Design, synthesis, and biological evaluation of a series of benzo[ de ][1,7]naphthyridin-7(8 H)-ones bearing a functionalized longer chain appendage as novel PARP1 inhibitors

Article abstract of DOI:10.1021/jm301825t

A series of benzo[de][1,7]naphthyridin-7(8H)-ones possessing a functionalized long-chain appendage have been designed and evaluated as novel PARP1 inhibitors. The initial effort led to the first-generation PARP1 inhibitor 26 bearing a terminal phthalazin-

Full text of DOI:10.1021/jm301825t

Subscriber access provided by the Library Service | University of Stellenbosch
Article
Design, Synthesis and Biological Evaluation of a Series of
Benzo[de][1,7]naphthyridin-7(8H)-ones Bearing a Functionalized
Longer Chain Appendage as Novel PARP1 Inhibitors
Na Ye, Chuan-Huizi Chen, Tian Tian Chen, Zilan Song, Jin-Xue He, Xia-Juan Huan,
Shanshan Song, Qiufeng Liu, Yi Chen, Jian Ding, Yechun Xu, Ze-Hong Miao, and Ao Zhang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm301825t • Publication Date (Web): 08 Mar 2013
Downloaded from http://pubs.acs.org on March 13, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis and Biological Evaluation of a Series of
Benzo[de][1,7]naphthyridinꢀ7(8H)ꢀones Bearing a Functionalized
Longer Chain Appendage as Novel PARP1 Inhibitors
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
,§
‡,§
ζ,§
†
‡
Na Ye, ChuanꢀHuizi Chen, TianTian Chen, Zilan Song, JinꢀXue He, XiaꢀJuan
‡
‡
ζ
‡
‡
ζ,*
Huan, ShanꢀShan Song, Qiufeng Liu, Yi Chen, Jian Ding, Yechun Xu, ZeꢀHong
‡
,*
†,
∗
Miao, and Ao Zhang
†
CAS Key Laboratory of Receptor Research, and Synthetic Organic & Medicinal Chemistry
‡
ζ
Laboratory, State Key Laboratory of Drug Research, and CAS Key Laboratory of Receptor Research,
and Drug Discovery and Design Center, Shanghai Institute of Materia Medica (SIMM), Chinese
§
Academy of Sciences, Shanghai, China 201203. These three authors contributed equally to this work.
∗To whom correspondence should be addressed. For Ao Zhang, aozhang@mail.shcnc.ac.cn; Tel:
8
8
6ꢀ21ꢀ50806035; Fax: 86ꢀ21ꢀ50806035; for ZeꢀHong Miao, zhmiao@mail.shcnc.ac.cn. Tel.:
6ꢀ21ꢀ50806820; Fax, 86ꢀ21ꢀ50806820; For Yechun Xu, ycxu@mail.shcnc.ac. cn; Tel. 86ꢀ21
50801267.
a
Abbreviations: ADP, adenosine 5’ꢀdiphosphate; PARP, poly(ADPꢀribose) polymerases; NAD+,
nicotinamide adenine dinucleotide; PAR, poly(ADP)ribose; BRCA1, breast cancer gene 1; DNA,
deoxyribonucleic
acid;
EDCI,
NOE,
1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl)carbodiimide;
Nuclear Overhauser effect;
HOBt,
1ꢀhydroxyꢀ1Hꢀbenzotriazole;
HBTU,
2
ꢀ(1HꢀBenzotriazoleꢀ1ꢀyl)ꢀ1,1,3,3ꢀtetramethyluronium hexafluorophosphate; PK, pharmacokinetics,
Boc, tertꢀbutyloxycarbonyl; DIPEA, diethylpropylamine; DMF, dimethylformamide; TFA,
trifluoroacetic acid; DAST, diethylaminosulfur trifluoride; DAPI, 4',6ꢀdiamidinoꢀ2ꢀphenylindole; MW,
microwave; FITC, fluorescein isothiocyanate; GAPDH, glyceraldehydeꢀ3ꢀphosphate dehydrogenase;
DMEM, Dulbecco's Modified Eagle's Medium; SRB, sulforhodamine B; FBS, fetal bovine serum;
OPD, oꢀphenylenediamine.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract.
A series of benzo[de][1,7]naphthyridinꢀ7(8H)ꢀones possessing a functionalized
longꢀchain appendage have been designed and evaluated as novel PARP1 inhibitors.
The initial effort led to the firstꢀgeneration PARP1 inhibitor 26 bearing a terminal
phthalazinꢀ1(2H)ꢀone framework and showing remarkably high PARP1 inhibitory
activity (0.31 nM) but only moderate potency in the cell. Further effort generated the
secondꢀgeneration lead 41 showing high potency against both the PARP1 enzyme
and BRCAꢀdeficient cells, especially for the BRCA1ꢀdeficient MDAꢀMBꢀ436 cells
(CC < 0.26 nM). Mechanistic studies revealed that the new PARP1 inhibitors
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
2 2
significantly inhibited H O ꢀtriggered PARylation in SKOV3 cells, induced cellular
accumulation of DNA doubleꢀstrand breaks and impaired cellꢀcycle progression in
BRCA2ꢀdeficient cells. Significant potentiation on the cytotoxicity of temozolomide
was also observed. The unique structural character and exceptionally high potency of
41 made it stand out as a promising drug candidate worthy for further evaluation.
2
ACS Paragon Plus Environment

Page 3 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
a
Poly(ADPꢀribose) polymerases (PARPs) are ADPꢀribosyltransferases that cleave
+
nicotinamide adenine dinucleotide (NAD ) into ADPꢀribose and nicotinamide, and
then transfer the ADPꢀribose units onto a variety of target proteins (such as histones,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
ꢀ3
topoisomerases, DNA polymerases, and DNA ligases) or themselves. The PARP
family is comprised of 18 members and regulates diverse cellular functions including
DNA recombination and repair, cellular proliferation, apoptosis in ischemic
4
condition and necrotic death. Among the PARP superfamily, PARP1 is the most
abundant and wellꢀcharacterized PARP enzyme. It accounts for more than 90% of
ADPꢀribosylation occurring in cells and is highly evolutionarily conserved in all
5
,6
advanced eukaryotes. PARP1 is an enzyme of 113ꢀkDa protein containing three
major structural domains, a 42ꢀkDa DNA binding domain at the Nꢀterminal region
with two zinc fingers, a 55ꢀkDa catalytic domain at the Cꢀterminal region, and an
7
ꢀ9
automodification domain at the central region. The DNA binding domain can
recognize and locate the damaged DNA singleꢀstrand breaks and then bind to them
to stimulate polymerization of ADPꢀribose leading to the unwinding of DNA from
histones and exposing the damaged DNA for repair. The function of the catalytic
+
domain of PARP1 is to utilize NAD as a substrate to construct linear and branched
polymers of ADPꢀribose onto its targets therefore enabling it to perform a pivotal
role in DNA damage repair. Among the PARP family, PARP2 has high structural
homology with PARP1 and shares similar functionality. Interestingly, the DNA
binding domain of PARP2 is distinct from PARP1 and lacks the central
automodification domain. PARP1 and to a lesser extent, PARP2, function as DNA
damage sensors by binding with high affinity to the site of single and double strand
DNA breaks. PARP1 is overꢀexpressed in a variety of cancers, and its expression has
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
been associated with overall prognosis in cancers, especially breast and ovarian
5
,8,9
cancers.
In view of the key role of PARP1 in maintaining the genomic integrity, in
particular the repair of single strand DNA lesions caused by ionizing radiation,
chemotherapy, UV light, or products of cellular and oxidative metabolism, PARP1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5,10
has emerged as an attractive anticancer drug target anchoring DNA damage.
The
significance of PARP1 in cancer was further strengthened recently by two seminal
11,12
preclinical studies
disclosing that cell lines deficient in BRCA1 and BRCA2 (two
known tumor suppressor genes whose mutations are associated with breast, ovary,
prostate, as well as pancreas cancers) were 1000ꢀfold more sensitive to PARP
inhibition in comparison to wildꢀtype or heterozygous mutant cells. Such landmark
discovery highlighted that PARP1 is the synthetic lethal partner of BRCA1 and
BRCA2, and inhibition of PARP1 can induce highly selective killing of
BRCA1/2ꢀdeficient tumor cells. In this regard, many small molecules targeting
PARP1 have been developed either as chemosensitizers in combination with ionizing
radiation or DNAꢀdamaging chemotherapeutic agents or as standꢀalone therapies
(synthetic lethal) to kill cancers defective in DNA repair mechanisms (esp.
1
3ꢀ22
BRCA1/2ꢀmutant breast or ovarian tumors).
Among them, over a dozen are being
23a,b
pursued into different stages of clinical trials, including arylamide 1
(BSIꢀ201,
24
25
iniparib), azepinoindolone 2 (AG014699, rucaparib), phthalazinone 3 (AZD2281,
26
olaparib) and benzo[d]imidazoleꢀ4ꢀcarboxamide 4 (ABTꢀ888, veliparib) (Figure 1).
All these drugs have shown promising singleꢀagent activity in recurrent epithelial
ovarian cancer patients with germline BRCA1 or BRCA2 mutations, especially for
compound 3 which provided clinical proofꢀofꢀconcept for the use of PARP1
inhibitors for the treatment of cancers bearing known BRCA mutations achieving an
4
ACS Paragon Plus Environment

Page 5 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
objective response rate of 28ꢀ33% and a median duration of response of 28ꢀ41 weeks
27ꢀ31
in phase I/II studies.
However, the enthusiasm of further development of this
compound was dampened by the discontinuation announcement made by
32
AstraZeneca very recently after an interim analysis of a Phase II study which
indicated that the previously reported progression free survival benefit was not
translated into an overall survival benefit, the definitive measure of patient benefit in
ovarian cancer. Therefore, although additional efforts on the working mechanism of
the inhibitor and the association between PARPꢀdependent pathways and
tumorꢀspecific defects are needed, the final fate of PARP1ꢀtargeting strategy has to
wait for the clinical outcomes of other PARP1 inhibitors or development of new
inhibitors bearing more structural diversity and showing improved efficacy, better
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20ꢀ22,33,34
safety and pharmacokinetic profiles.
Most currently available PARP1 inhibitors were designed to mimic the
+
15,18,35,36
substrateꢀprotein
interactions
of
NAD
with
PARP1.
3ꢀAmino/nitroꢀbenzamides (e.g. 1 and 4 in Figure 1) were the earliest PARP1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibitors structurally analogous to the PARylation byꢀproduct, nicotinamide,
2
3b
although these compounds may also hit other enzymes that bind the nicotinamide
+
derivative NAD . The key interactions between these inhibitors and PARP1 include
the Hꢀbonding networks formed between the carboxamido component and Ser904
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
15
and Gly863 of PARP1, and the π–π stacking between the aryl ring and Tyr907.
These compounds generally showed lowꢀtoꢀmoderate activity toward PARP1 and
nonꢀoptimal PK properties. The rest majority of PARP1 inhibitors are bicyclic or
tricyclic artworks where the key carboxamido moiety is buried into a ring system to
form a lactam. Meanwhile, an auxiliary appendage with a shorter (e.g. 2) or longer
(e.g. 3) linking chain is generally attached to the polycyclic core as a solvent
15,35,36
accessory region.
3
7ꢀ39
Our group
bioisosteric replacement of the catecholic component of antiparkinsonian drug
apomorphine (5) to generate new dopamine D receptor agonists with elevated
has long been working on the SAR of aporphine analogues by
2
metabolic stability. We recently found that replacement of the catechol function in 5
with a lactam led to benzo[de][1,7]naphthyridinꢀ7(8H)ꢀones 6 that were completely
40
inactive at any of dopamine receptors. Interestingly, Torrisi and coꢀauthors recently
reported that compound 6 with an Nꢀacyl group (e.g. R=pyrrolidinꢀ3ꢀyl) displayed
oneꢀdigit nanomolar PARP1 enzymatic potency but only moderate or no inhibition to
the PARylation in whole cells. This result promoted us to switch our focus on the
development of novel PARP1 inhibitors bearing the benzo[de][1,7]naphthyridinꢀ
7(8H)ꢀone scaffold with a longꢀchain appendage as that in clinical compound 3.
SAR optimization campaign on the nature and length of the sideꢀchain appendage
was expected to achieve novel compounds with high potency both on the PARP1
enzyme and in the BRCA1 or BRCA2 mutant cells.
6
ACS Paragon Plus Environment

Page 7 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CHEMICAL SYNTHESIS
As outlined in Scheme 1, 1,2,3,8,9,9aꢀhexahydroꢀ7Hꢀbenzo[de]ꢀ1,7ꢀnaphthyridin
41
ꢀ
7ꢀone (7), prepared from commercially available 1,2,3,4ꢀtetrahydroisoquinoline by
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
following a literature procedure, was used as the key intermediate. Alkylation of 7
with appropriate ethyl bromoalkylates provided corresponding esters 8aꢀc in 35ꢀ70%
yields. Hydrolysis of esters 8aꢀc yielded corresponding acids 9aꢀc. Compounds 10ꢀ13
bearing a bicyclic amine functionality were prepared by treating the key intermediate
9
a with appropriate cyclic amine substrates in the presence of EDCI and HOBt in
42
65% yields. Likewise, condensation of acids 9aꢀc with aminoacetamides 14, which
in turn were prepared by reaction of NꢀBocꢀglycine with corresponding amines,
yielded compounds 15ꢀ20 possessing a longꢀchain appendage in 65ꢀ85% yields.
Scheme 1. Synthesis of Amides 10-13 and 15-20.a
H
H
N
N
NH
1
0 (66%), X=
O
O
O
O
NH
NH
NH
NH
i
ii
from 9a
11 (64%), X=
12 (64%), X=
NH
O
O
iii
O
H
N
NH
N
N
N
N
O
OH
X
n
n
7
8a-c (35-70%)
9a-c
10-13
N
N
N
(
a: n=1; b; n=2; c: n=3)
13 (65%), X=
N
O
N
CF3
R
NH
X
Y
R
HO
Boc
iv
iii
R
X
Y
N
+
N
O
X
H
HN
NH2
O
N
Y
N
n
O
H
O
15 (80%): n=1, R=4-OMe-Ph, -X-Y- = -N-CH2-
14 (90-95%)
16 (85%): n=2, R=4-OMe-Ph, -X-Y- = -N-CH2-
17 (82%): n=3, R=4-OMe-Ph, -X-Y- = -N-CH2-
1
8 (65%): n=2, R=2-F-Ph, -X-Y- = -N-CH2-
19 (68%): n=2, R=pyrimidin-2-yl, -X-Y- = -N-CH2-
20 (82%): n=2, R=4-F-Ph, -X-Y- = -C=CH-
a
o
Reagents and Conditions: i) Br(CH ) COOEt, K CO , NaI, MeCN, 60 C, 8h to 24h; ii) LiOH, THF:H O = 3:1,
2
n
2
3
2
o
6
0 C, 10h; iii) EDCI, HOBt, Et N, DMF, rt, 12h; iv) EDCI, HOBt, Et N, DMF, rt, 12h, then TFA, CH Cl , rt, 12h.
3
3
2
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Meanwhile, condensation of 4ꢀ(4ꢀfluorophenyl)ꢀ1,2,3,6ꢀtetrahydropyridine with
NꢀBocꢀDꢀalanine and NꢀBocꢀLꢀalanine under the same condition provided Rꢀ21 and
Sꢀ21, subsequent deprotection and reꢀcondensation led to Rꢀ22 and Sꢀ22 in 74% and
6
5% overall yield, respectively. Similarly, condensation of acid 9a with NꢀBocꢀDꢀ
and Lꢀphenylalanine (Rꢀ23, Sꢀ23) followed by deprotection yielded Rꢀ24 and Sꢀ24 in
5% and 42% yield, respectively (Scheme 2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
Scheme 2. Synthesis of Amides 21,22,24.a
O
F
Boc
NH
NH
O
*
HO
N
F
9a
O
BocHN
O
N
N
ii, i
i
R-21 (99%)
S-21 (96%)
N
H
N
H
O
R-22 (74%); S-22 (65%)
F
F
O
O
Bn
*
HO
Boc
NH
NH
O
N
H
O
2
9a
O
Bn
Bn
+
F
3 (R, S)
i
N
N
N
N
N
H
N
i,ii
H
O
O
R-24 (35%)
S-24 (42%)
a
Reagents and Conditions: i) EDCI, HOBt, Et N, DMF, rt, 12h; ii) TFA, CH Cl , rt, 12h.
3
2
2
43
Amides 26ꢀ29 were prepared by condensation of amine 7 with acids 25a – 25d
as a new series of benzo[de][1,7]naphthyridinꢀ7(8H)ꢀones bearing a bulky chain in
2ꢀ60% yields. Inserting varies amino acid moieties (racemic proline was used for
preparation of compound 32) to the connecting linkage yielded compounds 31ꢀ33 in
5ꢀ70% overall yields (Scheme 3).
5
4
8
ACS Paragon Plus Environment

Page 9 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
4
During the Schmidt rearrangement of cyclopenta[ij]isoquinolinꢀ7(1H)ꢀone 34
under catalytic amount of conc. H SO , the key intermediate 7 was obtained as a
2
4
minor product in 18% yield, and the regiomer 7’ was produced as a dominant
product in 61% yield (Scheme 4). Optimization of the rearrangement conditions with
various acid catalysts (conc. HCl, CF COOH or MeSO H) established that by using
3
3
a mixture of solvent system (MeSO H:CH Cl = 1:1), compound 7 was obtained as
3
2
2
the major product in 78% yield. NꢀBenzylation of 7 followed by LiAlH
4
reduction
led to key intermediate 36 in 42% overall yield. NꢀBoc protection of 36 followed by
4
5
deꢀbenzylation provided 37 in 69% yield. Condensation of acid 25a with 37 under
HBTU/DIPEA/DMF followed by removal of NꢀBoc group yielded amide 38 in 54%
overall yield. Similarly, condensation of acid 25a with amine 36 followed by
hydrogenation provided amide 39 in 23% overall yield.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 4. Synthesis of Compounds 35-39.a
O
O
H
N
O
N
O
NH
i
ii
NH
+
NH
NH
NH
Ph
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
34
7' (11%)
7 (78%)
3
5 (62%)
H
N
O
iii
N
Boc
N
NH
NH
vi,vii
iv,v
N
Ph
NH
7 (69%)
N
3
36 (71%)
O
F
38 (54%)
vi,v
O
N
N
F
NH
O
NH
3
9 (23%)
a
o
Reagents and Conditions. i) NaN , MeSO H:CH Cl =1:1, 0 C to rt, 24h; ii) BnBr, K CO ,
3
3
2
2
2
3
o
MeCN, 60 C, 2h; iii) LiAlH , THF, reflux, 12h; iv) (Boc) O, Et N, CH Cl , rt, 1h; v) 10% Pd/C,
4
2
3
2
2
H , AcOH, EtOH, rt, 8h; vi) 25a, HBTU, DIPEA, DMF, rt, 12h; vii) TFA, CH Cl , rt, 12h.
2
2
2
To explore the impact of regiochemistry, compound 40 derived from regiomer 7’
was also prepared in a similar manner in 60% yield (Scheme 5). Meanwhile, the
precursor 34 used for Schmidt rearrangement was also treated with acid 25a led to
amide 41 lacking the lactam function in 54% yield. Further, reduction of 41 with
o
NaBH
4
provided alcohol 42 in 91% yield, which was then treated with DAST at 0 C
46
afforded fluoride 43 in 73% yield (Scheme 5). Reductive amination of 41 with
o
47
4 3
NH OAc and NaBH CN under microwave irradiation (5 min, 130 C) yielded
amine 44 in 72% yield.
1
0
ACS Paragon Plus Environment

Page 11 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 5. Synthesis of Compounds 40-44.a
H
N
O
N
H
N
O
i
7' + 25a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H
N
N
O
N
X
O
F
40 (60%)
N
H
ii
N
O
O
F
N
O
O
42 (91%, X=OH)
iii
i
4
3 (73%, X=F)
34 + 25a
iv
N
H
F
N
O
N
NH2
4
1 (54%)
N
O
F
44 (72%)
a
Reagents and Conditions.i) HBTU, DIPEA, DMF, rt, 12h; ii) NaBH , MeOH, rt, 12h; iii)
4
o
o
DAST, CH Cl , 0 C, 5 min; iv) NH OAc, NaCNBH , EtOH, MW, 130 C, 5 min.
2
2
4
3
2
,2,2ꢀTrifluoroꢀNꢀ(3ꢀoxoꢀ2,3ꢀdihydroꢀ1Hꢀindenꢀ1ꢀyl)acetamide (49), devoid of the
tetrahydroisoquinoline skeleton was prepared from commercially available
48
D/Lꢀβꢀphenylalanine in three steps following a literature procedure (Scheme 6).
Amide 46 was obtained in 51% overall yield by removing the trifluoroacetyl group of
4
9 followed by condensation with acid 25a. Reduction of ketone 46 with NaBH at rt
4
afforded cisꢀisomer 47 and transꢀisomer 48 in nearly identical yield (49%). The cisꢀ
and transꢀconfigurations were deduced by comparison of the NMR data with similar
4
8
compounds, and by NOE NMR experiments (see supporting information).
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2 3 4
Methylation of 49 with MeI, K CO and Bu NBr, followed by removal of
trifluoroacetyl group yielded Nꢀmethylamine 51 in 20% overall yield for two steps.
Subsequent condensation of 51 with acid 25a under HBTU/DIPEA/DMF afforded
amide 52 in 68% yields. Schmidt rearrangement of ketone 49 with CF COOH as the
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
49
50
acid catalyst led to lactam 53 as the major product in 61% yield. After deprotection
of 53 with 0.2 N Ba(OH) solution followed by standard condensation with acid 25a
2
or 25e, corresponding amides 55 and 56 were obtained in 49% and 61% yield,
respectively.
Scheme 6. Synthesis of Compounds 46-48 and 52, 55,56.a
HO
O
F
Ar
O
N
H
HN
N
O
O
O
F
47 (49%, cis-isomer)
2
5a
iii
ii
N
HO
+
H
NH2
45 (83%)
O
Ar
4
6 (61%)
i
N
H
F
O
O
O
4
8 (48%, trans-isomer)
iv
O
i
ii
Ar
O
NCOCF3
0 (60%)
N
H
NHCOCF3
N
49
5
51 (34%)
F
v
52 (68%)
O
O
H
O
i
25a or 25e
N
NH
NH
O
Ar
ii
N
H
NHCOCF3
NH2
X
53 (61%)
54
55 (49%, X=F)
56 (61%, X=H)
O
HOOC
H
O
N
HN
N
N
Ar=
'
25e
a
Reagents and Conditions. i) 0.2 N Ba(OH) , MeOH, rt, 12h; ii) HBTU, DIPEA, DMF, rt, 12h; iii)
2
NaBH , MeOH, rt, 12h; iv) MeI, K CO , Bu NBr, MeCN, rt, 24h; v) NaN , CF COOH, reflux, 3d.
4
2
3
4
3
3
1
2
ACS Paragon Plus Environment

Page 13 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
PARP1 Inhibition and Lead Generation. All the newly synthesized
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
benzo[de][1,7]naphthyridinꢀ7(8H)ꢀones were evaluated for their ability to inhibit
25,40
PARP1 enzymatic activity.
The clinical compound 3 was used as the reference
25
compound and our data were identical to that reported . First, compounds 10ꢀ13 with
a bicyclic amino group attached to the N1 of benzo[de][1,7]naphthyridin ꢀ7(8H)ꢀone
core through a 2ꢀoxoꢀethyl (acetamide) linker were prepared to increase molecular
flexibility. As shown in Table 1, all the new compounds displayed negligible activity
against PARP1 with IC50 values of approximately 10 ꢁM. Interestingly, switch of the
bicyclic amino appendage to arylpiperazinyl and inserting a second acetamide moiety
to the linker afforded compound 15, showing significant improvement in PARP1
activity with an IC50 value of 300 nM. To evaluate the impact of the length of the
linker on the PARP1 inhibition, compounds 16 and 17 were designed containing a
longer linker. However, both compounds lost PARP1 inhibitory activity.
Manipulation of the aryl portion in the arylpiperazinyl fragment with other
heterocycles or with various substitutions (e.g. compounds 18 and 19) led to no
improvement at all. It is of note that compound 20 with
ꢀ(4ꢀfluorophenyl)ꢀ1,2,3,6ꢀtetrahydropyridine moiety connecting to
a
4
the
benzo[de][1,7]naphthyridinꢀ7(8H)ꢀone core through a bisꢀacetamido linker showed
modest PARP1 activity with an IC50 value of 1.1 ꢁM. These results indicated that a
lipophilic terminal group (e.g. 4ꢀ(4ꢀfluorophenyl)ꢀ1,2,3,6ꢀtetrahydropyridine)
connected to the core template through a bisꢀacetamido linker was favorable for
PARP1 interaction.
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Table 1. PARP1 enzymatic assays of compounds 10ꢀ13 and 15ꢀ20.
Compound
X
IC50 (ꢁM)
8
.80
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
9
.38
.35
11
12
13
9
>10
1
1
5 (n=1, Ar=4ꢀMeOꢀPh)
6 (n=2, Ar=4ꢀMeOꢀPh)
0.30
8.72
8.89
8.96
8.52
17 (n=3, Ar=4ꢀMeOꢀPh)
8 (n=2, Ar=2ꢀFꢀPh)
1
1
9 (n=2, Ar=pyrimidinꢀ2ꢀyl)
20 (Ar=4ꢀFꢀPh)
1.10
0.21
ꢀ
ꢀ
Rꢀ22
Sꢀ22
0.019
1
4
ACS Paragon Plus Environment

Page 15 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
>
10
ꢀ
Rꢀ24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
.49
ꢀ
ꢀ
Sꢀ24
0.004
3
(AZD2281)
b
(
0.005 )
a
IC50s were calculated by Logit method from the results of at least three independent tests with six
b
concentrations each (standard deviations were within 25% of the mean values); Data from ref 25.
To further explore the impact of the linker bearing a stereogenic center on the
PARP1 activity, the second acetamido moiety in the linker of compound 20 was
replaced with Rꢀ and Sꢀalanine moieties to afford compounds Rꢀ22 and Sꢀ22 (Table 1).
Both compounds displayed significantly enhanced PARP1 activity, especially Sꢀ22
showing an IC50 value of 19 nM and being 10ꢀfold more potent than Rꢀ22. However,
replacing the second acetamido moiety in the linker of compound 20 with Rꢀ and
Sꢀphenylalanine moieties led to compounds Rꢀ24 and Sꢀ24, possessing significantly
different PARP1 inhibitory activity. Again, Sꢀisomer (Sꢀ24) showed higher potency
with an IC value of 490 nM, which was 20ꢀfold more potent than that of Rꢀ24 (>10
50
ꢂM). These results indicated that both the steric effect and stereochemistry of the
linker played roles in the ligandꢀPARP1 interactions. Relatively less bulky linker with
Sꢀconfiguration was favored.
Our second series of PARP1 inhibitors were designed by borrowing the core
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
structure (phthalazinꢀ1(2H)ꢀone) from clinical compound 3 or its bioisosteres as the
terminal appendage to mimic the arylꢀtetrahydropyridine moiety in Sꢀ22. Meanwhile,
a bulky linker ꢀ methylbenzoyl was employed for readily synthetic manipulation. As
shown in Table 2, compounds 26ꢀ29 bearing various terminal functional groups with
the same linker displayed high PARP1 potency with IC values ranging between
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
0
.31—210 nM. Compound 26 bearing the core structure ꢀ phthalazinꢀ1(2H)ꢀone of
clinical compound 3 displayed remarkably high PARP1 inhibitory activity. It has an
IC50 value of 0.31 nM and is 13ꢀfold more potent than compound 3 (4.0 nM).
Saturation of the aryl moiety in the phthalazinꢀ1(2H)ꢀone framework yielded
compound 27 retaining high affinity of 1.8 nM. It is 6ꢀfold less potent than compound
2
6, but still 2ꢀfold more potent than compound 3. Replacing the aryl in the
phthalazinꢀ1(2H)ꢀone component with one or two methyl groups led to compounds 28
and 29 showing much reduced enzymatic activity. To increase the flexibility of the
linker in compound 26, a few aminoacyl moieties were inserted to the linkage leading
to compounds 31ꢀ33. All the three compounds retained high PARP1 inhibitory
activity with IC50 values between 1.9—45 nM. The steric effect in the linker seems
beneficial to the enzymatic activity. Compound 33 with the most bulky linker showed
the highest PARP1 activity among the three compounds with an IC50 value of 1.9 nM,
compatible to that of 27 and 2ꢀfold more potent than that of 3.
1
6
ACS Paragon Plus Environment

Page 17 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Table 2. PARP1 enzymatic assays of compounds 26ꢀ29 and 31ꢀ33.
Compound
Ar
IC50
Compound
IC50 (ꢁM)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
ꢁM)
0
.00031
0
.033
.045
26
0.0018
3
3
1
2
27
0
0
.21
2
8
9
0.0019
0.036
3
3
2
a
IC50s were calculated by Logit method from the results of at least three independent tests with six
concentrations each (standard deviations were within 25% of the mean values).
Cell Proliferation Inhibition of Potent Compounds and Identification of
Firstꢀgeneration Lead Compound 26. From the results above, compounds 26, 27,
29 and 31ꢀ33 were identified as potent PARP1 inhibitors with IC values lower than
50
5
0 nM. Therefore, they were further evaluated for their inhibition on the cell
proliferation. As shown in Table 3, the selected compounds together with reference
compound 3 were tested in a matched pair of previously wellꢀcharacterized Chinese
51
hamster lung fibroblast V79 (wild type) and VꢀC8 (BRCA2 deficient) cells.
Compared to its high enzymatic potency, compound 26 showed much less cellular
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
activity. It has a CC50 value of 96.0 nM against the BRCA2ꢀdeficient VꢀC8 cells which
was 2ꢀfold more potent than that of reference 3. Similar cellular potency was observed
for compound 27 (CC50, 114 nM). Notably, the rest compounds (29, 31ꢀ33) did not
show appreciable inhibitory activity in the same assay in spite of their potent
enzymatic activity. As expected, all the new synthetic compounds together with 3
showed no proliferation inhibition to the wildꢀtype V79 cells confirming that the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
11,12
tested PARP1 inhibitors can selectively kill BRCAꢀdeficient cells.
Since our compound 26 displayed much higher potency in the PARP1 enzyme but
only moderate cytotoxicity in the cell (300ꢀfold discrepancy), a nonꢀoptimal profile
similar to that of 3, therefore, a followꢀup focused SAR on lead 26 was conducted
with the aim to enhance the cellular activity.
a
Table 3. Enzyme and cellular assays of potent compounds.
Compound
IC (enzyme, nM)
CC (wildꢀtype, V79)
CC (BRCA2,VꢀC8)
5
0
50
50
2
2
2
3
3
3
6
7
9
1
2
3
0.31
1.80
36.0
32.7
44.7
1.92
4.10
>10 ꢁM
>10ꢁM
>10 ꢁM
>10 ꢁM
>10 ꢁM
>10 ꢁM
>10 ꢁM
96.0 nM
114 nM
>1000 nM
>1000 nM
>1000 nM
>1000 nM
201 nM
3
a
Cytotoxic effect (CC50) means the concentration required to reduce cell proliferation and growth by
0%. Values were calculated by Logit method from the results of at least three independent tests with
5
six concentrations each (standard deviations were within 25% of the mean values).
Further SAR on Lead 26 and Discovery of Secondꢀgeneration PARP1 Inhibitors.
1
8
ACS Paragon Plus Environment

Page 19 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Although we proposed that the benzo[de][1,7]naphthyridinꢀ7(8H)ꢀone network in our
lead 26 should interact with the catalytic domain of PARP1 enzyme, it was not clear
whether the terminal group phthalazinꢀ1(2H)ꢀone competitively entered the same
1
5,25
domain
as
that
of
compound
3.
To
validate
the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
benzo[de][1,7]naphthyridinꢀ7(8H)ꢀone core as an effective scaffold of PARP1
inhibitors, a focused library was designed based on structural modification on the
benzo[de][1,7]naphthyridinꢀ7(8H)ꢀone component of 26, which may be also linked to
the nonꢀoptimal cellular activity.
a
Table 4. Structural Modification on Lead 26.
IC50 (nM)
CC50 (ꢁM)
CC50 (nM)
Compd
X
(
enzyme)
(wildꢀtype, V79) (BRCA2,VꢀC8)
2
6
0.31
>10
96.0
38
156
>10
>10
>10,000
630
3
9
1.39
40
387
>10
>10,000
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
4
1
3.46
>10
4.53
N
a
Values were calculated by Logit method from the results of at least three independent tests with six
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
concentrations each (standard deviations were within 25% of the mean values).
As shown in Table 4, isomerization of the lactam moiety in 26, the proposed
key pharmacophoric group interacting with the PARP1 catalytic domain, led to
compound 40 showing over 1000ꢀfold reduction in the PARP1 enzymatic activity.
Reduction of the lactam fragment to amine 38 also resulted in 500ꢀfold loss of activity.
Moreover, both compounds 40 and 38 had no appreciable proliferation inhibition
against BRCA2ꢀdeficient VꢀC8 cells, indicating that the lactam component in lead 26
is indeed essential for both enzymatic and cellular activity. Interestingly, shifting the
appendage linker in compound 38 from Cꢀring Nꢀatom to Bꢀring Nꢀatom led to
compound 39 showing an IC value of 1.39 nM on the PARP1 enzyme and a CC
50
50
value of 630 nM against the BRCA2ꢀdeficient cells. This compound was only 4ꢀ to
6ꢀfold less potent than lead 26 both in the PARP1 enzyme and the VꢀC8 cells. More
surprisingly, compound 41 bearing a cyclopenta[ij]isoquinolinꢀ7(1H)ꢀone skeleton,
the ketone precursor of lactam moiety of 26, showed equally high potency for both
the PARP1 enzyme and the BRCA2ꢀdeficient cells, with IC and CC values of 3.46
50
50
and 4.53 nM, respectively. Although slightly less potent on the PARP1 enzyme,
compound 41 was over 20ꢀ and 40ꢀfold more potent against the BRCA2ꢀdeficient
VꢀC8 cells than our firstꢀgeneration lead 26 and reference compound 3, respectively.
2
0
ACS Paragon Plus Environment

Page 21 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Table 5. Structural Modification on Lead 41.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
IC (nM)
CC (ꢁM)
CC (nM)
50
50
50
Compd
X
R
F
(
enzyme)
(wildꢀtype, V79) (BRCA2,VꢀC8)
4
4
4
1
2
3
3.46
>10
>10
>10
>10
4.53
F
F
F
86.2
0.56
76.6
>10,000
>10,000
>10,000
4
4
4
6
F
F
F
F
F
H
83.9
4.81
>10
1.49
4.26
>10
>10
>10
>10,000
418
47
4
5
5
8
2
5
0.57
2269
11.23
86.34
145.75
4680
>10,000
>10,000
56
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Values were calculated by Logit method from the results of at least three independent tests with six
concentrations each (standard deviations were within 25% of the mean values).
In view of the readily functional transformation from the ketone moiety in
compound 41, a series of analogues derived from 41 were further designed. As shown
in Table 5, reduction of ketone 41 to alcohol 42 yielded 25ꢀfold loss of PARP1
enzymatic activity (86 nM) and complete loss of cellular potency. Conversion of
alcohol 42 to fluoride 43 led to a 6ꢀfold enhancement on the enzyme but the cellular
potency was abolished. Further conversion of alcohol 42 to amine 44 retained
reasonable enzymatic activity (76.6 nM) but again lost cellular activity. Opening the
Nꢀcontaining ring of the cyclopenta[ij]isoquinolinꢀ7(1H)ꢀone network in compound
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
1 led to structurally simplified analog 46 which retained moderate enzymatic activity
(
83.9 nM). Further reduction of the ketone function in 46 produced cisꢀisomer 47 and
transꢀisomer 48 showing significantly different activity against PARP1 enzyme and
the BRCA2ꢀdeficient VꢀC8 cells as well. The cisꢀisomer 47 showed high PARP1
enzymatic activity with an IC value of 4.81 nM, and moderate activity (418 nM) in
50
the cells. Higher enzymatic potency was observed for the transꢀisomer 48 with an
IC value of 0.57 nM; however, it was much less potent in the cells. In addition, both
50
compounds showed somewhat cytotoxicity against wildꢀtype V79 cells (1.5 vs 4.2
ꢁ
M), indicating their potential nonꢀselective offꢀtarget toxicity. NꢀMethylation of
compound 46 resulted in analog 52 showing a 7ꢀfold improvement in the enzymatic
activity (11.2 nM), and modest cellular activity. Similarly, moderate enzymatic
activity was observed on the Schmidt rearranged products 55 and 56 having IC50
values of 86 and 146 nM, respectively, but they were inactive against the VꢀC8 cells.
On the basis of the results above, compound 41 emerged as our
secondꢀgeneration lead compound with high activity both for PARP1 enzyme and the
2
2
ACS Paragon Plus Environment

Page 23 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
BRCA2ꢀdeficient VꢀC8 cells.
Inhibition of Cell Proliferation of Potent Compounds on BRCA1 Deficient Cells.
In addition to BRCA2ꢀdeficient VꢀC8 cells, both lead compounds 26 and 41 were
further evaluated in human breast cancer MDAꢀMBꢀ436 cells carrying natural BRCA1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25
mutations. Meanwhile, compounds 33, 39, 43, 47 and 48 were also assayed for
comparison. As summarized in Table 6, the BRCA1ꢀdeficient MDAꢀMBꢀ436 cells
were hypersensitive to the PARP inhibition by most of the tested compounds,
therefore, higher potencies were generally observed in this cell line in comparison
with those in BRCA2ꢀdeficient VꢀC8 cells. Our firstꢀgeneration lead 26 had a CC50
value of 23.2 nM against MDAꢀMBꢀ436 cells, 2ꢀfold more potent than that of 3.
Extremely high potency was observed for the secondꢀgeneration lead 41 which
showed a CC50 value lower than 0.26 nM, and was respectively 100ꢀ and 160ꢀfold
more potent than compounds 26 and 3. This compound may be the most potent
PARP1 inhibitors reported in the literature so far against the MDAꢀMBꢀ436 cell line.
It showed much greater potency than any other of the reported clinical candidates in
this cell line. Interestingly, compound 33 was inactive in the BRCA2ꢀdeficient VꢀC8
cells, but showed moderate cytotoxicity (344 nM) against BRCA1ꢀdeficient
MDAꢀMBꢀ436 cells. In comparison to their high enzymatic potency, amine 40 only
displayed moderate activity and fluoride 43 was inactive in both cell lines. Again,
some effects of the stereochemistry were observed on the cisꢀ and transꢀisomers 47
and 48. The former compound showed moderate potency against the BRCA2ꢀdeficient
VꢀC8 cells, whereas the latter showed moderate potency against the BRCA1ꢀdeficient
MDAꢀMBꢀ436 cells.
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Table 6. Cell Proliferation Inhibition of Compounds on BRCA1 and BRCA 2 Deficient Cells.
a
a
IC50 (nM)
(PARP1)
CC50 (BRCA 2, nM)
CC50 (BRCA 1, nM)
Compd
(VꢀC8)
201
(MDAꢀMBꢀ436)
3
(AZD2281)
4.1
43.2
23.3
344
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6
3
0.31
1.92
1.39
3.46
0.56
4.81
0.57
96.0
3
>10,000
630
39
556
41
43
47
48
4.53
<0.26
4,300
4,520
775
>10,000
418
2,270
a
Values were calculated by Logit method from the results of at least three independent tests with
six concentrations each (standard deviations were within 25% of the mean values).
Pharmacokinetic (PK) Profiles of Leads 26 and 41. Compounds 26 and 41 have
molecular weight of less than 500 (468Da and 453Da, respectively), and good
lipophilicity (CLogP: 2.06 and 2.67 respectively). As summarized in Table 7, both
compounds have similar volume of plasma distribution in rats (Vss =4.3 and 4.6 L/kg,
respectively) and similar halfꢀlife. Compound 26 has lower plasma clearance (2.28
L/h/kg), good plasma exposure (AUC0ꢀ∞, 1818 ng.h/mL) and moderate bioavailability
(
21%). While compound 41 showed higher plasma clearance (7.67 L/h/kg), moderate
52
lower plasma exposure (AUC0ꢀ∞, 985 ng.h/mL) and lower bioavailability (~10%).
2
4
ACS Paragon Plus Environment

Page 25 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 7. In vitro pharmacokinetic profile of lead compounds in rats
a
a
iv (10 mg/kg)
po (20 mg/kg)
Compd
26
CL
Vss
T
1/2
C
max
T
max
AUC0ꢀ∞
F
(
(L/h/kg)
(L/kg)
(h)
(ng/mL)
(h)
(ngꢃh/mL)
1818±431
980±182
(%)
21.0
10.0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.28 ± 0.28 4.27 ± 0.66 1.73 ± 0.12 722 ± 203
7.67 ± 2.7 4.56 ± 1.86 1.11 ± 0.28 116 ± 10.0
0.42±0.14
0.44±0.13
41
a
Values are the average of three runs. Vehicle: DMSO, Tween 80, normal saline. CL, clearance; Vss,
volume of distribution; T1/2, halfꢀlife; Cmax, maximum concentration; Tmax, time of maximum
concentration; AUC0ꢀ∞, area under the plasma concentration time curve; F, oral bioavailability.
53
Meanwhile, compounds 26 and 41 were further evaluated for the inhibitory
liability on hERG K+ channels, a major safety concern during the development of
therapeutic drugs. Both compounds did not show significant inhibition at the
concentration of 5.0 ꢁM (17% and 34%, respectively).
Inhibition of Leads 26 and 41 on the H O ꢀTriggered PARylation. Since both leads
2
2
2
6 and 41 showed high potency for the PARP1 enzyme and for the BRCA1ꢀ and
2
ꢀdeficient cells, their intrinsic activity was further evaluated. We first compared their
ability to inhibit cellular PARP1/2 by detecting the synthesis of poly(ADPꢀribose)
(
2 2
PAR) triggered by the DNA damaging agent H O in human ovarian cancer
15,25
BRCAꢀproficient SKOV3 cells.
H O induces DNA singleꢀstrand breaks that
2 2
causes a series of subsequent responses including activation of PARP, which may lead
5
4ꢀ56
to PARylation of various proteins, i.e., the synthesis of PAR.
alone resulted in prominent PAR formation in the SKOV3 cells, and the
PARylation could be reduced by pretreatment with compound 3, 26 or 41. Compound
1 was found significantly more potent to suppress the H O ꢀdriven PAR formation
As shown in Figure 2,
2 2
H O
4
2
2
than compound 3 or 26, with IC50 values of 0.49 nM (for 41) versus 20.9 nM (for 3)
and 66.1 nM (for 26), respectively.
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6
ACS Paragon Plus Environment

Page 27 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
B
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2 2
Figure 2. Leads 26 and 41 reduced the PARylation triggered by the DNAꢀdamaging agent H O .
SKOV3 cells were treated and assayed as described in the Methods section. A. Representative images
of PAR formation inhibition. Nuclei stained with DAPI indicated the cell density; the FITC images
revealed the levels of cellular PAR. B. The concentrationꢀresponse curves quantitatively showed the
inhibition of 3, 26 and 41 on the formation of PAR. The data were expressed as mean ± SD, obtained
from the images in A and from three independent experiments.
Compounds 26 and 41 Caused Cellular Accumulation of DNA Doubleꢀstrand
Breaks and Impaired Cellꢀcycle Progression in BRCA2ꢀdeficient Cells. The
production and repair of DNA singleꢀstrand breaks are common phenomenon in the
daily life, and the doubleꢀstrand breaks would occur and accumulate in the cell if the
repair by PARP1/2 is not timely or the repair pathways are deficient, especially
56
BRCAꢀdeficient. Since γꢀH2A
X
has been regarded as a molecular marker for DNA
57
doubleꢀstrand breaks, we therefore examined whether compounds 26 and 41
increased the levels of cellular γꢀH2A . As shown in Figure 3, treatments with
compound 3, 26 and 41 at various concentrations led to significantly enhanced levels
of γꢀH2A in BRCA2ꢀdeficient VꢀC8 cells, while only slightly altered levels of
X
X
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
γꢀH2AX was observed in the BRCA2ꢀproficient wildꢀtype V79 cells. Moreover,
compound 41 was found more efficacious to enhance the level of γꢀH2AX than
compound 26 and the reference compound 3, which was consistent with their capacity
of PAR inhibition.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Leads 26 and 41 Increased the Levels of Cellular γꢀH2AX. Cells were exposed to the
corresponding compound at 37°C for 24 h, and then were subjected to standard Western blotting to
detect the levels of cellular γꢀH2AX.
Consistently, as shown in Figure 4, exposure of compounds 26 and 41 to the
BRCA2ꢀdeficient VꢀC8 cells significantly impaired the cellꢀcycle progression.
Moreover, our data also indicated that inhibition of cellular PARP by compounds 26
and 41 likely impaired the capacity of the cell to repair DNA singleꢀstrand breaks,
thus triggering a series of subsequent biological responses. These results confirmed
that our compounds caused DNA doubleꢀstrand breaks accumulation and this process
was induced by PARP1 inhibition.
2
8
ACS Paragon Plus Environment

Page 29 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Leads 26 and 41 Changed Cellꢀcycle Progression. Cells were exposed to the indicated
compound at 37°C for 24 h, and then were subjected to routine flow cytometry analyses to detect the
cell cycle progression.
Compounds 26 and 41 Potentiated Cytotoxicity of the DNA Damaging Agent
Temozolomide. To evaluate the potential application of compounds 26 and 41 as
combinations with other antitumor agents, we then explored the impact of these two
25,58ꢀ63
compounds in parallel with 3 on the cytotoxic efficacy of temozolomide.
As
shown in Table 8, at both concentrations of 0.5 and 1.0 ꢁM, all three compounds
significantly enhanced the cytotoxicity of temozolomide in SKOV3 cells that have
proficient repair for DNA doubleꢀstrand breaks. In agreement with the previous data,
compound 41 displayed higher potentiation on the efficacy of temozolomide than
compound 26 and reference compound 3 at the same concentrations.
Table 8. Leads 26 and 41 Potentiated Cytotoxicity of Temozolomide in SKOV3 Cells.
Temozolomide [IC50 (ꢂM) (mean ± SD) (n=3)]
Compound
alone
+ 0.5 ꢁM compound
206 ± 17.2 （9.3ꢀfold*）
98.0 ± 28.6 （19.5ꢀfold）
+1.0 ꢁM compound
2
6
1
1910 ± 6.45
1910 ± 4.04
180 ± 5.59 （10.6ꢀfold）
85.8 ± 18.7 （22.2ꢀfold）
4
2
9
ACS Paragon Plus Environment