Effect of the structure of adenosine mimic of bisubstrate-analog inhibitors on their activity towards basophilic protein kinases

Article abstract of DOI:10.1016/j.bmcl.2009.09.026

Previously reported structural fragments that associate with the ATP-binding pocket of basophilic protein kinases were conjugated with d-arginine-containing peptides. Inhibitory potency of the resulting bisubstrate-analog inhibitors towards PKA and ROCK-II extended to subnanomolar range. The conjugates incorporating 2-pyrimidyl-5-amidothiophene fragment had the highest activity and at 100 nM concentration exhibited over 80% inhibition of most of the tested basophilic kinases of the AGC group.

Full text of DOI:10.1016/j.bmcl.2009.09.026

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6098–6101
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Effect of the structure of adenosine mimic of bisubstrate-analog inhibitors
on their activity towards basophilic protein kinases
*
Erki Enkvist, Marie Kriisa, Mart Roben, Grete Kadak, Gerda Raidaru, Asko Uri
Institute of Chemistry, University of Tartu, 2 Jakobi St., 51014 Tartu, Estonia
a r t i c l e i n f o
a b s t r a c t
Article history:
Previously reported structural fragments that associate with the ATP-binding pocket of basophilic protein
Received 6 August 2009
Revised 4 September 2009
Accepted 5 September 2009
Available online 12 September 2009
kinases were conjugated with D-arginine-containing peptides. Inhibitory potency of the resulting bisub-
strate-analog inhibitors towards PKA and ROCK-II extended to subnanomolar range. The conjugates
incorporating 2-pyrimidyl-5-amidothiophene fragment had the highest activity and at 100 nM concen-
tration exhibited over 80% inhibition of most of the tested basophilic kinases of the AGC group.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Protein kinases
AGC kinases
PKA
PKB/Akt
ROCK
Bisubstrate inhibitors
Oligoarginine
In the last decade protein kinases (PKs) that regulate the activ-
ity of substrate proteins by catalyzing their phosphorylation have
emerged as important drug targets.1 Ten small-molecule inhibitors
of PKs have been approved for clinical use in cancer treatment and
more than 50 compounds are in clinical trials for drug develop-
ment for cancer and other diseases.2
Due to the apparent ease of development of high-affinity low
molecular weight inhibitors targeted to the well-defined hydro-
phobic adenine-binding cleft, most of the disclosed and studied
inhibitors of PKs bind to the ATP-binding site. However, the design
of ATP-competitive inhibitors has several downsides, including
selectivity problems (in addition to 500 coded PKs in human gen-
ome, 1500 other proteins bind adenine nucleotides3) and the
necessity to compete with high concentration of ATP in the cellular
milieu.
The structure of protein substrate-binding domain of PKs is less
conserved and several highly selective peptide inhibitors have
been described.4 However, longer peptidic structures are needed
to achieve nanomolar potency which leads to problems with cellu-
lar transport and stability of the compounds.
The combined approach represented by the development of
bisubstrate-analog (biligand) inhibitors that simultaneously asso-
ciate with both ATP- and protein-binding domains of PKs could
hence give selective and potent inhibitors of these two-substrate
enzymes.6,7
Conjugates of adenosine analogs and arginine-rich peptides
(ARCs)5 designed by our research group inhibit several basophilic
PKs from AGC group with nanomolar to subnanomolar potency.6
The incorporation of arginine-rich transport peptides into the
structure of ARCs provides them with cell-penetrative properties
and allows their use for modulating intracellular protein phosphor-
ylation.8 The very high affinity of ARCs is retained upon their label-
ing with fluorescent dyes or immobilization that supports their
application for biosensors.9,10
Thecrystalstructureofa complexofthecatalytic subunitofcAMP-
dependent protein kinase (PKAc) with an ARC-type inhibitor
Adc-Ahx-(D
-Arg)2-NH2 (ARC-1034; Adc—adenosine-40-dehydroxy-
methyl-40-carboxylic acid, Ahx—6-aminohexanoic acid), the pre-
sumed lead scaffold of previously reported ARC-type inhibitors, was
solved. According to the crystal structure, the nucleoside moiety of
ARC-1034 was bound to the adenosine binding pocket of PKAc as ex-
pected and the C-terminal amide group of 6-aminohexanoic acid lin-
ker interacted strongly with the glycine-rich loop at the enzyme
residues usually interacting with the b-phosphate group of ATP.6
This co-crystal structure served as a basis for the development
of ARC-type inhibitors of the third generation, incorporating 2 flex-
ible linkers and a chiral spacer (Fig. 1). Ahx was used as the first
(flanking Adc) linker whose structure was optimized according to
the results previous structure–activity studies.5a The chiral spacer,
* Corresponding author. Tel.: +372 5175593; fax: +372 7375275.
E-mail address: Asko.Uri@ut.ee (A. Uri).
a
D-amino acid residue, directs the peptide chain towards the
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.09.026

E. Enkvist et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6098–6101
NH2
6099
HN
free carboxylic acid group was coupled to the peptide N-terminus
on solid phase with HBTU/HOBt activation. Differently, N-terminal
acetylation of the peptide was performed with acetic anhydride.
Unstable precursor pre5 was synthesized via treatment of 4-ami-
nobenzamide with phosgene and it was used without separation
in the reaction with the resin-bound peptide to prepare the conju-
gate 5. Synthesis of pre6 followed mainly the previously described
procedure.11 5-Acetylthiophene-2-carboxylic acid was treated
with N,N-dimethylformamide dimethyl acetal followed by the
reaction with guanidine in the presence of a base (K2CO3 or CH3O-
Na) to form the aminopyrimidine ring. Hydrolysis of the ester gave
triethyl ammonium salt of the carboxylic acid (pre6) that was used
in the coupling reaction with the peptide on resin. A similar proce-
dure was used for synthesis of pre8 (starting from 3-acetylbenzoic
acid). For the synthesis of pre9, 6-chloropurine was reacted with
hydrazine, followed by reaction with sodium nitrite in acetic acid
to give 6-azidopurine. Ethyl propiolate, ascorbic acid and CuSO4
(0.1 equiv in water) were directly added to the resulting mixture.
The overall yield of this one-pot procedure (including three sepa-
rate steps) was only 7%. The low yield may be caused by instability
of 6-azidopurine intermediate that exists in different tautomers.12
Hydrolysis of the ester gave the carboxylic acid pre9 that was
thereafter coupled with the peptide. Other precursors were pur-
chased from commercial sources. Protection groups were removed
and the target conjugates were cleaved from the Rink-amide resin
by treatment with mixture of TFA/H2O/triisopropyl silane 90/5/5.
The activity of the synthesized compounds was determined to-
NH2
C
HN
Peptide
Chiral spacer
O
O
O
H
N
H
C
N
N
H
N
H
C
O
NH2
Ahx2
O
Ahx1
NH
C
Adc
O
N
HN
O
NH2
N
HN
NH2
N
N
OH
HO
Figure 1. Adc-Ahx-(
D
-Lys)-Ahx-(D-Arg)2-NH2, compound 1 (ARC-1012), was used
as the template for the design of the new bisubstrate inhibitors.
binding site for arginines of a PK substrate. The second linker (Ahx2
in Fig. 1) is the tether between the chiral spacer and the oligoarg-
inine fragment. The latter moiety associates with substrate pro-
teins/peptide binding site of basophilic PKs.6 Compound 1 (ARC-
1012, compound XV in Ref. 6; Fig. 1) inhibits several basophilic
protein kinases at nanomolar level (IC50-values 50–800 nM at
100 lM ATP concentration).
Here, we report on bisubstrate-analog inhibitors where Adc
moiety of compound 1 was replaced with fragments that according
to the literature and reported co-crystal structures bind to the ki-
nase in a similar way as adenosine. The structures of both the opti-
mized linker and the peptide were primarily left unchanged. From
this series of compounds the conjugate with the highest inhibitory
potency was chosen for the extension of the oligoarginine moiety
wards three PKs (PKAca
13, PKBc6,13 and ROCK-II9, Table 1). For
comparison, the data for the previously characterized compound
16 are also listed in the Table 1. Compound 2 incorporating an acet-
yl group caused almost no inhibition of PKAc- and PKBc-catalyzed
from (D-Arg)2-NH2 to (D-Arg)6-NH2 to increase the potency of the
reactions while weak binding to ROCK-II could be established for
this compound. Benzoyl derivative 3 was clearly a better inhibitor
than 2 and insertion of 4-amido group (compound 4) led to a 10-
fold increase of potency that points to the formation of hydrogen
bond with the hinge region of the kinase. The application of the
urethane tether in 5 instead of amide group in 4 led to the 10-fold
decrease in inhibitory potency. However the compound 4 is still a
30–500-fold weaker inhibitor than its adenosine counterpart 1.
The heterocyclic moieties used in the novel conjugates were
chosen according to the overlay of X-ray structures of the co-crys-
tals of complexes of PKAc with 5-(2-aminopyrimidin-4-yl)thio-
phene-2-carboxylic acid-based inhibitors,11 from one side, and
Adc-containing ARC-type bisubstrate inhibitors, from the other
side.6 As both groups of inhibitors have a carboxyl groups posi-
tioned in the same region of PKAc, the heterocyclic adenosine-
mimicking fragments could be efficiently conjugated with peptides
in similar way as it was performed in case of adenosine mimics in
ARCs, taking advantage of the optimized linker in the latter conju-
gates. Very high potency of the conjugates 6a, 7 and 8 towards
tested AGC kinases points to the adequacy of this hypothesis. Com-
pound 6a was a stronger inhibitor than its adenosine counterpart 1
towards all three tested kinases. Compound 7 revealed some
ROCK-II selectivity that is in accordance with the selectivity of
the adenosine-mimicking fragment.14 Compound 8 with similar
inhibitors to the subnanomolar region.6
Bisubstrate inhibitors and their precursors were prepared
according to the Scheme 1. Protected peptides were synthesized
on the Rink-amide resin according to the conventional Fmoc pep-
tide synthesis methodology. Adenosine mimic that contained a
i, j
R-COOH
1 - 4, 7
NH2
O
b, j
i, j
NH2
O
a
H2N
5
OCN
pre5
NH2
N
O
O
O
N
c, d, e
S
S
6a-c
8
HO
HO
pre6
NH2
O
O
N
N
O
i, j
c, d, e
HO
HO
pre8
structure was a weaker inhibitor of PKBc and ROCK-II leading to
NH
N
NH
N
some PKAc selectivity for this compound. Triazole-purine deriva-
tive 9 was a much weaker inhibitor than expected although being
structurally isosteric to the potent inhibitor 6a. The decreased
activity of compound 9 may be caused by tautomeric equilibria
leading to re-positioning of NH hydrogen bond donor group to
N-7 of the purine that eliminates its potential for a hydrogen bond
with the kinases.
i, j
f, g, h, e
O
N
9
Cl
N
N
N
N
HO
N
N
pre9
Scheme 1. Synthesis of adenosine mimic precursors and their conjugates with
arginine-rich peptides. Reagents and conditions: (a) triphosgene, DCE, 0 °C, 10 min;
(b) H2N-peptide–resin; (c) N,N-dimethylformamide dimethyl acetal, toluene,
reflux; (d) guanidine, K2CO3, DMF, reflux; (e) TEA, H2O; (f) hydrazine; (g) NaNO2,
AcOH; (h) ethyl propiolate, ascorbic acid, CuSO4; (i) H2N-peptide–resin, HBTU/
HOBt; (j) TFA/H2O/triisopropyl silane (90/5/5).
Based on the highest potency of the pre6-derived inhibitor in the
series of compounds containing the short (
D-Arg)2-NH2 peptide
(6a in Table 1) conjugates that comprised peptides containing 6

6100
E. Enkvist et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6098–6101
Table 1
Inhibitory potency and affinity of the conjugates incorporating various adenosine mimics and the (D-Arg)2-NH2 peptide
Compds
R
Inhibition of PKAc,
IC50 (nM)
a
Inhibition of PKB
IC50 (nM)
c
Displacement of ROCK-II
Kd (nM)
a,b
a,b
a
R-Ahx-(
D-Lys)-Ahx-(
D-Arg)2-NH2
NH2
N
N
N
O
1
54 66
774 136
<3
O
N
HO
O
OH
2
3
na 3 mM
>2 mM
123,000 (3.91 0.19)
8000 (5.10 0.15)
O
183,000 (3.74 0.18)
23,000 (4.64 0.14)
240,000 (3.62 0.25)
14 (7.85 0.18)
ꢀ360000
O
NH2
O
4
28000 (4.55 0.18)
177 (6.75 0.18)
6100 (5.21 0.28)
<3
O
NH2
5
ꢀ900000
N
O
H
NH2
N
O
N
6a
7
270 (6.57 0.14)
900 (6.05 0.31)
4600 (5.34 0.17)
S
N
O
390 (6.41 0.12)
61 (7.21 0.16)
3.8 (8.42 0.14)
36 (7.44 0.24)
NH2
N
O
N
8
NH
N
O
N
9
3250 (5.49 0.17)
27,000 (4.57 0.21)
800 (6.10 0.25)
N
N
N
N
a
The values of pIC50 or pKd with 95% confidence intervals (na = not active) are listed in the brackets.
100 lM ATP was used in kinetic assays, Km-values were 20 lM for PKA and 100 lM for PKBc.
b
the fluorescent probe ARC-583 (KD = 3.6 nM for ROCK-II)9 from the
complex with ROCK-II, but as the probe does not afford the determi-
nation of exact Kd-values for compounds with dissociation constants
Kd <3 nM9 the data are not listed in Table 2.Compound 6b was about
10–25-fold more potent than 6a towards PKAc and PKBc, which
correlates well with previous data for their adenosine counterparts.
D
-arginine residues were prepared. Table 2 lists the activity data for
these compounds. These compounds displaced at low concentration
Table 2
Inhibition potency of conjugates of 5-(2-aminopyrimidin-4-yl)thiophene-2-carbox-
ylic acid
Compounds 6b and 6c have similar inhibitory potency towards PKA
but 6b is 3.5-fold more active than 6c towards PKBc. Previously it
Compds
Inhibition
of PKAc,
IC50 (nM)
Inhibition
of PKB
IC50 (nM)
a
c
a
a,b
was shown that the D-alanine spacer between the linkers increased
the selectivity of ARCs towards PKAc.6
6a Pre6-Ahx(
D
-Lys)Ahx(
-Lys)Ahx(
-Ala)Ahx(
D
-Arg)2-NH2
-Arg)6-NH2
-Arg)6-( -Lys)-
82
270
(7.08 0.15)
5.5
(8.26 0.27)
4.9
(6.57 0.14)
12 (7.92 0.30)
Selectivity of 6b towards 50 PKs was tested in a panel of PKs
(Invitrogen, SelectScreen Biochemical Kinase Profiling, Z’-LYTE
Assay; Table 3) at 100 nM concentration of the inhibitor. It was
assumed that the basophilic kinases would constitute the ‘target
group’ of the inhibitors and the ‘negative controls’ would be repre-
sented by the acidophilic kinase CK1, kinases of the CMGC group
(CDK2, GSK isoforms), and the tyrosine kinase Src. The same
6b Pre6-Ahx(
D
D
6c Pre6-Ahx(
NH2
D
D
D
42 (7.38 0.19)
(8.31 0.23)
a
Values in the brackets express pIC50-s or pKd-s with 95% confidence intervals.
1000 lM ATP was used in kinetic assay of PKA to minimize the distortions
caused by the tight binding conditions.
b

E. Enkvist et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6098–6101
6101
Table 3
kinases were previously used for characterization of the adenosine
counterpart (ARC-1028, compound XXII in Ref. 6) of 6b.6
Generally, 6b inhibited strongly the same basophilic kinases as
Inhibition efficiency of 6b (100 nM) towards protein kinases
Protein kinase
Kinase group
% of inhibition
its adenosine counterpart ARC-1028. PKB
a, PKBc, PKG1, PKG2,
AMPK A1/B1/G1
Aurora A
CaMK-Id
CAMK
Other
CAMK
CAMK
CAMK
CMGC
CAMK
CAMK
CK1
CAMK
CMGC
CMGC
CAMK
AGC
AGC
AGC
STE
STE
STE
AGC
CAMK
CAMK
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
CAMK
CAMK
AGC
41 (2)a
46 (4)
25 (8)
10 (1)
22 (6)
6 (5)
82 (0)
33 (5)
ꢁ22 (2)
44 (3)
8 (1)
SGK-s, PKD-s, PIM-2 and PKC were more potently inhibited by
l
6b than ARC-1028, while the adenosine counterpart was more
active with CHK-s and PAK-s.6
CaMK-II
a
CaMK-IIb
CDK2/cyclin A
CHK1
In conclusion, several aromatic structures were used as adeno-
sine mimics to produce bisubstrate inhibitors for basophilic PKs
with up to subnanomolar potency. Novel compounds comprising
CHK2
CK1
DAPK3 (ZIPK)
GSK3
a1
synthetic non-natural fragments and
D-arginine residues are
expected to be resistant to enzymatic degradation and thus have
potential for application in experiments with live cells and tissues.
a
GSK3b
MELK
MSK1
MSK2
p70S6K
PAK3
PAK4
PAK6
PDK1
PIM1
4 (4)
83 (3)
112 (9)
112 (7)
99 (1)
69 (4)
21 (4)
70 (0)
65 (9)
97 (2)
95 (1)
104 (2)
97 (4)
80 (5)
100 (2)
92 (3)
100 (0)
104 (3)
97 (4)
103 (5)
108 (1)
106 (5)
24 (8)
93 (6)
91 (1)
38 (11)
79 (0)
90 (9)
105 (0)
78 (3)
106 (0)
106 (2)
106 (2)
97 (0)
103 (1)
97 (1)
97 (2)
91 (1)
26 (3)
Acknowledgments
The research was supported by grants from the Estonian Sci-
ence Foundation (6710) and the Estonian Ministry of Education
and Sciences (SF0180121s08).
References and notes
PIM2
PKA C
a
1. Cohen, P. Nat. Rev. Drug Disc. 2002, 1, 309.
2. Johnson, L. N. Q. Rev. Biophys. 2009, 42, 1.
3. Fischer, P. M. Curr. Med. Chem. 2004, 11, 1563.
4. Bogoyevitch, M. A.; Barr, R. K.; Ketterman, A. J. Biochim. Biophys. Acta 2005,
1754, 79.
PKBa
(Akt1)
PKBb (Akt2)
PKBc
(Akt3)
PKC
a
PKC bI
PKC bII
PKC d
5. (a) Enkvist, E.; Lavogina, D.; Raidaru, G.; Vaasa, A.; Viil, I.; Lust, M.; Viht, K.; Uri,
A. J. Med. Chem. 2006, 49, 7150; (b) Enkvist, E.; Raidaru, G.; Vaasa, A.; Pehk, T.;
Lavogina, D.; Uri, A. Bioorg. Med. Chem. Lett. 2007, 17, 5336.
6. Lavogina, D.; Lust, M.; Viil, I.; König, N.; Raidaru, G.; Rogozina, J.; Enkvist, E.;
Uri, A.; Bossemeyer, D. J. Med. Chem. 2009, 52, 308.
7. (a) Parang, K.; Cole, P. A. Pharmacol. Ther. 2002, 93, 145; (b) Lee, J. H.; Kumar, S.;
Lawrence, D. S. Chembiochem 2008, 9, 507; (c) Schneider, T. L.; Mathew, R. S.;
Rice, K. P.; Tamaki, K.; Wood, J. L.; Schepartz, A. Org. Lett. 2005, 7, 1695; (d) Ahn,
D.-R.; Han, K.-C.; Kwon, H. S.; Yang, E. G. Bioorg. Med. Chem. Lett. 2007, 17, 147.
8. Räägel, H.; Lust, M.; Uri, A.; Pooga, M. FEBS J. 2008, 275, 3608.
9. Vaasa, A.; Viil, I.; Enkvist, E.; Viht, K.; Raidaru, G.; Lavogina, D.; Uri, A. Anal.
Biochem. 2009, 385, 85.
10. Viht, K.; Schweinsberg, S.; Lust, M.; Vaasa, A.; Raidaru, G.; Lavogina, D.; Uri, A.;
Herberg, F. W. Anal. Biochem. 2007, 362, 268.
11. Lin, X.; Murray, J. M.; Rico, A. C.; Wang, M. X.; Chu, D. T.; Zhou, Y.; Del Rosario,
M.; Kaufman, S.; Ma, S.; Fang, E.; Crawford, K.; Jefferson, A. B. Bioorg. Med.
Chem. Lett. 2006, 16, 4163.
PKC
PKC
PKC
PKC
PKC
e
c
g
i
l
PKC h
PKCf
PKD2
PKD3
PKG1
PKG2
PRKX
ROCK-1
ROCK-II
RSK1
RSK2
RSK3
SGK1
SGK2
SRC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
AGC
12. Temple, C.; Kussner, C. L.; Montgomery, J. A. J. Org. Chem. 1966, 31, 2210.
13. Viht, K.; Vaasa, A.; Raidaru, G.; Enkvist, E.; Uri, A. Anal. Biochem. 2005, 340, 165.
14. Sessions, E. H.; Yin, Y.; Bannister, T. D.; Weiser, A.; Griffin, E.; Pocas, J.;
Cameron, M. D.; Ruiz, C.; Lin, L.; Schürer, S. C.; Schröter, T.; LoGrasso, P.; Feng,
Y. Bioorg. Med. Chem. Lett. 2008, 18, 6390.
AGC
TK
a
Percent of inhibition of the kinases (6b, C = 100 nM; ATP, C ꢂ Km), as relative to
that in control incubations where the inhibitor was omitted (means of duplicate
determinations). Values in parentheses are differences of results of duplicate
measurements.