Molecular design, synthesis and docking study of benz[b]oxepines and 12-oxobenzo[c]phenanthridinones as topoisomerase 1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.03.058

Benz[b]oxepines 4a-g and 12-oxobenzo[c]phenanthridines 5a-d were designed and synthesized as constrained forms of 3-arylisoquinolines through an intramolecular radical cyclization reaction. Radical cyclization of O-vinyl compounds preferentially led to th

Full text of DOI:10.1016/j.bmcl.2009.03.058

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2444–2447
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Molecular design, synthesis and docking study of benz[b]oxepines
and 12-oxobenzo[c]phenanthridinones as topoisomerase 1 inhibitors
Suh-Hee Lee a, Hue Thi My Van a, Su Hui Yang a, Kyung-Tae Lee b, Youngjoo Kwon c, Won-Jea Cho a,
*
a College of Pharmacy and Research Institute of Drug Development, Chonnam National University, Gwangju 500-757, Republic of Korea
b College of Pharmacy, Kyung-Hee University, Seoul 130-701, Republic of Korea
c College of Pharmacy, Ewha Womans University, Seoul 120-750, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Benz[b]oxepines 4a–g and 12-oxobenzo[c]phenanthridines 5a–d were designed and synthesized as con-
strained forms of 3-arylisoquinolines through an intramolecular radical cyclization reaction. Radical
cyclization of O-vinyl compounds preferentially led to the 7-endo-trig cyclization pathway to the
benz[b]oxepines and 12-oxobenzo[c]phenanthridines through 6-exo-trig path as minor products. Among
the synthesized compounds, benz[b]oxepine derivative 4e exhibited potent in vitro cytotoxicity against
three different tumor cell lines, as well as topoisomerase 1 inhibitory activity. A Surflex–Dock docking
study was performed to clarify the topoisomerase 1 activity of 4e.
Received 14 January 2009
Revised 10 March 2009
Accepted 13 March 2009
Available online 18 March 2009
Keywords:
Synthesis
Docking study
Topoisomerase 1
Inhibitor
Ó 2009 Elsevier Ltd. All rights reserved.
Molecular modeling
Radical cyclization
Cytotoxicity
Since topotecan and irinotecan were launched as anti-neoplas-
tic drugs, topoisomerase 1 (topo 1) has been considered a promis-
ing target in the development of anticancer drugs.1 The
cytotoxicities of these compounds come from their abilities to in-
hibit the religation process of the topo 1 reaction, thereby resulting
in the accumulation of DNA–drug–topo 1 ternary complex.2 Both
drugs are camptothecin (CPT) derivatives that were developed con-
sidering the physicochemical properties of CPT 1. However, they
have critical problems: (i) these drugs must be infused for long
periods for cancer treatment as they reverse the CPT-trapped
cleavage complexes within minutes; (ii) their inactive decomposed
carboxylates are in equilibrium with active lactone forms under
physiological conditions; and (iii) these analogs can cause resis-
tance because the drugs are also substrates for efflux transporters.3
So far, a number of trials to develop novel non-CPT topo 1 inhibi-
tors have been actively pursued.4–7
but also QSAR studies of 3-arylisoquinolines that showed topo 1
inhibition and cytotoxicity.9–12
Recently, we reported the synthesis and biological evaluation of
indenoisoquinolines 3 as constrained forms of 3-arylisoquinolines
2 as shown in Figure 1.9,12 Rigid structures are commonly consid-
ered to have little conformational entropy compared to flexible
structures and can be more efficiently fitted into the active site
of a receptor. In the previous study, we found that some indenoiso-
quinoline analogs show potent topo 1 inhibitory activities and
cytotoxicities.
Our research next focused on synthesizing new six- or seven-
membered heterocyclic rings containing the 3-arylisoquinoline
skeleton. Because benz[b]oxepine 4 and 12-oxobenzo[c]phenan-
thridines 5 have not yet been synthesized, we investigated their
syntheses and biological evaluation.
As depicted in Scheme 1, the benz[b]oxepine 4 and 12-oxo-
benzo[c]phenanthridinone 5 could be constructed through intra-
molecular radical cyclization of compound 6. The endo type (A)
or exo type (B) pathway would yield the benz[b]oxepine 4 as well
as 12-oxobenzo[c]phenanthridinone 5, respectively. O-vinyl com-
pound 6 could be synthesized via the toluamide-benzonitrile
cycloaddition reaction from 7 and 8.
The 3D structure determination of drug–DNA–topo 1 complex
provided valuable information of the drug binding mode in the ac-
tive site and a computer-aided molecular modeling study was uti-
lized to develop novel topo 1 inhibitors.8
We have investigated not only the syntheses of isoquinoline
alkaloids such as protoberberines and benzo[c]phenanthridines,
The previously reported lithiated toluamide-benzonitrile cyclo-
addition method was used to synthesize the 3-arylisoquinolines
9a–c.13 N-Diethyl-o-toluamides 7a–c were treated with n-BuLi to
give the anions, which were then reacted with benzonitrile 8 to af-
* Corresponding author. Tel.: +62 530 2933; fax: +62 530 2911.
E-mail address: wjcho@jnu.ac.kr (W.-J. Cho).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.058

S.-H. Lee et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2444–2447
2445
R'
R
NH
O
O
Indeno[1,2-c]isoquinolinone (3)
R
NH
O
Benz[b]oxepine (4)
OH
N
O
R
O
N
NH
Me
O
O
O
Camptothecin (1)
3-Arylisoquinoline (2)
R
NH
O
12-Oxobenzo[c]phenanthridinone (5)
Figure 1. Structure of CPT (1) and constrained form of 3-arylisoquinoline to benz[b]oxepine (4) and 12-oxobenzo[c]phenanthridinone (5).
O
R'
A
B
R
NH
O
O
B
Me
NHMe
R'
4
A
O
MOMO
NC
R
R
R'
+
NH
Me
O
R'
O
6
8
7
R
NH
O
5
Scheme 1. Retrosynthesis of benz[b]oxepine (4) and 12-oxobenzo[c]phenanthridinone (5).
ford the 3-arylisoquinolines 9a–c in moderate yield. Treatment of
9a–c with alkyl halides such as MeI or PMBCl in the presence of
NaH or K2CO3 provided the corresponding N-alkylated compounds
10a–g in good yield. Regioselective bromination of 10a–g on C-4
was accomplished by treatment with N-bromosuccinimide and
ACCN to yield the desired compounds 11a–g in good yield.14
Deprotection of the methoxymethyl group of 11a–g was achieved
by treatment with 10% HCl in THF to give the phenols 12a–g,
which were then reacted with tetravinyl tin and Cu(OAc)2 in the
presence of O2 to afford the desired O-vinyl compounds 13a–g.15
Radical cyclization reaction of 13a–g with n-Bu3SnH with AIBN
in benzene provided the desired benz[b]oxepine 4a–g and 12-
oxobenzo[c]phenanthridinone 5a–d in 56–86% yield.16,17 The ratio
of products depended on the substitution pattern of the starting
materials 13a–g. The yields of the final products were as follows:
4a:5a (56:0), 4b:5b (86:0), 4c:5c (61:0), 4d:5d (38:27), 4e:5e
(45:31), 4f:5f (40:25), 4g:5g (35:27). Interestingly, PMB-substi-
tuted isoquinolines 13a–c furnished exclusively 7-endo-trig path-
way products, benz[b]oxepines (Scheme 2).
The in vitro cytotoxicity experiments were performed using the
synthesized compounds against three human tumor cell lines com-
prising of A 549 (lung cancer), HL60 (leukemia), and HeLa (cervical
cancer) cells in sulforhodamine B (SRB) assays.18 The topo 1 inhib-
itory activity assay was carried out using the supercoiled DNA
unwinding method.19
3-Arylisoquinoline 9a,b displayed stronger cytotoxicities than
N-alkylated compounds 10a–13f against three tumor cell lines as
shown in Table 1. In general, potent cytotoxic activities were ob-
served for the non substituted 3-arylisoquinolines 9a,b at C2 posi-
tion relative to their N-Me or N-PMB compounds. The next worth
mentioning is the discrepancy between the cytotoxicity of 9a
(0.62 lM), 9b (0.16 lM) and their 0% topo 1 inhibition activity.
This result is not entirely without precedence, and it is noticeable
that the cytotoxicity of these molecules are not related to their
capability to inhibit topo 1, but must be dependent upon different
unknown biological target.
Unexpectedly, 12-oxobenzo[c]phenanthridine analogs 5a–d
exhibited weak cytotoxicity with the exception of dimethoxy-
substituted compound 5d (8.54–13.76 lV). Furthermore, most of
these compounds, except 5b (30.4%) and 5d (5.3%), had no topo 1
inhibitory activity. These results could not be explained by their
low aqueous solubility or poor membrane permeability. However,
dramatic enhancement of topo 1 inhibitory activity was observed
in benz[b]oxepines 4d (98.2%) and 4e (94.5%). Moreover,
benz[b]oxepine compounds showed relatively potent cytotoxicities
and topo 1 inhibitory activities compared to 12-oxobenzo[c]phe-
nanthridines. Among the benz[b]oxepines, the N-PMB-substituted
compounds 4e–g exhibited greater cytotoxicities than the N-Me-
substituted compounds 4a–d. Interestingly, compound 4e, which
contains the PMB group at C2 nitrogen, showed more potent cyto-
toxicities than the N-methyl substituted compound 4a.
The topo 1 inhibitory activity of the compounds is depicted in
Figure 2. A quantitative assay was carried out to assess the relative
topo 1 inhibitory potency of the compounds. Compound 11d had
half the potency (40.7%) of the reference CPT (86.2%). However,
compounds 4d and 4e exhibited much more potent inhibitory

2446
S.-H. Lee et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2444–2447
MOMO
R1
MOMO
N
MeI/NaH
or PMBCl/K2CO3
R1
R2
Me
NEt2
R1
R2
MOMO
n-BuLi
THF
+
NH
R2
R3
NC
O
7
O
O
8
9
10
O
MOMO
HO
Br
O
Br
Br
R1
R1
R2
R1
R2
Sn( )4, Cu(OAc)2
CH3CN, O2
NBS, ACCN
CCl4
10% HCl
THF
N
N
N
R2
R3
R3
R3
O
O
13
11
12
O
N
Me
O
N
R1
R2
R1
R2
n-Bu3SnH, AIBN
benzene
+
R3
R3
O
O
5
4
Scheme 2. The synthesis of benz[b]oxepine and 12-oxobenzo[c]phenanthridine analogs. (a) R1 = H, R2 = H, R3 = Me; (b) R1 = Me, R2 = H, R3 = Me; (c) R1 = H, R2 = Me, R3 = Me;
(d) R1 = OMe, R2 = OMe, R3 = Me; (e) R1 = H, R2 = H, R3 = PMB; (f) R1 = Me, R2 = H, R3 = PMB; (g) R1 = H, R2 = Me, R3 = PMB.
Table 1
Surflex–Dock docks ligands automatically into a receptor’s li-
Cytotoxicity (IC50, lM) and topo 1 inhibitory activity (% inhibition) of the compounds
gand binding site using a protomol-based method and an empiri-
cally derived scoring function. The protomol is a unique and
important factor of the docking algorithm and is a computational
representation of assumed ligands that interact with the binding
site. Surflex–Dock’s scoring function contains hydrophobic, polar,
repulsive, entropic, and solvation terms. In addition to the auto-
mated docking process, the function in Surflex–Dock has recently
been improved by incorporating a base portion matching algo-
rithm that allows a fragment of the ligand to be prepositioned as
it docks in the binding site. The fragment is allowed to convey from
its original position to a certain point during pose optimization.
This is noteworthy when the position of the base portion is not
completely set. Ligand docking with the base fragment-matching
characteristic is anticipated to yield docking and scoring of ligands
constrained to match an exact binding motif.
The structure of the inhibitor 4e was drawn into the Sybyl pack-
age with standard bond lengths and angles and minimized using
the conjugate gradient method. The Gasteiger–Huckel charge, with
a distance-dependent dielectric function, was applied for the min-
imization process. We chose the 1SC7 (PDB code) structure from
the Protein Data Bank and the structure was polished as follows.
The phosphoester bond of G12 in 1SC7 was rebuilt and the SH of
G11 on the scissile strand was altered to OH. After running Sur-
flex–Dock, the scores of 10 docked conformers were ranked in a
molecular spread sheet. We selected the best total score conformer
(5.89) and speculated regarding the detailed binding patterns in
the cavity. The resulting docking model revealed a binding mode
similar to the indenoisoquinoline model. In our model, the oxepine
ring intercalated between the À1 and +1 bases, parallel to the
plane of the base pairs as depicted in Figure 3. The oxygen of the
oxepine ring had an H-bond with the amino group of A 113, which
is considered an essential nucleotide that interacts with the ligand
in the DNA–top 1 active site. In our model, the isoquinoline ring
and seven-membered oxepine ring worked as DNA intercalator
as well as blocker of the religation step of the phosphoester. In this
docking study, we observed that the seven-membered oxepine ring
No
Compd
R1
R2
R3
A549
HL60
HELA
Topo 1
a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
4a
4b
4c
4d
4e
4f
4g
5a
5b
5c
5d
9a
H
Me
H
OMe
H
Me
H
H
Me
H
OMe
H
Me
H
Me
H
OMe
H
Me
H
Me
CPT
H
H
Me
OMe
H
H
Me
H
Me
Me
Me
Me
PMB
PMB
PMB
Me
Me
Me
Me
H
H
Me
Me
Me
Me
66.76
61.32
30.33
72.98
11.04
>100
4.56
65.52
46.00
66.89
13.76
23.6
>100
>100
99.76
82.00
>100
71.25
53.01
>100
67.76
0.058
37.59
33.57
14.12
34.82
4.49
9.26
4.40
30.05
29.13
ntb
19.66
41.57
8.76
38.06
4.78
10.26
4.54
40.08
28.32
21.88
9.02
—
—
11.9
98.2
94.5
—
—
—
30.4
—
5.3
—
—
—
—
—
40.7
—
—
—
—
86.2
H
Me
OMe
H
H
H
8.54
0.62
0.16
0.62
0.17
9b
10a
10b
11c
11d
12e
12f
13e
13f
59.48
14.28
17.17
37.79
31.59
24.45
59.47
29.17
0.089
64.58
28.58
17.81
99.28
36.78
20.37
36.47
14.77
0.072
H
Me
OMe
H
H
H
PMB
PMB
PMB
PMB
H
a
No inhibition activity.
Not tested.
b
activity than CPT. In many cases, the topo 1 inhibitory activity did
not correlate well with the cytotoxicity. However, compound 4e
showed potent cytotoxicity and potent topo 1 inhibitory activity.
Given the X-ray crystallographic structure of topo 1-DNA com-
plex with indenoisoquinoline, docking studies of indenoisoquino-
lines in the active site have been more convincing than those of
molecules to non-clarified binding sites.8 To understand the bind-
ing mode of action of the most potent topo 1 inhibitor 4e, we per-
formed a docking study using Surflex–Dock in Sybyl version 8.02
(Tripos Associates) operating under Red Hat Linux 4.0 on an IBM
computer (Intel Pentium 4, 2.8 GHz CPU, 1 GB memory).20

S.-H. Lee et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2444–2447
2447
Figure 2. Topo 1 inhibitory activities of compounds. Compounds were examined at a final concentration of 100
lM. Lane D: pBR322 only, Lane T: pBR322 + Topo 1, Lane C:
pBR322 + Topo 1 + CPT, Lane 1–10 or 1–14: pBR322 + Topo 1 + compounds.
Figure 3. Wall-eyed viewing docked model of compound 4e.
5. Morrell, A.; Antony, S.; Kohlhagen, G.; Pommier, Y.; Cushman, M. J. Med. Chem.
was positioned as a DNA intercalator, and the PMB group was
located in the cavity between DNA and topo 1.
2006, 49, 7740.
6. Nagarajan, M.; Morrell, A.; Ioanoviciu, A.; Antony, S.; Kohlhagen, G.; Agama, K.;
Hollingshead, M.; Pommier, Y.; Cushman, M. J. Med. Chem. 2006, 49, 6283.
7. Xiao, X. S.; Antony, S.; Pommier, Y.; Cushman, M. J. Med. Chem. 2005, 48, 3231.
8. Staker, B. L.; Feese, M. D.; Cushman, M.; Pommier, Y.; Zembower, D.; Stewart,
L.; Burgin, A. B. J. Med. Chem. 2005, 48, 2336.
9. Cho, W. J.; Le, Q. M.; My Van, H. T.; Youl Lee, K.; Kang, B. Y.; Lee, E. S.; Lee, S. K.;
Kwon, Y. Bioorg. Med. Chem. Lett. 2007, 17, 3531.
10. Le Thanh, N.; Van, H. T.; Lee, S. H.; Choi, H. J.; Lee, K. Y.; Kang, B. Y.; Cho, W. J.
Arch. Pharm. Res. 2008, 31, 6.
11. Le, T. N.; Gang, S. G.; Cho, W. J. J. Org. Chem. 2004, 69, 2768.
12. Van, H. T.; Le, Q. M.; Lee, K. Y.; Lee, E. S.; Kwon, Y.; Kim, T. S.; Le, T. N.; Lee, S. H.;
Cho, W. J. Bioorg. Med. Chem. Lett. 2007, 17, 5763.
13. Le, T. N.; Cho, W. J. Chem. Pharm. Bull. 2008, 56, 1026.
14. Cho, W. J.; Park, M. J.; Imanishi, T.; Chung, B. H. Chem. Pharm. Bull. 1999, 47, 900.
15. Blouin, M.; Frenette, R. J. Org. Chem. 2001, 66, 9043.
16. Padwa, A.; Rashatasakhon, P.; Ozdemir, A. D.; Willis, J. J. Org. Chem. 2005, 70, 519.
17. All synthesized compounds were fully characterized by spectroscopy. Selected data
for some compounds: 4a; mp: 153.4–157 0 °C, IR (cmÀ1): 1639 (C@O), 1H NMR
(300 MHz, CDCl3) d 8.57 (d, J = 7.8 Hz, 1H), 7.72–7.70 (m, 2H), 7.55–7.50 (m,
1H), 7.45–7.40 (m, 1H), 7.36–7.28 (m, 2H), 7.24 (dd, J = 1.1, J = 8.0 Hz, 1H),
4.58–4.53 (m, 2H), 3.55 (s, 1H), 3.15–3.08 (m, 1H), 2.53–2.41 (m, 1H). EI-MS m/
z (%) 277 (M+, 100). Compound 4b; mp: 169.7–173.9 °C, IR (cm-1): 1639 (C@O),
1H NMR (300 MHz, CDCl3) d 8.45 (d, J = 8.2 Hz, 1H), 7.49 (s, 1H), 7.42 (t,
J = 7.5 Hz, 1H), 7.36–7.28 (m, 3H), 7.23 (d, J = 8.1 Hz, 1H), 4.57–4.53 (m, 2H),
3.54 (s, 3H), 3.14–3.08 (m, 1H), 2.52 (s, 3H), 2.48–2.39 (m, 1H). EI-MS m/z (%)
291 (M+, 84). Compound 5a; mp: 173.1–176.9 0C, IR (cmÀ1): 1648 (C@O), 1H
NMR (300 MHz, CDCl3) d 8.51 (d, J = 7.9 Hz, 1H), 7.71 (t, J = 8.3 Hz, 1H), 7.59 (d,
J = 8.2 Hz, 1H), 7.52 (t, J = 7.6 Hz, 2H), 7.30 (t, J = 7.5 Hz, 1H), 7.11–7.06 (m, 2H),
5.80 (dd, J = 6.6, J = 13.3 Hz, 1H), 3.81 (s, 3H), 1.33 (d, J = 6.7 Hz, 3H). EI-MS m/z
(%) 277 (M+, 100). Compound 5b; mp: 146.2–150.7 °C, IR (cmÀ1): 1646 (C@O),
1H NMR (300 MHz, CDCl3) d 8.39 (d, J = 8.2 Hz, 1H), 7.59 (d, J = 8.3 Hz, 1H), 7.32
(t, J = 6.9 Hz, 2H), 7.10–7.05 (m, 3H), 5.78 (dd, J = 6.6, J = 13.1 Hz, 1H), 3.79 (s,
3H), 2.52 (s, 3H), 1.33 (d, J = 6.7 Hz, 3H). EI-MS m/z (%) 291 (M+, 88).
18. Rubinstein, L. V.; Shoemaker, R. H.; Paull, K. D.; Simon, R. M.; Tosini, S.; Skehan,
P.; Scudiero, D. A.; Monks, A.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82, 1113.
19. Zhao, L. X.; Moon, Y. S.; Basnet, A.; Kim, E. K.; Jahng, Y.; Park, J. G.; Jeong, T. C.;
Cho, W. J.; Choi, S. U.; Lee, C. O.; Lee, S. Y.; Lee, C. S.; Lee, E. S. Bioorg. Med. Chem.
Lett. 2004, 14, 1333.
In conclusion, we synthesized various benz[b]oxepine and 12-
oxobenzo[c]phenanthridines as analogs of constrained 3-aryliso-
quinolines structures. An intramolecular radical cycloaddition
reaction was employed to efficiently generate benz[b]oxepine
and 12-oxobenzo[c]phenanthridines. Although 3-arylisoquinoline
synthetic intermediates did not exhibit interesting biological activ-
ities, the oxepine compounds, that is, the constrained structures of
3-arylisoquinolines, exhibited more potent topo 1 inhibitory activ-
ity than CPT. Among the synthesized compounds, 4e had potent
topo 1 inhibitory activity as well as cytotoxicity against three dif-
ferent tumor cell lines. To further explain the topo 1 inhibitory
activity of 4e, molecular docking studies were carried out with
the Surflex–Dock program to give a reasonable binding mode of
the compound in the binding site of DNA and topo 1. For further
study of constrained structures of 3-arylisoquinolines, we are cur-
rently investigating diverse structural modifications in the synthe-
sis of 3-arylisoquinolines, and the results and structure–activity
relationships will be reported in due course.
Acknowledgment
This work was supported by a Korea Research Foundation Grant
(KRF-2007-521-E00190).
References and notes
1. Liew, S. T.; Yang, L. X. Curr. Pharm. Des. 2008, 14, 1078.
2. Creemers, G. J.; Lund, B.; Verweij, J. Cancer Treat. Rev. 1994, 20, 73.
3. Pommier, Y. Nat. Rev. Cancer 2006, 6, 789.
4. Meng, L. H.; Liao, Z. Y.; Pommier, Y. Curr. Top. Med. Chem. 2003, 3, 305.
20. Jain, A. N. J. Comput. Aided Mol. Des. 2007, 21, 281.