Identification of BR102910 as a selective fibroblast activation protein (FAP) inhibitor

Article abstract of DOI:10.1016/j.bmcl.2021.127846

Fibroblast activation protein (FAP) belongs to the family of prolyl-specific serine proteases and displays both exopeptidase and endopeptidase activities. FAP expression is undetectable in most normal adult tissues, but is greatly upregulated in sites of tissue remodeling, which include fibrosis, inflammation and cancer. Due to its restricted expression pattern and dual enzymatic activities, FAP inhibition is investigated as a therapeutic option for several diseases. In the present study, we described the structure–activity relationship of several synthesized compounds against DPPIV and prolyl oligopeptidase (PREP). In particular, BR102910 (compound 24) showed nanomolar potency and high selectivity. Moreover, the in vivo FAP inhibition study of BR102910 (compound 24) using C57BL/6J mice demonstrated exceptional profiles and satisfactory FAP inhibition efficacy. Based on excellent in vitro and in vivo profiles, the potential of BR102910 (compound 24) as a lead candidate for the treatment of type 2 diabetes is considered.

Full text of DOI:10.1016/j.bmcl.2021.127846

Bioorg. Med. Chem. Lett. 37 (2021) 127846
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of BR102910 as a selective fibroblast activation protein
(FAP) inhibitor
,
b
,
,*
Hui Jin Junga *, Eun Hye Nama, Jin Young Parka, Prithwish Ghoshb, In Su Kimb
a Research Center, Boryung Pharmaceuticals Co. Ltd., Ansan 15425, Republic of Korea
b School of Pharmacy, Sungkyunkwan University, Suwon 16419, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Fibroblast activation protein (FAP) belongs to the family of prolyl-specific serine proteases and displays both
exopeptidase and endopeptidase activities. FAP expression is undetectable in most normal adult tissues, but is
greatly upregulated in sites of tissue remodeling, which include fibrosis, inflammation and cancer. Due to its
restricted expression pattern and dual enzymatic activities, FAP inhibition is investigated as a therapeutic option
for several diseases. In the present study, we described the structure–activity relationship of several synthesized
compounds against DPPIV and prolyl oligopeptidase (PREP). In particular, BR102910 (compound 24) showed
nanomolar potency and high selectivity. Moreover, the in vivo FAP inhibition study of BR102910 (compound 24)
using C57BL/6J mice demonstrated exceptional profiles and satisfactory FAP inhibition efficacy. Based on
excellent in vitro and in vivo profiles, the potential of BR102910 (compound 24) as a lead candidate for the
treatment of type 2 diabetes is considered.
Fibroblast activation protein
Fibroblast growth factor 21
FAP inhibitor
DPPIV
PREP
Fibroblast activation protein (FAP, FAP-
α
, separase) is a type II
sensitivity, fat browning, and glucose uptake. Therefore, the develop-
ment of FAP inhibitor could be a promising therapeutic approach for
type 2 diabetes alternative to the recombinant FGF21 therapies.10
With the growing interest on FAP inhibition in drug discovery,
various FAP inhibitors have been reported.11 For examples, a non-
selective inhibitor, Val-boroPro (1) reached up to the phase II clinical
trial for cancer treatment. However, it was withdrawn due to both safety
and efficacy issues (Fig. 2).12 Linagliptin (2) was clinically approved as a
DPPIV inhibitor though it shows significant FAP affinity.13 Recently,
ARI-3099 (3) with a boronic acid moiety was also found as a FAP in-
hibitor.14 The selectivity of this compound is more than 350-fold to-
wards FAP over PREP. The affinity of this compound is negligible for the
DPPs. It is noted that a dipeptide linker between pyrrolidine moiety and
heterocycle framework is very crucial for improved FAP affinity.14 In
fact, the installations on dipeptide linker led to the loss in potency to-
ward FAP and the increased selectivity for PREP rather than FAP. It was
also observed that addition of halo substituents (F and Cl) to the pyridine
ring improved potency against FAP.
transmembrane glycoprotein consisting of 760 amino acids, which is
inducible in cultured fibroblasts using the monoclonal antibody (mAb)
F19.1 It belongs to the family of post-proline dipeptidyl aminopepti-
dases, which mainly cleave peptide substrates after proline residues.2
Dipeptidyl peptidases (DPP2, DPP4, DPP8, DPP9) and prolyl oligo-
peptidases (PREP, POP) are also parts of this family.3 FAP is a serine
protease that has all of the activities of endopeptidase, and is known to
be highly expressed in various lesion tissues such as epithelial cancer,
cirrhosis and pulmonary fibrosis.4 Since FAP is mainly expressed in fi-
broblasts, it induces tumor development by cancer growth and the in-
vasion into normal tissues.5 Moreover, FAP is known to affect type 2
diabetes and obesity through the inactivation of FGF21 (fibroblast
growth factor 21) by the N- and C-terminal cleavage process.6 As shown
in Fig. 1, FGF21 is a hormone secreted from liver and adipose tissues to
regulate sugar and lipid metabolism.7 In the case of human FGF21, the
C-terminal portion of FGF21 is readily cleaved by FAP, and the biolog-
ical action of FGF21 is reduced by about 400 times.8 In addition, the
inactivation of FGF21 is involved in the accumulation of triglycerides
and improvement of inflammation in the liver. Therefore, it is expected
to be a crucial target for the treatment of non-alcoholic steatohepatitis
(NASH).9 Moreover, FGF21 is highly related with the control of insulin
Compound
4
containing
a
quinolinoylglycyl(2-cyano-4,4-
difluoropyrrolidine) backbone has been recognized as one of the most
potent FAP inhibitors.15 Notably, the preferential accommodation of 2-
cyanopyrrolidine moiety with minimal steric bulk in the FAP’s S1 pocket
* Corresponding authors at: Research Center, Boryung Pharmaceuticals Co. Ltd., Ansan 15425, Republic of Korea (H.J. Jung).
E-mail addresses: yesjin00@naver.com (H.J. Jung), insukim@skku.edu (I.S. Kim).
https://doi.org/10.1016/j.bmcl.2021.127846
Received 21 December 2020; Received in revised form 23 January 2021; Accepted 30 January 2021
Available online 9 February 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

H.J. Jung et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127846
Fig. 3. General structure of synthesized FAP inhibitors.
All the compounds listed in Tables 2–5 were prepared according to
the general synthetic routes summarized in Scheme 1. The key inter-
mediate, 4,4-difluoro-2-glycylcyclopentane-1-carbonitrile 53, was pre-
pared in two steps from 4,4-difluoropyrrolidine-2-carbonitrile 51 via
amide formation followed by deprotection of Boc group under acidic
conditions. The Pd(0)-catalyzed Suzuki coupling of azolyl chloride 54
with aryl boronic acids provided the corresponding adducts 55. Subse-
quently, hydrolysis of ester group and EDC-catalyzed amidation reaction
furnished difluoropyrrolidines 5–16.
Fig. 1. Biological action of human FGF21.
Acyl chlorination of phenylacetic acids 56 with oxalyl chloride and
consecutive amination by ammonia afforded the corresponding phe-
nylacetamides 57, which was converted to the corresponding thio-
amides by treatment with Lawesson’s reagent. The intermediate
thioamides were subjected into intermolecular annulation with ethyl-3-
bromo-2-oxopropanoate to give 2-benzyl thiazoles 58. Difluor-
opyrrolidines 17–38 were obtained from the benzyl thiazoles 58 via
basic hydrolysis and amide formation in a similar manner as stated
before. Sulfinyl 35 and sulfonyl 36 derivatives were obtained through
oxidation of compound 34 with m-CPBA. Substituted acetic acids 59
were transformed into difluoropyrrolidines 39–50 by adopting a similar
sequence as that for compounds 17–38.
Fig. 2. Selected FAP inhibitors.
Compounds 5–10 containing a para-chlorophenyl ring were first
evaluated against inhibitory effects on FAP and PREP, as shown in
was described.16 The potency and selectivity of the synthesized com-
pounds were significantly decreased by exchanging cyano group into
other functional groups such as boronic acid and chloromethyl ketone.
More importantly, the improved hydrophobicity arising out of
difluorination might be playing an important role for high FAP inhibi-
tory potency, proficient ligand efficiency and superior FAP/PREP
selectivity of compound 4 (Table 1).
Table 2
In
vitro
potency
and
selectivity
of
compounds
Taking into consideration the structure activity relationship of
reference compounds 3 and 4, we conceived a general structure with
three fragments to design the novel and selective FAP inhibitors (Fig. 3).
Firstly, the 2-cyano-4,4-difluoropyrrolidine tail was retained in the
newly designed compounds. A N-oxoglycyl group as a second fragment
was selected to link a tail with a heterocycle framework. Then we started
the synthesis and evaluation of various aryl-substituted 5- and 6-
membered heterocycles, which have been rarely explored as the het-
erocyclic fragment. Moreover, different kind of linkers consisting of
alkyl and heteroalkyl moieties between aryl and heteroaryl rings were
also screened to identify the influence on the FAP inhibition.
5–10.
% Inhibition†
1uM
IC50
(μM)
SI*
‡
Entry
R
FAP
PREP
5
6
7
86.74
0.148
0.093
0.550
26.84
3.997
3.623
181.4
43.0
6.6
88.71
57.75
Table 1
IC50 values for reference compounds 1–4.
Compounds
IC50
(
μ
M)
8
85.31
63.98
80.64
0.104
0.795
0.100
0.624
0.197
0.108
6.0
0.2
1.1
FAP
DPP-4
PREP
DPP-2
DPP-9
N.D.a
1
2
3
4
0.07 ± 0.01
0.022 ±
0.001
0.98 ±
0.06
0.086 ±
0.0007
>100
9
0.37 ±
0.002
0.0020 ±
0.00002
>100
>100
>100
>100
>12.5
10
0.025 ±
0.001
0.99 ±
0.04
>100
>100
0.0033 ±
0.0004
>100
1.8 ±
0.2
*SI means selectivity index (calculated as [IC50(PREP)/IC50(FAP)]).
†
‡
Competitive binding assay (TR-FRET).
a
N.D. = not determined.
IC50 values by fluorometric assay using AMC substrates.
2

H.J. Jung et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127846
Table 3
Table 4
IC50 values of compounds (5 and 11–17) with
a
thiazole scaffold.
IC50 values of compounds 17–38 with various substituents on aryl
rings.
‡
Entry
R
% Inhibition†
IC50
(
μM)
SI* (PREP/FAP)
R1
% Inhibition†
1uM
IC50
FAP
(
μ
M)
SI*
‡
Entry
5
1uM
FAP
DPP-4
PREP
DPP-4
N.D
PREP
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4-Cl
98.58
93.71
97.52
99.96
90.64
97.95
99.35
98.97
97.20
94.47
95.42
95.87
97.84
86.32
99.27
98.70
0.021
0.050
0.035
0.016
0.078
0.026
0.004
0.002
0.001
0.055
0.047
0.060
0.037
0.022
0.014
0.012
>100
0.20
0.182
0.009
0.159
0.108
0.001
0.005
0.026
49.00
0.001
0.016
0.474
0.048
0.107
0.200
3.204
0.103
8.7
0.2
4.5
6.8
0.0
0.2
6.6
24.5
1.0
0.3
10.1
0.8
2.9
9.1
27.6
8.6
2-Br
86.74
0.148
0.077
0.06
26.84
1.377
0.154
181.4
17.9
2.6
3-Br
52.02
>100
>100
>100
>100
>100
>100
>100
>100
>100
68.85
>100
>100
>100
4-Br
2-I
11
98.96
94.64
>100
>100
4-OMe
2-naphthyl
3,4-di-Cl
2,4-di-Cl
2,5-di-Cl
3,4-di-F
2-Cl, 4-F
4-Me
12
13
14
15
90.49
76.45
84.75
0.087
0.214
0.414
N.D
N.D
N.D
1.597
1.182
10
18.4
5.5
4-OCF3
4-CN
24.2
33
93.84
0.145
>100
0.464
3.2
16
17
83.38
98.58
0.446
0.021
N.D
10
22.4
8.7
34
35
36
37
4-SMe
99.02
95.00
98.03
98.84
0.025
0.390
0.020
0.023
>100
0.060
1.205
0.048
0.076
2.4
3.1
2.4
3.3
4-SOMe
4-SO2Me
4.12
>100
0.182
10.110
>100
*SI means selectivity index (calculated as [IC50(PREP)/IC50(FAP)]).
†
38
98.31
0.028
18.71
0.410
14.6
Competitive binding assay (TR-FRET).
‡
IC50 values by fluorometric assay using AMC substrates.
*SI means selectivity index (calculated as [IC50(PREP)/IC50(FAP)]).
Table 2. After critically analyzing the inhibitory potency and selectivity
of the compounds comprising of 5- or 6-membered heterocycle, com-
pound 5 containing a thioazole ring was identified to display selectivity
more than 100 folds for FAP than PREP. Introduction of an oxazole or
pyrrole moiety (compounds 6–8) could not provide any significant
advantage to overall activity and selectivity. Major loss of selectivity
was observed with the increase in ring size in case of compounds 9 and
10 containing pyrazine and pyridine moieties respectively. Hence, the
thiazole moiety was chosen for further investigation based on the
selectivity of inhibition criteria.
†
Competitive binding assay (TR-FRET).
‡
IC50 values by fluorometric assay using AMC substrates.
Table 5
IC50
values
of
compounds
24
and
39–50
with
modified
linkers.
Next, the effect of the substituents on the aryl ring attached to
thiazole was evaluated, and the results are given in Table 3. A CF3-group
on compounds 11 and 12 increased the FAP inhibitory activity, but
caused a sharp reduction in the SI value compared to 5. Similar effect
was recorded in case of compound 13 with a 4-fluorophenyl substituent.
Decrease in both activity and selectivity were found for para-methoxy
and para-phenyl groups (compounds 14 and 15). Replacement of the
phenyl ring with a fused benzofuran ring in compound 16 could not
offer any improvement in this trend. Interestingly, para-chlorobenzyl
thiazole substituent in compound 17 dramatically enhanced the FAP
†
Entry
X
Y
n
IC50
FAP
(
μ
M)
SI* (PREP/FAP)
DPP-4
PREP
24
39
40
41
42
43
44
45
46
47
48
49
50
3,4-di-Cl
CH2
CH2
CH2
S
0
1
1
1
1
1
1
1
1
1
1
1
1
0.002
0.051
0.021
0.119
0.005
0.024
0.011
0.019
0.009
0.005
0.020
0.035
0.007
>100
>100
82.53
>100
>100
44.56
>100
>100
76.790
>100
>100
>100
>100
49.00
0.387
0.051
0.001
0.121
0.138
0.508
0.368
0.077
0.151
0.163
0.269
0.070
24.5
7.6
3,4-di-Cl
4-CN
2.4
3,4-di-Cl
4-Me
0.74
24.2
5.8
S
4-CN
S
inhibitory potency (0.021 μM). Determining the best scaffold with spe-
4-CF3
S
46.2
19.4
8.6
cific regard to affinity for FAP, regardless of selectivity, was the primary
goal of second step of scaffold optimization. Therefore, the next series of
compounds were constructed with benzyl thiazole as the heterocyclic
part and the derivatives were synthesized with different substitution of
various size and electronic properties at the phenyl ring (Table 4). The
IC50 values for FAP, DPPIV, and PREP inhibitions were determined for
these compounds. In general, monohalo-substituted compounds
demonstrated high potency and selectivity varied according to the po-
sition (compounds 17–21). Particularly, loss of selectivity was observed
3,4-di-Cl
4-CN
SO2
SO2
SO2
SO2
O
4-CF3
30.3
8.2
4-Me
3,4-di-Cl
4-CN
7.7
O
10.0
*SI means selectivity index (calculated as [IC50(PREP)/IC50(FAP)]).
†
IC50 values by fluorometric assay using AMC substrates.
3

H.J. Jung et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127846
Scheme 1. Reagents and conditions: (a) (tert-butoxycarbonyl)glycine, EDCI, HOBt, NMP, DMF; (b) TFA, rt; (c) Pd(PPh3)4, DPPF, K2CO3, aryl boronic acid, toluene,
90 ℃; (d) LiOH⋅H2O, THF/H2O, 80 ℃; (e) 53, EDCI, HOBt, NMP, DMF; (f) oxalyl chloride, CH2Cl2, DMF (cat.), rt; (g) NH3, THF, rt; (h) Lawesson’s reagent, THF, 80
℃; (i) ethyl-3-bromo-2-oxopropanoate, EtOH, 80 ℃; (j) m-CPBA, CH2Cl2, rt.
in ortho-halo compounds 18 and 21, and para-halo congeners 17 and 20
were found most active as well as selective. Particularly, the para-bromo
compound 20 displayed the highest percent inhibition value in this se-
ries. Although an electron-donating para-methoxy group on compound
22 could impart good activity against FAP, this analog almost lost
selectivity over PREP whereas the analog with an ortho-naphthyl group
on 23 displayed higher potency and selectivity in contrast to the ortho-
halo compounds. Among the dihalo derivatives 24–28, meta- and para-
substituted analogs were found to be more selective. Dichloro-
substituted compounds 24 and 25 were recorded to be highly active
with single-digit nanomolar FAP inhibitory concentration. It is noted
that 3,4-dichloro-substituted compound 24 was far more selective to
FAP than 2,4-dichloro isomer 25, which showed equal preference to
bind with FAP and PREP. An electron donating methyl group at the para-
position of compound 29 decreased the FAP activity, but increase in
selectivity was observed with respect to the para-methoxy analog 22.
Electron-withdrawing groups such as trifluoromethoxy and cyano at the
para-position (compounds 30 and 31) could not improve the activity
compared to 24, but para-cyano derivative 31 was found to be very
selective to FAP with the highest SI index in this series and a percent
inhibition value of 99.27. A better FAP potency was observed for para-
cyano phenyl substituted compound 32 though the SI index was similar
to that of para-chloro substituted compound 17. Significant loss of ac-
tivity and selectivity was observed for para-chloro phenyl substituted
compound 33. Remarkably, the FAP potency of the para-thiomethyl
derivative 34 was close to that of para-methoxy derivative 22, and SI
index was near to that of para-methyl derivative 29. The monoxide
derivative 35 showed manifold-diminished FAP inhibitory potency,
while the sulfone derivative 36 was more potent as compared to com-
pound 35. Pyrazole substituted compound 37 exhibited nearly same
potency to that of triazole substituted analog 38 regarding FAP inhibi-
tion. Taking into account both activity and selectivity, 3,4-dichloro-
substituted compound 24 was found to be the most suitable candidate
in this series.
The influence of different linkers between the 5-membered thiazole
and 6-membered phenyl rings at the heterocyclic fragment was inves-
tigated. Therefore, the effect of various linkers consisting of alkyl and
heteroalkyl moieties was screened and the results are listed in Table 5.
Additionally, some of the compounds with satisfactory potency and
selectivity from the former series were also evaluated with a modified
linker to analyze the effect in comparison with compound 24. Increasing
the length of the carbon chain was not beneficial. The immediate ho-
mologues 39 and 40 were found to be much inferior against FAP as
compared to the lower homologues 24 and 31, respectively. The thio-
methylene linker in 41 induced superior PREP inhibition and selectivity,
but FAP potency was measured to be negligible with respect to PREP.
Remarkably, replacing a methylene group with a thiomethylene group
made the analog 42 approximately 7-fold more active and 8-fold more
selective compared to 29. Contrasting outcome was recorded in case of
compound 43. The para-trifluoromethyl phenyl derivative 44 with a
thiomethylene linker stood out as the most selective (SI = 46.2) with
moderate FAP inhibitory action compared to 24. Generally, introduction
4

H.J. Jung et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127846
of sulfone in place of sulfur produced enhancement of FAP potency in
the compounds with electron withdrawing substituents, and reverse
impact was observed in case of compounds 45–48 with electron-
donating substituents. Intriguingly, complete oxidation of the thio-
methylene linker in 41 brought about an abrupt change in selectivity
making the sulfone derivative 45, a favorable inhibitor of FAP. The di-
oxide derivatives 47 was also found to be highly active and selective.
Loss of both activity and selectivity were observed in case of compound
48 as compared to 42. Replacement of sulfur with oxygen also provided
a drastic modification in affinity towards FAP in compound 49 as
compared to 41. The para-cyano derivative 50 containing a methyl-
eneoxy linker exhibited maximum FAP inhibitory potency and selec-
tivity among the similar members (40, 43, 46 and 50).
To evaluate the in vivo therapeutic effect of BR102910 (compound
24) as a FAP inhibitor, C57BL/6J mice (n = 5 per group) were orally
administered with two dosages such as 10 and 30 mg/kg. Blood samples
were collected, and the FAP inhibition was measured at five time in-
tervals. As shown in Fig. 4, BR102910 (compound 24) showed signifi-
cant FAP inhibition in a dose dependent manner.
In order to confirm the effect in treating type 2 diabetes, compound
24 was studied in oral glucose tolerance test (OGTT), as shown in Fig. 5.
Single oral administration of 24 with dosages as 20 and 50 mg/kg in two
groups of 8 weeks-old ob/ob mice, 60 min prior to glucose challenge
reduced the subsequent plasma glucose levels in a dose-dependent
manner compared to the control group (Fig. 5A). Since the cleavage
by FAP does not take place in mice, rhFGF21 was administered by
intraperitoneal injection into the control group after treating with
vehicle to figure out the reaction in actual human body. It is reported
that the amount of undigested FGF21 is augmented when the C-terminal
cleavage of FGF21 is inhibited. As shown in Fig. 5B, we observed the
similar result on the FAP inhibition effect in in vivo experiment by using
ob/ob mice. These results indicate that the FAP inhibition by treatment
of BR102910 (compound 24) is a promising therapeutic approach for
type 2 diabetes.
The binding mode of BR102910 (compound 24) in the x-ray crystal
Fig. 5. Blood glucose profiles in ob/ob mice via FAP inhibition by compound
24. 8-Week-old ob/ob mice were treated with compound 24 (BR102910) or
vehicle before ip treatment of rhFGF21. (A) Non fasting blood glucose levels
were measured at the indicated time points. (B) Plasma FAP activity was
measured by fluorometric assay using AMC substrate at 1, 3, 6 and 9 h.
structure of human fibroblast activation protein-
α
was also predicted
¨
using the Schrodinger program. Four main interactions are observed as
shown in Fig. 6. One of hydrogen bonding interactions is observed be-
tween the nitrogen of cyanide at the pyrrolidine ring of 24 and hydroxyl
of the amino acid (AA) residue S624 of FAP. Second hydrogen bonding
interaction is between the oxygen of carbonyl group attached to thiazole
moiety and one amine of guanidine at the AA residue R123. A third
hydrogen bonding interaction can be observed between the carbonyl
oxygen attached to the pyrrolidine ring of 24 and iminium hydrogen of
the histidine734 AA residue. The fourth interaction is a
π-stacking
interaction between thiazole ring of 24 and indole moiety of amino acid
residue W623.
In summary, we have successfully identified a novel FAP inhibitor,
BR102910 (compound 24), with high FAP inhibitory activity and
Fig. 6. Predicted binding mode of BR102910 in crystal structure of human FAP
(PDB code: 1Z68). Hydrogen bonding interactions are shown in red and
π
-stacking interaction in cyan. Docked molecule and the interacting AA residues
are highlighted by stick model.
selectivity (PREP/FAP, SI = 24.5). In addition, BR102910 (compound
24) demonstrated excellent level of in vivo FAP inhibition activity.
Moreover, this compound displayed the reduction of blood glucose
Fig. 4. Plasma FAP profile of compound 24 in C57BL/6J mice.
5

H.J. Jung et al.
Bioorganic & Medicinal Chemistry Letters 37 (2021) 127846
levels in animal model. Based on excellent in vitro and in vivo profiles,
BR102910 (compound 24) can be developed as a lead candidate for the
treatment of type 2 diabetes.
References
1
2
3
4
(a) Scanlan MJ, Raj BK, Calvo B, et al. PNAS. 1994;91:5657.(b) Welt S, Divgi CR,
Scott AM, et al. J Clin Oncol. 1994;12:1193.
(a) Kharitonenkov A, Adams AC. Mol Metab. 2014;3:221.(b) Zhen EY, Jin Z,
Ackermann BL, et al. Biochem J. 2016;473:605.
Declaration of Competing Interest
´
(a) Aertgeerts K, Levin I, Shi L, et al. J Biol Chem. 2005;280:19441.(b) Polgar L. Cell
Mol Life Sci. 2002;59:349.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
(a) Wang H, Wu Q, Liu Z, et al. Cell Death Dis. 2014;5, e1155.(b) Levy MT,
McCaughan GW, Abbott CA, et al. Hepatology. 1999;29:1768.(c) Fan M-H, Zhu Q,
Li H-H, et al. J Biol Chem. 2016;291:8070.(d) Liu R, Li H, Liu L, et al. Cancer Biol
Ther. 2012;13:123.
5
6
7
Cheng JD, Dunbrack Jr RL, Valianou M, et al. Cancer Res. 2002;62:4767.
Dunshee DR, Bainbridge TW, KIjavin, NM, et al. J Biol Chem. 2016;291:5986.
(a) Owen BM, Mangelsdorf DJ, Kliewer SA. Trends Endocrinol Metab. 2015;26:22.
Acknowledgments
´
(b) Sanchez-Garrido MA, Habegger KM, Clemmensen C, et al. Mol Metab. 2016;5:
This work was supported by Boryung Pharmaceuticals Co. Ltd. and
the National Research Foundation of Korea (NRF) grant funded by the
Korea government (MSIT) (2018M3A9C8021792, 2019R1A4A2001451,
and 2020R1A2C3005357).
1015.
8
9
Micanovic R, Raches DW, Dunbar JD, et al. J Cell Physiol. 2009;219:227.
Deng H, Wang C-L, Lai J, et al. Hepat Mon. 2016;16, e40263.
10 Coppage AL, Heard KR, DiMare MT, et al. PLoS ONE. 2016;11, e0151269.
11 For selected publications: (a) Tsai T-Y, Yeh T-K, Chen X, et al. J Med Chem. 2010;53:
6572; (b) Jansen K, Winter HD, Heirbaut L, et al. Med. Chem Commun 2014;5:1700;
(c) Ryabtsova O, Jansen K, Goethem SV, et al. Bioorg Med Chem Lett. 2012;22:3412.
12 (a) Narra K, Mullins SR, Lee HO, et al. Cancer Biol Ther. 2007;6:169.(b) Eager RM,
Cunningham CC, Senzer NN, et al. BMC Cancer. 2009;9:263.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.127846.
13 Thomas L, Eckhardt M, Langkopf E, et al. J Pharmacol Exp Ther. 2008;325:175.
14 Poplawski SE, Lai JH, Li Y, et al. J Med Chem. 2013;56:3467.
15 Jansen K, Heirburt L, Verkerk R, et al. J Med Chem. 2014;57:3053.
16 Jansen K, Heirburt L, Cheng JD, et al. ACS Med Chem Lett. 2013;4:49.
6