An aminopyridazine-based inhibitor of a pro-apoptotic protein kinase attenuates hypoxia-ischemia induced acute brain injury

Article abstract of DOI:10.1016/S0960-894X(03)00733-9

Death associated protein kinase (DAPK) is a calcium and calmodulin regulated enzyme that functions early in eukaryotic programmed cell death, or apoptosis. To validate DAPK as a potential drug discovery target for acute brain injury, the first small molecule DAPK inhibitor was synthesized and tested in vivo. A single injection of the aminopyridazine-based inhibitor administered 6 h after injury attenuated brain tissue or neuronal biomarker loss measured, respectively, 1 week and 3 days later. Because aminopyridazine is a privileged structure in neuropharmacology, we determined the high-resolution crystal structure of a binary complex between the kinase domain and a molecular fragment of the DAPK inhibitor. The co-crystal structure describes a structural basis for interaction and provides a firm foundation for structure-assisted design of lead compounds with appropriate molecular properties for future drug development.

Full text of DOI:10.1016/S0960-894X(03)00733-9

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3465–3470
An Aminopyridazine-Based Inhibitor of a Pro-apoptotic Protein
Kinase Attenuates Hypoxia-Ischemia Induced Acute Brain Injury
Anastasia V. Velentza,a,b Mark S. Wainwright,d Magdalena Zasadzki,a,b
Salida Mirzoeva,c Andrew M. Schumacher,a,b Jacques Haiech,e Pamela J. Focia,b
Martin Eglif and D. Martin Wattersona,*
aDrug Discovery Program, Northwestern University, 303 E. Chicago Avenue, Ward 8-196, Chicago, IL 60611, USA
bDepartment of Molecular Pharmacology and Biological Chemistry, Northwestern University, Chicago, IL 60611, USA
cDepartment of Medicine, Northwestern University, Chicago, IL 60611, USA
dDepartment of Pediatrics, Northwestern University, Chicago, IL 60611, USA
e
´
´
Institut G. Laustriat, Faculte de Pharmacie, Universite Louis Pasteur, Illkirch, France
fDepartment of Biochemistry, Vanderbilt University, Nashville, TN, USA
Received 25 April 2003; revised 14 July 2003; accepted 14 July 2003
Abstract—Death associated protein kinase (DAPK) is a calcium and calmodulin regulated enzyme that functions early in eukary-
otic programmed cell death, or apoptosis. To validate DAPK as a potential drug discovery target for acute brain injury, the first
small molecule DAPK inhibitor was synthesized and tested in vivo. A single injection of the aminopyridazine-based inhibitor
administered 6 h after injury attenuated brain tissue or neuronal biomarker loss measured, respectively, 1 week and 3 days later.
Because aminopyridazine is a privileged structure in neuropharmacology, we determined the high-resolution crystal structure of a
binary complex between the kinase domain and a molecular fragment of the DAPK inhibitor. The co-crystal structure describes a
structural basis for interaction and provides a firm foundation for structure-assisted design of lead compounds with appropriate
molecular properties for future drug development.
# 2003 Elsevier Ltd. All rights reserved.
Acute brain injuries, such as stroke, are leading causes
of death and disability in the US, with a continuing
unmet medical need for effective therapies.1 Evaluation
of recent failures in clinical trials has allowed a con-
sensus to evolve for features required of new com-
pounds and potential therapeutic targets.2 Key among
these are the requirement that treatment be amenable to
use within the therapeutic time window of patient pre-
sentation to the clinic, which is usually hours after
injury, and that targeted tissue and cellular responses be
events occurring within this time window. One such
response is neuronal apoptosis in the penumbra sur-
rounding the brain injury core. However, early steps in
the apoptotic pathway prior to cell commitment to
death must be targeted for potential use in therapeutic
development. DAPK functions early in neuronal apop-
tosis and the levels of DAPK increase in the brain prior
to the onset of neuronal death, raising the possibility
that DAPK could be a potential drug discovery target
for acute brain injury.3 However, there are no selective,
bioavailable small molecule inhibitors of DAPK to test
this hypothesis.
We report here the development of the first DAPK
small molecule inhibitor, an alkylated 3-amino-6-phe-
nylpyridazine. Remarkably, a single intraperitoneal (ip)
injection of the inhibitor is able to diminish in vivo
brain injury in an animal model when administered 6 h
after the insult, a minimal desired therapeutic time win-
dow. These proof of principle results demonstrate the
potential utility of targeting protein kinases that func-
tion at early steps in apoptosis. Further, aminopyr-
idazine is a privileged structure in neuropharmacology,4
yet there is no structure available for how this chemo-
type interacts with a protein kinase target. Therefore,
we determined the high-resolution co-crystal structure
of the DAPK catalytic domain in the presence of a 3-
amino-6-phenylpyridazine fragment of the inhibitor.
*Corresponding author. Tel.: +1-312-503-0656; fax: +1-312-503-
0007; e-mail: m-watterson@northwestern.edu
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00733-9

3466
A. V. Velentza et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3465–3470
The novel co-crystal structure provides a precedent for
how this privileged structure can interact with a protein
kinase active site, and a firm foundation for future
development of lead compounds using as a starting
point the fragment-based, structure guided synthesis
approach.5
protein kinase II (CaMKII). Assays were performed
and data analyses done as previously described.7,8
Compounds 1 and 2 are 3-amino-6-phenylpyridazine
derivatives that differ only at C(5) on the pyridazine
ring. Compound 1 exhibited (Table 1) more selective
inhibition for DAPK compared to compound 2. The
common feature of a 6-phenyl group among com-
pounds 1 and 2 (Table 1) is also found in previously
described4 aminopyridazine drugs. To test the impor-
tance of this common feature, compound 3, which has a
methyl in place of the phenyl group, was synthesized.
Compound 3 did not inhibit DAPK at concentrations
up to 100 mM (Table 1), indicating that the 6-phenyl
group is required for DAPK inhibitory activity. Deri-
vatives of 3-amino-6-phenylpyridazine have been
shown7 to be effective cell-permeable inhibitors whose
pharmacology depends on the nature of the chemical
diversification4 of the aminopyridazine privileged struc-
ture. These precedents and our finding of a common
requirement for the 6-phenyl modification of the 3-ami-
nopyridazine raised the possibility that currently
approved drugs for CNS disorders4,13 might be undis-
covered protein kinase inhibitors, similar to the dis-
covery that experimental anti-inflammatory compounds
are protein kinase inhibitors.6,14 To address this possi-
bility, the commercially available (Aldrich) compound
4, (4-methyl-6-phenyl-pyridazin-3-yl)-(2-morpholin-4-
yl-ethyl)-amine, also known as minaprine,4 was tested
for protein kinase inhibitory activity. Compound 4 does
not show protein kinase inhibitory activity comparable
to compounds 1 and 2 (Table 1). The basis of why
compound 4 is not a protein kinase inhibitor was not
pursued as part of this investigation. Regardless, we
concluded that the diversification of 3-amino-6-phe-
nylpyridazine required for selective protein kinase inhi-
bition is distinct from that of previously developed
drugs that use this chemotype. Based on these initial
results, compound 1 was chosen for further analysis.
DAPK inhibitor discovery was done using previously
described experimental approaches.6,7 Initially, we
developed a quantitative DAPK enzyme assay8 and
screened a proprietary, pharmacology-focused library
of compounds.6 An aminopyridazine with non-selective
kinase inhibitory activity6,7,9 was the only DAPK inhi-
bitor detected (IC50 100 mM). Subsequently, the parent
structure was diversified through parallel synthesis,7 and
in vitro screening for inhibition of DAPK activity.7,9
The procedure for the synthesis of compounds 1–310ꢀ12
is illustrated in Scheme 1. Briefly, esterification of the
commercially available 11-bromoundecanoic acid
(Aldrich) yielded the corresponding acid ethyl ester,
which was then alkylated with 5,6-dihydro-benzo[h]cin-
nolin-3-ylamine, 3-amino-6-phenyl-pyridazine or 3-
amino-6-methyl-pyridazine as previously described.7
The resulting ester was hydrolyzed to generate the cor-
responding carboxylic acid, which was coupled to 3-
amino-6-phenyl-pyridazine by the HOBt/EDC method.
3-amino-6-phenyl-pyridazine and 3-amino-6-methyl-
pyridazine were prepared from the corresponding 3-
chloro-pyridazines following previously described pro-
cedures.7,9 All products were screened for their ability to
inhibit DAPK and closely related protein kinases.8,9
The latter included signaling kinases key to CNS func-
tion, such as protein kinase C (PKC), protein kinase A
(PKA), and the closely related, calmodulin-dependent
The potential of a small molecule inhibitor of an intra-
cellular signal transduction enzyme to function as an in
vivo modulator of disease or injury depends on the
upstream activating events and the downstream effector
pathways. Based on this regulatory paradigm, extended
in vitro testing of a new compound for inhibition of
protein kinases past those with structural and reg-
ulatory similarity (e.g., CaMKII), or those previously
implicated in the same tissue injury and recovery (e.g.,
PKC, PKA), has limited utility if the ultimate goal is in
vivo modulation of disease relevant biological respon-
ses.15 Further, protein kinase inhibitors with inhibition
Table 1. IC50 values for compounds on protein kinase activities
Compd
DAPK
IC50 (mM)a
CAMKII
IC50 (mM)a
PKC
IC50 (mM)a
PKA
IC50 (mM)a
1
2
3
4
13
22
>100
>100
>100
>100
>100
>100
>100
3
31
>100
>100
>100
>100
Scheme 1. Reagents and conditions: (a) EtOH, 1 equiv of 4 N HCl, 48
h, 95%; (b) 0.75% equiv of aminopyridazine, DMF, 80 ꢁC, 94%; (c)
20% concentrated HCl in AcOH, 100 ꢁC, 10 h, 90–98%; (d) 1 equiv of
3-amino-6-phenyl-pyridazine, HOBt/EDC, NMP, 0–22 ꢁC, 21 h, 60–
98%.
>100
aValues are averages of two or more determinations. The SEM
between assays was <4.

A. V. Velentza et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3465–3470
3467
constants that approximate the Km value of substrates
have proven to be effective drug discovery com-
pounds,15 indicating that feasibility experiments to
evaluate compound 1 in vivo should be considered.
Kinetic analysis revealed mixed inhibition by compound
1 against ATPor peptide substrate, with apparent Ki
values of 20 and 15 mM, respectively. The kinetics are
consistent with prior docking studies9,16 which showed
that a 3-amino-6-phenylpyridazine can occupy the ATP
site of DAPK. Therefore, compound 1 has the minimal
in vitro properties to justify proceeding immediately to
in vivo screening.
given compound 4 (minaprine) were not protected from
brain injury (Fig. 1A). These animals showed a loss
(29.3ꢂ5.5%) similar to that seen with vehicle treated
animals. In contrast, animals (n=30) given compound 1
had only a 17.3ꢂ3.2% loss. The brain tissue loss in the
compound 1 treated animals is significantly (p<0.05)
less than that seen in the two control groups. To obtain
information about neuronal viability using an estab-
lished biomarker, western blot analysis was done 3 days
after injury using a monoclonal antibody selective for
microtubule associated protein, MAP-2. As shown in
the insert of panel A, levels decreased as expected in the
injured right (R) hemisphere of vehicle-treated mice,
whereas animals administered compound 1 did not
exhibit decreases in the neuronal biomarker.
Our initial goal was to probe the ability of compound 1
to exhibit in vivo neuroprotective function when admini-
stered outside the CNS in a standard screening proto-
col. We used an established rodent model16 that
measures brain hemisphere weight loss 1 week after
injury as a quantitative end point.17 Animals were trea-
ted with compounds (5 mg/kg body weight) or solvent
vehicle administered by IPinjection just prior to HI
injury. In the first set of experiments (Fig. 1A), control
animals given solvent (n=14) showed a 30.7ꢂ5.5% loss
of brain hemisphere weight (data expressed as mean%
hemisphere weight lossꢂSEM). Compound 4 (mina-
prine) lacks kinase inhibitory activity (Table 1), allowing
its use as a structural analogue control with CNS bio-
availability for in vivo studies. Control animals (n=10)
Although minaprine has been shown previously18 to
protect against brain ischemia induced neuronal death
in other animal models when administered prior to
injury, this was not observed with the vigorous injury
induced by HI in this study. In addition, no behavioral
changes were observed in response to administration of
Compound 1, and no overt signs of toxicity in these
groups were seen over a 7-day recovery period after
exposure to compound 1. In separate experiments,27 no
cardiotoxicity was found, determined by analysis on the
QT interval in the mouse, with exposure to compound 1
with doses up to 50 mg/kg. Specifically, the corrected Q-
T (Q-Tc) interval (values pressed in msꢂSEM) in con-
trol mice was 17.32ꢂ0.45, and that from mice treated
with Compound 1 was 18.52ꢂ0.33.
To determine if the in vivo function of compound 1
could be observed in a more therapeutically relevant
scenario, we next administered the compound 6 h after
the HI insult. Again, neurological injury was attenuated
in animals treated with a single injection of compound 1
(Fig. 1B). Control animals (n=8) had a 44.0ꢂ3.3% loss
of brain hemisphere weight. In contrast, animals (n=8)
treated with compound 1 had only a 21.6ꢂ4.6% loss of
brain hemisphere weight. This difference in brain injury
between control animals and those treated with com-
pound 1 six h after injury is significant (p<0.05).
These results are remarkable in that the protection is
afforded by only a single treatment given after injury
and at a site remote from the CNS.
Our proof of concept results with compound 1 are con-
sistent with DAPK being a potential therapeutic target
for acute brain injury, and raise the significance of
future investigations that focus on drug development
using the privileged aminopyridazine structure. How-
ever, the molecular properties of compound 1 do not
make it an ideal candidate for consideration in drug
development based on the history of the development
process.19 For example, the molecular weight is already
over 500 (534.6) and the calculated LogP(ACD logD
suite, v5.11, Toronto, ON, Canada) is over 5 (6.26 ꢂ0.76).
Although oral biovailability is not a primary consideration
in critical care medicine, where administration is often
intravenous, the future development of compounds with
attractive molecular properties and potential for oral
Figure 1. Fig. 1. (A) Reduction of HI induced brain injury by 3-
amino-6-phenylpyridazine DAPK inhibitor compd 1 but not by the 3-
amino-6-phenylpyridazine analogue compd 4 (the CNS active drug
minaprine). The weight of the ischemic (right) cerebral hemisphere
from treated (vehicle control, compd 1, or compd 4) rats was com-
pared seven days after injury. Compd 4 does not attenuate brain tissue
loss as seen with compd 1. *p<0.05 by student’s two-tailed t-test. The
insert shows protection afforded by compd 1 treatment using a neu-
ronal viability biomarker, microtubule associated protein-2 (MAP-2).
Homogenate supernatants from the left, L, or right, R, brain hemi-
spheres of animals were analyzed by western blot analysis (1:500 dilu-
tion of monoclonal antibody) three days after injury and treatment
with either vehicle or compd 1. (B) The attenuation of ischemic brain
tissue loss afforded by compd 1 is also seen when the DAPK inhibitor
is given 6 h after HI injury. The percent reduction in hemispheric
weights was compared by one-way analysis of variance (StatView 5.0,
SAS, Cary, NC, USA).

3468
A. V. Velentza et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3465–3470
bioavailability should be possible based on the prior use
of aminopyridazines in drug development and their
potential for chemical diversification at multiple posi-
tions.4 In this regard, an aminopyridazine fragment
would make an excellent starting point for lead com-
pound development using structure-assisted fragment
based approaches.5 In this approach, a low molecular
weight fragment containing the chemotype of interest is
used in determination of a high-resolution co-crystal
structure with the protein target as a starting point for
structure-guided synthesis. The starting fragment does
not need to have detectable activity. The goal is to use
structure guided design and in silico filtering for desired
molecular properties in the synthesis of compounds in
activity-based discovery. Therefore, we determined the
crystal structure of the complex between the DAPK
catalytic domain and the 3-amino-6-phenylpyridazine
fragment of compound 1 to obtain experimental evi-
dence for how this chemotype interacts with the enzyme
and to provide a firm foundation for future research
using the fragment-based approach. Crystal growth and
data processing were done essentially as previously
described20 but with the exceptions noted.21
potential H-bond distance of Asp161. The smaller
aminopyridazine molecular fragment, therefore, occu-
pies space in the enzyme active site that is used by the
ATPsubstrate, but atoms on the aminopyridazine frag-
ment are oriented such that they are amenable to che-
mical diversifications with standard chemistries and
generation of products that could exploit space not used
by ATP.
Several areas around the fragment can be potentially
filled by synthetic diversification that exploits com-
plementary hydrophobic or polar interactions. One
example is the available space between the phenyl ring
of the fragment and the carbonyl group of Glu94 in
DAPK (see Fig. 2). This protein carbonyl oxygen is
hydrogen-bonded to the exocyclic amino group of the
adenine ring of ATP,20 a conserved interaction found in
kinases. This interaction could be potentially exploited
in aminopyridazine derivatives that have H-bond donor
groups on the phenyl ring of the fragment. Combined
use of such conserved features with various unique
aspects of the structure within this area of the active site
can clearly be used in the design of future syntheses.
Current investigations are focused on the use of this
crystal structure of the binary complex as a starting point
for the development of DAPK inhibitors that utilize
potential chemical complementarity and available space.
Structure-assisted synthetic design can then be filtered for
the potential of the proposed products to be rule-of-five
compliant19 in their molecular properties, and, therefore,
more attractive candidates for further development.
The structure of the DAPK:aminopyridazine complex
was determined at 1.9 A resolution. The orientation of
the aminopyridazine fragment is stabilized by hydro-
phobic and hydrogen bond interactions with the protein,
and is well-defined within the electron density (Fig. 2).
Direct comparisons of the 3-amino-6-phenylpyridazine
complex to the previously described20 AMPPnP com-
plex of DAPK (PDB file 1JKL) revealed that the 3-
amino-6-phenylpyridazine occupies part of the ATP
binding site. For example, the phenyl ring of the frag-
ment is in the same plane and occupies part of the space
filled by the adenine ring of ATP, although interactions
with the adenine-binding region of the protein are not
observed. The 3-amino group of the pyridazine ring
occupies the area used by the alpha phosphate of ATP,
in proximity to the catalytic Lys42, and is also within
In conclusion, compound 1 is the first DAPK small
molecule inhibitor described. A single administration at
a tissue site remote from the CNS and at a thera-
peutically relevant time window of hours after injury is
able to attenuate CNS damage measured 1 week later. The
crystal structure of the binary complex between this pri-
vileged chemotype and the target protein kinase reveals
regions of potential chemical complementarity to be
exploited in future syntheses of compounds with desired
molecular properties. The results provide support for the
hypothesis that targeting protein kinases which function
early in programmed cell death pathways could identify
new therapeutic approaches to acute brain injury.
Acknowledgements
This work was supported by funds from the Institute for
the Study of Aging, the National Institutes of Health, the
Northwestern Drug Discovery Program, the National
Science Foundation, the State of Illinois, and the
Department of Energy, USA. We thank Zszislaw.
Wawrzak for technical assistance, and Dr. Lester I. Bin-
der for the kind gift of MAP-2 monoclonal antibody.
Figure 2. Co-crystal structure of 5,6-dihydro-benzo[h]cinnolin-3-yl-
amine inhibitor fragment bound in the ATPsite of DAPK catalytic
domain. The 2Fo-Fc electron density map for the binary complex
between DAPK and the aminopyridazine fragment is shown for the
region proximal to the conserved catalytic lysine-42 (K42) and glu-
tamic acid-94 (E94) in the ATPsite. DAPK amino acids are shown in
stick representation and the aminopyridazine is shown in ball and
stick representation. The structure is deposited in the Protein Data
Bank with PDB ID code 1P4F.
References and Notes
1. Gorelick, P. B. Stroke 2002, 33, 862.
2. Gladstone, D. J.; Black, S. E.; Hakim, A. M. Stroke 2002,
33, 2123.

A. V. Velentza et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3465–3470
3469
3. Schumacher, A. M.; Velentza, A. V.; Watterson, D. M.
Expert Opin. Ther. Targets 2002, 6, 497.
were calculated for each animal both directly and using the
À
Á
CꢀI
formula: 100ꢃ
¼ %Damage where I represents the weight
C
4. Wermuth, C.-G. J. Heterocyclic Chem. 1998, 35, 1091.
5. Carr, R.; Jhoti, H. Drug Discov. Today 2002, 7, 522.
6. Watterson, D. M.; Mirzoeva, S.; Guo, L.; Whyte, A.;
Bourguignon, J.-J.; Hibert, M.; Haiech, J.; Van Eldik, L. J.
Neurochem. Intl. 2001, 39, 459.
7. Mirzoeva, S.; Sawkar, A.; Zasadzki, M.; Guo, L.; Velentza,
A. V.; Dunlap, V.; Bourguignon, J. J.; Ramstrom, H.; Haiech,
J.; Van Eldik, L. J.; Watterson, D. M. J. Med. Chem. 2002, 45,
563.
8. Velentza, A. V.; Schumacher, A. M.; Weiss, C.; Egli, M.;
Watterson, D. M. J. Biol. Chem. 2001, 276, 38956.
9. Velentza, A. V.; Schumacher, M.; Watterson, D. M. Phar-
macol. Ther. 2002, 93, 217.
of the ipsilateral and C the weight of the contralateral hemi-
sphere. All data are expressed as meanꢂSEM. The degree of
HI induced brain injury in each treatment group was expres-
sed as the percentage of reduction in the tissue weight of the
hemisphere ipsilateral to carotid ligation. This was calculated
as a ratio of the right (ipsilateral, ischemic) to the left (con-
tralateral, non-ischemic) hemispheric weights. The percent
reduction in hemispheric weights was compared by one-way
analysis of variance (StatView 5.0, SAS, Cary, NC, USA). The
Mann–Whitney test was used to determine significance for
nonparametric data. Statistical significance was assumed when
p<0.05. In the first set of experiments where animals were
treated prior to HI induced injury, the weight of the ischemic
hemisphere in the vehicle-treated animals was 0.2878ꢂ0.0242
g. The weight of the ischemic hemisphere in the animals trea-
ted with compound 4 (minaprine) was 0.3181ꢂ0.0328 g. The
weight of the ischemic hemisphere in the animals treated with
compound 1 was 0.3869ꢂ0.0165 g. There was no significant
difference in the weight of the non-ischemic (left) hemisphere
10. The extent of reactions was monitored by analytical
HPLC and final compounds were purified by RP-HPLC
chromatography on a preparative Microsorb (Rainin Instru-
ments; Woburn, MA, USA) C18 column using gradients of
0.1% (v/v) TFA in water and 80% aqueous acetonitrile con-
taining 0.08% TFA. Mass spectrometry was done using a Per-
1
Septives (Foster City, CA, USA) Voyager DE-Pro system. H
NMR spectrum for compound 1 was obtained on an INOVA
500 (1H, 20–500 MHz tunable) spectrometer in CD3OD.
between vehicle-treated (0.4168ꢂ0.1248 g), compound
4
(minaprine)-treated (0.4462ꢂ0.0172 g) and compound 1-trea-
ted (0.4672ꢂ0.0083 gm) animals. In the second experiment
where treatments were done 6 h after HI injury, the weight of
ischemic hemisphere in the vehicle-treated animals
(0.2484ꢂ0.0103 g) was significantly different from the com-
pound 3-treated animals (0.3191ꢂ0.0228 g).
11. Synthesis and characterization of compounds: For the
synthesis of compound 1, 11-(3-imino-5,6-dihydro-3H-ben-
zo[h]cinnolin-2-yl)-undecanoic acid (0.88 g, 2.3 mmol) reacted
with 3-amino-6-phenyl-pyridazine (0.4 g, 2.3mmol) in 2 mL
DMF. HOBt (0.35 g, 2.3 mmol) and EDC (0.44 g, 2.3 mmol)
were used. 11-(3-imino-5,6-dihydro-3H-benzo[h]cinnolin-2-yl)-
undecanoic acid-(6-phenyl-pyridazin-3-yl)-amide was prepared
as a light yellow solid. The mass (m/z) of 534.9 was that
expected for the product (calcd for C33H38N6O 534.3). NMR
18. Karasawa, Y.; Araki, H.; Otomo, S. Physiol. Behavior
1992, 52, 141.
19. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney,
P . J.Adv. Drug. Deliv. Rev. 2001, 46, 3.
20. Tereshko, V.; Teplova, M.; Brunzelle, J.; Watterson,
D. M.; Egli, M. Nature Struct. Biol. 2001, 8, 899.
1
analysis was consistent with the expected product. H NMR
(CD3OD): d 1.262–1.4 (t, 16), 1.675–1.690 (t, 2), 1.916–1.929
(t, 2), 2.461–2.490 (t, 2), 2.974–3.030 (m, 4), 4.307–4.336 (t, 2),
5.468 (s, 4), 7.285–7.524 (m, 4), 7.971–8.087 (m, 1), 8.484–
8.502 (m, 2). For the synthesis of compound 2, 11-(6-imino-3-
phenyl-6H-pyridazin-1-yl)-undecanoic acid (0.88 g, 2.3 mmol)
reacted with 3-amino-6-phenyl-pyridazine (0.4 g, 2.3 mmol)
exactly as described for compound 1. The mass (m/z) of 509.1
was that expected for the product (calcd for C31H36N6O
508.3). For the synthesis of compound 3, 11-(6-imino-3-
methyl-6H-pyridazin-1-yl)-undecanoic acid (0.1 g, 0.35 mmol)
reacted with 3-amino-6-phenyl-pyridazine (0.06 g, 0.35 mmol)
exactly as described for compound 1. The mass (m/z) of 447.4
was that expected for the product (calcd for C26H34N6O
446.3).
21. Crystals were grown at 20 ꢁC by mixing the protein and
aminopyridazine solutions (6 mg mLꢀ1 protein, 30 mM 5,6-
dihydro-benzo[h]cinnolin-3-ylamine) with an equal volume of
1.1 M ammonium sulfate, 100 mM Hepes pH 7.5. The crystals
were transferred to a non-polar stabilization buffer of 20%
PEG 8000, 100 mM Hepes pH 7.5, containing 15 mM of the
5,6-dihydro-benzo[h]cinnolin-3-ylamine to ensure the binding
of the low affinity fragment. The crystals were cryoprotected
with 20% glycerol in the stabilization buffer and flash-cooled
in liquid nitrogen. Diffraction data were measured at the
DuPont-Northwestern-Dow Collaborative Access Team
(DND-CAT) beamline 5-ID-B, at the Advanced Photon
Source (APS), Argonne, IL, USA and processed with the HKL
suite of programs.22,23 The initial model, 1JKL20 with waters
and ligands removed, was positioned by rigid body refinement
using CNS.24 Model building was done using O.25 Further
refinement included simulated annealing, positional minimiza-
tion and temperature factor protocols using CNS, employing
the bulk solvent correction.22 The structure has been deposited
in the Protein Data Bank and the PDB ID code is 1P4F.
22. Data collection and refinement statistics for the crystal
structure of the complex of DAPK catalytic domain with 5,6-
Dihydro-benzo[h]cinnolin-3-ylamine: Space group P212121,
unit cell a=47.24 b=62.53 c=88.93 A, resolution range 30–
1.9 A (highest bin 1.97–1.90 A), completeness 97.9% (97.3%),
12. Contreras, J. M.; Parrot, I.; Sippl, W.; Rival, Y. M.;
Wermuth, C. G. J. Med. Chem. 2001, 44, 2707.
13. Worms, P.; Chambon, J. P.; Wermuth, C. G. Actual.
´
Chim. Ther. 1989, 16, 303.
14. Mirzoeva, S.; Koppal, T.; Petrova, T. V.; Lukas, T. J.;
Watterson, D. M.; Van Eldik, L. J. Brain Res. 1999, 844, 126.
15. Capdeville, R.; Buchdunger, E.; Zimmermann, J.; Matter,
A. Nat. Rev. Drug Discov. 2002, 1, 493.
16. Schumacher, A. M.; Velentza, A. V.; Watterson, D. M.;
Wainwright, M. S. Biochim. Biophys. Acta 2002, 1600, 128.
17. Quantification of brain injury: All experiments were per-
formed in accordance with the relevant National Institutes of
Health guidelines, and protocols were approved by the Insti-
tutional Animal Care and Use Committee of Northwestern
University, Chicago. The comparison of disparities in hemi-
spheric weight as a valid outcome measure of neurologic
injury in the newborn rat has been demonstrated in studies26
correlating changes in hemispheric weight following hypoxia-
ischemia with other measures of brain injury. Animals were
sacrificed seven days after injury and differences in weight
between the HI injured and control contralateral hemisphere
R
merge=0.049 (0.213), Average I/s(I)=25.8 (5.5). The final
model contains 2275 protein, 193 water, and 15 aminopyr-
idazine fragment atoms, has an Rcryst of 20.7% and Rfree of
25.6%, and mean temperature factors of 20.8 A2 for protein
atoms, 26.6 A2 for waters and 23.1 A2 for aminopyridazine
atoms.
23. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276,
307.
24. Brunger, A. T. Crystallography & NMR System (CNS),
¨
Version 0.9; Yale University: New Haven, 1998.

3470
A. V. Velentza et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3465–3470
25. Jones, T. A.; Zou, J. Y.; Cowan, S. W.; Kjeldgaard, M.
Acta Crystallogr. A 1991, 47, 110.
26. McDonald, J.; Roeser, N.; Silverstein, F.; Johnston, M.
Exp. Neurol. 1989, 106, 289.
27. Mice were anesthetized using ketamine and isoflurane
with temperature monitored by rectal probe and maintained at
37.0 ꢁC by surface heating and cooling. Anode, cathode and
ground leads were placed subcutaneously and the electro-
cardiogram (ECG) recorded using commercially available
software (Maclab, AD Intstruments, Colorado Springs, CO,
USA). After stabilization, compound 1 or control solutions
was administered and the ECG recorded for up to 60 min.
Mice treated with quinidine (positive control) showed an
increase in the Q-Tc to 21.48ꢂ1.11, significantly different
(p<0.001 by ANOVA) from the control mice (17.32ꢂ0.45) or
the compound 1 treated mice (18.52ꢂ0.33).