate cancer cell lines. Cells were treated

for 24 h.

intraperitoneal (IP) injection eﬀectively reduces the level of AR

protein, starting at 3 h and with the eﬀect persisting for at least

48 h. Consistent with the profound decrease of the AR protein,

the level of PSA protein is also eﬀectively reduced at the 3 h time

point with the eﬀect persisting for 24−48 h.

As shown in Scheme 2, compounds 8−17 were synthesized by

the amidation of intermediate 52 and the VHL ligand (49a).

Intermediate 52 was produced by amidation of compound 50

and a series of amines. Compound 50 was synthesized by the

hydrolysis of commercial enzalutamide (4). Compound 18 was

synthesized from the amidation reaction of compound 53, which

was prepared from intermediate 50. The synthesis of compound

26 was through several amidation reactions starting from

intermediate 50.

The synthesis of compound 19 is shown in Scheme 3. First,

the key intermediate (57) was synthesized by coupling of

compounds 55 and 56. Then, the Sonogashira coupling reaction

of 57 with 1-bromo-4-ethynylbenzene in the presence of

cuprous iodide (CuI) and PdCl₂(PPh₃)₂at 100 °C in DMF/

triethylamine (TEA) gave compound 58. The intermediate (59)

was obtained from 58 in the presence of 2-hydroxy-2-

methylpropanenitrile. The target compound (19) can be

obtained through the amidation reaction of VHL ligand (49a)

with 60, which in turn was derived from intermediate 59.

CHEMISTRY

■

The synthesis of VHL ligands (49a, 49b) is outlined in Scheme

1. Brieﬂy, (4-bromophenyl)methanamine (42a) was protected

by Boc₂O to generate 43a. An intermediate (44) was obtained

through the Heck reaction of 43a and 4-methylthiazole, and

subsequent deprotection of the Boc group gave compound 45a.

Amide coupling of 46 with 47 followed by hydrolysis of the

methyl ester produced the key intermediate (48). Amidation of

48 with 45a in the presence of 1-[bis(dimethylamino)-

methylene]-1H-1,2,3-triazolo[4,5-b]-pyridinium 3-oxide hexa-

ﬂuorophosphate (HATU) and N,N-diisopropylethylamine

(DIPEA) at room temperature (rt) in dimethylformamide

(DMF) gave the target compound (49a).

G

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 6. Suppression of AR-regulated gene expression in the LNCaP cell line by the AR degrader 34 (ARD-69) and the AR antagonist 6. LNCaP cells

were treated for 24 h and quantitative real-time polymerase chain reaction (qRT-PCR) was performed to determine the mRNA levels for AR and AR-

regulated genes.

Figure 7. Suppression of AR-regulated gene expression in the VCaP cell line by AR degrader 34 (ARD-69) and AR antagonist 6. VCaP cells were

treated for 24 h and qRT-PCR was performed to determine the mRNA levels for AR and AR-regulated genes. r, regular serum; css, charcoal stripped

serum.

As shown in Scheme 4, compounds 20−25, 27, and 28 were

synthesized according to the following procedure. A Heck

reaction of compounds 61 and 62 gave the key intermediate

(63). The other key intermediate (68) was made through four

steps from the starting material (64) according to a published

method.⁴⁰Then, a Sonogashira coupling reaction of 69 and 70

in the presence of dichlorobispalladium (PdCl₂(PPh₃)₂),

cuprous iodide (CuI) and triethylamine (TEA) in dry DMF

gave the key linker portion (71). A mixture of 71 and acetone

cyanohydrin was heated to 80 °C and stirred for 4 h. The

medium was concentrated and dried under vacuum to yield 72,

which was used in the next step without further puriﬁcation. A

mixture of 4-isothiocyanato-2-(triﬂuoromethyl)benzonitrile

and 72 in DMF was stirred overnight. Then, MeOH and 2 N

HCl were added to this mixture to give the cyclized intermediate

(73). A substitution reaction between 73 and tert-butyl 2-

bromoacetate in the presence of K₂CO₃and KI gave the key

intermediate (74). Amidation of 74 and the VHL ligand (49)

gave the target compounds (20−25). Compounds 27 and 28

were obtained through two amidation reactions.

H

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 8. Cell growth inhibition in LNCaP, VCaP, and 22Rv1 cells treated with AR degrader 34 (ARD-69) and two AR antagonists enzalutamide (4)

and 6. LNCaP and VCaP cells were treated with diﬀerent compounds in charcoal stripped medium in the presence of 0.1 nM of AR agonist R1881 for 7

days, and 22Rv1 cells were treated with a regular culture medium for 7 days. Cell viability was determined by a 2-(2-methoxy-4-nitrophenyl)-3-(4-

nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium (WST-8) assay.

Figure 9. Mechanistic investigation of AR degradation induced by 34 (ARD-69) in LNCaP cells. Cells were pretreated with AR antagonist 6, VHL-d,

MLN4924, and MG132, followed by 6 h treatment with ARD-69 at 100 nM.

Figure 10. Mechanistic investigation of AR degradation induced by 34 (ARD-69) in VCaP cells. Cells were pretreated with AR antagonist 6, VHL-d,

MLN4924, and MG132, followed by 6 h treatment with 34 (ARD-69) at 100 nM.

and TEA in dry DMF gave the key linker portion (77). A

mixture of 77 and acetone cyanohydrin was heated to 80 °C and

stirred for 4 h. The medium was concentrated and dried under

vacuum to give 78, which was used in the next step without

further puriﬁcation. A mixture of 4-isothiocyanato-2-

(triﬂuoromethyl)benzonitrile and 78 in DMF was stirred

overnight at rt. Then, MeOH and 2 N HCl were added to this

mixture to give the cyclized intermediate (79). A substitution

reaction of 79 with tert-butyl 4-bromobutanoate in the presence

of K₂CO₃and KI gave the key intermediate (80). Amidation of

80 and 63 gave the compound 81. In the last step, the

compounds 29 and 30 were obtained through amidation

reaction of the intermediate 81 with 68.

Figure 11. Pharmacodynamics (PD) study of AR degrader 34 (ARD-

Syntheses of compounds 32 and 34−36 are shown in Scheme

6. Compound 83 was synthesized from the Sonogashira

coupling reaction of the starting material (82) and 76.

Compound 84 was made through the substitution reaction of

intermediate 83 and tert-butyl 4-bromopiperidine-1-carboxy-

late. As shown in Scheme 6, the key intermediate (88) was

synthesized in two steps. Compound 89 can be obtained from

the amidation reaction of compound 88 with diﬀerent amino

69) in VCaP tumor tissue in mice. Severe combined immunodeﬁcient

(SCID) mice bearing xenograft VCaP tumors were treated with a single

dose of 34 (ARD-69) (IP, 50 mg/kg). Tumor tissues were harvested at

the indicated time points for immunoblotting.

The synthesis of compounds 29 and 30 is shown in Scheme 5.

A Sonogashira coupling reaction of 75 and 76 in the presence of

dichlorobispalladium (PdCl₂(PPh₃)₂), cuprous iodide (CuI),

I

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 1. VHL Ligands 49

a

Reaction conditions: (a) (Boc)₂O, NaHCO₃, EtOAc/H₂O; (b) 4-methylthiazole, Pd(OAc)₂, KOAc, 90 °C; (c) triﬂuoroacetyl (TFA),

dichloromethane (DCM), rt; (d) HATU, DIPEA, DMF, rt; (e) LiOH, tetrahydrofuran (THF), H₂O; (f) HATU, DIPEA, DMF, rt; (g) TFA,

DCM, rt.

a

Scheme 2. Compounds 8−18 and 26

a

Reaction conditions: (a) concd HCl, dioxane, reﬂux; (b) HATU, DIPEA, DMF, rt; (c) TFA, DCM, rt; (d) HATU, DIPEA, DMF, rt. (e) HATU,

DIPEA, DMF, rt; (f) TFA, DCM, rt; (g) HATU, DIPEA, DMF, rt; (h) HATU, DIPEA, DMF, rt; (i) HATU, DIPEA, DMF, rt; (j) TFA, DCM, rt;

(k) HATU, DIPEA, DMF, rt.

acids. Finally, the target compounds were obtained through the

and 91 gave the key intermediate (92). The key AR antagonist

portion (96) can be obtained from the reaction of compounds

93 and 94 followed by a further oxidation reaction. A

Sonogashira coupling reaction of compounds 96 and 92 gave

amidation of intermediate 89 and various VHL fragments.

Compound 31 was synthesized according to the method

shown in Scheme 7. A substitution reaction of compounds 90

J

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 3. Compound 19

a

Reaction conditions: (a) n-BuLi, THF, rt; (b) CuI, PdCl₂(PPh₃)₂, DMF/TEA, 100 °C; (c) 2-hydroxy-2-methylpropanenitrile, reﬂux; (d) 4-

isothiocyanato-2-(triﬂuoromethyl)benzonitrile, DMF, 80 °C; (e) MeOH, 2 N HCl, reﬂux; (f) TFA, DCM, rt; (g) HATU, DIPEA, DMF, rt.

compound 97, and the target compound 31 was obtained by two

amidation reactions.

capable of achieving complete AR degradation in these cell lines.

ARD-69 eﬀectively suppresses AR-regulated gene expression in

a dose-dependent manner and is eﬀective at concentrations as

low as 10 nM in the LNCaP and VCaP cell lines with 24 h

treatment time. It potently inhibits cell growth in the LNCaP,

VCaP, and 22Rv1 prostate cancer cell lines and is >100 times

more potent than the two AR antagonists that were tested. In

particular, ARD-69 achieves subnanomolar IC₅₀values in the

LNCaP and VCaP cell lines. A single dose of ARD-69 also

eﬀectively reduces AR and PSA proteins in VCaP xenograft

tumor tissues in mice for more than 48 h. Taken together, our

data demonstrate that ARD-69 is an extremely potent AR

degrader and that further optimization of ARD-69 may yield a

new class of drugs for the treatment of mCRPC.

The synthesis of compound 33 is shown in Scheme 8.

Compound 101 was synthesized by the Michael addition

reaction of compound 99 with 100, and compound 104 was

made by the reaction of compound 101 with NH₂NH₂. The

reaction of compound 102 with compound 85 gave the AR

antagonist (103). A Sonogashira coupling reaction of 103 with

92 in the presence of CuI and PdCl₂(PPh₃)₂at rt in DMF/TEA

gave compound 104. Finally, the target compound 33 was

obtained from an amidation reaction.

As shown in Scheme 9, compounds 39−41 were made

according the following procedure. In detail, compound 107 was

synthesized by the amidation reaction of compounds 87 and

106. Compound 108 was made by hydrolysis reaction of

intermediate 107, and intermediate 109 was synthesized by the

amidation of compound 108 with nonane-1,9-diamine. Finally,

the target compound (39) was synthesized by two amidation

reactions. Compound 41 was made through the substitution

reaction of 88 with 111, and compound 40 was synthesized by

the substitution reaction of 111 and 112, which was produced by

amidation reaction of decane-1,10-diamine with compound

108.

EXPERIMENTAL SECTION

■

Chemistry. General Experiment and Information. Unless

otherwise noted, all purchased reagents were used as received without

further puriﬁcation. ¹H NMR and ¹³C NMR spectra were recorded on a

Bruker Avance 400 MHz spectrometer. ¹H NMR spectra were reported

in parts per million (ppm) downﬁeld from tetramethylsilane. All ¹³C

NMR spectra were reported in ppm and obtained with ¹H decoupling.

In reported spectral data, the format (δ) chemical shift (multiplicity, J

values in Hz, integration) was used with the following abbreviations: s =

singlet, d = doublet, t = triplet, q = quartet, m = multiplet. Mass spectral

(MS) analysis was carried out with a Waters ultraperformance liquid

chromatography (UPLC) mass spectrometer. The ﬁnal compounds

were all puriﬁed by C18 reverse-phase preparative high-performance

liquid chromatography (HPLC) column with solvent A (0.1% TFA in

H₂O) and solvent B (0.1% TFA in CH₃CN) as eluents. The purity of all

of the ﬁnal compounds was conﬁrmed to be >95% by UPLC−MS or

UPLC.

In Scheme 10, compounds 37 and 38 were synthesized

according to the previously published methods as shown in

Scheme 4.

CONCLUSIONS

■

In this study, we have designed, synthesized, and evaluated a

series of PROTAC AR degraders using ﬁve diﬀerent AR

antagonists and ligands for VHL/cullin 2 and cereblon/cullin 4A

E3 ligases. We also performed extensive optimization of the

linker tethering the AR antagonist portion and the E3 ligand

portion in our designed AR degraders. Our eﬀorts have yielded

highly potent and eﬀective AR degraders, as exempliﬁed by

compound 34 (ARD-69). ARD-69 is eﬀective in inducing AR

degradation at concentrations lower than 1 nM in LNCaP and

VCaP prostate cancer cell lines with a 24 h treatment time and is

General Procedure for Synthesis of Compounds 8−17. Concd

HCl (5 mL) was added to a solution of enzalutamide (4) (4.64 g, 10

mmol) in dioxane (50 mL). The reaction mixture was reﬂuxed for 2 h.

The reaction mixture was quenched with H₂O and extracted with

DCM. The organic layer was separated, washed with brine, dried, and

evaporated. The ﬁnal compound 50 was obtained by ﬂash column

chromatography (hexane/EtOAc = 4:1) in 95% yield.

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

compound 50 (451 mg, 1 mmol) and a series of linear amines (1.1

K

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 4. Compounds 20−25, 27, and 28

a

Reaction conditions: (a) Pd(OAc)₂, KOAc, 90 °C; (b) MeOH, H₂SO₄, 70 °C; (c) 2-iodopropane, t-BuOK, THF, rt; MeOH, (d) LiOH, MeOH/

H₂O, rt; (e) benzyl (2S,4R)-4-hydroxypyrrolidine-2-carboxylate hydrochloride, HATU, DIPEA, DMF, rt; (f) Pd−C, H₂, MeOH, rt; (g) CuI,

PdCl₂(PPh₃)₂, DMF/TEA, 100 °C; (h) 2-hydroxy-2-methylpropanenitrile, reﬂux; (i) 4-isothiocyanato-2-(triﬂuoromethyl)benzonitrile, DMF, 80

°C; (j) MeOH, 2 N HCl, reﬂux; (k) TFA, DCM, rt; (l) K₂CO₃, KI, CH₃CN, reﬂux; (m) TFA, DCM, rt; (o) HATU, DIPEA, DMF, rt; (p) TFA,

DCM, rt; (q) HATU, DIPEA, DMF, rt.

equiv) in DMF (2 mL). After 30 min at rt, the mixture was subject to

HPLC puriﬁcation to aﬀord compound 51 in 80−90% yields.

A solution of compound 51 in 1:1 TFA/DCM was stirred at rt for 30

min. The solvents were evaporated under reduced pressure to give the

corresponding deprotected intermediate 52 (TFA salt) that was used in

the following reactions without further puriﬁcation (95% yield).

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

compound 52 (52.2 mg, 0.1 mmol) and compound 49a (1.1 equiv) in

DMF (2 mL). After 30 min at rt, the mixture was subject to HPLC

puriﬁcation to aﬀord compound 8 in 90% yield. Following the

procedures used to prepare compound 8, compounds 9−17 with

diﬀerent chain lengths were obtained by the same methods.

(4R)-1-((S)-2-(4-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

butanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (9). H NMR (400

MHz, MeOD-d₄) δ = 8.93 (s, 1H), 8.16 (d, J = 7.6 Hz, 2H), 7.99 (d, J =

8.0 Hz, 1H), 7.50−7.32 (m, 6H), 4.63 (s, 1H), 4.61−4.45 (m, 3H), 4.35

(d, J = 15.6 Hz, 1H), 3.91 (d, J = 10.0 Hz, 1H), 3.83−3.55 (m, 3H), 3.41

(t, J = 7.2 Hz, 2H), 2.48 (s, 3H), 2.33−2.15 (m, 3H), 2.12−2.00 (m,

3H), 1.60 (s, 6H), 1.04 (s, 9H). UPLC−MS calcd for C₄₆H₄₉F₄N₈O₆S₂

[M + H]⁺: 949.32, found: 949.58 (H⁺). UPLC-retention time: 4.92

min.

(4R)-1-((S)-2-(5-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

pentanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

(4R)-1-((S)-2-(3-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

propanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (10). H NMR (400

1

MHz, MeOD-d₄) δ = 8.97 (s, 1H), 8.16 (d, J = 7.4 Hz, 2H), 7.99 (d, J =

8.4 Hz, 1H), 7.84 (dd, J = 8.0 Hz, J = 7.6 Hz, 1H), 7.53−7.39 (m, 4H),

7.38−7.37 (m, 2H), 4.63 (s, 1H), 4.60−4.46 (m, 3H), 4.36 (d, J = 15.2

Hz, 1H), 3.90 (d, J = 10.8 Hz, 1H), 3.83−3.75 (m, 1H), 3.40 (t, J = 7.0

Hz, 2H), 2.51 (s, 3H), 2.38−2.15 (m, 3H), 2.13−2.02 (m, 1H), 1.69−

1.55 (m, 10H), 1.04 (s, 9H). UPLC−MS calcd for C₄₇H₅₁F₄N₈O₆S₂[M

+ H]⁺: 963.33, found: 963.54. UPLC-retention time: 5.15 min.

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (8). H NMR (400

MHz, MeOD-d₄) δ = 8.94 (s, 1H), 8.16 (d, J = 7.2 Hz, 2H), 7.99 (d, J =

8.4 Hz, 1H), 7.52−7.31 (m, 6H), 4.63 (s, 1H), 4.59−4.45 (m, 3H), 4.36

(d, J = 15.2 Hz, 1H), 3.90 (d, J = 10.0 Hz, 1H), 3.83−3.75 (m, 1H), 3.40

(t, J = 7.0 Hz, 2H), 2.49 (s, 3H), 2.44−2.00 (m, 4H), 1.60 (s, 6H), 1.04

(s, 9H). UPLC−MS calcd for C₄₅H₄₇F₄N₈O₆S₂[M + H]⁺: 935.30,

found: 935.52. UPLC-retention time: 4.77 min.

L

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 5. Compounds 29 and 30

a

Reaction conditions: (a) CuI, PdCl₂(PPh₃)₂, DMF/TEA, 100 °C; (b) reﬂux; (c) DMF, 80 °C; (d) MeOH, 2 N HCl, reﬂux; (e) TFA, DCM, rt;

(f) K₂CO₃, KI, CH₃CN, reﬂux; (g) TFA, DCM, rt; (h) HATU, DIPEA, DMF, rt; (e) TFA, DCM, rt; (j) HATU, DIPEA, DMF, rt.

a

Scheme 6. Compounds 32 and 34−36

a

Reaction conditions: (a) CuI, PdCl₂(PPh₃)₂, DMF/TEA, 100 °C; (b) TFA, DCM, rt; (c) K₂CO₃, KI, CH₃CN, reﬂux; (d) NaOH, MeOH/H₂O,

rt; (e) NaH, DMF, rt; (f) TFA, DCM, rt; (g) HATU, DIPEA, DMF, rt; (h) TFA, DCM, rt; (i) HATU, DIPEA, DMF, rt; (j) TFA, DCM, rt; (k)

HATU, DIPEA, DMF, rt.

(4R)-1-((S)-2-(6-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

hexanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

(m, 1H), 1.69−1.55 (m, 10H), 1.49−1.26 (m, 2H), 1.03 (s, 9H).

UPLC−MS calcd for C₄₈H₅₃F₄N₈O₆S₂[M + H]⁺: 991.35, found:

977.52. UPLC-retention time: 5.38 min.

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (11). H NMR (400

(4R)-1-((S)-2-(7-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

heptanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

MHz, MeOD-d₄) δ = 9.20 (s, 1H), 8.16 (d, J = 7.2 Hz, 2H), 7.99 (d, J =

8.4 Hz, 1H), 7.83 (dd, J = 8.0 Hz, J = 8.0 Hz, 1H), 7.52−7.40 (m, 4H),

7.35 (dd, J = 7.6 Hz, J = 8.0 Hz, 2H), 4.63 (s, 1H), 4.59−4.45 (m, 3H),

4.36 (d, J = 15.6 Hz, 1H), 3.90 (d, J = 10.4 Hz, 1H), 3.83−3.75 (m, 1H),

3.41 (t, J = 6.4 Hz, 2H), 2.51 (s, 3H), 2.38−2.15 (m, 3H), 2.13−2.02

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (12). H NMR (400

MHz, MeOD-d₄) δ = 9.38 (s, 1H), 8.16 (d, J = 7.2 Hz, 2H), 7.99 (d, J =

8.4 Hz, 1H), 7.82 (dd, J = 8.0 Hz, J = 7.6 Hz, 1H), 7.54−7.40 (m, 4H),

M

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 7. Compound 31

a

Reaction conditions: (a) CuI, PdCl₂(PPh₃)₂, DMF/TEA, 100 °C; (b) NaH, THF, rt; (c) H₂O₂, (CF₃CO)₂O; (d) CuI, PdCl₂(PPh₃)₂, DMF/

TEA, 100 °C; (e) TFA, DCM, rt; (f) HATU, DIPEA, DMF, rt; (g) TFA, DCM, rt; (h) HATU, DIPEA, DMF, rt.

7.40−7.30 (m, 2H), 4.63 (s, 1H), 4.61−4.47 (m, 3H), 4.37 (d, J = 15.6

Hz, 1H), 3.91 (d, J = 11.2 Hz, 1H), 3.83−3.77 (m, 1H), 3.40 (t, J = 7.2

Hz, 2H), 2.53 (s, 3H), 2.35−2.24 (m, 3H), 2.12−2.05 (m, 1H), 1.69−

1.55 (m, 10H), 1.46−1.34 (m, 4H), 1.03 (s, 9H). UPLC−MS calcd for

C₄₉H₅₅F₄N₈O₆S₂[M + H]⁺: 991.36, found: 991.53. UPLC-retention

time: 5.55 min.

undecanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (16). H NMR (400

MHz, MeOD-d₄) δ = 8.98 (s, 1H), 8.15 (d, J = 8.4 Hz, 2H), 8.03−7.94

(m, 1H), 7.87−7.75 (m, 1H), 7.49−7.35 (m, 4H), 7.36−7.31 (m, 2H),

4.63 (s, 1H), 4.60−4.46 (m, 3H), 4.35 (d, J = 15.6 Hz, 1H), 3.90 (d, J =

11.2 Hz, 1H), 3.83−3.55 (m, 3H), 3.39 (t, J = 7.0 Hz, 2H), 2.48 (s, 3H),

2.32−2.16 (m, 3H), 2.13−2.04 (m, 1H), 1.68−1.54 (m, 10H), 1.46−

1.30 (m, 12H), 1.03 (s, 9H). UPLC−MS calcd for C₅₃H₆₃F₄N₈O₆S₂[M

+ H]⁺: 1047.42, found: 1047.43. UPLC-retention time: 6.38 min.

(4R)-1-((S)-2-(12-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-

dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

dodecanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

(4R)-1-((S)-2-(8-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

octanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (13). H NMR (400

MHz, MeOD-d₄) δ = 9.24 (s, 1H), 8.16 (d, J = 7.6 Hz, 2H), 7.99 (d, J =

8.4 Hz, 1H), 7.82 (dd, J = 8.0 Hz, J = 7.6 Hz, 1H), 7.54−7.40 (m, 4H),

7.39−7.30 (m, 2H), 4.64 (s, 1H), 4.62−4.46 (m, 3H), 4.36 (d, J = 15.2

Hz, 1H), 3.90 (d, J = 10.8 Hz, 1H), 3.83−3.55 (m, 3H), 3.40 (t, J = 7.0

Hz, 2H), 2.51 (s, 3H), 2.35−2.17 (m, 3H), 2.12−2.01 (m, 1H), 1.68−

1.54 (m, 10H), 1.47−1.32 (m, 6H), 1.02 (s, 9H). UPLC−MS calcd for

C₅₀H₅₇F₄N₈O₆S₂[M + H]⁺: 1005.38, found: 1005.57. UPLC-retention

time: 5.74 min.

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (17). H NMR (400

MHz, MeOD-d₄) δ 9.30 (s, 1H), 8.20−8.12 (m, 2H), 8.00 (dd, J = 8.3,

1.7 Hz, 1H), 7.85 (t, J = 8.0 Hz, 1H), 7.54−7.45 (m, 4H), 7.40−7.34

(m, 2H), 4.65 (s, 1H), 4.57 (dt, J = 24.3, 10.3 Hz, 3H), 4.38 (d, J = 15.6

Hz, 1H), 3.93 (d, J = 11.1 Hz, 1H), 3.82 (dd, J = 11.0, 3.8 Hz, 1H), 3.42

(t, J = 7.0 Hz, 2H), 2.54 (d, J = 3.4 Hz, 3H), 2.41−2.17 (m, 4H), 2.09

(ddd, J = 19.6, 10.0, 5.4 Hz, 1H), 1.67−1.62 (m, 3H), 1.61 (s, 6H), 1.36

(d, J = 24.1 Hz, 14H), 1.05 (s, 9H). UPLC−MS calcd for

C₅₄H₆₅F₄N₈O₆S₂[M + H]⁺: 1061.44, found 1061.40. UPLC-retention

time: 6.5 min.

General Procedure for Synthesis of Compound 18. DIPEA (5

equiv) and HATU (1.2 equiv) were added to a solution of compound

50 (45.1 mg, 0.1 mmol) and tert-butyl 3-(2-(2-aminoethoxy)ethoxy)-

propanoate (1.1 equiv) in DMF (2 mL). After 30 min at rt, the mixture

was subject to HPLC puriﬁcation to aﬀord the tert-butyl protected

compound (53) in 92% yield. Then, compound 53 was obtained by a

deprotection reaction in TFA/DCM solvent.

(4R)-1-((S)-2-(9-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-di-

methyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

nonanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (14). H NMR (400

MHz, MeOD-d₄) δ = 8.97 (s, 1H), 8.16 (d, J = 8.0 Hz, 2H), 7.98 (d, J =

8.0 Hz, 1H), 7.82 (dd, J = 8.0 Hz, J = 7.6 Hz, 1H), 7.49−7.38 (m, 4H),

7.40−7.30 (m, 2H), 4.64 (s, 1H), 4.61−4.47 (m, 3H), 4.35 (d, J = 15.6

Hz, 1H), 3.91 (d, J = 10.8 Hz, 1H), 3.83−3.75 (m, 1H), 3.40 (t, J = 7.2

Hz, 2H), 2.48 (s, 3H), 2.35−2.15 (m, 3H), 2.13−2.05 (m, 1H), 1.67−

1.56 (m, 10H), 1.45−1.30 (m, 8H), 1.03 (s, 9H). UPLC−MS calcd for

C₅₁H₅₉F₄N₈O₆S₂[M + H]⁺: 1019.39, found: 1019.42. UPLC-retention

time: 5.93 min.

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

53 (48.8 mg, 0.08 mmol) and 49a (1.1 equiv) in DMF (2 mL). After 30

min at rt, the mixture was subject to HPLC puriﬁcation to aﬀord

compound 18 in 85% yield.

(4R)-1-((S)-2-(10-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-

dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

decanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-(4-methyl-

1

thiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (15). H NMR (400

MHz, MeOD-d₄) δ = 8.90 (s, 1H), 8.16 (d, J = 7.6 Hz, 2H), 8.03−7.94

(m, 1H), 7.85−7.78 (m, 1H), 7.50−7.30 (m, 6H), 4.63 (s, 1H), 4.61−

4.47 (m, 3H), 4.35 (d, J = 15.2 Hz, 1H), 3.90 (d, J = 11.6 Hz, 1H),

3.83−3.77 (m, 1H), 3.40 (t, J = 7.0 Hz, 2H), 2.47 (s, 3H), 2.32−2.18

(m, 3H), 2.14−2.03 (m, 1H), 1.70−1.55 (m, 10H), 1.47−1.25 (m,

10H), 1.03 (s, 9H). UPLC−MS calcd for C₅₂H₆₁F₄N₈O₆S₂[M + H]⁺:

1033.41, found: 1033.41. UPLC-retention time: 6.15 min.

(4R)-1-((S)-13-(tert-Butyl)-1-(4-(3-(4-cyano-3-(triﬂuoromethyl)-

phenyl)-5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuoro-

phenyl)-1,11-dioxo-5,8-dioxa-2,12-diazatetradecan-14-oyl)-4-hy-

droxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxa-

mide (18). ¹H NMR (400 MHz, MeOD-d₄) δ = 9.76 (s, 1H), 8.16 (d, J

= 8.4 Hz, 2H), 7.99 (d, J = 8.4 Hz, 1H), 7.86 (dd, J = 8.4 Hz, J = 8.0 Hz,

1H), 7.57−7.45 (m, 4H), 7.41−7.31 (m, 2H), 4.65 (s, 1H), 4.60−4.46

(m, 3H), 4.36 (d, J = 15.6 Hz, 1H), 3.92−3.56 (m, 12H), 2.57 (s, 3H),

2.58−2.41 (m, 2H), 2.22−2.00 (m, 2H), 1.59 (s, 6H), 1.39−1.32 (m,

(4R)-1-((S)-2-(11-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-

dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamido)-

N

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 8. Compound 33

a

Reaction conditions: (a) Na₂CO₃, acetone, reﬂux; (b) NH₂NH₂, EtOH, reﬂux; (c) NaH, DMF, rt; (d) CuI, PdCl₂(PPh₃)₂, DMF/TEA, 100 °C;

(e) TFA, DCM, rt; (f) HATU, DIPEA, DMF, rt; (g) TFA, DCM, rt; (h) HATU, DIPEA, DMF, rt.

3H), 1.02 (s, 9H). UPLC−MS calcd for C₄₉H₅₅F₄N₈O₈S₂[M + H]⁺:

1023.35, found 1023.50. UPLC-retention time: 5.2 min.

A solution of compound 59 (2.2 g, 5 mmol) and 4-isothiocyanato-2-

(triﬂuoromethyl)-benzonitrile (1.1 equiv) in 20 mL of DMF was stirred

at 80 °C for 8 h. Then, 10 mL of MeOH and 10 mL of 2 N HCl were

added and the mixture was reﬂuxed for another 4 h. After UPLC−MS

demonstrated the full conversion of starting materials, the reaction

mixture was cooled to rt and H₂O was added into the mixture. The

aqueous layer was extracted with EtOAc, and the combined organic

layers were washed with brine, then dried over anhydrous Na₂SO₄. The

solvent was removed on a rotary evaporator, and the residue was

puriﬁed by ﬂash column chromatography to aﬀord compound 60 in

60% yield.

General Procedure for Synthesis of Compound 19. n-BuLi (1

equiv) was added to a solution of compound 56 (2.5 g, 10 mmol) in

THF at −78 °C. Then, compound 55 (1 equiv) in THF was added at

−78 °C slowly. After 2 h at rt, the reaction mixture was quenched with

ice/H₂O and extracted with DCM. The organic layer was separated,

washed with brine, dried, and evaporated. The ﬁnal compound 57 was

obtained by ﬂash column chromatography (hexane/EtOAc = 2:1) in

70% yield.

Compound 57 was placed in a 25 mL round-bottom ﬂask (2.7 g, 7

mmol) with 4-iodoaniline (1 equiv), CuI (0.2 equiv), and

PdCl₂(PPh₃)₂(0.1 equiv) in DMF and TEA under Ar. Then, the

mixture was stirred for 4 h at 100 °C. After this time, H₂O was added

into the resulting complex, which was extracted with EtOAc three

times. The organic layer was again washed with H₂O before being dried

over MgSO₄, and the solvent was removed under vacuum, leaving the

crude product. The pure product 58 was obtained by ﬂash column

chromatography (hexane/EtOAc = 4:1) in 80% yield.

A solution of compound 58 (2.1 g, 5.6 mmol) in 2-hydroxy-2-

methylpropanenitrile was reﬂuxed for 8 h. The intermediate 59 was

obtained by removing the solvent under reduced pressure and used in

the next step without further puriﬁcation.

A solution of compound 60 in 1:1 TFA/DCM was stirred at rt for 30

min. The solvents were evaporated under reduced pressure to give the

corresponding deprotected intermediate 61 (TFA salt) that was used in

the following reactions without further puriﬁcation (95% yield).

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

61 (62 mg, 0.1 mmol) and compound 49a (1.1 equiv) in DMF (2 mL).

After 30 min at rt, the mixture was subject to HPLC puriﬁcation to

aﬀord compound 19 in 90% yield.

(4R)-1-((S)-2-(7-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-

5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)ethynyl)-

pyridin-2-yl)heptanamido)-3,3-dimethylbutanoyl)-4-hydroxy-N-(4-

(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxamide (19). ¹H

NMR (400 MHz, MeOD-d₄) δ 8.97 (s, 1H), 8.37−8.33 (m, 1H),

8.24−8.14 (m, 3H), 8.02 (dd, J = 8.3, 2.0 Hz, 1H), 7.75 (dd, J = 8.8, 2.3

O

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 9. Compounds 39−41

a

Reaction conditions: (a) HATU, DIPEA, DMF, rt; (b) NaOH, H₂O/MeOH, rt; (c) HATU, DIPEA, DMF, rt; (d) TFA, DCM, rt; (e) HATU,

DIPEA, DMF, rt; (f) TEA, dimethyl sulfoxide (DMSO), 100 °C; (g) HATU, DIPEA, DMF, rt; (h) HATU, DIPEA, DMF, rt.

Hz, 1H), 7.67 (dd, J = 9.1, 2.5 Hz, 2H), 7.51−7.42 (m, 7H), 6.95 (d, J =

8.9 Hz, 1H), 4.61−4.53 (m, 3H), 4.39 (d, J = 15.6 Hz, 1H), 3.96 (d, J =

11.0 Hz, 1H), 3.87−3.82 (m, 1H), 3.70−3.57 (m, 2H), 3.21 (d, J = 7.6

Hz, 3H), 2.72 (s, 5H), 2.49 (d, J = 2.2 Hz, 3H), 2.41 (t, J = 6.7 Hz, 2H),

2.29−2.21 (m, 1H), 2.12 (td, J = 9.1, 4.6 Hz, 1H), 1.82−1.74 (m, 3H),

1.60 (s, 6H), 1.08 (s, 9H). UPLC−MS calcd for C₅₅H₅₈F₃N₈O₅S₂[M +

H]⁺: 1031.39, found 1031.48. UPLC-retention time: 5.1 min.

General Procedure for Synthesis of Compounds 20−24. A

solution of 61 (3.43 g, 10 mmol), 4-methylthiazole (2 equiv), KOAc

(2 equiv), and Pd(OAc)₂(1%) in DMF/TEA was stirred at 80 °C for 4

h. After the reaction was complete, the TEA was removed under

reduced pressure, then H₂O was added into the mixture, and the

mixture was extracted three times by EtOAc. The solvent was collected,

dried with Na₂SO₄, and evaporated under reduced pressure to give the

corresponding intermediate 63 by ﬂash column chromatography

(hexane/EtOAc = 4:1) in 80% yield.

To a solution of 64 (1.41 mg, 10 mmol) in MeOH was added H₂SO₄

(1 equiv). Then, the mixture was stirred at 70 °C for 4 h. The mixture

was quenched with H₂O and extracted with EtOAc three times. The

organic layer was washed with H₂O and brine and dried with Na₂SO₄.

The product 65 was obtained by removing the solvent and used without

further puriﬁcation.

t-BuOK (1.5 equiv) was added slowly to a solution of 65 (1.55 g, 10

mmol) in THF, then 2-iodopropane (1.3 equiv) was added dropwise at

0 °C. After the addition was completed, the mixture was stirred at rt

overnight. After UPLC−MS demonstrated the full conversion of the

starting materials, the solvent was removed on a rotary evaporator and

the residue was puriﬁed by ﬂash column chromatography (hexane/

EtOAc = 6:1). The desired intermediate 66 was obtained through the

hydrolysis by LiOH in THF/H₂O (82% yield).

DIPEA (6 equiv) and HATU (1.2 equiv) were added to a solution of

66 (1.46 g, 8 mmol) and benzyl (2S,4R)-4-hydroxypyrrolidine-2-

carboxylate hydrochloride (1.1 equiv) in DMF. The mixture was stirred

at rt overnight and then the desired intermediate 70 was isolated

according to a published method.⁴⁰

Pd−C (10%) was added under H₂to a solution of 67 (386 mg, 1

mmol) in MeOH and stirred at rt for 2 h. Then, the solvent was

removed to aﬀord the product 68 without further puriﬁcation.

Compound 69 (1.17 g, 10 mmol), 70 (1 equiv), CuI (0.2 equiv), and

PdCl₂(PPh₃)₂(0.1 equiv) in DMF and TEA solvent were placed in a 25

mL round-bottom ﬂask under Ar. Then, the mixture was stirred for 4 h

at 100 °C. After this time, H₂O was added into the resulting complex

and extracted with EtOAc three times. The organic layer was again

washed with H₂O before being dried over MgSO₄, and the solvent was

removed under vacuum leaving the crude product. The pure product 71

was obtained by ﬂash column chromatography (hexane/EtOAc = 4:1)

in 30% yield.

A solution of 71 and tert-butyl 4-((4-aminophenyl)ethynyl)-

piperidine-1-carboxylate (1.1 g, 3 mmol) in 2-hydroxy-2-methylpropa-

nenitrile was reﬂuxed for 8 h. The intermediate (72) was obtained by

removing the solvent under reduced pressure and was used in the next

step without further puriﬁcation.

P

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 10. Compounds 37−38

a

Reaction conditions: (a) Pd(OAc)₂, KOAc, 90 °C; (b) HATU, DIPEA, DMF, rt; (c) TFA, DCM, rt; (d) HATU, DIPEA, DMF, rt; (e) HATU,

DIPEA, DMF, rt.

A solution of 72 (1.3 g, 3 mmol) and 4-isothiocyanato-2-

(triﬂuoromethyl)benzonitrile (1.1 equiv) in 20 mL of DMF was stirred

at 80 °C for 8 h. Then, 10 mL of MeOH and 10 mL of 2 N HCl were

added and the mixture was reﬂuxed for another 4 h. After UPLC−MS

demonstrated the full conversion of the starting materials, the reaction

mixture was cooled to rt and H₂O was added into the mixture. The

aqueous layer was extracted with EtOAc, and the combined organic

layers were washed with brine and then dried over anhydrous Na₂SO₄.

The solvent was removed on a rotary evaporator, and the residue was

puriﬁed by ﬂash column chromatography. The desired intermediate

(73) was isolated in 80% yield by deprotection in TFA/DCM.

K₂CO₃(1.2 equiv) and KI (0.2 equiv) were added to a solution of the

intermediate 73 (57.4 mg, 0.1 mmol) and tert-butyl 2-bromoacetate

(1.2 equiv) in CH₃CN. After stirring the mixture overnight at 100 °C,

the solvents were evaporated under reduced pressure to aﬀord the

corresponding crude compound that was puriﬁed by ﬂash column

chromatography (DCM/MeOH = 20:1). Compound 74 was obtained

through deprotection by TFA in DCM (75% yield).

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

74 (45 mg, 0.075 mmol) and compound 49a (1.1 equiv) in DMF (2

mL). After 30 min at rt, the mixture was subject to HPLC puriﬁcation to

aﬀord compound 24 in 88% yield. Following the procedures used to

prepare 24, compounds 20−23 with diﬀerent chain lengths were

obtained by the same methods.

(4R)-1-((S)-2-(4-(4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-

5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)ethynyl)-

pyridin-2-yl)piperazin-1-yl)butanamido)-3,3-dimethylbutanoyl)-4-

hydroxy-N-(4-(4-methylthiazol-5-yl)benzyl)pyrrolidine-2-carboxa-

mide (20). ¹H NMR (400 MHz, DMSO-d₆) δ = 9.0 (s, 1H, CONH),

8.6 (t, J = 4.0 Hz, 1H), 8.4 (m, 2H), 8.3 (m, 2H), 8.10 (m, 2H), 7.8 (m,

1H), 7.7 (m, 1H), 7.6 (m, 1H), 7.4 (m, 4H), 7.2 (s, 1H), 7.0 (s, CONH,

1H), 4.5 (d, 1H), 4.4 (d, 1H), 4.3 (m, 4H), 4.2 (m, 2H), 3.2 (m, 3H),

3.1 (m, 3H), 2.7 (m, 1H), 2.4 (s, 3H), 2.3 (m, 1H), 2.1 (m, 1H), 1.9 (m,

3H), 1.5 (s, 6H), 1.5 (s, OH, 1H), 1.3 (m, 2H), 1.2 (m, 1H), 0.9 (s, t-

butyl, 9H). UPLC−MS calcd for C₅₆H₆₀F₃N₁₀O₅S₂[M + H]⁺: 1073.41,

found 1073.54. UPLC-retention time: 5.0 min.

hydroxy-N-((S)-1-(4-(4-methylthiazol-5-yl)phenyl)ethyl)pyrrolidine-

1

2-carboxamide (21). H NMR (400 MHz, MeOD-d₄) δ 9.00−8.93

(m, 1H), 8.38 (d, J = 2.3 Hz, 1H), 8.18 (d, J = 8.8 Hz, 2H), 8.02 (dd, J =

8.3, 1.9 Hz, 1H), 7.79 (dt, J = 8.9, 1.9 Hz, 1H), 7.67 (dd, J = 8.5, 5.4 Hz,

2H), 7.50−7.41 (m, 6H), 6.99 (d, J = 9.0 Hz, 1H), 5.04 (d, J = 6.9 Hz,

1H), 4.61−4.53 (m, 3H), 4.45 (d, J = 4.8 Hz, 1H), 3.95 (d, J = 10.9 Hz,

1H), 3.78−3.63 (m, 3H), 3.53−3.46 (m, 1H), 3.28−3.14 (m, 5H),

2.70−2.57 (m, 2H), 2.51 (t, J = 2.5 Hz, 3H), 2.24−2.16 (m, 1H), 2.14−

1.93 (m, 4H), 1.61 (d, J = 3.6 Hz, 6H), 1.54 (d, J = 6.9 Hz, 3H), 1.39

(dd, J = 7.0, 3.6 Hz, 1H), 1.08 (d, J = 17.4 Hz, 9H). UPLC−MS calcd

for C₅₇H₆₂F₃N₁₀O₅S₂[M + H]⁺: 1087.43, found 1087.55. UPLC-

retention time: 5.2 min.

(2S,4R)-1-((S)-2-(5-(4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)-

phenyl)-5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)-

ethynyl)pyridin-2-yl)piperazin-1-yl)pentanamido)-3,3-dimethylbu-

tanoyl)-4-hydroxy-N-((S)-1-(4-(4-methylthiazol-5-yl)phenyl)ethyl)-

1

pyrrolidine-2-carboxamide (22). H NMR (400 MHz, MeOD-d₄) δ

9.01 (d, J = 2.5 Hz, 1H), 8.37 (d, J = 2.3 Hz, 1H), 8.20−8.16 (m, 2H),

8.02 (dd, J = 8.3, 2.1 Hz, 1H), 7.82−7.75 (m, 1H), 7.71−7.58 (m, 3H),

7.51−7.41 (m, 7H), 6.99 (d, J = 8.9 Hz, 1H), 5.03 (d, J = 7.0 Hz, 2H),

4.73−4.52 (m, 4H), 4.47 (s, 1H), 3.93 (d, J = 11.1 Hz, 1H), 3.78 (dd, J

= 11.0, 3.9 Hz, 1H), 3.64 (d, J = 12.8 Hz, 1H), 3.52−3.46 (m, 1H), 3.23

(s, 4H), 2.50 (d, J = 5.1 Hz, 3H), 2.42 (t, J = 6.9 Hz, 2H), 2.22 (dd, J =

12.8, 8.3 Hz, 1H), 2.00 (td, J = 8.9, 4.6 Hz, 1H), 1.79 (dt, J = 27.5, 7.7

Hz, 5H), 1.61 (q, J = 2.5 Hz, 6H), 1.53 (d, J = 6.9 Hz, 3H), 1.07 (d, J =

13.3 Hz, 9H). UPLC−MS calcd for C₅₈H₆₄F₃N₁₀O₅S₂[M + H]⁺:

1101.45, found 1101.53. UPLC-retention time: 4.9 min.

(4R)-1-((S)-2-(3-(4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-

5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)ethynyl)-

pyridin-2-yl)piperazin-1-yl)propanamido)-3,3-dimethylbutanoyl)-

4-hydroxy-N-((S)-1-(4-(4-methylthiazol-5-yl)phenyl)ethyl)-

1

pyrrolidine-2-carboxamide (23). H NMR (400 MHz, MeOD-d₄) δ

8.89 (d, J = 9.1 Hz, 1H), 8.53 (d, J = 7.5 Hz, 1H), 8.38 (d, J = 2.3 Hz,

1H), 8.19 (d, J = 1.6 Hz, 2H), 8.02 (dt, J = 8.2, 2.2 Hz, 1H), 7.78 (dd, J =

8.8, 2.3 Hz, 1H), 7.69−7.66 (m, 2H), 7.44 (dd, J = 9.2, 3.0 Hz, 6H),

6.98 (d, J = 8.9 Hz, 1H), 5.04 (t, J = 7.1 Hz, 1H), 4.57 (d, J = 5.3 Hz,

2H), 4.46 (s, 1H), 3.95 (d, J = 11.0 Hz, 1H), 3.78−3.71 (m, 2H), 3.63

(d, J = 11.2 Hz, 1H), 3.52−3.46 (m, 1H), 3.30−3.15 (m, 5H), 2.60 (t, J

= 6.3 Hz, 2H), 2.50 (d, J = 1.9 Hz, 3H), 2.21 (dd, J = 13.1, 7.6 Hz, 1H),

(4R)-1-((S)-2-(4-(4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-

5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)ethynyl)-

pyridin-2-yl)piperazin-1-yl)butanamido)-3,3-dimethylbutanoyl)-4-

Q

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

2.09 (t, J = 6.9 Hz, 2H), 1.98 (td, J = 9.2, 4.6 Hz, 1H), 1.61 (d, J = 3.6

Hz, 6H), 1.53 (d, J = 7.0 Hz, 3H), 1.39 (dd, J = 6.7, 3.5 Hz, 1H), 1.10 (s,

9H). UPLC−MS calcd for C₅₆H₆₀F₃N₁₀O₅S₂[M + H]⁺: 1073.41,

found 1073.49. UPLC-retention time: 4.8 min.

Hz, 1H), 1.98 (ddd, J = 13.1, 8.5, 5.1 Hz, 1H), 1.61 (s, 6H), 1.47−1.33

(m, 6H), 1.25 (d, J = 31.7 Hz, 6H), 1.07 (d, J = 6.5 Hz, 3H), 0.89 (dd, J

= 12.4, 6.6 Hz, 3H). UPLC−MS calcd for C₅₆H₆₄F₄N₉O₇S₂[M + H]⁺:

1114.43, found 1114.49. UPLC-retention time: 6.0 min.

General Procedure for Synthesis of Compounds 27 and 28.

K₂CO₃(1.2 equiv) and KI (0.2 equiv) were added to a solution of the

intermediate 73 (57.4 mg, 0.1 mmol) and tert-butyl (3-bromopropyl)-

carbamate (1.2 equiv) in CH₃CN was added. After stirring the mixture

overnight at 100 °C, the solvents were evaporated under reduced

pressure to aﬀord the corresponding crude compound that was puriﬁed

by ﬂash column chromatography (DCM/MeOH = 20:1). Then, 4-(3-

(4-((6-(4-(3-aminopropyl)piperazin-1-yl)pyridin-3-yl)ethynyl)-

phenyl)-4,4-dimethyl-5-oxo-2-thioxoimidazolidin-1-yl)-2-

(triﬂuoromethyl)benzonitrile was obtained through deprotection by

TFA in DCM (72% yield).

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

the compound 4-(3-(4-((6-(4-(3-aminopropyl)piperazin-1-yl)pyridin-

3-yl)ethynyl)phenyl)-4,4-dimethyl-5-oxo-2-thioxoimidazolidin-1-yl)-

2-(triﬂuoromethyl)benzonitrile (44 mg, 0.07 mmol) and 63 (1.1 equiv)

in DMF (2 mL). After 30 min at rt, the mixture was subject to HPLC

puriﬁcation to aﬀord compound (S)-3-amino-N-(3-(4-(5-((4-(3-(4-

cyano-3-(triﬂuoromethyl)phenyl)-5,5-dimethyl-4-oxo-2-thioxoimida-

zolidin-1-yl)phenyl)ethynyl)pyridin-2-yl)piperazin-1-yl)propyl)-3-(4-

(4-methylthiazol-5-yl)phenyl)propanamide in 81% yield after depro-

tection with TFA/DCM.

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

(S)-3-amino-N-(3-(4-(5-((4-(3-(4-cyano-3-(triﬂuoromethyl)phenyl)-

5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)ethynyl)-

pyridin-2-yl)piperazin-1-yl)propyl)-3-(4-(4-methylthiazol-5-yl)-

phenyl)propanamide (44 mg, 0.05 mmol) and 68 (1.1 equiv) in DMF

(2 mL). After 30 min at rt, the mixture was subject to HPLC

puriﬁcation to aﬀord compound 27 in 84% yield. Following the

procedures used to prepare compound 27, compound 28 with diﬀerent

chain lengths was obtained with the same methods.

(2S,4R)-N-((S)-3-((3-(4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)-

phenyl)-5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)-

ethynyl)pyridin-2-yl)piperazin-1-yl)propyl)amino)-1-(4-(4-methyl-

thiazol-5-yl)phenyl)-3-oxopropyl)-4-hydroxy-1-((R)-3-methyl-2-(3-

methylisoxazol-5-yl)butanoyl)pyrrolidine-2-carboxamide (27). ¹H

NMR (400 MHz, MeOD-d₄) δ 9.02−8.94 (m, 1H), 8.38 (t, J = 4.2 Hz,

1H), 8.22−8.14 (m, 2H), 8.02 (dd, J = 8.2, 2.0 Hz, 1H), 7.79 (td, J = 8.4,

2.4 Hz, 1H), 7.73−7.67 (m, 2H), 7.64 (s, 1H), 7.58−7.42 (m, 4H), 6.98

(dd, J = 26.1, 8.9 Hz, 1H), 5.44−5.32 (m, 1H), 4.72−4.66 (m, 1H),

4.60 (s, 1H), 4.49 (s, 1H), 3.89 (d, J = 11.0 Hz, 1H), 3.77 (dd, J = 11.0,

4.2 Hz, 1H), 3.68−3.49 (m, 2H), 3.31−2.83 (m, 7H), 2.53 (d, J = 5.3

Hz, 3H), 2.30−2.11 (m, 2H), 2.00 (s, 3H), 1.78 (d, J = 8.1 Hz, 2H),

1.66 (d, J = 6.6 Hz, 1H), 1.61 (s, 6H), 1.48−1.27 (m, 2H), 1.08 (s, 1H),

1.00 (s, 6H). UPLC−MS calcd for C₆₀H₆₃F₃N₁₁O₆S₂[M + H]⁺:

1154.44, found 1154.56. UPLC-retention time: 5.6 min.

(4R)-1-((S)-2-(2-(4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-

5,5-dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)phenyl)ethynyl)-

pyridin-2-yl)piperazin-1-yl)acetamido)-3,3-dimethylbutanoyl)-4-

hydroxy-N-((S)-1-(4-(4-methylthiazol-5-yl)phenyl)ethyl)pyrrolidine-

2-carboxamide (24). ¹H NMR (400 MHz, MeOD-d₄) δ 10.00−9.92

(m, 1H), 8.36−8.29 (m, 1H), 8.23−8.10 (m, 3H), 8.10−7.96 (m, 2H),

7.74 (dd, J = 8.5, 2.6 Hz, 2H), 7.60−7.47 (m, 6H), 7.42 (dd, J = 6.6, 3.5

Hz, 1H), 5.07 (d, J = 6.3 Hz, 1H), 4.68 (s, 1H), 4.65−4.58 (m, 1H),

4.49 (s, 1H), 4.23−4.05 (m, 4H), 3.95 (d, J = 11.0 Hz, 1H), 3.78 (dd, J

= 11.2, 4.2 Hz, 1H), 3.64 (s, 2H), 3.02 (s, 1H), 2.89 (s, 1H), 2.63 (s,

3H), 2.26 (dd, J = 13.2, 7.8 Hz, 1H), 2.06−1.94 (m, 1H), 1.61 (s, 6H),

1.54 (d, J = 7.0 Hz, 3H), 1.39 (dd, J = 6.7, 3.5 Hz, 3H), 1.09 (d, J = 9.9

Hz, 9H). UPLC−MS calcd for C₅₅H₅₈F₃N₁₀O₅S₂[M + H]⁺: 1059.40,

found 1059.45. UPLC-retention time: 5.2 min.

General Procedure for Synthesis of Compound 25. Triphosgene (1

equiv) was added to a solution of compound 49b (44.4 mg, 0.1 mmol)

in DCM and DIPEA (4 equiv) at rt. The reaction mixture was stirred at

rt for 30 min, and then compound 73 (1 equiv) was added. After an

additional 1 h at rt, the mixture was subject to HPLC puriﬁcation to

aﬀord compound 25 in 83% yield.

4-(5-((4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-5,5-dimethyl-4-

oxo-2-thioxoimid-azolidin-1-yl)phenyl)ethynyl)pyridin-2-yl)-N-

((2S)-1-((4R)-4-hydroxy-2-(((S)-1-(4-(4-methylthiazol-5-yl)phenyl)-

ethyl)carbamoyl)pyrrolidin-1-yl)-3,3-dimethyl-1-oxobutan-2-yl)-

1

piperazine-1-carboxamide (25). H NMR (400 MHz, MeOD-d₄) δ

9.01 (s, 1H), 8.29 (d, J = 2.3 Hz, 1H), 8.23−8.16 (m, 2H), 8.02 (dd, J =

8.3, 2.0 Hz, 1H), 7.92 (dt, J = 9.3, 2.5 Hz, 1H), 7.75−7.67 (m, 2H), 7.55

(d, J = 8.5 Hz, 1H), 7.50−7.42 (m, 5H), 7.15 (dd, J = 9.4, 3.5 Hz, 1H),

6.84 (dd, J = 8.5, 2.3 Hz, 1H), 5.04 (d, J = 7.0 Hz, 1H), 4.60 (d, J = 8.5

Hz, 1H), 4.47 (d, J = 4.8 Hz, 1H), 3.95 (d, J = 11.2 Hz, 1H), 3.78 (q, J =

3.9 Hz, 4H), 3.69 (d, J = 5.0 Hz, 3H), 3.65−3.59 (m, 1H), 3.23 (d, J =

7.4 Hz, 1H), 2.51 (s, 2H), 2.26−2.18 (m, 1H), 1.99 (ddd, J = 13.1, 9.0,

4.4 Hz, 1H), 1.61 (s, 6H), 1.54 (d, J = 7.1 Hz, 3H), 1.39 (dd, J = 6.7, 3.5

Hz, 3H), 1.20 (t, J = 7.1 Hz, 1H), 1.09 (s, 9H). UPLC−MS calcd for

C₅₄H₅₆F₃N₁₀O₅S₂[M + H]⁺: 1045.38, found 1045.42. UPLC-retention

time: 6.4 min.

General Procedure for Synthesis of Compound 26. DIPEA (5

equiv) and HATU (1.2 equiv) were added to a solution of compound

50 (45.1 mg, 0.1 mmol) and tert-butyl-(9-aminononyl)carbamate (1.1

equiv) in DMF (2 mL). After 30 min at rt, the mixture was subject to

HPLC puriﬁcation to aﬀord Boc protected compound 54 in 90% yield.

Then, compound 54 was obtained by a deprotection reaction in TFA/

DCM solvent.

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

54 (47 mg, 0.08 mmol) and 63 (1.1 equiv) in DMF (2 mL). After 30

min at rt, the mixture was subject to HPLC puriﬁcation to aﬀord

compound (S)-N-(9-(3-amino-3-(4-(4-methylthiazol-5-yl)phenyl)-

propanamido)nonyl)-4-(3-(4-cyano-3-(triﬂuoromethyl)phenyl)-5,5-

dimethyl-4-oxo-2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamide in 80%

yield after deprotection in TFA/DCM.

(2S,4R)-N-((S)-3-(4-(5-((4-(((1r,3r)-3-(3-Chloro-4-cyanophenoxy)-

2,2,4,4-tetramethylcyclobutyl)carbamoyl)phenyl)ethynyl)pyridin-

2-yl)piperazin-1-yl)-1-(4-(4-methylthiazol-5-yl)phenyl)-3-oxoprop-

yl)-4-hydroxy-1-((R)-3-methyl-2-(3-methylisoxazol-5-yl)butanoyl)-

1

pyrrolidine-2-carboxamide (28). H NMR (400 MHz, MeOD-d₄) δ

9.01 (s, 1H), 8.28 (s, 1H), 7.86 (d, J = 7.8 Hz, 3H), 7.75 (d, J = 8.7 Hz,

1H), 7.65 (d, J = 7.9 Hz, 2H), 7.57−7.41 (m, 4H), 7.16 (s, 1H), 7.10 (d,

J = 8.2 Hz, 1H), 7.01 (d, J = 8.8 Hz, 1H), 6.24 (d, J = 28.0 Hz, 1H), 5.45

(d, J = 26.8 Hz, 1H), 4.61−4.46 (m, 2H), 4.32 (s, 1H), 4.20 (s, 1H),

3.93−3.87 (m, 1H), 3.80−3.48 (m, 10H), 3.07 (ddd, J = 22.9, 19.4, 11.3

Hz, 2H), 2.48 (s, 3H), 2.39 (d, J = 6.2 Hz, 1H), 2.26 (d, J = 6.0 Hz, 3H),

2.05−1.97 (m, 1H), 1.32 (s, 6H), 1.26 (s, 6H), 1.11−1.02 (m, 3H),

0.92−0.83 (m, 3H). UPLC−MS calcd for C₆₀H₆₅ClN₉O₇S [M + H]⁺:

1090.44, found 1090.56. UPLC-retention time: 6.6 min.

General Procedure for Synthesis of Compounds 29 and 30.

Compound 75 (2.19 g, 10 mmol), 76 (1.1 equiv), CuI (0.2 equiv), and

PdCl₂(PPh₃)₂(0.1 equiv) in DMF and TEA were placed in a 25 mL

round-bottom ﬂask under Ar. Then, the mixture was stirred for 4 h at

100 °C. After this time, H₂O was added into the resulting complex and

extracted with EtOAc three times. The organic layer was again washed

with H₂O before being dried over MgSO₄, and the solvent was removed

under vacuum leaving the crude product. The pure product (77) was

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

(S)-N-(9-(3-amino-3-(4-(4-methylthiazol-5-yl)phenyl)propanamido)-

nonyl)-4-(3-(4-cyano-3-(triﬂuoromethyl)phenyl)-5,5-dimethyl-4-oxo-

2-thioxoimidazolidin-1-yl)-2-ﬂuorobenzamide (66 mg, 0.08 mmol)

and 68 (1.1 equiv) in DMF (2 mL). After 30 min at rt, the mixture was

subject to HPLC puriﬁcation to aﬀord compound 26 in 84% yield.

(2S,4R)-N-((S)-3-((9-(4-(3-(4-Cyano-3-(triﬂuoromethyl)phenyl)-

5 , 5 - di m e t h y l - 4 - o x o - 2 - t h i o x o i mi d a z o l i d in - 1 - y l ) - 2 -

ﬂuorobenzamido)nonyl)amino)-1-(4-(4-methylthiazol-5-yl)-

phenyl)-3-oxopropyl)-4-hydroxy-1-((R)-3-methyl-2-(3-methylisoxa-

zol-5-yl)butanoyl)pyrrolidine-2-carboxamide (26). ¹H NMR (400

MHz, MeOD-d₄) δ 9.78 (s, 1H), 8.26−8.15 (m, 2H), 8.02 (dd, J = 8.3,

2.0 Hz, 1H), 7.85 (t, J = 8.1 Hz, 1H), 7.56 (s, 4H), 7.40 (d, J = 10.1 Hz,

1H), 6.24 (s, 1H), 5.37 (dd, J = 8.0, 6.0 Hz, 1H), 4.49 (dd, J = 17.4, 9.4

Hz, 2H), 3.97−3.71 (m, 2H), 3.67−3.57 (m, 1H), 3.41 (t, J = 7.0 Hz,

3H), 3.11 (td, J = 6.8, 3.9 Hz, 2H), 3.02 (s, 1H), 2.94−2.69 (m, 3H),

2.60 (s, 3H), 2.43 (q, J = 9.4, 8.6 Hz, 1H), 2.27 (s, 2H), 2.18 (t, J = 10.8

R

DOI: 10.1021/acs.jmedchem.8b01631

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

obtained by ﬂash column chromatography (hexane/EtOAc = 4:1) in

80% yield.

A solution of 77 (2.4 g, 8 mmol) in 2-hydroxy-2-methylpropaneni-

trile was reﬂuxed for 8 h. Then, by removing the solvent under reduced

pressure, the intermediate 78 was obtained and used in the next step

without further puriﬁcation.

A solution of 94 (1.3 equiv) in THF was added to a mixture of NaH

(1.3 equiv) in THF at 0 °C and stirred for 5 min. Then, a solution of 93

(2.7 g, 10 mmol) in THF was added slowly. The mixture was stirred at

rt for 3 h. After UPLC−MS demonstrated the full conversion of starting

materials, the solvent THF was distilled oﬀ and some EtOAc was added

and the solution was washed with brine and H₂O. The combined

organic layers were dried over anhydrous Na₂SO₄. The solvent was

removed on a rotary evaporator giving the desired intermediate 95,

which was used without further puriﬁcation.

Compound 95 (460 mg, 1 mmol) was dissolved in DCM and 30%

H₂O₂(8 equiv) was added; then, the mixture was cooled to −55 °C and

triﬂuoroacetic anhydride (6 equiv) was added very slowly, keeping the

reaction temperature below 0 °C. After the addition was complete, the

reaction mixture was stirred at rt for 16 h. After UPLC−MS

demonstrated the full conversion of starting materials, ice/H₂O and

brine were added into the mixture, which was stirred for another 20

min; then, the organic layer was collected and dried with Na₂SO_4.The

solvent was removed on a rotary evaporator giving the desired

intermediate 96, which was puriﬁed by ﬂash column chromatography

(hexane/EtOAc = 2:1) in 80% yield.

Compounds 96 (390 mg, 0.8 mmol) and 92, tert-butyl 4-ethynyl-

[1,4′-bipiperidine]-1′-carboxylate (1.1 equiv), CuI (0.2 equiv), and

PdCl₂(PPh₃)₂(0.1 equiv) in DMF and TEA solvent were placed in a 25

mL round-bottom ﬂask under Ar. Then, the mixture was stirred for 4 h

at 100 °C. Then, H₂O was added into the resulting complex, which was

extracted with EtOAc three times. The organic layer was again washed

with H₂O before being dried over MgSO₄and the solvent was removed

under vacuum leaving the crude product. The pure product was

obtained by ﬂash column chromatography (DCM/MeOH = 20:1).

Then, compound 97 was obtained through the deprotection by TFA in

DCM (82% yield).

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

compounds 97 (60.2 mg, 0.1 mmol) and 63 (1.1 equiv) in DMF (2

mL). After 30 min at rt, the mixture was subject to HPLC puriﬁcation to

aﬀord compound 98 in 88% yield after deprotection in TFA/DCM.

DIPEA (5 equiv) and HATU (1.2 equiv) were added to a solution of

compounds 98 (67.7 mg, 0.08 mmol) and 68 (1.1 equiv) in DMF (2

mL). After 30 min at rt, the mixture was subject to HPLC puriﬁcation to

aﬀord compound 31 in 83% yield.

(2S,4R)-N-((1S)-3-(4-((4-((3-((4-Cyano-3-(triﬂuoromethyl)phenyl)-

amino)-2-hydroxy-2-methyl-3-oxopropyl)sulﬁnyl)phenyl)ethynyl)-

[1,4′-bipiperidin]-1′-yl)-1-(4-(4-methylthiazol-5-yl)phenyl)-3-oxo-

propyl)-4-hydroxy-1-((R)-3-methyl-2-(3-methylisoxazol-5-yl)-

butanoyl)pyrrolidine-2-carboxamide (31). ¹H NMR (400 MHz,

MeOD-d₄) δ 8.93 (s, 1H), 8.61 (dd, J = 16.4, 8.6 Hz, 1H), 8.26 (s, 1H),

8.07−7.83 (m, 4H), 7.47 (d, J = 28.4 Hz, 5H), 6.31−6.20 (m, 1H), 5.45

(d, J = 25.2 Hz, 2H), 4.71 (s, 1H), 4.59−4.41 (m, 2H), 4.29 (s, 2H),

4.12 (dd, J = 14.8, 1.7 Hz, 1H), 3.96−3.70 (m, 3H), 3.70−3.44 (m,

6H), 3.27−3.05 (m, 4H), 2.95 (s, 2H), 2.70−2.60 (m, 1H), 2.52 (t, J =

5.1 Hz, 3H), 2.39 (d, J = 32.8 Hz, 2H), 2.27 (d, J = 6.3 Hz, 3H), 2.16 (s,

2H), 2.03−1.89 (m, 2H), 1.79−1.66 (m, 1H), 1.58−1.47 (m, 3H),

1.41−1.27 (m, 2H), 1.12−0.96 (m, 3H), 0.89 (s, 3H). UPLC−MS

calcd for C₅₆H₆₄F₄N₉O₇S₂[M + H]⁺: 1125.42, found 1125.49. UPLC-

retention time: 4.0 min.

General Procedure for Synthesis of Compounds 32. Compound 82

(2.62 g, 10 mmol), 76 (1.1 equiv), CuI (0.2 equiv), and PdCl₂(PPh₃)₂

(0.1 equiv) in DMF and TEA solvent were placed in a 25 mL round-

bottom ﬂask under Ar. Then, the mixture was stirred for 4 h at 100 °C.

After this time, H₂O was added into the resulting complex and extracted

with EtOAc three times. The organic layer was again washed with H₂O

before being dried over MgSO₄, and the solvent was removed under

vacuum leaving the crude product. The pure product was obtained by

ﬂash column chromatography (hexane/EtOAc = 4:1). Then, 83 was

obtained through deprotection by TFA in DCM in 90% yield.

K₂CO₃(1.2 equiv) and KI (0.2 equiv) were added to a solution of the

deprotected intermediate 83 (1.94 g, 8 mmol) tert-butyl 4-

bromopiperidine-1-carboxylate (1.2 equiv) in CH₃CN. After stirring

the mixture overnight at 100 °C, the solvents were evaporated under

reduced pressure to aﬀord the corresponding crude compound 4-((4-

(methoxycarbonyl)phenyl)ethynyl)-[1,4′-bipiperidine]-1′-carboxy-

A solution of 78 (2.9 g, 8 mmol) and 4-isothiocyanato-2-