Design, synthesis, X-ray crystallographic analysis, and biological evaluation of thiazole derivatives as potent and selective inhibitors of human dihydroorotate dehydrogenase

Article abstract of DOI:10.1021/jm501127s

Human dihydroorotate dehydrogenase (HsDHODH) is a flavin-dependent mitochondrial enzyme that has been certified as a potential therapeutic target for the treatment of rheumatoid arthritis and other autoimmune diseases. On the basis of lead compound 4, which was previously identified as potential HsDHODH inhibitor, a novel series of thiazole derivatives were designed and synthesized. The X-ray complex structures of the promising analogues 12 and 33 confirmed that these inhibitors bind at the putative ubiquinone binding tunnel and guided us to explore more potent inhibitors, such as compounds 44, 46, and 47 which showed double digit nanomolar activities of 26, 18, and 29 nM, respectively. Moreover, 44 presented considerable anti-inflammation effect in vivo and significantly alleviated foot swelling in a dose-dependent manner, which disclosed that thiazole-scaffold analogues can be developed into the drug candidates for the treatment of rheumatoid arthritis by suppressing the bioactivity of HsDHODH.

Full text of DOI:10.1021/jm501127s

Subscriber access provided by Northern Illinois University
Article
Design, Synthesis, X-ray Crystallographic Analysis, and
Biological Evaluation of Thiazole-Derivatives as Potent and
Selective Inhibitors of Human Dihydroorotate Dehydrogenase
Junsheng Zhu, Le Han, Yanyan Diao, Xiaoli Ren, Minghao Xu, Liu-Xin Xu, shiliang li, Qiang Li, Dong Dong,
Jin Huang, Xiaofeng Liu, zhenjiang zhao, Rui Wang, Lili Zhu, Yufang Xu, Xuhong Qian, and Honglin Li
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501127s • Publication Date (Web): 12 Jan 2015
Downloaded from http://pubs.acs.org on January 24, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
Design, Synthesis, Xꢀray Crystallographic Analysis, and
Biological Evaluation of ThiazoleꢀDerivatives as Potent
and Selective Inhibitors of Human Dihydroorotate
Dehydrogenase
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
5
1
,†
1,2,†
1,†
1
1
1
Junsheng Zhu , Le Han ,Yanyan Diao , Xiaoli Ren , Minghao Xu , Liuxin Xu ,
1
1
1
1
1
1
6
7
8
9
0
1
Shiliang Li , Qiang Li , Dong Dong , Jin Huang , Xiaofeng Liu , Zhenjiang Zhao , Rui
1
1,*
1
2,*
1,*
Wang , Lili Zhu ,Yufang Xu , Xuhong Qian , Honglin Li
1
State Key Laboratory of Bioreactor Engineering, Shanghai Key Laboratory of New
2
Drug Design, Shanghai Key Laboratory of Chemical Biology
,
School of Pharmacy,
1
1
East China University of Science and Technology, Shanghai 200237, China.
†
12
13
Authors contributed equally to this work
* To whom correspondence should be addressed. Email: zhulfl@ecust.edu.cn,
xhqian@ecust.edu.cn, hlli@ecust.edu.cn
1
1
1
1
1
4
5
6
7
8
Please address correspondence and requests for reprints to:
Prof. Honglin Li
School of Pharmacy, East China University of Science and Technology
ACS Paragon Plus Environment

Page 3 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
130 Mei Long Road, Shanghai 200237
Phone/Fax: +86ꢀ21ꢀ64250213
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
Abstract. Human Dihydroorotate dehydrogenase (HsDHODH) is a flavinꢀdependent
mitochondrial enzyme, which has been certified as a potential therapeutic target for the
treatment of rheumatoid arthritis and other autoimmune diseases. Based on lead
compound 4, which was previously identified as potential HsDHODH inhibitor, a novel
series of thiazole derivatives were designed and synthesized. The Xꢀray complex
structures of the promising analogues 12 and 33 confirmed that these inhibitors bind at
the putative ubiquinone binding tunnel and guided us to explore more potent inhibitors,
such as compounds 44, 46 and 47 which showed double digit nanomolar activities of 26
nM, 18 nM and 29 nM, respectively. Moreover, 44 presented considerable
antiꢀinflammation effect in vivo and significantly alleviated foot swelling in a
doseꢀdependent manner, which disclosed that thiazoleꢀscaffold analogues can be
possibly developed into the drug candidates for the treatment of rheumatoid arthritis by
suppressing the bioactivity of HsDHODH.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
12
13
1
1
1
1
1
4
5
6
7
8
Keywords: Human dihydroorotate dehydrogenase (HsDHODH), thiazole derivatives,
structureꢀactivity relationship (SAR).
ACS Paragon Plus Environment

Page 5 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
Abbreviations: DHOase, dihydroorotase dihydroorotate; DHODH, dehydrogenase;
HsDHODH, Human dihydroorotate dehydrogenase; PfDHODH, Plasmodium
falciparum dihydroorotate dehydrogenase; DHO, dihydroorotate; ORO, Orotate; CoQ,
ubiquinone; FMN, flavin mononucleotide; RA, rheumatoid arthritis; SAR,
structureꢀactivity relationship; CIA, collagenꢀinduced arthritis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
Introduction
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
Pyrimidines serve essential functions in biosynthesis of DNA, RNA, and are linked by
1
4
5
6
7
8
9
0
1
phosphodiester bridges to purine nucleotides in doubleꢀstrand DNA. Dihydroorotate
dehydrogenase (DHODH), as the fourth enzyme of pyrimidine de novo synthesis,
catalyzes the transformation of dihydroorotate (DHO) to orotate (ORO) with the
participation of coꢀfactors flavin mononucleotide (FMN) and ubiquinone (CoQ),
2
representing the rate limiting step in pyrimidine biosynthesis. DHODHs are divided
into two classes in terms of cellular localization, amino acid sequence, and
3
1
1
substrate/cofactor dependence. Human dihydrooroate dehydrogenase (HsDHODH)
belongs to the family class 2 enzymes, characterized as a membraneꢀassociated protein
4
ꢀ6
12
13
in eukaryotes that utilizes respiratory quinones as terminal electron acceptors, which
+
4, 5, 7
is unlike the family class 1 utilizing fumarate or NAD in prokaryotes instead.
The
1
1
1
1
1
4
5
6
7
8
location of HsDHODH at the inner mitochondrial leaflet is functionally connected to
the respiratory chain, ensuring the catalysis proceeds efficiently.
Generally, humans obtain pyrimidine by recycling existing ones in most cells,
1
restricting the demand for the energetic de novo biosynthesis. However, the gap
ACS Paragon Plus Environment

Page 7 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
between pyrimidine supply and demand in rapidly proliferating cells including tumor
cells and the activated Tꢀlymphocytes, Bꢀlymphocytes, suggests their requirements on
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
both de novo and salvage pathways. The metabolism and multiplication modulated by
DHODH determines it being a promising target for the development of new drug
2
candidate. The primary function of the mitochondrial electron transport chain further
demonstrates the significance of developing new chemotherapeutic agents against
HsDHODH. Controlling HsDHODH activities has been corroborated beneficial for the
9ꢀ11
treatment of various diseases such as rheumatoid arthritis (RA), cancer and sclerosis.
Leflunomide (1) and brequinar (3) are two representative DHODH inhibitors (Figure 1).
The former, a prodrug of its metabolite A771726 (2) (Figure 1), was implemented in
1
1
8, 12ꢀ14
treatment of RA in clinical trial.
While the latter was initially applied in prevention
12
13
of organ transplant rejection therapy and then regarded as antitumor and
15
immunosuppressive agent in phase II clinical treatment. However, the occurrence of
severe side effects on both molecules limited their application. Longꢀterm use of
leflunomide would lead to reversible alopecia, hypertension, rash, diarrhea and
1
1
1
1
1
4
5
6
7
8
14, 16ꢀ18
abnormalities in liver enzymes.
Oral administration of brequinar results in toxic
19ꢀ21
effects when given in combination with cyclosporine or cisꢀplatin.
Although
exponential growth of researches was witnessed to explore the therapeutic potential of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2ꢀ30
1
2
3
4
5
6
7
8
9
0
1
HsDHODH inhibitors in the past decades,
there are not successful cases reported yet
2
with promising results in clinical trials. As the therapeutic result, developing new
chemotherapeutic agents targeting HsDHODH remains a compelling therapeutic goal.
We recently discovered several HsDHODH inhibitors with different chemotypes by
structureꢀbased virtual screening, of which compound 4 (Figure 1) was selected as a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
31
promising structure for further optimization. The initial structural modification of
compound 4 produced compound 12 with the inhibitory activity improved by about 6
folds. The crystal complex structure of 12 and HsDHODH was solved to confirm the
putative binding poses of the thiazoleꢀscaffold analogues in the ubiquinone binding
tunnel, which led to further structureꢀbased modification to improve the in vitro
inhibitory activities. The most potent compounds achieved doubleꢀdigit nanomolar
activities and displayed the significance of antiꢀinflammation effect, suggesting their
prospective roles as immunosuppressive and antiproliferative agents.
1
1
12
13
1
4
15 Results and Discussion
1
1
1
6
7
8
Chemistry
The synthesis route of compounds used in this study is shown in Scheme 1. Reaction of
3ꢀchloropentaneꢀ2,4ꢀdione or ethyl 2ꢀchloroꢀ3ꢀoxobutanoate with derived thioureas,
ACS Paragon Plus Environment

Page 9 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
which were prepared in a facile threeꢀstep sequence using previously described
3
2ꢀ34
methods,
gave 4ꢀ13 or 14ꢀ18 in good to excellent yields. Isothiocyanates were
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
served as starting reactants to achieve compounds 19ꢀ20, which also facilitated the
synthesis of compounds 21ꢀ24 in alkaline conditions with the corresponding acyl
chloride. Alternatively, to generate compounds 25ꢀ56, we adopted oneꢀpot strategies by
the
treatment
of
ethyl
3ꢀcyclopropylꢀ3ꢀoxopropanoate
or
ethyl
4,4ꢀdimethylꢀ3ꢀoxopentanoate or ethyl 3ꢀoxoꢀ3ꢀphenylpropanoate with NBS in water in
the presence of βꢀcyclodextrin, providing bromineꢀsubstituted βꢀketo esters as
intermediates, which then undergo cyclization with substituted thioureas similarly as
compounds 4ꢀ13 and 14ꢀ18.
1
1
12
13
For compounds 49ꢀ53 and 54ꢀ56, it was decided to synthesize the relevant aryl amine
35
precursors 49cꢀ53c and 54bꢀ56b (Schemes 2, 3). Compounds 49cꢀ53c were achieved
under a palladium catalyzed crossꢀcoupling reaction, which were conducted in a mixed
solvent of water, ethanol and benzene, using appropriate phenylboronic acid, aniline,
potassium carbonate and Pd(PPh ) . Compounds 54bꢀ56b were provided via
1
1
1
1
1
4
5
6
7
8
3
4
etherification, using proper aromatic substituted compounds and requisite aromatic
alcohol or phenol under alkaline conditions in acetonitrile. Then the Feꢀmediated
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
reduction of nitro group to amino group was carried out in a solution of ammonium
chloride and ethanol, which give the precursors 54bꢀ56b.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
Preliminary Hydrophobic Group Modification on R
The lead compound 4 with a thiazole scaffold, a ubiquinoneꢀbinding competitive
inhibitor of HsDHODH, has recently been identified as a potential HsDHODH inhibitor
with an IC value of 3.937 ꢁM (Table 1) through structureꢀbased virtual screening in
5
0
31
our previous work. Compared with the reported DHODH inhibitors (Compounds
2,3,22ꢀ30
1ꢀ3),
the compound 4 shared similar “thiazole moiety” as hydrogen bondꢀforming
3
1
1
“head” linked via a nitrogen to an aromatic group as hydrophobic “tail”. According to
the predicted binding pose in our previous work, the thiazole moiety of compound 4
located at the S2ꢀS4 subsites tethered by several hydrogen bonds with the polar residues
12
13
(Arg136, Gln47) in this hydrophilic region, and the 4ꢀCH ꢀ3ꢀCl substituted phenyl ring
3
31
1
1
1
1
1
4
5
6
7
8
then partially occupied the large hydrophobic region in S1 subsite
,
which inspired us
to increase the size and hydrophobicity of this moiety to occupy the rest of S1 subsite
for leading improved inhibitory activity.
To identify more potent compounds, a systematic modification was performed by
ACS Paragon Plus Environment

Page 11 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
dividing the lead compound 4 into three fragments characterized by different
1
2
pharmacophore features (Table 1): the sections of acetyl group (R ), methyl group (R )
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
and the aromatic motif (R , the referred “tail group”). Through the above analysis
against the large hydrophobic S1 subsite, various substituents with some larger and
3
more hydrophobic groups were introduced on R as shown in Table 1. Unsubstituted or
single pꢀposition substituent (5, 6, and 7) did not show improved activities or deplete
the activities comparing with lead compound 4. Disubstituents at pꢀ and mꢀ position
such as 3,4ꢀdiꢀCH (8), 3ꢀCF ꢀ4ꢀCl (10) and 3ꢀCF ꢀ4ꢀBr (11) showed a comparative
3
3
3
3
potency to the original lead compound (4), as their R section have similar
1
1
hydrophobicity except 3ꢀFꢀ4ꢀCH (9) which knocked down the activity probably
3
because of distinctive hydrophilic and electrostatic features of fluorine atom against
3
12
13
other halogens. Augmenting the size and hydrophobicity of R substituent with naphthyl
moiety (12) significantly improved the inhibitory activity IC50 by almost 6 folds to
0.562 ꢁM, rendering compound 12 as the most potent candidate against HsDHODH
identified in the preliminary SAR optimization attempt. However, the replacement of
naphthyl 12 with anthracenyl 13 completely depleted the activity, suggesting the volume
size of S1 subsite is limited and can not tolerate very bulky hydrophobic moieties.
1
1
1
1
1
4
5
6
7
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
Binding mode analysis through the crystal structure of HsDHODH in complex with
compound 12
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In order to investigate the binding mode of the thiazoleꢀbased inhibitors, the most potent
compound 12 so far was selected to coꢀcrystallize with HsDHODH. The structure was
refined to 2.05 Å with excellent electron density (Table 2, Figure 2) and the coordinate
of the complex structure had been deposited in the Protein Data Bank as the entry 4JGD.
As shown in Figure 2A, HsDHODH possesses a large Cꢀterminal domain and a small
Nꢀterminal domain, linked by an extended loop. The large Cꢀterminal harbors the redox
site, where FMN and substrate dihydroorotate bind. On the other hand, The Nꢀterminal
3
1
1
extension consists of two helices α1 and α2, which features family class 2 enzymes. A
slot is formed between the two helices, within the short α1ꢀα2 loop at the narrow end,
contributing to the formation of a tunnel that ends at FMN cavity near the short α1ꢀα2
loop. This tunnel narrows as it approaches to the proximal redox site. Previous study
suggested ubiquinone probably inserted into the tunnel and thus easily approached FMN
12
13
1
1
1
1
1
4
5
6
7
8
6
, 36ꢀ39
for the redox reaction.
It is clearly noticed that the binding site of inhibitor 12 is
3, 40
formed between the two Nꢀterminal helices, along with the reported inhibitors,
again
supporting the “putative ubiquinone tunnel” is a rational target for development of novel
chemotherapeutics.
ACS Paragon Plus Environment

Page 13 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
The solved structure showed in Figure 2B provided us further explanation for its potent
activity. The binding site can be roughly divided into an exclusive hydrophobic entrance
and a rather polar narrow end. The thiazole ring occupied the polar subsite in a planar
conformation with the adjacent amide bond. There were totally two buried water
molecules in the binding pocket: one formed a bridge between thiazole ring of the
compound 12 and the guanidine group of Arg136, and the other one was hydrogen
bonded to Gln47 and Thr360. Additionally, the carbonyl moiety on thiazole ring formed
a direct hydrogen bond with Tyr356. The naththyl ring, occupied the bulky hydrophobic
entrance as expected, exhibited several hydrophobic contacts with apolar residues like
Met43, Leu46, Ala59, Phe62, Leu68, Phe98, Leu359 and Pro364, consisting with the
fact that the entrance of the binding tunnel is almost made up by hydrophobic amino
acids, as family class 2 enzymes are membrane associated.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
12
13
1
1
1
1
1
4
5
6
7
8
1
2
Modification of R Acyl group and R methyl group
The coꢀcrystal structure with 12 indicated that the inhibitor fits well in the binding
region. But there was still some space for further modification to fit better in the binding
site. For example, S4 subsite capped by Val134 and Val143 accessed to the proximal
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
2
3
4
5
6
7
8
9
0
1
FMN cofactor (Figure 3A) was near the R position on compound 12. This cavity can
be presumably occupied by an alternative bulky group like an ethyl ester substituent.
This hypothesis led to compounds 14ꢀ18 (Table 3) with maintained (18 vs 12) or
improved inhibitory activities by 2ꢀ4 folds (14 vs 6, 15 vs 8, 16 vs 9, and 17 vs 4). The
results summarized above indicated that the ethyl ester group is easily accommodated
with the inhibitor binding site herein and brings compounds closer to the FMN cofactor.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
The impacts on R from the size and polarity of the substituents were further assessed
(Table 3). The replacement of ꢀCH with ꢀNH (compounds 19 and 20) was strongly
3
2
1
1
disfavored possibly because the occupied cavity is surrounded by the relatively
hydrophobic environment involving side chains Pro52 and Val134 (Figure 3A).
2
12
13
However, no improvement was achieved for the amide hydrophobic R substituents,
neither cyclopropane (21 and 22) nor phenyl (23 and 24). The unexpected activity loss
1
1
1
1
1
4
5
6
7
8
may be caused by the formation of intramolecular hydrogen bond between the carbonyl
1
2
oxygen of R and the amide of R , which hindered the direct hydrogen bond to Tyr356.
2
In contrast to the amide derivatives, the introduction of totally hydrophobic group at R
seemed to slightly improve the activities. The replacement by cyclopropyl (25, 26 and
27) and tertꢀbutyl (28, 29 and 30) presented a 6 to 12ꢀfold improvement in inhibitory
ACS Paragon Plus Environment

Page 15 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
activities compared with compound 4. Interestingly, the phenyl substituents (31, 32 and
33) significantly improved activity to double digit nanomolar level (IC = 58 nM, 49
5
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
nM, and 35 nM respectively).
5
6
7
8
9
0
To investigate the reason of the activities improvement for the phenyl substituents,
compound 33 was selected to coꢀcrystallize with HsDHODH. The Xꢀray coꢀcrystal
2
structure of 33 with HsDHODH (Figure 3A) indicated that when the size of R is
enlarged to the size of phenyl group, the “lower” side of subsite S4 is more suitable than
the tiny “upper” hydrophobic pocket formed by Pro52 and Val143 to accommodate the
larger phenyl moiety, and owing to the flexible characteristic of the amine linker, the
binding mode of the thiazole moiety is shifted 180 degree compared with that of 12.
Specifically, the ethyl ester group of 33 formed hydrogen bond with Arg136 other than
1
11
12
13
2
Tyr356, and the larger R phenyl group was oriented toward Tyr356 and made close
1
1
1
1
1
4
5
6
7
8
contact with Val134 and FMN. Considering that compound 33 displayed 100 folds more
2
potency than lead 4, it was concluded that the larger R phenyl group induced binding
style flip of thiazole moiety is beneficial for improving binding affinity, and the newly
presented polar contacts in subsite S2 as well as the strengthened hydrophobic
interactions in subsite S4 are key contributors for the potency boost that reached double
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
digit nanomolar level.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
4
5
6
7
8
9
0
3
Revisit R modification for optimal hydrophobicity
23 new compounds were synthesized by coupling different amines moieties to the
3
thiazole scaffold at R in terms of various hydrophobicity and size to fineꢀtune the
1
2
hydrophobicity effects on the activities, where R was fixed as ethyl ester group and R
3
as phenyl group (Table 4). The inhibitory activity decreased sharply when the R pheny
group was unsubstituted (34) or only methoxyl substituted at pꢀ position (35), indicating
3
that the hydrophobic groups are exclusively favored at R . Consequently, the binding
1
potency of the compounds in this round correlated well with their hydrophobicity
except compound 36, which showed similar activity to compounds 37 and 38 with
chloride substitute. pꢀ and mꢀ disubstitutions yielded compounds 39ꢀ41 with
subꢀmicromolar activities, while the 3,5ꢀdisubstitution (42) showed a slightly reduced
inhibitory activity (42 vs 41). However, oꢀ and pꢀ disubstitution depleted activity
completely (43), corresponding to the limited space of the narrow channel formed
11
12
13
1
1
1
1
1
4
5
6
7
8
3
between aryl amine and thiazole ring as shown in figure 3A. We also evaluated the R
aromatic ring substituents’ impacts on activities (Table 4). Compounds with fused
3
bicyclic system R substituents (44ꢀ47) showed improved activity especially for
ACS Paragon Plus Environment

Page 17 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
compounds 44, 46 and 47 (IC = 26 nM, 18 nM and 29 nM, respectively). The
50
replacement of 5ꢀindanyl (44) with 5ꢀbenzodioxolyl (45) reduced the activity (IC =
5
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.026 ꢁM vs 1.108 ꢁM, respectively), again corresponding to the hydrophobic nature of
the binding cavity. Fused tricyclic substituents seemed unfavorable since compound 48
showed about 10ꢀfold activity decrease comparing with compound 47. However, almost
all the biphenyl or biphenyl with additional linker system depleted inhibitory activity
completely (compounds 49ꢀ52, 54ꢀ56), suggesting the steric constraints of the binding
3
pocket in this area. These efforts on R modification aforementioned led to remarkable
improvement in inhibitory activity over those tested in the preliminary modification.
1
1
To thoroughly inspect the binding modes of the most potent compounds, a flexible
docking study was performed for compound 47 based on the crystal structures reported
in this study, using the Induced Fit Docking protocol in Maestro 9.0 (Schrödinger LLC).
Just like the binding mode in the coꢀcrystal structure of 33, the carbanyl moiety of the
ethyl ester group in 47 formed a hydrogen bond to Arg136 (Figure 3B) with the
conformation of Met43 and Leu46 changed slightly to accommodate the ethyl ester
group, while the phenyl fragment adopted a favorable configuration due to its well fitted
size and hydrophobility with Pro52, Val134 andVal143. Meanwhile, the docking score
12
13
1
1
1
1
1
4
5
6
7
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
of compound 47 was topꢀranked (ꢀ10.17 kcal/mol), which is consistent with its
inhibition activity (Table S1, Supporting Information). Moreover, by comparing the
structures, binding modes and inhibition potencies of 12 and 47, it was once again
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
proved that the flip of thiazole moiety caused by the introduction of phenyl as R in
1
subsite S4 is favorable and the hydrogen bond between the R ethyl ester group and
Arg136 is also preferable for ameliorating binding affinity.
Species Selectivity Analysis of the ThiazoleꢀDerivatives
Interestingly, compound 4 also presented a slightly inhibition potency against
3
1
1
1
pfDHODH with an IC value of 0.630 ꢁM. To evaluate the species selectivity of the
50
thiazoleꢀderivatives, in vitro enzyme assay against Plasmodium falciparum
dihydroorotate dehydrogenase (PfDHODH) was performed (Tables 1, 3 and 4). One of
the most potent compound 12 displayed comparable activities towards the two enzymes
12
13
1
1
1
1
1
4
5
6
7
8
(IC = 0.562 ꢁM vs 0.871 ꢁM, respectively). In the preliminary optimization process,
50
3
the hydrophobic group R was modified and only two compounds (7 and 11) showed
obvious selectivity over PfDHODH. However, when larger substitutions were
1
2
introduced to R or R group (14ꢀ48), the activities against PfDHODH of almost all the
compounds were totally lost. Remarkably, the inhibitors with larger volumes tend to
ACS Paragon Plus Environment

Page 19 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
exhibit preferential inhibition effects against HsDHODH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further explore the structural basis of species selectivity of these thiazoleꢀderivatives,
the crystal structure of HsDHODH in complex with compound 12 was aligned to
41
reported PfDHODH structure (PDB code: 3I65). In PfDHODH, residues Phe171 and
Met536 were the corresponding residues Leu42 and Pro364 in HsDHODH, respectively,
which seal off the hydrophobic channel where the 3ꢀchloroꢀ4ꢀmethylphenyl group binds
42
(Figure 4). On the other hand, the substitutions of Ile263 for Val134 and Ile272 for
2
Val143 made the R ꢀoccupied cavity smaller, which couldn’t tolerate larger substituents
1
1
such as tertiary butyl and phenyl groups. In conclusion, the species selectivity profile of
these inhibitors was caused by the size constraints of the smaller PfDHODH pocket,
31, 43
12
13
which was consistent with our previous study.
1
1
1
1
1
4
5
6
7
8
The initial thiazoleꢀbased lead compound 4 identified from virtual screening approach
showed modest potency against HsDHODH. In order to explore more and better potent
inhibitors, a set of new derivatives were synthesized. The replacement of ethyl ester
1
2
group on R and phenyl group on R led to the most potent series in our study. Naphthyl
analogues (44, 46 and 47) showed double digit nanomolar inhibitory activities against
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
HsDHODH. By comparison of the more potent inhibitors with that displayed poor
activity ones, we anticipated to obtain structural clues for future rational design study.
First, the orientation of the thiazole moiety was largely influenced by the molecular size
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
of hydrophobic groups at R in subsite S4, which determines whether the polar
1
substitute R is hydrogen bonded to Arg136 in S2 or Tyr356 in S3. And a wellꢀfitted
2
1
phenyl group was recommended for R . As for R , some other polar groups like
carbonyl and carboxyl were preferred to generate hydrogen bond in the hydrophilic
3
region of S2 and S3. Then, the aromatic system on R with two fused rings was favored.
Aniline with mono bulky hydrophobic moiety at para position or combination of para
and meta substitution were also well tolerated. Usually, the diꢀhalogen substitutents on
benzene were preferred. Additionally, the introduction at ortho postion on aniline
should be avoidable since the binding site between thiazole and aryl amine is narrow.
Finally, the sizable conformational change in S1 subsite may be paid more attention.
Generally, it was the major factor to switch the selectivity between HsDHODH and
PfDHODH.
1
1
12
13
1
1
1
1
1
4
5
6
7
8
Physicochemical Characterization
The physicochemical properties of all compounds were analyzed by Jaguar pKa
ACS Paragon Plus Environment

Page 21 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
4
45
46
1
2
3
4
5
6
7
8
9
0
1
prediction module, XLOGP and XLOGS software. Selected inhibitors (32, 33, 44,
and 47) were also measured by Sirius T3 Station. The pKa values of the derivatives
(5ꢀ13) varying between 0.5 and 2.3 were closely comparable to that of the hit compound
4 with predicted pKa at 1.6 (Table 1). The estimated logP showed variation between
3.25 and 5.75 where the logS of these derivatives (5ꢀ13) possessed values between ꢀ5.33
and ꢀ2.56 (Table 1). These values indicated that compunds 5ꢀ13 were low basic and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
shared poor solubility in water. The replacement of the acyl groups (R ) with esters
(compounds 14ꢀ18) increase the pKa and logP values (8 vs 15, 12 vs 18) albeit at a
2
similar level of solubility. Modification on methyl group (R ) with polar substituents
1
1
such as amines, amides (19ꢀ24), and bulkiers (25ꢀ30) also did not improve solubility
2
(Table 3). Finally, as discribed in table 4, compounds 34ꢀ56 bearing phenyl at R and
3
12
13
different hydrophobic substituents at R reduced the pKa and logS whereas exhibited
higher logP (33 vs 17, 47 vs 18). All these calculated or measured molecular descriptors
for these compounds will be used as the indicators for further optimizatation of the
physicochemical and drugꢀlike properties.
1
1
1
1
1
4
5
6
7
8
In vivo Antiꢀarthritic Effect of Compound 44
In this experiment, Wistar rats were treated with bovine type II collagen to induced
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
arthritis (collagen ꢀ induced arthritis, CIA), then injected intraperitoneally with the
compound 44 and methotrexate once per day for 28 days. The effect of compound 44
was evaluated by the arthritis swelling score and morphological observation of joint
tissue of rats (Figure 5).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
During experiment, control rats had shiny hair and normal eating. The body weight
continuously increased, while the rats in model group showed a rough and dull hair,
with a mild diarrhea at the beginning. The growth of body weight was markedly slower
than that in normal group since day 9 (Figure 5A). Treatment of compound 44 and
methotrexate had no obvious effect on the growth of body weight. Moreover, compound
44 displayed significant antiꢀarthritic efficacy in vivo (p < 0.05) and markedly alleviated
foot swelling in a doseꢀdependent manner. Substantial antiꢀinflammation effects were
be clearly observed after day 12 (both 5 mg/kg and 30 mg/kg, Figure 5B). After day 18,
the arthritis scores in CIA rats reached to the peak and kept stable (Figure 5B).
Compared to CIA rats, Methotrexate and compound 44 treated rats (in both dosage)
significantly decreased the arthritis swelling score, indicated by obviously alleviated
foot swelling in a dose related manner (Figures 5B and 5C). Moreover, hematoxylin and
eosin (H&E) staining demonstrated that CIA rats exhibited histological changes of
1
1
12
13
1
1
1
1
1
4
5
6
7
8
ACS Paragon Plus Environment

Page 23 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
severe arthritis, showing substantial infiltration of inflammatory cells, synovial space
exudation, synovial hyperplasia and cartilage erosion in joint tissues. Aboveꢀmentioned
pathological features were markedly alleviated in rats treated with compound 44 in a
doseꢀdependent manner (Figure 5D), indicating the significant of antiꢀinflammation
effects. Briefly, the in vivo antiꢀarthritic effect of compound 44 were quite encouraging
that it might serve as a promising lead compound for further development as
immunosuppressant and antiproliferative agent targeting HsDHODH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
Conclusion
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
4
5
6
7
8
9
0
1
2
3
4
5
6
We successfully identified a series of novel and promising thiazoleꢀderivatives as
inhibitors against HsDHODH through structureꢀbased lead optimization. Preliminary
SAR deconvolution on lead compound 4 led to a potent compound 12, then the crystal
structures of the complex of HsDHODH with compounds 12 and 33 were solved and
analyzed to guide further SAR optimization aiming to increase the inhibitory activity.
The SAR study reported here corresponds well to the structural analysis of the binding
site and interaction modes observed from the crystal structure and leads to several
promising compounds with double digit nanomolar activities, especially compounds 33,
44, 46 and 47 with IC50 values of 35nM, 26 nM, 18 nM, and 29 nM respectively,
whereas their drugꢀlike properties along with pharmacokinetic characteristics will be
optimized further in the future work. Moreover, in vivo efficacy study demonstrated that
compound 44 displayed considerable antiꢀarthritic effect and significantly alleviated
foot swelling in a doseꢀdependent manner. In conjunction with the cocrystal structure
data, these results further suggested that HsDHODH is an effective target for
rheumatoid arthritis chemotherapy and novel scaffolds designed in this work might lead
to the discovery of new immunosuppressant and antiproliferative agents targeting
HsDHODH.
1
1
1
1
1
1
1
17
18
19
ACS Paragon Plus Environment

Page 25 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
Experimental Section
In Vitro Enzyme Assay.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4
5
6
7
8
9
0
1
The pasmids coding for human and Plasmodium falciparum DHODH were kindly
provided by Prof. Jon Clardy (Harvard Medical School). HsDHODH (Met30ꢀArg396)
and PfDHODH (Phe158ꢀSer569) plasmid construction, protein expression and
4
1, 47
purification were followed the protocols of Liu and Deng respectively.
The
DHODH inhibition assays were carried out by using a DCIP assay method. The purified
HsDHODH was diluted into a final concentration of 10 nM with an assay buffer
containing 50 mM HEPES pH 8.0, 150 mM KCl, then UQ and DCIP were
0
1
1
supplemented to the assay buffer to the final concentration of 100 ꢁM and 120 ꢁM
48
respectively. For PfDHODH, additional 0.1% triton Xꢀ100 was added. The mixture
was transferred into a 96ꢀwell plate and incubated for 5 min at room temperature. In the
following step, the dihydroorotate was added to a final concentration of 500 ꢁM to
initiate the reaction. The reaction was monitored by measuring the decrease of DCIP in
the absorption at 600 nm for each 30 s over a period of 6 min. Inhibition studies were
performed in this assay with additional variable amounts of compounds. A771726 and
DSM1 were also measured as the positive control for HsDHODH and PfDHODH
respectively. Percent inhibition relative to the no inhibitor control was calculated from
12
13
1
1
1
1
1
4
5
6
7
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
(1ꢀV /V ) ×100%. For the determination of the IC values, 8ꢀ9 different concentrations
i
0
50
were applied. Each inhibitor concentration point was tested in triplicate. IC50 values
were calculated using the sigmoidal fitting option of the program Origin 8.0.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Co-crystallization of HsDHODH with compounds 12 and 33
The complex crystals were cocrystallized by hangingꢀdrop vapor diffusion method at 20
o
C. Drops were formed on glass coverslips by mixing 1.5 ꢁL of a 20 mg/mL protein
solution in 50 mM HEPES pH 7.8, 400 mM NaCl, 30% glycerol, 1 mM EDTA, 10 mM
N,NꢀdimethylundecylaminꢀNꢀoxide (C DAO), 2 mM dihydroorotate (DHO), and 1
11
1
1
mM compound 12 or 33 with an equal volume of precipitant solution consisting of 0.1
M acetate pH 4.8, 40 mM C DAO, 20.8 mM N,NꢀdimethyldecylamineꢀNꢀoxide
11
12
13
(DDAO), 1.6ꢀ1.8 M ammonium sulfate. The drops were incubated against 1 mL of
reservoir of 0.1 M acetate pH 4.8, 1.6ꢀ1.8 M ammonium sulfate and 30% glycerol.
Crystals usually appeared as small yellow cubes within 3 days and reached a full size of
1
1
1
1
1
4
5
6
7
8
3
0.2×0.2 ×0.2 mm within 3 weeks.
Data Collection, Structure Determination, and Refinement.
Xꢀray diffraction data were collected at 100K at the synchrotron beamline BL17U1 of
ACS Paragon Plus Environment

Page 27 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
SSRF (Shanghai, China). Statistics of data collection, processing, and refinement were
49
summarized in Table 2. The data were processed with MOSFLM, and scaled using the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50
SCALA program from the CCP4 suite. Structural elucidation and refinement were
5
0, 51
carried out using the CCP4 suite of programs.
The crystal structure was determined
by molecular replacement, using PDB entry 1D3G (without ligands and water
52
molecules) as the template. REFMAC was employed for structure refinement. The
53
computer graphics program Coot, implemented in the CCP4 suite, was used for
interpretation of the electron density map and model building. The molecular graphics
package PyMOL (DeLano, 2002) was used to generate the figures.
9
0
1
1
1
In vivo Efficacy Study.
1
1
1
1
1
2
3
4
5
6
Wistar rats (male, 130 ± 20 g), purchased from Shanghai SLRC Laboratory animal
company, were housed in the condition of 20ꢀ25 °C, 50ꢀ60% relative humidity with a
12ꢀhour light and night cycle. Food and water could be freely obtained. All procedures
conform to the Chinese government guidelines for animal experiments. To establish
CIA model, equal volume of bovine type II collagen (CII, Chondrex, USA) and
incomplete Freund's adjuvant (IFA, Chondrex, USA) were emulsified using a
17
18
19
54
highꢀspeed homogenizer (IKA Co., Germany) on ice. Compound 44 was freshly
dissolved in a cosolvent consisted of 5% DMSO, 45% saline and 50% PEG200. 45 rats
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2
3
4
5
6
7
8
9
0
1
were randomly attributed into normal control (n = 9), CIA model (n = 8), Methotrexate
treated group (n = 10) and 5 mg/kg (n = 10), 30 mg/kg (n = 8) compound 44 treated
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
5
groups. Methotrexate, reported in the treatment of rheumatoid arthritis, was used as
positive control in present study.
At the beginning (day 0) of experiment, the rats were intradermally injected at the back
base of the tail with 200 ꢁL of the emulsified mixture containing 200 ꢁg of CII. A
booster injection of 100 ꢁg CII was given at day 7 after primary immunization. Normal
control received no treatment. The rats were injected intraperitoneally of compound 44
(5 mg/kg and 30 mg/kg) and Methotrexate (0.3 mg/kg), respectively, once per day for
28 days. At the meantime, the control group and model group were given the same
amount of solvent.
1
1
12
13
During experiment, the weight of rats was recorded every three days. After the onset of
arthritis, the foot swelling scores were recorded every two days under blinded
conditions. The criteria of evaluation: 0 = no swelling; 1 = slight swelling; 2 = swelling;
3 = significant swelling; 4 = severe swelling. Since hind foot articular showed more
pronounced swelling than forepaw, we used scoring criteria of maximum 8 points that
was the total score of two hind feet to evaluate the incidence of rats’ arthritis.
All rats were sacrificed after 28 day’s treatment of compounds. Whole knee joints were
1
1
1
1
1
4
5
6
7
8
ACS Paragon Plus Environment

Page 29 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
0
1
fixed in 4% formalin for pathological detection. Tissue samples were decalcified in 10%
EDTA for at least two weeks, then processed paraffin embedding, slice chopping and
hematoxylin and eosin (H&E) staining. Stained sections were observed under optical
microscope and photographed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Data were expressed as mean ± standard error (mean ± SEM). SPSS statistical
software was used to analyze the statistical significance of differences among groups,
by oneꢀway ANOVA following a multiꢀcomparison analysis (LSD). P ꢀ 0.05 was
56
considered to be statistically significant.
1
1
Chemistry General Methods.
All chemical reagents and solvents were obtained from commercial sources and used
without further purification. Thinꢀlayer chromatography (TLC) was carried out to
monitor the process of reactions. Purification of compounds was achieved by column
12
13
1
13
1
1
1
1
1
4
5
6
7
8
chromatography with silica gel (Hailang, Qingdao) 200ꢀ300 mesh. H NMR and
C
NMR spectra were recorded on a Bruker AMꢀ400 spectrometer with chemical shifts
expressed as ppm (in CDCl , Me Si as internal standard). Melting points were analyzed
3
4
on a WRSꢀ1Bꢀdigital melting point apparatus. The mass spectra were measured at The
Institute of Fine Chemistry of ECUST. Purity was determined using highꢀperformance
ACS Paragon Plus Environment