Synthesis and SAR of novel tricyclic quinoxalinone inhibitors of poly(ADP-ribose)polymerase-1 (PARP-1)

Article abstract of DOI:10.1016/j.bmcl.2009.06.016

Based on screening hit 1, a series of tricyclic quinoxalinones have been designed and evaluated for inhibition of PARP-1. Substitutions at the 7- and 8-positions of the quinoxalinone ring led to a number of compounds with good enzymatic and cellular poten

Full text of DOI:10.1016/j.bmcl.2009.06.016

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4050–4054
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR of novel tricyclic quinoxalinone inhibitors
of poly(ADP-ribose)polymerase-1 (PARP-1)
a
b
a
a
a
Julie Miyashiro a, , Keith W. Woods , Chang H. Park , Xuesong Liu , Yan Shi , Eric F. Johnson ,
Jennifer J. Bouska a, Amanda M. Olson a, Yan Luo a, Elizabeth H. Fry b, Vincent L. Giranda a,
Thomas D. Penning a
*
a Cancer Research, GPRD, Abbott Laboratories, 100 Abbott Park Rd., Abbott Park, IL 60064, USA
b Structural Biology, GPRD, Abbott Laboratories, 100 Abbott Park Rd., Abbott Park, IL 60064, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Based on screening hit 1, a series of tricyclic quinoxalinones have been designed and evaluated for inhi-
bition of PARP-1. Substitutions at the 7- and 8-positions of the quinoxalinone ring led to a number of
compounds with good enzymatic and cellular potency. The tricyclic quinoxalinone class is sensitive to
modifications of both the amine substituent and the tricyclic core. The synthesis and structure–activity
relationship studies are presented.
Received 21 April 2009
Revised 3 June 2009
Accepted 5 June 2009
Available online 13 June 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
PARP-1
Poly(ADP-ribose)polymerase
Quinoxalinone
Nuclear enzyme
DNA repair
Programmed cell death
Nicotinamide
Poly(ADP-ribose)polymerases (PARPs) are a family of nuclear en-
zymes (17 members) involved in the regulation of a number of cel-
lular processes involving DNA repair and programmed cell death.
PARP-1, the main isoform, is one of the most abundant proteins in
the nucleus and can be activated up to 100-fold by DNA strand
breaks. This ubiquitous 116-kDa protein consists of 3 domains: an
amino(N)-terminal DNA-binding domain, an automodification do-
main, and a carboxy(C)-terminal domain. The N-terminal zinc fin-
gers of PARP-1 recognize single and double-stranded DNA breaks.
The oxidative stress induced by DNA damage catalyzes the synthesis
of poly ADP-ribose polymers on acceptor proteins with NAD+ as the
substrate. This cellular ADP-ribose transfer is an essential compo-
nent of DNA repair and the maintenance of genomic integrity.1
PARP-1 is primarily responsible for the catastrophic depletion of
NAD+ and ATP observed under high oxidative stress which culmi-
nates in cell dysfunction and death. Since tumor cells often have
compromised DNA repair mechanisms, they are more dependent
than normal cells on PARP-1 for DNA repair. Thus, using a PARP-1
inhibitor to shut down the DNA repair mechanism has the potential
to enhance the therapeutic benefit of DNA-damaging anticancer
drugs or ionizing radiation.2–7
O
NH2
N
7
H
N
H
N
O
O
N
R
8
N
N
N
H
1
2
9c
A considerable number of novel PARP inhibitors have been de-
scribed in the literature over the past decade.8 Many of the recent
medicinal chemistry efforts have focused on tricyclic and tetracy-
clic carboxamide scaffolds. In general, these inhibitors mimic the
binding mode of nicotinamide with the carboxamide group form-
ing hydrogen bonds with Gly-863 and Ser-904 in the PARP en-
zyme.4–9 Utilizing screening hit 1, whose core has a potency
comparable to the previously described benzimidazole carboxam-
ide core 2 (R = H, PARP-1 Ki = 0.240 l
M),10 a series of tricyclic qui-
noxalinone analogues were explored leading to the identification
of 9c. This compound displayed excellent enzymatic and cellular
potency. As seen in the PARP-1 X-ray co-crystal structure (Fig. 1),
the quinoxalinone scaffold accesses a binding region previously
not filled by the benzimidazole inhibitors. Access to this new
‘northern’ binding pocket may provide insights into the activity
of 9c and its analogues.
The synthesis of pyrroloquinoxalinone 9c and its analogues is
shown in Scheme 1. Attempts to arylate methyl pyrrole-2-carbox-
* Corresponding author. Tel.: +1 847 938 3686.
E-mail address: julie.miyashiro@abbott.com (J. Miyashiro).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.06.016

J. Miyashiro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4050–4054
4051
CO2CH3
NO2
CO2CH3
F
N
F
N
H
5
NO2
a
NO2
b
CO2H
CO2Me
CO2Me
3a
12
11
c
O
N
O
O
N
N
d
NHR1R2
NH
NH
NH
e
R1
CO2H
CO2Me
O
N
R2
15a-c
13
14
Figure 1. X-ray of 2 (R = 4-(N-propylpiperidine), purple) and 9c (orange) in PARP-1
binding site.11
f
ylate (5) with fluoronitrobenzoic acid 3a or 3b were unsuccessful.
The reaction was facilitated by conversion of the benzoic acid to
either the Weinreb amide 4 (Scheme 1) or the methyl ester 11
(Scheme 2). Weinreb amide 4 and methyl pyrrole-2-carboxylate
(5) were heated under basic conditions to give arylpyrrole 6.
Reduction of the nitro group and subsequent cyclization formed
the quinoxalinone core. The Weinreb amide 7 was reduced to alde-
hyde 8 with lithium aluminum hydride. Introduction of the amine
substituents at the 7- and 8-positions was made via reductive
aminations of aldehyde 8 with various amines to give compounds
9a–f and 10a–c (Table 1).
O
O
O
N
N
N
NH
h
NH
NH
g
R1
N
Br
OH
R2
17
16
9g-h
Scheme 2. Reagents and conditions: (a) SOCl2, MeOH; (b) Cs2CO3, DMF, 80 °C; (c)
10% Pd/C, EtOH; (d) 5 equiv LiOH, 5:1 THF/MeOH; (e) HATU, DIPEA, DMF; (f) LAH,
0 °C to rt; (g) PBr3, dioxane, 0 °C to rt; (h) 3 equiv amine, CH3CN.
Alternatively, compounds 9g–h were synthesized via bromide
17 shown in Scheme 2. Methyl ester 13 was reduced to alcohol
16 then converted to bromide 17 with PBr3. Reaction with amines
resulted in compounds 9g–h (Table 1). Saponification of ester 13 to
give carboxylic acid 14 followed by amide formation gave the de-
sired amides 15a–c (Table 1).
Imidazoquinoxalinones 23a–b were prepared from N-arylation
of ethyl imidazole-2-carboxylate (18) with methyl 4-fluoro-3-
nitrobenzoate (11) as shown in Scheme 3. As in the pyrroloquinox-
alinone series, the quinoxalinone core was constructed by
reduction of the nitro group followed by cyclization. Reduction of
methyl ester 20 followed by reaction with PBr3 gave bromide 22
which was reacted with various amines to give compounds 23a–
b (Table 1).
Ring expansion of the tricylic core to give the pyrrolodiazepi-
none scaffold is shown in Scheme 4. The acetal protected fluorocy-
anobenzaldehyde 25 was heated under basic conditions with
pyrrole 6 to give N-arylated pyrrole 26. The key intermediate 27
was obtained after Raney nickel reduction of the nitrile, cyclization
and deprotection of the aldehyde. Reductive amination of aldehyde
27 with the requisite amines gave compounds 28a–b.
CO2Me
CO2Me
NO2
F
F
N
H
N
NO2
NO2
a
5
b
CO2H
CON(OCH3)CH3
CON(OCH3)CH3
c
4
6
3a =4-fluoro-3nitrobenzoic acid
3b=5-fluoro-4-nitrobenzoic acid
NHR1R2
O
O
N
O
N
N
d
NH
R1
NH
e
NH
N
CHO
CON(OCH3)CH3
R2
8
7
9a-f 7-substituted quinoxalinone
10a-b 8-substituted quinoxalinone
Scheme 1. Reagents and conditions: (a) (1) (COCl)2, CH2Cl2; (2) NH(OMe)MeÁHCl; (b) Cs2CO3, DMF, 80 °C; (c) 10% Pd/C, EtOH; (d) 1 M LAH in THF, 0 °C for 0.5 h; (e)
Na(OAc)3BH, HOAc, CH2Cl2.

4052
J. Miyashiro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4050–4054
Table 1
Table 1 (continued)
SAR of tricyclic quinoxalinones
Compds
–NR1R2
PARP-1a Ki (
lM)
Cellulara EC50
—
(lM)
28a
0.589
N
O
N
O
O
O
N
N
N
NH
NH
NH
NH
N
N
28b
37a
37b
38a
1.12
—
R1
R1
N
R2
R1
O
N
N
R2
R2
0.096
0.190
0.115
0.063
>1
1
10a-c
15a-c
9a-h
N
O
0.646
0.058
0.039
N
N
O
O
O
N
N
N
N
NH
NH
NH
NH
R1
R1
R1
R1
N
N
R2
38a-b
N
N
R2
R2
23a-b
38b
R2
28a-b
N
37a-b
a
Mean of at least two determinations.
Compds
–NR1R2
—
PARP-1a Ki (
l
M)
Cellulara EC50
(lM)
1
0.238
—
H
N
9a
9b
9c
0.037
0.009
0.005
0.026
0.161
0.698
0.133
0.023
0.031
0.039
N
N
N
N
F
CO2CH2CH3
18
O
NH
N
H
CO2CH2CH3
NO2
N
NO2
N
N
0.011
0.006
0.014
0.164
—
b
a
CO2Me
CO2Me
CO2Me
20
11
19
c
9d
9e
9f
NH
N
H
N
N
N
N
N
N
O
O
O
NH
NHR1R2
e
O
N
d
NH
NH
R1
N
Br
22
OH
21
N
H
R2
9g
9h
10a
0.171
0.040
0.025
N
23a-b
N
H
N
Scheme 3. Reagents and conditions: (a) Cs2CO3, DMF, 55 °C; (b) 10% Pd/C, EtOH; (c)
1 M LAH in THF; (d) PBr3, dioxane, 0 °C to rt; (e) 3 equiv amine, CH3CN.
O
H
N
Analogs 37a–b and 38a–b were prepared as shown in Scheme
5. N-arylation of pyrrolidine 29 and piperidine 30 gave compounds
31 and 32 which were reduced and cyclized to give the tetra-
hydropyrroloquinoxalinone and tetrahydropyridoquinoxalinone
cores, respectively. Methyl esters 33 and 34 were reduced with
LAH to give the corresponding alcohols 35 and 36. Conversion of
the alcohol to the amines via the bromide was unsuccessful due
to the apparent instability of the bromide. The amines were there-
fore prepared from the alcohol using standard Mitsunobu condi-
tions to give 37a–b and 38a–b (Table 1).
N
N
10b
10c
0.047
0.048
0.031
0.029
N
H
15a
15b
15c
0.149
0.023
0.239
—
N
Compound 1 was identified as a hit from a high-throughput
screening assay (PARP-1 Ki = 0.238 lM). The tricyclic quinoxali-
none core possesses potency equivalent to the benzimidazole car-
N
0.280
—
boxamide core 2 which we have investigated extensively in our
N
N
H
labs, (R = H, PARP-1 Ki = 0.240 l
M).10 In an attempt to gain access
to a new binding pocket and potentially obtain potency through
interactions with additional residues in this pocket, we functional-
ized the 7- and 8-positions of the tricyclic quinoxalinone core with
an amine containing side chain.
N
N
23a
23b
0.280
0.519
—
—
The compounds were initially evaluated for inhibition of PARP-
1 enzyme activity.12 Inhibitors with sufficient enzymatic activity

J. Miyashiro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4050–4054
4053
bonyl group as in amides 15a–c had deleterious effects on the
PARP-1 activity (Ki = 0.023–0.239 M).
The 8-position of the quinoxalinone core was also explored as
demonstrated in 10a–c. In general, the analogues substituted in
the 7-position were more potent with 9b and 9c displaying the
best activity.
CO2Me
N
H
O
l
F
F
N
6
OCH3
CN
a
CN
CN
b
O
H
EtO
OEt
EtO
OEt
Modification of the quinoxalinone core by the installation of a
second nitrogen as in the imidazoquinoxalinones 23a–b dramati-
26
24
25
cally decreased the enzymatic potency (Ki = 0.280
Ki = 0.519 M, respectively). These analogues were even less active
than the initial screening hit 1 (Ki = 0.238 M). Increasing the ring
size of the lactam by one carbon also led to a marked decrease in
activity as seen in 28a–b (Ki = 0.589 M and Ki = 1.12 M, respec-
lM and
c
l
l
O
O
l
l
N
N
NHR1R2
d
NH
NH
tively). The pyrrolidine analogues 37a–b and the piperidine ana-
logues 38a–b displayed diminished activity and were only
slightly more potent than the screening hit (Ki = 0.063–0.190 lM).
R1
An overlay of the X-ray co-crystal structures of 9c and a benz-
imidazole carboxamide 2 (R = 4-(N-propylpiperidine) is shown in
Figure 1. The pyrrolidine nitrogen of 9c is positioned to access a
new binding pocket through a water-mediated hydrogen bond
interaction with Ser-864. Analogues substituted in the 8-position
are not in the proper orientation to access this binding region
and in general, they show a sixfold decrease in activity compared
to 9b and 9c. The previously described benzimidazole carboxamide
series does not occupy this new binding region. Within the quinox-
alinone series, it is clear that the Ser-864 interaction is important.
However, it should be noted that 9c (Ki = 5 nM) and 2 (Ki = 8 nM)10a
have comparable enzyme inhibitory activities (see also 3a in Ref.
10b). Thus, this interaction is series dependent and not required
for potent inhibition.
In summary, we have modified a novel screening hit 1, to
lead to compounds with improved enzymatic and cellular
potency against PARP-1. The quinoxalinone scaffold appears to
be very sensitive to modifications to both the amine side chain
and the tricyclic core. Acceptable enzymatic and cellular potency
in this class has been limited. Enlarging the ring system or the
addition of nitrogens was not tolerated. Further exploitation of
the new ‘northern’ binding pocket will be addressed in future
publications.
N
O
H
R2
27
28a-b
Scheme 4. Reagents and conditions: (a) CH(OEt)3, NH4NO3, EtOH,
DMF, 55 °C; (c) (1) Raney Ni, 7 M NH3 in MeOH; (2) 2 M HCl, THF; (d) Na(OAc)3BH,
D
; (b) Cs2CO3,
HOAc, CH2Cl2.
29 n=0
30 n=1
n
n
n
O
F
N
N
CO2CH3
NO2
N
CO2CH3
NO2
NH
H
b
a
CO2Me
CO2Me
CO2Me
33 n=0
34 n=1
11
31 n=0
32 n=1
c
n
n
O
O
NHR1R2
d
N
N
NH
NH
References and notes
R1
N
R2
OH
1. Virág, L.; Szabó, C. Pharmacol. Rev. 2002, 54, 375.
35 n=0
36 n=1
2. (a) Munoz-Gamez, J. A.; Martin-Oliva, D.; Aguilar-Quesada, R.; Canuelo, A.;
Nunez, M. I.; Valenzuela, M. T.; Ruiz de Almodovar, J. M.; de Murcia, G.; Oliver,
J. A. Biochem. J. 2005, 386, 119; (b) Tentori, l.; Graziani, G. Pharmacol. Res. 2005,
52, 25; (c) Shirou, S.; Nomura, F.; Tomonaga, T.; Sunaga, M.; Noda, M.; Ebara,
M.; Saisho, H. Oncol. Rep. 2004, 12, 821; (d) Griffin, R. J.; Curtin, N. J.; Newell, D.
R.; Golding, B. T.; Durkacz, B. W.; Calvert, A. H. Biochemie 1995, 77, 408.
3. (a) Plummer, E. R. Curr. Opin. Pharmacol. 2006, 6, 364; (b) Horváth, E. M.; Szabó,
C. Drug News Perspect. 2007, 20, 171; (c) Ratnam, K.; Low, J. A. Clin. Cancer Res.
2007, 13, 1383.
4. Donawho, C. K.; Luo, Y.; Luo, Y.; Penning, T. D.; Bauch, J. L.; bouska, J. J.;
Bontcheva-Diaz, V. D.; Cox, B. F.; DeWeese, T. L.; Dillehay, L. E.; Ferguson, D. C.;
Ghoreishi-Haack, N. S.; Grimm, D. R.; Guan, R.; Han, E. K.; Holley-Shanks, R.;
Hristov, B.; Idler, K. B.; Jarvis, K.; Johnson, E. F.; Kleinberg, L. E.; Klinghofer, V.;
Lasko, L. M.; Liu, X.; Marsh, K. C.; MarshMcGonigal, T. P.; Meulbroek, J. A.;
Olson, A. M.; Palma, J. P.; Rodriguez, L. E.; Shi, Y.; Stavropoulos, J. A.; Tsurutani,
A. C.; Zhu, G.-D.; Rosenberg, S. H.; Giranda, V. L.; Frost, D. J. Clin. Cancer Res.
2007, 13, 2728.
37a-b n=0
38a-b n=1
Scheme 5. Reagents and conditions: (a) Cs2CO3, DMF, 60 °C; (b) 10% Pd/C, EtOH; (c)
1 M LAH in THF, 0 °C to rt; (d) PPh3, DBAD, THF.
(Ki 6 0.200 lM) were further investigated in a PARP-1 cellular as-
say.13 The PARP-1 enzymatic and cellular data are summarized in
Table 1.
Both the isopropyl analogue 9d and the cyclopropyl analogue
9a are relatively potent inhibitors of PARP-1, showing more than
sixfold improvement in enzyme activity (Ki = 0.026
lM and
Ki = 0.037 M, respectively) as compared to the hit 1. Although
l
5. Lapidus, R. G.; Tentori, L.; Graziani, G.; Leonetti, C.; Scarsella, M.; Vergati, M.;
Muzi, A.; Zhang, J. J. Clin. Oncol. 2005, 23, 3136.
9d and 9a display comparable activity, the isopropyl analogue is
three times more potent in the cellular assay. Replacement of the
alkyl side chain with a benzyl group as in 9e resulted in the loss
6. Calabrese, C. R.; Almassy, R.; Barton, S.; Batey, M. A.; Calvert, A. H.; Canan-Koch,
S.; Durkacz, B. W.; Hostomsky, Z.; Kumpf, R. A.; Kyle, S.; Li, J.; Maegley, K.;
Newell, D. R.; Notarianni, E.; Stratford, I. J.; Skalitsky, D.; Thomas, H. D.; Wang,
L.-Z.; Webber, S. E.; Williams, K. J.; Curtin, N. J. J. Natl. Cancer Inst. 2004, 96, 56.
7. Tentori, L.; Leonetti, C.; Scarsella, M.; d’Amati, G.; Vergati, M.; Portarena, I.; Xu,
W.; Kalish, V.; Zupi, G.; Zhang, J.; Graziani, G. Clin. Cancer Res. 2003, 9, 5370.
8. For reviews, see: (a) Zaremba, T.; Curtin, N. Anti-Cancer Agents Med. Chem. 2007,
7, 515; (b) Ratnam, K.; Low, J. Clin. Cancer Res. 2007, 13, 1383; (c) Peukert, S.;
Schwahn, U. Exp. Opin. 2004, 14, 1531; (d) Southan, G.; Szabó, C. Curr. Med.
Chem. 2003, 10, 321.
of both enzymatic and cellular activity (Ki = 0.161
EC50 = 0.164 M). Cyclic amines 9b and 9c showed a significant in-
crease in potency (Ki = 0.009 M and Ki = 0.005 M, respectively),
whereas the addition of a carbonyl as in lactam 9f dramatically de-
creased the potency (Ki = 0.698 M). Increasing the length of the
side chain as illustrated by 9g–h, as well as, installation of a car-
lM and
l
l
l
l

4054
J. Miyashiro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4050–4054
9. (a) Li, J.-H.; Zhang, J. IDrugs 2001, 4, 804; (b) Cockcroft, X.; Dillon, K. J.; Dixon, L.;
Drezewiecki, J.; Kerrigan, F.; Loh, V. M.; Martin, N. M. B.; Menear, K. A.; Smith,
G. C. M. Bioorg. Med. Chem. Lett. 2006, 16, 1040.
11. The coordinates of compound 9c have been deposited in the RCSB Protein Data
Bank (PDB ID code 3GJW).
12. The enzyme assay was conducted in buffer containing 50 mM tris pH 8.0,
1 mM DTT and 4 mM MgCl2. PARP reactions contained 1.5 mM [3H]-NAD+,
200 nM biotinylated histone H1, 200 nM slDNA and 1 nM PARP-1 enzyme. The
details are described in Ref. 10b.
13. C41 cells were treated with test compound for 30 min. PARP was activated by
damaging DNA with 1 mM H2O2 for 10 min. The details are described in Ref.
10b.
10. (a) Penning, T. D.; Zhu, G.-D.; Gandhi, V. B.; Gong, J.; Thomas, S.; Lubisch, W.;
Grandel, R.; Wernet, W.; Park, C. H.; Fry, E. H.; Liu, X.; Shi, Y.; Klinghofer, V.;
Johnson, E. F.; Donawho, C. K.; Frost, D. J.; Bontcheva-Diaz, V.; Bouska, J. J.;
Olson, A. M.; March, K. C.; Luo, Y.; Rosenberg, S. H.; Giranda, V. L. Bioorg. Med.
Chem. 2008, 16, 6965; (b) Penning, T. D.; Zhu, G.-D.; Gong, J.; Gandhi, V. B.; Luo,
Y.; Liu, X.; Shi, Y.; Klinghofer, V.; Johnson, E. F.; Frost, D.; Donawho, C. K.;
Rodriguez, L.; Bukofzer, G.; Jarvis, K.; Bouska, J. J.; Osterling, D. J.; Olson, A. M.;
Marsh, K. C.; Rosenberg, S. H.; Giranda, V. L. J. Med. Chem. 2009, 52, 514.