Substituted uracil derivatives as potent inhibitors of poly(ADP-ribose)polymerase-1 (PARP-1)

Article abstract of DOI:10.1016/S0960-894X(02)00602-9

A new class of PARP-1 inhibitors, namely substituted fused uracil derivatives were synthesised. Starting from a derivative with an IC₅₀=2 μM the chemical optimisation program led to compounds with more than a 100-fold increase in potency (IC

Full text of DOI:10.1016/S0960-894X(02)00602-9

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3187–3190
Substituted Uracil Derivatives as Potent Inhibitors of
Poly(ADP-ribose)polymerase-1 (PARP-1)
Henning Steinhagen,a,* Michael Gerisch,a Joachim Mittendorf,a Karl-Heinz Schlemmerb
and Barbara Albrechtc
aInstitute of Medicinal Chemistry, Pharma Research Centre, Bayer AG, D-42096Wuppertal, FRG
bInstitute of Pharmacokinetics, Pharma Research Centre, Bayer AG, D-42096Wuppertal, FRG
cInstitute of Cardiovascular Research, Pharma Research Centre, Bayer AG, D-42096Wuppertal, FRG
Received 22 April 2002; revised 2 July 2002; accepted 19 July 2002
Abstract—A new class of PARP-1 inhibitors, namely substituted fused uracil derivatives were synthesised. Starting from a deriva-
tive with an IC50=2 mM the chemical optimisation program led to compounds with more than a 100-fold increase in potency
(IC50<20 nM). Additionally, physicochemical and pharmacokinetic properties were evaluated. It could be shown that compounds
bearing a piperazine or phenyl substituted bAla-Gly side chain exhibited the best overall profile.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
inhibition with respect to benzamides. In this
communication we wish to report the optimisation of a
novel class of PARP-1 inhibitors, namely fused uracils
II (Fig. 1).
The nuclear enzyme poly-(ADP-ribose) polymerase-1
(PARP-1), also known as poly-(ADP-ribose) synthetase
(PARS), is involved in a variety of physiological pro-
cesses related to DNA repair.1 PARP-1 is activated by
DNA strand breaks and uses nicotinamide dinucleotide
(NAD) as a substrate for poly-ADP ribosylation of
several proteins. The activation of PARP-1 plays an
important role in N-methyl-d-aspartate (NMDA) and
NO induced neurotoxicity,2 cerebral ischemia3 and
ischemia/reperfusion injury.4 PARP-1 is further
involved in a variety of transcriptional mechanisms
regulating inflammation and apoptosis.1c Thus, PARP-1
became an attractive target for possible clinical use and
consequently, a variety of PARP-1 inhibitors were dis-
covered. Many of the known inhibitors fall in the class
of benzamides I (Fig. 1), which are structural analogues
of NAD and are thought to compete with NAD at the
catalytic domain.5 Compounds I were shown to be
active in the low micromolar to nanomolar range.6 In
general, cyclic derivatives of benzamides I, including
dihydroisoquinolin-1-ones,7 quinazolin-4-ones,8 benzi-
midazoles,9 phenanthridin-6-ones,10 and 1,8-naphthyli-
mides11 showed an increased potency in PARP-1
This novel class of PARP-1 inhibitors could be identi-
fied by similarity searching in our corporate compound
collection. One of the identified compounds was the tri-
fluoromethyl substituted derivative 1 that showed an
IC50 value of 2 mM. Based upon this promising finding a
lead optimisation program was started with the goal to
improve potency, physicochemical and pharmacokinetic
(PK) properties.
Chemistry
In order to systematically optimise these properties an
approach for the synthesis of the fused uracils II was
developed that allowed a broad variation of both the
N-side chain R2 and also of the condensed (hetero) alkyl
moiety. The key reaction in assembling the heterocyclic
core was the cyclisation of an imine with chlorocarbonyl
isocyanate12 as exemplified in Scheme 1. Typically, the
imine can be generated in situ by reacting the appro-
priate ketone with a primary amine followed by direct
conversion in toluene at reflux to the desired uracils.
Aqueous workup and purification by chromatography
led to pure products. Based on this method nearly 900
*Corresponding author. Tel.: +49-202-36-5171; fax: +49-202-36-
4624; e-mail: henning.steinhagen@aventis.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00602-9

3188
H. Steinhagen et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3187–3190
Figure 1. Substituted benzamides (I) and new uracil based (II, 1)
PARP-1 inhibitors.
Scheme 3. (a) Chloroacetyl chloride, triethylamine, methylene chlo-
ride, 0 ꢀC, 30 min; (b) hexamethylene tetramine, chloroform, rt, 18 h
then ethanol, HCl, 50 ꢀC, 3 h; (c) IV, EDC, HOBt, DMAP, 4-methyl-
morpholine, DMF, rt, 18 h.
compounds were synthesised and screened for inhibition
on PARP-1. The most interesting and potent derivatives
of this series are discussed in this communication.
Biological Results and Discussion
For most derivatives an amino alkanoate was used as
the amine and the carboxylic acids IV could be obtained
after saponification of III (Scheme 1). Subsequently, the
acids IV were coupled with the primary amines V or VI
to the corresponding amides 2–12 and 23–34 using
standard conditions (Schemes 2 and 3). The amines V
that were used for coupling can be prepared from the
corresponding commercially available bromo aceto-
phenones via amination according to published meth-
ods.13 The piperazine substituted amines VI can be
prepared by a simple two step sequence (acylation and
amination) as exemplified in Scheme 3.
1
All compounds were fully characterised by H NMR,
MS and HPLC and have been tested as inhibitors for
human recombinant PARP-1. Solubility was deter-
mined in phosphate buffered saline (1% DMSO) at pH
6.5 (24 h equilibration). IC50 and EC50 values were
determined in dose–response inhibition16 and cell pro-
tection assays.17 The structure–activity relationships
(SARs) of the novel PARP-1 inhibitors were system-
atically explored by varying both side chain R2 and the
(hetero) cyclohexene moiety. In the course of the studies
it could be found that compounds bearing a ring system
connected via a b-alanine-glycine (bAla-Gly) spacer
gave an optimum in potency (Tables 1–3). The attached
ring systems leading to most potent compounds were
either substituted phenyls or piperazines. In addition, it
was shown that a 3,6-dihydro-2-thiopyrane subunit
(X=S) resulted in a three to tenfold increase in potency
compared to a cyclohexenyl moiety (X=C).
Additionally to the products that were directly acces-
sible via amide formation, the aryl bromides 7–9 could
be further derivatised to the biarylic compounds 13–22
using Pd catalysed Stille or Suzuki type cross couplings
(Scheme 2). Compounds bearing spacer variations
(Table 3) were prepared via similar methods.14 Com-
pound 44 was prepared using 1,8-nonadien-5-one and
converted to 45 using by RCM reaction15 with Grubbs
catalyst (Scheme 4).
In a first optimisation round the phenyl ring of 2 was
widely varied (Table 1). Compared to 2 (IC50=450 nM)
all of the given substituents led to an improvement in
potency. The SAR regarding R3 is somewhat flat with
some EC50 values being higher than expected (e.g., 9,
14). This might be explained by a selective transport
mechanism of the compounds for membrane passage.
Regarding the overall profile (IC50, EC50 and solubility)
the biarylic compounds 19 and 20 proved to be an
optimum. Since solubility was a crucial parameter upon
optimisation
(iv
application
anticipated)
20
(sol.=305 mg/l) was considered for PK-studies. In order
to further improve solubility the phenyl ring was
replaced by the inherently better soluble piperazine ring
(Table 2). As anticipated, several potent inhibitors with
increased solubility could be identified. Similar to the
substituted phenyl series (Table 1) the SAR regarding
Scheme 1. (a) Campher sulfonic acid, toluene, reflux, 2 h, then
Cl(CO)NCO, toluene, reflux, 1 h; (b) LiOH, ethanol/water, reflux, 1 h.
Scheme 2. (a) EDC, HOBt, DMAP, 4-methylmorpholine, DMF, rt,
18 h; (b) for 7–9 (R3=Br): aryl/hetaryl-SnBu3, 5 mol% [PdCl2(PPh3)2],
DMF, 120 ꢀC, 18 h, or aryl/hetaryl-B(OH)2, 2 M Na2CO3, 5 mol%
[Pd(PPh3)4], dioxane, 80 ꢀC, 12 h.
Scheme 4. (a) 5 Mol% [RuCl2(PCy3)2(=CHPh)] (Grubbs catalyst),
methylene chloride, reflux, 72 h, 80%.

H. Steinhagen et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3187–3190
3189
R3 is relatively broad with selected EC50 values being
higher than expected (e.g., 27, 33).
the efficiency of elimination from the systemic circula-
tion (clearance). Table 5 shows PK-parameters for a
number of selected compounds. Medium clearance
values were determined for 11, 20, 29 and 32 whereas 24,
25, 26 and 28 showed high clearances. The determined
The most promising piperazine based compounds 24,
25, 26, 28, 29 and 32 were further investigated in PK
studies. Besides various substitutions at the phenyl and
piperazine group (R3 and R4) different spacer moieties
have also been examined (Table 3). It could be shown
that the ßAla-Gly type spacer in 2–34 proved to be an
optimum whereas modification led to a moderate (35–
37, 39) or drastic loss (38, 40) in activity. Inversion of
the ßAla-Gly spacer to Gly-ßAla as in 36 resulted only
in a threefold loss of activity. These findings indicate
that the spacer is crucial for remaining biological activ-
ity. The reason behind this might be a better stabilisa-
tion of a favourable conformation of R2 in an inhibitor-
enzyme complex.
Table 2. Structure–activity relationshipfor comopunds bearing
bAla-Gly-piperazine based side chains
Compd
X, R4
IC50 (nM) EC50 (nM) Sol. (mg/L)
23
24
25
26
27
28
29
30
31
32
33
34
C, phenyl
S, phenyl
S, 4-(MeO)-phenyl
S, 2-pyridinyl
S, 4-pyridinyl
S, 3-(CF3)-2-pyridinyl
S, 4-(CF3)-2-pyridinyl
S, 3-(NH2)-2-pyridinyl
C, 2-pyrimidinyl
S, 2-pyrimidinyl
S, pyridazinyl
200
30
20
35
40
40
20
70
90
20
20
50
n.d.
100
70
36
330
150
470
110
100
117
n.d.
n.d.
380
n.d.
n.d.
Apart from side chain R2 the (hetero) cyclohexene moi-
ety was also widely varied (Table 4). Substituted cyclo-
hexenones resulted in a drastic loss of potency and as
shown in Tables 1 and 2 it could be confirmed that only
a 3,6-dihydro-2-thiopyrane subunit (X=S) as in 32
resulted in an improvement of biological activity com-
pared to the cyclohexene core as in 31. All other repla-
cements including exchange of heteroatoms (e.g., 41)
and variation of the ring size (e.g., 42, 43) led to a
decrease in potency. This also includes 45 in which the
ethylene moiety was introduced as a potential biosteric
replacement for sulfur in 32.18
100
>1000
100
100
700
300
100
400
600
S, triazinyl
n.d.: Not determined.
Table 3. Structure–activity relationshipfor compounds bearing dif-
ferent spacer moieties
Some of the most potent derivatives were administered
intravenously to male Wistar rats in order to determine
Table 1. Structure–activity relationshipfor comopunds bearing
bAla-Gly-phenyl based side chains
Compds
X
S
R2
IC50 (nM)
20
32
35
36
37
38
11
39
40
S
S
150
60
Compd
X, R3
IC50 (nM) EC50 (nM) Sol. (mg/L)
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
C, H
C, p-OMe
S, p-OMe
S, p-Cl
S, p-OCHF2
C, p-Br
S, p-Br
450
300
25
60
30
80
15
70
12
25
8.5
20
70
30
30
15
12
15
25
25
20
n.d.
n.d.
80
500
200
900
70
3500
100
800
80
500
2000
500
500
200
200
100
70
n.d.
n.d.
29
n.d.
130
10
34
n.d.
15
18
10
<1
n.d.
10
2
<1
34
S
250
S, m-Br
S, p-N-pyrrolidinyl
C
C
C
C
3000
25
0
C, 00Ph-R30 =2-naphthyl
S, Ph-R3 =2-naphthyl
S, p-Ph
S, m-Ph
S, p-(O-phenyl)
S, p-(OCO-phenyl)
S, p-(4-OMe)-phenyl
S, p-(3-pyridinyl)
S, p-(4-pyridinyl)
S, p-pyridazinyl
S, p-(5-pyrimidinyl)
S, p-(3-NH2)-phenyl
400
112
305
n.d.
18
>5000
400
70
n.d.: Not determined.

3190
H. Steinhagen et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3187–3190
Table 4. Structure–activity relationshipfor compounds bearing core
variations
Moskowitz, M. A. J. Cereb. Blood Flow Metab. 1997, 17,
1143. (b) Takahashi, K.; Pieper, A. A.; Croul, S. E.; Zhang, J.;
Snyder, S. H.; Greenberg, J. H. Brain Res. 1999, 829, 46.
4. (a) Szabo, C.; Dawson, V. L. Trends Pharmacol. Sci. 1998,
19, 287. (b) Thiemermann, C.; Bowes, J.; Myint, F. P.; Vane,
J. R. Proc. Natl. Acad. Sci. 1997, 94, 679. (c) Plaschke, K.;
Kopitz, J.; Weigand, M. A.; Martin, E.; Bardenheuer, H. J.
Neurosci. Lett. 2000, 284, 109. (d) Takahashi, K.; Greenberg,
J. H. NeuroReport 1999, 10, 2017.
5. For a recent review on PARP-1 inhibitors see Larner, A. J.
Expert Opin. Ther. Patents 2002, 12, 481.
6. Costantino, G.; Macchiarulo, A.; Camaioni, E.; Pellicciari,
R. J. Med. Chem. 2001, 44, 3786.
7. Suto, M. J.; Turner, W. R.; Arundel-Suto, C. M.; Werbel,
L. M.; Sebolt-Leopold, J. S. Anti-Cancer Drug Des. 1991, 6,
107.
8. Griffin, R. J.; Srinivasan, S.; Bowman, K.; Calvert, A. H.;
Curtin, N. J.; Newell, D. R.; Pemberton, L. C.; Golding, B. T.
J. Med. Chem. 1998, 41, 5247.
Compd
X
IC50 (nM)
31
32
41
42
43
44
45
C
S
O
90
20
800
3000
900
>5000
900
—
CH2-CH2
2Â(CH¼CH2)
CH¼CH
Table 5. Pharmacokinetic parameters of selected compounds in male
Wistar rats (1 mg/kg iv)
9. White, A. W.; Almassy, R.; Calvert, A. H.; Curtin, N. J.;
Griffin, R. J.; Hostomsky, Z.; Maegley, K.; Newell, D. R.;
Srinivasan, S.; Golding, B. T. J. Med. Chem. 2000, 43, 4084.
10. (a) Soriano, F. G.; Virag, L.; Jagtap, P.; Szabo, E.; Mab-
ley, J. G.; Liaudet, L.; Marton, A.; Hoyt, D. G.; Murthy, K.
G. K.; Salzman, A. L.; Southan, G. J.; Szabo, C. Nature Med.
2001, 7, 108. (b) Li, J.-H.; Serdyuk, L.; Ferraris, D. V.; Xiao,
G.; Tays, K. L.; Kletzly, P. W.; Li, W.; Lautar, S.; Zhang, J.;
Kalish, V. J. Bioorg. Med. Chem. Lett. 2001, 11, 1687.
11. Banasik, M.; Komura, H.; Shimoyama, M.; Ueda, K. J.
Biol. Chem. 1992, 267, 1569.
Compd
AUC (mgÂh/L)
Cl (L/hÂkg)
Vss (L/kg)
t1/2 (h)
11
20
24
25
26
28
29
32
0.60
0.61
0.23
0.16
0.32
0.23
0.45
0.52
1.70
1.64
4.33
>4.50
3.11
4.30
2.21
2.01
0.50
0.27
1.39
1.01
1.19
n.c.
0.7
0.3
0.4
0.2
1.2
n.c.
0.3
0.2
0.42
0.31
12. For a review see: Gorbatenko, V. I. Tetrahedron 1993, 49,
3227.
n.c.: Not calculated.
13. Temple, C., Jr.; Wheeler, G. P.; Elliott, R. D.; Rose, J. D.;
Kussner, C. L.; Comber, R. N.; Montgomery, J. A. J. Med.
Chem. 1982, 25, 1045. (b) Stegmann, H. B.; Deuschle, G.;
Schuler, P. J. Chem. Soc., Perkin Trans. 2 1999, 547.
14. 36/39: p rep aration fromV and the corresponding amines;
38: preparation from 4-[4-(2-pyrimidinyl)-1-piperazinyl]butyl-
amine; 40: preparation from ethyl 3-aminobutanoate.
15. For a leading reference see: Furstner, A. Angew. Chem.,
Int. Ed. Engl. 2000, 39, 3012.
16. Purified recombinant human PARP-1 (Bac-To-Bac
expression system; Life Technologies); Assay volume=110 mL,
62.5 mM Tris–HCl (pH 7.4), 6.5 mM MgCl2, 0.65 mM dithio-
threitol, 100 mg/mL sonicated calf thymus DNA, 1 mg/mL
type IIA histone, 60 mM [14C]NAD+ (2.8 kBq), 50 mg/mL
PARP-1, compound to be tested; Procedure: (a) incubation of
reaction mix for 10 min at rt, (b) completion by adding 1 mL
ice cold 10% (w/v) trichloroacetic acid, (c) transfer of the
precipitate onto a filter (Printed Filter Mat A; Wallac), wash-
ing with ethanol and drying, d) determination of radioactivity
by scintillation counting; see also: Iti, S.; Shizuta, Y.;
Hayaishi, O. J. Biol. Chem. 1979, 254, 3647.
values for t1/2 were relatively short, but compatible with
the anticipated application route (iv infusion). Regard-
ing the overall profile 20 and 32 are the most promising
candidates and are under further evaluation.
In conclusion, we have discovered and optimised sub-
stituted uracil derivatives as a new class of PARP-1
inhibitors. Apart from high potency, compounds with
good solubility and medium clearance in rat could be
identified.
Acknowledgements
The authors are grateful to Dr. M. Harter, Dr. G.
Handke, Dr. A. Jensen and Dr. J. Keldenich for helpful
discussions.
17. 7500 MHEC5-T cells/well (DSM ACC 336; german col-
lection of microorganisms and cell cultures) were seeded on 96
well plates and incubated for 24 h at 37 ꢀC, 5% CO2. Then
3 mM H2O2, 6% Alamar blue and the compound to be tested
were added. Cells were incubated for 5 h (37 ꢀC, 5% CO2).
Protection of cells was measured as amount of generated
fluorescent dye from Alamar blue. Fluorescence was measured
at absorption 560 nm/emission 590 nm; see also: Zhang, J.;
Lautar, S.; Huang, S.; Ramsey, C.; Cheung, A.; Li, J.-H. Bio-
chem. Biophys. Res. Commun. 2000, 278, 590.
References and Notes
1. (a) Ha, H. C.; Snyder, S. H. Neurobiol. Dis. 2000, 7, 225.
(b) Pieper, A. A.; Verma, A.; Zhang, J.; Snyder, S. H. Trends
Pharmacol. Sci. 1999, 20, 171. (c) Chiarugi, A. Trends Phar-
macol. Sci. 2002, 23, 122.
2. (a) Zhang, J.; Dawson, V. L.; Dawson, T. M.; Snyder, S. H.
Science 1994, 263, 687. (b) Lo, E. H.; Bosque-Hamilton, P.;
Meng, W. Stroke 1998, 29, 830.
18. For a review on bioisosterism see: Patani, G. A.; LaVoie,
E. J. Chem. Rev. 1996, 96, 3147.
3. (a) Endres, M.; Wang, Z. Q.; Namura, S.; Waeber, C.;