Structure-based design and biological evaluation of novel 2-(indol-2-yl) thiazole derivatives as xanthine oxidase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2015.12.055

Inhibition of xanthine oxidase (XO) has obviously been a central concept for controlling hyperuricemia, which causes serious and painful inflammatory arthritis disease such as gout. We discovered a series of novel 2-(indol-2-yl)thiazole derivatives as XO inhibitors at the level of nanomolar activity. Structure-guided design using molecular modeling program (Accelrys Software program) provided an excellent basis for optimization of 2-(indol-2-yl)thiazole compounds. Structure-activity relationship indicated that hydrophobic alkoxy group (isopropoxy, cyclopentoxy) at 5-position and hydrogen binding acceptor (NO₂, CN) at 7-position of indole ring appear as critical functional groups. Among the compounds, 2-(7-nitro-5-isopropoxy-indol-2-yl)-4-methylthiazole-5-carboxylic acid (9m) exhibits the most potent XO inhibitory activity (IC₅₀value: 5.1 nM) and the excellent uric acid lowering activity in potassium oxonate induced hyperuricemic rat model.

Full text of DOI:10.1016/j.bmcl.2015.12.055

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure-based design and biological evaluation of novel 2-(indol-2-
yl) thiazole derivatives as xanthine oxidase inhibitors
Jeong Uk Song a,b, Jae Wan Jang b, Tae Hun Kim b, Heuisul Park b, Wan Su Park b, Sang-Hun Jung a,
,
⇑
Geun Tae Kim b,
⇑
a College of Pharmacy, Chungnam National University, Daejeon 305-764, Republic of Korea
b Research & Development, LG Life Sciences Ltd., 188 Munji-ro, Yuseong-gu, Daejeon 305-380, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Inhibition of xanthine oxidase (XO) has obviously been a central concept for controlling hyperuricemia,
which causes serious and painful inflammatory arthritis disease such as gout. We discovered a series of
novel 2-(indol-2-yl)thiazole derivatives as XO inhibitors at the level of nanomolar activity. Structure-
guided design using molecular modeling program (Accelrys Software program) provided an excellent
basis for optimization of 2-(indol-2-yl)thiazole compounds. Structure–activity relationship indicated that
hydrophobic alkoxy group (isopropoxy, cyclopentoxy) at 5-position and hydrogen binding acceptor (NO2,
CN) at 7-position of indole ring appear as critical functional groups. Among the compounds, 2-(7-nitro-5-
isopropoxy-indol-2-yl)-4-methylthiazole-5-carboxylic acid (9m) exhibits the most potent XO inhibitory
activity (IC50 value: 5.1 nM) and the excellent uric acid lowering activity in potassium oxonate induced
hyperuricemic rat model.
Received 15 September 2015
Revised 27 November 2015
Accepted 17 December 2015
Available online xxxx
Keywords:
Xanthine oxidase inhibitors
Hyperuricemia
Molecular modeling
Structure–activity relationship
2-(Indol-2-yl)thiazole
Ó 2016 Published by Elsevier Ltd.
The most common symptom of gout is painful acute gouty
arthritis and gout is apparently associated with hyperuricemia
which is defined as extracellular fluid urate supersaturation. Hype-
ruricemia shows a expansion of uric acid in the body and increases
the risk for precipitation of monosodium urate (MSU) crystals in
joints, connective tissues and organs including the kidneys.1,2 Gout
results from inflammatory responses to deposition of microcrystals
of monosodium urate derived from body fluids saturated with
urate.3 Reduction of the uric acid would appear the most important
in the treatment of gout. Therefore, antihyperuricemic drugs such
as allopurinol and uricosuric agents, have been widely prescribed
for patients with gout.3,4 Xanthine oxidase catalyzes the oxidation
of hypoxanthine to xanthine and can further catalyze the oxidation
of xanthine to uric acid.5,6 This enzyme plays an important role in
the metabolism of purines in humans. Inhibitors of xanthine oxi-
dase (XO) block conversion of xanthine to uric acid and are poten-
tially useful for treatment of hyperuricemia or gout.7,8
is an inhibitor of xanthine oxidase, it is structurally quite different
from allopurinol and has an alternate mechanism of enzyme
inhibition.10
Higher serum uric acid levels have been known to be associated
with metabolic syndrome. Because the excess of uric acid level
causes insulin resistance, hyperuricemia is related with metabolic
syndrome and type 2 diabetes. Also, an elevation of serum uric acid
in body is strongly associated with the development of hyperten-
sion and renal disease.11,12 Because hyperuricemia becomes the
cause of serious health problem, it requires more diversity of treat-
ment option. In the recent years, many synthetic scaffolds compris-
ing triazole,13 isocytosines,14,15 pyrimidone16 and xanthones17
have been reported to display XO inhibition in the literatures. For
this reason we have made an effort to discover novel XO inhibitors
based on 2-(indol-2-yl)thiazole 9 as shown in Figure 1.
Indole is an important scaffold of many natural and synthetic
products and is widely used in medicinal chemistry.18 Indole
derivatives are an useful class of heterocyclic compounds with a
wide range of biological activities in antidiabetic drug, anticancer,
antimicrobial, anti-inflammatory, antidiabetic agents.19,20 We have
applied the indole as a pharmacophore of XO inhibitor.
In recent years, febuxostat, a nonpurine selective inhibitor is
approved. Febuxostat inhibited both the oxidized and reduced
forms of xanthine oxidase, and was used for the management of
hyperuricemia in adults with chronic gout.9,10 Although febuxostat
Our continuous effort to find novel XO inhibitors, we conducted
a high-throughput screening (HTS) of in-house compound library.
We found a hit compound 9a with an IC50 value of 2.2
contains novel 2-(indol-2-yl)thiazole scaffold. In our previous
lM. This
⇑
Corresponding authors at: Tel.: +82 42 821 5939; fax: +82 42 823 6566 (S.-H.J.);
tel.: +82 42 866 2247; fax: +82 42 861 2566 (G.T.K.).
E-mail addresses: jungshh@cnu.ac.kr (S.-H. Jung), gtakim@lgls.com (G.T. Kim).
http://dx.doi.org/10.1016/j.bmcl.2015.12.055
0960-894X/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Song, J. U.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2015.12.055

2
J. U. Song et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
OH
N
HO
N
N
N
CN
O
N
S
N
O
N
N
N
H
N
H
CN
Febuxostat
(IC50= 3.1-5.5 nM)
FYX-051
(IC50= 5.3 nM)
Allopurinol
(IC50= 3.74 ¥ì )
M
CN
O
N
R1
NO2
R1
H
H
H3C
R2
N
S
H3C
H3C
HO
HN
N
S
N
N
S
N
N
R3
R3
HO
O
HO
N
R3
O
H
R2
O
O
Compound 10
(IC50= 3.0-12.3 nM)
Compound 9m
(IC50= 5.1 nM)
Pyrimidone
(IC50= 4.8-7.6 nM)
Compound 9
(IC50= 3.5-2200 nM)
Figure 1. Structures of allopurinol, febuxostat, FYX-051, pyrimidone, compound 9 and compound 9m as xanthine oxidase inhibitors.
study, 2-(indol-5-yl)thiazole derivatives, compound 10 were
investigated as XO inhibitor, which differ in structure from
2-(indol-2-yl)thiazole derivatives.21 The structural difference
enables 2-(indol-2-yl)thiazole derivatives to have additional
hydrogen bond interactions with the binding site residues. The
indole NH of 2-(indol-2-yl)thiazole derivatives interacts with
Glu802 of active site via hydrogen bond interaction, whereas that
of 2-(indol-5-yl)thiazole derivatives substituted by various
moieties heads toward the solvent. Thus, 2-(indol-2-yl)thiazole
derivatives benefit from the interaction energy in the binding site.
Furthermore, the interaction between cyanide (9j) and Asn768 did
not deteriorate through the positional changes of indole scaffold,
which again tells us its superiority (Fig. 2).
The general synthetic route to 9 is shown in Scheme 1. 1 was
reacted with concd HCl/NaNO2 and concd HCl/SnCl at room tem-
perature to furnish compound 2 in 45–50% yield. 2 and methyl
pyruvate were reacted with sodium acetate in methanol to afford
3 in 70–78% yield, and 3 were reacted with polyphosphoric acid
at 70 °C to provide 4 in 60–67% yield. Ester 4 was hydrolyzed with
sodium hydroxide in aqueous THF and methanol to obtain 5 in 90%
yield. Amide 6 was synthesized from acid 5 using HBTU [2-(1H-
benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate]
and ammonium chloride in 74–90% yield. Amide 6 was converted
to thioamide 7 by refluxing with Lawesson’s agent in THF in
87–95% yield. Thioamide 7 and ethyl 2-chloroacetoacetate were
reacted with a catalytic amount of pyridine by refluxing in ethanol
to afford indole-2-thiazole 8 in a reasonable yield of 80–85%. Ester
8 was hydrolyzed with sodium hydroxide in aqueous THF and
methanol, and the final compound 9 was obtained in a good yield.
In order to understand the binding conformation of the hit com-
pound 9a as a XO inhibitor, flexible molecular docking was carried
out into the active site of bovine milk XO.26,27 The compounds were
docked into the substrate binding pocket using CDOCKER module
in DiscoveryStudio4.1 (Accelrys Software Inc., US).27 Figure 3
shows the docking conformations of the hit compound at the bind-
ing site of XO. Docking score of compound 9a (À32.64 kcal/mol)
was less than score of febuxostat (À38.14 kcal/mol). Both com-
pounds shows important interactions between carboxylic acid
moiety and the residues known to be important for binding of
endogeneous substrate of XO, Arg880, and Thr1010. We can see
almost identical conformation of the carboxylic acid and thiazole
moiety of the hit to those of febuxostat. However, the conforma-
tion of indole moiety and its side chain is quite different from that
of febuxostat. As shown in Figure 3, cyano group of febuxostat
interacts with Asn768 and Lys771 via hydrogen bond interactions,
whereas the hit compound does not have such moiety. The lack of
interactions is attributed to the lower activity of the hit compound
at the XO binding pocket compared to febuxostat.
The established synthesis led us to evaluate in vitro xanthine
oxidase inhibition of 9. Assay of in vitro XO inhibition activity
was performed by measuring the inhibitor concentration needed
for 50% inhibition.22 Tables 1 and 2 represents the inhibitory activ-
ities against XO of the test compounds.
The proposed docking model provides us with a reasonable
explanation of modification in compounds. In Table 1, substituents
of chloro, bromo, nitro and cyano groups at R1-position led to more
than 10-fold gain in potency.25 First of all, in order to examine the
effect of the substitution at R2 and R3 positions of the indole ring,
analogs 9a–f were prepared to investigate their XO inhibitory
activities. 9a with hydrogen for R1 and methoxy group for R2
exhibited low IC50 value of 2200 nM, and 9b with methoxy group
for R3 slightly increased activity (IC50 = 792.0 nM). 9c (R2 = isobu-
toxy) exhibited IC50 value of 481.0 nM, and 9d (R3 = isobutoxy)
increased activity (IC50 = 209.0 nM). 9f (R3 = benzyloxy) with bulky
group exhibited IC50 value of 522.0 nM, and 9e (R3 = phenoxy)
increased activity (IC50 = 233.0 nM). Substitution for R3-position
rather than R2-position much increased in vitro activities. That
can be explained by hydrophobic space of proteins at R3-position.
Therefore, R3-position was fixed as phenoxy group for comparison.
In order to examine the effect of the substitution of hydrogen at
R1-position, analogs 9g–9j were prepared and their XO inhibitory
activities were investigated. Based on the docking model, chloro,
Figure 2. Binding of 9j (green) and the compound 10 shown in our previous Letter
(gray)21 at bovine milk XOR. The hydrogen bond distance between NH of indole and
Glu802 is shown as green color, and interacting residues are shown as stick form.
Please cite this article in press as: Song, J. U.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2015.12.055

J. U. Song et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
R1
R1
R1
O
H
N
R2
R3
R2
R3
R2
c
H2N
a
b
H2NHN
O
O
+
N
R3
O
O
1
3
2
R1
1H
N
R1
R
H
N
H
N
R2
R2
R3
O
R2
R3
O
d
O
e
f
R3
H2N
O
HO
4
7
5
6
R1
R1
R1
H
N
H
N
H
N
O
R2
R3
g
R2
R3
H3C
HO
O
R2
R3
h
H3C
S
N
S
N
S
+
O
O
H2N
Cl
O
O
9
8
Scheme 1. Reagents and conditions: (a) (i) Concd HCl/NaNO2, (ii) Concd HCl/SnCl2, 45–50%; (b) methyl pyruvate, sodium acetate, 70–78%; (c) polyphosphoric acid, 70 °C, 60–
67%; (d) THF/MeOH = 1:1, 1 N NaOH, 90%; (e) HBTU, NH4Cl, Et3N, DMF, 74–90%; (f) Lawesson’s agent, THF, reflux, 87–95%; (g) ethyl 2-chloroacetoacetate, pyridine, EtOH,
reflux, 80–85%; (h) THF/MeOH = 1:1, 1 N NaOH, 85–90%.
inhibitory activities of analogs 9i–t and their uric acid inhibitions
in rats were tested as shown in Table 2.
9i with nitro group for R1 and phenoxy group for R3 exhibited
IC50 value of 15.9 nM, and 9k with methoxy group for R3 showed
much increased inhibitory activity (IC50 = 7.0 nM). 9i showed weak
uric acid lowering activity (7.1%) because of low IC50 value of
15.9 nM and moderate Cmax of 3.20
acid lowering activity (4.0%) due to poor plasma exposure
(0.10
g/mL) after 1 h. 9l with sulfonamide group for R3 exhibited
lg/mL. 9k showed weak uric
l
the most potent IC50 value of 3.6 nM. However, low in vivo efficacy
(0.0%) was observed, which is caused by low plasma exposure
(0.017 l
g/mL) after 1 h. 9m with nitro group for R1 with iso-
propoxy group for R3 also displayed potent inhibitory activities
Table 1
IC50 and metabolic stability of compound 9 with various R1, R2 and R3
R1
H
Figure 3. Predicted binding mode of compound 9a and febuxostat. Overlap of
compound 9a (green) and febuxostat (gray) at the substrate binding pocket of
bovine milk XOR (PDB code: 1N5X). The substrate binding pocket is represented by
hydrophobicity plot, and the legend is displayed on the left. The interacting residues
are shown as stick representation. Docking results are visualized using CDOCKER
module in DiscoveryStudio4.1.27
R2
R3
H3C
HO
N
S
N
O
Compound R1
R2
R3
In vitro
Metabolic
IC50 (nM) stabilitya (half life,
min)
bromo, nitro and cyano groups are capable of forming hydrogen
bonds with Asn768, explaining a significant increase in inhibition
for 9g, 9h, 9i and 9j. Thiazole acid appears to have week hydrogen
bond between Arg880 and Thr1010. The thiazole ring that interacts
9a
9b
H
H
H
H
2200.0
792.0
NTb
NT
O
H
O
with Glu802 appears to have p–p stacking interaction with Phe914
O
9c
9d
9e
H
H
H
481.0
209.0
233.0
NT
and Phe1009. Non-bonded interactions are also observed with
amino acid residues, Leu648, Phe649, Leu873, Val1011, Phe1013
and Leu1014 in the active site of XO. 9g (IC50 = 14.3 nM) and 9h
(IC50 = 17.0 nM) with chloro or bromo for R1 increased activity
more than 10-fold. Also, 9i (IC50 = 15.9 nM) and 9j (IC50 = 8.9 nM)
with nitro or cyano for R1 remarkably enhanced in vitro activities.
When metabolic stability was studied in rat liver microsomes, 9i
and 9j showed good stability. Therefore, we focused on the modi-
fication of 9i and 9j comprising nitro and cyano groups with vari-
ous substituents for R3-position.
O
H
H
96.8
116.3
O
O
9f
H
H
522.0
NT
9g
9h
9i
Cl
H
H
H
H
14.3
17.0
15.9
281.7
91.4
O
O
O
O
Br
In order to examine the effect of the substitution at R1 and
R3-positions of the indole ring, analogs 9i–t were prepared. In addition
to exploring the effect of nitro and cyano groups at R1-position, we
were also interested in examining substitution of alkoxy and
amine series at R3-position. The hydrophilic groups would be
appropriate for the solvent-exposed region in XO enzyme. XO
NO2
CN
242.0
9j
8.9
220.5
174.0
Febuxostat
3.1–5.5
a
Metabolic stability was studied in rat liver microsomes (RLM).
NT: not tested.
b
Please cite this article in press as: Song, J. U.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2015.12.055

4
J. U. Song et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Table 2
The effect of R1 and R3 in compound 9 on xanthine oxidase inhibition
R1
H
H3C
N
S
N
HO
R3
O
Compound R1
R3
In
Metabolic
stabilitya
(half life,
min)
Uric acid
vitro
IC50
(nM)
inhibition (%)b
at 1 h, 10 mg/kg
9i
NO2
15.9
7.0
242.0
24.1
7.1
4.0
O
9k
NO2
NO2
O
H
N
S
9l
3.6
873.2
0.0
O
O
9m
9n
9o
9p
NO2
CN
5.1
247.6
130.2
124.1
NTc
43.0
19.0
2.3
O
O
Figure 4. Predicted binding mode of compound 9m and febuxostat. Overlap of
compound 9m (red) and febuxostat (gray). The substrate binding pocket is
represented by hydrophobicity plot, and the legend is displayed on the left. The
interacting residues are shown as stick representation. Docking results are
visualized using CDOCKER module in DiscoveryStudio4.1.27
6.1
CN
5.7
O
CN
31.8
NT
NH2
9q
9r
CN
CN
CN
9.4
9.3
6.9
177.8
90.2
NT
NT
NT
N
H
Table 3
Pharmacokinetic data of 9m in SD ratsa
N
H
Pharmacokinetic parameters
Dose (mg/kg)
p.o.
10
N
9s
150.5
Cmax (
AUC0–last
AUC0–infinity
Tmax (h)
l
g/mL)
4.94
Febuxostat
3.1–
5.5
(l
g h/mL)
8.23
13.14
0.42
(l
g h/mL)
a
b
c
Metabolic stability was studied in rat liver microsomes (RLM).
Uric acid inhibition was studied in SD rats at 10 mg/kg by oral route.
NT: not tested.
t1/2 (h)
3.45
a
n = 3.
(IC50 = 5.1 nM). When 9m was orally administrated in rats, high
Cmax, 4.94 g/mL was observed. Therefore, 9m exhibited strong
uric acid lowering activity (43.0%). We turned our interest to XO
inhibition activity of the substitution with cyano group at R1 posi-
tion. 9n with cyano group for R1 and isopropoxy group for R3
exhibited good IC50 value of 6.1 nM. But, relatively low Cmax of
4
3
2
1
0
l
3.75
lg/mL compared to 9m was observed, so it showed low
***
in vivo efficacy (19.0%). 9o with cyclopentoxy group for R3 exhib-
***
ited potent IC50 value of 5.7 nM. But low Cmax of 0.72 lg/mL com-
pare to 9m was observed. 9p with amino group for R3 exhibited
poor activity (IC50 = 31.8 nM). 9q (R3 = cycloropentyl amine)
showed moderate inhibitory activity of 9.4 nM. Moderate IC50 val-
ues of 9.3 nM and 6.9 nM were observed in 9r (R3 = benzyl amine)
and 9s (R3 = piperidine), respectively. 9q and 9s indicated low Cmax
values of 0.30 lg/mL and 0.74 lg/mL, respectively. Among test
compounds for XO inhibition, 9m with nitro group for R1 and iso-
propoxy group for R3 was found to display well-balanced and
excellent in vitro activity and in vivo efficacy.
The moiety that gives compound 9m with the highest activity
among the analogs is nitro group at R1-position, which shows
hydrogen bonding interactions with Asn768. Branched chains for
R3 showed better docking mode and activity than R2-position
because they are well placed in hydrophobic region of proteins.
The docked mode of compound 9l, 9m, 9n and 9o exhibited
geometric interactions similar to those of febuxostat. As shown
in Figure 4, compound 9m efficiently matched the substrate
binding pocket of the enzyme. Overall binding mode of 9m was
found to be similar to that observed with febuxostat.27 As reflected
in our binding energy, 9m (À35.50 kcal/mol) also docked in the
binding pocket in a more stabilized fashion than 9a. The hydrogen
Figure 5. Plasma uric acid concentration (mg/dl) in hyperuricemic rats at 10 mg/kg,
orally administered. (***P <0.001: ANOVA and post Dunnet’s test).
bond interactions between cyano group and the residues, such as
Asn768 and Lys771, shown in febuxostat, are also shown in
compound 9m. This important interaction of 9m was derived from
nitro group at R1. In addition, the hydrophobic region occupied by
isobutoxy group of febuxostat is filled by R3 substituent of 9m,
which is also hydrophobic. Thus, the binding mode effectively
shows that R1 and R3-positions of indole moiety in our scaffold
are important for inhibitory activity.
Please cite this article in press as: Song, J. U.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2015.12.055

J. U. Song et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
5
13. Sato, T.; Ashizawa, N.; Iwanaga, T.; Nakamura, H.; Matsumoto, K.; Inoue, T.;
Nagata, O. Bioorg. Med. Chem. Lett. 2009, 19, 184.
Pharmacokinetic profile of 9m was investigated in Sprague–
Dawley (SD) rats at a single 10 mg/kg dose by oral route (p.o.) over
24 h. Pharmacokinetic parameters of 9m were described in Table 3,
showing high Cmax of 4.94 lg/mL, AUC0–last of 8.23 lg h/mL and
half-life (t1/2) of 3.45 h. 9m exhibited excellent oral bioavailability
14. Khanna, S.; Burudkar, S.; Bajaj, K.; Shah, P.; Keche, A.; Ghosh, U.; Chandrika, B.-
Rao; Sharma, R.; Sivaramakrishnan, H. Bioorg. Med. Chem. Lett. 2012, 22, 7543.
15. Bajaj, K.; Burudkar, S.; Shah, P.; Keche, A.; Ghosh, U.; Tannu, P.; Khanna, S.;
Chandrika, B.-Rao; Sharma, R.; Sivaramakrishnan, H. Bioorg. Med. Chem. Lett.
2013, 23, 834.
16. Evenas, J.; Edfeldt, F.; Lepisto, M.; Svitacheva, N.; Synnergren, A.; Lundquist, B.;
Granse, M.; Tjornebo, A.; Narjes, F. Bioorg. Med. Chem. Lett. 2014, 24, 1315.
17. Hu, L.; Hu, H.; Wu, W.; Chai, X.; Luo, J.; Wu, Q. Bioorg. Med. Chem. Lett. 2011, 21,
4013.
18. Zhang, M.-Z.; Mulholland, N.; Beattie, D.; Irwin, D.; Gu, Y.-C.; Chen, Q.; Yang, G.-
F.; Clough, J. Eur. J. Med. Chem. 2013, 63, 22.
19. Acton, J. J.; Akiyama, T. E.; Chang, C. H.; Colwell, L.; Debenham, S.; Doebber, T.;
Einstein, M.; Liu, K.; McCann, M. E.; Moller, D. E.; Muise, E. S.; Tan, Y.;
Thompson, J. R.; Wong, K. K.; Wu, M.; Xu, L.; Meinke, P. T.; Berger, J. P.; Wood,
H. B. J. Med. Chem. 2009, 52, 3846.
20. Hwang, I. K.; Yoo, K. Y.; Li, H.; Park, O. K.; Lee, C. H.; Choi, J. H.; Won, M.-H. J.
Neurosci. Res. 2009, 87, 2126.
21. Song, J. U.; Choi, S. P.; Kim, T. H.; Jung, C. K.; Lee, J. Y.; Jung, S. H.; Kim, G. T.
Bioorg. Med. Chem. Lett. 2015, 25, 1254.
22. Assay of xanthine oxidase inhibition activity was performed as follows.
Xanthine oxidase originated from butter milk (Sigma) was adjusted to 0.05 U/
mL by using 50 mM phosphate buffer, and then added to 96-well plate at
and good in vivo efficacy.
In order to estimate the plasma uric acid-lowering ability,
experiments were carried out using oxonic acid-induced high-uric
acid model.23,24 Figure 5 represents plasma uric acid concentration.
Each uric acid inhibition rate of 9l, 9m, 9n and febuxostat (10 mg/
kg) at 1 h after oral administration (putting the plasma uric acid
level of normal group as 100% inhibition, and plasma uric acid level
of control group as 0% inhibition) was 0.0%, 43.0%, 19.0% and 60%.
As revealed by the experiment, 9m exerts an excellent inhibitory
effect on XO. Therefore, 9m can be used as an agent for the treat-
ment and prevention of the diseases associated with human xan-
thine oxidase such as hyperuricemia and gout.
In conclusion, we have developed a novel class of potent XO
inhibitors using structure based design and the activity data of
all compounds are summarized in Tables 1 and 2. The important
interaction with nitro and cyano groups of compound 9 and
Asn768 was critical for the inhibition of XO. We also demonstrated
that computational modeling with known binding site information
could provide an useful strategy to discover new inhibitor classes
incorporating 2-(indol-2-yl)thiazole scaffolds.
20
compounds of various concentrations (10% DMSO phosphate buffer) were
added at 20 L/well, and pre-incubated for 3 min at room temperature.
Immediately after 200 M xanthine solution was added at 20 L/well, the
lL/well. 50 mM phosphate buffer was added at 140 lL/well, and test
l
l
l
initial velocity of uric acid formation was determined at 293 nm using
microplate reader SpectraMax spectrophotometer. Linear Vmax during 5 min
for each test compound concentration was determined as initial velocity. In
addition, the initial velocity of the test compound at each concentration was
converted to % inhibition rate on the basis of the initial velocity in the absence
of the inhibitor, thereby the inhibitor concentration needed for 50% inhibition
was calculated as IC50 values. IC50 values of 9l, 9m, 9n and 9o showed the
similar levels to febuxostat. Tables 1 and 2 represent the inhibitory activities
against XO of the test compound.
Acknowledgment
We would like to thank Dr. Dongchul Lim for his continuous
support to this structure-guided drug discovery project.
23. Experiments were carried out using oxonic acid-induced high-uric acid model.
Potassium oxonate (300 mg/kg) suspended in 0.8% carboxymethylcellulose
solution was intraperitoneally administered to male SD rats of body weight
200 g. One hour after the oxonic acid administration, test compounds dissolved
References and notes
in
a solution of poly(ethylene glycol) 400 and ethanol (2:1) were orally
administered, and after the time period of 1 h, blood was sampled. Plasma was
separated from obtained blood and the uric acid concentration in the plasma
was quantified utilizing LC–MS/MS.
1. Shoji, A.; Yamanaka, H.; Kamatani, N. Arthritis Rheum. 2004, 51, 321.
2. Choi, H. K.; Ford, E. S. Am. J. Med. 2007, 120, 442.
3. Adams, J. U. Nat. Biotechnol. 2009, 27, 309.
24. Osada, Y.; Tsuchimoto, M.; Fukushima, H.; Takahashi, K.; Kondo, S.; Hasegawa,
M.; Komoriya, K. Eur. J. Pharmacol. 1993, 241, 183.
4. Suresh, E. Postgrad. Med. J. 2005, 81, 572.
25. Hachioji, S. K.; Hino, H. F.; Hino, M. H.; Hino, M. T.; Hino I. N.; Hino Y. O.;
Hachioji, K. K.; Hino H. Y. U.S. Patent 5,614,520, 1997.
26. Okamoto, K.; Eger, B. T.; Nishino, T.; Kondo, S.; Pai, E. F. J. Biol. Chem. 2003, 278,
1848.
27. The docking study of 9m was performed using the X-ray crystal structure of
bovine milk XO (PDB code 1N5X). Since the residues around the binding pocket
are highly conserved in bovine and human XO, our findings are also applicable
to human XO. The compounds were docked into the substrate binding pocket
using CDOCKER module in DiscoveryStudio4.1 (Accelrys Software Inc., US).
5. Emmerson, B. T. N. Engl. J. Med. 1996, 334, 445.
6. Okamoto, K.; Eger, B. T.; Nishino, T.; Pai, E. F.; Nishino, T. Nucleosides,
Nucleotides Nucleic Acids 2008, 27, 888.
7. Pacher, P.; Nivorozhkin, A.; Szabó, C. Pharmacol. Rev. 2006, 58, 87.
8. Okamoto, K.; Eger, B. T.; Nishino, T.; Kondo, S.; Pai, E. F.; Nishino, T. J. Biol. Chem.
2003, 278, 1848.
9. Takano, Y.; Hase-Aoki, K.; Horiuchi, H.; Zhao, L.; Kasahara, Y.; Kondo, S.; Becker,
M. A. Life Sci. 2005, 76, 1835.
10. Ernst, M. E.; Fravel, M. A. Clin. Ther. 2009, 31, 2503.
11. Li, C.; Hsieh, M.-C.; Chang, S.-J. Curr. Opin. Rheumatol. 2013, 25, 210.
12. Johnson, R. J.; Kang, D.-H.; Feig, D.; Kivlighn, S.; Kanellis, J.; Watanabe, S.;
Tuttle, K. R.; Rodriguez-Iturbe, B.; Herrera-Acosta, J.; Mazzali, M. Hypertension
2003, 41, 1183.
Please cite this article in press as: Song, J. U.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2015.12.055