Aryl-indolyl maleimides as inhibitors of CaMKIIδ. Part 2: SAR of the amine tether

Article abstract of DOI:10.1016/j.bmcl.2008.02.058

A family of aryl-substituted maleimides was prepared and studied for their activity against calmodulin-dependant kinase. Inhibitory activities against the enzyme ranged from 34 nM to >20 μM and were dependant upon both the nature of the aryl group and the tether joining the basic amine to the indolyl maleimide core. Key interactions with the kinase ATP site and hinge region, predicted by homology modeling, were confirmed.

Full text of DOI:10.1016/j.bmcl.2008.02.058

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2395–2398
Aryl-indolyl maleimides as inhibitors of CaMKIId.
Part 2: SAR of the amine tether
Daniel E. Levy,* Dan-Xiong Wang, Qing Lu, Zheng Chen, John Perumattam,
Yong-jin Xu, Jeffrey Higaki, Hanmin Dong, Albert Liclican,
Maureen Laney, Babu Mavunkel and Sundeep Dugar
Scios, Inc., 6500 Paseo Padre Parkway, Fremont, CA 94555, USA
Received 6 December 2007; revised 21 February 2008; accepted 22 February 2008
Available online 4 March 2008
Abstract—A family of aryl-substituted maleimides was prepared and studied for their activity against calmodulin-dependant kinase.
Inhibitory activities against the enzyme ranged from 34 nM to >20 lM and were dependant upon both the nature of the aryl group
and the tether joining the basic amine to the indolyl maleimide core. Key interactions with the kinase ATP site and hinge region,
predicted by homology modeling, were confirmed.
Ó 2008 Elsevier Ltd. All rights reserved.
H
Calcium is critical to cardiac1–3 and neuropathic2,4,5 sig-
N
O
O
naling pathways. Additionally, calcium is an important
second messenger controlling apoptosis, cell cycle regu-
lation, gene expression, and hormone signaling pro-
cesses.6 To induce these responses, calcium forms a
complex with calmodulin in order to and activate the
family of Ca2+/calmodulin-dependant protein kinases
(CaMKs).6
Br
N
N
H
H2N
1
The family of CaMKs consists of three types, among
which multiple tissue specific isoforms are known. For
example, CaMKIIa and b are found primarily in neural
tissue7 while CaMKIId is found in cardiac tissue.8 Previ-
ous work in our group9 identified compound 1 (Fig. 1)
as a 34 nM inhibitor of CaMKIId.
Figure 1. Lead aryl-indolyl maleimide.
Glu100
O
N
H
H
Hydrophobic
Pocket
As shown in Figure 2, compound 1 was docked into a
homology model of CaMKIId and key interactions
within the ATP binding site and hinge region, previously
identified, were verified.9,10 Specific interactions include
dual hydrogen bonds between the Glu100 hinge back-
bone and the maleimide as well as a salt bridge between
Glu106 and the terminal amine. Additionally, optimal
utilization of the hydrophobic binding pocket by the
bromoindole was noted.
N
O
O
Br
N
N
H
Hydrophobic
Pocket
H
H
N
O
Glu106
H
O
Keywords: Calmodulin; Kinase; Maleimides; Inhibitors.
*
Figure 2. Interactions between compound 1 and the CaMKIId
catalytic site as indicated by homology modeling.
Corresponding author. Tel.: +1 650 704 3051; e-mail: del345@
gmail.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.058

2396
D. E. Levy et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2395–2398
Table 1. SAR indicating the relationship between the amino group
and potency
Based on the observation that the amine is tethered to
the indolyl maleimide core via a flexible linker, our
group became interested in the possibility of improving
inhibitory activity by constraining the tether or utilizing
additional substituents. In order to realize this goal, pre-
viously described chemistry9 was employed where aryl
glyoxalates were reacted with aryl acetamides to yield
the desired maleimides under basic conditions (Scheme
1).11 This chemistry reduced the synthetic problems to
the preparation of suitably functionalized indole-3-
acetamides or methyl indole-3-glyoxalates.
H
N
O
O
R1
N
R2
Compound R1
R2
IC50 (lM)
7a13
7g13
CH2CH2CH2NH2 0.38 (n = 1)
1.19 (n = 1)
Referring to the structures listed in Tables 1 and 2, com-
pounds 6a–c, 7b–f, and compounds 7h–i were prepared
according to Scheme 1 using acetamides prepared from
commercially available arylacetic acids as shown in
Scheme 2. Of the required glyoxylates, commercially
available methyl 3-indoleglyoxylate and methyl (1-meth-
ylindole)-3-glyoxylate were utilized. The 1-(3-aminopro-
pylindole) glyoxylate required for compounds 7a–d was
prepared as previously described.9
H
N
H
Br
7b
7e
CH2CH2CH2NH2 0.034 0.009 (n = 3)
CH3 0.87 (n = 1)
N
H
7c
7f
CH2CH2CH2NH2 1.35 0.64 (n = 2)
CH3
H
>20 (n = 1)
>20 (n = 1)
Regarding the structures in Table 4, the required indole-
glyoxylates were prepared from commercially available
3-(1-indole)- and 4-(1-indole)piperidine. As shown in
Scheme 3, protection of the piperidine nitrogen was fol-
lowed by treatment with oxalyl chloride and methanol
giving the desired compounds 2.
7h
7d
7i
CH2CH2CH2NH2 0.18 0.06 (n = 3)
3.53 (n = 1)
Br
H
Having addressed various patterns of rigidity and the
structural requirements of the amine-tether combina-
tion, interest turned to actual substitutions on the tether
itself. In accessing this SAR, compound 7q (Table 5) was
prepared using the chemistry described in Scheme 4. As
Table 2. SAR indicating the dependence of amine basicity on potency
H
N
O
O
R1
OCH3
O
H2N
O
O
N
R2
+
Aryl
Compound R1
R2
IC50 (lM)
H
3
N
N
O
O
(a)
(a)
Tether
7a13
6a
CH2CH2CH2NH2
0.38 (n = 1)
Boc/Bn-Amine
CH2CH2CH2NHBoc 3.81 0.25 (n = 2)
Aryl
2
N
H
NH2
H3CO
O
O
O
N
Br
Tether
7b
6b
CH2CH2CH2NH2
0.034 0.009 (n = 3)
+
Aryl
Boc/Bn-Amine
CH2CH2CH2NHBoc 7.54 (n = 1)
6
N
H
5
N
Tether
Boc/Bn-Amine
4
(b) or (c)
7c
6c
CH2CH2CH2NH2
1.35 0.64 (n = 2)
H
CH2CH2CH2NHBoc >20 (n = 2)
N
O
O
Aryl
illustrated, compound 7h (prepared using previously de-
scribed chemistry)9 was first reacted with epibromohy-
drin and then with ammonia thus providing the
desired 1,2-amino alcohol.
N
Tether
Amine . HCl
7
IC50 values12 for all tested compounds are listed in the
following tables. As shown in Table 1, the importance
Scheme 1. Reagents: (a) tBuOK, THF, 40–80%; (b) HCl, dioxane,
MeOH, 100%; (c) H2, 10% Pd/C, MeOH, 50–80%.

D. E. Levy et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2395–2398
2397
Table 3. SAR related to the incorporation of cyclic amines connected
to the indole via flexible tethers
HO
H2N
O
O
H
N
(a)
Aryl
Aryl
O
O
8
3
Scheme 2. Reagents: (a) CDI, THF then NH3 (0.5 M in MeOH), 94%.
N
N
H
OCH3
O
R
Compound
R
IC50 (lM)
O
7a13
0.38 (n = 1)
(a)
(b)
N
N
N
NH2
(CH2)y
(CH2)y
(CH2)y
NBn
(CH2)x
(CH2)x
(CH2)x
NH
NBn
7j14
7k14
7l13
2.02 (n = 1)
1.45 (n = 1)
1.89 (n = 1)
N
9
10
2
NH
Scheme 3. x = 1, y = 2 or x = 2, y = 1. Reagents: (a) BnCl, K2CO3,
H2O, CH2Cl2, 60–70%; (b) oxalyl chloride, CH2Cl2 then MeOH, 70–
80%.
N
H
H
N
H
N
O
O
O
O
N
CH3
(a) then (b)
N
N
H
HO
into a more rigid cyclic structure. The data, summarized
in Table 3, indicate that such modifications result in po-
tency losses ranging from 5- to 8-fold based on the par-
ent tethered primary amine. These results may simply
result from the inhibitors’ inabilities to present the
amine groups to Glu106 while maintaining low-energy
conformations.
7h
7q
H2N
Scheme 4. Reagents and condition: (a) NaH, DMF, epibromohydrin;
(b) NH3/H2O, 100 °C, sealed tube, 24% (2 steps).
of an amine tethered to the indole moiety was confirmed
via the study of a series of amines as compared to in-
doles with N–CH3 or N–H substitutions. The data illus-
trate a drop in potency from 3- to 20-fold over the
corresponding amine-based compounds.
Continuing with the exploration into the entropic effects
of flexible tethers, rigidified tethers were incorporated
into our studies. As illustrated in Table 4, such modifica-
tions were generally well tolerated with equipotent activ-
ities noted between the parent structure and the 3-
piperidyl analog. Conversion of a cyclic secondary
amine to a bulky tertiary amine did result in further loss
in potency presumably due to an interruption in the
ability of the inhibitor to present the rigid amine to
Glu106 in a low-energy conformation.
As an extension to the importance of an amino group
tethered to the indole, the importance of basicity was
addressed via a comparison of the potency of amine-
based analogs to their corresponding NH-Boc deriva-
tives. The data, shown in Table 2 and summarized from
a previous report,9 demonstrate that elimination of the
amine basicity results in potency losses ranging from
10- to 220-fold. While one can argue that the steric bulk
of the Boc group could be interfering with the ability of
the inhibitors to bind in the active site, homology mod-
eling indicates that the salt bridge is near the solvent
front and that there is sufficient space available to
accommodate large groups. Furthermore, if binding
was truly impaired, one might expect substantially larger
reductions in potency for parent compounds that do not
possess optimized hydrophobic groups such as 5-
bromoindole.
Finally, with respect to direct substitution of the tether
with additional functional groups, addition of a hydro-
xyl group demonstrated the potential for increased
activity. As shown in Table 5, incorporation of this
group onto the parent phenyl maleimide resulted in a
3-fold increase in potency. While this is only a represen-
tative example, this result supports further exploration
in this area. Specifically, the stereochemistry and the
nature of the functional group substitution are of
interest.
In summary, synthetic and commercially available aryl-
indolyl maleimides were studied for their activity
against CaMKIId. A specific emphasis was placed on
In dealing with entropic factors, a study was conducted
in order to elucidate the effect of tying the amine group

2398
D. E. Levy et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2395–2398
Table 4. SAR related to the incorporation of rigid cyclic amines
directly connected to the indole
amines, rigid amine-tether combinations exhibited equi-
potent activity to the previously identified leads. In one
specific example, hydroxylation of the tether indicated
that enhanced activity over the parent may be
achieved.
H
N
O
O
References and notes
N
N
R
CH3
1. Thorneloe, K. S.; Nelson, M. T. Can. J. Physiol.
Pharmacol. 2005, 83, 215.
Compound
R
IC50 (lM)
2. Rossier, M. F. Recent Res. Dev. Biochem. 2003, 4, 13.
3. Takano, H.; Zou, Y.; Akazawa, H.; Nagai, T.; Mizukami,
M.; Toko, H.; Komuro, I. Prog. Exp. Cardiol. 2003, 7, 85.
4. Verkhratsky, A. Physiol. Rev. 2005, 85, 201.
5. Putney, J. W. Cell Calcium 2003, 34, 339.
7m13
0.31 (n = 1)
NH2
6. Hook, S. S.; Means, A. R. Annu. Rev. Pharmacol. Toxicol.
2001, 41, 471.
7. Hudmon, A.; Schulman, H. Biochem. J. 2002, 364, 593.
8. Edman, C. F.; Schulman, H. Biochim. Biophys. Acta 1994,
1221, 89.
9. Levy, D. E.; Wang, D.-X.; Lu, Q.; Chen, Z.; Perumattam,
J.; Xu, Y.-J.; Liclican, A.; Higaki, J.; Dong, H.; Laney,
M.; Mavunkel, B.; Dugar, S. Bioorg. Med. Chem. Lett.
2008, 18, 2390.
10. An homology model of autoinhibited CaMKIId was built
based on crystal structure 1A06 of autoinhibited rat
CaMKI. Because rat CaMKI shows high sequence
homology with CaMKIId, this model was used to study
inhibitors that were not ATP competitive. Due to the lack
of availability of a crystal structure of activated CaMKIId,
homology models were built based on crystal structures
1CDK, 1PHK, and 1KOB. From these homology models,
we selected the one that best explained the SAR. The
accuracy of this model was validated using point mutation
studies.
7n
7o
0.30 (n = 1)
0.64 (n = 1)
NH
N
H
7p
1.87 0.18 (n = 2)
N
11. Faul, M. M.; Winneroski, L. L.; Krumrich, C. A. J. Org.
Chem. 1998, 63, 6053.
Table 5. Effect of hydroxyl substitution on tether
12. Assays were performed with inhibitor or suitable control
solvent added 10 ll per well in a 96-well microtiter plate
(Corning, NY). CaMKIId was diluted in enzyme buffer
(50 mM PIPES pH 7, 0.2 mg/ml BSA, 1 mM DTT) and
added 10 ll per well. Reactions were initiated with 30 ll
reaction buffer (62.5 mM PIPES pH 7, 0.25 mg/ml BSA,
33.3 mM MgCl2, 83 lM ATP, 0.4 mM CaCl2, 8.3 lg/ml
calmodulin, 25 lM [His 5] autocamtide-2, 120 nM [g-
33P]ATP) and incubated at rt for 3 min. Reactions were
terminated by transferring 25 ll to a UNIFILTER 96-well
P81microplate (Whatman, UK), pre-wet with 15 ll 1%
phosphoric acid. After 10 min, the plate was washed three
times with 1% phosphoric acid and one time with 95%
ethanol on a BiomekFX (Beckman Coulter, CA) equipped
with a vacuum manifold. Plates were dried for approxi-
mately 60 min, scintillant was added to the wells, and the
plates were read on a TopCount NXT Microplate
Scintillation and Luminescence Counter (Perkin-Elmer,
MA).
H
N
O
O
N
R
Compound
R
IC50 (lM)
7c
1.35 0.64 (n = 2)
NH2
NH2
7q
0.45 (n = 1)
OH
13. The following compounds were obtained from Calbio-
chem–Novabiochem: 7a (cat # 203294), 7g (cat # 203297),
7l (cat # 203292) and 7m (cat # 557508).
14. The following compounds were obtained from Sigma: 7j
(cat # B3681) and 7k (cat # B3556).
observing the influence of the amine tether on potency.
While primary acyclic amines generally demonstrated
better activity than cyclic secondary or cyclic tertiary