Discovery and structure-activity relationship studies of 1-aryl-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione derivatives as potent dual inhibitors of indoleamine 2,3-dioxygenase 1 (IDO1) and trytophan 2,3-dioxygenase (TDO)

Article abstract of DOI:10.1016/j.ejmech.2020.112703

Indoleamine 2,3-dioxygenase 1 (IDO1) and tryptophan 2,3-dioxygenase (TDO), which mediate kynurenine pathway of tryptophan degradation, have emerged as potential new targets in immunotherapy for treatment of cancer because of their critical role in immunosuppression in the tumor microenvironment. In this investigation, we report the structural optimization and structure-activity relationship studies of 1-phenyl-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione derivatives as a new class of IDO1/TDO dual inhibitors. Among all the obtained dual inhibitors, 1-(3-chloro-4-fluorophenyl)-6-fluoro-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione (38) displayed the most potent IDO1 and TDO inhibitory activities with IC₅₀ (half-maximal inhibitory concentration) values of 5 nM for IDO1 and 4 nM for TDO. It turned out that compound 38 was not a PAINS compound. Compound 38 could efficiently inhibit the biofunction of IDO1 and TDO in intact cells. In LL2 (Lewis lung cancer) and Hepa1-6 (hepatic carcinoma) allograft mouse models, this compound also showed considerable in vivo anti-tumor activity and no obvious toxicity was observed. Therefore, 38 could be a good lead compound for cancer immunotherapy and deserving further investigation.

Full text of DOI:10.1016/j.ejmech.2020.112703

Journal Pre-proof
Discovery and Structure-Activity Relationship Studies of 1-Aryl-1H-naphtho[2,3-d]
[1,2,3]triazole-4,9-dione Derivatives as Potent Dual Inhibitors of Indoleamine 2,3-
Dioxygenase 1 (IDO1) and Trytophan 2,3-Dioxygenase (TDO)
Shulei Pan, Yangli Zhou, Qiusheng Wang, Yanlin Wang, Chenyu Tian, Tianqi Wang,
Luyi Huang, Jinshan Nan, Linli Li, Shengyong Yang
PII:
S0223-5234(20)30675-9
DOI:
https://doi.org/10.1016/j.ejmech.2020.112703
EJMECH 112703
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 27 April 2020
Revised Date: 21 July 2020
Accepted Date: 27 July 2020
Please cite this article as: S. Pan, Y. Zhou, Q. Wang, Y. Wang, C. Tian, T. Wang, L. Huang, J. Nan,
L. Li, S. Yang, Discovery and Structure-Activity Relationship Studies of 1-Aryl-1H-naphtho[2,3-d]
[1,2,3]triazole-4,9-dione Derivatives as Potent Dual Inhibitors of Indoleamine 2,3-Dioxygenase 1
(IDO1) and Trytophan 2,3-Dioxygenase (TDO), European Journal of Medicinal Chemistry, https://
doi.org/10.1016/j.ejmech.2020.112703.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2020 Elsevier Masson SAS. All rights reserved.

1
2
3
4
Discovery and Structure-Activity Relationship Studies of
1-Aryl-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione Derivatives as Potent Dual
Inhibitors of Indoleamine 2,3-Dioxygenase 1 (IDO1) and Trytophan
2,3-Dioxygenase (TDO)
5
a
a
a
a
a
6
7
8
9
0
1
2
Shulei Pan , Yangli Zhou , Qiusheng Wang , Yanlin Wang , Chenyu Tian , Tianqi
a
a
a
b,
a,
Wang , Luyi Huang , Jinshan Nan , Linli Li *, Shengyong Yang *
a
State Key Laboratory of Biotherapy and Cancer Center, West China Hospital,
Sichuan University, Chengdu, Sichuan 610041, China.
b
1
1
1
Key Laboratory of Drug Targeting and Drug Delivery System of Ministry of
Education, West China School of Pharmacy, Sichuan University, Sichuan 610041,
China.
1
1
1
1
1
1
1
2
2
3
4
5
6
7
8
9
0
1
* Corresponding authors.
E-mail addresses: yangsy@scu.edu.cn (S Yang), ysylilinli@sina.com (L Li)
1

1
2
3
2

1
Abstract
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Indoleamine 2,3-dioxygenase 1 (IDO1) and tryptophan 2,3-dioxygenase (TDO),
which mediate kynurenine pathway of tryptophan degradation, have emerged as
potential new targets in immunotherapy for treatment of cancer because of their
critical role in immunosuppression in the tumor microenvironment. In this
investigation, we report the structural optimization and structure-activity relationship
studies of 1-phenyl-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione derivatives as a new
class of IDO1/TDO dual inhibitors. Among all the obtained dual inhibitors,
1-(3-chloro-4-fluorophenyl)-6-fluoro-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione (38)
displayed the most potent IDO1 and TDO inhibitory activities with IC50
(half-maximal inhibitory concentration) values of 5 nM for IDO1 and 4 nM for TDO.
It turned out that compound 38 was not a PAINS compound. Compound 38 could
efficiently inhibit the biofunction of IDO1 and TDO in intact cells. In LL2 (Lewis
lung cancer) and Hepa1-6 (hepatic carcinoma) allograft mouse models, this
compound also showed considerable in vivo anti-tumor activity and no obvious
toxicity was observed. Therefore, 38 could be a good lead compound for cancer
immunotherapy and deserving further investigation.
1
1
1
1
1
1
1
1
18
1
2
9
0
Keywords: IDO1; TDO; dual inhibitor; structure-activity relationship; cancer
immunotherapy
2
2
1
2
3

1
1．Introduction
2
3
4
5
6
7
8
9
0
1
2
Cancer immunotherapy has attracted much attention in the past years and is
thought as the future direction of cancer treatment [1, 2]. This great upsurge is largely
due to the success of drugs targeting immune checkpoints in clinical cancer therapy,
for example, anti-CTLA4 mAb (Yervoy), anti-PD1 mAb (Keytrude), and anti-PD1
mAb (Opdivo) [3]. Despite the success, many patients across a wide range of
malignancies still do not respond to the immune checkpoint drugs [3]. A possible
reason is that tumor cells could evade immune destruction through evolving various
tactics, in addition to the immune checkpoint system, to avoid, subvert, or manipulate
innate and adaptive immunity [4,5]. Therefore, discovery of more new cancer
immunotherapy strategies that could restore or activate the immune system's response
to cancer is necessary and also important at present.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
3
4
5
6
7
8
9
0
1
2
3
The kynurenine system, namely tryptophan is metabolized to form kynurenine, is
another important tumor immune escape pathway. The kynurenine pathway of
tryptophan metabolism leads to depletion of tryptophan and production of kynurenine
metabolites, both responsible for local immunosuppression [6]. Activation of the
kynurenine pathway results in a dampening of the immune response, leading to tumor
progression and recurrence in a number of ways: blocking T cell activation, increasing
the conversion of naive T cells into regulatory T cells and inducing T cell cycle arrest
and apoptosis, thereby decreasing the presence of effector T cells. Targeting the
kynurenine pathway of tryptophan metabolism may be an alternative therapeutic
strategy in cancer immunotherapy. Indoleamine 2,3-dioxygenase 1(IDO1) and
tryptophan 2,3-dioxygenase (TDO), which are heme-containing dioxygenases, are the
4

1
2
3
4
5
6
7
8
9
initial and rate-limiting enzymes in the kynurenine pathway that catalyze the
oxidative cleavage of tryptophan to N-formyl kynurenine [7]. After the first step of
tryptophan metabolism, N-formyl kynurenine is subsequently converted to a series of
biologically active metabolites including kynurenine, kynurenic acid, quinolinic acid,
and NAD+ [8], which are important contributors to immunosuppression. Importantly,
the up-regulation of IDO1 and/or TDO has been frequently detected in multiple tumor
types, including colorectal, non-small-cell lung carcinoma, ovarian, melanoma, and
hepatocarcinoma [9-13], and is associated with a poor clinical outcome [14,15].
Therefore, IDO1 and TDO are thought as potential targets for cancer immunotherapy.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
Discovery of IDO1 or TDO inhibitors has attracted much attention in both
academic and industry communities in the past several years. A number of potent
IDO1 [16-68] or TDO inhibitors [69-77] have been reported. Typical examples are
displayed in Fig. 1, including 1-MT, Cpd 1, NLG 919, INCB024360\Epacdostat,
Amg-1 and EOS200271/PF-0684003 for IDO1, and LM-10, Cpd 2 and Cpd 3 for
TDO. Several IDO1 inhibitors have entered clinical trials including NLG 919,
INCB024360\Epacdostat and EOS200271/PF-0684003. However, the fail of IDO1
inhibitor Epacadostat in a phase III clinical trial (ECHO-301) for the treatment of
metastatic melanoma significantly reduced the enthusiasm of drug research and
development targeting IDO1.[78] It is speculated that inhibition of IDO1 probably
leads to TDO activation in tumors; TDO has similar biological function as IDO1,
implying that TDO might compensate for the increases ration of Trp/Kyn caused by
IDO1 inhibition. Therefore, agents that are able to inhibit both IDO1 and TDO may
become a new strategy for tumor immunotherapy.
5

1
2
3
4
5
To date, a few of dual IDO1/TDO inhibitors have been reported in literature
[79-85] and typical examples are shown in Figure 1 (Cpd 4 [80] and Cpd 5 [81]).
Nevertheless, most of these inhibitors have just showed moderate potency against
IDO1 and/or TDO. Therefore, discovery of more new dual IDO1/TDO inhibitors,
particularly with high potency against both IDO1 and TDO, is still desired.
O
H N
2
O
O
N
N
Br
N
OH
F
O
N
H N
N
F
2
S
NH
N
N
H
HO
O
H
N
O
N
N
OH
1
-MT
Cpd 1
IDO1 inhibitor
NLG 919
IDO1 inhibitor
INCB024360/Epacdostat
IDO1 inhibitor
IDO1 inhibitor
O
N
N
O
O
N
S
NH
O
NH
N
NH2
O
O
F
O
N
S
F
S
N
H
O
N
H
N
N
N
F
N
H
H
Amg-1
EOS200271/PF-0684003
IDO1 inhibitor
LM-10
Cpd 2
IDO1 inhibitor
TDO inhibitor
TDO inhibitor
O
N
O
N
O
N
N
N
N
Br
R1
O
R2
N
N
NH
N
O
O
HN
N
Cpd 3
Cpd 4
Cpd 5
IDO1/TDO inhibitor
TDO inhibitor
IDO1/TDO inhibitor
6
7
Figure 1. Structures of representative selective IDO1, TDO and dual IDO1/TDO
8
inhibitors.
9
0
1
2
3
The objective of this study is to discover potent dual inhibitors targeting IDO1
and TDO. To this end, we firstly screened our in-house chemical library containing
approximately 5000 compounds with fluorescence method. This screening led to the
retrieval of a compound, 1-phenyl-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione (1,
1
1
1
1
Figure 2), that showed moderate enzymatic inhibitory activities against both IDO1
6

1
2
3
4
and TDO with IC values of 0.100 μM and 0.070 μM, respectively. In the following
50
study, We will perform structural optimization and structure-activity relation analyses
to compound 1. To the most potent compound, further in vivo anti-tumor studies will
be carried out.
5
6
7
Figure 2. Chemical structure of hit compound 1 and regions that are the focus of
structural modifications
8
2. Results and discussion
9
0
1
2
3
4
5
6
2.1. Chemical syntheses of 1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione derivatives
In order to improve the potency of the hit compound and explore its
structure-activity relationship (SAR), a series of derivatives of compound 1 were
synthesized. Scheme 1 shows the reaction route for compounds 3 and 4, which is
similar with literature [73]. Commercially available starting material benzyl bromide
and 3-chlorobenzyl bromide reacted with sodium azide to give benzyl azides 2
through a neutral nucleophilic substitution reaction. 1,3-Cycloaddition of azide 2 and
1,4-naphthoquinone generated final products 3 and 4 in yields of 20-30%.
1
1
1
1
1
1
1
17
Scheme 1. Synthesis of compounds 3 and 4.
7

1
2
3
Reagents and conditions: (i)NaN , DMSO, rt, overnight, 80-95%; (ii) EtOAc, reflux,
3
overnight, 20-30%.
4
5
6
7
8
9
0
The preparation of 1-phenyl-1H-benzo[d][1,2,3]triazole derivatives 7-33 were
undertaken following the synthetic route summarized in Scheme 2. Direct
.
diazotization of commercially available arylamine with tert-butyl nitrite and BF Et O
3
2
under low temperature afforded diazonium salts 5, which were subsequently
undergone azidation with sodium azide to yield aryl azides 6. [3+2] Cycloaddition of
the aryl azides 6 with 2-hydroxynaphthalene-1,4-dione provided the final compounds
7-33 with 20-80% yields.
1
1
1
Scheme 2. Synthesis of compounds 7-33.
8

O
N
N
NH2
Ar
i
N2+BF4-
Ar
ii
N3
Ar
iii
N
Ar
O
5
6
7-33
Cl
N
N
7
8
: Ar =
14: Ar =
15: Ar =
21: Ar =
28: Ar =
29: Ar =
N
H
N
N
H N
NO2
Cl
N
: Ar =
22: Ar =
N
H
Cl
F
F
Cl
Cl
N
9
1
: Ar =
16: Ar =
17: Ar =
23: Ar =
24: Ar =
30: Ar =
31: Ar =
CN
CF3
Cl
N
N
0: Ar =
NO2
Cl
Cl
1
1
1
1: Ar =
2: Ar =
3: Ar =
18: Ar =
19: Ar =
25: Ar =
26: Ar =
27: Ar =
32: Ar =
33: Ar =
Cl
Cl
F
F
COOMe
Cl
Cl
NHAc
NO2
Cl
N
2
SO NH2
COOH
20: Ar =
F
N
Cl
N
1
.
o
2
3
4
Reagents and conditions: (i) tert-Butyl nitrite, BF Et O, THF, -15 C, 1 h, 88-98%; (ii)
3
2
o
NaN , MeCN/H O, 0 C, 2 h, 80-90%; (iii) 2-Hydroxy-1,4-naphoquinone, DBU,
3
2
o
MeCN, 80 C, 12 h, 20-80%.
5
6
7
8
9
Compounds 36-41 were prepared according to the route outlined in Scheme 3.
Commercially available 3,4-dihydronaphthalen-1(2H)-one analogues were oxidized
using oxygen to generate 2-hydroxynaphthalene-1,4-dione analogues 35a-35f with
30-88% yields. Under the same conditions as shown in Scheme 2, 35a-35f were
converted to the desired derivatives 36-41 with 40-80% yields.
10
Scheme 2. Synthesis of compounds 36-41.
9

1
2
3
4
Reagents and conditions: (i) tert-BuOK, tert-BuOH, O , rt, 1-16 h, 30-88 %; (ii)
2
.
o
o
tert-Butyl nitrite, BF Et O, THF, -15 C, 1 h, 88-98 %; (iii) NaN , MeCN/H O, 0 C,
3
2
3
2
o
2 h, 80-90%; (iv) DBU, MeCN, 80 C, 12 h, 40-81%.
5
6
7
8
9
0
1
2
2.2. SAR analyses of 1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione derivatives
In order to study the SAR, enzymatic inhibitory activities of the synthesized
compounds against IDO1 and TDO were measured. For comparison, Epacadostat
and
1-(3-(4-methylpiperazine-1-carbonyl)benzyl)-3a,9a-dihydro-1H-naphtho[2,3-d][1,2,3]
triazole-4,9-dione (Cpd 3), were used as positive controls. The SAR analyses for the
synthesized compounds will mainly focus on two regions (see Figure 2): the terminal
phenyl group (region I) and the aromatic ring (region II).
1
1
1
13
2.2.1. Effect of the terminal phenyl group (region I)
1
1
1
4
5
6
To investigate the impact of different region I groups on the bioactivity, we fixed
region II as its original moiety and varied region I with various substituted aryl,
benzyl or heteroaryl groups. Twenty-nine compounds (3, 4, 7-33) were synthesized
1
0

1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
and evaluated for their inhibitory activity against IDO1 and TDO. Bioactivities of
these compounds are summarized in Table 1. Compounds 3 and 4 with benzyl or
substituted benzyl at N-1 of triazole ring showed remarkably reduced bioactivity
compared with 1 bearing a phenyl group. We then examined the influence of
monosubstituted phenyl groups. Compounds 7 and 8 with mete- and pare-chlorinated
phenyl groups respectively displayed an increased potency. In contrast,
1
ortho-chlorinated phenyl group in R (compound 9) led to a complete loss of
bioactivity against both IDO1 and TDO, which is probably due to the steric collision
at this site with the protein receptors. We thus used various other mete- or pare-
monosubstituted phenyl groups at N-1 of triazole ring, and the generated compounds
(10-21) all showed excellent bioactivity, except compound 20, which has a bulkyl
N-methyl piperazinyl-phenly group. Next, various mete- and pare- disubstituted or
trisubstituted phenyl groups were tested. The derived compounds 22-27 exhibited
considerable potency, and the most potent one corresponds to 22, which showed IC50
values of 13 nM and 14 nM against IDO1 and TDO, respectively. Finally, various
(substituted-) heteroaromatic rings (including monocycle and bicyclic) at N-1 of
triazole ring were examined. All the generated compounds 28-33 showed high
potency, but did not exceed 22.
1
1
1
1
1
1
1
1
1
19
Table 1. Structure and Biological Activities of Compounds 3, 4, and 7-33.
20
a
Cpd
R
IC50 (μM)
1
1

IDO1
TDO
0.070±0.006
>10
1
0.100±0.005
0
.075±0.004
Epacadostat
-
-
b
b
(
0.073)
(>10)
0
.658±0.031
0.036±0.005
Cpd 3
3
c
c
(
0.640±0.013)
(0.030±0.002)
0.886±0.056
0.858±0.036
0.769±0.020
0.712±0.012
4
c
c
(
0.767±0.026)
0.040±0.001
(0.711±0.026)
Cl
7
8
0.036±0.005
0.015±0.001
0.023±0.004
>10
9
>10
1
0
1
0.052±0.008
0.058±0.008
0.101±0.009
0.049±0.011
1
12
0.081±0.012
0.132±0.007
13
14
15
0.114±0.008
0.122±0.005
0.029±0.003
0.071±0.006
0.064±0.010
0.028±0.003
1
2

1
6
7
0.026±0.005
0.033±0.006
0.066±0.005
0.048±0.004
1
1
8
9
0.034±0.001
0.030±0.007
0.058±0.003
0.048±0.013
1
20
>10
>10
21
22
23
24
25
26
27
0.060±0.010
0.013±0.002
0.030±0.003
0.023±0.004
0.033±0.006
0.015±0.002
0.020±0.003
0.025±0.007
0.014±0.003
0.025±0.010
0.018±0.002
0.034±0.002
0.015±0.002
0.018±0.002
1
3

28
29
30
31
0.078±0.014
0.045±0.004
0.040±0.008
0.023±0.002
0.104±0.022
0.042±0.009
0.034±0.009
0.021±0.008
3
2
3
0.113±0.015
0.027±0.003
0.248±0.024
0.025±0.001
3
a
1
2
3
The data are averages of three independent experiments and reported as the mean
b
c
values ± SD. Data for Epacadostat from ref. 32. Data for compound Cpd 3 from
c
ref. 73. Data for compound 4 from ref. 73.
4
2.2.2. Influence of the aromatic ring (Region II)
5
6
7
8
9
0
1
2
3
4
5
We then fixed 3-chloro-4-fluorophenyl as the optimal group obtained in the last
step and optimized substitution in region II. Firstly, hydroxyl, methoxyl and fluorine
were
introduced
to
the
6-position
of
1-(3-chloro-4-fluorophenyl)-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione scaffold to
search for more potent IDO1/TDO dual inhibitor. Compounds 36 and 37 with
6-hydroxyl and 6-methoxyl in aromatic ring showed a decreased potency compared
with 22. Interestingly, compound 38 with a fluorine substituent displayed an
obviously increased activity (IDO1: 5 nM; TDO: 4 nM) compared to compound 22
(IDO1: 13 nM; TDO: 14 nM) . Then, the same groups were applied to the 7-postion
1
1
1
1
1
1
of
the
scaffold
1-(3-chloro-4-fluorophenyl)-1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione. Relative to
1
4

1
2
3
the corresponding compounds with 6-position substation, the 7-position substation
counterparts exhibited a weaker potency (39 vs 36; 40 vs 37; 41 vs 38). Compound 40
with 7-methoxyl even almost lost its activity (IC > 10 μM).
5
0
4
Table 2. Structure and Biological Activities of Compounds 36-41.
5
a
IC (μM)
5
0
1
Cpd.
R
IDO1
TDO
3
3
6
7
6-OH
0.144±0.020
0.163±0.020
0.171±0.010
6-OMe
0.105±0.020
3
3
4
4
8
9
0
1
6-F
7-OH
7-OMe
7-F
0.005±0.003
0.355±0.107
>10
0.004±0.001
0.384±0.1110
>10
0.010±0.002
0.010±0.003
6
7
a
The data are averages of three independent experiments and reported as the mean
8
values ± SD.
9
0
1
In the above structural optimization and SAR analyses, we obtained a potent
1
1
compound, 38, which showed IC50 values of 5 nM and 4 nM against IDO1 and TDO,
respectively. As far as we know, this compound is the most active one among all the
1
5

1
2
3
4
5
6
7
8
9
0
1
reported dual IDO1/TDO inhibitors. One may notice that compounds containing
quinones substructures have been reported to be IDO1 inhibitors.[17,36,39,62]
Compared with the previously reported quinones for IDO1 inhibitors, compound 38
has an additional triazole ring, which has been demonstrated to benefit to eliminate
the
strong
oxidability
of
quinones.[73]
On
the
other
hand,
1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione derivatives have also been reported to be
selective TDO inhibitors.[73] Compound 38 is different from these selective TDO
inhibitors in the N-1 substitution of 1H-naphtho[2,3-d][1,2,3]triazole-4,9-dione
(phenyl in compound 38 vs benzyl in the TDO inhibitors). In the following studies,
we will perform further evaluations to compound 38, including PAINS-liable property,
effects on IDO1 and TDO in intact cells, and in vivo anti-tumor activity.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
3
4
5
6
7
8
9
0
1
2
3
4
5
2.3. Evaluation of PAINS-Liable Property.
Compounds containing a quinones substructure are often thought as possible pan
assay interference compounds (PAINS) [86]. To investigate the PAINS-liable
properties of compound 38, we performed several experiments. Firstly, we evaluated
the radical scavenging ability of compound 38 using the diphenylpicrylhydrazyl
(DPPH) assay, which is for revealing whether the IDO1/TDO inhibition is due to its
antioxidant activity. As shown in Figure 3A, compound 38 did not reduce DPPH
radical, indicating no reducibility. Secondly, a horseradish peroxidase-phenol red
(HRP-PR) H O detection assay [87,89] was designed to evaluate the oxidizing ability
2
2
of compound 38. When incubation of 50 μM compound 38 with 800 μM dithiothreitol
(DTT), 38 produced 12 % H O with an activation concentration (AC) of >50 μM
2
2
(Figure 3B), indicating that 38 is not a redox cycling compound. Thirdly, we carried
out an iron ion chelation assay to explore whether inhibition of IDO1 or TDO
activities was due to its metal-chelation effect. The result, displayed in Figure 3C,
1
6

1
2
3
4
5
showed that compound 38 did not have metal-chelation ability. Fourthly,
3-(N-Maleimidylpropionyl)biocytin (MPB) labeling [89] and Western blotting assays
were performed to investigate whether compound 38 is a covalent modifier. As
displayed in Figure 3D, the tested compound 38 did not modify any cysteine residues
of IDO1 or TDO, implying not a covalent modifier.
6
7
8
9
0
1
2
3
4
5
6
Finally, to examine whether our compound exhibited promiscuous bioactivity
profiles, a panel of 422 human protein kinase inhibition assay was performed with
compound 38 by the Eurofins kinase profiling service. As shown in Table S1,
compound 38 did not display activity against all the tested kinases (IC > 10 μM). In
5
0
1
1
1
1
1
1
1
addition, we also tested the bioactivities of compound 38 against other biological
targets, including N6-adenosine-methyltransferase (METTL3), Jumonji
domain-containing protein 6 (JMJD6), NOP2/Sun domain family member 2 (NSUN2),
and Histone acetyltransferase 1(HAT1); these biological targets were chosen because
their bioactivity assays are available in our lab. Compound 38 exhibited weak or no
activity in these assays (Table S2). Collectively, all the above experiments clearly
demonstrated that compound 38 is unlikely a PAINS compound.
1
7

1
2
3
4
5
6
7
8
9
0
1
2
3
Figure 3. Experimental results for clarifying the PAINS-liable properties of
compound 38. (A) Reduction of the DPPH radical by 38 and ascorbic acid (positive
control). The used concentration of DPPH was 50 μM. Data are the means ± SEM,
from triplicates. (B) Incubation of 38 (50 μM) with DTT (800 μM) produced 12 % of
H O with AC >50 μM, indicated that 38 was not a RCC. Data are the means ±SEM,
2
2
from triplicates. (P value *: * P < 0.05: ** P < 0.01: *** P < 0.001, by Dunnett’s test).
(C) 38 did not have chelation effect. 10 % Sodium citrate, an iron chelating agent, was
selected as positive control. Data are the means ±SEM, from triplicates. (P value *: *
P < 0.05: ** P < 0.01: *** P < 0.001, by Dunnett’s test). (D) 38 did not inhibit hIDO1
or hTDO by reactivity toward cysteine residues through Michael addition. The
selenazal drug Ebselen potently inhibited hIDO1 by targeting enzyme cysteine
residues and was selected as positive control.
1
1
1
1
1
4
5
2.4. Effects of 38 on IDO1 and TDO in intact cells
1
In the above structural optimization and SAR analyses, we obtained a series of
1
8

1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
potent IDO1/TDO dual inhibitors and compound 38 was the most potent one. Here we
explored the effects of 38 on IDO1 and TDO in intact cells. To this end, two cell lines,
HeLa and A172, were chosen. HeLa cells can overexpress IDO1 upon stimulation by
IFN-γ, and A172 cells have a high level of TDO expression. Though the two cell lines
might not be specific tools, they have been routinely used for the cellular IDO1 and
TDO inhibition assays, respectively. As shown in Figure 4A, 38 significantly and
dose-dependently inhibited the production of kynurenine in both the cellular IDO1
(HeLa cells stimulated by IFN-γ) and TDO (A172 cells) assays. The calculated IC50
values are 98 nM and 59 nM for IDO1 and TDO, respectively. Meanwhile, the
viability of HeLa and A172 cells did not change when treated with compound 38 at
0.1 μM, 1 μM and 10 μM (Figure 4B and 4C). In addition, 38 had no effect on the
expressions of IDO1 and TDO proteins (Figure 4D). Overall, the results indicated that
the reduction of kynurenine level was not due to the cytotoxicity of 38, and the
inhibitory activity of compound 38 was not due to the down-regulation of IDO1 and
TDO expression.
1
1
1
1
1
1
1
9

1
2
3
4
5
6
Figure 4. Effects of compound 38 on IDO1 and TDO in intact cells. (A)
38 dose-dependently inhibits IDO1 activity in INF-γ treated HeLa cells, and TDO
activity in A172 cells. (B) Effects of compound 38 (10 μM, 1 μM and 0.1 μM) on
viability of HeLa cells and production of kynurenine. DMSO was selected as control.
Data are the means ±SEM, from triplicates. (P value *: * P < 0.05: ** P < 0.01: ***
7
8
9
P < 0.001 , by Dunnett’s test) (C) Effects of compound 38 (10 μM, 1 μM and 0.1 μM)
on viability of A172 cells and production of kynurenine. DMSO was selected as
control. Data are the means ±SEM, from triplicates. (P value *: * P < 0.05: ** P <
1
1
0
1
0.01: *** P < 0.001 , by Dunnett’s test) (D) Western blot showed that compound 38
treatment did not impact the expression of IDO1 and TDO.
1
1
1
2
3
4
2.5. In vivo anti-tumor activity of 38 in LL2 and Hepa1-6 allograft models
In vivo anti-tumor activity of 38 was first evaluated on the LL2 allograft mouse
model; LL2 cells bear a high level of IDO1 expression. In this experiment, the
2
0

1
2
3
4
5
6
7
8
9
0
1
2
3
4
selective IDO1 inhibitor Epacdostat was chosen as a positive control. When the LL2
3
tumor grew to a mean volume of around 200 mm , the mice were intraperitoneally
treated with vehicle, 30 mg/kg/day of 38 and 30 mg/kg/day of Epacdostat for 11 days;
38 and Epacdostat were dissolved in soybean oils. As displayed in Figure 5, 38
showed moderate anti-tumor activity with a tumor growth inhibitory (TGI) rate of
62.5%, which was comparable with Epacdostat. Then another in vivo anti-tumor
experiment was conducted with the Hepa1-6 allograft mouse model; Hepa1-6 cells
harbor a high level of both IDO1 and TDO. In this experiment, 38 displayed more
potent anti-tumor activity than Epacdostat (TGI: 80% vs 50%), suggesting an
additive effect of dual target inhibition. It is also necessary to point out that the
compound with <10 nM inhibitor efficiency does not properly reflect its in vivo data.
One possible explanation is that this compound has a poor water solubility (0.16 µM),
and hence a poor bioavailability. In both experiment, no obvious change was found
for the body weights of mice, implying low toxicity.
1
1
1
1
1
15
2
1

1
2
Figure 5. Antitumor activity of 38 in LL2 and Hepa1-6 allograft models.
3. Conclusion
3
4
5
6
7
8
9
0
1
2
3
4
5
In the present study, structural optimization towards hit compound 1 led to the
discovery of a series of potent IDO1 and TDO dual inhibitors. Among these
compounds, compound 38 displayed the highest potency in enzymatic inhibition
activity against both IDO1 and TDO with IC values of 5 nM and 4 nM, respectively.
5
0
In cellular IDO1 and TDO inhibition assays, compound 38 also showed potent
activities with IC values of 98 nM and 59 nM, respectively. In in vivo assays, it
5
0
exhibited considerable anti-tumor activity in LL2 and Hepa1-6 mice allograft tumor
models. To the best of our knowledge, compound 38 is the most potent dual
IDO1/TDO inhibitor at the moment. Nevertheless, we have to mention that this
compound might not be a good drug candidate because its poor water solubility (0.16
µM), and hence poor oral bioavailability, and a further structural optimization is
needed. Overall, this study provides a good starting point for subsequent drug
discovery targeting IDO1/TDO.
1
1
1
1
1
1
16
4. Experimental Section
1
1
1
2
2
2
2
7
8
9
0
1
2
3
4.1. IDO1 enzymatic assay
The hIDO1 enzymatic assay was performed as described previously [24]. Briefly,
a standard reaction mixture (0.5 mL) containing 50 mM potassium phosphate buffer
(pH 6.5), 40 mM ascorbic acid, 20 μM methylene blue, 200 μg/mL catalase, 400 μM
L-tryptophan and (4-10 μg) hIDO1 was added to the solution containing the test
sample at a determined concentration. The reaction carried out at 37 °C for 30 min
and stopped by adding 200 μL of 30% (w/v) trichloroacetic acid. After heating at
2
2

1
2
3
4
5
6
7
65 °C for 15 min, the reaction mixture was centrifuged at 12 000 rpm for 10 min. The
supernatant (100 μL) was transferred into a well of a 96-well microplate and mixed
with 100 μL of 2% (w/v) p-dimethylaminobenzaldehyde in acetic acid. The yellow
pigment derived from kynurenine was recorded by measuring absorbance 492 nm
using a SpectraMax250 microplate reader (Molecular Devices, USA). Enzymatic IC50
values were calculated by using GraphPad Prism (version 6.07, from GraphPad
Software Inc. MA).
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4.2. TDO enzymatic assay
The hTDO enzymatic assay was carried out as described previously [76].
Briefly, a standard reaction mixture (0.5 mL) containing 50 mM potassium phosphate
buffer (pH 7.0), 40 mM ascorbic acid, 20 μM methylene blue, 200 μg/mL catalase,
1
1
1
1
1
1
1
1
1
1
2
2
400 μM tryptophan and (4-10 μg) TDO was added to the solution containing the test
L-
sample at a determined concentration. The reaction was undergone at 37 °C for 30
min and stopped by adding 200 μL of 30% (w/v) trichloroacetic acid. After heating at
65 °C for 15 min, the reaction mixture was centrifuged at 12 000 rpm for 10 min. The
supernatant (100 μL) was transferred into a well of a 96-well microplate and mixed
with 100 μL of 2% (w/v) p-dimethylaminobenzaldehyde in acetic acid. The yellow
pigment derived from kynurenine was recorded by measuring absorbance at 492 nm
using a SpectraMax250 microplate reader (Molecular Devices, USA). Enzymatic IC50
values were calculated by using GraphPad Prism (version 6.07, from GraphPad
Software Inc. MA).
2
2
2
2
3
4
4.3. IDO1 cellular assay.
In vitro cellular IDO1 inhibition assay was performed as described previously
[24]. HeLa cells were cultured in DMEM with 10% FBS, 1% pen/strep, 1%
2
3

1
2
3
4
5
6
7
8
9
0
1
2
3
nonessential amino acid. Cells were seeded in 96-well microplates at a density of 5
000 per well. On the next day, human IFN-γ (100 ng/mL) and compound in a total
volume of 200 μL culture medium containing 15 μL/mL of tryptophan were added
L-
to the cells. After incubation at 37 °C for 24 h, 140 μL of medium was removed into a
new 96-well plate and 10 μL of 6.1 N trichloroactic acid was added. The plate was
incubated for 30 min at 50 °C. The plate was then centrifuged at 25 000 rpm for 10
min to remove sediments. 100 μL of the supernatant per was transferred to another
96-well plate and mixed with 100 μL of 2% (w/v) p-dimethylaminobenzaldehyde in
acetic acid. The plate was incubated at rt for 10 min. The yellow pigment derived
from kynurenine was recorded by measuring absorbance at 492 nm using a
SpectraMax250 microplate reader (Molecular Devices, USA). Cellular IC50 values
were calculated by using GraphPad Prism (version 6.07, from GraphPad Software Inc.
MA).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
4
5
6
7
8
9
0
1
2
3
4
5
4.4. TDO cellular assay.
In vitro cellular TDO inhibition assay was performed as described previously
[90]. A172 cells were cultured in RPMI 1640 containing 10% (v/v) BCS. The
compounds were dissolved in DMSO and diluted in RPMI 1640. The assay was
performed in 96-well flat bottom plates seeded with 5 000 cells in a final volume of
200 μL. On the next day, a series dilution of the tested compounds containing 15
μg/mL of tryptophan in a total volume of 200 μL culture medium were added to the
L-
cells. After incubation for 24 h, 140 μL of the supernatant was mixed with 10 μL of
6.1 N trichloroacetic acid and the mixture was incubated for 30 min at 50 °C. The
reaction mixture was then centrifuged for 10 min at 4000 rpm to remove sediments.
100 μL of the supernatant was mixed with 100 μL of 2% (w/v)
p-dimethylaminobenzaldehyde in acetic acid. The plate was incubated at rt for 10
2
4

1
2
3
4
min. The yellow color derived from kynurenine was recorded by measured at 492 nm
using a SpectraMax250 microplate reader (Molecular Devices, USA). Cellular IC50
values were calculated by using GraphPad Prism (version 6.07, from GraphPad
Software Inc. MA).
5
6
7
8
9
0
1
4.5. Diphenylpicrylhydrazyl (DPPH) assay
The antioxidant activities of compound 38 was assessed as described previously
[24]. An amount of 100 μL of 5, 10, 20, 40, 80, or 160 μM 38 in DMSO was added to
100 μL of a solution of the stable free radical DPPH in ethanol (100 μM), buffered
with acetate to pH 5.5, in a 96-well plate. The absorbance was recorded at 517 nm
after a 30 min incubation under gentle shaking in the dark. Ascorbic acid was used as
reference compound. All analyses were performed in triplicate.
1
1
1
1
1
1
1
1
1
1
2
2
3
4
5
6
7
8
9
0
4. 6. Horseradish peroxidase-phenol red (HRP-PR) H O detection assay
2
2
The horseradish peroxidase-phenol red (HRP-PR) H O detection assay was
2
2
performed by measured H O generated in the reaction using catalase activity
2
2
colorimetric/fluorometric assay kit (BioVision, USA) according to manufacturer’s
instructions and Soares et al. [88] with minor modification. Briefly, compound 38 at
50 μM were incubated with 0.8 mM dithiothreitol for 15 min at room temperature.
The reaction was terminated by the stop solution, and a developer mix was then added
to the mixture for 10 min. The signal was measured by the spectrophotometer at the
absorbance of 560 nm.
2
2
2
2
1
2
3
4
4. 7. Chelation detection assay
The chelation of 38 was detected by the reaction of phenanthridazine with
ferrous iron [91]. Ferrozine can form a colored complex with ferrous iron with an
absorption peak at 562 nm. The ferrozine and FeSO were dissolved in purified water
4
2
5

1
2
3
4
5
6
7
to a dose of 5 mM and 1mM, respectively. Test compounds were dissolved into 10
mM in DMSO. Then 1 μL solution of test compounds, 5 μL FeSO solution and 20
4
μL ferrozine solution were added to 100 μL purified water, shaking well for five
minutes after mixing at room temperature. Finally, the absorbance at 562 nm of
samples was detected using the Multiskan MK3 ELISA photometer (Thermo
Scientific, USA). All values were normalized to background (DMSO and purified
water). The experiment was repeated twice.
8
9
0
1
2
3
4
4. 8. Chemical labeling of IDO1 and TDO cysteine residues
To monitor reactive cysteine residues present on hIDO1 and TDO, we employed
the thiol alkylating agent 3-(N-maleimidylpropionyl)biocytin (MPB) [87]. For this,
native or ebselen-treated hIDO1 or hTDO (10 μM) was incubated with 1 mM MPB
for 30 min prior to addition of SDS loading buffer. Protein samples were then
resolved, transferred, and probed with anti-streptavidin (Invitrogen) using Western
blotting to assess MPB labeling.
1
1
1
1
1
1
1
1
1
1
2
2
2
5
6
7
8
9
0
1
2
4. 9. MTT assay of cell viability
HeLa and A172 cells were seeded in 96-well culture plates at a density of 5 000
per well. After overnight incubation, solution of compounds in DMSO (200 μM) was
added to the cells. Cells treated with DMSO 1% were used as negative control. After
further incubation for 24 h, 20 μL MTT solution (5 mg/mL in PBS) was added to each
well. After 4 h of incubation at 37 °C, MTT solution was removed and 150 μL of
DMSO was added to dissolve the crystals formed. Then, absorbance at 570 nm was
read using a microplate reader.
2
2
3
4
4. 10. Western blot analysis of the expression of IDO1 and TDO
Western blot assay was performed to analyze the expression of IDO1 and TDO
2
6

1
2
3
4
5
6
7
8
9
using a standard procedure. Anti-TDO (Proteintech, Cat. No. 15880-1-AP) and
anti-IDO1 (CST, Cat. No. 68572) antibodies were added in a primary antibody
dilution buffer (Beyotime) and allowed to incubate overnight at 4 °C. Then, the
membranes were washed with PBST (PBS containing 3‰ Tween-20) 3 times (5 min,
10 min, 15 min) before the secondary antibodies (Beyotime, goat antirabbit second
antibody, goat anti-mouse second antibody) were added and the incubation continued
at room temperature for 1 h. Then the membranes were washed with PBST 3 times
(10 min each time). High-signal ECL Western blotting substrate (Tanon) was added
and the membrane was immediately imaged using a chemi scope (clinx).
10
4. 11. In vitro kinase activity assays.
1
1
1
1
2
3
The in vitro kinase activity assays were conducted by the Kinase Profiling
Services provided by Eurofins (Eurofins, UK) and the protocols can be found in
http://www.eurofins.com/pharmadiscovery.
14
4. 12. In vivo models.
1
1
1
1
1
2
2
2
5
6
7
8
9
0
1
2
The 6- to 8-week-old C57BL/6 rats used for the study were purchased from
Chendu Dossy Experimental Aniamls CO., Ltd. All animals were maintained in a
specific pathogen-free facility. The procedures related to animal handling, care, and
treatment in this article were performed in compliance with Agreement of the
Experimental Animal Management Committee of Sichuan University. All animal
experiments were carried out in accordance with the National Institutes of Health
guide for the care and use of laboratory animals. LLC tumor-bearing mice and
Hepa1-6 tumor-bearing mice were constructed as described previously with some
2
7

1
2
3
4
5
6
7
8
9
modifications [24]. LLC and Hepa1-6 cells were harvested during exponential growth
6
and washed 3 times with serum-free medium. Approximately 2 × 10 cells were
implanted subcutaneously into the right flank in a total volume of 150 μL/mice. Mice
were randomized into vehicle and treated groups with a group mean tumor size of
3
150-200 mm . For efficacy studies, mice were dosed once daily intraperitoneally with
either vehicle or test compounds using the indicated doses for a maximum of 20 days.
Mice were monitored for side effects every day. Tumor size and body weight were
measured every 3 or 4 days. The volume was calculated as follows: tumor size = a ×
2
b /2 (a, long diameter; b, short diameter).
10
11
12
13
14
4. 13. Statistical analysis.
GraphPad Prism (version 6.07, from GraphPad Software Inc) was used for
statistical analysis. The statistical significance of results in all of the experiments was
determined by the Student t test and ANOVA. P values < 0.05 were considered
statistically significant.
1
1
1
1
1
2
2
2
2
5
6
7
8
9
0
1
2
3
4. 14. Chemistry
All the chemical solvents and reagents were obtained from commercial suppliers
and used without further purification. All reactions were monitored by TLC.
1
13
Chromatography was performed using silica gel (300-400 mesh). H NMR and C
NMR spectra were recorded on a Bruker AV-400 spectrometer at 400 MHz and 100
MHz, respectively. Coupling constants (J) are expressed in hertz (Hz). Chemical
shifts (δ) of NMR are reported in parts per million (ppm) units. Low resolution
ESI-MS were recorded on Agilent 1200-G6410A mass spectrometer. HRMS were
recorded on Agilent 1200-G6210 TOF mass spectrometer. HPLC analysis was
2
8