Design and discovery of new pyrimidine coupled nitrogen aromatic rings as chelating groups of JMJD3 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.01.006

The histone methylation on lysine residues is one of the most studied post-translational modifications, and its aberrant states have been associated with many human diseases. In 2012, Kruidenier et al. reported GSK-J1 as a selective Jumonji H3K27 demethyl

Full text of DOI:10.1016/j.bmcl.2016.01.006

Accepted Manuscript
Design and Discovery of new pyrimidine coupled nitrogen aromatic rings as
chelating groups of JMJD3 Inhibitors
Jianping Hu, Xin Wang, Lin Chen, Min Huang, Wei Tang, Jianping Zuo, Yu-
Chih Liu, Zhe Shi, Rongfeng Liu, Jingkang Shen, Bing Xiong
PII:
S0960-894X(16)30008-7
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.01.006
BMCL 23471
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
15 October 2015
31 December 2015
5 January 2016
Please cite this article as: Hu, J., Wang, X., Chen, L., Huang, M., Tang, W., Zuo, J., Liu, Y-C., Shi, Z., Liu, R.,
Shen, J., Xiong, B., Design and Discovery of new pyrimidine coupled nitrogen aromatic rings as chelating groups
of JMJD3 Inhibitors, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.
2016.01.006
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Design and Discovery of new pyrimidine coupled nitrogen
aromatic rings as chelating groups of JMJD3 Inhibitors
Jianping Hua Xin Wanga Lin Chena Min Huangb Wei Tangb
Jianping Zuob
Yu-Chih Liuc Zhe Shic Rongfeng Liuc Jingkang Shena,* Bing Xionga,*
aDepartment of Medicinal Chemistry, State Key Laboratory of Drug Research, Shanghai Institute
of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road, Shanghai 201203,
China
bDepartment of pharmacology, Shanghai Institute of Materia Medica, Chinese Academy of
Sciences, 555 Zuchongzhi Road, Shanghai 201203, China
cShanghai ChemPartner Co., LTD. Building 5, 998 Halei Road, Zhangjiang Hi-Tech Park,
Pudong New Area, Shanghai, 201203, P.R.China
*Corresponding Authors: (B Xiong) E-mail: bxiong@simm.ac.cn, Tel: (86)021-50806600-5412;
(JK Shen) jkshen@simm.ac.cn, Tel: (86)021-50806600-5410
Abstract
The histone methylation on lysine residues is one of the most studied
post-translational modifications, and its aberrant states have been associated with
many human diseases. In 2012, Kruidenier et al. reported GSK-J1 as a selective
Jumonji H3K27 demethylase (JMJD3 and UTX) inhibitor. However, there is limited
information on the structure-activity relationship of this series of compounds.
Moreover, there are few scaffolds reported as chelating groups for Fe(II) ion in

Jumonji demethylase inhibitors development. To further elaborate the
structure-activity relationship of selective JMJD3 inhibitors and to explore the novel
chelating groups for Fe(II) ion, we initialized a medicinal chemistry modification
based on the GSK-J1 structure. Finally, we found that several compounds bearing
different chelating groups showed similar activities with respect to GSK-J1 and
excellent metabolic stability in liver microsomes. The ethyl ester prodrugs of these
inhibitors also showed a better activity than GSK-J4 for inhibition of TNF-α
production in LPS-stimulated murine macrophage cell line Raw 264.7 cells. Taking
together, the current study not only discoveried alternative potent JMJD3 inhibitors,
but also can benefit other researchers to design new series of Jumonji demethylase
inhibitors based on the identified chelating groups.
Keywords: epigenetics; histone demethylase; JMJD3 inhibitors; GSK-J1; chelating
group

Epigenetics studies heritable changes in gene function without changes in DNA
sequence. Based on the fundamental roles in diverse biological processes and human
diseases, epigenetics catches considerable efforts in current medical and
pharmaceutical researches.1-4 Decades-long studies revealed that epigenetic regulation
of gene expression is mainly associated with DNA methylation, noncoding RNA,
nucleosome remodeling and post-translational modification of histones.5 The detailed
mechanisms of histone modification were gradually emerged and a large number of
enzymes and recognition modules were discovered.6-8
Basically, the proteins involved in post-translational modification of histones can be
classified into three categories: the “writers”—enzymes that create these
modifications, the “readers”—proteins that recognize the modifications, and the
“erasers”—enzymes that remove these modifications. Among them, the histone
methylation on lysine residues is one of the most studied post-translational
modifications, and its aberrant states have been associated with many human
diseases.9-12
Until a decade ago, histone methylation on lysine residues was believed to be a stable
modification. However, the discovery of lysine specific demethylase 1 (LSD1)
elucidated the new direction in histone demethylation.13 And subsequent studies led to
the identification of a family of enzymes containing the catalytic Jumonji C (JMJC)
domain, which utilizes the Fe(II) and 2-oxoglutarate to hydroxylate the ξ-methyl
groups of methylated lysine substrates, allowing for the demethylation of lysine in all
15
three methylation states(mono-, di- and tri-methylated lysines).14,
In 2012,

Kruidenier et al.16 reported a selective Jumonji H3K27 demethylase (JMJD3 and
UTX) inhibitor GSK-J1 (Figure 1A), which bearing a 2-(pyridine-2-yl)pyrimidine
scaffold and acting as 2-oxoglutarate competitive inhibitor by using its two nitrogen
atoms to chelate the Fe(II) ion. Although GSK-J1 was discovered as a potent JMJD3
inhibitor, there is limited information on the structure-activity relationship of the
series of compounds. Moreover, there are few scaffolds reported as chelating groups
for Fe(II) in Jumonji demethylase inhibitors development. 17, 18
To further elaborate the structure-activity relationship of selective JMJD3 inhibitors
and to explore the novel chelating groups for Fe(II), we initialized a medicinal
chemistry modification based on the GSK-J1 structure. We first synthesized
compound 1, which preserves the important pharmacophores of the beta amino acid
and benzoazepine group, and only the chelating scaffold was modified to
2-(pyrimidin-4-yl)thiazole. From the AlphaLISA immunodetection assay, it was found
that compound 1 showed very similar activity with respect to GSK-J1 (Table 1).
Therefore, the 2-(pyrimidin-4-yl)thiazole was utilized as an alternate scaffold to
further studied on the SAR of JMJD3 inhibitors. Scrutinizing the co-crystal structure
of JMJD3 bound with GSK-J1 (PDB entry 4ASK), one can found that the
benzoazepine group is located at the solvent exposing part of the JMJD3 catalytic
domain, indicating the benzoazepine group might be feasible for modification (Figure
1B).

A.
B.
Figure 1. (A) The chemical structure of GSK-J1 and (B) the co-crystal structure of
GSK-J1 bound to JMJD3 (PDB entry: 4ASK).
Along this direction, we prepared other seventeen analogs to investigate the SAR. The
synthetic route for thiazolopyrimidine derivatives 1-18 is shown in Scheme 1. As
shown in Table 1, aromatic rings coupled with piperazine or piperidine were generally
tolerable in this subpocket. Substituting with benzylamine groups (3, 5 and 13) or
longer group (6) dramatically reduced the enzymatic activity, which indicated that the
size of group is critical for the binding. Replacing the aromatic ring with alkyl
carbamate or amide groups also attenuated the binding activity considerably. By
comparing to GSK-J1, it was found that fused tetrahydroisoquinoline ring (14)
reduced the activity about six folds (IC50=0.99 μM). Compound 9 replaced benzene
by pyridine showed very similar activity with respect to compound 15. Substituted the
benzene group with ortho-methoxyl group showed about 3-fold less potent than
compound 8, while the compound (15) with dichloro-substituted benzene group
showed similar activity. Therefore, the modification of benzene group was tolerable
for the binding (7, 8, 9 and 15) and consistent with the hypothesis that this part is
close to the solvent –accessible area, and can be utilized for modification.
Table 1. The activity of thiazolopyrimidine derivatives on inhibition of

JMJD3 from AlphaLISA immunodetection assay.
Compound
number
Compound
R1
JMJD3 IC50(μM)
R1
JMJD3 IC50(μM)
number
10
/
0.15
0.2
-8.4%@10μM
87.5%@10μM
32.9%@10μM
GSK-J1
1
2
11
1.2
12
1.0%@10μM
9.8%@10μM
-0.6%@10μM
-8.7%@10μM
0.99
3
4
5
13
14
15
0.94
10%@10μM
-5.1%@10μM
21.1%@10μM
6
7
16
17
2.1
0.6
0.9
43.9%@10μM
8
9
18

Scheme 1. Reagents and conditions: a) sodium hydride, diethyl carbonate, 90°C; b)
Thiourea, K2CO3, 105°C; c) chloroacetic acid, concentrated HCl, reflux; d) POCl3,
PCl5, 105°C; e) DIPEA, dioxane, beta-alanine ethyl ester hydrochloride, 80°C; f)
DIPEA, isopropanol, R1, 106°C; g) LiOH, THF/H2O, rt.
With this encouraging result, we start to explore more chelating groups by replacing
the thiazole to pyrazole and two kinds of triazoles. As listed in Table 2, the diverse
scaffolds can maintain similar activities with respect to GSK-J1. Comparing 19, 26
and 33 to GSK-J1, it was found that the pyrazole compound (19) had excellent
potency as GSK-J1 and were slightly more active than the triazole compounds(1).
This tendency can also be deduced from the activities of 20-24. We also prepared a
series of 2-(pyrimidin-4-yl)-2H-indazole compounds as listed in Table 2. From the
assay, it indicated that the larger and more hydrophobic indazole group may impair
the binding and decreased the enzymatic activity. The synthetic route for
azolepyrimidine derivatives 19-43 is shown in Scheme 2.

Table 2. The activity of azolepyrimidine derivatives on inhibition of JMJD3 from
AlphaLISA immunodetection assay.
Compound
number
JMJD3
Compound
number
JMJD3
R
R2
R
R2
IC50(μM)
IC50(μM)
0.15
0.75
0.55
1.00
0.74
0.49
>20
19
20
21
22
23
24
32
0.21
33
74%@10μM
34
2.1
1.4
10
35
36
37
>20
2.2
25
26
38
0.27
30%@10μM
39

1.5
5.8
27
40
41
42
1.3
1.4
>100
28
6.1
29
30
1.2
54%@10μM
43
2.6
/
/
0.15
31
GSK-J1
Scheme 2. Reagents and conditions: a) R, THF, DIPEA, 80°C or 0°C; b) DIPEA,
dioxane, beta-alanine ethyl ester hydrochloride, 80°C; c) DIPEA, isopropanol, R2,

106°C; g) LiOH, THF/H2O, rt.
To get better understanding the structure-activity relationship, the GSK-J1 and other
five analogs with diversified chelating scaffolds were subjected to calculations of
physicochemical properties by applying the chemBiooffice software.19 As shown in
Table 3, the logP and tPSA were not relevant to the enzymatic activities, while the
total partial charges on two nitrogen atoms (to chelate Fe2+ ion) were correlated to the
tendency of the activities, which reinforced the chelating group was the dominant
factor for binding affinity.
Table 3. The calculated physicochemical properties of six JMJD3 inhibitors.
Compound
Partial charge Partial charge
total
logP
tPSA
number
N_1
-0.75
-0.82
-0.84
-0.79
-0.84
-0.84
N_2
-0.64
-0.63
-0.51
-0.48
-0.36
-0.27
4.52
3.87
3.37
2.99
2.98
4.79
89.65
89.65
92.89
105.25
105.25
92.89
-1.39
-1.45
-1.35
-1.27
-1.20
-1.11
GSk-J1
1
19
26
33
38
As reported by Kruidenier et al. that the highly polar carboxylic acid group of GSK-J1
restricts cellular permeability.16 Then the prodrug approach was utilized to prepare an
ethyl ester of GSK-J1 compound (GSK-J4) to test the cellular effects on inflammatory
response. From the reported study, the expression of JMJD3 in macrophages is

rapidly induced by proinflammatory stimuli and correlated with the expression of
nuclear factor-κB (NF-κB). Therefore, we examined the efficacy of our JMJD3
inhibitors 49, 75, 82 and 89 (ethyl ester prodrugs of 1, 19, 26 and 33 respectively) as
well as the control compound GSK-J4 for inhibition of TNF-α production in
LPS-stimulated murine macrophage cell line Raw 264.7 cells. The results presented in
Figure 2 illustrate the dose-dependent inhibitory effect of compounds 75 and 82 on
TNF-α production in LPS-stimulated Raw 264.7 cells, implicating the excellent
inhibitory effect on LPS-triggered immune responses. By comparing with GSK-J4, it
was found that compounds 49, 75 and 82 had higher cellular activity at concentration
0.82 μM, while compound 89 showed similar activity.
Figure 2. JMJD3 inhibitors (ethyl ester prodrug) as well as the control
compound GSK-J4 for inhibition of TNF-α production in LPS-stimulated
murine macrophage cell line Raw 264.7 cells. The results were averaged from
three times of experiments.
To complete the investigation of these groups acting as the Fe(II) chelating groups, we
checked the metabolic stability of identified potent inhibitors. The result from human

liver microsome stability assay indicated that these new chelating scaffolds were all
very stable for metabolic transformation in liver microsome, especially the
compounds with pyrazole or triazole groups. From direct inhibition studies on the five
common CYP enzymes, all compounds including GSK-J1 showed low inhibition ratio
at the compound concentration 10 μM, indicating less possibility of metabolism and
drug-drug interaction issues.
Table 4. In vitro metabolic stability assay of selected potent JMJD3 inhibitors
Compoun
d
HLM Stability
Clint(ul/min/mg
protein)
Direct inhibition （%）
HLM
t1/2(min)
3A4
2D6
2C9
1A2
2C19
number
5
0
266.54
12%
8%
5%
9%
9%
5%
13%
1
>693.00
3%
5%
3%
19
1
0
2310
7%
7%
6%
6%
4%
1%
4%
1%
4%
8%
26
33
>693.00
5
3
256.67
495
4%
0%
17%
5%
8%
11%
16%
38
13%
20%
10%
GSK-J1
In summary, to check whether other scaffolds can be used as the chelating groups for
JMJD demethylases, we initialized medicinal chemical modification based on JMJD3
inhibitor GSK-J1. We preserved the important substitutions of the beta amino acid
and changed the chelating scaffold 2-(pyridine-2-yl)pyrimidine in GSK-J1 to other
five scaffolds including 2-(pyrimidin-4-yl)thiazole, 4-(1H-pyrazol-1-yl)pyrimidine,
4-(2H-1,2,3-triazol-2-yl)pyrimidine,
4-(1H-1,2,3-triazol-1-yl)pyrimidine
and
2-(pyrimidin-4-yl)-2H-indazole, and found that the diverse scaffolds except

2-(pyrimidin-4-yl)-2H-indazole showed similar enzymatic activities with respect to
GSK-J1. From cellular effect studys, we found that the ethyl esters prodrug
compounds 49, 75 and 82 showed better activity at TNF-α production in
LPS-stimulated murine macrophage cell line Raw 264.7 cells than GSK-J4. Finally,
the result from human liver microsome stability assay indicated that these new
chelating scaffolds have excellent metabolic stability, especially the compounds with
pyrazole or triazole groups. From direct inhibition studies of the five common CYP
enzymes, all compounds including GSK-J1 showed low inhibition ratio at the
compound concentration 10 μM. Taking together, these results can enable other
researchers to design novel inhibitors not only for JMJD3, but also the whole JMJD
demethylase family.
Acknowledgements
We are grateful for financial support from the National Natural Science
Foundation of China (Grant No. 81273368, 81330076 and 81473094); The
National Science & Technology Major Project "Key New Drug Creation and
Manufacturing Program" of China (Grant No. 2014ZX09507-002 and
2013ZX09507001).
References and notes
1. Egger, G.; Liang, G.; Aparicio, A.; Jones, P. A. Nature 2004, 429, 457.
2. Handel, A. E.; Ebers, G. C.; Ramagopalan, S. V. Trends Mol Med 2010, 16, 7.

3. Portela, A.; Esteller, M. Nat Biotechnol 2010, 28, 1057.
4. Eccleston, A.; Cesari, F.; Skipper, M. Nature 2013, 502, 461.
5. Helin, K.; Dhanak, D. Nature 2013, 502, 480.
6. Winter, S.; Fischle, W. Essays Biochem 2010, 48, 45.
7. Hojfeldt, J. W.; Agger, K.; Helin, K. Nat Rev Drug Discov 2013, 12, 917.
8. Sanchez, R.; Meslamani, J.; Zhou, M. M. Biochim Biophys Acta 2014, 1839, 676.
9. Zhang, Y. Z.; Zhang, Q. H.; Ye, H.; Zhang, Y.; Luo, Y. M.; Ji, X. M.; Su, Y. Y. Neurosci Res 2010,
68, 66.
10. Feng, Y.; Jankovic, J.; Wu, Y. C. J Neurol Sci 2015, 349, 3.
11. Okada, Y.; Tateishi, K.; Zhang, Y. J Androl 2010, 31, 75.
12. Guerra-Calderas, L.; Gonzalez-Barrios, R.; Herrera, L. A.; Cantu de Leon, D.; Soto-Reyes, E.
Cancer Genet 2015, 208, 215.
13. Shi, Y.; Lan, F.; Matson, C.; Mulligan, P.; Whetstine, J. R.; Cole, P. A.; Casero, R. A. Cell 2004,
119, 941.
14. Maes, T.; Carceller, E.; Salas, J.; Ortega, A.; Buesa, C. Curr Opin Pharmacol 2015, 23, 52.
15. Pilka, E. S.; James, T.; Lisztwan, J. H. Drug Discov Today 2015, 20, 743.
16. Kruidenier, L.; Chung, C. W.; Cheng, Z.; Liddle, J.; Che, K.; Joberty, G.; Bantscheff, M.; Bountra,
C.; Bridges, A.; Diallo, H.; Eberhard, D.; Hutchinson, S.; Jones, E.; Katso, R.; Leveridge, M.; Mander,
P. K.; Mosley, J.; Ramirez-Molina, C.; Rowland, P.; Schofield, C. J.; Sheppard, R. J.; Smith, J. E.;
Swales, C.; Tanner, R.; Thomas, P.; Tumber, A.; Drewes, G.; Oppermann, U.; Patel, D. J.; Lee, K.;
Wilson, D. M. Nature 2012, 488, 404.
17. Rose, N. R.; Woon, E. C.; Kingham, G. L.; King, O. N.; Mecinovic, J.; Clifton, I. J.; Ng, S. S.;
Talib-Hardy, J.; Oppermann, U.; McDonough, M. A.; Schofield, C. J. J Med Chem 2010, 53, 1810.
18. Hamada, S.; Suzuki, T.; Mino, K.; Koseki, K.; Oehme, F.; Flamme, I.; Ozasa, H.; Itoh, Y.;
Ogasawara, D.; Komaarashi, H.; Kato, A.; Tsumoto, H.; Nakagawa, H.; Hasegawa, M.; Sasaki, R.;
Mizukami, T.; Miyata, N. J Med Chem 2010, 53, 5629.
19. ChemBioOffice. http://www.cambridgesoft.com/

Graphical Abstract