Syntheses, Biological Evaluations, and Mechanistic Studies of Benzo[c][1,2,5]oxadiazole Derivatives as Potent PD-L1 Inhibitors within VivoAntitumor Activity

pubs.acs.org/jmc

Article

Syntheses, Biological Evaluations, and Mechanistic Studies of

Benzo[c][1,2,5]oxadiazole Derivatives as Potent PD-L1 Inhibitors

with In Vivo Antitumor Activity

Liu Liu,^∥Zhiying Yao,^∥Shijun Wang,^∥Tao Xie, Guoqing Wu, Honghan Zhang, Pu Zhang, Yaojun Wu,

Haoliang Yuan,* and Hongbin Sun *

Cite This: J. Med. Chem. 2021, 64, 8391 −8409

Read Online

sı

*

Metrics & More

Article Recommendations

Supporting Information

ACCESS

ABSTRACT: A series of novel benzo[c][1,2,5]oxadiazole deriva-

tives were designed, synthesized, and biologically evaluated as

inhibitors of PD-L1. Among them, compound L7 exhibited 1.8 nM

IC₅₀value in a homogeneous time-resolved ﬂuorescence (HTRF)

assay, which was 20-fold more potent than the lead compound

BMS-1016. In the surface plasmon resonance (SPR) assay, L7

bound to human PD-L1 (hPD-L1) with a K_Dvalue of 3.34 nM,

without showing any binding to hPD-1. In the cell-based coculture

assay, L7 blocked PD-1/PD-L1 interaction with an EC₅₀value of

375 nM, while BMS-1016 had an EC₅₀value of 2075 nM. Moreover, compound L24, an ester prodrug of L7, was orally bioavailable

and displayed signiﬁcant antitumor eﬀects in tumor models of syngeneic and PD-L1 humanized mice. Mechanistically, L24 exhibited

signiﬁcant in vivo antitumor eﬀects probably through promoting antitumor immunity. Together, this series of benzoxadiazole PD-L1

inhibitors holds promise for tumor immunotherapy. Preclinical trials with selected compounds are ongoing in our laboratory.

like extracellular domains, essentially via hydrophobic and

polar interactions involving two front β-sheets and connecting

loops.¹¹Like most of the B7 protein family members, PD-L1

could form a homodimer in the solution or crystalline state,

either in the presence or absence of its ligand.¹¹⁻¹³Although

the structural studies on the hPD-1/hPD-L1 complex pointed

to the 1:1 stoichiometry, it cannot be ruled out the possibility

that the PD-L1 homodimer binds to PD-1 since no steric

hindrance seems to exist in this context.¹³By far, some

nonpeptide small-molecule and peptidomimetic inhibitors of

the PD-1/PD-L1 protein−protein interaction have been

developed (Figure 1).^8,14−17Bristol-Myers Squibb (BMS)

has developed a series of biphenyl compounds including BMS-

202 (1)¹⁸and BMS-1016 (2),¹⁹which are among the most

potent lead compounds of PD-L1 inhibitors. Cocrystal

structures of the complexes of PD-L1 bound to BMS-202

and related compounds show that a single BMS compound

binds to two PD-L1 protein molecules, indicating the

formation of PD-L1 homodimers induced and stabilized by

the small-molecule compounds.^10,11,20PD-L1 inhibitors 3/4/5

INTRODUCTION

■

In the past decade, cancer immunotherapy based on the

blockade of immune checkpoints has made great strides.^1,2In

particular, monoclonal antibodies (mAbs) of programmed cell

death protein 1 (PD-1) and programmed cell death-ligand 1

(PD-L1) have been very successful for the treatment of various

tumors, including nonsmall cell lung cancer (NSCLC),

urothelial cancer, melanoma, head and neck squamous cell

cancer, and lymphoma.³⁻⁵Recently, encouraging results from

several phase III clinical trials showed that combinations of

PD-1/PD-L1 antibodies and antiangiogenic agents are very

eﬀective for treating NSCLC (adenocarcinoma) and hepato-

cellular carcinoma.^6,7Until now, there are six PD-1/PD-L1

antibodies (anti-PD-1, nivolumab, pembrolizumab, and

cemiplimab; anti-PD-L1, atezolizumab, avelumab, and durva-

lumab) approved by the US Food and Drug Administration

(FDA) for clinical applications.⁸Despite their broad

applicability in cancer treatment, the shortcomings of PD-1/

PD-L1 antibodies exist, such as low response rates with a range

of 10−30% in most cancers, autoimmune symptoms, and

tumor hyperprogression.^8,9On the other hand, small-molecule

inhibitors of the PD-1/PD-L1 protein−protein interaction

have attracted much attention due to their absent or weak

immunogenicity, improved tumor penetration, better oral

bioavailability, manipulative immune-related adverse eﬀects

(irAEs), and low cost.^8,10

Received: March 4, 2021

Published: June 11, 2021

Both PD-1 and PD-L1 are membrane proteins and interact

with each other through their canonical immunoglobulin (Ig)-

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

8391

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 1. Chemical structures of small-molecule inhibitors of PD-1/PD-L1 interaction.

Figure 2. Binding modes of BMS-1016 (A) and L1 (B) in the pocket of the PD-L1 homodimer predicted by molecular docking. The π−π stacking

interactions were shown as a green dash line, π−cation interactions were shown as an orange dash line, and hydrogen-bonding interactions were

shown as a magenta dash line. (C) Alignment of BMS-1016 (gray) and L1 (green) in the cleft formed by the PD-L1 dimer (the docking model was

based on the dimeric PD-L1 protein taken from PDB ID: 5J89). (D) Design of benzo[c][1,2,5]oxadiazole small-molecule inhibitors of PD-1/PD-

L1 protein−protein interaction.

were developed on the basis of BMS compounds. Peptidomi-

metic inhibitors 6/7 were developed by Curis and Aurigene.

One of the most advanced small-molecule immune checkpoint

inhibitors is the peptidomimetic compound CA-170, which is

thought to be a dual inhibitor of PD-L1 and V-domain Ig

suppressor of T-cell activation (VISTA). In a phase II clinical

trial on multitumors, CA-170 was well tolerated and showed

clinical beneﬁt across tumor types with an incidence of irAEs

(https://www.curis.com/pipeline/ca-170/). Nevertheless, the

results of a comprehensive characterization of small-molecule

PD-L1 inhibitors demonstrated that CA-170 had neither direct

binding to PD-L1 nor interference on PD-1/PD-L1

interaction.²¹Other small-molecule PD-L1 inhibitors entering

clinical trials include INCB86550 (Incyte Corp.), GS-4224

(Gilead), MAX-10181 (MaxiNovel Pharma., China), etc.

The challenges of developing small-molecule PD-L1

inhibitors for clinical applications still exist, such as inadequate

eﬃcacy and speciﬁcity compared to mAbs, lack of mechanistic

insight into their in vivo immunoactivation activity, and

incomplete structural information for ligand recognition. In

this study, a series of benzo[c][1,2,5]oxadiazole derivatives (8)

based on the BMS scaﬀold were designed and synthesized as

new PD-L1 inhibitors. The results of in vitro assays, including

homogeneous time-resolved ﬂuorescence (HTRF), surface

plasmon resonance (SPR), coculture assay, identiﬁed com-

pound L7 as a potent PD-L1 inhibitor that blocked PD-1/PD-

L1 interaction at both protein and cellular levels. An orally

bioavailable prodrug of L7 exhibited signiﬁcant antitumor

activity in tumor models of syngeneic and PD-L1 humanized

mice. Mechanistically, this prodrug (L24) suppressed tumor

8392

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

growth likely through upregulating cytotoxic T and NK cells

while downregulating immunosuppressive cells.

Table 1. PD-1/PD-L1 Blockade Eﬀect of Benzoheterocycle

Compounds L1−L4

a

RESULTS AND DISCUSSION

■

Molecular Design. The cocrystal structure of BMS-202

bound to PD-L1 (PDB ID: 5J89) reveals that the biphenyl

moiety of BMS-202 is located in a deep, cylindrical,

hydrophobic pocket created by the interface of the dimeric

PD-L1 protein to generate multiple hydrophobic interactions

with surrounding amino acid residues.²⁰Multiple experiments

have proved that the biphenyl moiety is essential for the

inhibitors to block the interaction of PD-1/PD-L1.¹¹With

using BMS-1016 as a lead compound, we hoped to develop

more potent PD-L1 inhibitors with new scaﬀolds. Molecular

docking of BMS-1016 to the PD-L1 homodimer revealed that

π−π stacking interactions were formed between ring A and

Tyr56_Aand ring C and Tyr56_B(Figure 2A). The hydrophilic

D-serine moiety formed hydrogen-bonding interactions with

adjacent residues Thr20_A, Asp122_A, and Lys124_A. Ring D

occupied a pocket surrounded by Tyr123_A, Lys124_A, and

Arg125_A. A single hydrogen bond was generated between the

CN group of 3-cyanobenzyl substituent and Arg125_A, while π-

cation interaction was found between ring D and Lys124_A

(Figure 2A). Of note, the pharmacophoric features of this

pocket indicated that it was possible to ﬁnd more potent PD-

L1 inhibitors by replacing the 3-cyanobenzyl side chain with

groups, forming additional protein−ligand interactions. Con-

sidering the aromatic property of the Tyr123_Aresidue, along

with the positive charged Arg125_Aand Lys124_A, we focused on

using benzoheterocyclic groups as the alternative substituents

to achieve stronger binding aﬃnity with better binding mode.

Several benzoheterocyclic compounds L1−L4 (Table 1) were

thus designed as close analogues of BMS-1016. Among them,

benzoxadiazole compound L1 showed well alignment of the

binding mode with BMS-1016 (Figure 2B,C). Importantly, the

benzoxadiazole moiety showed additional π−π stacking

interaction with Tyr123_Awhile keeping all of the other

interactions. The Glide docking scores of BMS-1016 and L1

were −13.085 and −13.195 kcal/mol, respectively, indicating

the good potential of L1 as a novel lead compound of PD-L1

inhibitors.

Chemistry. Synthesis of compounds L1−L4 is depicted in

Scheme 1. Selective O-alkylation of 2,4-dihydroxybenzalde-

hyde with benzyl bromide 9 in the presence of NaHCO₃in

reﬂuxing acetonitrile aﬀorded phenol-aldehyde 10 in 83%

yield. Further O-alkylation of 10 with benzoheterocyclic benzyl

bromide in the presence of cesium carbonate gave aldehydes

11a−d in 92−99% yields (see the Supporting Information, SI).

Borch reductive amination of 11a−d with ^D-serine ethyl ester

in the presence of sodium triacetoxyborohydride followed by

hydrolysis of the resulting esters provided the target

compounds L1−L4.

In a similar pattern, compounds L5−L12 and L24−L42

were synthesized (Scheme 2). Therefore, dihydroxybenzalde-

hydes X1−3 were selectively alkylated with benzylbromides (9,

12a−d) to aﬀord hydroxybenzaldehydes 13a−h, which were

further alkylated with 5-(bromomethyl)-2,1,3-benzoxadiazole

to furnish aldehydes 14a−h in 89−93% yields (see the SI).

Reductive amination of 14a−h with ^D-serine ethyl ester

followed by hydrolysis of the resulting esters gave compounds

L5−L12. Reductive amination of 14c with ^D-serine ethyl ester

furnished ester prodrug L24. Reductive amination of 14c with

L-serine ethyl ester furnished compound L25 as an enantiomer

a

The data were generated from two independent experiments.

of L7. Reductive amination of 14c or 14d with other amino

acid esters followed by hydrolysis of the resulting esters gave

compounds L26, L28, L30, L37, L41, and L42. Reductive

amination of 14c or 14d with other amines aﬀorded

compounds L27, L29, L31−36, and L38−40.

Synthesis of compounds L13−L23 is depicted in Scheme 3.

Selective O-alkylation of 2,4-dihydroxy-5-chlorobenzaldehyde

with 3-iodo-benzylbromides 15a−b aﬀorded phenol-aldehydes

16a−b in 65−75% yields. The Suzuki−Miyaura coupling

reaction between 16a−b and phenylboronic acid derivatives

provided biphenyl compounds 17a−k. O-Alkylation of 17a−k

with 5-(bromomethyl)-2,1,3-benzoxadiazole gave benzoxadia-

zole compounds 18a−k (see the SI). Reductive amination of

18a−k with ^D-serine ethyl ester followed by hydrolysis of the

resulting esters provided compounds L13−L23 in 19−33

yields (two steps).

HTRF Assay for PD-1/PD-L1 Blockade and Structure−

Activity Relationship (SAR) Analysis. The blockade eﬀect

of the compounds on PD-1/PD-L1 interaction was examined

by the HTRF assay. As shown in Table 1, benzoheterocyclic

compounds L1−L4 had substantially varying potency with

IC₅₀values ranging from nanomoles to micromoles. Among

them, L1 with a benzo[c][1,2,5]oxadiazol-5-ylmethoxy moiety

exhibited potency (IC₅₀= 33.7 nM) comparable to that of

BMS-1016 (IC₅₀= 35.8 nM). The potency of L2 (IC₅₀

=

2159.5 nM) with a benzo[c][1,2,5]oxadiazol-4-ylmethoxy

moiety and L4 (IC₅₀= 2343.5 nM) with a benzofuran-6-

ylmethoxy moiety signiﬁcantly decreased. L3 (IC₅₀= 205.6

8393

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Scheme 1. Synthesis of Compounds L1−L4

a

Reagents and conditions: (a) 2,4-dihydroxybenzaldehyde, NaHCO₃, MeCN, reﬂux, 83%. (b) R-Br, Cs₂CO₃, N,N-dimethylformamide (DMF), rt,

92−99%. (c) ^D-Serine ethyl ester hydrochloride, NaBH(OAc)₃, N,N-diisopropylethylamine (DIPEA), dichloromethane (DCM), AcOH, rt. (d)

LiOH, MeOH, rt, then neutralized by 1 N HCl, two steps, 13−26%.

a

Scheme 2. Synthesis of Compounds L5−L12 and L24−L42

a

Reagents and conditions: (a) NaHCO₃, MeCN, reﬂux, 63−82%. (b) 5-(Bromomethyl)-2,1,3-benzoxadiazole, Cs₂CO₃, DMF, rt, 89−93%. (c) D-

Serine ethyl ester hydrochloride or other amines, NaBH(OAc)₃, DCM, AcOH, rt, 19−30%. (d) In case amino acid esters were used for the

reductive amination reaction, hydrolysis of the resulting esters was performed using the following conditions: LiOH, MeOH, rt, then neutralized by

1 N HCl.

a

Scheme 3. Synthesis of Compounds L13−L23

a

Reagents and conditions: (a) 2,4-dihydroxy-5-chlorobenzaldehyde, NaHCO₃, MeCN, reﬂux, 65−75%. (b) Phenylboronic acid derivatives,

Pd(dppf)Cl₂·CH₂Cl₂, K₂CO₃, PhMe, EtOH, H₂O, reﬂux, 40−48%. (c) 5-(Bromomethyl)-2,1,3-benzoxadiazole, Cs₂CO₃, DMF, rt, 82−90%. (d) D-

Serine ethyl ester hydrochloride, NaBH(OAc)₃, DIPEA, DCM, AcOH, rt. (e) LiOH, MeOH, rt, then neutralized by 1 N HCl, two steps, 19−33%.

nM) with a benzo[c][1,2,5]thiadiazol-5-ylmethoxy moiety was

sixfold less potent than L1. Together, the benzo[c][1,2,5]-

oxadiazol-5-ylmethoxy unit of L1 was proved to be a privileged

scaﬀold for the design of new PD-1/PD-L1 inhibitors.

8394

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

With L1 as a novel lead compound, structural modiﬁcations

on the biphenyl unit (B ring) and the central benzene ring (C

ring) were conducted to provide compounds L5−L12 (Table

2). HTRF assay results showed that replacement of the methyl

group of L1 with triﬂuoromethyl group caused a sharp

decrease in potency (L5: IC₅₀= 1384.5 nM). Introduction of

5-Cl substitution at the central benzene ring of L1 led to

increased potency (L6: IC₅₀= 16.6 nM). Notably, replacement

of the methyl group of L6 with bromine, chlorine, or ﬂuorine

caused signiﬁcantly improved potency (L7: IC₅₀= 1.8 nM; L8:

IC₅₀= 3.3 nM; L9: IC₅₀= 3.5 nM). Upon 5-Cl substitution of

L7 switched into 5-Br or 5-F substitution, the potency was

signiﬁcantly decreased (L10: IC₅₀= 27.5 nM; L11: IC₅₀= 7.8

nM). Replacement of 5-Cl of L8 with 5-F also caused

decreased potency (L12: IC₅₀= 9.5 nM).

Table 2. PD-1/PD-L1 Blockade Eﬀect of Compounds L5−

a

L23 with Substitutions on A−C Rings

Further SAR exploration based on the terminal benzene ring

(A ring) of L7 and L8 was conducted, resulting in compounds

L13−L23 (Table 2). Introduction of substituents at the

terminal benzene ring generally led to decreased activity. For

example, introduction of a para-methylsulfonyl group resulted

in a big loss of activity (L13: IC₅₀> 1 μM). Of note, the

compounds with meta-substitution (L15: IC₅₀= 6.5 nM) were

more potent than the compounds with para-substitution (L14:

IC₅₀= 56.1 nM).

The hydrophilic ^D-serine tail of L7 and related compounds

served as hydrogen-bond donors and played an important role

in binding with PD-L1. To further investigate the variability of

this hydrophilic tail, compounds L24−L42 with various

hydrogen-bond donors or acceptors replacing the ^D-serine

moiety were synthesized and biologically evaluated (Table 3).

Compound L24, an ethyl ester of L7, showed signiﬁcantly

decreased potency (IC₅₀= 425.9 nM), indicating the critical

role of the free acid group in binding with PD-L1. Compound

L25, a L-serine analogue of L7, showed similar potency (IC₅₀

= 2.6 nM) to L7, suggesting that the chiral conﬁguration of the

serine moiety had little eﬀect on the activity. Other

compounds (L33−L42) with diﬀerent hydrophilic side chains

showed variable potency (IC₅₀= 13.3−3592.5 nM) but were

much less potent than L7.

Docking Study on L7. Molecular docking was used to

predict the binding mode of L7 (Figure 3). In comparison with

the binding mode of L1, the diﬀerent substituted halogens (Br

on ring B and Cl on ring C) of L7 strengthened the conserved

π−π stacking interactions with Tyr56_Aand Tyr56_B. The

hydrogen-bonding interactions with Thr20_A, Asp122_A, and

Lys124_Awere also kept. The π−π stacking interaction and π-

cation interaction dominated the protein−ligand interactions

between the benzoxadiazole unit and the pocket surrounded by

Tyr123_A, Lys124_A, and Arg125_A, which was diﬀerent from the

binding mode of L1. The Glide docking score of L7 (−13.379

kcal/mol) also indicated the more potent binding aﬃnity of L7

to PD-L1 (Table S1).

Binding Aﬃnity of L7 to hPD-1 and hPD-L1. The

kinetic analysis of the interaction of L7 with hPD-L1 or hPD-1

was performed through a surface plasmon resonance (SPR)

method. As shown in Figure 4, the calculated K_Dvalue for L7

binding to the hPD-L1 extracellular segment was 3.34 nM.

Notably, there was no binding for hPD-1 upon treatment with

various concentrations of L7, indicating that this compound

took eﬀects through binding to PD-L1 but not PD-1.

Cytotoxicity Evaluation. To exclude the cytotoxic eﬀects

of the test compounds on the subsequent in vitro and in vivo

assays, the cytotoxicity of the compounds (L7, L8, L9, L11,

a

The data were generated from two independent experiments.

8395

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Table 3. PD-1/PD-L1 Blockade Eﬀect of Compounds L24−L42 with Hydrophilic Tail Substitutions

a

The data were generated from two independent experiments.

L12, L15, L20, L21, L28, BMS-202, and BMS-1016) was

evaluated in PD-1⁺/NFAT-luc/Jurkat cells, PD-L1⁺/aAPC/

CHO-K1 cells, and MC38 cells (Figure S1). In the above cell

lines, L7, L9, L11, L15, and L20 did not aﬀect cell viability at

various concentrations (0.1−10 μM), while L8, L12, L21, and

L28 only showed slight inhibition of cell viability at 10 μM. Of

note, BMS-202 and BMS-1016 signiﬁcantly decreased cell

viability at 10 μM. These results indicated that the

benzoxadiazole compounds had a lower cytotoxic eﬀect than

the BMS compounds, and the inﬂuence of cytotoxicity of our

compounds on the subsequent cellular assays and in vivo

antitumor experiments could be excluded.

8396

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Table 4. Data of Cell-Based PD-1/PD-L1 Signaling

a

Blockade Assay

compound

EC₅₀(nM)

RLU_MAX

atezolizumab

L7

BMS-1016

BMS-202

0.258 0.043

375 84.3

2075 573

>10 000

2.40 0.261

1.81 0.028

1.83 0.079

b

ND

a

b

The data were generated from two independent experiments. ND =

not determined.

segment of PD-L1 was ﬁxed on the cell membrane, thus

requiring more binding force to induce dimerization.

Figure 3. Binding mode of L7 in the pocket of the PD-L1 dimer

predicted by molecular docking (the docking model was based on the

dimeric PD-L1 protein taken from PDB ID: 5J89). The π−π stacking

interactions were shown as a cyan dash line, π-cation interactions

were shown as an orange dash line, and hydrogen-bonding

interactions were shown as a magenta dash line.

Pharmacokinetic Studies. Given the fact that L7

exhibited promising potency at both protein and cellular

levels, we examined the in vivo pharmacokinetic (PK)

properties of L7 and its ester prodrug L24. As shown in

Table 5, although the PK parameters of L7 intravenously

administered in male Sprague−Dawley (SD) rats were

reasonable, this compound had very poor oral bioavailability

(0.35%). This might be due to its low cellular permeability

(data not shown). To improve the oral exposure of L7, we

examined the PK property of L24, a potential prodrug of L7.

As expected, L24 was readily converted into L7 in vivo (Table

5 and Figure S2). Notably, oral administration of L24 resulted

in a signiﬁcant increase in bioavailability of L7 (9.12% in SD

rats; 7.70% in C57BL/6 mice). Moreover, oral half-lives of L24

detected in the form of L7 were 3.43 h in SD rats and 1.55 h in

mice. Together, L7 could be orally bioavailable through its

ester prodrug L24 and could readily achieve plasma

concentrations above the cellular eﬃcacious threshold.

Toxicity Studies. Before moving on to the in vivo

antitumor experiments, we examined the safety proﬁles of

L24 and L7. Acute and subacute toxicity studies on L24 were

conducted in healthy C57BL/6 mice. In an acute toxicity

study, oral administration of L24 at a single dose of 2000 mg/

kg resulted in no mortality and obvious body weight loss

during the 30 days observation period (Figure S3). In subacute

toxicity study, multidose administration of L24 for 30 days did

not lead to mortality, body weight loss, organ (liver, kidneys,

lungs, heart, and brain) weight changes, or macroscopic

pathologic ﬁndings at any dose levels (50, 100, and 300 mg/

kg) (Figure S4). Notably, the spleen weight slightly decreased

in the low- and medium-dose groups (data not shown),

implying possible immunomodulatory eﬀects of the com-

pound. Serum indicators such as aspartate aminotransferase

(AST), alanine aminotransferase (ALT), creatine kinase (CK),

Cell-Based PD-1/PD-L1 Signaling Blockade Assay.

The cell-based assay was conducted in a coculture system

containing PD-1⁺/NFAT-luc/Jurkat cells and PD-L1⁺/aAPC/

CHO-K1 cells. PD-1⁺/NFAT-luc/Jurkat cells are T lympho-

cyte-like cell lines (Jurkat), which are modiﬁed by carrying a

luciferase reporter gene driven by a T-cell receptor (TCR)-

inducible NFAT response element and constitutively express

hPD-1. PD-L1⁺/aAPC/CHO-K1 cells are antigen-presenting

surrogate CHO cells, which constitutively express the TCR

agonist and hPD-L1. TCR signaling could be transferred from

CHO cells to Jurkat cells but repressed by the interaction of

PD-1/PD-L1. The compounds that inhibit PD-1/PD-L1

interaction could release the TCR-mediated activation of the

Jurkat cells, evidenced by the increased activity of luciferase.

The ﬂuorescence intensity is proportional to the inhibition

potency. Atezolizumab, a clinically relevant anti-PD-L1 anti-

body,^22,23was shown to inhibit the PD-1/PD-L1 interaction

with an EC₅₀value of 0.258 nM and a RLU_maxvalue of 2.395

(Table 4 and Figure 5A). L7 exhibited an EC₅₀value of 376

nM and was much potent than BMS-1016 (EC₅₀= 2075 nM)

and BMS-202 (EC₅₀> 10 000 nM) (Table 4 and Figure 5B).

The signiﬁcant diﬀerence in potency of L7 determined by the

HTRF assay (IC₅₀= 1.8 nM) and the cell-based assay (EC₅₀

=

376 nM) might be attributed to the diﬀerential capability for

L7 to induce PD-L1 dimerization under assay conditions. In

the HTRF assay, the PD-L1 protein was dissolved in the

buﬀer, which could be more easily dimerized upon contacting

with L7, while in the cell coculture assay, the extracellular

Figure 4. Binding aﬃnity of L7 to His-hPD-L1 (A) and Fc-hPD-1 (B) was evaluated using the SPR technique. His-hPD-L1 or Fc-hPD-1 was

captured on CM5 Chips and various concentrations of L7 ﬂowed through it. The K_Dvalues were determined by Biacore Evaluation software as

mean SEM from two to three independent experiments.

8397

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 5. Cell-based PD-1/PD-L1 signaling blockade assay. PD-L1 antibody atezolizumab (A) and small-molecule inhibitors (L7, BMS-1016,

BMS-202) (B) were evaluated for their cellular blockade activity on PD-1/PD-L1 interaction in PD-L1⁺/aAPC/CHO-K1 cells cocultured with PD-

1⁺/NFAT-luc/Jurkat cells for 6 h in the presence of varying concentrations of each inhibitor. The relative potency of the bioactivity assay was

expressed as the ratio of EC₅₀determined on the basis of the ﬁtting curve.

a

Table 5. Pharmacokinetic Parameters of L7 and L24

b

compound

species

L7

L24

SD rat

C57BL/6 mice

dosage

2 mg/kg (iv)

10 mg/kg (po)

9 mg/kg (iv)

45 mg/kg (po)

2 mg/kg (iv)

10 mg/kg (po)

t_1/2(h)

T_MAX(h)

C_MAX(ng/mL)

AUC_0−t(h·ng/mL)

CL (mL/h/kg)

F (%)

4.04 0.58

0.08

2087 583

860 233

2377 534

4.17 0.03

2.33 3.18

4.46 2.06

15.2 5.50

5.02 0.36

0.08

12971 577

6280 218

1426 48.5

3.43 0.53

4

488 79.7

2863 517

2.05 0.19

0.08

2821 308

1153 100

1722 144

1.55 0.26

1

150 27

444 71

0.35 0.13

9.12 1.65

7.70 1.24

a

b

The data were generated as mean SEM (n = 3). The detection form was L7. t_1/2, terminal elimination half-life; T_MAX, peak time; C_MAX, peak

concentration; AUC, area under drug time curve; CL, clearance; F, oral bioavailability.

and blood urea nitrogen (BUN) did not change signiﬁcantly

(Figures S4 −S6). H&E staining of major tissues demonstrated

no signiﬁcant toxic eﬀects (data not shown). To further

evaluate the potential cardiotoxicity of L7, the hERG assay was

conducted and showed that L7 had no inhibitory eﬀect on the

hERG potassium channel (IC₅₀> 30 μM). Together, L24

together with L7 had very good safety proﬁles.

at low dosage (50 mg/kg) (TGI = 50.93%) and medium

dosage (150 mg/kg) (TGI = 58.17%) but not at high dosage

(300 mg/kg) (TGI = 36.47%). On day 18, all three dosages

exhibited signiﬁcant antitumor eﬃcacy, although the high-

dosage regimen (TGI = 48.88%) was still less eﬀective than the

medium-dosage regimen (TGI = 55.09%). In fact, we have

repeated this kind of in vivo antitumor experiments many

times, and the phenomenon that higher-dosage regimens

might not end up with better antitumor eﬃcacy did exist.

Notably, the results from phase II clinical trials on PD-L1/

VISTA dual inhibitor CA-170 indicated that a higher incidence

of clinical beneﬁts and irAEs was observed at a lower dosage

(400 mg) instead of a higher dosage (800 mg) (https://www.

curis.com/pipeline/ca-170/). These ﬁndings suggested that

higher-dosage regimens of small-molecule immunostimulatory

agents might not result in more eﬀective antitumor immunity.

Further studies on the dose−eﬃcacy relationship and the

underlying mechanisms are ongoing in our laboratory.

Antitumor Eﬃcacy in the PD-L1 Humanized Tumor

Mouse Model. Our ﬁndings demonstrated that L24 exhibited

signiﬁcant antitumor eﬀects via its active metabolite L7 in

syngeneic mouse tumor models. Although other studies

conﬁrmed the antitumor eﬀect of BMS compound analogues

in syngeneic mouse models,¹⁶we thought that such syngeneic

models might be insuﬃcient to evaluate the in vivo antitumor

activity of L24, considering that human PD-L1 (hPD-L1) has

77% sequence homology with mouse PD-L1 (mPD-L1).²⁴In

fact, we found that L7 had no binding to mPD-L1 in the SPR

assay (Figure S7). Therefore, we hoped to evaluate the

antitumor activity of L24 in PD-L1-humanized tumor models

to recapitulate human biology.²⁵PD-L1-humanized MC38

cells (MC38-hPD-L1) and PD-L1-humanized mice were

Antitumor Eﬃcacy in Tumor Models of Syngeneic

Mice. Based on the promising results of the above studies on

the selected compounds, we evaluated the in vivo antitumor

eﬀects of L24 in B16-F10 melanoma and MC38 colon cancer

models of C57BL/6J mice. In the B16-F10 melanoma model,

tumor implantation and treatment started on the same day

(day 1). L24 (15 mg/kg, 45 mg/kg) and cyclophosphamide

(CTX, 80 mg/kg) were administered via oral gavage once a

day for 18 days. As shown in Figure 6, the tumor volume was

signiﬁcantly reduced from day 11. After 18 days of treatment,

the average tumor growth inhibition (TGI) rates of the L24-

treated groups were 45.77% (15 mg/kg group) and 70.70%

(45 mg/kg group) (Figure 6A). The average tumor weights of

the L24-treated groups were 1.27 g (15 mg/kg group) and

0.79 g (45 mg/kg group), compared to 1.76 g of the vehicle

control (Figure 6B). The CTX-treated group (positive

control) showed much better tumor growth inhibition than

the L24-treated groups, but CTX caused severe weight loss in

mice due to its toxicity (Figure 6C). In contrast, L24-treated

mice exhibited stable body weights and spleen weights (Figure

6C,D). Together, L24 exhibited a signiﬁcant preventive

antitumor eﬀect against melanoma tumor growth.

In the MC38 colon cancer model, after the implanted tumor

reached 50 mm³, the treatment started (day 1). As shown in

Figure 7, on day 14, L24 showed a signiﬁcant antitumor eﬀect

8398

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 6. L24 treatment inhibited tumor growth in the B16-F10 melanoma mouse model. B16-F10 melanoma tumors were established by injecting

1 × 10⁶B16-F10 cells into the dorsal area of mice. Treatment with CMC-Na (vehicle), L24, or CTX started on the same day when the tumor cells

were injected (day 1). On day 18, the mice were sacriﬁced and the tumors were isolated. The tumor volume curve (A), tumor weight (B), body

weight (C), and spleen weight (D) of the tumor-bearing mice are presented. For tumor volume and body weight, statistical analysis was performed

by two-way analysis of variance (ANOVA) multiple comparisons. For tumor weight and spleen weight, statistical analysis was performed by one-

way ANOVA multiple comparisons. All quantitative data were represented as means SEM (n = 8−10); *p < 0.1, **p < 0.01, and ***p < 0.001 vs

vehicle group.

Figure 7. L24 treatment inhibited tumor growth in the MC38 colon cancer mouse model. C57BL/6 mice were injected with 1 × 10⁶MC38 cells

into the dorsal area. When the tumor volume reached 50 mm³, the mice bearing tumors were randomly divided into four groups and treated with

CMC-Na (vehicle), L24 (50 mg/kg), L24 (150 mg/kg), and L24 (300 mg/kg) by oral gavage daily. After 18 days of treatment, the mice were

sacriﬁced and the tumors were isolated. Tumor volume curve (A) and body weight (B) of the tumor-bearing mice are presented. Statistical analysis

was performed by two-way ANOVA multiple comparisons. All quantitative data were represented as means SEM (n = 8−10); *p < 0.1, **p <

0.01, and ***p < 0.001 vs vehicle group.

constructed by replacing mouse PD-L1 extracellular segment

with the same part of human PD-L1. Tumor implantation was

carried out by injection of MC38-hPD-L1 cells into PD-L1

humanized C57BL/6 mice. As shown in Figure 8, L24

exhibited signiﬁcant antitumor eﬃcacy in this model with TGIs

of 44.2% at a dose of 25 mg/kg and 30.8% at a dose of 50 mg/

kg. Interestingly, two tumor volume curves crossed during the

period of treatment. High dosage (50 mg/kg) showed a

stronger antitumor eﬀect in the early stage of administration,

while after day 23, low dosage (25 mg/kg) exhibited a better

eﬀect. This phenomenon suggests that the dosage of L24

might be a key factor for its antitumor eﬀect.

To further investigate the optimal dosage, a lower dosage

was used in the PD-L1 humanized tumor mouse model. As

shown in Figure 9, on the last day of treatment, L24 exhibited

signiﬁcant antitumor eﬃcacy at a dose of 25 mg/kg (TGI =

8399

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 8. L24 inhibited tumor growth in the PD-L1 humanized tumor mouse model. PD-L1 humanized C57BL/6 mice were injected with 1 × 10⁶

MC38-hPD-L1(Tg)-mPDL1(KO) cells into the right ﬂank area. When the tumor volume reached 50 mm³, the mice bearing tumors were randomly

divided into three groups and treated with CMC-Na (vehicle) and L24 (25 and 50 mg/kg) by oral gavage daily. After 30 days of treatment, the

mice were sacriﬁced and the tumors were isolated. Tumor volume curve (A) and tumor weight (B) are presented. Statistical analysis was performed

by two-way ANOVA multiple comparisons. All quantitative data were represented as means SEM (n = 6); *p < 0.05 and ***p < 0.001 vs vehicle

control group.

Figure 9. L24 treatment inhibited tumor growth in PD-L1 humanized tumor mouse model. PD-L1 humanized C57BL/6 mice were injected with 1

× 10⁶MC38-hPD-L1(Tg)-mPDL1(KO) cells into the right ﬂank area. When the tumor volume reached 50 mm³, the mice bearing tumors were

randomly divided into three groups and treated with CMC-Na (vehicle) and L24 (5 and 25 mg/kg) by oral gavage daily. After 27 days of

treatment, the mice were sacriﬁced and the tumors were isolated. Tumor volume curve (A) and tumor weight (B) are presented. Statistical analysis

was performed by two-way ANOVA multiple comparisons. All quantitative data were represented as means SEM (n = 8); *p < 0.05 and ***p <

0.001 vs vehicle control group.

32.5%) but not at a dose of 5 mg/kg (TGI = 19.3%). Of note,

in the 25 mg/kg dosage group, one mouse exhibited complete

tumor regression.

ment.^28,29We found that L24 (25 mg/kg) was able to reduce

the proportions of mMDSCs (monocytic MDSCs), gMDSCs

(granulocytic MDSCs), and macrophages (CD11b⁺F4/80⁺)

(Figure 10). However, no signiﬁcant changes in regulatory T

(Treg) cells were observed (Figure 10). Together, L24

suppressed tumor growth probably through modulating the

proportions of tumor-inﬁltrating lymphocytes (TILs) with

upregulated cytotoxic T and NK cells and downregulated

immunosuppressive cells.

To investigate the mechanism of the antitumor eﬀect of L24,

cell counts for T-cell subsets, myeloid-derived suppressor cell

(MDSC) subsets, macrophages (MΦs), and natural killer

(NK) cells in tumor tissues were determined by ﬂow

cytometry. As shown in Figure 10, L24 treatment altered the

phenotype of tumor-inﬁltrating immune cells. The CD3⁺T

cells were increased in the 25 mg/kg dosage group compared

to the vehicle and 5 mg/kg dosage groups. Among CD3⁺T

cells in the 25 mg/kg dosage group, the ratio of CD8⁺T cells

to CD4⁺T cells increased. (Figure 11). NK cells play an

important role in killing tumor cells as a result of the antitumor

immune response,^26,27and it was found that the percentage of

NK cells increased in the 25 mg/kg dosage group (Figure 10).

On the other hand, immunosuppressive cells, such as MDSC

subsets and tumor-associated macrophages (TAMs), are

critical players within the immunosuppressive microenviron-

CONCLUSIONS

■

In this study, using BMS-1016 as a lead compound, a series of

novel benzoxadiazole derivatives have been designed, synthe-

sized, and biologically evaluated as PD-L1 inhibitors. SAR

analysis and docking studies suggested that the benzoxadiazole

unit and the serine side chain played key roles in binding with

PD-L1. Among the selected compounds, L7 (HTRF assay:

IC₅₀= 1.8 nM; cell coculture assay: EC₅₀= 375 nM) exhibited

much more potent PD-1/PD-L1 blockade activity than BMS-

8400

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 10. Flow cytometry analysis of the lymphocyte proﬁle within the tumor. Plots showing abundance of CD3⁺T cells (A), CD4⁺T cells (B),

CD8⁺T cells (C), Treg cells (D), gMDSCs (E), mMDSCs (F), macrophages (G), and NK cells (h). Statistical analysis was performed by one-way

ANOVA multiple comparisons. All quantitative data were represented as means SEM (n = 7−8); *p < 0.05 and **p < 0.01. mMDSC, monocytic

MDSC; gMDSC, granulocytic MDSC.

Figure 11. (A) Flow cytometry analysis of CD4⁺T cells and CD8⁺T cells within tumor tissues of representative individuals from three groups

(vehicle control, L24 5 mg/kg and L24 25 mg/kg). (B) Ratio of CD8⁺T cells to CD4⁺T cells. The data were represented as means SEM (n =

7−8).

DMSO-d₆at 2.50 ppm). Proton coupling patterns were described as

follows: singlet (s), doublet (d), doublet of doublets (dd), triplet (t),

triplet of doublets (td), quartet (q), multiplet (m), and broad (br).

¹³C NMR chemical shifts are reported in ppm from tetramethylsilane

(TMS) with the solvent resonance as the internal standard (CDCl₃at

77.20 ppm, DMSO-d₆at 39.51 ppm). The purity of all target

compounds was more than 95% as conﬁrmed by a Waters ACQUITY

UPLC with a BEH C18 column (2.1 × 50 mm², 1.7 μm). The elution

condition was a linear gradient from 0 to 100% B over 2 min and then

100% B for 3 min (A = H₂O, B = MeOH with 0.1% AcOH, ﬂow rate

0.3 mL/min).

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-methyl-[1,1′-bi-

phenyl]-3-yl)methoxy)benzyl)-^D-serine (L1). Compound 11a (see

the Supporting Information, SI) (292 mg, 0.65 mmol) was added into

a solution of ^D-serine ethyl ester hydrochloride (330 mg, 1.95 mmol),

ethyldiisopropylamine (DIPEA, 323 μL, 1.95 mmol), and one drop of

glacial acetic acid in anhydrous DCM (4 mL). The solution was

stirred at room temperature for 2 h. Sodium triacetoxyborohydride

(612 mg, 2.76 mmol) was then added into the reaction mixture and

stirred at room temperature for approximately 20 h. The reaction

mixture was concentrated under reduced pressure, and saturated

NaHCO₃(10 mL) was added, which was then extracted with DCM

(3 × 20 mL). The combined organic layer was washed with water (3

× 30 mL) and brine (20 mL), dried over Na₂SO₄, ﬁltered, and

concentrated. The resulting residue was puriﬁed by ﬂash chromatog-

raphy (DCM/MeOH = 50:1) to get a yellow solid (150 mg). The

1016 (HTRF assay: IC₅₀= 35.8 nM; cell coculture assay: EC₅₀

= 2075 nM) at protein and cellular levels. Binding aﬃnity assay

results showed that L7 had strong binding to hPD-L1 (SPR

assay: K_D= 3.34 nM) without binding to hPD-1. PK studies

demonstrated that L7 was orally bioavailable in the form of its

ester prodrug L24. Importantly, L24 exhibited signiﬁcant

antitumor eﬀects in tumor models of syngeneic and PD-L1-

humanized mice. Preliminary mechanistic studies indicated

that promotion of the antitumor immune response might

contribute to the antitumor eﬀects of L24. Given the good

safety proﬁles of L24 and L7, this class of benzoxadiazole PD-

L1 inhibitors holds promise for small-molecule-based tumor

immunotherapy.

EXPERIMENTAL SECTION

■

General Chemistry. Reagents and solvents were purchased from

commercial suppliers and used as received without further

puriﬁcation. All of the reactions were monitored by analytical thin-

layer chromatography (TLC) using silica gel 60 F254 plates (Qingdao

Ocean Chemical Company, China). Silica gel column chromatog-

raphy was performed using silica gel (200−300 mesh, Qingdao Ocean

1

Chemical Company, China). H NMR spectra were recorded on an

ACF* 300Q Bruker spectrometer. ¹H NMR chemical shifts are

reported in ppm relative to tetramethylsilane (TMS) with the solvent

resonance employed as the internal standard (CDCl₃at 7.26 ppm,

8401

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

yellow solid was dissolved in MeOH (5 mL), and lithium hydroxide

(22 mg, 0.52 mmol) was added to react at room temperature for 6 h.

The mixture was concentrated under reduced pressure, and water (3

mL) was added. The pH value was adjusted to 7 by 1 N HCl. The

resulted precipitate was ﬁltered, washed with water, and dried to yield

Hz, 1H), 6.60 (dd, J = 8.5, 2.1 Hz, 1H), 5.29 (s, 2H), 5.26 (s, 2H),

3.74 (d, J = 3.1 Hz, 2H), 3.48 (dd, J = 13.9, 5.6 Hz, 1H), 3.35 (dd, J =

13.9, 7.1 Hz, 2H), 2.76 (dd, J = 7.8, 5.6 Hz, 1H). ¹³C NMR (126

MHz, DMSO-d₆) δ 174.83, 174.49, 158.98, 157.23, 149.33, 142.78,

142.45, 136.53, 133.07, 132.71, 132.12, 131.18, 130.30, 128.68,

128.36, 127.95, 125.48 (q, J = 28.5 Hz), 125.05 (d, J = 276.7 Hz),

120.66, 116.73, 113.01, 106.23, 100.99, 69.00, 68.17, 62.80, 46.49,

23.76. HRMS (ESI): exact mass calculated for C₃₁H₂₇F₃N₃O₆[M +

H]⁺594.1846, found 594.1842. Purity, 96.18%.

1

a white solid (132 mg, 13% for two steps) H NMR (300 MHz,

DMSO-d₆) δ 8.17 (s, 1H), 8.07 (d, J = 9.3 Hz, 1H), 7.73 (dd, J = 9.3,

1.3 Hz, 1H), 7.49−7.41 (m, 3H), 7.40−7.33 (m, 2H), 7.32−7.27 (m,

2H), 7.27−7.14 (m, 2H), 6.87 (d, J = 2.3 Hz, 1H), 6.73 (dd, J = 8.4,

2.3 Hz, 1H), 5.33 (s, 2H), 5.15 (s, 2H), 4.10 (d, J = 5.3 Hz, 2H), 3.75

(dd, J = 11.2, 4.6 Hz, 1H), 3.64 (dd, J = 11.2, 6.6 Hz, 1H), 3.20 (dd, J

= 6.6, 4.6 Hz, 1H), 2.17 (s, 3H). ¹³C NMR (126 MHz, DMSO-d₆) δ

142.65, 142.46, 141.81, 134.34, 133.10, 132.47, 130.10, 129.62,

128.71, 128.41, 127.43, 127.43, 125.95, 116.73, 113.20, 106.76,

101.07, 69.18, 62.73, 61.01, 45.76, 16.30. HRMS (ESI): exact mass

calculated for C₃₁H₃₀N₃O₆[M + H]⁺540.21291, found 540.21305.

Purity, 96.47%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-meth-

yl-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine (L6). L6 was pre-

pared starting with compound 14b (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

1

(19 mg, 21% for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.15

(s, 1H), 8.08 (d, J = 9.2 Hz, 1H), 7.73 (d, J = 9.4 Hz, 1H), 7.54−7.43

(m, 4H), 7.39 (d, J = 6.9 Hz, 1H), 7.31 (d, J = 7.3 Hz, 2H), 7.25−

7.15 (m, 2H), 7.14 (s, 1H), 5.40 (s, 2H), 5.27 (s, 2H), 3.93 (s, 2H),

3.59 (d, J = 5.8 Hz, 2H), 3.06 (s, 1H), 2.22 (s, 3H). ¹³C NMR (126

MHz, DMSO-d₆) δ 171.98, 156.13, 154.36, 149.35, 142.65, 142.32,

141.73, 135.47, 133.00, 131.47, 130.21, 129.63, 128.72, 128.26,

127.44, 125.94, 119.19, 116.81, 113.51, 113.29, 100.89, 70.12, 69.69,

62.91, 62.00, 45.37, 16.32. HRMS (ESI): exact mass calculated for

C₃₁H₂₉ClN₃O₆[M + H]⁺574.17394, found 574.17529. Purity,

98.19%.

(2-(Benzo[c][1,2,5]oxadiazol-4-ylmethoxy)-4-((2-methyl-[1,1′-bi-

phenyl]-3-yl)methoxy)benzyl)-^D-serine (L2). L2 was prepared

starting with compound 11b (see SI) according to the procedure

described for preparation of L1 to yield a white solid (48 mg, 26% for

1

two steps). H NMR (300 MHz, DMSO-d₆) δ 8.01 (d, J = 9.0 Hz,

1H), 7.79 (d, J = 6.5 Hz, 1H), 7.66−7.58 (m, 1H), 7.44 (d, J = 7.2

Hz, 3H), 7.41−7.34 (m, 2H), 7.31 (d, J = 7.8 Hz, 2H), 7.22 (dd, J =

18.0, 7.5 Hz, 2H), 6.96 (s, 1H), 6.73 (d, J = 7.9 Hz, 1H), 5.58 (s, 2H),

5.17 (s, 2H), 4.08 (s, 2H), 3.71 (dd, J = 5.6 Hz, 1H), 3.62 (dd, J = 5.6

Hz, 1H), 3.20 (t, J = 5.5 Hz, 1H), 2.18 (s, 3H). ¹³C NMR (126 MHz,

DMSO-d₆) δ 169.10, 160.78, 157.81, 149.55, 148.09, 142.66, 135.84,

134.38, 133.21, 132.80, 130.36, 130.13, 129.63, 128.71, 128.46,

127.44, 125.98, 115.92, 114.60, 107.09, 100.99, 69.21, 66.35, 63.05,

60.53, 45.55, 16.32. HRMS (ESI): exact mass calculated for

C₃₁H₃₀N₃O₆[M + H]⁺540.21291, found 540.21309. Purity, 95.77%.

(2-(Benzo[c][1,2,5]thiadiazol-5-ylmethoxy)-5-chloro-4-((2-meth-

yl-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine (L3). L3 was pre-

pared starting with compound 11c (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

(126 mg, 23% for two steps). ¹H NMR (300 MHz, DMSO-d₆) δ 8.23

(s, 1H), 8.11 (d, J = 9.0 Hz, 1H), 7.86 (dd, J = 9.1, 1.7 Hz, 1H),

7.50−7.36 (m, 5H), 7.34−7.27 (m, 2H), 7.26−7.15 (m, 2H), 6.91 (d,

J = 2.3 Hz, 1H), 6.75 (dd, J = 8.4, 2.3 Hz, 1H), 5.42 (s, 2H), 5.17 (s,

2H), 4.24 (s, 2H), 3.90 (dd, J = 12.0, 3.3 Hz, 1H), 3.81 (dd, J = 12.0,

4.3 Hz, 1H), 3.75 (d, J = 3.8 Hz, 1H), 2.17 (s, 3H). ¹³C NMR (126

MHz, DMSO-d₆) δ 169.06, 161.09, 158.20, 154.79, 142.66, 141.78,

139.51, 135.75, 133.64, 130.34, 130.12, 129.61, 128.71, 128.43,

127.43, 125.94, 121.75, 119.61, 106.85, 100.99, 69.53, 69.20, 60.90,

59.31, 44.22, 16.29. HRMS (ESI): exact mass calculated for

C₃₁H₃₀N₃O₅S [M + H]⁺556.19007, found 556.19012. Purity, 98.54%.

(2-(Benzofuran-6-ylmethoxy)-4-((2-methyl-[1,1′-biphenyl]-3-yl)-

methoxy)benzyl)-^D-serine (L4). L4 was prepared starting with

compound 11d (see the SI) according to the procedure described

for the preparation of L1 to yield a white solid (20 mg, 23% for two

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-bi-

phenyl]-3-yl)methoxy)-5-chlorobenzyl)-^D-serine (L7). L7 was pre-

pared starting with compound 14c (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

1

(36 mg, 24% for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.15

(s, 1H), 8.07 (d, J = 9.3 Hz, 1H), 7.72 (d, J = 9.3 Hz, 1H), 7.64−7.59

(m, 1H), 7.53 (s, 1H), 7.52−7.42 (m, 4H), 7.41−7.30 (m, 3H), 7.09

(s, 1H), 5.40 (s, 2H), 5.30 (s, 2H), 3.99 (s, 2H), 3.75−3.55 (m, 2H),

3.17 (t, J = 5.5 Hz, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 170.33,

155.80, 154.01, 148.87, 148.59, 143.00, 141.77, 140.74, 136.25,

132.50, 131.45, 131.07, 129.22, 128.81, 128.16, 127.74, 127.70,

122.78, 117.68, 116.36, 113.12, 112.89, 100.38, 71.02, 69.29, 62.67,

60.98, 44.58. HRMS (ESI): exact mass calculated for

C₃₀H₂₆BrClN₃O₆[M + H]⁺640.06676, found 640.06584. Purity,

95.71%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine (L8). L8 was

prepared starting with compound 14d (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

1

(12 mg, 20% for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.15

(s, 1H), 8.07 (d, J = 9.1 Hz, 1H), 7.71 (d, J = 9.5 Hz, 1H), 7.64 (d, J =

6.2 Hz, 1H), 7.54 (s, 1H), 7.51−7.35 (m, 7H), 7.12 (s, 1H), 5.40 (s,

2H), 5.34 (s, 2H), 4.01 (s, 2H), 3.74−3.60 (m, 2H), 3.21 (t, J = 5.4

Hz, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 156.24, 149.35, 142.26,

135.12, 132.99, 131.70, 129.74, 129.21, 128.72, 128.27, 127.66,

118.51, 116.81, 113.33, 100.91, 69.74, 69.07, 62.95, 61.52, 45.04.

HRMS (ESI): exact mass calculated for C₃₀H₂₆Cl₂N₃O₆[M + H]⁺

594.11932, found 594.11915. Purity, 95.13%.

1

steps). H NMR (300 MHz, DMSO-d₆) δ 8.00 (d, J = 2.2 Hz, 1H),

7.83 (d, J = 1.7 Hz, 1H), 7.60 (d, J = 8.5 Hz, 1H), 7.51−7.34 (m,

6H), 7.33−7.28 (m, 2H), 7.27−7.17 (m, 2H), 6.92 (dd, J = 18.5, 2.2

Hz, 2H), 6.71 (dd, J = 8.3, 2.3 Hz, 1H), 5.27 (d, J = 2.6 Hz, 2H), 5.15

(s, 2H), 4.18−4.06 (m, 2H), 3.83 (dd, J = 11.6, 3.8 Hz, 1H), 3.73

(dd, J = 11.7, 5.1 Hz, 1H), 3.51 (d, J = 4.5 Hz, 1H), 2.18 (s, 3H). ¹³C

NMR (126 MHz, DMSO-d₆) δ 169.10, 160.87, 158.30, 154.46,

146.96, 142.66, 141.80, 135.82, 134.36, 132.92, 131.79, 130.13,

129.62, 128.71, 128.46, 127.77, 127.44, 125.98, 124.90, 121.31,

111.69, 107.27, 106.48, 101.08, 70.37, 69.14, 61.73, 59.86, 45.00,

16.30. HRMS (ESI): exact mass calculated for C₃₃H₃₂NO₆[M + H]⁺

538.2224, found 538.2225. Purity, 97.21%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-ﬂuoro-

[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine (L9). L9 was prepared

starting with compound 14e (see the SI) according to the procedure

described for the preparation of L1 to yield a white solid (23 mg, 21%

1

for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.14 (s, 1H), 8.07

(d, J = 9.3 Hz, 1H), 7.70 (d, J = 9.2 Hz, 1H), 7.59−7.40 (m, 8H),

7.28 (t, J = 7.6 Hz, 1H), 7.15 (s, 1H), 5.40 (s, 2H), 5.36 (s, 2H), 4.02

(s, 2H), 3.72−3.63 (m, 2H), 3.24 (t, 1H). ¹³C NMR (126 MHz,

DMSO-d₆) δ 157.53 (d, J = 248.9 Hz), 156.20, 154.39, 149.34,

142.22, 135.27, 132.93, 131.71, 131.41, 129.92, 129.32, 129.30,

129.12, 128.86 (d, J = 13.2 Hz), 128.45, 125.16, 124.50 (d, J = 15.5

Hz), 116.79, 113.59, 113.18, 100.90, 69.68, 65.54, 63.02, 61.56, 45.07.

HRMS (ESI): exact mass calculated for C₃₀H₂₆ClFN₃O₆[M + H]⁺

578.14887, found 578.14837. Purity, 99.12%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-bromo-4-((2-

bromo-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine (L10). L10

was prepared starting with compound 14f (see the SI) according to

the procedure described for preparation of L1 to yield a white solid

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-(triﬂuorometh-

yl)-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine (L5). L5 was pre-

pared starting with compound 14a (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

1

(12 mg, 20% for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.11

(s, 1H), 8.05 (d, J = 9.4 Hz, 1H), 7.74 (t, J = 7.6 Hz, 2H), 7.66 (t, J =

7.7 Hz, 1H), 7.48−7.36 (m, 3H), 7.36−7.20 (m, 4H), 6.77 (d, J = 2.3

8402

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(28 mg, 26% for two steps). 1H NMR (300 MHz, DMSO-d₆) δ 8.16

(s, 1H), 8.07 (d, J = 9.3 Hz, 1H), 7.73 (d, J = 9.5 Hz, 1H), 7.70−7.62

(m, 2H), 7.54−7.41 (m, 4H), 7.41−7.31 (m, 3H), 7.07 (s, 1H), 5.41

(s, 2H), 5.30 (s, 2H), 4.00 (s, 2H), 3.71−3.61 (m, 2H), 3.17 (d, J =

5.5 Hz, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 170.85, 156.93,

155.39, 149.36, 149.08, 143.45, 142.27, 141.22, 136.79, 134.67,

133.01, 131.47, 129.71, 129.14, 128.64, 128.22, 128.16, 123.09,

119.08, 116.85, 113.40, 102.31, 100.80, 71.50, 69.69, 63.04, 61.53,

45.04. HRMS (ESI): exact mass calculated for C₃₀H₂₆Br₂N₃O₆[M +

H]⁺684.01624, found 684.01634. Purity, 98.16%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-3′-ﬂuoro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine-

(L15). L15 was prepared starting with compound 18c (see the SI)

according to the procedure described for the preparation of L1 to

yield a white solid (33 mg, 24% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 8.14 (s, 1H), 8.06 (d, J = 9.3 Hz, 1H), 7.70 (d, J = 9.5

Hz, 1H), 7.65 (d, J = 7.0 Hz, 1H), 7.58−7.48 (m, 2H), 7.46−7.37 (m,

2H), 7.31−7.22 (m, 3H), 7.11 (s, 1H), 5.39 (s, 2H), 5.33 (s, 2H),

4.02 (s, 2H), 3.74−3.60 (m, 2H), 3.25 (t, J = 5.3 Hz, 1H). ¹³C NMR

(126 MHz, DMSO-d₆) δ 170.46, 162.26 (d, J = 243.6 Hz), 156.28,

154.47, 149.34, 149.05, 142.23, 141.39 (d, J = 7.6 Hz), 139.81,

135.23, 132.97, 131.89, 131.65, 130.99, 130.75 (d, J = 8.5 Hz),

129.61, 127.75, 126.08, 118.15, 116.80, 116.71 (d, J = 23.4 Hz),

115.19 (d, J = 21.2 Hz), 113.59, 113.33, 100.89, 69.75, 68.98, 63.01,

61.37, 44.95. HRMS (ESI): exact mass calculated for

C₃₀H₂₅Cl₂FN₃O₆[M + H]⁺612.10990 or 614.10695, found

614.10838. Purity, 98.97%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-3′-hydroxy-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine

(L16). L16 was prepared starting with compound 18d (see the SI)

according to the procedure described for the preparation of L1 to

yield a white solid (36 mg, 22% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 9.60 (s, 1H), 8.16 (s, 1H), 8.07 (d, J = 9.2 Hz, 1H), 7.71

(d, J = 9.3 Hz, 1H), 7.62 (d, J = 7.2 Hz, 1H), 7.54 (s, 1H), 7.45−7.31

(m, 2H), 7.26 (t, J = 7.9 Hz, 1H), 7.12 (s, 1H), 6.87−6.74 (m, 3H),

5.40 (s, 2H), 5.33 (s, 2H), 4.02 (s, 2H), 3.76−3.60 (m, 2H), 3.22 (t,

1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 170.33, 157.61, 156.34,

154.61, 149.36, 149.07, 142.19, 141.31, 140.48, 135.04, 132.99,

132.00, 131.54, 131.08, 129.75, 129.13, 127.56, 120.35, 117.75,

116.81, 116.68, 115.24, 113.62, 113.36, 100.89, 69.77, 69.11, 63.04,

61.26, 44.91. HRMS (ESI): exact mass calculated for C₃₀H₂₆Cl₂N₃O₇

[M + H]⁺610.11423, found 610.11387. Purity, 99.05%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-bi-

phenyl]-3-yl)methoxy)-5-ﬂuorobenzyl)-^D-serine (L11). L11 was

prepared starting with compound 14g (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

1

(18 mg, 21% for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.14

(s, 1H), 8.07 (d, J = 9.3 Hz, 1H), 7.70 (d, J = 9.3 Hz, 1H), 7.59 (d, J =

7.3 Hz, 1H), 7.51−7.40 (m, 4H), 7.41−7.31 (m, 4H), 7.12 (d, J = 7.1

Hz, 1H), 5.35 (s, 2H), 5.29 (s, 2H), 3.98 (s, 2H), 3.69−3.61 (m, 2H),

3.17 (t, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 170.74, 152.94,

149.34, 149.05, 146.57 (d, J = 11.6 Hz), 146.54 (d, J = 237.2 Hz),

143.54, 142.40, 141.24, 136.68, 133.01, 131.69, 129.75, 129.66,

128.61, 128.19, 128.15, 123.64, 118.30 (d, J = 20.2 Hz), 117.13,

116.77, 113.27, 102.15, 71.87, 70.00, 63.03, 61.44, 45.00. HRMS

(ESI): exact mass calculated for C₃₀H₂₆BrFN₃O₆[M + H]⁺

624.09631, found 624.09704. Purity, 96.70%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-chloro-[1,1′-bi-

phenyl]-3-yl)methoxy)-5-ﬂuorobenzyl)-^D-serine (L12). L12 was

prepared starting with compound 14h (see the SI) according to the

procedure described for the preparation of L1 to yield a white solid

1

(55 mg, 26% for two steps). H NMR (300 MHz, DMSO-d₆) δ 8.14

(s, 1H), 8.07 (d, J = 9.3 Hz, 1H), 7.70 (d, J = 9.3 Hz, 1H), 7.64−7.58

(m, 1H), 7.52−7.34 (m, 8H), 7.15 (d, J = 7.2 Hz, 1H), 5.35 (s, 2H),

5.33 (s, 2H), 4.01 (s, 2H), 3.73−3.61 (m, 2H), 3.21 (t, J = 5.3 Hz,

1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 170.43, 153.02, 149.36,

149.06, 147.49, 146.74, 146.64, 145.60, 142.39, δ 140.24 (d, J = 245.8

Hz), 135.05, 133.01, 131.89, 131.45, 129.74, 129.71, 128.72, 128.28,

127.66, 118.47, 118.31, 116.76, 113.26, 102.15, 70.00, 69.45, 63.00,

61.28, 44.91. HRMS (ESI): exact mass calculated for C₃₀H₂₆ClFN₃O₆

[M + H]⁺578.14887, found 578.14936. Purity, 98.19%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-4′-(methylsulfonyl)-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-

serine (L13). L13 was prepared starting with compound 18a (see the

SI) according to the procedure described for the preparation of L1 to

yield a white solid (30 mg, 20% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 8.15 (s, 1H), 8.06 (t, J = 8.0 Hz, 3H), 7.72 (d, J = 8.5

Hz, 4H), 7.54 (s, 1H), 7.52−7.39 (m, 2H), 7.12 (s, 1H), 5.41 (s, 2H),

5.36 (s, 2H), 4.01 (s, 2H), 3.67 (m, J = 11.0, 5.3 Hz, 2H), 3.18 (t, J =

5.3 Hz, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 172.96, 156.14,

154.05, 149.41, 149.12, 144.37, 142.42, 141.13, 139.75, 135.71,

132.85, 131.43, 131.15, 130.76, 130.71, 129.94, 127.76, 127.23,

127.18, 116.78, 114.71, 113.43, 102.42, 70.27, 70.21, 69.53, 69.47,

63.25, 62.49, 45.64, 44.18, 44.13. HRMS (ESI): exact mass calculated

for C₃₁H₂₈Cl₂N₃O₈S [M + H]⁺672.0969, found 672.0967. Purity,

98.87%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-4′-ﬂuoro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine

(L14). L14 was prepared starting with compound 18b (see the SI)

according to the procedure described for the preparation of L1 to

yield a white solid (26 mg, 19% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 8.14 (s, 1H), 8.07 (d, J = 9.2 Hz, 1H), 7.72 (d, J = 9.3

Hz, 1H), 7.64 (d, J = 6.7 Hz, 1H), 7.53 (s, 1H), 7.51−7.37 (m, 4H),

7.32 (m, J = 8.8 Hz, 2H), 7.11 (s, 1H), 5.40 (s, 2H), 5.33 (s, 2H),

3.98 (s, 2H), 3.73−3.58 (m, 2H), 3.16 (t, 1H). ¹³C NMR (126 MHz,

DMSO-d₆) δ 171.15, 162.25 (d, J = 244.4 Hz), 156.22, 154.34,

149.34, 149.05, 142.28, 140.10, 135.53, 135.20, 132.97, 131.93,

131.79 (d, J = 18.9 Hz), 131.16, 129.33, 127.69, 118.81, 116.81,

115.71, 115.54, 113.58, 113.31, 100.92, 69.73, 69.03, 63.01, 61.66,

45.13. HRMS (ESI): exact mass calculated for C₃₀H₂₅Cl₂FN₃O₆[M +

H]⁺612.10990, found 612.10784. Purity, 97.91%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-2′,3′-diﬂuoro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^D-serine

(L17). L17 was prepared starting with compound 18e (see the SI)

according to the procedure described for the preparation of L1 to

1

yield a white solid (8 mg, 19% for two steps). H NMR (300 MHz,

DMSO-d₆) δ 8.21 (s, 1H), 8.11 (d, J = 9.0 Hz, 1H), 7.91−7.82 (m,

1H), 7.77−7.68 (m, 1H), 7.61−7.49 (m, 2H), 7.49−7.41 (m, 2H),

7.40−7.30 (m, 1H), 7.22 (t, J = 6.8 Hz, 1H), 7.15 (s, 1H), 5.46 (s,

2H), 5.35 (s, 2H), 3.99 (s, 2H), 3.72−3.61 (m, 2H), 3.19 (t, 1H). ¹³C

NMR (126 MHz, DMSO-d₆) δ 172.94, 156.32, 154.94, 154.61,

153.99, 150.49 (dd, J = 246.3, 13.0 Hz), 147.60 (dd, J = 246.7, 12.3

Hz), 139.66, 135.52, 134.42, 132.16, 131.75, 131.02, 130.30, 130.11,

129.24 (d, J = 12.8 Hz), 127.64, 127.13, 125.24, 122.63, 121.78,

119.53, 117.81 (d, J = 17.1 Hz), 114.57, 102.52, 70.55, 69.44, 63.26,

62.48, 45.63. HRMS (ESI): exact mass calculated for

C₃₀H₂₄O₆N₃Cl₂F₂[M + H]⁺630.10047, found 630.09869. Purity,

95.70%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-3′-(2-hydroxyethoxy)-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-

D-serine (L18). L18 was prepared starting with compound 18f (see

the SI) according to the procedure described for the preparation of L1

1

to yield a white solid (29 mg, 33% for two steps). H NMR (300

MHz, DMSO-d₆) δ 8.16 (s, 1H), 8.08 (d, J = 9.3 Hz, 1H), 7.72 (d, J =

9.4 Hz, 1H), 7.68−7.60 (m, 2H), 7.48−7.32 (m, 3H), 7.14 (s, 1H),

7.04−6.84 (m, 3H), 5.41 (s, 2H), 5.36 (s, 2H), 4.16 (s, 2H), 4.04 (t, J

= 4.9 Hz, 2H), 3.89−3.76 (m, 2H), 3.73 (t, J = 4.8 Hz, 2H), 3.60 (t,

1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 169.63, 158.87, 156.65,

155.09, 149.34, 149.06, 142.08, 141.03, 140.50, 135.00, 132.99,

132.84, 131.66, 131.13, 129.82, 129.26, 127.60, 121.99, 116.76,

115.94, 115.50, 114.36, 113.57, 113.40, 100.71, 70.05, 69.83, 69.10,

62.15, 60.18, 60.01, 44.15. HRMS (ESI): exact mass calculated for

C₃₂H₃₀Cl₂N₃O₈[M + H]⁺654.1404, found 654.1402. Purity, 96.47%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-3′-(3-hydroxypropoxy)-[1,1′-biphenyl]-3-yl)methoxy)-

benzyl)-^D-serine (L19). L19 was prepared starting with compound

18g (see the SI) according to the procedure described for the

preparation of L1 to yield a white solid (47 mg, 26% for two steps).

¹H NMR (300 MHz, DMSO-d₆) δ 8.16 (s, 1H), 8.08 (d, J = 9.3 Hz,

1H), 7.72 (dd, J = 9.3 Hz, 1H), 7.64 (dd, 1H), 7.55 (s, 1H), 7.48−

8403

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

7.31 (m, 3H), 7.12 (s, 1H), 7.05−6.89 (m, 3H), 5.41 (s, 2H), 5.35 (s,

2H), 4.09 (t, J = 6.4 Hz, 2H), 4.03 (s, 2H), 3.80−3.61 (m, 2H), 3.57

(t, J = 6.2 Hz, 2H), 3.21 (t, 1H), 1.96−1.73 (m, 2H). ¹³C NMR (126

MHz, DMSO-d₆) δ 170.19, 158.88, 156.39, 154.71, 149.34, 149.05,

142.18, 141.04, 140.54, 135.07, 132.99, 132.19, 131.62, 131.08,

129.81, 129.20, 127.60, 121.93, 117.19, 116.79, 115.86, 114.36,

113.60, 113.38, 100.84, 69.80, 69.07, 65.16, 63.22, 61.10, 57.79, 44.88,

32.60. HRMS (ESI): exact mass calculated for C₃₃H₃₂Cl₂N₃O₈[M +

H]⁺668.1561, found 668.1559. Purity, 98.90%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-3′-ﬂuo-

ro-[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)-^D-serine (L20).

L20 was prepared starting with compound 18h (see the SI) according

to the procedure described for the preparation of L1 to yield a white

solid (17 mg, 21% for two steps). ¹H NMR (300 MHz, DMSO-d₆) δ

8.16 (s, 1H), 8.08 (d, J = 9.4 Hz, 1H), 7.72 (dd, J = 9.3 Hz, 1H), 7.64

(dd, J = 7.3 Hz, 1H), 7.56−7.45 (m, 3H), 7.37 (d, J = 7.6 Hz, 1H),

7.30−7.19 (m, 3H), 7.10 (s, 1H), 5.40 (s, 2H), 5.31 (s, 2H), 4.02 (s,

2H), 3.76−3.66 (m, 2H), 3.24 (t, 1H). ¹³C NMR (126 MHz, DMSO-

d₆) δ 170.81, 162.18 (d, J = 243.8 Hz), 156.24, 154.39, 149.35,

149.06, 143.40, 142.26, 142.15, 136.85, 132.99, 131.82, 131.47,

130.71, 129.68, 128.25, 126.06, 123.04, 118.51, 116.83, 116.68 (d, J =

22.4 Hz), 115.12 (d, J = 20.4 Hz), 113.58, 113.37, 100.89, 71.40,

69.74, 63.01, 61.50, 45.06. HRMS (ESI): exact mass calculated for

C₃₀H₂₅BrClFN₃O₆[M + H]⁺656.0594, found 656.0595. Purity,

96.43%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-3′-me-

thoxy-[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)-^D-serine

(L21). L21 was prepared starting with compound 18i (see the SI)

according to the procedure described for the preparation of L1 to

yield a white solid (17 mg, 27% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 8.16 (s, 1H), 8.07 (d, J = 9.3 Hz, 1H), 7.72 (d, J = 9.2

Hz, 1H), 7.61 (d, J = 7.4 Hz, 1H), 7.55 (s, 1H), 7.45 (t, J = 7.6 Hz,

1H), 7.42−7.32 (m, 2H), 7.09 (s, 1H), 6.99 (d, J = 8.4 Hz, 1H),

6.96−6.89 (m, 2H), 5.40 (s, 2H), 5.30 (s, 2H), 4.03 (s, 2H), 3.80 (s,

3H), 3.77−3.62 (m, 2H), 3.25 (t, 1H). ¹³C NMR (126 MHz, DMSO-

d₆) δ 170.54, 159.35, 156.24, 154.44, 149.36, 149.07, 143.33, 142.51,

142.25, 136.71, 133.00, 131.83, 131.47, 129.74, 129.31, 128.11,

123.20, 122.01, 118.36, 116.83, 115.42, 113.69, 113.57, 113.39,

100.88, 71.47, 69.74, 62.98, 61.45, 55.61, 44.98. HRMS (ESI): exact

mass calculated for C₃₁H₂₈BrClN₃O₇[M + H]⁺668.0794, found

668.0799. Purity, 97.96%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-3′-

chloro-[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)-^D-serine

(L22). L22 was prepared starting with compound 18j (see the SI)

according to the procedure described for the preparation of L1 to

yield a white solid (12 mg, 20% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 8.13 (s, 1H), 8.05 (d, J = 9.4 Hz, 1H), 7.72 (d, J = 9.3

Hz, 1H), 7.63 (d, J = 7.7 Hz, 1H), 7.53−7.46 (m, 4H), 7.44 (d, J =

3.0 Hz, 1H), 7.39−7.32 (m, 2H), 7.06 (s, 1H), 5.38 (s, 2H), 5.28 (s,

2H), 3.91 (s, 2H), 3.58 (d, J = 5.9 Hz, 2H), 3.04 (t, J = 5.9 Hz, 1H).

¹³C NMR (126 MHz, DMSO-d₆) δ 172.86, 156.16, 154.07, 149.42,

HRMS (ESI): exact mass calculated for C₃₃H₃₂BrClN₃O₈[M + H]⁺

712.10558 or 714.10354, found 714.10327. Purity, 98.22%.

Ethyl(2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-

[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)-^D-serinate hydro-

chloride (L24). A solution of 14c (120 mg, 0.22 mmol), ^D-serine

ethyl ester hydrochloride (150 mg, 0.88 mmol), and one drop of

glacial acetic acid in anhydrous DCM (4 mL) was stirred in an ice

bath for 2 h. Sodium triacetoxyborohydride (140 mg, 0.66 mmol) was

then added into the reaction mixture and stirred at room temperature

for approximately 20 h. The reaction mixture was concentrated under

reduced pressure, and the residue was taken up in saturated NaHCO₃

(10 mL), which was extracted with DCM (3 × 20 mL). The

combined organic layer was washed with water (3 × 30 mL) and

brine (20 mL), dried over Na₂SO₄, ﬁltered, and concentrated. The

resulting residue was puriﬁed by ﬂash chromatography (solvent A,

DCM; solvent B, MeOH; 0−40% B) to give a colorless oil. The oil

product was added into a hydrochloric acid ethanol solution and

precipitated as a white solid. The precipitate was ﬁltered, washed with

CH₃OH (1.5 mL), and dried to yield a white solid (32 mg, 21%). ¹H

NMR (300 MHz, DMSO-d₆) δ 9.45 (d, 2H), 8.19 (s, 1H), 8.11 (d, J

= 8.9 Hz, 1H), 7.78−7.67 (m, 2H), 7.63 (dd, J = 7.6, 1.8 Hz, 1H),

7.54−7.42 (m, 4H), 7.42−7.32 (m, 3H), 7.14 (s, 1H), 5.65 (s, 1H),

5.43 (s, 2H), 5.35 (s, 2H), 4.28 (s, 2H), 4.12 (m, J = 10.7, 7.1, 3.8 Hz,

3H), 4.01−3.84 (m, 2H), 1.17 (t, J = 7.1 Hz, 3H). ¹³C NMR (75

MHz, DMSO-d₆) δ 169.34, 158.72, 157.31, 143.82, 138.41, 135.40,

134.83, 133.26, 131.37, 131.06, 130.30, 129.89, 129.84, 118.45,

115.88, 115.73, 115.36, 102.83, 73.48, 71.88, 64.02, 63.13, 61.18,

45.44, 16.02. HRMS (ESI): exact mass calculated for

C₃₂H₃₀BrClN₃O₆[M + H]⁺668.09806, found 668.09807. Purity,

98.29%. Enatiomeric excess (ee), 99.64%, chiral high-performance

liquid chromatography (HPLC) analysis: CHIRALCEL ODH0CE-

SD069 column, 35 °C, (n-hexane/isopropanol/diethylamine =

60:40:0.1), 0.8 mL/min.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-bi-

phenyl]-3-yl)methoxy)-5-chlorobenzyl)-^L-serine (L25). L25 was

prepared starting with compound 14c and ^L-serine ethyl ester

hydrochloride according to the procedure described for preparation of

1

L1 to yield a white solid (19 mg, 23% for two steps). H NMR (300

MHz, DMSO-d₆) δ 8.15 (s, 1H), 8.05 (d, J = 9.0 Hz, 1H), 7.71 (d, J =

9.2 Hz, 1H), 7.60 (d, J = 5.2 Hz, 2H), 7.52−7.39 (m, 4H), 7.39−7.28

(m, 3H), 7.10 (s, 1H), 5.40 (s, 2H), 5.29 (s, 2H), 4.11 (s, 2H), 3.83−

3.68 (m, 2H), 3.38 (s, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ

172.93, 156.18, 154.13, 149.45, 149.17, 143.73, 142.44, 141.47,

137.11, 132.88, 131.37, 131.23, 129.65, 129.10, 128.52, 128.12,

128.05, 123.10, 122.50, 116.82, 114.76, 113.50, 102.43, 72.13, 70.33,

63.32, 62.50, 45.67. HRMS (ESI): exact mass calculated for

C₃₀H₂₆BrClN₃O₆[M + H]⁺638.06880 or 640.06676, found

640.06711. Purity, 95.53%.

(S)-1-(2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-

[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)piperidine-2-carbox-

ylic acid (L26). L26 was prepared starting with compound 14c and ^L-

pipecolate methyl ester according to the procedure described for

preparation of L1 to yield a white solid (12 mg, 30% for two steps).

¹H NMR (300 MHz, DMSO-d₆) δ 8.13 (s, 1H), 8.09 (d, J = 9.4 Hz,

1H), 7.64 (td, J = 8.3, 7.5, 1.5 Hz, 2H), 7.53−7.30 (m, 8H), 7.07 (s,

1H), 5.38 (s, 2H), 5.29 (s, 2H), 3.87−3.65 (m, 2H), 3.22 (dd, J = 7.5,

4.3 Hz, 1H), 2.94 (m, J = 11.8 Hz, 1H), 2.41−2.28 (m, 1H), 1.79 (m,

2H), 1.50 (m, 3H), 0.85 (m, J = 6.8 Hz, 1H). ¹³C NMR (126 MHz,

DMSO-d₆) δ 173.81, 156.18, 153.87, 149.33, 143.45, 142.54, 141.22,

136.81, 132.89, 131.71, 131.50, 129.68, 129.27, 128.62, 128.19,

128.16, 119.81, 116.84, 113.73, 113.20, 101.03, 71.48, 69.57, 64.19,

52.82, 49.61, 28.91, 24.74, 22.11. HRMS (ESI): exact mass calculated

for C₃₃H₃0BrClN₃O₅[M + H]⁺662.10519 or 664.10314, found

664.10296. Purity, 99.18%.

2-((2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-

biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)ethane-1-sulfonic

acid (L27). L27 was prepared starting with compound 14c and 2-

aminoethane-1-sulfonic acid according to the Borch reductive

amination procedure described for the preparation of L1 to yield a

white solid (28 mg, 21%). ¹H NMR (300 MHz, DMSO-d₆) δ 8.62 (s,

149.15, 143.34, 142.42, 142.17, 137.24, 133.37, 132.87, 131.29,

131.21, 130.42, 129.60, 129.47, 128.47, 128.19, 122.88, 122.51,

116.81, 114.73, 113.49, 102.42, 72.01, 70.30, 63.28, 62.46, 45.64.

HRMS (ESI): exact mass calculated for C₃₀H₂₄BrCl₂N₃NaO₆[M +

Na]⁺694.01177 or 696.00973, found 696.00934. Purity, 97.99%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-3′-(3-hy-

droxypropoxy)-[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)-^D-

serine (L23). L23 was prepared starting with compound 18k (see the

SI) according to the procedure described for the preparation of L1 to

yield a white solid (26 mg, 17% for two steps). ¹H NMR (300 MHz,

DMSO-d₆) δ 8.15 (s, 1H), 8.07 (d, 1H), 7.72 (d, 1H), 7.65−7.51 (m,

2H), 7.50−7.28 (m, 3H), 7.09 (s, 1H), 7.05−6.82 (m, 3H), 5.39 (s,

2H), 5.30 (s, 2H), 4.07 (t, 2H), 4.01 (s, 2H), 3.78−3.66 (m, 2H),

3.64−3.52 (m, 2H), 3.22 (t, 1H), 1.97−1.78 (m, 2H). ¹³C NMR (75

MHz, DMSO-d₆) δ 171.38, 158.84, 156.17, 154.28, 149.35, 149.07,

143.39, 142.54, 142.28, 136.77, 132.95, 131.39, 129.68, 129.19,

128.07, 123.09, 121.89, 119.57, 116.81, 115.91, 114.36, 113.83,

113.40, 101.21, 71.60, 69.83, 65.27, 63.06, 61.78, 57.84, 45.19, 32.64.

8404

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

1H), 8.16 (s, 1H), 8.08 (d, J = 9.3 Hz, 1H), 7.74 (d, J = 9.4 Hz, 1H),

7.66−7.55 (m, 2H), 7.52−7.28 (m, 7H), 7.13 (s, 1H), 5.44 (s, 2H),

5.32 (s, 2H), 4.23 (s, 2H), 3.21 (t, J = 6.5 Hz, 2H), 2.82 (t, J = 6.4 Hz,

2H). ¹³C NMR (126 MHz, DMSO-d₆) δ 156.62, 149.36, 149.07,

142.04, 141.17, 136.51, 133.01, 132.95, 131.62, 129.67, 129.37,

128.63, 128.22, 128.16, 123.36, 116.91, 114.17, 113.57, 113.51,

100.74, 71.54, 69.79, 46.98, 45.00, 44.09. HRMS (ESI): exact mass

calculated for C₂₉H₂₆BrClN₃O₆S [M + H]⁺658.04087 or 660.03883,

found 660.03822. Purity, 96.13%.

(2R,4R)-1-(2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-

bromo-[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)-4-hydroxy-

pyrrolidine-2-carboxylic acid (L28). L28 was prepared starting with

compound 14c and (2R,4R)-methyl 4-hydroxypyrrolidine-2-carbox-

ylate hydrochloride according to the procedure described for the

preparation of L1 to yield a white solid (20 mg, 30% for two steps).

¹H NMR (300 MHz, DMSO-d₆) δ 8.17 (s, 1H), 8.09 (d, J = 9.3 Hz,

1H), 7.72 (d, J = 9.3 Hz, 1H), 7.67−7.61 (m, 2H), 7.53−7.42 (m,

4H), 7.41−7.32 (m, 3H), 7.15 (s, 1H), 5.44 (s, 2H), 5.33 (s, 2H),

4.30 (d, J = 6.7 Hz, 2H), 4.14 (d, J = 13.2 Hz, 1H), 3.93 (s, 1H), 3.25

(d, J = 10.7 Hz, 1H), 3.10−3.02 (m, 1H), 1.99−1.89 (m, 1H), 0.88−

0.81 (m, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 171.78, 156.73,

155.21, 149.38, 149.09, 143.51, 142.06, 141.21, 136.62, 133.34,

133.07, 131.63, 129.69, 129.47, 128.64, 128.21, 123.39, 116.92,

113.73, 113.61, 100.84, 71.60, 69.93, 68.46, 65.22, 61.74, 51.93, 38.26.

HRMS (ESI): exact mass calculated for C₃₂H₂₈BrClN₃O₆[M + H]⁺

664.08445 or 666.08241, found 666.08260. Purity, 96.12%.

2.29 (s, 3H). HRMS (ESI): exact mass calculated for

C₃₂H₃₁BrClN₄O₃[M + H]⁺633.1263, found 633.1263. Purity,

97.26%.

(R)-3-((2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-

[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)propane-1,2-

diol (L33). L33 was prepared starting with compound 14c and (R)-3-

amino-1,2-propanediol according to the Borch reductive amination

procedure described for the preparation of L1 to yield a white solid

1

(22 mg, 29%). H NMR (300 MHz, DMSO-d₆) δ 8.15−8.04 (m,

2H), 7.72−7.59 (m, 2H), 7.52−7.42 (m, 5H), 7.41−7.31 (m, 3H),

7.07 (s, 1H), 5.39 (s, 2H), 5.30 (s, 2H), 4.66 (s, 2H), 3.80 (s, 3H),

3.59 (s, 2H), 2.68 (dd, J = 8.0 Hz, 1H), 2.53 (dd, 1H). HRMS (ESI):

exact mass calculated for C₃₀H₂₈BrClN₃O₅[M + H]⁺624.0895, found

624.0884. Purity, 96.27%.

(S)-3-((2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-

[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)propane-1,2-

diol (L34). L34 was prepared starting with compound 14c and (S)-3-

amino-1,2-propanediol according to the Borch reductive amination

procedure described for the preparation of L1 to yield a white solid

1

(11 mg, 27%). H NMR (300 MHz, DMSO-d₆) δ 8.09 (m, J = 8.9

Hz, 2H), 7.68 (d, J = 9.5 Hz, 1H), 7.63 (d, J = 7.6 Hz, 1H), 7.44 (d, J

= 13.5 Hz, 5H), 7.41−7.29 (m, 3H), 7.08 (s, 1H), 5.39 (s, 2H), 5.30

(s, 2H), 3.84 (s, 2H), 3.62 (m, 3H), 2.72 (dd, J = 11.7 Hz, 1H), 2.54

(dd, 1H). ¹³C NMR (126 MHz, DMSO-d₆) δ 156.78, 155.36, 149.35,

149.07, 143.51, 142.15, 141.21, 136.60, 133.17, 131.59, 129.66,

129.43, 128.61, 128.19, 128.16, 123.38, 116.84, 114.20, 113.65,

113.57, 100.78, 71.59, 69.89, 67.63, 63.89, 50.27, 44.65. HRMS

(ESI): exact mass calculated for C₃₀H₂₈BrClN₃O₅[M + H]⁺

624.0895, found 624.0902. Purity, 95.76%.

4-((2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-

biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)cyclohexan-1-ol

hydrochloride (L35). L35 was prepared starting with compound 14c

and 4-aminocyclohexanol according to the Borch reductive amination

procedure described for the preparation of L1 to give a colorless oil.

The oil product was added to a hydrochloric acid ethanol solution and

precipitated as a white solid. The precipitate was ﬁltered, washed with

CH₃OH (1.5 mL), and dried to yield a white solid (29 mg, 30%). ¹H

NMR (300 MHz, DMSO-d₆) δ 8.86 (s, 2H), 8.19 (s, 1H), 8.13 (d, J =

9.3 Hz, 1H), 7.76 (d, J = 9.1 Hz, 1H), 7.69−7.60 (m, 2H), 7.47 (dt, J

= 10.9, 7.0 Hz, 4H), 7.41−7.31 (m, 3H), 7.16 (s, 1H), 5.41 (s, 2H),

5.36 (s, 2H), 4.13 (s, 2H), 3.49−3.43 (m, 1H), 3.03−2.89 (m, 1H),

2.05 (d, J = 11.4 Hz, 2H), 1.78 (d, J = 10.0 Hz, 2H), 1.38 (dd, J =

23.2, 9.7 Hz, 2H), 1.19−1.02 (m, 2H). ¹³C NMR (126 MHz, DMSO-

d₆) δ 156.77, 155.41, 149.33, 149.07, 143.55, 142.04, 141.19, 136.60,

133.34, 133.19, 131.61, 129.66, 129.38, 128.62, 128.21, 128.18,

123.37, 116.79, 114.62, 113.94, 113.49, 100.65, 71.58, 69.94, 68.02,

56.15, 42.01, 33.37, 27.06. HRMS (ESI): exact mass calculated for

C₃₃H₃₂BrClN₃O₄[M + H]⁺648.1259, found 648.1261. Purity,

97.78%.

N-(2-((2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-

[1,1′-biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)ethyl)-

acetamide hydrochloride (L36). L36 was prepared starting with

compound 14c and N-(2-aminoethyl)acetamide according to the

Borch reductive amination procedure described for the preparation of

L1 to give a colorless oil. The oil product was added to a hydrochloric

acid ethanol solution and precipitated as a white solid. The precipitate

was ﬁltered, washed with CH₃OH (1.5 mL), and dried to yield a

white solid (26 mg, 28%). ¹H NMR (300 MHz, DMSO-d₆) δ 9.00 (s,

2H), 8.18 (s, 2H), 8.11 (d, J = 9.4 Hz, 1H), 7.74 (d, J = 9.4 Hz, 1H),

7.67 (s, 1H), 7.62 (d, J = 7.5 Hz, 1H), 7.52−7.41 (m, 4H), 7.40−7.31

(m, 3H), 7.15 (s, 1H), 5.45 (s, 2H), 5.34 (s, 2H), 4.20 (s, 2H), 3.38−

3.33 (m, 2H), 3.07−2.94 (m, 2H), 1.81 (s, 3H). ¹³C NMR (126

MHz, DMSO-d₆) δ 170.56, 156.64, 155.40, 149.37, 149.09, 143.53,

142.17, 141.20, 136.57, 133.10, 133.02, 131.64, 129.69, 129.43,

128.65, 128.23, 128.19, 123.41, 116.88, 114.20, 113.57, 100.74, 71.57,

69.85, 46.75, 44.33, 35.63, 23.04. HRMS (ESI): exact mass calculated

for C₃₁H₂₉BrClN₄O₄[M + H]⁺635.1055, found 635.1062. Purity,

98.09%.

2-((2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-

biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)ethan-1-ol (L29).

L29 was prepared starting with compound 14c and ethanolamine

according to the Borch reductive amination procedure described for

1

the preparation of L1 to yield a white solid (51 mg, 25%). H NMR

(300 MHz, DMSO-d₆) δ 8.27−7.95 (m, 2H), 7.65 (d, J = 9.7 Hz,

1H), 7.61 (d, J = 7.2 Hz, 1H), 7.53−7.25 (m, 8H), 7.04 (s, 1H), 5.35

(s, 2H), 5.27 (s, 2H), 3.72 (s, 2H), 3.46 (t, J = 5.7 Hz, 2H), 2.57 (t, J

= 5.7 Hz, 2H). ¹³C NMR (126 MHz, DMSO-d₆) δ 173.58, 155.67,

153.30, 149.32, 149.06, 143.44, 142.63, 141.23, 136.93, 132.86,

131.44, 130.44, 129.69, 129.15, 128.62, 128.19, 128.14, 123.37,

123.14, 116.90, 113.61, 113.09, 101.06, 71.46, 69.49, 60.75, 51.45,

47.01. HRMS (ESI): exact mass calculated for C₂₉H₂₆BrClN₃O₄[M +

H]⁺594.0790, found 594.0790. Purity, 97.95%.

1-(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-

biphenyl]-3-yl)methoxy)-5-chlorobenzyl)azetidine-3-carboxylic

acid (L30). L30 was prepared starting with compound 14c and 3-

azetidinecarboxylic acid according to the Borch reductive amination

procedure described for the preparation of L1 to yield a white solid (8

mg, 19%). ¹H NMR (300 MHz, DMSO-d₆) δ 8.13 (s, 1H), 8.10 (d, J

= 9.3 Hz, 1H), 7.71 (d, J = 9.3 Hz, 1H), 7.62 (d, J = 7.7 Hz, 1H),

7.52−7.41 (m, 4H), 7.41−7.31 (m, 4H), 7.09 (s, 1H), 5.40 (s, 2H),

5.30 (s, 2H), 3.79 (s, 2H), 3.63 (m, 2H), 3.49 (m, 2H), 3.34−3.27

(m, 1H). HRMS (ESI): exact mass calculated for C₃₁H₂₆BrClN₃O₅

[M + H]⁺634.07389 or 636.07184, found 636.07083. Purity, 99.00%.

1-(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-

biphenyl]-3-yl)methoxy)-5-chlorobenzyl)piperidin-4-ol (L31). L31

was prepared starting with compound 14c and 4-hydroxypiperidine

according to the Borch reductive amination procedure described for

1

the preparation of L1 to yield a white solid (9 mg, 20%). H NMR

(300 MHz, DMSO-d₆) δ 8.13 (s, 1H), 8.09 (d, J = 9.3 Hz, 1H),

7.70−7.61 (m, 2H), 7.52−7.43 (m, 4H), 7.42−7.32 (m, 4H), 7.09 (s,

1H), 5.39 (s, 2H), 5.30 (s, 2H), 4.59 (s, 1H), 3.62−3.49 (m, 3H),

2.76 (t, 2H), 2.18 (t, 1H), 2.00 (t, J = 7.6 Hz, 1H), 1.79−1.70 (m,

2H), 1.51−1.44 (m, 2H). HRMS (ESI): exact mass calculated for

C₃₂H₃₀BrClN₃O₄[M + H]⁺634.1103, found 634.1104. Purity,

96.84%.

5-((5-((2-Bromo-[1,1′-biphenyl]-3-yl)methoxy)-4-chloro-2-((4-

methylpiperazin-1-yl)methyl)phenoxy)methyl)benzo[c][1,2,5]-

oxadiazole (L32). L32 was prepared starting with compound 14c and

1-methylpiperazine according to the Borch reductive amination

procedure described for the preparation of L1 to yield a white solid

(14 mg, 21%). ¹H NMR (300 MHz, DMSO-d₆) δ 8.14 (s, 1H), 8.10

(d, J = 9.1 Hz, 1H), 7.69−7.61 (m, 2H), 7.52−7.41 (m, 4H), 7.40−

7.31 (m, 4H), 7.09 (s, 1H), 5.39 (s, 2H), 5.29 (s, 2H), 3.51 (s, 2H),

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-bi-

phenyl]-3-yl)methoxy)-5-chlorobenzyl)glycine (L37). L37 was pre-

pared starting with compound 14c and glycine methyl ester according

8405

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

to the procedure described for the preparation of L1 to yield a white

solid (16 mg, 24% for two steps). ¹H NMR (300 MHz, DMSO-d₆) δ

8.16 (s, 1H), 8.09 (d, J = 9.6 Hz, 1H), 7.71 (d, J = 9.6 Hz, 1H), 7.62

(d, J = 7.1 Hz, 1H), 7.53 (s, 1H), 7.46 (dd, J = 12.1, 6.2 Hz, 4H),

7.41−7.31 (m, 3H), 7.10 (s, 1H), 5.40 (s, 2H), 5.31 (s, 2H), 3.99 (s,

2H), 3.15 (s, 2H). ¹³C NMR (126 MHz, DMSO-d₆) δ 170.66,

156.41, 154.63, 149.42, 149.14, 143.72, 142.32, 141.42, 136.98,

132.85, 131.79, 131.40, 129.61, 129.13, 128.52, 123.13, 119.90,

116.82, 114.71, 113.51, 102.23, 72.10, 70.31, 49.51, 46.07. HRMS

(ESI): exact mass calculated for C₂₉H₂₄BrClN₃O₅[M + H]⁺

608.0582, found 608.0589. Purity, 97.98%.

to yield a white solid (28 mg, 21%). ¹H NMR (300 MHz, DMSO-d₆)

δ 8.18−7.97 (m, 3H), 7.77−7.59 (m, 2H), 7.55−7.29 (m, 8H), 7.09

(s, 1H), 5.38 (s, 2H), 5.33 (s, 2H), 4.21 (t, 1H), 3.79 (s, 2H), 3.73

(m, 1H), 3.54 (m, J = 10.4, 4.1 Hz, 1H), 3.44 (dd, J = 3.9 Hz, 1H),

3.38−3.25 (m, 3H), 3.16 (m, 1H), 1.04 (d, J = 2.8 Hz, 9H). ¹³C

NMR (126 MHz, DMSO-d₆) δ 172.56, 171.66, 155.91, 153.54,

149.31, 149.05, 142.56, 141.11, 139.26, 135.30, 132.83, 131.61,

131.01, 130.86, 129.75, 129.08, 128.72, 128.26, 127.64, 122.81,

116.93, 113.65, 113.06, 101.10, 73.21, 69.49, 69.04, 62.98, 62.51,

62.14, 54.51, 45.92, 27.58. HRMS (ESI): exact mass calculated for

C₃₇H₃₉Cl₂N₄O₈[M + H]⁺737.2139, found 737.2141. Purity, 95.96%.

(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-^L-leucyl-D-leucine

(L42). L42 was prepared starting with compound 14d and ethyl ^L-

leucyl-^D-leucinate according to the Borch reductive amination

procedure described for the preparation of L1 to yield a white solid

(29 mg, 25%). ¹H NMR (300 MHz, DMSO-d₆) δ 8.06 (t, 3H), 7.72−

7.59 (m, 2H), 7.54−7.30 (m, 8H), 7.06 (s, 1H), 5.35 (s, 2H), 5.32 (s,

2H), 4.34−4.20 (m, 1H), 3.65 (dd, J = 33.8, 13.9 Hz, 2H), 3.10−3.01

(m, 1H), 1.77−1.63 (m, 1H), 1.62−1.52 (m, 1H), 1.50 (dd, J = 6.6

Hz, 2H), 1.32 (dd, J = 8.2 Hz, 2H), 0.90−0.67 (m, 12H). ¹³C NMR

(126 MHz, DMSO-d₆) δ 174.95, 174.72, 155.68, 153.35, 149.34,

149.07, 142.62, 141.12, 139.27, 135.34, 132.90, 131.60, 131.02,

130.36, 129.75, 129.06, 128.72, 128.27, 127.63, 123.32, 116.84,

113.66, 113.05, 101.00, 69.49, 69.04, 60.38, 50.51, 45.45, 43.18, 24.87,

24.70, 23.49, 23.34, 22.56, 21.67. HRMS (ESI): exact mass calculated

for C₃₉H₄₃Cl₂N₄O₆[M + H]⁺733.2554, found 733.2549. Purity,

96.08%.

N-(2-(((2-methoxy-6-((2-methyl-[1,1′-biphenyl]-3-yl)methoxy)-

pyridin-3-yl)methyl)amino)ethyl)acetamide (BMS-202). BMS-202

was synthesized according to the procedures described in BMS

patents.¹⁸¹H NMR (300 MHz, DMSO-d₆) δ 7.81 (s, 1H), 7.63 (d, J

= 7.9 Hz, 1H), 7.52−7.40 (m, 3H), 7.39 (d, J = 6.9 Hz, 1H), 7.31

(dd, J = 8.2, 1.6 Hz, 2H), 7.25 (d, J = 7.5 Hz, 1H), 7.23−7.13 (m,

1H), 6.43 (d, J = 7.8 Hz, 1H), 5.40 (s, 2H), 3.89 (s, 2H), 3.59 (s,

2H), 3.12 (q, J = 6.2 Hz, 2H), 2.58−2.50 (m, 2H), 2.21 (s, 3H), 1.78

(s, 3H). ¹³C NMR (126 MHz, DMSO-d₆) δ 169.68, 161.17, 141.88,

141.67, 136.44, 134.25, 129.94, 129.61, 128.68, 128.63, 127.39,

125.91, 113.68, 101.21, 66.61, 53.64, 48.42, 46.40, 38.91, 23.09, 16.37.

HRMS (ESI): exact mass calculated for C₂₅H₃₀N₃O₃[M + H]⁺

420.2242, found 420.2288. Purity, 99.17%.

2-((2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-4-((2-bromo-[1,1′-

biphenyl]-3-yl)methoxy)-5-chlorobenzyl)amino)ethane-1-sulfona-

mide (L38). L38 was prepared starting with compound 14c and 2-

aminoethanesulfonamide according to the Borch reductive amination

procedure described for the preparation of L1 to yield a white solid

1

(28 mg, 24%). H NMR (300 MHz, DMSO-d₆) δ 8.14−8.05 (m,

2H), 7.67 (d, J = 9.5 Hz, 1H), 7.62 (d, J = 6.2 Hz, 1H), 7.53−7.30

(m, 8H), 7.06 (s, 1H), 6.80 (s, 2H), 5.38 (s, 2H), 5.28 (s, 2H), 3.73

(s, 2H), 3.16 (t, J = 7.0 Hz, 2H), 2.92 (t, J = 7.1 Hz, 2H). ¹³C NMR

(126 MHz, DMSO) δ 155.68, 153.35, 149.34, 149.09, 143.47, 142.65,

141.26, 136.94, 132.91, 131.45, 130.33, 129.69, 129.16, 128.62,

128.20, 128.15, 123.41, 123.14, 116.93, 113.70, 113.21, 101.24, 71.52,

69.57, 55.00, 46.71, 43.87. HRMS (ESI): exact mass calculated for

C₂₉H₂₇BrClN₄O₅S [M + H]⁺657.0569, found 657.0571. Purity,

97.31%.

N-(2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-2-(piperidin-1-yl)-

ethan-1-amine dihydrochloride (L39). L39 was prepared starting

with compound 14d and 2-piperidin-1-ylethanamine according to the

Borch reductive amination procedure described for the preparation of

L1 to yield a colorless oil. The oil product was added to a

hydrochloric acid ethanol solution (4 mL) and precipitated as a white

solid. The precipitate was ﬁltered, washed with CH₃OH (1.5 mL),

1

and dried to yield a white solid (17 mg, 26%). H NMR (300 MHz,

DMSO-d₆) δ 10.69 (s, 1H), 9.58 (s, 2H), 8.22 (s, 1H), 8.11 (d, J =

9.3 Hz, 1H), 7.78 (d, J = 9.3 Hz, 1H), 7.73 (s, 1H), 7.64 (dd, J = 7.3,

1.9 Hz, 1H), 7.54−7.46 (m, 3H), 7.45−7.34 (m, 4H), 7.18 (s, 1H),

5.48 (s, 2H), 5.38 (s, 2H), 4.25 (s, 2H), 3.54−3.42 (m, 6H), 2.91 (t,

2H), 1.87−1.73 (m, 4H), 1.71−1.64 (m, 1H), 1.50−1.33 (m, 1H).

¹³C NMR (126 MHz, DMSO-d₆) δ 156.55, 155.44, 149.37, 149.07,

142.22, 141.16, 139.20, 134.95, 133.19, 132.85, 131.78, 129.73,

129.32, 128.72, 128.28, 127.66, 116.80, 113.99, 113.54, 100.77, 69.84,

69.16, 53.06, 52.48, 44.57, 41.45, 22.77, 21.57. HRMS (ESI): exact

mass calculated for C₃₄H₃₅Cl₂N₄O₃[M + H]⁺617.2081, found

617.2080. Purity, 95.57%.

(2-((3-Cyanobenzyl)oxy)-4-((2-methyl-[1,1′-biphenyl]-3-yl)-

methoxy)benzyl)-^D-serine (BMS-1016). BMS-1016 was synthesized

according to the procedures described in BMS patents.¹⁹¹H NMR

(300 MHz, DMSO-d₆) δ 8.02 (s, 1H), 7.91 (d, J = 7.9 Hz, 1H), 7.81

(d, J = 7.6 Hz, 1H), 7.61 (t, J = 7.8 Hz, 1H), 7.53−7.10 (m, 9H), 6.83

(d, J = 2.2 Hz, 1H), 6.72 (d, J = 8.4 Hz, 1H), 5.25 (s, 2H), 5.15 (s,

2H), 4.16−3.97 (m, 2H), 3.82−3.54 (m, 4H), 3.17 (t, J = 5.7 Hz,

1H), 2.18 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆) δ 168.78, 159.99,

157.23, 142.15, 141.31, 138.46, 135.34, 133.82, 132.37, 131.86,

131.63, 130.93, 129.66, 129.58, 129.09, 128.18, 127.91, 126.90,

125.45, 118.63, 115.02, 111.45, 106.13, 100.58, 68.65, 68.50, 62.16,

60.42, 45.24, 15.77. HRMS (ESI): exact mass calculated for

C₃₂H₃₁N₂O₅[M + H]⁺523.2188, found 523.2234. Purity, 99.40%.

(S)-5-(((2-(Benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)amino)methyl)-

pyrrolidin-2-one hydrochloride (L40). L40 was prepared starting

with compound 14d and (5S)-5-(aminomethyl)pyrrolidin-2-one

according to the Borch reductive amination procedure described for

the preparation of L1 to yield a colorless oil. The oil product was

added to a hydrochloric acid ethanol solution (2 mL) and precipitated

as a white solid. The precipitate was ﬁltered, washed with CH₃OH

1

(1.5 mL), and dried to yield a white solid (10 mg, 19%). H NMR

Docking Method. Molecular docking was completed with Glide

(300 MHz, DMSO-d₆) δ 9.14 (s, 2H), 8.18 (s, 1H), 8.11 (d, J = 9.3

Hz, 1H), 7.77 (s, 2H), 7.72 (d, J = 9.3 Hz, 1H), 7.64 (dd, 1H), 7.44

(m, J = 16.5, 7.5, 4.3 Hz, 7H), 7.18 (s, 1H), 5.45 (s, 2H), 5.39 (s,

2H), 4.21 (s, 2H), 4.01−3.77 (m, 1H), 3.16−2.90 (m, 2H), 2.25−

2.05 (m, 3H), 1.83−1.65 (m, 1H). ¹³C NMR (126 MHz, DMSO-d₆)

δ 176.78, 156.68, 155.42, 149.34, 149.07, 142.11, 141.17, 139.20,

134.97, 133.15, 133.07, 131.77, 131.21, 129.73, 129.29, 128.72,

128.28, 127.67, 116.85, 114.05, 113.64, 113.55, 100.69, 69.87, 69.15,

52.37, 50.52, 44.78, 29.70, 24.94. HRMS (ESI): exact mass calculated

for C₃₂H₂₉Cl₂N₄O₄[M + H]⁺603.15604, found 603.15564. Purity,

98.12%.

30

̈

5.5 implemented in Schrodinger2009. The crystal structure (PDB

ID: 5J89) of PD-L1 in complex with BMS-202 was obtained from the

Protein Data Bank (http://www.rcsb.org/). The crystal structure was

prepared with the Protein Preparation Wizard workﬂow. All of the

water molecules were removed. The grid ﬁle for molecular docking

was generated based on the binding site, which was deﬁned by a box

centered on the centroid of the crystal ligand and in similar size to it.

Compounds were prepared with LigPrep and docked using the Glide

extra-precision (XP) mode. Default settings were used for all of the

other parameters.

PD-1/PD-L1 Homogeneous Time-Resolved Fluorescence

(HTRF) Binding Assay. The blockade activity of compounds against

PD-1/PD-L1 interaction was determined by the PD-1/PD-L1 binding

assay kit (cisbio, 64ICP01PEG). Brieﬂy, the test compound was

diluted to a series of appropriate concentrations (0.1 nM to 10 μM)

N-(2-(benzo[c][1,2,5]oxadiazol-5-ylmethoxy)-5-chloro-4-((2-

chloro-[1,1′-biphenyl]-3-yl)methoxy)benzyl)-O-(tert-butyl)-^D-seryl-

D-serine (L41). L41 was prepared starting with compound 14d and

ethyl O-(tert-butyl)-^D-seryl-D-serinate according to the Borch

reductive amination procedure described for the preparation of L1

8406

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

by diluent buﬀer and applied to the 384-well plate in triplicate (2 μL

volume). Then, 4 μL of Tag1-PD-L1 protein and 4 μL of Tag2-PD-1

protein were added and incubated for 15 min at room temperature.

Then, 10 μL of premixed anti-Tag1-Eu³⁺and anti-Tag2-XL665 was

added and the plate was sealed with 2 h incubation at room

temperature. At last, the signals were read on an EnVision multilabel

plate reader (PerkinElmer, Waltham, MA) with excitation at 337 nm

and emission at 665 and 610 nm. HTRF signals were calculated as a

ratio as follows: (signal of 665 nm)/(signal of 615 nm) × 10 000. The

inhibition rates and the half-maximal inhibitory concentrations (IC₅₀)

were calculated using GraphPad Prism 7.0 version.

Cell Viability Assay. The toxicity of the compounds was tested in

PD-L1⁺/aAPC/CHO-K1 cells and MC38 cells by the methyl thiazolyl

tetrazolium (MTT) assay and in PD-1⁺/NFAT-luc/Jurkat cells by the

CCK8 assay. Test compounds were dissolved in dimethyl sulfoxide

(DMSO) and subsequently diluted in a culture medium with a 1%

ﬁnal DMSO concentration. Brieﬂy, PD-L1⁺/aAPC/CHO-K1 cells or

MC38 cells were seeded in 96-well microplates at an appropriate

density (PD-L1⁺/aAPC/CHO-K1, 2000 cells/well; MC38, 5000

cells/well) in 100 μL of medium and allowed to adhere for 12 h. The

cells were treated with test compounds and incubated for 24 h. For

the PD-1⁺/NFAT-luc/Jurkat cell assay, the cells were seeded in 96-

well microplates at a density of 20 000 cells per well in 100 μL of

medium, treated with the test compounds, and incubated for 24 h.

After incubation, CCK8 and MTT were added into the well according

to the manufacturer’s protocol and incubated for 4 h. The absorbance

of the CCK8 assay was read on an automated microplate

spectrophotometer (EnSpire, PerkinElmer). In the MTT assay, the

culture medium should be replaced with DMSO to test the

absorbance. Statistical analysis was carried out using GraphPad

Prism 7.0 version.

Cell-Based PD-1/PD-L1 Signaling Blockade Assay. The PD-

1⁺/NFAT-luc/Jurkat cells and PD-L1⁺/aAPC/CHO-K1 cells were

purchased from Genomeditech (Shanghai, China). Anti-PD-L1 mAb

(Atezolizumab) was purchased from Selleckchem (Shanghai, China).

The PD-L1⁺/aAPC/CHO-K1 cells were seeded at a cell density of

25 000 cells/well into 96-well plates 12 h prior to the assay.

Compounds were dissolved in diluted culture medium with 1% ﬁnal

DMSO concentration. The culture medium in the wells was removed,

and 50 μL of compound dilutions was added. The PD-1⁺/NFAT-luc/

Jurkat cells were then seeded at 100 000 cells/well in 50 μL of culture

medium and incubated at 37 °C and in 5% CO₂for 6 h. After

incubation, Bio-Glo reagent was added to each well, and luminescence

was quantiﬁed using an automated microplate spectrophotometer

(EnSpire, PerkinElmer). EC₅₀values and maximal luminescence

values (RLUmax) were calculated by GraphPad Prism 7.0 version.

Binding Aﬃnity Assay. The binding aﬃnity of compounds with

the human PD-L1 IgV domain (human PD-L1/B7-H1 protein, His

tag, PD-1-H5229) and the human PD-1 IgV domain (human PD-1/

PDCD1 protein, Fc tag, ACRO, PD-1-H5257) was examined on a

Biacore T200 (GE Healthcare Biosciences, Sweden). Brieﬂy, His-PD-

L1 or Fc-PD-1 was diluted using 1× phosphate-buﬀered saline

(PBST) buﬀer (99.95% PBS + 0.05% Tween 20) and immobilized

onto one CM5 chip, all in 10 mM sodium acetate (pH 4.5) by a

standard amine-coupling procedure. Compounds were serially diluted

with the phosphate-buﬀered saline (PBS) buﬀer containing 0.05%

Tween 20 and 5% DMSO to appropriate concentrations and injected

at a ﬂow through the chip following the preset program. The K_Dvalue

was determined with Biacore T200 evaluation software (GE

Healthcare).

B16-F10 Melanoma Mouse Model. All animal studies were

conducted in accordance with the guiding principles of the Animal

Care and Use Committee of the China Pharmaceutical University

(Nanjing, China).

C57BL/6J mice (female, 8 week old) were purchased from Beijing

Vital River Laboratory Animal Technology (Beijing, China) and

maintained under a controlled environment at 25 °C for a 12 h light/

dark cycle, with free access to food and water. B16-F10 tumors were

established by injecting 1 × 10⁶B16-F10 cells in PBS into the dorsal

area of mice. The mice bearing tumors were randomly divided into

four groups and treated with 0.5% sodium carboxymethyl cellulose

(CMC-Na) (vehicle), L24 (15 mg/kg), L24 (45 mg/kg), and CTX

(80 mg/kg) by oral gavage daily. Treatment started on the same day

when the tumor cells were injected (day 1). Tumor dimensions and

animal body weights were measured every day. Tumor volumes were

calculated according to the following formula: volume (mm³) = 0.5 ×

length (mm) × width (mm) × width (mm). On day 18, the mice

were sacriﬁced and the tumors were isolated. Tumor growth

inhibition (TGI) rates were calculated using the following formula:

TGI (%) = [1 − V_t/V_v] × 100%, where V_tand V_vare the mean tumor

volumes of the treatment group and vehicle control, respectively.

MC38 Tumor Mouse Model. C57BL/6J mice (female, 8 week

old) were purchased from Beijing Vital River Laboratory Animal

Technology (Beijing, China) and maintained under a controlled

environment at 25 °C in a 12 h light/dark cycle, with free access to

food and water. MC38 tumors were established by injecting 1 × 10⁶

MC38 cells in PBS into the dorsal area of mice. When the tumor

volume reached 50 mm³, the mice bearing tumors were randomly

divided into four groups and treated with CMC-Na (vehicle), L24

(50 mg/kg), L24 (150 mg/kg), and L24 (300 mg/kg) by oral gavage

daily. Tumor dimensions and animal body weights were measured

every two days. Tumor volumes were calculated according to the

following formula: volume (mm³) = 0.5 × length (mm) × width

(mm) × width (mm). After 18 days of treatment, the mice were

sacriﬁced and the tumors were isolated. Tumor growth inhibition

(TGI) rates were calculated as mentioned above.

PD-L1 Humanized Tumor Mouse Model. The PD-L1

humanized C57BL/6 mice (female, 6 week old) and MC38-hPD-

L1(Tg)-mPDL1(KO) cells were purchased from GemPharmatech

(Nanjing, China). PD-L1-humanized C57BL/6 mice were injected

with 1 × 10⁶MC38-hPD-L1(Tg)-mPDL1(KO) cells in PBS into the

right ﬂank area. When the tumor volume reached 50 mm³, the mice

bearing tumors were randomly divided into three groups and treated

with CMC-Na (vehicle) and L24 (5 and 25 mg/kg) by oral gavage

daily. Tumor dimensions and animal body weights were measured

twice a week. Tumor volumes were calculated according to the

following formula: volume (mm³) = 0.5 × length (mm) × width

(mm) × width (mm). On day 27, the mice were sacriﬁced and the

tumors were isolated. Tumor growth inhibition (TGI) rates were

calculated as mentioned above.

Flow Cytometry. Antibodies were purchased from BD Bio-

sciences, eBioscience, or Invitrogen. The tumor tissue was isolated

and digested into a single-cell suspension. Red blood cells were lysed

by ACK lysis buﬀer and incubated with Fc block antibodies. For cell

surface staining, the cells were incubated with antibodies anti-CD45

(BV510), anti-CD3 (BV711), anti-CD4 (APC-H7), anti-CD8 (PE),

anti-CD335 (AF647), anti-CD11b (PE-CF594), anti-Ly6G (BV421),

anti-Ly6C (FITC), anti-F4/80 (Super Bright 600) for 60 min, and

after washing twice with PBS, sytoxAAD (Invitrogen, S10274) was

added before being read on ﬂow cytometry. As for intracellular

staining, cells were labeled with FVS780 (BD, 565388), permeabilized

with Transcription Factor Staining Buﬀer (eBioscience, 00-5523)

after surface staining, and incubated with anti-FOXP3 (PE) for 60

min. All samples were read on an Attune NxT (Thermo).

Pharmacokinetic Study. The pharmacokinetic studies on

selected compounds were conducted by Medicilon (Shanghai,

China). Brieﬂy, male Sprague−Dawley rats or C57BL/6 mice were

prohibited diet 12 h before the experiment. Blood samples (0.2 mL)

were collected from the jugular vein at 0.0833, 0.25, 0.5, 1, 1.5, 2, 4, 8,

12, and 24 h after oral or intravenous administration. The data was

detected by liquid chromatography−mass spectrometry (LC−MS;

TQ6500+ Triple quad), and PK parameters were determined by

Phoenix WinNonlin7.0.

ASSOCIATED CONTENT

■

sı

* Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.jmedchem.1c00392.

8407

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

¹H NMR, ¹³C NMR, and UPLC spectra of target

compounds; ﬁgures of pharmacokinetic studies; toxicity

data in vivo and in vitro (PDF)

ACKNOWLEDGMENTS

■

Financial support from the Jiangsu Natural Science Foundation

(BK20202009) is gratefully acknowledged. This research was

also supported by the National Natural Science Foundation of

China (81730094, 81773584, and 21977118).

Docking models (5J89-BMS-1016-XP) (PDB)

Docking models (5J89-L1-XP) (PDB)

Docking models (5J89-L7-XP-reﬁne) (PDB)

Molecular formula strings (CSV)

ABBREVIATIONS

■

PD-1, programmed cell death-1; PD-L1, programmed cell

death-ligand 1; HTRF, homogeneous time-resolved ﬂuores-

cence; SPR, surface plasmon resonance; mAbs, monoclonal

antibodies; FDA, Food and Drug Administration; irAEs,

immune-related adverse eﬀects; VISTA, V-domain Ig suppres-

sor of T-cell activation; DMF, N,N-dimethylformamide;

DIPEA, N,N-diisopropylethylamine; Pd(dppf)Cl₂.CH₂Cl₂,

[1,1′-bis(diphenylphosphino) ferrocene] dichloropalladium

(II), complex with dichloromethane; NFAT, nuclear factor

of activated T cells; TCR, T-cell receptor; PK, pharmacoki-

netic; AST, aspartate aminotransferase; ALT, alanine amino-

transferase; CK, creatine kinase; BUN, blood urea nitrogen;

TGI, tumor growth inhibition; CTX, cyclophosphamide;

MDSC, myeloid-derived suppressor cell; MΦs, macrophages;

NK cells, natural killer cells; TAMs, tumor-associated macro-

phages; TILs, tumor-inﬁltrating lymphocytes

AUTHOR INFORMATION

Corresponding Authors

■

Haoliang Yuan − Jiangsu Key Laboratory of Drug Discovery

for Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China; orcid.org/0000-0003-0248-4347;

Email: yhl@cpu.edu.cn

Hongbin Sun − Jiangsu Key Laboratory of Drug Discovery for

Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China; State Key Laboratory for the Chemistry and

Molecular Engineering of Medicinal Resources, School of

Chemistry and Pharmacy, Guangxi Normal University,

Guilin 541004, China; orcid.org/0000-0002-3452-7674;

Phone: +86-25-83271196; Email: hongbinsun@

cpu.edu.cn

REFERENCES

■

Authors

(1) Ribas, A.; Wolchok, J. D. Cancer immunotherapy using

checkpoint blockade. Science 2018, 359, 1350−1355.

(2) Sharma, P.; Allison, J. P. The future of immune checkpoint

therapy. Science 2015, 348, 56−61.

Liu Liu − Jiangsu Key Laboratory of Drug Discovery for

Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China

Zhiying Yao − Jiangsu Key Laboratory of Drug Discovery for

Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China

Shijun Wang − Jiangsu Key Laboratory of Drug Discovery for

Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China

Tao Xie − Jiangsu Key Laboratory of Drug Discovery for

Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China

Guoqing Wu − Jiangsu Key Laboratory of Drug Discovery for

Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China

Honghan Zhang − Jiangsu Key Laboratory of Drug Discovery

for Metabolic Disease, State Key Laboratory of Natural

Medicines, China Pharmaceutical University, Nanjing

210009, China

(3) Chen, Y.; Pei, Y.; Luo, J.; Huang, Z.; Yu, J.; Meng, X. Looking for

the optimal PD-1/PD-L1 inhibitor in cancer treatment: a comparison

in basic structure, function, and clinical practice. Front. Immunol.

2020, 11, No. 1088.

(4) Hodi, F. S.; O’Day, S. J.; McDermott, D. F.; Weber, R. W.;

Sosman, J. A.; Haanen, J. B.; Gonzalez, R.; Robert, C.; Schadendorf,

D.; Hassel, J. C.; Akerley, W.; van den Eertwegh, A. J. M.; Lutzky, J.;

Lorigan, P.; Vaubel, J. M.; Linette, G. P.; Hogg, D.; Ottensmeier, C.

H.; Lebbé, C.; Peschel, C.; Quirt, I.; Clark, J. I.; Wolchok, J. D.;

Weber, J. S.; Tian, J.; Yellin, M. J.; Nichol, G. M.; Hoos, A.; Urba, W.

J. Improved survival with ipilimumab in patients with metastatic

melanoma. N. Engl. J. Med. 2010, 363, 711−723.

(5) Boussiotis, V. A. Molecular and biochemical aspects of the PD-1

checkpoint pathway. N. Engl. J. Med. 2016, 375, 1767−1778.

(6) Finn, R. S.; Qin, S.; Ikeda, M.; Galle, P. R.; Ducreux, M.; Kim, T.

Y.; Kudo, M.; Breder, V.; Merle, P.; Kaseb, A. O.; Li, D.; Verret, W.;

Xu, D. Z.; Hernandez, S.; Liu, J.; Huang, C.; Mulla, S.; Wang, Y.; Lim,

H. Y.; Zhu, A. X.; Cheng, A. L.; Investigators, I.M.. Atezolizumab plus

bevacizumab in unresectable hepatocellular carcinoma. N. Engl. J.

Med. 2020, 382, 1894−1905.

(7) Hack, S. P.; Zhu, A. X.; Wang, Y. Augmenting anticancer

immunity through combined targeting of angiogenic and PD-1/PD-

L1 pathways: Challenges and opportunities. Front. Immunol. 2020, 11,

No. 598877.

Pu Zhang − Jiangsu Flag Chemical Industry Co., Ltd., Nanjing

211500, China

Yaojun Wu − Jiangsu Flag Chemical Industry Co., Ltd.,

Nanjing 211500, China

(8) Yang, J.; Hu, L. Immunomodulators targeting the PD-1/PD-L1

protein-protein interaction: from antibodies to small molecules. Med.

Res. Rev. 2019, 39, 265−301.

(9) Kamada, T.; Togashi, Y.; Tay, C.; Ha, D.; Sasaki, A.; Nakamura,

Y.; Sato, E.; Fukuoka, S.; Tada, Y.; Tanaka, A.; Morikawa, H.;

Kawazoe, A.; Kinoshita, T.; Shitara, K.; Sakaguchi, S.; Nishikawa, H.

PD-1(+) regulatory T cells amplified by PD-1 blockade promote

hyperprogression of cancer. Proc. Natl. Acad. Sci. U.S.A. 2019, 116,

9999−10008.

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.1c00392

Author Contributions

^∥L.L., Z.Y., and S.W. contributed equally to this work.

(10) Bailly, C.; Vergoten, G. Protein homodimer sequestration with

small molecules: Focus on PD-L1. Biochem. Pharmacol. 2020, 174,

No. 113821.

Notes

The authors declare no competing ﬁnancial interest.

8408

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(11) Zak, K. M.; Grudnik, P.; Magiera, K.; Domling, A.; Dubin, G.;

Holak, T. A. Structural biology of the immune checkpoint receptor

PD-1 and its ligands PD-L1/PD-L2. Structure 2017, 25, 1163−1174.

(12) Zak, K. M.; Kitel, R.; Przetocka, S.; Golik, P.; Guzik, K.;

Musielak, B.; Domling, A.; Dubin, G.; Holak, T. A. Structure of the

complex of human programmed death 1, PD-1, and its ligand PD-L1.

Structure 2015, 23, 2341−2348.

C.; Ribas, A. PD-1 blockade induces responses by inhibiting adaptive

immune resistance. Nature 2014, 515, 568−571.

(28) Qian, B. Z.; Pollard, J. W. Macrophage diversity enhances

tumor progression and metastasis. Cell 2010, 141, 39−51.

(29) Noman, M. Z.; Desantis, G.; Janji, B.; Hasmim, M.; Karray, S.;

Dessen, P.; Bronte, V.; Chouaib, S. PD-L1 is a novel direct target of

HIF-1alpha, and its blockade under hypoxia enhanced MDSC-

mediated T cell activation. J. Exp. Med. 2014, 211, 781−790.

(30) Kontoyianni, M.; McClellan, L. M.; Sokol, G. S. Evaluation of

docking performance: comparative data on docking algorithms. J.

Med. Chem. 2004, 47, 558−565.

(13) Chen, Y.; Liu, P.; Gao, F.; Cheng, H.; Qi, J.; Gao, G. F. A

dimeric structure of PD-L1: functional units or evolutionary relics?

Protein Cell 2010, 1, 153−160.

(14) Li, K.; Tian, H. Development of small-molecule immune

checkpoint inhibitors of PD-1/PD-L1 as a new therapeutic strategy

for tumour immunotherapy. J. Drug Targeting 2019, 27, 244−256.

(15) Wang, T.; Wu, X.; Guo, C.; Zhang, K.; Xu, J.; Li, Z.; Jiang, S.

Development of inhibitors of the programmed cell death-1/

programmed cell death-ligand 1 signaling pathway. J. Med. Chem.

2019, 62, 1715−1730.

(16) Cheng, B.; Ren, Y.; Niu, X.; Wang, W.; Wang, S.; Tu, Y.; Liu,

S.; Wang, J.; Yang, D.; Liao, G.; Chen, J. Discovery of novel resorcinol

dibenzyl ethers targeting the programmed cell death-1/programmed

cell death-ligand 1 interaction as potential anticancer agents. J. Med.

Chem. 2020, 63, 8338−8358.

(17) Qin, M.; Cao, Q.; Zheng, S.; Tian, Y.; Zhang, H.; Xie, J.; Xie,

H.; Liu, Y.; Zhao, Y.; Gong, P. Discovery of [1,2,4]Triazolo[4,3-

a]pyridines as potent inhibitors targeting the programmed cell death-

1/programmed cell death-ligand 1 interaction. J. Med. Chem. 2019,

62, 4703−4715.

(18) Chupak, L. S.; Zheng, X. Compounds Useful as Immunomo-

dulators. Bristol-Myers Squibb Co., WO2015034820, 2015.

(19) Chupak, L. S.; Ding, M.; Martin, S. W.; Zheng, X.; Hewawasam,

P.; Connoly, T. P.; Xu, N.; Yeung, K.-S.; Zhu, J.; Langley, D. R.;

Tenney, D. J.; Scola, P.M. Compounds useful as immunomodulators.

Bristol-Myers Squibb Co., WO2015160641, 2015.

(20) Zak, K. M.; Grudnik, P.; Guzik, K.; Zieba, B. J.; Musielak, B.;

Dömling, A.; Dubin, G.; Holak, T. A. Structural basis for small

molecule targeting of the programmed death ligand 1 (PD-L1).

Oncotarget 2016, 7, 30323−30335.

(21) Ganesan, A.; Ahmed, M.; Okoye, I.; Arutyunova, E.; Babu, D.;

Turnbull, W. L.; Kundu, J. K.; Shields, J.; Agopsowicz, K. C.; Xu, L.;

Tabana, Y.; Srivastava, N.; Zhang, G.; Moon, T. C.; Belovodskiy, A.;

Hena, M.; Kandadai, A. S.; Hosseini, S. N.; Hitt, M.; Walker, J.;

Smylie, M.; West, F. G.; Siraki, A. G.; Lemieux, M. J.; Elahi, S.;

Nieman, J. A.; Tyrrell, D. L.; Houghton, M.; Barakat, K.

Comprehensive in vitro characterization of PD-L1 small molecule

inhibitors. Sci. Rep. 2019, 9, No. 12392.

(22) Lee, H. T.; Lee, J. Y.; Lim, H.; Lee, S. H.; Moon, Y. J.; Pyo, H.

J.; Ryu, S. E.; Shin, W.; Heo, Y. S. Molecular mechanism of PD-1/PD-

L1 blockade via anti-PD-L1 antibodies atezolizumab and durvalumab.

Sci. Rep. 2017, 7, 5532.

(23) Inman, B. A.; Longo, T. A.; Ramalingam, S.; Harrison, M. R.

Atezolizumab: a PD-L1-blocking antibody for bladder cancer. Clin.

Cancer Res. 2017, 23, 1886−1890.

(24) Lin, D. Y- W.; Tanaka, Y.; Iwasaki, M.; Gittis, A. G.; Su, H. P.;

Mikami, B.; Okazaki, T.; Honjo, T.; Minato, N.; Garboczi, D. N. The

PD-1/PD-L1 complex resembles the antigen-binding Fv domains of

antibodies and T cell receptors. Proc. Natl. Acad. Sci. U.S.A. 2008, 105,

3011−3016.

(25) De La Rochere, P.; Guil-Luna, S.; Decaudin, D.; Azar, G.;

Sidhu, S. S.; Piaggio, E. Humanized mice for the study of immuno-

oncology. Trends Immunol. 2018, 39, 748−763.

(26) Long, E. O.; Kim, H. S.; Liu, D.; Peterson, M. E.; Rajagopalan,

S. Controlling natural killer cell responses: integration of signals for

activation and inhibition. Annu. Rev. Immunol. 2013, 31, 227−258.

(27) Tumeh, P. C.; Harview, C. L.; Yearley, J. H.; Shintaku, I. P.;

Taylor, E. J.; Robert, L.; Chmielowski, B.; Spasic, M.; Henry, G.;

Ciobanu, V.; West, A. N.; Carmona, M.; Kivork, C.; Seja, E.; Cherry,

G.; Gutierrez, A. J.; Grogan, T. R.; Mateus, C.; Tomasic, G.; Glaspy, J.

A.; Emerson, R. O.; Robins, H.; Pierce, R. H.; Elashoff, D. A.; Robert,

8409

https://doi.org/10.1021/acs.jmedchem.1c00392

J. Med. Chem. 2021, 64, 8391−8409

Article Doi

Syntheses, Biological Evaluations, and Mechanistic Studies of Benzo[c][1,2,5]oxadiazole Derivatives as Potent PD-L1 Inhibitors within VivoAntitumor Activity

DOI: 10.1021/acs.jmedchem.1c00392

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.1c00392

Full text of DOI:10.1021/acs.jmedchem.1c00392

Products guided by the article

R&D Labs maybe for 1251950-91-0

Relevant to this article

Hot Product