Structure and activity relationship of 2-(substituted benzoyl)- hydroxyindoles as novel CaMKII inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.01.012

A series of novel 2-substituted-5-hydroxyindoles were synthesized and evaluated for their inhibitory activity against CaMKII. Structure and activity relationship results indicated that potent inhibitory activity could be achieved by modification at the para-position of the phenyl ring of the high throughput screening hit compound 2. Among the prepared compounds, we identified 14 as a novel CaMKII inhibitor with an activity stronger than that of KN-93, a known CaMKII inhibitor.

Full text of DOI:10.1016/j.bmcl.2011.01.012

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1456–1458
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure and activity relationship of 2-(substituted benzoyl)-hydroxyindoles
as novel CaMKII inhibitors
⇑
Masafumi Komiya , Shigehiro Asano, Nobuyuki Koike, Erina Koga, Junetsu Igarashi, Shogo Nakatani,
Yoshiaki Isobe
Chemistry Research Laboratories, Dainippon Sumitomo Pharma Co., Ltd, 3-1-98 Kasugade Naka, Konohana-ku, Osaka 554-0022, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel 2-substituted-5-hydroxyindoles were synthesized and evaluated for their inhibitory
activity against CaMKII. Structure and activity relationship results indicated that potent inhibitory activ-
ity could be achieved by modification at the para-position of the phenyl ring of the high throughput
screening hit compound 2. Among the prepared compounds, we identified 14 as a novel CaMKII inhibitor
with an activity stronger than that of KN-93, a known CaMKII inhibitor.
Received 8 October 2010
Revised 27 December 2010
Accepted 5 January 2011
Available online 8 January 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
CaMKII
Indole
Kinase
Inhibitor
Calmodulin
Calcium (Ca2+) is an important intracellular messenger, control-
ling a diverse range of cellular processes, such as apoptosis, ion
channel and cell cycle regulation, and cellular response to oxida-
tive stress.1,2 A rise in intracellular Ca2+ concentration leads to
binding of Ca2+ ions to calmodulin (CaM), which binds to and acti-
vates Ca2+/CaM-dependent protein kinases (CaMKs). CaMKs, which
are ubiquitous serine/threonine kinases classified into three sub-
types (I, II, and IV), modulate many cellular functions in response
to intracellular Ca2+ levels.3,4 CaMKII, a member of CaMKs family,
assembles into a complex of dodecamers with four isoforms
CaMKII was sixfold weaker than that of 1 (2; IC50 = 10
lM vs 1;
IC50 = 1.6 M). Based on the structure of compound 2, we initiated
l
a structure and activity relationship (SAR) study to improve the
inhibitory activity for CaMKII. Here, we report the synthesis and
biological evaluation of a novel series of 2-(substituted benzoyl)-
hydroxyindoles.
The synthetic routes for the indole compounds are described in
Schemes 1 and 2. The various indole compounds were prepared
from the commercially available indoles 3–5. According to the re-
ported method,16 NH group in 3–5 was protected by using carbon
dioxide, followed by treatment with tert-butyllithium to afford 2-
lithium carbanion. Nucleophilic addition of the carbanion to an
appropriate benzaldehyde (R2–C6H4CHO), followed by oxidation
of the hydroxyl group with MnO2 produced the 2-acylindole deriv-
atives. The 2-acylindole derivatives except 16–18 underwent
deprotection of tert-butyldimethylsilyl (TBS) group with tetrabu-
tylammonium fluoride (TBAF) to afford 6–15. Hydrogenation of
compound 18, which was prepared in a similar manner above, gave
the key intermediate 19 in good yield. Mitsunobu reaction of 19
(a, b, c, and d), having each a subunit composed of three main
parts; catalytic, regulatory and association domains.5–7 CaMKII is
well established for its modulating effects on synaptic plasticity
and processes like learning and memory.8 In addition, CaMKII plays
a role in osteoclast differentiation and bone resorption,9 and active
CaMKII is known to enhance proliferation and cytotoxic activity of
T cells.10
CaM-competitive inhibitor KN(93) (1)11 and autocamtide-2-re-
lated inhibitory peptide (AIP)12 are well-known CaMKII inhibitors.
Recently, Ca2+/CaM antagonists13 and CaM non-competitive inhib-
itors14 were reported. We considered CaMKII to be a good target
for anti-inflammatory agents. In search for potent CaMKII inhibi-
tors, we have recently run a high throughput screening (HTS) cam-
paign of our library compounds and found compound 215 as a hit
with a novel structure (Fig. 1). The inhibitory activity of 2 against
O O
N
HO
N
S
O
OMe
HN
OH
Cl
1 (KN-93)
2
⇑
Corresponding author.
E-mail address: masafumi-komiya@ds-pharma.co.jp (M. Komiya).
Figure 1. Structures of 1 and HTS hit compound 2.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.012

M. Komiya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1456–1458
1457
O
O
2'
R1
X
OMe
3'
5
a,b,c
R1
5
X
Y
a,b
OHC
R2
Y
HN
HN
N
H
OH
6
N
H
4'
O
6
32
33
34
35
(X=N, Y= H)
(X=CH, Y=N)
(X=N, Y= H)
1
1
2
3
4
R =H
6
7
R =5-OH, R =4'-F
(X=CH, Y=N)
1
1
2
R =5-OTBS
R =5-OH, R =4'-Cl
5 R1=6-OTBS
8 R1=5-OH, R2=4'-Me
Scheme 3. Reagents and conditions: (a) 4-bromodiphenylether (8.0 equiv), n-BuLi
(8.2 equiv), THF, ꢀ10 °C, 2 h, 38–74%; (b) MnO2 (10 equiv), CHCl3, rt, 1 h, 50–81%;
(c) BBr3 (4.0 equiv), CH2Cl2, rt, 6 h, 12–71%.
9 R1=5-OH, R2=4'-OMe
1
2
10
11
12
R =5-OH, R =4'-OCF3
R =5-OH, R =4'-Ph
1
2
1
2
R =5-OH, R =2'-OPh
a
13 R1=5-OH, R2=3'-OPh
The inhibitory activity of the synthesized indole compounds to-
ward CaMKII is shown in Table 1.18,19 First, we examined the effect
of a substituent on the phenyl moiety (A). An electron-donating or
electron-withdrawing group was introduced to investigate for sub-
stitution at the para position, but no clear electronic effect was ob-
served (6–10). However, the inhibitory activity for CaMKII was
slightly improved in the para phenyl compound 11 compared to
14 R1=5-OH, R2=4'-OPh
1
2
15
R =6-OH, R =4'-OPh
O
O
d,b
R1
HN
HN
R2
R2
OH
1
2
16
17
R =H, R =OPh
2
the hit compound 2 (11; IC50 = 6.4 lM). Surprisingly, the phenoxy
20
O
O
R =
1
2
compound 1420 demonstrated the most potent inhibitory activity
with an IC50 value about 10-fold greater than that of 11 (14;
R =OTBS, R =OPh
18 R1=OTBS, R2=OBn
19 R1=OTBS, R2=OH
21 R2=
c
IC50 = 0.61 lM). From these results we speculated that sterically
2
22
23
R = O
large alkoxy substituents are favorable for better activity. Regard
to the position of the phenoxy group in the phenyl ring (A), the
ortho 12 and meta 13 showed decreased activity compared with
para 14, indicating that the substituent should be in the para-posi-
tion. Based on these findings, we prepared and evaluated various
alkoxy compounds at the para-position, in particular, those with
a sterically large alkoxy group. Compounds 20 and 21 exhibited
moderate activity with IC50 values in the single micromolar range,
whereas compounds with small alkoxy groups, such as a methoxy
(9) or trifluoromethoxy group (10), showed abolished activity. This
confirmed our hypothesis that introduction of sterically large alk-
oxy substituents improves the activity. As both the phenoxy and
cycloalkoxy groups are lipophilic, we therefore changed these
O
2
O
O
O
NBoc
R =
e
2
NH
24 R =
2
25
NMe
NAc
R =
26 R2=
O
Scheme 1. Reagents and conditions: (a) (i) n-BuLi (1.2 equiv), CO2(s), THF, ꢀ78 °C,
1 h; (ii) t-BuLi (1.1 equiv); (iii) R2–C6H4CHO (1.1 equiv), ꢀ78 °C to rt, 2 h; (iv) MnO2
(10 equiv), acetone, rt, overnight, 30–83%; (b) TBAF (1.1 equiv), THF, rt, 1 h, 45–95%;
(c) 10% Pd/C (25 wt %), H2, THF, EtOAc, rt, 4 h, 64%; (d) alcohols (1.5 equiv), PPh3
(2.0 equiv), diisopropyl azodicarboxylate (2.0 equiv), THF, rt, overnight, 48–85%;
(e) trifluoroacetic acid, CHCl3, rt, overnight, 68%.
Table 1
Inhibitory activity of indole derivatives against CaMKII
O
O
O
A
R2
B
2'
3'
a,b
R1
5
HN
17
N
OTBS
OH
HN
O
O
R3
4'
6
3
27
28
29
R =SO2Ph
3
Compd
R1
R2
CaMKII IC50 (lM)
R =Me
3
R =CH2CH2OH
1
2
6
7
8
1.6
10
>10
>10
8.8
>10
>10
6.4
1.8
1.7
0.61
2.3
7.5
30 R3=CH2CO2Et
31 R3=CH2CO2H
5-OH
5-OH
5-OH
5-OH
5-OH
5-OH
5-OH
5-OH
5-OH
5-OH
6-OH
H
H
40-F
40-Cl
40-Me
40-OMe
40-OCF3
40-Ph
20-OPh
30-OPh
40-OPh
40-OPh
40-OPh
c
Scheme 2. Reagents and conditions: (a) R3X (1.1 equiv), NaH (1.1 equiv), DMF,
0–50 °C, 4 h, 26–75% in two steps; (b) TBAF (1.1 equiv), THF, rt, 1 h; (c) 1 M NaOH aq
(1.2 equiv), THF, MeOH, rt, overnight, 51%.
9
10
11
12
13
14
15
16
with an appropriate alcohol followed by deprotection of TBS and/or
Boc afforded 20–26 (Scheme 1). Benzenesulfonylation or alkylation
of 17 with the corresponding halides followed by deprotection of
TBS group gave 27–30, and then hydrolysis of the formed ethyl es-
ter compound 30 gave compound 31 (Scheme 2).
The synthetic route for the azaindole compounds 34 and 35 is
shown in Scheme 3. Aldehyde 32 and 33 were prepared according
to a reported method.17 The aldehyde derivatives were reacted
with 4-lithium diphenylether, which generated by halogen–lith-
ium exchange reaction using n-BuLi. Oxidation of the formed hy-
droxyl group followed by deprotection of the methyl group with
BBr3 gave 34 and 35.
20
21
5-OH
5-OH
3.9
2.3
4'-O
4'-O
4'-O
22
5-OH
6.8
O
24
25
26
5-OH
5-OH
5-OH
>10
>10
9.9
4'-O
4'-O
4'-O
NH
NMe
NAc

1458
M. Komiya et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1456–1458
Table 2
tained by HTS identified the optimal substituents, which are the
hydroxyl group as R1 and the para-phenoxy group as R2. Actually,
compound 14 showed the most potent inhibitory activity against
Inhibitory activity of substituted 5-hydroxyindole derivatives against CaMKII
O
CaMKII enzyme with IC50 value of 0.61 lM, and the inhibitory
activity was stronger than that of 1, a known CaMKII inhibitor.
X
N
OH
O
R3
Y
R3
X
Y
CaMKII IC50 (lM)
References and notes
Compd
14
27
28
29
30
31
34
35
1
H
SO2Ph
Me
CH2CH2OH
CH2CO2Et
CH2CO2H
H
CH
CH
CH
CH
CH
CH
N
CH
CH
CH
CH
CH
CH
CH
N
0.61
>10
1.3
1.9
0.77
1.8
>10
6.8
1.6
1. Bootman, M. D.; Collins, T. J.; Peppiatt, C. M.; Prothero, L. S.; MacKenzie, L.; De
Smet, P.; Travers, M.; Tovey, S. C.; Seo, J. T.; Berridge, M. J.; Ciccolini, F.; Lipp, P.
Semin. Cell Dev. Biol. 2001, 12, 3.
2. Howe, C. J.; Lahair, M. M.; Mccubrey, J. A.; Franklin, R. A. J. Biol. Chem. 2004, 279,
44573.
3. Hook, S. S. Annu. Rev. Pharmacol. Toxicol. 2001, 41, 471.
4. Braun, A. P.; Schulman, H. Ann. Rev. Physiol. 1995, 57, 417.
5. Hanson, P. I.; Schulman, H. Annu. Rev. Biochem. 1992, 12, 559.
6. Colbran, R. J.; Soderling, T. R. Curr. Top. Cell. Regul. 1990, 279, 181.
7. Gaertner, T. R.; Kolodziej, S. J.; Wang, D.; Kobayashi, R.; Koomen, J. M.; Stoops, J.
K.; Waxham, M. N. J. Biol. Chem. 2004, 279, 12484.
H
CH
8. Wayman, G. A.; Lee, Y. S.; Tokumitsu, H.; Silva, A.; Soderling, T. R. Neuron 2008,
59, 914.
9. Ang, E. S.; Zhang, P.; Steer, J. H.; Tan, J. W.; Yip, K.; Zheng, M. H.; Joyce, D. A.; Xu,
J. J. Cell. Physiol. 2007, 212, 787.
10. Lin, M. Y.; Zal, T.; Ch’en, I. L.; Gascoigne, N. R. J.; Hedrick, S. M. J. Immunol. 2005,
174, 5583.
11. Sumi, M.; Kiuchi, K.; Ishikawa, T.; Ishii, A.; Hagiwara, M.; Nagatsu, T.; Hidaka, H.
Biochem. Biophys. Res. Commun. 1991, 181, 968.
12. Ishida, A.; Kameshita, I.; Okuno, S.; Kitani, T.; Fujisawa, H. Biochem. Biophys. Res.
Commun. 1995, 212, 806.
13. Colomer, J.; Schmitt, A. A.; Toone, E. J.; Means, A. R. Biochemistry 2010, 49, 4244.
14. Asano, S.; Komiya, M.; Koike, N.; Koga, E.; Nakatani, S.; Isobe, Y. Bioorg. Med.
Chem. Lett. 2010, 20, 6696.
groups to find out whether substituent size or lipophilicity is
important for the activity. Compound 22 where the carbon atom
in the cyclopentyl ring was displaced with an oxygen atom, gave
a twofold decreased activity. Conversion of the cyclohexyl ring to
a piperidine or substituted piperidine resulted in a large drop in
the activity (21 vs 24–26). These results indicate that lipophilic
large substituents are preferable as R2 substituents. In addition,
the results of 14 and 21 suggest that the planarity of R2 substituent
is important for CaMKII inhibition. On the other hand, regard to the
position of the hydroxyl group in the indole moiety (B), compound
15 with the hydroxyl group at 6th position resulted in fourfold
drop in activity, and replacement with a hydrogen atom also gave
decreased activity (12-fold). These findings confirmed that the hy-
droxyl group in the indole moiety is important for CaMKII inhibi-
tion, and that its placement in the 5th position is better.
Second, we focused on the effects of a substitution in the indole
moiety (R3), and the results are shown in Table 2. Substitution with
an electron-withdrawing group, such as a benzensulfonyl group
(27), showed in diminished inhibitory activity. Among the substi-
tuted alkyl compounds (28–31), the methyl (28), 2-hydroxyethyl
(29), and carboxymethyl (31) showed a 2–3-fold loss of activity,
leading us to speculate that the acidic proton at the N(1)-position
is important for CaMKII inhibition. However, IC50 value of the
ethoxycarbonylmethyl (30) was almost equivalent to that of 14,
suggesting that further optimization studies are necessary to find
the best substituent.
15. Igarashi, S.; Isaka, M.; Inami, H.; Hara, H.; Kamitoku, H. JP 07145147.
16. Katritzky, A. R.; Akutagawa, K. Tetrahedron Lett. 1985, 26, 5935.
17. Frydman, B.; Reil, S. J.; Boned, J.; Rapoport, H. J. Org. Chem. 1968, 33, 3762.
18. Expression and purification of CaMKII: CaMKII was purified according to the
method of Brickey et al. (Biochem. Biophys. Res. Commun. 1990, 173, 578) with
some modifications. cDNA encoding CaMKII was sub-cloned into the transfer
vector pFastBac1. Bacmids were generated in DH10Bac cells, and CaMKII
recombinant baculovirus stocks were prepared according to the protocol of
Bac-to-Bac Baculovirus Expression System (invitrogen, USA). Sf9 cells were
maintained in Sf900 II medium and infected with CaMKII recombinant
baculovirus stocks. The cells were harvested 3 days post-infection and
CaMKII was affinity purified using Calmodulin Sepharose 4B (GE healthcare).
19. CaMKII kinase assay: Autocamtide-2 and CaM were purchased from Millipore,
and
c-
33P ATP was obtained from GE Healthcare. ATP was purchased from
Sigma–Aldrich. The test substances and purified CaMKII were added to the
assay buffer containing 25 mM Tris–HCl, pH 7.5, 2 mM dithiothreitol, 10 mM
MgCl2, 0.1% CHAPS, and 1 mM CaCl2. Autocamtide-2 was diluted to a final
concentration of 10
CaM (500 M, 100
ATP (final concentration 10
added to the diluted test substances and the whole was incubated for 20 min at
30 °C. After incubation, the reaction was terminated by transfer of a 16
aliquot onto the appropriate area of a P30 Filtermat (PerkinElmer, USA), and
the labeled substrate was captured by negatively charged filtermat.
l
l
M with assay buffer containing varying concentrations of
M, 50 M, 25 M, 12.5 M, 6.25 M, and 3.125 M), and
and 350–1500 cpm/pmol
33P ATP) was
l
l
l
l
l
l
lM
c-
lL
Finally, in order to clarify the importance of the hydroxyl group
at the 5th position, we modified the indole ring to azaindole. Both
34 and 35 exhibited decreased or diminished activity compared to
the indole 14. We therefore considered two possibilities to explain
these results. One is that the lipophilic ring is better for the activity,
and the other is that the pyridone form is generally stable in 34 and
35. Further optimization work to confirm these two ideas is still
ongoing and will be presented in a future paper.
a
Phosphorylation was linear with respect to time under these conditions. The
filtermat was washed three times, each for 5 min with 75 mM phosphoric acid,
dried, and then spotted by MicroScintO (PerkinElmer, USA). The radioactivity
was counted in TopCount NXT (PerkinElmer, USA). CPM counts were calculated
for transfer of a phosphate group per minute per 1 mg CaMKII.
20. Compound 14: Mp = 162–163 °C; 1H NMR (CDCl3, 300 MHz) d 4.73 (1H, s),
6.91–7.24 (8H, m), 7.34–7.45 (3H, m), 8.01 (2H, m), 9.16 (1H, s); 13C NMR
(CDCl3, 75 MHz) d 106.1, 111.5, 113.1, 117.4, 120.2, 124.6, 128.3, 130.1, 131.5,
132.3, 133.0, 135.1, 150.2, 155.6, 161.6, 185.8; IR (ATR) 3300, 1618 cmꢀ1; MS
(ESI) m/z 330 (M+1); Anal. Calcd for C21H15NO3ꢁ0.2H2O: C, 75.75; H, 4.66; N,
4.21. Found: C, 75.96; H, 4.55; N, 4.27.
In summary, we disclosed in this report a novel class of CaMKII
inhibitors. An extensive SAR study based on hit compound 2 ob-