Structural analysis of ARC-type inhibitor (ARC-1034) binding to protein kinase A catalytic subunit and rational design of bisubstrate analogue inhibitors of basophilic protein kinases

Article abstract of DOI:10.1021/jm800797n

The crystal structure of a complex of the catalytic subunit (type α) of cAMP-dependent protein kinase (PKA Cα) with ARC-type inhibitor (ARC-1034), the presumed lead scaffold of previously reported adenosineoligo-arginine conjugate-based (ARC-type) inhibitors, was solved. Structural elements important for interaction with the kinase were established with specifically modified derivatives of the lead compound. On the basis of this knowledge, a new generation of inhibitors, conjugates of adenosine-4′-dehydroxymethyl-4′-carboxylic acid moiety and oligo(D-arginine), was developed with inhibitory constants well into the subnanomolar range. The structural determinants of selectivity of the new compounds were established in assays with ROCK-II and PKBy.

Full text of DOI:10.1021/jm800797n

308
J. Med. Chem. 2009, 52, 308–321
Structural Analysis of ARC-Type Inhibitor (ARC-1034) Binding to Protein Kinase A
Catalytic Subunit and Rational Design of Bisubstrate Analogue Inhibitors of Basophilic Protein
Kinases
Darja Lavogina,† Marje Lust,† Indrek Viil,† Norbert Ko¨nig,‡ Gerda Raidaru,† Jevgenia Rogozina,† Erki Enkvist,† Asko Uri,*,†
and Dirk Bossemeyer‡
Institute of Chemistry, 2 Jakobi Street, 51014 Tartu, Estonia, Group of Structural Biochemistry, German Cancer Research Centre,
Im Neuenheimer Feld 280, 69120 Heidelberg, Germany
ReceiVed July 2, 2008
The crystal structure of a complex of the catalytic subunit (type R) of cAMP-dependent protein kinase
(PKA CR) with ARC-type inhibitor (ARC-1034), the presumed lead scaffold of previously reported adenosine-
oligo-arginine conjugate-based (ARC-type) inhibitors, was solved. Structural elements important for interaction
with the kinase were established with specifically modified derivatives of the lead compound. On the basis
of this knowledge, a new generation of inhibitors, conjugates of adenosine-4′-dehydroxymethyl-4′-carboxylic
acid moiety and oligo(D-arginine), was developed with inhibitory constants well into the subnanomolar
range. The structural determinants of selectivity of the new compounds were established in assays with
ROCK-II and PKBγ.
Introduction
istics, but their selectivity suffers from the high homology of
ATP-binding sites of protein kinases.10,11 Other target sites have
been explored,12 and recently several highly selective peptide-
based inhibitors have been developed;13,14 for nanomolar
potency, however, longer peptidic structures are required,
making cellular transport and stability of the compounds an
issue.
Protein kinases (PKsa) belong to the transferase class of
enzymes and constitute a large gene family comprising ca. 2%
of the mammalian and 4% of the plant genome.1,2 Their
common function, catalysis of protein phosphorylation, is a key
mechanism of cell regulation, switching cellular systems as
responses to extra- and intracellular stimuli.3 Aberrant activity
of PKs due to mutation, overexpression, or disabled cellular
inhibition can result in serious diseases including diabetes,
cancer, and immune system deficiencies;4,5 some protein kinases
contribute to disease also during their normal cellular function-
ing. Because of the established important role of kinases in
tumor genesis and progression more than 70 small-molecule
kinase inhibitors are at various stages of clinical trials as
anticancer drugs.6
Bisubstrate-analogue, or biligand inhibitors target both the
binding site for the cosubstrate ATP and the binding site for
the substrate protein/peptide. For this they combine two inhibi-
tory domains, each mimicking one of the substrate types, and
simultaneously associate with binding sites of both substrates.
First applied to the adenylate kinase in 1972,15 such a bisub-
strate-analogue inhibitor approach has given several potent
inhibitors of protein kinases.12,16 Variations have addressed both
the structures of the incorporated single-site inhibitors/pseu-
dosubstrates and the linking chain. The principle advantage of
bisubstrate-analogue inhibitors is their potential for a synergistic
effect on affinity compared to the affinities of the single
components17,18 due to entropic and enthalpic factors. The
additional prospect of a possible higher tolerance of such
systems toward mutations of the so-called gate-keeper residue,
a major factor in the emergence of therapy resistance, has been
suggested.19
Consequently, a considerable amount of research is dedicated
to the development, screening, and improvement of protein
kinase inhibitors. The most common type of inhibitors is ATP-
site directed.7-9 They have generally good affinity character-
* To whom correspondence should be addressed. Phone: +3725175593.
Fax: +3727375275. E-mail: asko.uri@ut.ee.
†
Institute of Chemistry.
Group of Structural Biochemistry, German Cancer Research Centre.
‡
a Abbreviations: Abu, 4-aminobutanoic acid moiety; Adc, adenosine-
4′-dehydroxymethyl-4′-carboxylic acid; Ahx, 6-aminohexanoic acid moiety;
AMP-PNP, adenylyl-5′-imidodiphosphate; Aoc, 8-aminooctanoic acid
moiety; ARC, adenosine-arginine conjugate; BisTris, 1,3-Bis[tris(hy-
droxymethyl)methylamino]; Boc, tert-butoxycarbonyl; BOP, benzotriazol-
1-yl-oxytris(dimethylamino)-phosphonium hexafluorophosphate; Bz, benzyl;
Bzo, benzoyl; DCE, 1,2-dichloroethane; DIEA, diisopropylethylamine;
Fmoc, 9-fluorenylmethoxycarbonyl; GST, glutathione S-transferase; H1152P,
(S)-(+)-4-methyl-5-(2-methyl-[1,4]diazepane-1-sulfonyl)-isoquinoline; H89,
N-[2-((p-bromocinnamyl)amino)ethyl]-5-isoquinolinesulfonamide dihydro-
chloride; HOBt, N-hydroxybenzotriazole; MALDI-TOF, matrix-assisted
laser desorption/ionization-time-of-flight; MBHA, 4-methylbenzhydry-
lamine; Mes, 2-(N-morpholine)-ethanesulfonic acid; Pbf, 2,2,4,6,7-pentam-
ethyl-dihydrobenzofurane-5-sulfonyl; MS, mass spectrometry; PDB, Protein
Data Bank; PK, protein kinase; PKA, cAMP-dependent protein kinase; PKA
CR, cAMP-dependent protein kinase catalytic subunit type R; PKBγ, protein
kinase B isoform type γ (Akt3); PKI, heat-stable protein kinase inhibitor;
RIIR, cAMP-dependent protein kinase regulatory subunit isoform IIR;
ROCK-II, GST-tagged Rho-dependent protein kinase; TAMRA, carbox-
ytetramethylrhodamine; TEA, N-triethylamine; TFA, trifluoroacetic acid.
Previously, we reported the development of conjugates of
oligo-(D-arginine) peptides with adenosine-4′-dehydroxymethyl-
4′-carboxylic acid (ARCs), 5-isoquinolinesulfonic acid,7 or
carbocyclic analogue of 3′-deoxyadenosine.20 These compounds
are highly potent inhibitors of protein kinase A (also known as
cAMP-dependent protein kinase, cAPK or PKA) with inhibitory
constants in the low nanomolar region. This high potency of
the conjugates contrasts to merely submillimolar activities of
their individual single site-targeted constituents. This is an
indication of the true bisubstrate character of the inhibitors,
although kinetic analysis reveals competitiveness with only one
substrate, ATP. The bisubstrate functioning was further con-
firmed in surface plasmon resonance binding studies, showing
a competitive displacement of the compounds from their
complex with the kinase by ligands targeting either the ATP-
10.1021/jm800797n CCC: $40.75
2009 American Chemical Society
Published on Web 12/29/2008

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 309
binding pocket (such as ATP, H89, or H1152P) or the
protein-substrate-binding site (RIIR and GST-PKIR).21 Further,
the selectivity profiling with a wide kinase panel (altogether 52
kinases) revealed a strong tendency of the arginine-rich con-
jugates to inhibit basophilic PKs, corresponding to an active
contribution of both functional moieties in the formation of the
binary complex with the kinase.7,20
Protein kinases share the bilobal overall structure of a highly
conserved catalytic core, although details of ligand binding
patterns and recognition sites are different.22,23 PKA that was
used as the target protein kinase in this work is one of the best
characterized protein kinases. Its structural similarities to certain
other protein kinases led to its use as an “ersatz”-kinase, e.g.,
structural models in the form of chimeras can be constructed
carrying specific point mutations to mimic inhibitor binding sites
of related kinases.24,25 Protein kinase A R-catalytic subunit (PKA
CR) has been crystallized as an apoenzyme and cocrystallized
in a variety of complexes with a peptide derived from the
pseudosubstrate inhibitor PKI, with protein kinase C inhibitors,26
Rho-kinase inhibitors,24,27 general PK inhibitors like stauro-
sporine28 and Hidaka’s inhibitors,27,29 and a variety of balanol
derivatives.30,31 Here we present the crystal structure of the
complex of the catalytic subunit of PKA with a representative
of ARC-type inhibitors, the compound 1 (ARC-1034). To our
knowledge, bisubstrate-analogue inhibitors have been so far
cocrystallized with four target protein kinases, representing both
tyrosine kinases (the core tyrosine kinase domain of the insulin
receptor, at 2.7 Å resolution;16 the epidermal growth factor
receptor kinase domain, 2.6 Å resolution;32 and Abelson tyrosine
kinase, 1.8 Å resolution33), and serine/threonine kinases (CMGC
family, cyclin-dependent kinase 2, 1.7 Å resolution34). A key
feature of the compound 1 disclosed by the structure is the linker
region connecting the nucleosidic part and the peptidic part of
the molecule, which connects well to the protein kinase glycine
flap by utilizing a conformational induced fit change, at the same
time orienting the attached oligo-D-arginine peptide for position-
ing of distal inhibitor side-chains. On the basis of the insights
from the crystal structure, a new generation of adenosine
analogue-oligo-(D-arginine) inhibitors was designed, synthe-
sized, and characterized in PKA, PKBγ, and ROCK-II inhibition
and binding assays. For the best of these novel ARC compounds,
inhibition constants in the subnanomolar range were determined.
Figure 1. The structure of the compound 1. The nucleosidic moiety
is in red, the linker in cyan, and the peptidic moiety in blue.
Table 1. General Crystallographic Data of PKA CR-Compound 1
Complex
data collection
X-ray source
Deutsches Elektronen
Synchrotron BW6, Hamburg
P212121
space group
cell dimensions (a, b, c) (Å)
resolution range (Å)
completeness (%) (last shell)
Rsym (last shell)
71.6, 73.2, 78.0
53.38-2.28
99.9 (100)
0.119 (0.387)
48.31
solvent (%)
refinement
number of atoms used in refinement 3098
R factor (%)
free R factor (%)
free R factor value test size (%)
reflections used
0.199
0.249
5.1
17407
standard deviation from ideal values
0.017
bond length (Å)
bond angles (deg)
1.598
temperature factors
mean B value
25.27
Results and Discussion
BJW3, see also Supporting Information), locating one (“inner”)
molecule to the catalytic cleft and the ATP-binding site and
the other (“outer”) one to the surface of the enzyme in the region
wedged between two symmetry related PKA molecules. The
presence of the “outer” compound 1 molecule is most likely
caused by high inhibitor concentration at crystallization condi-
tions, and it is not related to the mechanism of inhibition.
Interestingly, the surface-bound “outer” compound 1 molecule
contacts two phosphoryl groups from enzyme residues, Thr197P
from one enzyme molecule, and Ser139P from the symmetry
mate. The electron density for the inner compound 1 molecule
in the ATP-binding site allowed unambiguous modeling for the
adenosine moiety and the 6-aminohexanoic acid linker group.
The electron density for the peptidic part of the inhibitor, facing
the opening of the catalytic cleft, was discontinuous but allowed
to model the two arginine residues (Figure 3). The “outer”
compound 1 molecule is more exposed to the solvent, lying
near the loop between helices F and G of the C-terminal lobe
with its adenosine moiety at the hydrophobic surface formed
by Tyr229, Tyr235, Phe238, and Phe239, and one of the
arginines interacting with the phosphoryl moiety of the auto-
General Crystallographic Data. The chemical structure of
the compound 1 (Figure 1) used for crystallization as the lead
compound representing ARC-inhibitors was optimized on the
basis of PKA inhibitory potential during previous studies.7
The compound 1 is a conjugate of adenosine-4′-dehydroxy-
methyl-4′-carboxylic acid moiety (Adc) and a D-arginine dipep-
tide linked by 6-aminohexanoic acid moiety (Ahx). The IC50
value of 2.6 µM (at 100 µM ATP) was determined in a PKA
inhibition fluorometric TLC assay.35
The general crystallographic data of the PKA CR-compound
1 complex (3BWJ) are presented in Table 1 and the electron
density maps of the inhibitor binding pockets in Figure 2.
The complex crystallized in space group P212121 to a
resolution of 2.29 Å. This space group is common for various
complexes of PKA, including those with the PKI(5-24)
inhibitor peptide. The unit cell dimensions are minimally smaller
(71.6/73.2/78.0 Å) than for typical PKA/PKI(5-24) complexes
(PDB 1YDS: 73.6/76.3/80.6 Å).
The electron density map indicated the presence of the two
compound 1 molecules bound to the kinase molecule (PDB

310 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
LaVogina et al.
Figure 2. Electron density map within 1.6 Å around the compound 1 molecules contoured at 1 σ. (A) “Inner” compound 1 molecule in the active
site. Enzyme residues of the binding site are shown with light red carbon atoms. H-bonds between the contacting groups of the inhibitor and the
enzyme are depicted. (B) “Outer” compound 1 molecule in surface bound location. Residues of both symmetry mates within 4 Å from the inhibitor
are shown. H-bond contacts are depicted. Symmetry molecules differ by light red or marine blue carbon atom colors. Figures were created with
Pymol.
relatively low B-factors. The glycine flap appears to be fixed
by interacting with the Ahx linker of the bound inner compound
1 molecule. With this, the overall structure is almost identical
to the structure of the PKA complex with the isoquinoline
sulfonamide inhibitor H89 and PKI(5-24) peptide (PDB 1YDT)
with the rmsd value of 0.22 Å for CR atoms. In 1YDT, the
large bromocinnamyl side-chain binds to the glycine flap.29
Analysis of Interactions. The ATP-mimicking part of the
inhibitor interacts with the enzyme as typical for an ATP
molecule: Glu121 is an acceptor for the adenine N6-donated
hydrogen bond, Val123 is a hydrogen bond donor for N1 of
the purine ring, and the Thr183 hydroxyl group forms a H-bond
to the purine N7. Also, the 2′ and 3′ hydroxyl groups of the
ribose make typical contacts to the Glu127 side-chain and to
the Glu170 backbone carbonyl. A new interaction, however, is
formed between the O5′ oxygen of the carbonyl group connected
to the C4′ of the ribose and water 404 in the ATP-binding site.
This water molecule itself is well connected to the protonated
amino group of the side-chain of Lys72 and to the hydroxyl
group of Thr183. This indirectly connects O5′ to enzyme
residues, thus possibly explaining the better inhibitory potency
of conjugates containing an adenosine-4′-dehydroxymethyl-4′-
carboxylic acid moiety, compared to conjugates with corre-
sponding adenine and adenosine moieties, as was previously
observed.7
Figure 3. Interactions of the inner compound 1 molecule with amino
acid residues of PKA. Enzyme carbons are light red, inhibitor carbons
are in green. H-bonds are depicted. Water molecules are shown as red
balls. Water 404 is noted.
A characteristic feature of the bound compound 1 molecule
is the binding mode of the Ahx linker group. The main
determinant is the interaction of the carbonyl O28 with the
backbone amide groups of three residues from the glycine flap,
Ser53, Phe54, and Gly55. Similar H-bond contacts have been
observed in the complex with bound AMP-PNP or ATP between
the ꢀ- and γ-phosphoryl groups and the same amide groups.36,37
The flexible aliphatic chain of the linker packs underneath the
glycine flap and makes van der Waals contacts with Val57. In
phosphorylated Ser139 (Figure 4), the other one with the Thr197
phosphoryl group of a symmetry related kinase molecule.
The enzyme is in an active conformation, which can be
considered intermediate (half-closed) because of the more
open conformation of the glycine flap structure. While the open
conformation of the glycine flap is often found among inhibitor
cocrystal structures of protein kinases, the open position of the
glycine flap in the PKA CR-compound 1 structure is different
from most of these as it is relatively well fixed, as indicated by

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 311
Figure 4. (A) Part of the unit cell showing two of the symmetry kinase molecules with bound compound 1 molecules. The interactions of the outer
compound 1 molecule with the enzyme residues are indicated. (B) Binding site of the outer compound 1 inhibitor formed by two symmetry related
enzyme molecules. H-bond interactions with the phosphoryl groups are shown. Symmetry molecules differ by color of carbon atoms (light red
versus marine blue).
Figure 5. Overlay of structures in the region of the binding pockets of the compound 1 and H89 (1YDT). The compound 1 is shown with green
carbon atoms, H89 with light red carbon atoms.
contrast to the AMP-PNP-complex, here the glycine flap is in
a more open conformation, which may result from a fit of the
flexible 6-carbon chain with the ribose on one side and the O28
position for multiple interactions with the backbone amides on
the other side, accommodating the linker by a rotational
movement of the glycine flap around a hinge defined by the C-R
atoms of residues Gly50 and Val57. In this respect, the
conformation of the glycine flap very much resembles the
structure of 1YDT (H89 bound to PKA), whereas H-bonds with
O28 replace hydrophobic interaction of the bromine atom with
the glycine flap in 1YDT (Figure 5).
In contrast to the very good definition of the electron density
for the nucleosidic part and for the aliphatic nonpolar flexible
linker part of the “inner” compound 1 molecule, the electron
density for the peptidic moiety is not continuous. This most
likely indicates conformational mobility of the arginine residues.
The terminal distal arginine residue is exposed to the solvent
and does not appear to make significant contacts to enzyme
residues, with the exception of an H-bond from its carboxyl
group to the side-chain of Ser53. The other, proximal arginine
residue, shows clear electron density for its guanidinium group,
which binds in a bidentate manner to the carboxyl group of
Asp166, and additionally to the hydroxyl group of Thr201. For
Asp166, which is invariantly conserved among all protein
kinases, a role in catalysis is suggested, either as the catalytic
base for the phosphotransfer reaction, or as having a function
in orienting the substrate hydroxyl group toward the γ-phos-
phorus for nucleophilic attack. Both of these functions contribute
to the substrate binding, corroborated by the fact that mutation
of Asp166 to Glu causes an increase in Km for the peptide, but
not in Km for ATP.38 Thr201, less conserved, has been disputed
about as a sensor for the presence of substrate, together with

312 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
LaVogina et al.
Table 3. Structures and PKA Inhibition Data of Novel ARC-Type
Table 2. Structures and PKA Inhibition Data of Novel ARC-Type
Compounds (First Set)
Compounds (Second Set)
compd
no.
nucleosidic
moiety
compd nucleosidic first
chiral second
peptidic
moietya
linker
peptidic moiety
Kd (µM)b
no.
moiety
linker spacer linker
Kd (nM)b
a
15
16
17
18
19
20
21
22
23
24
25
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Ahx- (D-Lys)- Ahx- (D-Arg)2-NH2 7.6 ( 0.1
Ahx- (D-Lys)- Abu- (D-Arg)2-NH2 12.6 ( 0.1
Ahx- (D-Lys)- Aoc- (D-Arg)2-NH2 9.5 ( 0.1
1
2
3
4
5
6
7
8
Adc-
Adc-
Adc-
Ahx- (D-Arg)2-NH2
0.319 ( 0.057
55.8 ( 0.8
2.18 ( 0.09
98.7 ( 1.5
52.1 ( 1.3
0.249 ( 0.049
0.443 ( 0.011
8.23 ( 0.78
6.05 ( 0.52
1.64 ( 0.31
18.7 ( 0.4
13.5 ( 0.6
over 100
Ahx- OHa
a
Ahx- NH2
a
Ahx-
Ahx-
Ahx-
Ahx- (D-Arg)2-NH2 83.0 ( 1.4
Abu- (D-Arg)2-NH2 247 ( 11
Aoc- (D-Arg)2-NH2 95.4 ( 0.9
Ado(CdO)- Abu- NH2
Ado(CdO)- Ahx- NH2
a
a
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Adc-
Ahx- (D-Arg)-NH2
Ahx- (D-Lys)- Ahx- (D-Arg)4-NH2 ∼0.5
Ahx- (D-Lys)- Ahx- (D-Arg)6-NH2 below 0.5
Ahx- (D-Lys)- Ahx- (L-Arg)2-NH2 8.7 ( 0.5
Ahx- (L-Lys)- Ahx- (D-Arg)2-NH2 594 ( 12
Ahx- (D-Ala)- Ahx- (D-Arg)2-NH2 5.8 ( 0.1
a
Ahx- (D-Lys)-NH2
Ahx- NH(CH2)2NH2
Ahx- NH(CH2)3NH2
Ahx- NH(CH2)4NH2
Ahx- NH(CH2)2OH
Ahx- NHBz
Ahx- [L-Phe(Bzo)]-NH2
Ahx- [D-Phe(Bzo)]-NH2
9
10
11
12
13
14
a -NH2 stands for amide group at the C-terminus of the inhibitors.
b Measured in a fluorescence polarization binding/displacement assay.
a
a
1.96 ( 0.32
a -OH stands for a carboxylic group and -NH2 for an amide group at
the C-terminus of the inhibitors. b Measured in fluorescence polarization
assay.40
amino acid residues. Therefore, we tried the effect of basic
amino acids (compounds 6-7), aliphatic amines and diamines,
and/or hydrogen bond donors (compounds 8-11) on inhibitory
potency and affinity of the conjugates. Second, PKA has a
hydrophobic P + 1 pocket, which in the crystal structure is
positioned near the C-terminus of the inhibitor. Therefore, to
analyze possible binding of inhibitors with C-terminal aromatic
groups to the aromatic rings of Phe54 and Phe187, we inserted
aromatic amines/amino acids (compounds 12-14) as moieties
to the C-terminus of the linker.
Asp166.39 Another contact from the peptidic part of the inhibitor,
although less defined, is from O77 to the ammonium group of
Lys168sthis residue is invariantly conserved in the Ser/Thr
subfamily of protein kinases and is important for catalysis.36
Prior to cocrystallization experiments, there was some
expectation that the two arginine residues of the compound 1
might have a similar positioning and function as the two Arg
residues (P-2 and P-3) in the PKA substrate’s consensus
sequence, Arg-Arg-Xaa-Ser/Thr-Yaa. There is no indication,
however, of a direct interaction of the arginines of the compound
1 with the known substrate recognition residues of PKA, i.e.,
Glu127, Glu170, and Glu230. In this respect, the compound 1
seems only in part to represent the previously reported ARC-
type inhibitors of high potency,7,20 which contain four or six
arginines. A possible scenario is that the first two D-arginine
residues of the conjugates with 4 or 6 arginines provide suitable
stereochemical geometry to serve as a joining chain between
the linker and the C-terminal arginines, which then are
responsible for the interaction with basophilic kinases, including
PKA.
Rational Design of Inhibitors. On the basis of the crystal
structure of the PKA CR-compound 1 complex, we designed
and synthesized two sets of novel ARC-type inhibitors.
The crystal structure indicates a certain conformational
mobility of the arginine moieties of the compound 1, with the
exception of the guanidinium group of the proximal arginine
residue, which is bound to the enzyme. This suggests that a
significant contribution to the binding is provided by the Ahx
linker moiety, at least in the case of compounds with short
peptidic moieties. To test the role of the linker for binding, we
designed a set of derivatives either without any peptide portion
or with different amine or amino acid extensions.
The second set of inhibitors (compounds 15-25, Table 3)
was designed to afford hydrogen bonds and charge-charge
interactions between the arginine residues of the peptidic part
of the inhibitors and the peptide substrate or pseudosubstrate
recognition site of PKA CR that is known to include glutamate
residues Glu127, Glu170, Glu230, and Glu203. The former three
glutamate residues usually interact with the P-2, and P-3 arginine
residues in the consensus sequence of PKA substrates and
pseudosubstrates; Glu203 interacts with the P-6 arginine residue
of the protein kinase inhibitor PKI.41 In the crystal structure of
PKA CR-compound 1, the arginine residues do not interact
with any of these specific glutamate residues. Three of the
glutamate residues are too far away for an interaction, only
Glu170 theoretically could be within reach of one of the two
arginine residues, but apparently such an interaction is not
favored. Thus, we decided to elongate the linker part of the
conjugates by introducing a second linker. The conformation
of the arginine moieties of the compound 1 in the crystal
structure points to the possibility that the peptidic part could
serve as a suitable chiral spacer for the binding of the conjugates
incorporating longer peptides. Hence, in compounds 15-20 we
varied the length of the second linker and checked whether the
chiral spacer between the two linkers could affect the binding.
We used a lysine residue as the chiral spacer, a conservative
replacement for arginine and, from the future prospective, suited
for the attachment of various cargoes to the conjugate via its
amino group. We have also tested the effect of the L-enantiomers
of amino acid residues in the peptidic moiety (compound 23)
or the chiral spacer (compound 24). In compound 25, a D-alanine
residue was used as the chiral spacer to assess the effect of
side-chain functionality of the chiral spacer on the affinity of
the inhibitor.
The compounds of the first set (Table 2) have all an
adenosine-4′-dehydroxymethyl-4′-carboxylic acid or -carbamoyl
moiety attached to the N-terminus of Ahx linker (compounds
2-5), and 1-2 small amine or amino acid moieties connected
to the C-terminus of the linker (compounds 6-14).
Within this set, we also tested the tolerance of PKA for a
negative charge at the C-terminus of the linker (the compound
2) and tested the effect of the structure of adenosine-linker
connection without changing the total length of the conjugate
(compound 4) or without changing the structure of the linker
(compound 5).
The best ARC-type inhibitors reported so far contained six
D-arginine residues,7,20 therefore, in compounds 21 and 22, we
increased the number of arginine residues to four and six,
respectively.
Biological Evaluation of the Inhibitors. Out of the 14
compounds from the first set of inhibitors (Table 2), the best
inhibition constants were measured for compounds 6 and 7
Two additional aspects were explored. PKA CR has a
preference for substrates or pseudosubstrates incorporating basic

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 313
featuring D-arginine and D-lysine at the C-terminus of the
inhibitor (Kd of 249 and 443 nM, respectively). This result
matches with the observed enzyme contacts of the guanidinium
group of the first arginine residue of the compound 1 in the
crystal structuresi.e., to the carboxylate group of Asp166 and
to Thr201. The interpretation was further supported by the
observed tendency of compounds 8-10 to show increasing
affinity when the distance between the end of the linker and
the protonated amine group increases (Kd 8.23 µM for compound
8 versus Kd 1.64 µM for compound 10). Possibly, the hydrogen
bonds and ionic interactions of the ammonium group with the
carboxylate group of Asp166 are favored in case of longer and
more flexible chain. The preference for a positive charge in this
region of the inhibitor is further supported by a 2-fold increase
in Kd value when the hydroxyl group is substituted for the amine
group (compound 11) compared to the corresponding amine-
containing counterpart (compound 8).
Considering the small molecular weight of the compound 3
(MW ) 394.2, see Supporting Information), it possesses a
remarkable Kd value of 2.18 µM (binding energy approximately
0.28 kcal per heavy atom). In contrast, its carboxylate coun-
terpart (compound 2) has a 25-fold higher Kd. There is no
obvious explanation for this effect of the negatively charged
group on binding affinity, as the crystal structure does not
elucidate any obvious advantage of an amide group over a
carboxylate group. Thus, right now it cannot be excluded that
the carboxyl group at the C-terminus of the linker may form
interactions to amino acids of PKA C that change the binding
pattern of the inhibitor while decreasing its net affinity. A
possible “candidate” for such an interaction could be Lys72 of
PKA C; in the crystal structure of PKA CR-compound 1,
however, it lies 6.46 Å away from the nitrogen of the amide
group of the compound 1, a closer contact would most certainly
require differences in the binding pattern of the linker region.
13), however, increases the Kd dramatically, which points to
the steric conflicts caused by change of the stereochemistry.
These results demonstrate the great sensitivity of the ef-
ficiency of the interaction with PKA C to changes in the inhibitor
linker region, both at its N-terminus and C-terminus. The highest
affinity of the conjugate is achieved when the 6-aminohexanoic
acid linker (Ahx) is attached to the carboxyl group of the
nucleosidic fragment, adenosine-4′-dehydroxymethyl-4′-car-
boxylic acid. Basic amino acids (especially arginine) or aromatic
residues big enough to protrude to the hydrophobic pocket are
preferred at the C-terminus of the conjugate. The stereochemical
configuration of the residue inserted directly C-terminal from
the linker seems extremely important; D-amino acids are strongly
preferred over L-isomers here.
The largely different positioning of the two arginine residues
of the compound 1 from these of the P-3 and P-2 arginine
residues of basic protein/peptide substrates and pseudosubstrates
in the complex with PKA CR points to possible benefits of
elongation of the linker region of the conjugate. In agreement
with this, the Kd value of the compound 15 (Kd ) 7.60 nM),
which possesses two linkers and D-lysine as the chiral spacer,
is over 30-fold lower than that of the compound 6. The
comparison of the affinities of the compounds 15-17 indicates
that the 6-aminohexanoic acid residue has an optimal length
for the “second” linker. The Ahx linker thus may facilitate an
interaction of the arginine residues from the peptidic part of
inhibitor with the acidic peptide recognition residues of the
enzyme (i.e., Glu127, Glu170, Glu203, or Glu230).
The significantly lower inhibitory potency of the compounds
18-20 compared to their counterparts with an additional
D-lysine residue as the spacer between the two linker moieties
(compounds 15-17) reveals the importance of such a chiral
spacer. The requirement for the defined stereochemistry is
further demonstrated by the much lower binding affinity of the
compound 24, which contains L-lysine (Kd ) 594 nM) instead
of D-lysine of the compound 15 (Kd ) 7.60 nM). In the peptidic
part of the inhibitor, however, L-amino acid residues are tolerated
quite well, as indicated by the binding characteristics of the
compound 23. D-Amino acid-containing compounds, however,
are generally preferred in inhibitor design because their incor-
poration makes compounds more stable against proteolytic
degradation.7 Compound 25 with D-alanine as the chiral spacer
showed slightly better binding characteristics than the corre-
sponding D-lysine containing compound 15. This indicates that
in case of longer conjugates, the stereochemical nature of the
spacer that disposes the second linker and the C-terminal
D-arginine residues is more important than the direct interaction
of the spacer with the enzyme, supposedly with residues Asp166
and Thr201.
The positioning of the carbonyl group between ribose and
the linker (compound 4 versus compound 3) is also very
important. In compound 4, the carbonyl is located two bonds
further away from the ribose. Obviously, this change causes
the loss of the water-mediated interactions to Lys72 and Thr183,
which results in weakening of the interaction. Further, the plane
formed by C4′, C5′O5′, and the amide group of the linker is
shifted in position by two carbon atoms. The curvature of the
flexible linker underneath the glycine flap, as observed in the
PKA-compound 1 structure, which occurred probably also in
case of the compound 3, is interrupted in the compound 4 by
the insertion of the planar carbamoyl group; it is likely that
this will result in a change of the interactions of the linker with
the glycine flap, possibly leading to the lower binding affinity
found for the compound 4. The compound 5, being analogical
to compound 4 in the nucleosidic part but incorporating
aminohexanoic acid moiety as a linker and hence endowed with
a possibility to develop interactions with Gly-rich flap probably
analogical to the compound 1, demonstrated 2-fold lower Kd
value than the compound 4. Thus, Ahx moiety served as the
optimal linker in all compounds of the given set.
Finally, the conjugates 21 and 22, incorporating adenosine-
4′-dehydroxymethyl-4′-carboxylic acid as nucleoside moiety,
two linkers, D-lysine as the chiral spacer and four and six
D-arginine residues in the peptidic moiety, respectively, had
inhibition constants well into the subnanomolar range. Hence,
the inhibitory potency of these ARC-type inhibitors reaches that
of the best inhibitors of PKA reported so far.42 Overall, the
inhibitors developed in the present work demonstrate the
advantages of the rational design based on structures of
cocrystals. For example, the most potent of novel ARC-
inhibitors (the compounds 21 and 22) possess approximately
1000-fold higher inhibitory potency toward their biological
target PKA C, compared to the compounds designed with the
aid of the recently described method of stepwise combinatorial
In the case of the compound 12, the incorporation of an
aromatic moiety to the C-terminus of the linker resulted in a
6-fold increase of Kd compared to the compound 3. The aromatic
group was introduced to utilize possible hydrophobic interactions
with aromatic amino acid residues of PKA, namely Phe54 and
Phe187. Apparently, this requires a larger distance from the
linker as shown with the compound 14, where the incorporation
of a more bulky aromatic system (para-benzoyl-D-phenylalanine)
is tolerated very well. The use of the L-counterpart (compound

314 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Table 4. Activity Data of ARC-Type Compoundsa
LaVogina et al.
compd no.
H89
Kd, PKA (nM)c
Kd, ROCK-II (nM)c
IC50, PKA (nM)
IC50, PKBγ (nM)
662 ( 6d
23 ( 2
319 ( 57
31 ( 3
100 ( 20d 850 ( 90e
1 (ARC-1034)
2,600 ( 300d
ARC-902b
8.3 ( 1.5d 109 ( 21e
5.3 ( 0.7d 67 ( 19e
2.41 ( 0.10d 12.9 ( 2.0e
26000 ( 1000d
2100 ( 500d
36.9 ( 2.3d
30.7 ( 8.2d
43.1 ( 4.4d
over 1000000d
1610 ( 110d
774 ( 13d
ARC-903b
ARC-659-1bb
3
6
15
21
22
25
26
2180 ( 90
249 ( 49
7.6 ( 0.1
∼0.5
4480 ( 260
89.7 ( 2.4
1.4 ( 0.1
below 1
below 1
119 ( 3
187 ( 3
54 ( 6d
13.4 ( 2.0e
56.6 ( 1.0d
14.6 ( 0.4d
3,750 ( 100d
2,038 ( 480d
below 0.5
5.8 ( 0.1
1.8 ( 0.5
5.32 ( 0.03e
97 ( 9d
21.1 ( 2d
a The binding constants were determined by fluorescence-polarization binding/displacement assay with PKA and ROCK-II and inhibition characteristics
by fluorometric TLC assay with PKA and PKBγ. b Previously reported compounds,7,20 see structures in Supporting Information. c Measured in fluorescence
polarization binding/displacement assay. d Measured in fluorometric TLC-supported inhibition assay at 100 µM ATP and 30 µM TAMRA-Kemptide or 10
µM TAMRA labeled AKT/PKB/Rac-protein kinase substrate concentration. e Measured in fluorometric TLC-supported inhibition assay at 1 mM ATP and
30 µM TAMRA-Kemptide or 10 µM TAMRA-PKB substrate concentration.
Figure 6. Sequence alignment of the kinase regions of PKA, ROCK-I, and ROCK-II. The loops of PKA and the corresponding regions of the
Rho-associated kinase isoforms are colored as follows: gray, glycine-rich loop; green, linker responsible for binding of nucleoside moiety of ATP;
yellow, catalytic loop; red, Mg-positioning loop, cyan, activation loop; magenta, hydrophobic P + 1 pocket.
evolution for the development of Akt/PKB bisubstrate inhibitors
ROCK-II),24 the Kd values for binding of the compound 3 were
quite similar for both enzymes.
(Ki ) 260 nM).43
Crystal structures are available for ROCK-I with various small
molecule inhibitors. In the kinetic analysis, ROCK-II instead
of ROCK-I was used; however, 92% of the residues constituting
the kinase domains are conserved between the two isoforms46
(Figure6).OverlayofthecrystalstructuresofPKACR-compound
1 and ROCK-I-Fasudil (PDB 2ESM) does not suggest more
major interactions between specific amino acid residues of
ROCK and the non-nucleosidic part of the compound 1 molecule
than for Ser53 (PKA), which is alanine in both ROCK-isoforms.
Formation of a weak hydrogen bond between Ser53 and the
carbonyl group between two arginine moieties in the compound
3 would not be possible in ROCK (Figure 7). The absence of
significant differences is in agreement with the similar Kd values
of the compound 3 with both kinases.
Comparison of Binding and Inhibitory Properties of
ARC-Inhibitors in Assays with PKA, PKBγ, and ROCK-
II. The inhibitory and binding properties of several novel ARC-
type compounds were analyzed, in addition to PKA, also with
two other AGC-kinases, PKBγ, and ROCK-II. For comparison,
three previously reported potent ARC-type inhibitors ARC-902,
ARC-903, and ARC-659-1b were included in the analysis and
their inhibitory potency toward PKA and PKBγ was determined
(Table 4). For PKA, H89 was used as a reference inhibitor with
dissociation constant value determined by fluorescence-polariza-
tion binding/displacement assay (Kd ) 23 nM), fitting well into
the range of inhibition constants reported for this inhibitor (Ki
) 6.1 nM,44 Ki ) 48 nM45).
Despite the kinase-specific sequence differences in the ATP-
binding pockets of ROCK-II and PKA (Leu49, Val123, Glu127,
Thr183 in PKA versus Ile98, Met172, Asp176, Ala231 in
The compounds 6 and 15 demonstrate 3-5-fold better binding
characteristics for ROCK-II than for PKA. The common feature

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 315
Figure 7. Overlay of crystal structures of PKA CR-compound 1 (green, the compound 1 is shown with orange carbon atoms) and ROCK-
I-Fasudil (PDB 2ESM, blue). Alignment of residues Thr48-Leu59 of PKA and Val81-Leu92 of ROCK-I; the hydrogen bond between the carbonyl
of the first arginine of the compound 1 and the side-chain of Ser53 is shown. The residues are labeled according to PKA numbering.
of these compounds is the presence of a D-amino acid residue
(arginine or lysine) following the linker moiety. This makes it
likely that this residue is involved in ROCK-specific interactions.
The comparison of the residues lying near Asp166 of PKA with
the corresponding region of ROCK (Figure 6, see catalytic loop)
shows a somewhat larger polar pocket in ROCK compared to
PKA, which may be suited to accommodate basic moieties of
the inhibitors. Also, in PKA, this space is partly filled with the
hydrophobic side-chain of Phe187 from the Mg-positioning loop,
while ROCK-II has a threonine residue in the corresponding
position (Thr235).
In this connection, the binding parameters exhibited by the
compound 25 are of special interest. This conjugate, possessing
D-Ala residue as the chiral spacer, has over 20-fold higher Kd
value for ROCK-II than for PKA. This further supports the
conclusion that the lysine spacer inserted between the linkers
in the compound 15 gives a specific binding interaction with
ROCK-II.
Compounds 21 and 22 exhibit good binding to both PKA
and ROCK, and due to the high affinity of the conjugates it is
not possible to determine the exact Kd values. The arginine
residues lying close to the C-terminus of the inhibitor probably
form more potent interactions with PKA, as both kinases
recognize arginines/lysines at substrate positions P-2 and P-3,
but the residues at positions P-5 and P-6 are more important
for binding to PKA C than to ROCK.47,48
Thus, it can be concluded that the structure of the linker
moiety and the origin of the basic amino acid residue positioned
directly after this linker are the most critical features for binding
of the inhibitor to ROCK. The introduction of a small, preferably
hydrophobic, residue as the first residue after the linker makes
the compounds selective for PKA.
) 2.6 µM) within the limits of experimental error. This result
is in agreement with the electron density map of the PKA
CR-compound 1 complex that showed missing density for the
amide terminus of the molecule and only partial density for the
distal arginine of the compound 1, which most likely indicates
that the latter is not bound strongly to PKA C molecule and
does not develop significant hydrogen bonds or van der Waals
interactions with the glycine flap of PKA C. In the case of the
most potent compounds (21 and 22), 1 mM concentration of
ATP was used in the assay with PKA in order to avoid tight-
binding conditions (IC50 e Ckinase) and to scale the IC50 values
to the reasonable range, allowing assessment of the inhibitory
properties more precisely. The best inhibitory characteristics
belong to the compound 22 that inhibited PKA with an IC50
value of 5.3 nM at 1 mM concentration of ATP, resulting in
2-fold better inhibition for PKA C than that previously reported
for the most potent ARCs.7,20
The IC50 value determined by fluorometric TLC method35
for inhibition of PKBγ by Hidaka’s inhibitor H89 was 662 nM.
This is comparable to the IC50 value reported in the literature
for another PKB/Akt isoform PKBꢀ (IC50 ) 590 nM for
inhibition of PKBꢀ by H89).49 The sequence identity is over
80% between PKBꢀ and PKBγ50 (Figure 8). Our previously
reported ARC-compounds exhibited 5-fold lower IC50 values
for PKA compared to PKBγ, with the exception of the deoxy-
compound ARC-659-1b, which showed more than 10-fold
selectivity for PKA over PKBγ.
Surprisingly, compound 3 exhibited extremely poor inhibition
of PKBγ, with millimolar IC50 value. On the other hand, the
addition of one arginine moiety resulted in steep increase of
inhibitory potency (the IC50 value for PKBγ inhibition by the
compound 6 was 1.61 µM). Such a dramatic change points to
the different modes of binding of these two inhibitors to the
kinase.
Inhibition constants were determined in fluorometric TLC
assays35 with PKA and PKBγ. For PKA, the IC50 value for
inhibiting the phosphorylation reaction by the compound 6 is
2.1 µM, which is almost equal to that of the compound 1 (IC50
The comparison of IC50 values for compounds 15 and 25
confirms that PKBγ has preference for a basic residue right after

316 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
LaVogina et al.
Figure 8. Sequence alignment of the kinase regions of PKA, PKBꢀ, and PKBγ. The loops of PKA and the corresponding regions of the PKB
isoforms are colored as follows: gray, glycine-rich loop; green, linker responsible for binding of nucleoside moiety of ATP; yellow, catalytic loop;
red, Mg-positioning loop, cyan, activation loop; magenta, hydrophobic P + 1 pocket.
Figure 9. Overlay of crystal structures of PKA CR-compound 1 (green) and PKBꢀ complex with H89 and GSK3ꢀ (PDB 2JDO, wheat) aligned
at PKA residues 48-59 and PKBꢀ residues 157-168. The compound 1 is shown with orange carbon atoms. The residues are labeled according
to PKA numbering.
the first linker. The latter finding could result from the combined
effect of several factors, e.g., when Thr165 of PKBγ is
substituted for Ser53 of PKA, the first linker of the inhibitor
adopts a shifted position favorable for interaction between the
basic amino acid residue and Asp278 of PKBγ (Figure 9; the
cocrystal of PKBꢀ was chosen for the overlay, as PKBγ has
not been cocrystallized so far). However, the effect is not as
dramatic as in case of ROCK-II.
With the exception of the compound 6, the novel ARC-
inhibitors are less potent toward PKBγ than for PKA, with the

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 317
Table 5. Percent of Inhibition of the Protein Kinases in the Presence of Compound 22 (100 nM)
a Percent of inhibition of the kinases (at 100 nM concentration of the inhibitor and ATP concentration close to the KM value of the given kinase toward
ATP), as relative to that in control incubations where the inhibitor was omitted (means of duplicate determinations).
greatest difference in case of the compounds 21 and 22. The
inhibitory properties measured in PKBγ inhibition assays were
remarkably weaker than in PKA inhibition assays, especially
taking into consideration that 10 times lower ATP concentration
was used in the assay with PKBγ (100 µM in assay with PKB
versus1mMinassaywithPKA).AccordingtotheCheng-Prusoff
equation,51 a 10-fold increase of ATP concentration results in
10-fold decrease of inhibition constant Ki.
Furthermore, the inhibition constant is also linearly dependent
on the KM for ATP, and the KM value of PKBγ for ATP (KM
)
88 µM)52 is ca. 5-fold higher than that for PKA C (KM ) 18.7
µM).35 Thus, the 3-fold difference of IC50 values presented in
Table 4 for the inhibition of PKA and PKBγ by most potent
novel ARCs (the compounds 21 and 22) means approximately
150-fold difference in corresponding inhibition constants Ki in
favor of PKA.
Finally, to confirm our assumptions about the structural
determinants of kinase selectivity of ARC-type inhibitors, we
designed the compound 26 (ARC-1044, see Figure 10). The
nucleosidic part of the compound 26 is a cyclopentane-based
carbocyclic analogue of 3′-deoxyadenosine that, if conjugated
with hexa-(D-arginine) (corresponding compound ARC-659-1b),
was previously shown to enhance selectivity toward PKA in a
panel of 34 protein kinases.20 The second structural element
responsible for selectivity of the compound 26 was D-alanine
residue used as the chiral spacer between the linkers.
Figure 10. The structure of the compound 26.
The inhibitory properties of the compound 26 confirmed our
predictions, as it exhibited over 100-fold selectivity toward PKA
over ROCK-II and PKBγ (Table 4).
Selectivity Studies. Selectivity testing was performed toward
a panel of 50 kinases (on commercial basis, Invitrogen Se-
lectScreen Biochemical Kinase Profiling, Z’-LYTE Assay) of
the most potent of the novel ARC-type compounds, the

318 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
LaVogina et al.
Figure 11. (A) Displacement of compound 27 from its complex with PKA by H89; (B) displacement of compound 27 from its complex with PKA
by RIIR. The analogical displacement of compound 27 from its complex with PKA by Ala-Kemptide is presented in the Supporting Information.
compound 22 (Table 5). The kinases were chosen according to
our previous knowledge, expecting that the basophilic kinases
will constitute the “target group” and the “negative controls”
represented by acidophilic CK1, representatives of CMGC group
(CDK2, GSK isoforms), and tyrosine kinase Src.
compound 27 were as follows: D-arginine was used as the chiral
spacer between the linkers and one D-lysine residue was attached
to the C-terminus of the peptidic part of the inhibitor. The
fluorescent label 5-TAMRA was fused with the side-chain of
D-lysine.
The dissociation constant of 0.3 nM was determined for
compound 27 from titration with PKA (see Supporting Informa-
tion). The high affinity of the fluorescent probe enabled
performance of the displacement of compound 27 from the
complex with PKA by either H89 (Figure 11A) or PKA RIIR
(Figure 11B).
The fluorescent probe could be successfully displaced from
the complex with PKA by both H89 (targeted to ATP-binding
pocket of ATP) and RIIR (regulatory subunit of PKA, known
to compete with protein/peptide substrates), thus proving the
bisubstrate character of the fluorescent probe and novel ARC-
type inhibitors. The displacement IC50 value obtained for RII
subunit was 1.69 nM (the corresponding Kd value below 0.3
nM), and the displacement IC50 for H89 was 109.6 nM (the
corresponding Kd ) 11.4 nM).
As expected, the compound 22 inhibited most potently the
AGC family of basophilic protein kinases and did not inhibit
at all the acidophilic casein kinase (CK1). Overall, the most
inhibited protein kinases (over 90% inhibition) were PKA C,
PKC (except isoforms ι, ꢁ, and µ), ROCK isoforms, and
ribosomal S6-kinases (RSK, MSK, p70S6K), all from the AGC
kinase family. PKBR, PKBγ, SGK (all from AGC family), and
PAK isoforms (STE family) were also well inhibited (over
80%).
The inhibition preferences of the compound 22 were generally
similar to that of previously reported compound ARC-9027
(possessing six arginine moieties and one linker, kinase screen-
ing performed at 1 µM concentration of the inhibitor), although
the compound 22 inhibited PKA and PKC isoforms better and
the representatives of CAMK family (e.g., CaMK isoforms and
PKD) worse than ARC-902.
Conclusions
The inhibition potential of the compound 22 determined for
PKA C, PKBγ, and ROCK-II was in agreement with the results
of our measurements (Table 4).
The crystal structure of the complex PKA CR-compound 1
demonstrated the principal binding pattern of ARC-type protein
kinase inhibitors and paved the way for the rational design of
highly potent ARC-type bisubstrate analogues. Structural ele-
ments responsible for affinity and selectivity were established
using newly synthesized derivatives with corresponding modi-
fications in inhibition and binding assays. A prominent and
important feature of the compound 1 molecule is the Ahx linker,
joining the nucleosidic and peptidic part of the inhibitor. Its
length, shape, and electronic properties together with backbone
flexibility are obviously well suited to support multiple favorable
interactions with the glycine flap of the kinase, thus providing
an explanation for the good inhibitory potency of ARC-type
compounds. The elongation of the linker facilitated the interac-
tion of the peptidic part of bisubstrate-analogue inhibitors with
those amino acid residues of PKA C that are responsible for
substrate consensus sequence recognition. This was achieved
by adding the second Ahx-linker. The highest affinity toward
PKA C was obtained using the orienting effect of a D-amino
acid as the chiral spacer between the two Ahx-moieties. Out of
24 inhibitors synthesized proceeding from the structure of the
compound 1 cocrystal, the highest inhibitory potency was
obtained for the compounds 21 and 22, consisting of adenosine-
Confirmation of the Bisubstrate Character of Novel
Inhibitors. Previously, we have demonstrated that the inhibition
of PKA by ARC-type inhibitors is competitive with ATP and
noncompetitive with the peptide substrate TAMRA-Kemptide.7
This result probably indicates an ordered mechanism of inhibi-
tion where the interaction of the nucleosidic moiety of the
bisubstrate-analogue inhibitor with the ATP-binding site is first
required to allow the interaction of the peptidic fragment of
inhibitor with the protein/peptide-binding site of PKA.14,18
However, the complete displacement of the ARC-type inhibitor
from its complex with PKA C either by ATP or by protein/
peptide substrate confirms the bisubstrate character of the
inhibitor.21
Therefore, we performed the displacement of a novel ARC-
type inhibitor from its complex with PKA CR by either PKA
RIIR subunit binding to the peptide substrate site, or ATP-
competitive inhibitor H89. A fluorescent probe, compound 27
(ARC-1042, see structure in the Supporting Information), was
designed originating from the compound 22 to perform a
fluorescence-polarization binding/displacement assay.40 Com-
pared to the compound 22, the changes in the structure of

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 319
by use of a baculovirus system53 by means of a glutathione-
Sepharose column.
4′-dehydroxymethyl-4′-carboxylic acid as nucleoside group, two
linkers (both 6-aminohexanoic acid moieties), D-lysine as a
chiral spacer, and four or six D-arginine residues in the peptidic
moiety, respectively. With inhibition constants in the subnano-
molar range, they appear to be the highest affinity synthetic
small-molecular PKA inhibitors reported so far. Comparison
of the inhibitory activity of the novel ARC-inhibitors toward
three closely related protein kinases of the AGC group PKA,
PKBγ, and ROCK-II from the AGC-group revealed structural
elements responsible for the selectivity.
Crystallization. The compound 1 was cocrystallized with bovine
CR catalytic subunit of the cAMP-dependent protein kinase using
the hanging drop vapor diffusion method. Droplets contained 18
mg/mL protein, 25 mM Mes-BisTris, 75 mM LiCl, 1 mM
dithiothreitol, pH 6.1-6.8, and 1.5 mM inhibitor and were
equilibrated at 4 °C against 13-20% (v/v) methanol.
Data Collection and Structure Determination. Diffraction data
were measured from frozen crystals at the beamline BW6, Deut-
sches Elektronen Synchrotron (DESY), Hamburg. The data were
processed with the programs MOSFLM and SCALA. The structures
were determined by molecular replacement using AMoRe from the
CCP4 program suite (www.ccp4.ac.uk/main/html). As the search
model, we chose a binary PKA-PKI(5-24) complex, omitting the
peptide in the search coordinates (unpublished). The structure was
refined by one round of rigid body refinement and repetitive rounds
of restrained refinement using Refmac5. Model building was
performed using the ligand builder tool of the CCP4 suite and Moloc
(www.moloc.ch). Phosphorylation sites were found at Ser-139, Thr-
197, and Ser-338. Water molecules were automatically inserted
using the CCP4 programs PEAKMAX and WATPEAK and visually
inspected. Finally, the inhibitor molecules were built and modeled
into the inner and the outer difference electron densities, and the
whole complex was further refined.
Materials and Methods
Materials. All chemicals were obtained commercially unless
otherwise noted. PKA CR and PKA RIIR for fluorometric TLC
assay and fluorescence polarization-based assay was obtained from
Biaffin. PKB (Akt3/PKBγ, S472D, active) was obtained from
Invitrogen, and 5-TAMRA labeled AKT/PKB/Rac-protein kinase
substrate (ARKRERTYSFGHHA) was obtained from Anaspec.
Solvents were from Rathburn and Fluka. Fmoc Rink Amide MBHA
resin and other peptide synthesis chemicals were from Neosystem,
Novabiochem, Advanced ChemTech, and AnaSpec. The synthetic
peptide substrate 5-TAMRA labeled Kemptide (LRRASLG) was
synthesized as previously described.35 The other chemicals were
from Sigma-Aldrich. 1H and 13C NMR spectra were taken on Bruker
AC 200P spectrometer (see Supporting Information). Mass spectra
of all synthesized compounds were measured with MALDI-TOF
mass spectrometer Voyager DE-Pro (Applied Biosystems) and mass
spectra of conjugates 1-26 was recorded on Thermo Electron LTQ
Orbitrap mass spectrometer (see Supporting Information). Unicam
UV 300 (ThermoSpectronic) spectrometer was used for measuring
UV-vis spectra and quantification of the products. Fluorescence
anisotropy readings were taken on a PHERAstar microplate reader
(BMG Labtech) with optic module FP 540-20 590-20. Corning
black low volume 384-well NBS plates (code 3676) were used for
fluorescence anisotropy readings. The solution-phase reactions were
monitored by thin-layer chromatography (TLC) on Polygram Sil
G/UV254 plates (Macherey-Nagel), and a UV-lamp was used for
the visualization of the products. Fluorescence imaging of TLC
plates was performed by Molecular Imager FX Pro Plus (Bio-Rad
Laboratories; excitation at 532 nm, 555 nm LP emission filter, 100
µm resolution), and scanned images were processed with Quantity
One software (Bio-Rad). Column chromatography was performed
on silica gel 60 (0.04-0.063 mm), purchased from Fluka. The final
products were purified with Gilson HPLC system using C18 reverse-
phase column (GL Sciences) Inertsil ODS-3 (5 µm, 25 cm × 0.46
cm), with monitoring at 260 or 220 nm (peptides). Elution was
performed with water-acetonitrile gradient (0.1% TFA) with a flow
rate of 1 mL/min. The separated products were freeze-dried. The
products 1-21 and 23-25 were characterized by ion exchange
chromatography on Luna SCX 5 µ 150 mm × 4.6 mm column
(Phenomenex) by using the eluent systems of A (20 mM phosphate
buffer, pH 7.6, 18% isopropanol) and B (40 mM phosphate buffer,
pH 7.6, 1.2 M NaCl, 18% isopropanol) with a flow rate of 1 mL/
min and detection at 260 nm. Inhibitors 22 and 26 were character-
ized by ion exchange chromatography on Mono S HR 5/5 column
(Pharmacia Biotech) by using the eluent systems of A (20 mM
phosphate buffer, pH 7.6, 18% isopropanol) and B (40 mM
phosphate buffer, pH 7.6, 1.2 M NaCl, 18% isopropanol) with a
flow rate of 1 mL/min and detection at 260 nm. All compounds
used in biological tests were >95% pure by HPLC (detected at
260 nm). Concentrations of compounds for biological testing were
measured by UV spectroscopy (based on molar extinction coef-
ficient of 15000 M-1 cm-1 at 260 nm for adenosine derivatives).
Protein Expression and Purification. Bovine wild-type CR
catalytic subunit of the cAMP-dependent protein kinase was solubly
expressed in Escherichia coli BL21(DE3) cells and then purified
via affinity chromatography and ion exchange chromatography as
previously described.29 Constitutively active bovine GST-Rho-
kinase (ROCK-II, amino acids 6-553) was purified from Sf9 cells
Structural Analysis. For a visual inspection of the crystal
structure and comparison of conformational changes, structures were
aligned using the programs Moloc and Pymol (www.pymol.org).
Figures were prepared with Moloc or Pymol.
Chemical Synthesis. Inhibitors 1, 3-6, and 13-26 were
synthesized by Fmoc solid phase peptide synthesis strategy and
purified by reverse phase HPLC as previously described.7,35 The
synthesis of peptide fragments was carried out on Fmoc Rink Amide
MBHA resin. Protected amino acids (3 equiv, Pbf and Boc as side-
chain protective groups for arginine and lysine, respectively) were
dissolved in DMF/NMM and activated with HOBT/BOP mixture
(2.9 equiv). The coupling solutions were added to the resin and
agitated for 90 min. Completeness of each reaction was monitored
by Kaiser test, followed by deprotection of Fmoc group with 20%
piperidine in DMF (20 min). Fmoc-protected linkers were attached
to the peptide following the same protocol. The nucleoside part
was introduced by application of either 2′,3′-O-isopropylidenead-
enosine-4′-dehydroxymethyl-4′-carboxylic acid (1.5 equiv) solution
in DMF/DIEA after activation with HOBT/BOP (1.47 equiv) or in
case of the compounds 4 and 5 by 4′-nitrophenyloxycarbonylimino-
4′-dehydroxymethyl-2′,3′-O-isopropylideneadenosine (1.2-1.3 equiv)
solution in DMF/DIEA. In the former case, three-hour reaction time
was used, whereas in the latter case, the reaction mixture was
agitated overnight. The washing procedure with three solvents
(DMF, isopropanol, DCE) followed and after freeze-drying the
derivated resin was treated with cleavage cocktail (TFA/TIPS/H2O
mixture) for 2.5 h. The collected product was evaporated to dryness,
washed with methyl-tert-butyl ether, and subjected to further
purification by HPLC.
The precursor reagents of inhibitors 2 and 8-12 were synthesized
in solution and the corresponding full schemes of synthesis are
presented in the Supporting Information.
Fluorometric TLC Kinase Assay. The IC50 values of the
inhibitors corresponding to the concentration of the inhibitor
decreasing the enzyme activity 2-fold were measured as previously
described.35 The inhibitors in various concentrations were incubated
at 30 °C in Hepes buffer (50 mM, pH 7.5) containing PKA CR (1
nM) or PKBγ (0.5 nM), TAMRA-Kemptide (30 µM) or TAMRA-
labeled AKT/PKB/Rac-protein kinase substrate (10 µM), ATP (100
µM or 1 mM in case of inhibitors with very high affinity),
magnesium acetate (10 mM), and bovine serum albumin (0.2 mg/
mL). The phosphorylation reaction was initiated by the addition of
ATP. At fixed time points, the reaction was stopped by a 20-fold
dilution with 75 mM phosphoric acid and samples were analyzed
by normal phase TLC (without fluorescence indicator, eluted with
1-butanol/pyridine/acetic acid/water, 15/10/12/12 by volume). The

320 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
LaVogina et al.
(15) Lienhard, G. E.; Secemski, I. I. P1, P5-Di(adenosine-5′)pentaphosphate,
a potent multisubstrate inhibitor of adenylate kinase. J. Biol. Chem.
1973, 248, 1121–1123.
(16) Parang, K.; Till, J. H.; Ablooglu, A. J.; Kohanski, R. A.; Hubbard,
S. R.; Cole, P. A. Mechanism-based design of a protein kinase
inhibitor. Nat. Struct. Biol. 2001, 8, 37–41.
visualization and quantification of the fluorescent spots were carried
out by fluorescence imaging. The data were processed with
Graphpad Prism software version 4.03.
Fluorescence Polarization-Based Binding/displacement Kinase
Assay. All concentration-dependent binding experiments were
performed as previously described40 using 3-fold dilution series in
the assay buffer (150 mM NaCl, 50 mM HEPES hemisodium salt
pH 7.5, 5 mM DTT, 0.5 g/L BSA) with 10 min incubation time.
The concentration of fluorescent ligand was 2 nM, and the
concentration of enzyme was 3 nM for PKA and 20 nM for ROCK-
II; the final solution volume in the wells was 20 µL. The binding
curves were fitted using GraphPad Prism version 4.03 using
sigmoidal dose-response variable slope regression functions. The
dissociation/displacement constants Kd were calculated with the aid
of the online calculator for fluorescence-based competitive binding
assays (http://sw16.im.med.umich.edu/software/calc_ki/index.jsp54).
(17) Jencks, W. P. On the attribution and additivity of binding energies.
Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 4046–4050.
(18) Ricouart, A.; Gesquiere, J. C.; Tartar, A.; Sergheraert, C. Design of
potent protein kinase inhibitors using the bisubstrate approach. J. Med.
Chem. 1991, 34, 73–78.
(19) Garber, K. The second wave in kinase cancer drugs. Nat. Biotechnol.
2006, 24, 127–130.
(20) Enkvist, E.; Raidaru, G.; Vaasa, A.; Pehk, T.; Lavogina, D.; Uri, A.
Carbocyclic 3′-deoxyadenosine-based highly potent bisubstrate-analog
inhibitor of basophilic protein kinases. Bioorg. Med. Chem. Lett. 2007,
17, 5336–5339.
(21) Viht, K.; Schweinsberg, S.; Lust, M.; Vaasa, A.; Raidaru, G.; Lavogina,
D.; Uri, A.; Herberg, F. W. Surface-plasmon-resonance-based bio-
sensor with immobilized bisubstrate analog inhibitor for the determi-
nation of affinities of ATP- and protein-competitive ligands of cAMP-
dependent protein kinase. Anal. Biochem. 2007, 362, 268–277.
(22) Taylor, S. S.; Yang, J.; Wu, J.; Haste, N. M.; Radzio-Andzelm, E.;
Anand, G. PKA: a portrait of protein kinase dynamics. Biochim.
Biophys. Acta 2004, 1697, 259–269.
(23) Taylor, S. S.; Kim, C.; Vigil, D.; Haste, N. M.; Yang, J.; Wu, J.;
Anand, G. S. Dynamics of signaling by PKA. Biochim. Biophys. Acta
2005, 1754, 25–37.
(24) Bonn, S.; Herrero, S.; Breitenlechner, C. B.; Elbruch, A.; Lehmann,
W.; Engh, R. A.; Gassel, M.; Bossemeyer, D. Structural analysis of
protein kinase A mutants with Rho-kinase inhibitor specificity. J. Biol.
Chem. 2006, 281, 24818–24830.
(25) Gassel, M.; Breitenlechner, B. C.; Ru¨ger, P.; Jucknischke, U.;
Schneider, T.; Huber, R.; Bossemeyer, D.; Engh, R. A. Mutants of
protein kinase A that mimic the ATP-binding site of protein kinase B
(AKT). J. Mol. Biol. 2003, 329, 1021–1034.
Acknowledgment. We thank Ursel Soomets (Department
of Biochemistry, University of Tartu) for the support with
MALDI TOF MS analysis, and Sulev Ko˝ks (Department of
Physiology, University of Tartu) with the use of Molecular
Imager. The work was supported by grants from the Estonian
Science Foundation (6710) and the Estonian Ministry of
Education and Sciences (SF0180121s08).
Supporting Information Available: View of the total crystal
structure of PKA CR-compound 1 complex, tables listing structures
and analytical data of the inhibitors, and descriptions of synthesis
routes of inhibitors 2 and 8-12 and their precursor compounds.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(26) Gassel, M.; Breitenlechner, C. B.; Ko¨nig, N.; Huber, R.; Engh, R. A.;
Bossemeyer, D. The protein kinase C inhibitor bisindolyl maleimide
2 binds with reversed orientations to different conformations of protein
kinase A. J. Biol. Chem. 2004, 279, 23679–23690.
(27) Breitenlechner, C.; Gassel, M.; Hidaka, H.; Kinzel, V.; Huber, R.;
Engh, R. A.; Bossenmeyer, D. Protein kinase A in complex with Rho-
kinase inhibitors Y-27632, Fasudil, and H-1152P: structural basis of
selectivity. Structure 2003, 11, 1595–1607.
(28) Prade, L.; Engh, R. A.; Girod, A.; Kinzel, V.; Huber, R.; Bossemeyer,
D. Staurosporine-induced conformational changes of cAMP-dependent
protein kinase catalytic subunit explain inhibitory potential. Structure
1997, 5, 1627–1637.
(29) Engh, R. A.; Girod, A.; Kinzel, V.; Huber, R.; Bossemeyer, D. Crystal
structures of catalytic subunit of cAMP-dependent protein kinase in
complex with isoquinolinesulfonyl protein kinase inhibitors H7, H8,
and H89. Structural implications for selectivity. J. Biol. Chem. 1996,
271, 26157–26164.
(30) Akamine, P.; Madhusudan; Brunton, L. L.; Ou, H. D.; Canaves, J. M.;
Xuong, N. H.; Taylor, S. S. Balanol analogues probe specificity
determinants and the conformational malleability of the cyclic 3′,5′-
adenosine monophosphate-dependent protein kinase catalytic subunit.
Biochemistry 2004, 43, 85–96.
(31) Breitenlechner, C. B.; Wegge, T.; Berillon, L.; Graul, K.; Marzenell,
K.; Friebe, W. G.; Thomas, U.; Schumacher, R.; Huber, R.; Engh,
R. A.; Masjost, B. Structure-based optimization of novel azepane
derivatives as PKB inhibitors. J. Med. Chem. 2004, 47, 1375–1390.
(32) Zhang, X.; Gureasko, J.; Shen, K.; Cole, P. A.; Kuriyan, J. An allosteric
mechanism for activation of the kinase domain of epidermal growth
factor receptor. Cell. 2006, 125, 1137–1149.
(33) Levinson, N. M.; Kuchment, O.; Shen, K.; Young, M. A.; Koldobskiy,
M.; Karplus, M.; Cole, P. A.; Kuriyan, J. A Src-like inactive
conformation in the abl tyrosine kinase domain. PLoS Biol. 2006, 4,
753–767.
(34) Cheng, K. Y.; Noble, M. E.; Skamnaki, V.; Brown, N. R.; Lowe, E. D.;
Kontogiannis, L.; Shen, K.; Cole, P. A.; Siligardi, G.; Johnson, L. N.
The role of the phospho-CDK2/cyclin A recruitment site in substrate
recognition. J. Biol. Chem. 2006, 281, 23167–23179.
References
(1) Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam,
S. The protein kinase complement of the human genome. Science 2002,
298, 1912–1934.
(2) Breitenlechner, C.; Gassel, M.; Engh, R.; Bossenmeyer, D. Structural
insights into AGC kinase inhibition. Oncol. Res. 2004, 14, 267–278.
(3) Cohen, P. Protein kinasessthe major drug targets of the twenty-first
century. Nat. ReV. Drug DiscoVery 2002, 1, 309–315.
(4) Noble, M. E.; Endicott, J. A.; Johnson, L. N. Protein kinase inhibitors:
insights into drug design from structure. Science 2004, 303, 1800–
1805.
(5) Pelech, S. Tracking cell signaling protein expression and phosphory-
lation by innovative proteomic solutions. Curr. Pharm. Biotechnol.
2004, 5, 69–77.
(6) Gill, A. L.; Verdonk, M.; Boyle, R. G.; Taylor, R. A comparison of
physicochemical property profiles of marketed oral drugs and orally
bioavailable anti-cancer protein kinase inhibitors in clinical develop-
ment. Curr. Top. Med. Chem. 2007, 7, 1408–1422.
(7) Enkvist, E.; Lavogina, D.; Raidaru, G.; Vaasa, A.; Viil, I.; Lust, M.;
Viht, K.; Uri, A. Conjugation of adenosine and hexa-(D-arginine) leads
to a nanomolar bisubstrate-analog inhibitor of basophilic protein
kinases. J. Med. Chem. 2006, 49, 7150–7159.
(8) Nam, N.-H.; Lee, S.; Ye, G.; Sun, G.; Parang, K. ATP-phosphopeptide
conjugates as inhibitors of Src tyrosine kinases. Bioorg. Med. Chem.
2004, 12, 5753–5766.
(9) Shen, K.; Cole, P. A. Conversion of a tyrosine kinase protein substrate
to a high affinity ligand by ATP linkage. J. Am. Chem. Soc. 2003,
125, 16172–16173.
(10) Fischer, P. M. The design of drug candidate molecules as selective
inhibitors of therapeutically relevant protein kinases. Curr. Med. Chem.
2004, 11, 1563–1583.
(11) Knight, Z. A.; Shokat, K. M. Features of selective kinase inhibitors.
Chem. Biol. 2005, 12, 621–637.
(12) Parang, K.; Cole, P. A. Designing bisubstrate analog inhibitors for
protein kinases. Pharm. Ther. 2002, 93, 145–157.
(35) Viht, K.; Vaasa, A.; Raidaru, G.; Enkvist, E.; Uri, A. Fluorometric
TLC assay for evaluation of protein kinase inhibitors. Anal. Biochem.
2005, 340, 165–170.
(36) Bossemeyer, D.; Engh, R. A.; Kinzel, V.; Ponstingl, H.; Huber, R.
Phosphotransferase and substrate binding mechanism of the cAMP-
dependent protein kinase catalytic subunit from porcine heart as
deduced from the 2.0 A structure of the complex with Mn2+ adenylyl
imidodiphosphate and inhibitor peptide PKI(5-24). EMBO J. 1993,
12, 849–859.
(13) Bogoyevitch, M. A.; Barr, R. K.; Ketterman, A. J. Peptide inhibitors
of protein kinasessdiscovery, characterisation and use. Biochim.
Biophys. Acta 2005, 1754, 79–99.
(14) Lawrence, D. S. New design strategies for ligands that target protein
kinase-mediated protein-protein interactions. In Handbook of Ex-
perimental Pharmacology, Inhibitors of Protein Kinases and Protein
Phosphatases; Pinna, A. L., Cohen, P. T. W., Eds.; Springer-Verlag:
Berlin, 2005; Vol. 167, pp 11-44.

ARC-Type Inhibitor Binding to PK A Catalytic-Subunit
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 321
(37) Zheng, J.; Knighton, D. R.; ten Eyck, L. F.; Karlsson, R.; Xuong, N.;
Taylor, S. S.; Sowadski, J. M. Crystal structure of the catalytic subunit
of cAMP-dependent protein kinase complexed with MgATP and
peptide inhibitor. Biochemistry 1993, 32, 2154–2161.
(38) Gibbs, C. S.; Zoller, M. J. Rational scanning mutagenesis of a protein
kinase identifies functional regions involved in catalysis and substrate
interactions. J. Biol. Chem. 1991, 266, 8923–8931.
(39) Yang, J.; Ten Eyck, L. F.; Xuong, N. H.; Taylor, S. S. Crystal structure
of a cAMP-dependent protein kinase mutant at 1.26A: new insights
into the catalytic mechanism. J. Mol. Biol. 2004, 336, 473–487.
(40) Vaasa, A.; Viil, I.; Enkvist, E.; Viht, K.; Raidaru, G.; Lavogina, D.;
Uri, A. High-affinity bisubstrate probe for fluorescence anisotropy
binding/displacement assays with protein kinases PKA and ROCK.
Anal. Biochem. 2008, in press, doi:, 10.1016/j.ab.2008.10.030.
(41) Knighton, D. R.; Zheng, J. H.; Ten Eyck, L. F.; Xuong, N. H.; Taylor,
S. S.; Sowadski, J. M. Structure of a peptide inhibitor bound to the
catalytic subunit of cyclic adenosine monophosphate-dependent protein
kinase. Science 1991, 253, 414–420.
(42) Schneider, T. L.; Mathew, R. S.; Rice, K. P.; Tamaki, K.; Wood, J. L.;
Schepartz, A. Increasing the kinase specificity of k252a by protein
surface recognition. Org. Lett. 2005, 7, 1695–1698.
(43) Lee, J. H.; Kumar, S.; Lawrence, D. S. Stepwise combinatorial
evolution of Akt bisubstrate inhibitors. ChemBioChem. 2008, 9, 507–
509.
(44) Miick, S. M.; Jalali, S.; Dwyer, B. P.; Havens, J.; Thomas, D.; Jimenez,
M. A.; Simpson, M. T.; Zile, B.; Huss, K. L.; Campbell, R. M.
Development of a microplate-based, electrophoretic fluorescent protein
kinase A assay: comparison with filter-binding and fluorescence
polarization assay formats. J. Biomol. Screening 2005, 10, 329–338.
(45) Chijiwa, T.; Mishima, A.; Hagiwara, M.; Sano, M.; Hayashi, K.; Inoue,
T.; Naito, K.; Toshioka, T.; Hidaka, H. Inhibition of forskolin-induced
neurite outgrowth and protein phosphorylation by a newly synthesized
selective inhibitor of cyclic AMP-dependent protein kinase, N-[2-(p-
bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide (H-89), of
PC12D pheochromocytoma cells. J. Biol. Chem. 1990, 265, 5267–
5272.
(46) Nakagawa, O.; Fujisawa, K.; Ishizaki, T.; Saito, Y.; Nakao, K.;
Narumiya, S. ROCK-I and ROCK-II, two isoforms of Rho-associated
coiled-coil forming protein serine/threonine kinase in mice. FEBS Lett.
1996, 392, 189–193.
(47) Shi, J.; Wei, L. Rho kinase in the regulation of cell death and survival.
Arch. Immunol. Ther. Exp. 2007, 55, 61–75.
(48) Kang, J. H.; Jiang, Y.; Toita, R.; Oishi, J.; Kawamura, K.; Han, A.;
Mori, T.; Niidome, T.; Ishida, M.; Tatematsu, K.; Tanizawa, K.;
Katayama, Y. Phosphorylation of Rho-associated kinase (Rho-kinase/
ROCK/ROK) substrates by protein kinases A and C. Biochimie 2007,
89, 39–47.
(49) Collins, I.; Caldwell, J.; Fonseca, T.; Donald, A.; Bavetsias, V.; Hunter,
L. J.; Garrett, M. D.; Rowlands, M. G.; Aherne, G. W.; Davies, T. G.;
Berdini, V.; Woodhead, S. J.; Davis, D.; Seavers, L. C.; Wyatt, P. G.;
Workman, P.; McDonald, E. Structure-based design of isoquinoline-
5-sulfonamide inhibitors of protein kinase B. Bioorg. Med. Chem.
2006, 14, 1255–1273.
(50) Laine, J.; Ku¨nstle, G.; Obata, T.; Noguchi, M. Differential Regulation
of Akt Kinase Isoforms by the Members of the TCL1 Oncogene
Family. J. Biol. Chem. 2002, 277, 3743–3751.
(51) Cheng, Y.; Prusoff, W. H. Relationship between inhibition constant
(Ki) and concentration of inhibitor which causes 50% inhibition (IC50)
of an enzymatic reaction. Biochem. Pharmacol. 1973, 22, 3099–3108.
(52) Zhang, X.; Zhang, S.; Yamane, H.; Wahl, R.; Ali, A.; Lofgren, J. A.;
Kendall, R. L. Kinetic mechanism of AKT/PKB enzyme family.
J. Biol. Chem. 2006, 281, 13949–13956.
(53) Ito, K.; Shimomura, E.; Iwanaga, T.; Shiraishi, M.; Shindo, K.;
Nakamura, J.; Nagumo, H.; Seto, M.; Sasaki, Y.; Takuwa, Y. J.
Phosphorylation and activation of myosin by Rho-associated kinase
(Rho-kinase). Physiology 2003, 546, 823–836.
(54) Nikolovska-Coleska, Z.; Wang, R.; Fang, X.; Pan, H.; Tomita, Y.;
Li, P.; Roller, P. P.; Krajewski, K.; Saito, N.; Stuckey, J.; Wang, S.
Development and optimization of a binding assay for the XIAP BIR3
domain using fluorescence polarization. Anal. Biochem. 2004, 332,
261–273.
JM800797N