Structural modification of 3-arylisoquinolines to isoindolo[2,1-b]isoquinolinones for the development of novel topoisomerase 1 inhibitors with molecular docking study

Article abstract of DOI:10.1016/j.bmcl.2009.03.042

Isoindolo[2,1-b]isoquinolinones 9a-i were designed and synthesized as constrained forms of 3-arylisoquinolines through an intramolecular cyclization reaction. Among the synthesized compounds, 9d exhibited potent topoisomerase 1 inhibitory activity with cy

Full text of DOI:10.1016/j.bmcl.2009.03.042

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2551–2554
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structural modification of 3-arylisoquinolines to isoindolo[2,1-b]isoquinoli-
nones for the development of novel topoisomerase 1 inhibitors with
molecular docking study
*
Hue Thi My Van, Won-Jea Cho
College of Pharmacy and Research Institute of Drug Development, Chonnam National University, Gwangju 500-757, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Isoindolo[2,1-b]isoquinolinones 9a–i were designed and synthesized as constrained forms of 3-aryliso-
quinolines through an intramolecular cyclization reaction. Among the synthesized compounds, 9d exhib-
ited potent topoisomerase 1 inhibitory activity with cytotoxicities against three different tumor cell lines.
A Surflex-dock docking study was performed to clarify the topoisomerase 1 inhibitory activity of 9d.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 29 January 2009
Revised 4 March 2009
Accepted 7 March 2009
Available online 16 March 2009
Keywords:
Isoindolo[2,1-b]isoquinoline
Synthesis
Topoisomerase 1 inhibitor
Docking study
Cytotoxicity
Camptothecin (CPT) 1 was the representative compound identi-
fied as a topoisomerase 1 (top 1) inhibitor.1,2 Topotecan and irino-
tecan have been used clinically as CPT analogues3,4 and these two
drugs have a hydrolysis susceptible lactone ring incorporated
within their structures. Ring-cleavage of the CPT lactone moiety
yields an inactive structure that has high affinity for human serum
albumin. Moreover, the fact that these drugs are substrates for ef-
flux transporters associated with resistance has prompted further
development of novel top 1 inhibitors.5–7
The binding mode of a drug to its receptor site is governed by
subtle electronic or stereo factors, and these two functions play a
critical role in the bioactive conformation of the drug molecule.8
This conformation is the one that a drug molecule adopts when
it binds to the target, and this interaction is greatly influenced by
entropic factors that are related to the structural flexibility of the
molecule.
In view of these considerations, we designed the constrained
molecules to mimic the bioactive conformation as the basis for
designing potent top 1 inhibitors.9,10 As evident from its structure,
the free rotation of the 3-aryl ring could conceivably be con-
strained to indeno[1,2-c]isoquinolinone 3 via route A ring forma-
tion of 3-arylisoquinoline 2. The synthesis, top 1 activity, and
cytotoxicity of 3 were already reported.9 We then focused on iso-
indolo[2,1-b]isoquinolin-5-one 4, which could be synthesized
through route B as shown in Figure 1. Because isoindolo[2,1-b]iso-
quinolin-5-ones have not yet been studied, we investigated their
R'
R
a
NH
A
R'
O
R
Indeno[1,2-
c
]isoquinolinone (3)
NH
b
O
R'
D
B
3-Arylisoquinoline (2)
A
R
N
O
7H-Isoindolo[2,1-
b]isoquinolin-5-one (4)
N
HO
O
O
N
N
O
O
Camptothecin (1)
* Corresponding author. Tel.: +82 62 530 2933; fax: +82 62 530 2911.
E-mail address: wjcho@chonnam.ac.kr (W.-J. Cho).
Figure 1. Structure of CPT
isoindolo[2,1-b]isoquinolin-5-one 4.
1 and constrained form of 3-arylisoquinoline to
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.042

2552
H. T. M. Van, W.-J. Cho / Bioorg. Med. Chem. Lett. 19 (2009) 2551–2554
synthesis and evaluated them biologically. Moreover, the struc-
tural similarity between isoindolo[2,1-b]isoquinolin-5-one and
CPT prompted us to investigate these compounds.
DMSO. The DNA–top 1 activity was determined by assessing the
relaxation of supercoiled DNA pBR322. A mixture of 200 ng of plas-
mid pBR322 DNA and 2 units of calf thymus DNA–top 1 (Fermen-
tas, USA) was incubated without and with the prepared
compounds at 37 °C for 30 min in relaxation buffer [35 mM Tris–
HCl (pH 8.0), 72 mM KCl, 5 mM MgCl2, 5 mM dithiothreitol,
2 mM spermidine, 0.01% bovine serum albumin]. The reaction (fi-
Since the 3D structure determination of drug-DNA–top 1 com-
plex has been reported, computer-aided molecular modeling stud-
ies afford valuable information of drug binding mode in the active
site of DNA–top 1.11 We investigated not only the syntheses of iso-
quinoline alkaloids such as protoberberines and benzo[c]phenan-
thridines, but also conducted molecular modeling studies of
indeno[1,2-c]isoquinoline analogs that showed top 1 inhibition
and cytotoxicity.
As depicted in Scheme 1, the C ring of isoindolo[2,1-b]isoquin-
olin-5-one could be constructed by an intramolecular SN2 reaction.
The mesyl compound could be synthesized via toluamide–benzo-
nitrile cycloaddition reaction.
nal volume of 20 lL) was terminated by adding 5 lL of stop solu-
tion containing 10% SDS, 0.2% bromophenol blue, 0.2% xylene
cyanol and 30% glycerol. DNA samples were then electrophoresed
on 1% agarose gel at 15 V for 6 h with TAE running buffer. Gels
were stained for 30 min in an aqueous solution of ethidium bro-
mide (0.5 lg/mL). DNA bands were visualized by transillumination
with UV light and were quantitated using AlphaImagerTM (Alpha
Innotech Corporation).
The previously reported lithiated toluamide–benzonitrile cyclo-
addition method was used to synthesize 3-arylisoquinolines 7a–
i.12 N-Methyl-o-toluamides 5a–c or N,N-diethyl-o-toluamides 5d–
f were treated with n-BuLi to give the anions, which were then re-
acted with benzonitrile 6a–c to afford the corresponding 3-aryliso-
quinolines 7a–i in moderate yields. Treatment of 7a–c with DDQ
provided the corresponding hydroxymethyl compounds 8a–c.
Deprotection of MOM-substituted compounds 7d–i was conducted
with 10% HCl to give the desired products. Hydroxymethyl com-
pounds 8a–i were reacted with MsCl/Et3N in methylene chloride
to provide the desired isoindolo[2,1-b]isoquinolines 9a–i in mod-
erate yields as shown in Scheme 2.13
The in vitro cytotoxicity experiments were performed with the
synthesized compounds against three human tumor cell lines,
including A 549 (lung), HCT15 (colon), and OV-3 (ovarian), using
sulforhodamine B (SRB) assays.14 The top 1 inhibitory activity as-
says were conducted using the supercoiled DNA unwinding meth-
od.15 Stock solutions of the compounds (20 mM) were prepared in
From the biological data presented in Table 1, general features
are clear from the comparison of 3-arylisoquinolines. PMB-pro-
tected 3-arylisoquinoline 7a–c displayed stronger cytotoxicites
against the three tumor cell lines than MOM-protected com-
pounds 7d–i. The hydroxyl methyl-substituted compounds 8a–i
exhibited stronger cytotoxicity than the protected compounds
7a–i (0.21–22.10
erted potent cytotoxicities (3.70–49.09
l
g/mL). Isoindolo[2,1-b]isoquinolin-5-ones ex-
g/mL). Unexpectedly,
l
some isoindoloisoquinolines 9e–i exhibited poor top 1 inhibitory
activity as depicted in Figure 2. The methyl substituted com-
pounds 9b and 9c on A ring showed less top 1 activity than
unsubstituted ones and this could be explained by the steric
interaction between methyl group with side chain of Thr 718.
The semi-quantitative assay was conducted to show the relative
top 1 potency of the compounds. Compounds 9a and 9d had sim-
ilar potency (++++) compared to the reference CPT (++++). How-
ever, compounds 9b–c exhibited less potent inhibitory activity
than CPT (Fig. 3).
R'
R'
Me
NEt2
MOMO
R
+
R'
R
R
N
NH
NC
OMs
O
O
O
Scheme 1. Retrosynthesis of isoindolo[2,1-b]isoquinolin-5-one 4.
R4
R3
R1
R2
Me
R3
R2
R1
R2
R1O
NC
n
-BuLi
R3
NH
THF
OR5
O
O
7a-i
6a: R1=PMB, R2=H, R3=H
6b: R1=MOM, R2=OMe, R3=H
6c: R1=MOM, R2=R3=OMe
5a: R1=H, R2=H, R3=NHMe
5b: R1=Me, R2=H, R3=NHMe
5c: R1=H, R2=Me, R3=NHMe
5d: R1=H, R2=H, R3=NEt2
5e: R1=Me, R2=H, R3=NEt2
5f: R1=H, R2=Me, R3=NEt2
R4
R4
R4
R3
R3
R3
R1
R1
R2
R1
R2
MsCl, Et3N
CH2Cl2
DDQ
or 10% HCl
NH
NH
N
R2
OH
OMs
O
O
O
9a-i
8a-i
Scheme 2. The synthesis of isoindolo[2,1-b]isoquinoline analogs.

H. T. M. Van, W.-J. Cho / Bioorg. Med. Chem. Lett. 19 (2009) 2551–2554
2553
Table 1
Chemical yield, cytotoxicity (lg/mL and top 1 inhibitory activity of the compounds
No.
Compd
R1
R2
R3
R4
R5
Yield
A549
HCT15
OV-3
Top 1a
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
7a
7b
7c
7d
7e
7f
7g
7h
7i
8a
8b
8c
8e
8e
8f
8g
8h
8i
9a
9b
9c
9d
9e
9f
9g
9h
9i
H
Me
H
H
H
Me
Me
H
H
H
H
H
Me
H
H
H
H
Me
Me
H
H
Me
H
H
H
H
H
H
OMe
OMe
OMe
OMe
OMe
OMe
H
H
H
H
H
OMe
H
OMe
H
OMe
H
H
H
H
OMe
H
OMe
H
OMe
H
H
H
H
PMB
PMB
PMB
MOM
MOM
MOM
MOM
MOM
MOM
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
44
38
52
41
58
43
77
42
47
89
47
59
45
23
30
12
50
11
51
38
40
58
35
32
52
49
36
26.56
7.44
3.33
32.98
16.04
80.22
43.65
66.52
25.02
6.89
17.39
13.27
4.15
21.84
24.70
29.26
34.55
40.05
31.13
3.88
3.34
23.11
0.21
19.68
4.28
7.17
24.66
2.37
4.76
ntb
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
++++
++
++
++++
À
17.06
12.68
31.26
34.54
23.08
28.22
21.98
8.22
0.22
0.37
12.58
14.59
7.80
Me
H
H
H
H
3.66
12.78
3.65
OMe
OMe
OMe
OMe
OMe
OMe
H
H
7.32
Me
Me
H
H
H
Me
H
H
19.43
22.10
10.11
15.15
3.70
15.94
41.44
9.25
36.75
5.95
49.09
45.12
24.13
0.072
H
Me
Me
H
H
Me
H
H
H
H
Me
Me
7.39
9.69
0.67
22.45
43.93
46.35
35.27
7.29
9.12
8.78
34.98
34.33
6.56
14.84
44.12
48.92
12.01
12.69
5.52
36.15
15.34
7.23
H
H
OMe
OMe
OMe
OMe
OMe
OMe
H
OMe
H
OMe
H
Me
Me
H
+
À
+
H
OMe
—
+
CPT
0.089
0.035
++++
a
Activity is expressed semi-quantitatively as follows: À, very weak activity; + weak activity; ++++ similar activity to CPT.
b
nt means not tested.
Figure 2. Top 1 inhibitory activities of the compounds. The DNA–top 1 activity was determined by assessing the relaxation of supercoiled DNA pBR322. The upper band is
relaxed form and the lower is supercoiled DNA. Lane 1, 12: pBR322 only, Lane 2: pBR322 + Top 1, Lane 3: pBR322 + Top 1 + CPT (100 g/mL), Lane 4: pBR322 + Top 1 + CPT
(100 g/mL), Lane 5–11, compound (100 g/mL) + pBR322 + Top 1, 5 (9a), 6 (9b), 7 (9c), 8 (9d), 9 (9f), 10 (9e), 11 (9g).
l
l
l
Figure 3. Wall-eyed viewing docked model of compound 9d.
In many cases, the top 1 inhibitory activity does not correlate
well with cytotoxicity and these results can not be explained by
low aqueous solubility or poor membrane permeability.
To understand the binding mode of action of the most potent
top 1 inhibitor 9d, we performed a docking study using Surflex-
Dock in Sybyl version 8.05 by Tripos Associates, operating under
Red Hat Linux 4.0 with an IBM computer (Intel Pentium 4,
2.8 GHz CPU, 1 GB memory).
Surflex-Dock docks ligands automatically into a receptor’s li-
gand binding site using a protomol-based method and an empiri-
cally derived scoring function.16 The protomol is a unique and
important element of the docking algorithm and means a compu-
tational representation of putative ligands that interact with the
binding site. Surflex-Dock’s scoring function contains hydrophobic,
polar, repulsive, entropic, and solvation terms. In addition to the
automated docking procedure, the function of Surflex-Dock has re-

2554
H. T. M. Van, W.-J. Cho / Bioorg. Med. Chem. Lett. 19 (2009) 2551–2554
activity with potent cytotoxicities. In particular, 9a and 9d had po-
tent top 1 activity as well as potent cytotoxicity against three dif-
ferent tumor cell lines. For explanation of the top 1 activity of 9d,
molecular docking studies were carried out with the Surflex–Dock
program to give the reasonable binding mode of the compound
into the binding sites of DNA and top 1. In further studies of the
other constrained structures of 3-arylisoquinolines, diverse struc-
tural modifications are currently being investigated through syn-
thesis with computer modeling and the results will be given in
due course.
Acknowledgment
This work was supported by Korea Research Foundation grant
(KRF-2006-312-E00171).
Figure 4. Superimposition of compound 9d with Topotecan.
References and notes
cently been improved by incorporating a base portion matching
algorithm that allows prepositioning of a fragment of the ligand
being docked in the binding site. The fragment is allowed to trans-
fer from its original position to a certain point during pose optimi-
zation. This is significant when the position of the base portion is
not completely set. Ligand docking with the base fragment match-
ing characteristic is proposed to yield docking and scoring of li-
gands constrained to match an exact binding motif.
1. Creemers, G. J.; Lund, B.; Verweij, J. Cancer Treat. Rev. 1994, 20, 73.
2. Liew, S. T.; Yang, L. X. Curr. Pharm. Des. 2008, 14, 1078.
3. Meng, L. H.; Liao, Z. Y.; Pommier, Y. Curr. Top. Med. Chem. 2003, 3, 305.
4. Pommier, Y. Nat. Rev. Cancer 2006, 6, 789.
5. Cinelli, M. A.; Morrell, A.; Dexheimer, T. S.; Scher, E. S.; Pommier, Y.; Cushman,
M. J. Med. Chem. 2008, 51, 4609.
6. Antony, S.; Agama, K. K.; Miao, Z. H.; Takagi, K.; Wright, M. H.; Robles, A. I.;
Varticovski, L.; Nagarajan, M.; Morrell, A.; Cushman, M.; Pommier, Y. Cancer
Res. 2007, 67, 10397.
The structure of the inhibitor 9d was drawn into the Sybyl pack-
age with standard bond lengths and angles and minimized using
the conjugate gradient method until the gradient was 0.001 kcal/
mol with the Tripos force field. The Gasteiger–Huckel charge, with
a distance-dependent dielectric function, was applied for the min-
imization process. We chose the 1SC7 (PDB code) structure in Pro-
tein Data Bank and the structure was refined as follows.11 The
phosphoester bond of G12 in 1SC7 was rebuilt, and the SH of
G11 on the scissile strand was changed to OH. After running Sur-
flex-Dock, 10 docked conformers were displayed in a molecular
spread sheet to rank the scores. We selected the best total score
(4.85) conformer and speculated regarding the detailed binding
patterns in the cavity. The resulting docking model revealed a sim-
ilar binding mode as the indenoisoquinoline model. In our model,
the isoquinoline ring intercalated between the À1 and +1 bases,
parallel to the plane of the base pairs, and the amide carbonyl
group had H-bond to Asn 722, which is considered an important
amino acid that interacts with the ligand in the DNA–top 1 active
site. In our model, the isoindoloisoquinoline ring worked as a DNA
intercalator as a blocker of the rewinding step of the phosphoester.
The binding geometry of CPT in the DNA–top 1 complex was inves-
tigated by an ab initio quantum mechanics calculation to give the
7. Morrell, A.; Placzek, M.; Parmley, S.; Antony, S.; Dexheimer, T. S.; Pommier, Y.;
Cushman, M. J. Med. Chem. 2007, 50, 4419.
8. Zhang, A.; Zhou, G.; Hoepping, A.; Mukhopadhyaya, J.; Johnson, K. M.; Zhang,
M.; Kozikowski, A. P. J. Med. Chem. 2002, 45, 1930.
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. 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.
11. Xiao, X. S.; Antony, S.; Pommier, Y.; Cushman, M. J. Med. Chem. 2005, 48, 3231.
12. Le, T. N.; Gang, S. G.; Cho, W. J. J. Org. Chem. 2004, 69, 2768.
13. Synthetic experimental of the representative compounds; 7d: A solution of N,N-
diethylbenzamide 5d (1.68 g, 8.8 mmol) and benzonitrile 6b (1.52 g, 7.3 mmol)
in dry THF (20 mL) were added dropwise to a solution of n-butyllithium (6 mL
of 2.5 M in hexane, 15 mmol) in THF (20 mL) at À70 °C, and then the reaction
mixture was stirred at the same temperature for 6 h. The reaction was
quenched with water, extracted with ethyl acetate and dried over sodium
sulfate. After removal of the solvent, the residue was purified by column
chromatography with n-hexane–ethyl acetate (1:1) to afford compound 7d as
yellow oil (985 mg, 41%). IR (cmÀ1): 3447 (NH), 1655 (C@O). 1H NMR
(300 MHz, CDCl3) d: 9.79 (s, 1H), 8.40 (d, 1H), 7.67 (m, 1H), 7.56 (m, 1H),
7.48 (m, 2H), 7.03 (m, 1H), 6.97 (m, 1H), 6.52 (s, 1H), 4.80 (s, 2H), 4.56 (s, 2H),
3.87 (s, 3H), 3.43 (s, 3H). EIMS m/z (%) 325 (M+, 100). 9a: To a solution of
compound 8a (50 mg, 0.2 mmol) in CH2Cl2 was added triethylamine (46 mg,
0.4 mmol) followed by methanesulfonylchloride (101 mg, 1.0 mmol) at 0 °C.
The reaction mixture was warmed-up to room temperature and stirred
overnight. The reaction was quenched with water and extracted with CH2Cl2.
The organic layers were washed with water, brine, and dried over sodium
sulfate. After removing the solvent, the residue was purified by column
chromatography
with
n-hexane–ethyl
acetate
to
afford
the
fact that the
p–p stacking interactions, DNA intercalating forces
isoindoloisoquinoline 9a as yellow solid (24 mg, 51 %). Mp: 151.7–154.1 °C.
IR (cmÀ1): 1657 (C@O). 1H NMR (300 MHz, CDCl3) d: 8.48 (d, J = 8.1 Hz, 1H),
7.75–7.43 (m, 7H), 6.98 (s, 1H), 5.15 (s, 2H). EIMS m/z (%) 233 (M+, 100). 9b;
mp: 157.8–160.2 °C. IR (cmÀ1): 1657 (C@O). 1H NMR (300 MHz, CDCl3) d: 8.38
(d, J = 8.1 Hz, 1H), 7.80 (m, 1H), 7.57 (m, 1H), 7.48 (m, 2H), 7.42 (s, 1H), 7.31 (d,
1H), 6.97 (s, 1H), 5.19 (s, 2H), 2.50 (s, 3H). EIMS m/z (%) 247 (M+, 92). 9c; mp:
155.2–158.8 °C. IR (cmÀ1): 1657 (C@O). 1H NMR (300 MHz, CDCl3) d: 8.30 (s,
1H), 7.79 (m, 1H), 7.57–7.45 (m, 5H), 7.02 (s, 1H), 5.20 (s, 2H), 2.51 (s, 3H).
EIMS m/z (%) 247 (M+, 100).
was much more important than hydrogen bonding of the ligand
to the surrounding amino acid residues of the protein, or to the
base pairs.17 Our molecular docking study proved the importance
of DNA intercalation of 9d and it was clarified by the superimposi-
tion of 9d with Topotecan as shown in Figure 4.
In conclusion, we accomplished concise synthesis of various
isoindolo[2,1-b]isoquinolin-5-one as analogs of constrained 3-aryl-
isoquinolines structures. An intramolecular cycloaddition reaction
was employed to efficiently generate isoindolo[2,1-b]isoquinolin-
5-one. 3-Arylisoquinoline synthetic intermediates exerted rela-
tively strong cytotoxicities and most of the newly synthesized iso-
14. 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.
15. 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.
16. Jain, A. N. J. Comput. Aided Mol. Des. 2007, 21, 281.
17. Xiao, X. S.; Cushman, M. J. Am. Chem. Soc. 2005, 127, 9960.
indolo[2,1-b]isoquinolines exhibited potent top
1 inhibitory