Identification and characterization of potent, selective and metabolically stable IKKβ inhibitor

Article abstract of DOI:10.1016/j.bmcl.2016.01.065

We have previously reported the identification of a rhodanine compound (1) with well-balanced inhibitory activity against IKKβ and collagen-induced TNFα activated cells. However, we need more optimized compounds because of its instability over plasma and microsome. As part of a program directed toward the optimization of IKKβ inhibitor, we modified a substituent of parent compound to a series of functional groups. Among substituted compounds, fluorine substituent (12) on the para position of phenyl ring restored the stability toward plasma and microsome while retaining inhibitory potency and selectivity against IKKβ over other kinases. Also, we have demonstrated that compound 12 is an ATP non-competitive inhibitor and safe enough to apply to animal experiment from an acute toxicity test.

Full text of DOI:10.1016/j.bmcl.2016.01.065

Accepted Manuscript
Identification and characterization of potent, selective and metabolically stable
IKKβ inhibitor
Doyeon Kim, Yun Gyeong Kim, Jae Hong Seo, Kye Jung Shin
PII:
S0960-894X(16)30066-X
http://dx.doi.org/10.1016/j.bmcl.2016.01.065
BMCL 23530
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
22 December 2015
19 January 2016
22 January 2016
Please cite this article as: Kim, D., Kim, Y.G., Seo, J.H., Shin, K.J., Identification and characterization of potent,
selective and metabolically stable IKKβ inhibitor, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://
dx.doi.org/10.1016/j.bmcl.2016.01.065
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.

Bioorganic & Medicinal Chemistry Letters
Identification and characterization of potent, selective and
metabolically stable IKK inhibitor
Doyeon Kim, Yun Gyeong Kim, Jae Hong Seo and Kye Jung Shin*
Integrated Research Institute of Pharmaceutical Sciences, College of Pharmacy, The Catholic University of
Korea, 43 Jibong-ro, Wonmi-gu, Bucheon, Gyeonggi-do, 14662, Republic of Korea
Correspondence to:
Name
: Prof. Kye Jung Shin
Tel/Fax No.
E-mail Address
: +82-2-2164-4060/ +82-2-2164-4059
: kyejung@catholic.ac.kr

Abstract
We have previously reported the identification of a rhodanine compound (1) with well-
balanced inhibitory activity against IKK and collagen-induced TNF activated cells.
However, we need more optimized compounds because of its instability over plasma and
microsome. As part of a program directed toward the optimization of IKK inhibitor, we
modified a substituent of parent compound to a series of functional groups. Among
substituted compounds, fluorine substituent (12) on the para position of phenyl ring restored
the stability toward plasma and microsome while retaining inhibitory potency and selectivity
against IKK over other kinases. Also, we have demonstrated that compound 12 is an ATP
non-competitive inhibitor and safe enough to apply to animal experiment from an acute
toxicity test.

The transcription factor called the nuclear factor-B (NF-B) is related to inflammatory and
immune response, and the regulation of NF-B-mediated transcription is effective in the
treatment of autoimmune and inflammatory diseases such as rheumatoid arthritis (RA).1 NF-
B signaling pathway is triggered when the activation of IKK complex is induced by viral
infection, lipopolysaccharide (LPS) and pro-inflammatory cytokines such as TNF, IL-1.2 In
unstimulated cell, NF-B is isolated in the cytoplasm as an inactive complex with inhibitory
protein, IB.3 IB is phosphorylated by the IKK complex which is composed of three
subunits: the catalytic subunits IKK and IKK, and the regulatory subunit IKK (NEMO)4,
leading to poly-ubiquitination and degradation through the 26S proteosome.5 Among IKK
complex, IKKis responsible for phosphorylation of the IB in the signal induced pathway
leading to NF-B activation.6 Also, it is known that IKK rather than IKK is attractive
target for treatment of the inflammatory and autoimmune diseases because of a dominant role
in NF-B signaling pathway.7
In the previous communication,8 we described the identification of rhodanine compound 1 as
a novel IKK inhibitor by hit-to-lead strategy. Compound 1 showed good selective inhibition
against IKK over other kinases, and well-balanced inhibitory activity against IKK- and
collagen-induced TNF activated cells. Prior to in vivo animal study with compound 1 in
type-II collagen-induced rheumatoid arthritis animal model, we performed the rat plasma and
human hepatic microsomal stability test. However, the stability of compound 1 in plasma and
microsome was less than satisfactory to the application for in vivo test; the remaining % of
compound 1 after 120 min in rat plasma stability test was 58%, and the half life of compound
1 in human microsomal stability test was 72 min. Thus, we carried out the structural
modification of compound 1 to obtain more stable compounds in plasma and microsome
without the loss of selectivity and inhibitory activity against IKKIn this paper, we describe

the synthesis and biological properties of a series of rhodanine analogues. We also report a
few evaluation studies to determine the possibility of drug candidate for a representative
compound 12.
N
N
O
O
S
NH2
N
S
part B
O
O
1
part A
Figure 1. The structure of compound 1
We focused on our initial efforts to replace the carboxamido substituent of phenylether (part
A) in compound 1 with a variety of functional groups since the instability of compound 1 in
plasma and microsome was considered to be caused by the hydrolysis of carboxamide to acid.
In the previous paper,8 4-methylpiperazinopropyloxyphenyl moiety in part B was the
optimized structure in the SAR studies. On the basis of these results, we synthesized a series
of modified compounds in part A using similar method described in the earlier
communication with the fixed 4-methylpiperazinylpropyloxyphenyl moiety in part B
(Scheme 1), and their biological activities against IKK and TNF activated cells were
outlined in Table1.

N
N
a
N
O
N
O
S
N
NH2
S
3
2
O
N
b
N
O
S
R
N
S
O
O
4 ~ 19
Scheme 1. Reagents and conditions: (a) bis(carboxymethyl)trithiocarbonate, EtOH/H2O (4:3), reflux, 56%; (b)
aldehydes, NaOAc, AcOH, reflux, 59~86%.
N
N
O
S
R
N
S
O
O
Table 1. IKK inhibitory activities of compounds 4-19
enzyme assay9
cell-based assay8
compd
R
% inhibition of
% inhibition of
TNFIC50
(M)
IKKIC50
(µM)
IKK (10 µM)
NF-B (10 M)
COOH
CON(CH3)2
SO2NH2
CN
50.2
85.7
83.7
82.3
75.3
67.0
17.0
27.2
93.3
88.7
78.9
80.8
9.88
2.10
0.60
0.64
0.84
1.13
-
4
5
-
-
20
-
4
32.3
-
6
55.0
51.6
50.1
-
7
OCH3
CH3
4
8
5
9
CF3
-
-
0.5
3
10
11
12
13
14
15
t-Bu
-
-
F
0.79
0.82
0.96
0.90
84.8
66.6
76.6
60.9
Cl
Br
6
SO2CH3
4

NO2
NH2
89.4
83.8
86.6
91.6
90.1
0.79
0.85
0.84
0.82
0.35
61.1
68.8
48.0
54.0
65.4
2
6
16
17
18
19
1
NHCOCH3
NHSO2CH3
CONH2
20
8
2
The assay methods for enzyme and cell-lines are described in ref. 8 and 9.
As shown in Table 1, compound 4 which is a hydrolyzed form of compound 1 greatly
reduced the potency against IKK compared to compound 1 and showed no activity against
cell-lines. The replacement of carboxamide with dimethylcarboxamide (5) which is more
stable and bulkier than simple amide 1 was detrimental to enzymatic and cellular activities.
Bioisosteric transformation to sulfonamide (6) was found to inhibit IKKenzyme at low
concentration comparable with compound 1, however, its cellular activity was not observed
in cell-based assay. A series of compounds (7-17) were prepared for investigating the change
of activity according to the electronic effect. Substitution with electron withdrawing groups
such as nitrile (7), halogens (12-14), methanesulfonyl (15) and nitro (16) except for
trifluoromethyl group (10) maintained the activities against enzyme and cell-lines, especially,
fluoro substituent compound (12) showed better cellular activity than compound 1. Electron
donating substituents (8, 9, 17) exhibited good activities with IC50 values ranging from 1 to
0.8 M except for bulky t-butyl substituent. These results indicated that the electronic effects
on para position did not impact significantly the potency of parent compound 1, but rather
the bulkiness was more likely to affect on activities. Chemical modification of compound 1 to
reversed amides (18, 19) was tolerated in enzyme activities, but led to somewhat decrease in
cellular potency. In the study on substituent effect at part A, fluorine substituted compound 12
was the most potent representative of this small series against enzyme and cell.
To determine whether the substituted compounds instead of carboxamido functional group

could enhance the stability over plasma and microsome, the selected compounds (7, 8, 12-16)
which revealed good activities were performed to stability tests (Table 2).
Table 2. The stability test data of selected compound
Plasma stabilitya Microsomal stabilityb
Compd
(%)
49
(min)
72.9
7
73
87
78
72
72
45
58
223.5
239.0
126.0
157.5
70.5
8
12
13
14
15
16
1
57.8
72.0
aThe remaining % of compound after 120 min in rat plasma stability test
bThe half life of compound in human microsomal stability test
As illustrated in Table 2, the stability of selected compounds was evaluated with rat plasma
and human microsome. Nitrile (7), methanesulfonyl (15) and nitro (16) substituents were
very unstable to microsome and(or) plasma presumably due to the high polarity and reactivity
toward liver microsome, meanwhile, the remarkable metabolic stability enhancement
compared to parent molecule 1 was observed in methoxy (8) and halogen (12-14) substituents,
especially, fluorine substituent (12) had stable enough to plasma and 3-fold longer half-life to
microsome compared with compound 1. From the results of inhibitory activity and metabolic
stability, fluorine compound 12 had equipotent IKK inhibitory activity and excellent
metabolic stability in comparison with compound 1. Taken results together, compound 1210
was selected to be a program lead for further biological evaluation.
In the drug discovery process of kinase inhibitor, kinase selectivity profiling is an essential
element to allow more accurate evaluation of on- and off-target effects of the inhibitor that
can be expected in relation to efficacy and toxicity possibly working on other kinases.11 Next

efforts were made to perform the kinase panel test against more than 40 other kinases with
the lead compound 12 in 10 M concentration, and the results of % inhibition data are
summarized in Table 3.
Table 3. Kinase selectivity of compound 12
Kinase
% inhibitiona
Kinase
% inhibitiona
Abl1
Aurora A
Aurora B
b-Raf
20.0
0
0
2.0
0
IKK
JNK1
0
0
JNK2
JNK3
0
c-Src
14.0
58.6
3.1
6.4
17.3
5.5
12.9
2.5
0
LCK
0
LYN/LYN A
MAPKAPK2
MAPKAPK5
MNK1
27.8
42.0
5.2
0
33.1
27.1
0
CAMK1
CDK1/cyclin B
CDK2/cyclin A
CDK4/cyclin D1
CDK5/p25
CDK6/cyclin D1
CDK7/cyclin H
CDK9/cyclin T1
CHK1
MNK2
MSK1
MSK2
P8a/MAPK14
P70S6K
PDK1
14.8
24.9
0
0
47.0
0
CHK2
PIM1
0
CK1
0
PIM2
0
CK2
EGFR
0
PKA
0
EPHA2
29.0
6.8
20.5
0
18.4
61.1
0
57.5
24.7
93.3
PKC
ROCK1
TBK1
TIE2/TEK
TPKA
ERK2/MAPK2
FGFR1
GSK3
IKK
0
IKK
aDuplicate mode at 10 M of compound 12
As described in Table 3, compound 12 showed excellent selectivity over most of kinases
regardless of serine/threonine or tyrosine kinase. It is noteworthy that there was a big
difference between IKK and IKK in inhibitory activity of compound 12. Generally, the
majority of kinase inhibitors located in the ATP-binding site, and bind to the hinge region
through hydrogen bonding which involves a bidentate interaction with a donor-acceptor pair
pattern. However, compound 12 structurally does not contain the functional group to

participate in a bidentate hydrogen bonding. We conceived that the excellent selectivity of
compound 12 came from binding to allosteric region of IKK. To determine what type of
inhibition was occurring to IKK, enzyme kinetics study was carried out. We examined the
Lineweaver-Burk plot to determine whether compound 12 of IKK–catalyzed reaction was
competitive or non-competitive (allosteric). The reciprocals of the reaction rate and ATP
concentration were plotted with compound 12 being present, and its plot was displayed in
Figure 2.
Reaction time: 1 h, Detection time: 15 h
Substrate (FAM-IKBα-derived peptide): 10 μM
Compound 12: 0, 0.5, 1 μM
ATP: 0.625, 1.25, 2.5, 5, 10 μM
Figure 2. The double reciprocal plot for ATP concetration in IKK inhibition by compound 12
In the Lineweaver-Burk plot of compound 12, the lines had the same intercept point on the x-
axis, but had different slopes and different intercepts on the y-axis, which is a typical plot of
non-competitive inhibitor. With this result, we could infer that compound 12 is a non-
competitive IKK inhibitor and its excellent selectivity over other kinases is brought by

binding to allosteric region not ATP-binding site.
Next, to observe acute toxicity and animal mortality, and provide a reference for clinical trials,
compound 12 were orally administered to ICR mice. Twenty ICR mice were given 2000
mg/kg of compound 12 by single oral dose and closely observed 3 hours after administration
for 14-day period, and were weighed daily before and after administration. After the
observation period, all surviving animals underwent gross anatomy. Throughout the
experiment period, no animal was died, all animals did not show any abnormal symptoms,
and the average weight of male and female animals were normal (Table 4). Major organs had
no visible changes via gross anatomy in all of sacrificed animals at the15th day. From this
acute toxicity experiment, compound 12 was found to be safe in the single 2000 mg/kg dose
by oral administration for ICR mice.
Table 4. Average body weight after single oral dose of compound 12 in ICR mice（g， ±SD）
Before
administration
Group Number
Day1
Day2
Day3
Day7
Day10
Day14
Female
Male
10
10
18.9±0.9
19.7±0.9 21.7±1.4 23.3±1.2 28.9±1.3 31.3±1.1 35.1±1.4
21.5±1.1 22.1±1.2 23.2±1.0 26.2±1.6 27.4±1.7 29.2±1.5
20.3±0.4
To examine in vivo kinetics of compound 12, the pharmacokinetic properties, after dissolving
in 20% HP-β-cyclodextron, were evaluated intravenously and orally in rat (Table 5).
Compound 12 had moderate clearance, large volume distribution, and long half-life after
intravenous and oral administration, while had a relatively low oral AUC value resulting from
low Cmax level compared to intravenous administration. The bioavailability of compound 12
was 23% which was a possible level to administer orally.

Table 5. Pharmacokinetic profiles of compound 12
Parameter
Intravenousa
Oralb
AUC0–∞ (ng h/mL)
T1/2 (h)
175.6±41.3
5.50±2.20
93.99±9.96
5.88±1.22
44.33±10.35
79.4±9.8
5.81±2.16
8.6±2.5
Cmax (ng/mL)
CL (L/h/kg)
V (L/kg)
5.86±0.74
50.58±25.54
F (%)
23.0
aIV dose: 1 mg/kg
bPO dose: 2 mg/kg
Formulation: 20% HP-β-cyclodextron (yellow transparent solution)
Species: male SD rat (n = 3 per 0.25, 0.5, 1, 2, 4, 8, 12, 24 h time point)
In summary, we established the structure activity relationship for a series of rhodanine
compounds for IKK inhibitors. We showed that the metabolic instability of compound 1
could be overcome by the introduction of appropriate substituents on the phenyl ring instead
of carboxamido substituent which was a metabolically labile substituent. Among substituted
compounds, fluorine substituent (12) on the para position of phenyl ring restored the stability
toward plasma and microsome while retaining inhibitory potency and selectivity against
IKK over other kinases. Also, we have demonstrated that compound 12 is an ATP non-
competitive inhibitor and safe enough to apply to animal experiment from an acute toxicity
test. In conclusion, compound 12 has a high potential of drug candidate in the development of
therapeutics for the treatment of NF-B associated immune disease such as rheumatoid
arthritis, Crohn’s disease, chronic obstructive pulmonary disease (COPD), and cancer.
Acknowledgements
This work was financially supported by Research Fund 2012 of The Catholic University of
Korea and the Korean Health Technology R&D Project, Ministry of Health & Welfare,
Republic of Korea. (HI12C1087).

References and notes
1. (a) Breedveld, F. C.; Dayer, J. M. Ann. Rheum. Dis. 2000, 59, 841. (b) Yamamoto, Y.;
Gaynor, R. B. Trends Biochem. Sci. 2004, 29, 72, (c) Karin, M.; Yamamoto, Y.; Wang,
Q. M. Nat. Rev. Drug Discov. 2004, 3, 17, (d) Strnad, J.; Burke, J. R. Trends Pharmacol.
Sci. 2007, 28, 142, (e) Coish, P. D. G.; Wickens, P. L.; Lowinger, T. B. Expert Opin. Ther.
Patents 2006, 16, 1.
2. (a) Barnes, P. J.; Karin, M. N. Engl. J. Med. 1997, 336, 1066. (b) Morwick, T.; Berry, A.;
Brickwood, J.; Cardozo, M.; Catron, K.; DeTuri, M.; Emeigh, J.; Homon, C.; Hrapchak,
M.; Jacober, S.; Jakes, S.; Kaplita, P.; Kelly, T. A.; Ksiazek, J.; Liuzzi, M.; Magolda, R.;
Mao, C.; Marshall, D.; McNeil, D.; Prokopowicz, A. 3rd; Sarko, C.; Scouten, E.;
Sledziona, C.; Sun, S.; Watrous, J.; Wu, J. P.; Cywin, C. L. J. Med. Chem. 2006, 49,
2898.
3. (a) Tak, P. P.; Firestein, G. S. J. Clin. Invest. 2001, 107, 7. (b) Ghosh, S.; Karin, M. Cell
2002, 109, S81.
4. Gilmore, T. D. Oncogene 2006, 25, 6680.
5. Karin, M. Oncogene 1999, 18, 6867.
6. Li, Z. W.; Chu, W. M.; Hu, Y. L.; Delhase, M.; Deerinck, T.; Ellisman, M.; Johnson, R.;
Karin, M. J. Exp. Med. 1999, 189, 1839.
7. Schmid, J. A.; Birbach, A. Cytokine Growth Factor Rev. 2008, 19, 157.
8. Song , H.; Lee, Y. S.; Roh, E. J.; Seo, J. H.; Oh, K. S.; Lee, B. H.; Han, H.; Shin, K. J.
Bioorg. Med. Chem. Lett. 2012, 22, 5668.
9. Oh, K. S.; Lee, S.; Choi, J. K.; Lee, B. H. Comb. Chem. High Throughput Screening
2010, 13, 790.
10. Compound 12 (2HCl salt): 1H NMR (500 MHz, DMSO-d6)  2.26 (bs, 2H), 2.83 (s, 3H),
3.33−3.83 (m, 10H), 4.16 (t, J = 5.7 Hz, 2H), 7.11 (d, J = 8.0 Hz, 2H), 7.13 (d, J = 8.9
Hz, 2H), 7.23 (dd, J = 4.6, 9.1 Hz, 2H), 7.30−7.35 (m, 4H), 7.72 (d, J = 8.9 Hz, 2H), 7.83

13
(s, 1H), 12.07 (bs, 2H); C NMR (125 MHz, DMSO-d6)  23.81, 42.55, 48.71, 50.11,
53.65, 65.66, 115.48, 117.35, 117.53, 118.44, 121.93, 122.57, 122.64, 128.19, 128.26,
130.42, 132.58, 133.49, 151.38, 151.40, 158.49, 159.23, 160.17, 160.41, 167.54, 194.55.
MS: calcd for C30H30FN3O3 S2+ H+, 564.18; found [M + H]+, 564.14.
11. (a) Davies, S. P.; Reddy, H.; Caivano, M.; Cohen, P. Biochem. J. 2000, 351, 95. (b) Bain,
J.; Plater, L.; Elliott, M.; Shpiro, N.; Hastie, C. J.; Mclauchlan, H.; Klevernic, I.; Arthur,
J. S. C.; Alessi, D. R.; Cohen, P. Biochem. J. 2007, 408, 297.

Identification and characterization of potent, selective and metabolically stable IKK
inhibitor
Doyeon Kim, Yun Gyeong Kim, Jae Hong Seo and Kye Jung Shin
We describe an optimized compound 12 to have desired activity against IKK, good stability to
plasma and microsome, and excellent selectivity over other kinases.
N
N
O
S
F
N
S
O
O
12
Remaining % of rat plasma stability after 120 min: 87%
Half life in human microsomal stability: 239 min
IKK IC50: 0.79 M, TNF IC50: 0.5 M
Acute toxicity: >2,000 mg/kg