2'-Substituted 5-phenylterbenzimidazoles as topoisomerase I poisons

Article abstract of DOI:10.1016/S0968-0896(00)00054-7

5-Phenylterbenzimidazole (1) is active as a topoisomerase I poison (topo I) and is cytotoxic to human tumor cells. No cross-resistance was observed for 1 when it was evaluated against the camptothecin-resistant cell line, CPT-K5. Derivatives of 1 substituted at the 2'-position, however, did exhibit cross-resistance to this cell line. The basis for the resistance of this cell line towards CPT is that it possesses a mutant form of topo I. These results suggest that substituents at the 2'-position may be in proximity to the wild-type enzyme. Therefore, we hypothesized that terbenzimidazoles with 2'-substituents could be capable of interacting with the enzyme and thereby influence activity within this class of topo I poisons. 5-Phenylterbenzimidazoles with a hydroxy, hydroxymethyl, mercapto, amino, N-benzoylaminomethyl, chloro, and trifluoromethyl group at the 2'-position were synthesized. In addition, several 2'-ethyl-5-phenylterbenzimidazoles were prepared containing either a methoxy, hydroxy, amino, or N-acetylamino group at the 2-position of the ethyl side-chain. These 2'-substituted 5-phenylterbenzimidazoles were evaluated as topo I poisons and for cytotoxic activity. The presence of a strong electron-withdrawing group at the 2'-position, such as a chloro or trifluoromethyl group, did enhance both topo I poisoning activity and cytotoxicity. Studies on the relative DNA binding affinity of 1 to its 2'-amino and 2'-trifluoromethyl derivatives did exhibit a correlation with their relative differences in biological activity. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00054-7

Bioorganic & Medicinal Chemistry 8 (2000) 1371±1382
200-Substituted 5-Phenylterbenzimidazoles as Topoisomerase I
Poisons
Meera Rangarajan,a Jung Sun Kim,a Song Jin,a Sai-Peng Sim,b Angela Liu,b,c
Daniel S. Pilch,b,c Leroy F. Liub,c and Edmond J. LaVoiea,c,*
aDepartment of Pharmaceutical Chemistry, Rutgers, The State University of New Jersey, Piscataway,
NJ 08854, USA
bDepartment of Pharmacology, The University of Medicine and Dentistry of New Jersey,
Robert Wood Johnson Medical School, Piscataway, New Jersey 08854, USA
cThe Cancer Institute of New Jersey, New Brunswick, New Jersey 08901, USA
Received 11 January 2000; accepted 9 February 2000
AbstractÐ5-Phenylterbenzimidazole (1) is active as a topoisomerase I poison (topo I) and is cytotoxic to human tumor cells. No
cross-resistance was observed for 1 when it was evaluated against the camptothecin-resistant cell line, CPT-K5. Derivatives of 1
substituted at the 200-position, however, did exhibit cross-resistance to this cell line. The basis for the resistance of this cell line
towards CPT is that it possesses a mutant form of topo I. These results suggest that substituents at the 200-position may be in
proximity to the wild-type enzyme. Therefore, we hypothesized that terbenzimidazoles with 200-substituents could be capable of
interacting with the enzyme and thereby in¯uence activity within this class of topo I poisons. 5-Phenylterbenzimidazoles with a
hydroxy, hydroxymethyl, mercapto, amino, N-benzoylaminomethyl, chloro, and tri¯uoromethyl group at the 200-position were
synthesized. In addition, several 200-ethyl-5-phenylterbenzimidazoles were prepared containing either a methoxy, hydroxy, amino, or
N-acetylamino group at the 2-position of the ethyl side-chain. These 200-substituted 5-phenylterbenzimidazoles were evaluated as
topo I poisons and for cytotoxic activity. The presence of a strong electron-withdrawing group at the 200-position, such as a chloro
or tri¯uoromethyl group, did enhance both topo I poisoning activity and cytotoxicity. Studies on the relative DNA binding anity
of 1 to its 200-amino and 200-tri¯uoromethyl derivatives did exhibit a correlation with their relative di€erences in biological activity.
# 2000 Elsevier Science Ltd. All rights reserved.
Introduction
daunomycin. Crystal structures of human topoisomerase
I (topo I) in covalent and non-covalent complexes with
DNA have been identi®ed.8 These ®ndings have aided
in constructing several design principles for camptothe-
cin analogues. All of the topo I poisons approved for
clinical use are analogues of camptothecin (CPT).9
There is substantial evidence that camptothecin binds
reversibly to the putative covalent reaction intermediate,
the cleavable complex.10,11 It has been well established
that stabilization of the cleavable complex is critically
linked to the cytotoxicity and antitumor activity of these
CPT analogues.
The chemotherapeutic action of a number of anticancer
agents has been linked to their ability to inhibit nuclear
DNA topoisomerases. DNA topoisomerases are
enzymes that are present in the nuclei of cells and cata-
lyze the breaking and rejoining of DNA strands, thus
regulating the topological state of DNA.1 4 Recent stu-
dies suggest that topoisomerases are also involved in
controlling template supercoiling during RNA tran-
scription and helicase movement.5 7 Topoisomerases
have been classi®ed into type I or II depending on their
ability to produce transient protein-bridged single-strand
or double-strand DNA breaks. There are a diversity of
topoisomerase II (topo II) poisons that have potent
activity and are available clinically, e.g., etoposide (VP-
16), teniposide (VM-26), mitoxantrone, doxorubicin and
Hoechst 33258, 20-(4-hydroxyphenyl)-5-(4-methylpiper-
azinyl)-2,50-bi-1H-benzimidazole (NSC 32291, pibenzi-
mol), and Hoechst 33342, 20-(4-ethoxyphenyl)-5-(4-
methylpiperazinyl)-2,50-bi-1H-benzimidazole, represent
a structurally-unique class of topo I poisons.12 14 These
agents bind to the minor groove of DNA and have been
shown to bind with AT+topo I speci®city.15,16 More
recently, terbenzimidazoles have been identi®ed as topo
*Corresponding author. Tel.: +1-732-445-2674; fax: +1-732-445-
6312; e-mail: elavoie@rci.rutgers.edu
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00054-7

1372
M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
I poisons.17 In a study with 5-phenyl-200-substituted ter-
benzimidazoles, alkyl groups at the 200-position retained
activity as topo I poisons and exhibited good cytotoxi-
city against RPMI 8402, a human lymphoblastoma cell
line.18 Several terbenzimidazoles were also evaluated in
CPT-K5 cells, a camptothecin-resistant variant of
RPMI 8402. The basis for resistance of this cell line
towards CPT has been postulated to be associated with
the fact that it possesses a mutant form of topo I.19
Studies in our laboratory indicated that several terben-
zimidazole derivatives unsubstituted at the 200-position
did not exhibit signi®cant cross-resistance in CPT-K5
cells.17 However, all the terbenzimidazole derivatives
with varied alkyl and aryl groups at the 200-position
evaluated for cytotoxic activity were cross-resistant in
CPT-K5 cells (unpublished results). These results raise
the possibility that substituents at the 200-position may
be in close enough proximity to interact with speci®c
functional groups on the wild-type enzyme. Based upon
these results, we hypothesized that terbenzimidazoles
with 200-substituents capable of interacting with the
enzyme would signi®cantly in¯uence activity within this
class of topo I poisons. Several 5-phenyl-200-substituted
terbenzimidazoles with varied physicochemical proper-
ties were synthesized. The resulting analogues (listed in
Table 1) were evaluated for their relative topo I poi-
soning activities and cytotoxicities.
Treatment of 16 with the potassium salt of ethylxanthic
acid in ethanol provided 3 in 69% yield. Cyanogen
bromide was reacted with 16 to give 4 in 47% yield.
Compounds 5, 6, 8, 9, 11 and 14 were prepared as shown
in Scheme 2. Compound 17 was synthesized from 2-(3,4-
dinitrophenyl)-5-phenylbenzimidazole.18 Reaction of 17
with the corresponding aldehydes 18±23 in nitrobenzene
at 145 ꢀC gave the desired terbenzimidazoles. Treatment
of 11 with BBr3 provided 12.20 The N-acetyl moiety on
compound 14 was hydrolyzed by treatment with 2 N
HCl to give the amine 13 in quantitative yield.
Aldehydes 19±23 were prepared by reduction and
hydrolysis of their corresponding 5-cyanobenzimida-
zoles using Ni±Al in formic acid as outlined in Scheme
3. Several of these 5-cyanobenzimidazoles 26±30 were
prepared using the Phillips procedure.21,22 Compound 24
was synthesized from the commercially available 4-amino-
3-nitrobenzonitrile by hydrogenation-reduction using Pd/
C catalyst. Compound 25 was synthesized by condensa-
tion of 24 with urea in DMF in 81% yield. 2-Tri-
¯uoromethyl-5-cyanoterbenzimidazole (26) was prepared
by re¯uxing a solution of 24 and tri¯uoroacetic acid
Chemistry
Synthetic methods for the preparation of 1, 7, 10 and 15
have been previously described.18 The methods used for
the preparation of 2±4 are outlined in Scheme 1. Con-
densation of 16 with urea/DMF gave 2 in 88% yield.
Scheme 1.
Table 1.
Compound
n
X
1
2
3
4
5
6
7
8
0
0
0
0
0
0
1
1
1
2
2
2
2
2
2
H
OH
SH
NH
Cl
CF3
H
OH
9
NHCOC6H5
H
10
11
12
13
14
15
OCH3
OH
NH2
NHCOCH3
CH3
Scheme 2.

M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
1373
Table 2. Summary of the cyano derivatives and their yields prepared
from compound 24
No.
n
X
a
Conditions
Yield
(%)
25
26
27
28
29
30
0
0
1
1
2
2
OH
CF3
OH
H2NCONH2
CF3COOH
CH2OHCOOH
150 ꢀC, DMF
Re¯ux
4 N HCl, re¯ux
160 ꢀC, fuse
81
88
66
86
50
46
NHCOPh HOOCCH2NHCOPh
OCH3
NH2
HOOCCH2CH2OCH3 4 N HCl, re¯ux
HOOCCH2CH2NH2 6 N HCl, re¯ux
Table 3. Summary of the 5-formylbenzimidazoles synthesized and
the yields obtained from their corresponding cyano derivatives
Scheme 3.
(neat). Reaction of 24 with glycolic acid in 4 N HCl gave
2-hydroxymethyl-5-cyanobenzimidazole (27) in 66%
yield. The method of Lions and Hughes23 was followed to
prepare the 5-cyano-2-[(N-benzoyl)aminomethyl]benzi-
midazole 28 in 86% yield. Hippuric acid was ®nely
ground along with 24 and heated to obtain a melt.
Further heating at 165 ꢀC for 3 h gave the product.
Compound 24 was re¯uxed with 3-methoxypropionic
acid in 4 N HCl according to Phillips' procedure to give
the 5-cyano-2-(2-methoxyethyl)benzimidazole (29).
No.
n
X
Yield (%)
18
19
20
21
22
23
0
0
1
1
2
2
Cl
CF3
OH
34a
41
61
26
65
80
NHCOPh
OCH3
NHCOCH3
aThis represents the overall yield from 25.
its camptothecin-resistant variant CPT-K5 cells. The
relative topo I poisoning activities and cytotoxicities in
RPMI 8402 cells of these terbenzimidazole analogues 1±
15 are summarized in Table 4.
Reaction of 24 with b-alanine in 6 N HCl gave the 5-
cyano-2-(2-aminoethyl)benzimidazole (30). The amino
group was protected as the acetamide to prevent self-
condensation upon reduction of nitrile to the aldehyde.
5-Cyano-2-(2-acetamidoethyl)benzimidazole (32) was
formed from 30 with acetic anhydride/TEA and used as
the precursor to form the 5-formyl analogue. Table 2
gives a summary of the yields of the cyano derivatives.
The nitriles 25±30 and 32 were reduced in good yields to
the 5-formyl derivatives (18±23) using HCOOH/Ni±Al
catalyst. Table 3 gives a summary of the aldehydes
obtained from their corresponding cyano derivatives.
Aldehydes 19±23 were further used for the condensation
reaction with 17 as shown in Scheme 2. Compound 31
was converted to 18 using POCl3 in 62% yield, which
was then used for the condensation reaction.
Topo I poisoning activity
Inspection of the data in Table 4 reveals that analogues
1±15 were all active as topo I poisons. Quantitative
comparisons of the relative topo I poisoning eciencies
of terbenzimidazoles are frequently confounded by the
fact that the induced DNA cleavage pattern can di€er
from compound to compound. An example of this e€ect
can be seen in the DNA cleavage pattern observed for 8
as compared to 11 in the representative gel shown in
Figure 1. Speci®cally, some cleavage bands induced by
11 are not suciently intense in the cleavage pattern
induced by 8 to quantify accurately. In addition, strong
DNA binding by terbenzimidazoles can inhibit DNA
binding by topo I, thereby resulting in diminished
enzyme-mediated cleavage. Inhibition of DNA cleavage
is frequently observed within this class of topo I poisons
at high concentrations of the test compound. This e€ect
can be seen in Figure 1. Speci®cally, as the concentra-
tion of 1 is increased from 1 to 10 mM, a decrease in the
intensity of its major DNA cleavage product (denoted
with an arrow) is observed. Despite these confounding
e€ects, a dominant DNA fragment band (as indicated
with an arrow in Figure 1) typically results from the
interaction of most of the terbenzimidazoles studied to
date. The relative intensity of this band at concentra-
tions of compound that do not inhibit cleavage was
used to provide an assessment of the relative potency of
these terbenzimidazoles as topo I poisons.
Results and Discussion
Terbenzimidazoles represent a structurally unique class
of topo I poisons. Comparative biological and biophy-
sical data have suggested that formation of a ternary
enzyme±DNA±terbenzimidazole complex may involve
alignment of the terbenzimidazole molecule such that its
200-end is in close proximity to the enzyme. Analogues of
5-phenylterbenzimidazole with various 200-substituents
that may potentially interact with the enzyme and fur-
ther stabilize the cleavable form of the ternary complex
were synthesized. Comparative assays were performed
on these analogues to assess their relative cytotoxicities
and potencies as topo I poisons. Cytotoxicity was tested
against the human lymphoblast cell line RPMI 8402 and

1374
M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
Table 4. Biological activity of 200-substituted terbenzimidazoles
the negative in¯uence of the amino group with respect
to topo I poisoning activity was diminished. In the case
of the N-acetyl analogue of the 2-aminoethyl derivative
14 similar topo I poisoning activity to 1 was observed.
The 200-methoxyethyl analogue 11 and the 200-(2-hydro-
xyethyl) derivative 12 of 5-phenylterbenzimidazole had
similar activity to 1 as a topo I poison. The reason for
the slight increase in activity observed for the 200-hydro-
xymethyl analogue 8 is not clear.
Compound
Relative topo
I-mediated DNA
cleavagea
Cytotoxicity
RPMI 8402b
(IC50, mM)
1
2
3
4
5
6
7
8
1.0
2.0
1.0
5.0
0.1
0.5
2.0
0.5
2.0
2.0
1.0
2.0
1.0
1.0
0.5
0.27
0.20
0.12
>10
0.07
0.07
0.11
0.54
9.5
The 200-chloro derivative 5 and the 200-tri¯uoromethyl
derivative 6 were among the more potent topo I poisons
evaluated, having greater potency than either 5-phe-
nylterbenzimidazole 1 or its 200-methyl derivative 7.
Previous studies by Kim et al.24 demonstrated that the
presence of a methyl, ethyl, or n-propyl substituent at
the 200-position of the 5-phenylterbenzimidazole as in 7,
10 and 15 resulted in relative potencies as a topo I poi-
sons that ranged from one-half to twice that observed
for the 200-unsubstituted analogue 1.18 In aggregate,
these data suggest that steric factors alone do not sig-
ni®cantly alter activity. The presence of functional
groups capable of participating in hydrogen bonding
interactions did not result in a signi®cant enhancement
in topoisomerase I poisoning activity.
9
10
11
12
13
14
15
0.11
0.29
2.55
9.26
0.1
0.06
aTopoisomerase I cleavage values are reported in relative e€ective
concentration (REC) values, as compared to compound 1, whose
value is assumed to be arbitrarily 1.0, that produce a similar degree of
DNA cleavage. Cleavage is calculated from the intensity of the speci®c
band indicated by an arrow in Figure 1. Campothecin is approxi-
mately 10 times more potent than 5 based upon comparisons of DNA
cleavage at the lower concentrations of drug where detectable levels
are initially observed. As the DNA fragmentation pattern and DNA
binding anity are very di€erent between terbenzimidazoles and
camptothecin, one should interpret this comparison with caution (see
Experimental).
Increased potency as topoisomerase I poisons was
observed for those terbenzimidazoles that possessed
electron withdrawing substituents at the 200-position,
such as a tri¯uoromethyl group. The strong electron
withdrawing e€ect of the tri¯uoromethyl group may
in¯uence the binding of 6 to either the DNA or the
enzyme by altering the hydrogen bonding capacity of
the terminal benzimidazole. This altered binding a-
nity, in turn, may contribute to its enhanced topo I
poisoning eciency.
bCytotoxicity values, as their IC50 values in mM, are reported after 4
days of continuous exposure. Camptothecin has an IC50 of 0.006 mM
toward RPMI 8402 and an IC50 of 60 mM toward CPT-K5 cells.
DNA binding studies
Spectroscopic and calorimetric techniques were used to
characterize the relative DNA binding strengths of com-
pounds 1, 4 and 6 to assess the impact of di€ering 200-
substituents on terbenzimidazole±DNA recognition. The
.
poly(dA) poly(dT) duplex was chosen as the DNA target
in these studies, since our previous studies15,16 have
revealed that terbenzimidazoles exhibit high anities for
.
A T-rich tracts in duplex DNA.
Figure 2 shows the UV melting curves for the poly
.
(dA) poly(dT) duplex in the absence and presence of
Figure 1. Representative gel illustrating stimulation of enzyme-mediated
DNA cleavage by 1 and 8±11 using human topoisomerase I. The left-
most lane is the DNA control without topoisomerase I. The second
lane from left is the control with topoisomerase I alone. The rest of the
lanes are with topoisomerase I and serially (10-fold each) diluted 1, 8,
9, 10 and 11.
compounds 1, 4 and 6 at the concentrations required to
saturate the host DNA. Note that the presence of each
of the three terbenzimidazole analogues enhances the
.
thermal stability of poly(dA) poly(dT) duplex. These
analogue-induced changes in duplex thermal stability
are consistent with all three ligands binding to the host
duplex, with a preference for the duplex versus the sin-
gle-stranded state.25,26 Further inspection of Figure 2
reveals that the extent of binding-induced enhancement
in duplex thermal stability and/or the concentration of
ligand required to saturate the host duplex di€er from
analogue to analogue. Speci®cally, the same concentra-
tions of compounds 1 and 6 are required to saturate the
The 200-hydroxy derivative 2 had approximately half of
the potency of 1 as a topo I poison. The 200-mercapto
derivative 3 possessed similar intrinsic activity to that of
1. The presence of an NH2 group at the 200-position of 4
substantially reduced topo I poisoning activity relative
to 1. When the amino moiety was separated by two
methylene groups from the benzimidazole ring as in 13

M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
1375
‰bpŠ
L ˆ Ltot
2n
ꢁ2†
where Ltot is the total ligand concentration and [bp] is
the total concentration of DNA base pairs.
Table 5 summarizes the KT values that we have calcu-
lated using eqs (1) and (2m). Inspection of these data
reveals that the apparent binding anities of com-
.
pounds 1, 4 and 6 for the poly(dA) poly(dT) host duplex
follow the hierarchy 6>1>4. Note the agreement
between this hierarchy and that noted above for the
topo I poisoning activities of the three compounds (see
Table 4). Thus, the di€erential topo I poisoning activities
of the terbenzimidazole analogues studied here may be
due, at least in part, to their di€erential DNA binding
properties. Contrary to our initial hypothesis, the greater
impact of 200-substituents in regard to stabilization of the
cleavable complex may not be mediated by an enhance-
ment enzyme interaction, but rather by further increasing
the binding interaction with DNA.
.
Figure 2. UV melting pro®les for the poly(dA) poly(dT) duplex (®lled
circles) and its complexes with compounds 1 (open circles), 4 (open
squares) and 6 (®lled squares) at the indicated (total ligand to base
pair) (rbp) ratios. The DNA concentration was 10 mM in base pair and
the solution conditions were 10 mM sodium cacodylate (pH 7.0),
25 mM NaCl and 0.1 mM disodium EDTA. For clarity of presenta-
tion, the melting pro®les, which were acquired at 260 nm, were nor-
malized by subtraction of the upper and lower baselines to yield plots
of fraction single strand versus temperature.29,30
Cytotoxicity
Several of the 5-phenylterbenzimidazole derivatives that
were among the more potent topo I poisons did exhibit
greater cytotoxicity toward RPMI 8402 cells. The more
potent topo I poisons, 5, 6 and 15, were among the most
cytotoxic derivatives. Compounds 4 and 9 were sig-
ni®cantly less potent than 1 as a topo I poison and each
was also substantially less cytotoxic against RPMI 8402
cells. These associations between the subcellular assay
and the cytoxicity assay are interesting, but di€erences
between compounds in regard to cellular absorption or
metabolism can signi®cantly reduce the probability of
there being an absolute correlation between these
observed pharmacological activities. The decreased cyto-
toxicity of 12 and 13 relative to 1 against RPMI 8402 cells,
for example, would not have been anticipated on the basis
of their intrinsic topo I poisoning activity. In CPT-K5
cells, the campthothecin-resistant variant of RPMI 8402
cells, no signi®cant cross-resistance was observed for
compounds 1±3. The IC50 values of 1±3 in this cell line
were 1.7, 0.6 and 0.7 mM, respectively. We did, however,
observe signi®cant cross-resistance for compounds 4±15
in CPT-K5 cells. In each instance the IC50 observed in
this cell line was at least 6 mM and frequently >10 mM.
These data support the hypothesis that the primary site
of action associated with the cytotoxicity of these ter-
.
host poly(dA) poly(dT) duplex. However, compound 6
.
binding enhances the thermal stability (Tm) of poly(dA)
poly(dT) by 24.5 ꢀC, while compound 1 binding results in
a duplex thermal enhancement of only 21.8 ꢀC. Ten times
more of compound 4 is required to saturate the host
.
poly(dA) poly(dT) duplex than either compound 1 or 6.
However, the binding of compound 4 enhances the ther-
mal stability of the host duplex to a greater extent
(ÁTm=29.4 ꢀC) than either compound 1 or 6. Thus, as
.
measured by di€erences in ÁTm, the poly(dA) poly(dT)
duplex is able to distinguish between compounds 1, 4 and
6, which di€er only with respect to their 200-substituents.
We used the ÁTm method described below to assess the
relative anities of compounds 1, 4 and 6 for the host
.
poly(dA) poly(dT) duplex. Measured ligand-induced
.
changes in the thermal stability of the poly(dA) poly(dT)
duplex (see Fig. 2) were used in conjunction with the 6
base pair binding site size (n) revealed by the X-ray crys-
tallographic studies27 to estimate apparent ligand±DNA
association constants at Tm (KT ) from the expression:25
m
1
1
R
ˆ
lnꢁ1 ‡ KT L†
ꢁ1†
Table 5. ÁTm-Derived binding anities of compounds 1, 4 and 6 for
.
the poly(dA) poly(dT) duplex
m
Tꢀm Tm nꢁÁHWC
†
Compound
Tꢀm (ꢀC)a
Tm (ꢀC)
KT Â108 (M
)
1
m
In this equation, Tꢀm and Tm are the melting tempera-
tures of the ligand-free and ligand-saturated duplexes,
respectively; ÁHWC is the enthalpy change for the melt-
ing of a Watson±Crick base pair in the absence of bound
ligand (a value we determined independently for the
1
4
6
59.7
59.7
59.7
81.5
89.1
84.2
2.8
1.1
5.8
aTm values were derived from the UV melting pro®les shown in Figure
2, as described previously.29,30 Solution conditions were 10 mM
sodium cacodylate (pH 7.0), 25 mM NaCl and 0.1 mM disodium
EDTA. Binding constants were determined using eqs 1 and 2, as well
as a calorimetrically determined duplex-to-single strand transition
enthalpy (ÁHWC) of 10.4 kcal/mol base pair.28
.
poly(dA) poly(dT) duplex using di€erential scanning
calorimetry (DSC));28 and L is the free ligand con-
centration at Tm. We estimated the value of L using the
following relationship:

1376
M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
benzimidazoles is topo I. In a comparison of the cyto-
toxicity of these compounds in another part of camp-
tothecin-sensitive and camptothecin-resistant cell lines,
U937 and U937/CR, however, no similar cross-resis-
tance was observed in the cytotoxicity assays performed
with 1±15. In the case of CPT-K5 and U937/CR, there
are di€erent mutant forms of topo I.19,31 The basis for
this di€erence in response could be linked to the speci®c
topo I mutants that exist in each cell type. Despite the
fact that the target enzyme is the same, cross-resistance
of one mutant form of the enzyme to one series of
compounds (camptothecin-like) may not necessarily be
observed with a distinctly di€erent series of compounds
(terbenzimidazoles). In addition, one cannot preclude
that other mechanisms, such as drug uptake, could be
linked to resistance.
(Harrison Research, CA). Infrared spectral data (IR)
were obtained on a Perkin±Elmer 1600 Fourier trans-
1
form spectrophotometer and are reported in cm
.
Proton (1H NMR) and carbon (13C NMR) nuclear
magnetic resonance were recorded on a Varian Gemini-
200 Fourier transform spectrometer. NMR spectra
1
(200 MHz H and 50 MHz 13C) were recorded in the
deuterated solvent indicated with chemical shifts repor-
ted in d units down®eld from tetramethylsilane (TMS).
Coupling constants are reported in hertz (Hz). A few
drops of CF3COOH improved 13C NMR spectra by
allowing for increased solubility and formation of the
protonated form of the terbenzimidazoles, thereby
eliminating tautomeric di€erences among carbon atoms.
Mass spectra were obtained from Washington Uni-
versity Resource for Biomedical and Bio-organic Mass
Spectrometry within the Department of Chemistry at
Washington University, St. Louis, MO. The purity of
all compounds for which HRMS data are provided was
determined by analytical reverse-phase HPLC. Com-
pounds were analyzed using both of the following
conditions: (method A) a Vydac C-18 column (The
Separations Group) using methanol:H2O (87:13) with a
¯ow rate of 1 mL/min; (method B) a Microsorb C-8
column (Rainin Instrument Co., Inc.) using metha-
nol:0.1 M potassium phosphate bu€er (pH 7.0) (70:30)
with a ¯ow rate of 1 mL/min. HPLC analyses were per-
formed with a Hewlett±Packard 1090 liquid chromato-
graph equipped with a diode array UV detection
monitoring at 254 and 335 nm. The % purity of these
compounds was calculated from the peak area assuming
that the extinction coecient of the compound of inter-
est and the impurity is the same. On the basis of these
analyses, all the compounds were found to be 98.0±99%
pure in these systems. Combustion analyses were per-
formed by Atlantic Microlabs, Inc., Norcross, GA, and
were within ‹0.4% of the theoretical value. The synth-
eses of 1, 7 and 15 have been detailed in the literature.18
It is our view that in the case of CPT-K5 cells the spe-
ci®c mutant form of topo I imparts cross-resistance to
terbenzimidazoles. Compound 1, which lacks any 200-
substitution, does not exhibit cross-resistance to the
CPT-K5 cells. It is evident that the presence of a carbon,
chlorine, nitrogen, or sulfur atom at the 200-position results
in a very precipitous loss in cytotoxicity speci®cally
against CPT-K5 cells. Cytotoxicity studies revealed that,
with the exception of compounds 2 and 3, 200-substituted
terbenzimidazoles exhibit a strong cross-resistance in this
camptothecin-resistant variant cell line. It is presumed
that a single point mutation of aspartate to glycine
(Asp533Gly) in CPT-K5 cells makes the terbenzimida-
zoles resistant to this cell line.19
These 200-substituted terbenzimidazoles may be unable
to interact with the mutant form of the enzyme in an
appropriate fashion to stabilize the cleaved form of the
ternary complex. Such may not be the case with the
mutant form of topo I in U97/CR cells.
These studies con®rm that substituents at the 200-end can
in¯uence both topo I poisoning activity and cytotoxicity
of terbenzimidazoles. These data also suggest that the
presence of electronegative substituents at the 200-posi-
tion is associated with enhanced topo I poisoning activity
and cytotoxicity. There was no evidence to speci®cally
suggest that this enhanced activity was related to an
increased interaction with enzyme. The increase in
potency for such substituted terbenzimidazoles may be
based, in part, on enhanced binding to DNA that could
stabilize the cleavable ternary complex comprised of
enzyme, drug, and DNA. While cross-resistance is
observed for 200-substituted terbenzimidazoles with
CPT-K5 cells, this is not observed for all camptothecin-
resistant tumor cells.
5-Phenyl-2-[20(200-hydroxybenzimidazol-500-yl)benzimida-
zol-50 -yl]benzimidazole (2). 5-Phenyl-2-[20-(3,4-diamino-
phenyl)benzimidazol-50-yl]benzimidazole (29.7 mg, 0.07
mmol) and urea (6 mg, 0.1 mmol) were dissolved in DMF
(0.4 mL). The mixture was re¯uxed at 150 ꢀC for 7 h. The
cooled reaction mixture when concentrated in vacuo
provided 27.6 mg (88%) of pure yellow solid: mp
>280 ꢀC; IR (KBr) 3382, 3133, 1693, 1475, 1444, 1279;
1H NMR (DMSO-d6+3 drops CF3COOH) d 7.3 (d, 1H,
J=8.14 Hz), 7.46±7.80 (m, 3H), 7.80 (d, 2H, J=8.1 Hz),
7.89±8.01 (m, 4H), 8.05±8.14 (m, 2H), 8.32 (d, 1H, J=8.3
Hz), 8.63 (s, 1H), 11.35 (s, 1H); 13C NMR (DMSO-d6+3
drops CF3COOH) d 108.1, 109.6, 111.8, 114.0, 114.6,
114.6, 115.3, 120.2, 122.5, 125.5, 125.8, 127.4, 128.1,
129.3, 130.8, 131.6, 132.7, 132.9, 135.2, 135.3, 139.2,
139.6, 149.3, 153.1, 155.7; HRMS (FAB) calcd for
C27H19N6O (MH+) 443.1620, found 443.1625.
Experimental
Melting points were determined with a Thomas±Hoover
Unimelt capillary melting point apparatus. Column chro-
matography refers to ¯ash chromatography conducted on
SiliTech 32±63 mm (ICN Biomedicals, Eschwegge, Ger.)
using the solvent systems indicated. Radial chromato-
graphy refers to the use of a Model 8924 chromatotron
5-Phenyl-2-[20-(200-mercaptobenzimidazol-500-yl)benzimi-
dazol-50-yl] benzimidazole (3). 5-Phenyl-2-[20-(3,4-diami-
nophenyl)benzimidazol-50-yl]benzimidazole (59 mg, 0.14
mmol) and ethylxanthic acid potassium salt (25 mg,

M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
1377
0.16 mmol) were re¯uxed in ethanol (1 mL) and distilled
water (0.1 ml) overnight. The cooled reaction mixture
was acidi®ed to pH 3 with glacial acetic acid, volume
condensed in vacuo and puri®ed directly by column
chromatography. Elution with (40±100%) ethyl acetate/
n-hexanes provided 69% of yellow solid; mp >260 ꢀC;
IR (KBr) 3089, 2926, 2851, 1712, 1624, 1549, 1449, 1380,
by ¯ash column chromatography. (1±16%) Methanol/
ethyl acetate provided 40% pure yellow compound: mp
>280 ꢀC; IR (KBr) 3047, 2927, 1698, 1626, 1543, 1440,
1
1287, 1158; H NMR (DMSO-d6+3 drops CF3COOH)
d 7.44±7.59 (m, 3H), 7.79 (d, 2H, J=7 Hz), 7.87±8.01
(m, 2H), 8.05±8.09 (m, 2H), 8.15±8.19 (m, 1H), 8.29±
8.35 (m, 2H), 8.70 (m, 2H); 13C NMR (DMSO-d6+3
drops CF3COOH) d 111.8, 114.7, 114.9, 115.9, 117.5,
117.6, 118.4, 118.41, 119.6, 120.5, 124.1, 124.8, 125.7,
125.74, 127.5, 128.2, 129.4, 131.7, 133.0, 134.9, 137.4,
139.0, 139.0, 139.1, 139.7, 140.3, 149.8, 153.5; HRMS
(FAB) calcd for C28H18N6F3 (MH+) 495.1545, found
495.1543.
1
1274, 1186, 1079; H NMR NMR (DMSO-d6+3 drops
CF3COOH) d 7.33±7.42 (m, 2H), 7.48±7.63 (t, 4H), 7.71±
7.82 (m, 4H), 7.89 (s, 1H), 8.04±8.06 (m, 1H), 8.1±8.24
(m, 1H), 8.42(d, 1H, J=2.4 Hz); 13C NMR (DMSO-d6+3
drops CF3COOH) d 107.4, 107.2, 109.2, 109.7, 111.8,
111.7, 113.3, 113.5, 114.6, 114.6, 114.7, 115.6, 125.3,
127.4, 128.2, 129.3, 131.6, 132.9, 133.3, 139.1, 136.6,
139.6, 149.5, 152.9, 171.1; HRMS (FAB) calcd for
C27H19N6S (MH+) 459.1392, found 459.1403.
5-Phenyl-2-[20 -(200 -hydroxymethylbenzimidazol-500 -yl)]-
benzimidazol-50-yl]benzimidazole (8). 5-Phenyl-2-[20-(3,4-
diaminophenyl)benzimidazol-50-yl]benzimidazole (178
mg, 0.6 mmol) and 5-formyl-2-hydroxymethyl benzimi-
dazole (104.2 mg, 0.6 mmol) were condensed in nitro-
benzene (10 mL) overnight at 145 ꢀC. Nitrobenzene was
removed from the reaction mixture with a Kugelrohr
and the compound was puri®ed directly by column
chromatography. Elution with (1±18%) methanol/ethyl
acetate provided 62% of yellow solid : mp >260 ꢀC; IR
5-Phenyl-2-[20-(200-aminobenzimidazol-500-yl)benzimidazol-
50 -yl] benzimidazole (4). 5-Phenyl-2-[20-(3,4-diaminophe-
nyl)benzimidazol-50-yl]benzimidazole (66 mg, 0.16 mmol)
was dissolved in DMF (0.2 ml) and methanol (1 mL),
and was added to cyanogen bromide (10% solution in
water, 0.33 ml, 0.63 mmol). The reaction mixture was
stirred overnight at room temperature. The mixture was
concentrated under reduced pressure. The compound
was recrystallized from methanol to give 47% of pale
brown solid: mp >260ꢀC; IR (KBr) 3352, 3052, 2926,
1680, 1624, 1574, 1461, 1261,679; 1H NMR (DMSO-d6+
3 drops CF3COOH) d 7.46±7.62 (m, 3H), 7.79 (d, 2H,
J=7.3 Hz), 7.87±7.96 (m, 2H), 7.99±8.01 (m, 2H), 8.20±
8.26 (m, 2H), 8.31 (s, 1H), 8.72 (d, 2H, J=9 Hz); 13C
NMR (DMSO-d6+3 drops CF3COOH) d 107.4, 110.9,
111.8, 111.8, 114.7, 115.1, 115.2, 115.9, 123.6, 125.6, 126.8,
127.5, 127.53, 127.57, 127.8, 128.2, 129.4, 130.9, 131.9,
133.1, 138.7, 139.7, 150.1, 152.1, 153.4; HRMS (FAB)
calcd for C27H20N7 (MH+) 442.1780, found 442.1776.
1
(Nujol) 3406, 2922, 2725, 1631, 1553, 1461, 1377; H
NMR (DMSO-d6+ 3 drops CF3COOH) d 5.10 (s, 2H),
7.45±7.59 (m, 3H), 7.79 (d, 2H, J=7 Hz), 8.23±8.28 (m,
1H), 8.47 (d, 1H, J=8.6 Hz), 8.68 (s, 2H); 13C NMR
(DMSO-d6+ 3 drops CF3COOH) d 55.8, 111.7, 113.7,
114.6, 115.3, 115.8, 116.2, 118.3, 123.8, 124.5, 125.2,
125.6, 127.5, 127.5, 127.5, 128.1, 129.4, 131.6, 131.9,
132.9, 133.5, 137.7, 138.9, 139.7, 140.1, 150.3, 153.4,
157.6; HRMS (FAB) calcd for C28H21N6O (MH+)
457.1777, found 457.1774.
5-Phenyl-2-[20 -[200 -[2-(N-benzoyl)aminomethyl]benzim-
idazol - 500 - yl]benzimidazol - 50 - yl]benzimidazole (9). 5-
5-Phenyl-2-[20(200-chlorobenzimidazol-500-yl)benzimidazol-
50-yl]benzimidazole (5). 2-(3,4-Diaminophenyl)-5-phenyl-
benzimidazole (185 mg, 0.62 mmol) was heated with the 5-
formyl-2-chlorobenzimidazole (110 mg, 0.62 mmol) in
nitrobenzene (5 mL) at 145 ꢀC overnight. Nitrobenzene
was removed using a Kugelrohr and the compound pur-
i®ed by ¯ash column chromatography. (1±5%) Methanol/
ethyl acetate provided 45% pure yellow compound: mp
>260 ꢀC; IR (KBr) 3165, 2943, 1687, 1584, 1438, 1316,
1150; 1H NMR (DMSO-d6+3 drops CF3COOH) d
7.43±7.59 (m, 4H), 7.76±7.95 (m, 4H), 8.06 (s, 1H), 8.12±
8.19 (m, 2H), 8.29 (m, 1H), 8.52 (s, 1H), 8.64 (m, 1H);
13C NMR (DMSO-d6+3 drops CF3COOH) d 106.7,
111.8, 112.5, 114.7, 115.8, 115.9, 118.2, 118.7, 119.6,
122.8, 123.9, 124.8, 125.7, 127.5, 129.4, 131.8, 133.1,
134.7, 137.2, 138.9, 139.5, 139.7, 141.5, 142.3, 149.8,
153.6; HRMS (FAB) calcd for C27H18N6Cl (MH+)
461.1281, found 461.1278.
Phenyl-2-[3,4-diaminophenyl]benzimidazole
(75 mg,
0.25 mmol) and 5-formyl-2-[(N-benzoyl)aminomethyl]-
benzimidazole (70 mg, 0.25 mmol) were stirred together
in nitrobenzene (6 mL) at 14 ꢀC overnight. Nitrobenzene
was removed with a Kugelrohr and the compound was
loaded on a column. (2±20%) Methanol/ethyl acetate
gave 45% of a yellowish compound: mp >260 ꢀC; IR
(KBr) 3204, 1637, 1542, 1442, 1384, 1292, 1026, 818; 1H
NMR (DMSO-d6+3 drops CF3COOH) d 7.46±7.66 (m,
6H), 7.80 (d, 2H, J=7 Hz), 7.87±8.1 (m, 7H), 8.21±8.25
(m, 1H), 8.43±8.48 (m, 1H), 8.67 (s, 2H); 13C NMR
(DMSO-d6+3 drops CF3COOH) d 36.7, 107.4, 111.7,
114.5, 114.6, 115.3, 115.5, 116.0, 119.5, 122.5, 124.8,
125.6, 126.7, 127.4, 127.8, 128.0, 128.1, 128.6, 129.3,
131.6, 131.9, 132.1, 132.8, 133.2, 134.1, 136.2, 137.8,
139.1, 139.7, 149.7, 152.6, 155.6, 167.6; HRMS (FAB)
calcd for C35H26N7O (MH+) 560.2199, found 560.2209.
5-Phenyl-2-[20-[200-(2-methoxyethyl)benzimidazol-500-yl]-
benzimidazol-50-yl]benzimidazole (11). 5-Phenyl-2-[3,4-
diaminophenyl]benzimidazole (170mg, 0.35mmol) and 5-
formyl-2-(2-methoxyethyl)benzimidazole (120mg, 0.25
mmol) were heated together in nitrobenzene (5 mL) for
15 h at 145 ꢀC. Nitrobenzene was removed with a
Kugelrohr and the compound was loaded onto a col-
umn. (1±12%) Methanol/ethyl acetate gave 80% pure
5-Phenyl-2-[20(200-tri¯uoromethylbenzimdazol-500-yl)benzi-
midazol-50-yl]benzimidazole (6). 5-Phenyl-2-[20-(3,4-di-
aminophenyl)benzimidazol-50-yl]benzimidazole (0.16 g,
0.32 mmol) was stirred with the 5-formyl-2-tri-
¯uoromethylbenzimidazole (0.12 g, 0.54 mmol) in nitro-
benzene (4 mL) at 145 ꢀC overnight. Nitrobenzene was
removed using a Kugelrohr and the compound puri®ed

1378
M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
pale yellow product; mp >280 ꢀC; IR (KBr) 3157, 2933,
1629, 1551, 1441, 1385, 1288, 1108, 818, 760, 698; H
nitrobenzene (7 mL) overnight at 145 ꢀC. Nitrobenzene
was removed with the aid of a Kugelrohr and the com-
pound puri®ed directly by ¯ash column chromato-
graphy. Elution with (1±15%) methanol/ethyl acetate
provided 69% of yellow solid: mp >260 ꢀC; IR (KBr)
3064, 2954, 2862, 1654, 1556, 1442, 1286; 1H NMR
(DMSO-d6+3 drops CF3COOH) d 1.81 (s, 3H), 3.18±
3.35 (m, 2H), 3.61±3.64 (m, 2H), 7.45±7.6 (m, 3H), 7.79 (d,
2H, J=7 Hz), 7.87±8.00 (m, 2H), 8.07±8.30 (m, 4H), 8.46
(d, 1H, J=8.5 Hz), 8.71 (s, 2H); 13 C NMR (DMSO-d6+3
drops CF3COOH) d 22.6, 27.9, 36.4, 111.7, 113.7, 114.6,
115.2, 115.7, 116.2, 118.6, 124.0, 125.2, 125.7, 127.5, 127.6,
128.1, 128.2, 129.4, 131.6, 131.8, 132.9, 133.7, 137.1, 138.9,
139.5, 139.7, 150.2, 153.24, 155.4, 158.5, 170.3; HRMS
(FAB) calcd for C31H25N7O (MH+) 512.2199, found
512.2201.
1
NMR (DMSO-d6+3 drops CF3COOH) d 3.33 (s, 3H),
3.45 (t, 2H), 3.89 (t, 2H), 7.45±7.59 (m, 3H), 7.79 (d, 2H,
J=7 Hz), 7.86±7.99 (m, 2H), 8.06±8.14 (m, 3H), 8.26±8.31
(dd, 1H, J=1.3, 8.4 Hz), 8.44±8.49 (dd, 1H, J=1,8.8 Hz),
8.72 (s, 2H); 13C NMR (DMSO-d6+3 drops CF3COOH)
27.7, 58.3, 68.2, 111.7, 113.7, 114.6, 115.2, 115.7, 116.2,
118.5, 123.9, 124.2, 125.3, 125.6, 127.5, 128.1, 129.4, 131.7,
131.7, 132.9, 133.5, 137.2, 138.7, 139.7, 150.2, 153.2, 155.3;
HRMS (FAB) calcd for C30H25N6O (MH+) 485.2089,
found 485.2089.
5-Phenyl-2-[20 -[200 -(2-hydroxyethyl)benzimidazol-500 -yl]-
benzimidazol-50-yl]benzimidazole (12). 5-Phenyl-2-[20-[200-
(2-methoxyethyl)benzimidazol-500 -yl]benzimidazol -50 -
yl]benzimidazole (30 mg, 0.06 mmol) was suspended in
freshly distilled ethyl acetate (30 mL) to which was
added 10 equivalents of BBr3 (0.62 mL, 1.0 M) at
78 ꢀC. After stirring at room temperature overnight,
the reaction was quenched by adding water (10 mL).
The mixture was basi®ed to pH 9.0 with ammonium
hydroxide and extracted with ethyl acetate and dried
(anhyd. Na2SO4) and concentrated in vacuo. The crude
mixture was separated on a chromatotron, eluting with
(0±30%) methanol:ethyl acetate to give 10.6 mg
(0.02 mmol) of the product in 33% yield as a brown solid:
mp 260 ꢀC; IR (KBr) 3401, 3150, 2876, 1632, 1540, 1442,
5-Phenyl-2-[20 -(3,4-diaminophenyl)benzimidazol-50 -yl]-
benzimidazole (16). A solution of 5-phenyl-2-[20-(3,4-
dinitrophenyl)benzimidazol-50-yl]benzimidazole (75 mg,
0.16 mmol) in ethyl acetate (50 mL) was reduced by
hydrogenation over 10% Pd/C (15 mg) for 90 min. The
resulting solution was passed through a bed of Celite
and the ethyl acetate was removed to give the diamine,
which was used further without puri®cation.
5-Phenyl-2-[20-(3,4-dinitrophenyl)benzimidazol-50-yl]benz-
imidazole. 2-(3,4-Diaminophenyl)-5-phenyl-benzimida-
zole (0.48 g, 1.59 mmol) and 3,4-dinitrobenzaldehyde
(0.28 g, 1.45 mmol) were stirred together in 12 mL
nitrobenzene and the mixture was heated overnight at
145 ꢀC. The cooled reaction mixture was then puri®ed
directly by column chromatography and elution with 10±
95% ethyl acetate/hexanes gave 0.44g (0.92 mmol, 63%)
of a yellow solid: mp >260 ꢀC; IR (KBr) 3045, 2893, 1615,
1
1388, 1268, 1013, 823; H NMR (DMSO-d6+3 drops
CF3COOH) d 3.35 (t, 2H), 3.98 (t, 2H), 7.42±7.56 (m, 3H),
7.76 (d, 2H, J=7.7 Hz), 7.85±7.98 (m, 2H), 8.06±8.19 (m,
3H), 8.33 (d, 1H, J=8.5 Hz), 8.44 (d, 1H, J=8.8Hz), 8.73
(s, 2H); 13C NMR (DMSO-d6+3 drops CF3COOH) d
30.8, 58.1, 111.7, 114.4, 114.5, 115.3, 116.0, 116.02,
118.3,119.4, 122.4, 124.7, 124.8, 125.4, 125.7, 127.4, 128.1,
129.3, 131.6, 131.6, 132.8, 134.0, 135.3, 137.8, 139.1, 139.6,
149.7, 152.6, 156.2; HRMS (FAB) calcd for C29H23N6O
(MH+) 471.1933, found 471.1935.
1
1548, 1432, 1363, 1276; H NMR (DMSO-d6+3 drops
CF3COOH) d 7.37±7.41 (m, 1H), 7.51±7.54 (m, 3H), 7.73±
7.87 (m, 4H), 7.90±7.96 (m, 1H), 8.18±8.29 (m, 1H), 8.45±
8.51 (m, 2H), 8.74 (dd, 1H, J=1.7, 8.48 Hz), 8.97 (d, 1H,
J=1.5 Hz); 13C NMR (DMSO-d6+3 drops CF3COOH)
d 111.6, 114.6, 117.8, 123.9, 125.6, 127.5, 128.2, 128.5,
129.4, 129.6, 131.4, 132.3, 132.9, 134.6, 134.9, 138.7, 139.7,
142.5, 142.8, 150.5, 150.9, 157.6, 158.3; HRMS (FAB)
calcd for C26H17N6O4 (MH+) 477.1311, found 477.1307.
5-Phenyl-2-[20-[200-(2-aminoethyl)benzimidazol-500-yl]benz-
imidazol-50-yl] benzimidazole (13). 5-Phenyl-2-[20-(200-(2-
acetamidoethyl)benzimidazol - 500 - yl)]benzimidazol - 50]-
benzimidazole (35 mg, 0.067 mmol) was hydrolyzed with
2 N HCl (5 mL) at 110 ꢀC for 3 h. The cooled reaction
mixture was basi®ed with ammonium hydroxide to pH
9 and puri®ed by column chromatography. Elution with
50:50 methanol:ethyl acetate provided 75% yellow solid:
mp >260 ꢀC; IR (Nujol) 3375, 2719, 1553, 1461, 1377;
1H NMR (DMSO-d6+ 3 drops CF3COOH) d 3.45±3.52
(m, 4H), 7.45±7.56 (m, 2H), 7.79 (d, 2H, J=7 Hz), 7.88±
8.00 (m, 2H), 8.07±8.27 (m, 5H), 8.42 (d, 1H, J=8.9 Hz),
8.68 (d, 2H, J=5 Hz); 13C NMR (DMSO-d6+3 drops
CF3COOH) d 25.5, 36.4, 111.7, 113.8, 114.6, 115.3,
115.8, 116.2, 118.3, 123.8, 123.9, 124.8, 125.6, 127.5,
128.2, 129.4, 131.7, 133.0, 133.1, 134.8, 137.7, 138.8,
139.7, 140.1, 150.3, 153.4, 153.6; HRMS (FAB) calcd for
C29H23N7 (MH+) 470.2093, found 470.2089.
2-(3,4-Diaminophenyl)-5-phenylbenzimidazole (17). 2-
(3,4-Dinitrophenyl)-5-phenylbenzimidazole18
(0.2 g,
0.05 mmol) was dissolved in ethyl acetate (50 mL) and
reduced by hydrogenation using Pd/C 10% (50 mg) as
the catalyst for 1.5 h. The catalyst was removed by ®l-
tration through Celite. The ethyl acetate was con-
centrated in vacuo and the residue was dried to yield
0.13 g of crude solid in 80% yield (0.44 mmol). This
1
crude solid was further used without puri®cation: H
NMR (DMSO-d6+3 drops CF3COOH) d 4.65 (brs,
2H), 5.01 (brs, 2H), 6.62 (d, 1H, J=8 Hz), 7.41±7.48 (m,
8H), 7.69 (d, 2H, J=7 Hz).
5-Phenyl-2-[20 -(200 -(2-acetamidoethyl)benzimidazol-500 -
yl)]benzimidazol - 50]benzimidazole (14). 5-Phenyl-2-[20-
(3,4 - diaminophenyl)benzimidazol - 500 - yl]benzimidazole
(124mg, 0.41mmol) and 5-formyl-2-(2-acetamidoethyl)-
benzimidazole (95 mg, 0.41 mmol) were condensed in
5-Formyl-2-chlorobenzimidazole (18). A mixture of
200 mg (1.23 mmol) of 5-formyl-2-hydroxybenzimidazole
and POCl3 (4mL) was heated under re¯ux overnight.
Upon cooling to ambient temperature, the solution was
added to a mixture of ice and water (20 mL) with vigorous

M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
1379
stirring. The acidic, aqueous solution was basi®ed to pH
10 by the addition of concentrated NH4OH and extrac-
ted with ethyl acetate. The extracts were dried (anhyd
Na2SO4) and concentrated in vacuo. The residue was
puri®ed by column chromatography using (45%) ethyl
acetate/hexanes to give 133 mg (0.74 mmol) white
colored product in 62% yield: mp 165±168 ꢀC; IR (KBr)
3105, 2813, 1703, 1620, 1439, 1283, 991; 1H NMR
(DMSO-d6+3 drops CF3COOH) d 7.66±7.70 (d, 1H,
J=8.5 Hz), 7.78±7.83 (dd, 1H, J=1.5, 8.4 Hz), 8.15 (d,
1H, J=1.5 Hz), 10.15 (s, 1H); 13C NMR (DMSO-d6+3
drops CF3COOH) d 115.01, 118.2, 118.6, 123.5, 131.7,
139.1, 142.1, 192.8) HRMS (EI) calcd for C8H5N2OCl
(m/z) 180.0090, found 180.0084.
extracted with ethyl acetate. The extractions were dried
(anhyd Na2SO4) and concentrated in vacuo. The residue
was puri®ed over column chromatography using (50±
100%) ethyl acetate/hexanes to give 26% of pure cream
colored product: mp 108±109 ꢀC; IR (KBr) 3314, 2933,
1
1687, 1646, 1539, 1289, 806; H NMR (DMSO-d6+3
drops CF3COOH) d 4.99 (d, 1H, J=5 Hz), 7.54±7.63 (m,
4H), 7.93±8.07 (m, 3H), 8.34 (s, 1H), 10.15 (s, 1H); 13C
NMR (DMSO-d6+3 drops CF3COOH) 36.8, 115.1,
117.2, 125.8, 127.9, 128.7, 132.3, 132.3, 133.2, 133.7,
155.8, 167.5, 192.5; HRMS (EI) calcd for C16H13N3O2
m/z 279.1008, found 279.1003.
5-Formyl-2-(2-methoxyethyl)benzimidazole (22). 5-
Cyano-2-(2-methoxyethyl)benzimidazole (0.12 g, 0.61
mmol) was mixed in formic acid (9 mL) and water
(1 mL), to which was added Ni±Al (0.6 g). The mixture
was heated at 95 ꢀC for 5.5 h. The mixture was passed
through a Celite bed while hot and the bed was washed
with methanol and the ®ltrate concentrated. After neu-
tralization with 2 N NaOH and extraction with CHCl3,
the extracts were dried (anhyd Na2SO4) and con-
centrated in vacuo. The residue was puri®ed by column
chromatography using a gradient of 15±100% ethyl
acetate/hexanes to give 80 mg (0.39 mmol) white colored
product in 65% yield: mp 84±84 ꢀC; IR (Nujol) 2846,
5-Formyl-2-tri¯uoromethylbenzimidazole (19). 5-Cyano-
2-tri¯uoromethylbenzimidazole (196 mg, 0.93 mmol)
was re¯uxed with HCOOH (14 mL), H2O (5 mL) and
Ni±Al (0.9 g) catalyst for 6 h. The hot reaction mixture
was ®ltered through Celite and concentrated in vacuo.
The solution was basi®ed with 2 N NaOH. The solution
was extracted with ethyl acetate, dried (anhyd Na2SO4)
and concentrated in vacuo to give the crude aldehyde.
Puri®cation was achieved by column chromatography.
Elution with (90:10) hexanes:ethyl acetate gave 41% of
pure white compound: mp 178±179 ꢀC; IR (KBr) 3210,
1
1
2737, 1699, 1552, 1328, 1187, 986; H NMR (DMSO-
2728, 1692, 1290, 1109, 816; H NMR (DMSO-d6+3
d6+3 drops CF3COOH) d 7.87±7.95 (m, 2H), 8.34 (s, 1H),
10.11 (s, 1H); 13C NMR (DMSO-d6+3 drops CF3
COOH) d 106.6, 112.4, 116.7, 118.1, 121.9, 123.8, 124.0,
133.0, 192.7; HRMS (EI) calcd for C9H5N2OF3 m/z
214.0354, found 214.0348.
drops CF3COOH) d 3.31 (s, 3H), 3.44 (t, 2H), 3.86 (t,
2H), 7.79±8.09 (m, 2H), 8.36 (s, 1H), 10.16 (s, 1H); 13C
NMR (DMSO-d6+3 drops CF3COOH) d 27.7, 58.3,
68.1, 114.9, 116.9, 125.9, 131.6, 133.9, 135.2, 155.8,
192.4; anal. calcd for C11H12N2O2 1/4 H2O: C, 64.69; H,
5.92; N, 13.71; found: C, 64.76; H, 6.02; N, 13.60.
.
5-Formyl-2-hydroxymethylbenzimidazole (20). Ni±Al
(1.2 g) was added to a solution of 5-cyano-2-hydro-
xymethylbenzimidazole (0.17 g, 0.97mmol) in formic acid
(17 mL) and water (1.6mL). The reaction mixture was
heated at 95 ꢀC for 5 h. The hot mixture was ®ltered
through a bed of Celite and the reaction ¯ask and Celite
bed rinsed with water and then methanol. The combined
solution was concentrated in vacuo to dryness. After
addition of water to this residue, a white precipitate was
formed. The pH of this suspension was adjusted to 9 by
dropwise addition of 2 N NaOH. The product was
obtained by extraction with ethyl acetate. The ethyl acetate
extract was dried (anhyd Na2SO4) and concentrated in
vacuo to give 0.11g (0.63 mmol) of a pure white product
(61% yield): mp 189±190 ꢀC; IR (KBr) 3309, 2924, 1674,
5-Formyl-2-(2-acetamidoethyl)benzimidazole (23). 5-
Cyano-2-(2-acetamidoethyl)benzimidazole (94.6 mg, 0.41
mmol) was re¯uxed in HCOOH (5.9 mL), water
(0.5 mL) and Ni±Al (0.4 g) catalyst for 6 h. The mixture
was ®ltered hot through Celite bed and the bed washed
with methanol. The washings were concentrated in vacuo.
The concentrate was basi®ed with 2 N NaOH to pH 9 and
extracted with ethyl acetate. The ethyl acetate layer was
dried (anhyd Na2SO4) and puri®ed by ¯ash column chro-
matography using 0±15% methanol/ethyl acetate to give
80% (70.10mg, 0.3 mmol) of pure white product: mp 220±
221 ꢀC; IR (KBr) 3222, 3047, 1656, 1575, 1439, 1289, 1110,
1058, 816; 1H NMR (DMSO-d6+3 drops CF3COOH) d
1.79 (s, 3H), 3.29 (t, 2H), 3.54±3.64 (q, J=6.7, 12.6 Hz
2H), 7.97±8.09 (m, 2H), 8.19 (t, 1H), 8.38 (s, 1H), 10.17
(s, 1H); 13C NMR (DMSO-d6+3 drops CF3COOH) d
22.7, 27.9, 36.8, 114.9, 116.9, 125.8, 131.7, 133.8, 135.3,
155.9, 170.1, 192.5; anal. calcd for C12H13N3O2: C,
62.32; H, 5.67; N, 18.17; found: C, 62.10; H, 5.7; N, 17.92.
1
1619, 1434, 1291, 1073, 808; H NMR (DMSO-d6+3
drops CF3COOH) d 5.07 (s, 2H), 7.94 (d, 1H, J=8.7 Hz),
8.04±8.09 (dd, 1H, J=1.3, 8.5 Hz), 8.32 (s, 1H), 10.4 (s,
1H); 13C NMR (DMSO-d6+3 drops CF3COOH) d 55.8,
115.0, 116.8, 126.0, 131.5, 133.9, 135.1, 192.3; HRMS (EI)
calcd for C9H8N2O2 m/z 176.0586, found 176.0586.
5-Cyano-2-hydroxybenzimidazole (25). 3,4-Diamino-
benzonitrile (0.55 g, 4.14 mmol) was heated with urea
(0.27 g, 4.5 mmol) in DMF (5 mL) for 6 h at 150 ꢀC. The
cooled solution was concentrated in vacuo to give a yellow
colored solid. The compound was puri®ed by column
chromatography. Elution with (70±100%) ethyl acetate/
hexanes gave 81% (0.53 g, 3.35mmol) of pure compound:
mp 162±163 ꢀC; IR (KBr) 3235, 2924, 2214, 1645, 1610,
1388, 1332; 1H NMR (DMSO-d6+3 drops CF3COOH) d
5-Formyl-2-[(N-benzoyl)aminomethyl]benzimidazole (21).
5-Cyano-2-[(N-benzoyl)aminomethyl]benzimidazole (0.33
g, 1.2 mmol) was dissolved in formic acid (18 mL).
Water (5 mL) and Ni±Al (1.14 g) were added. The mix-
ture was heated under nitrogen at 95 ꢀC for 6 h and
while hot passed through a Celite bed. The bed was
washed with methanol and the washings concentrated in
vacuo. The pH was adjusted with 2 N NaOH to 9.0 and

1380
M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
7.05±7.09 (d, 1H, J=8 Hz), 7.31±7.4 (m, 2H); 13C NMR
(DMSO-d6+3 drops CF3COOH) d 109.34, 111.61,
112.39, 118.13, 120.03, 126.01, 130.20, 133.94; HRMS
(EI) calcd for C8H5N3O m/z 159.0433, found 159.0431.
yield (0.15 g, 0.76 mmol) of pure white product: mp 122±
123 ꢀC; IR (KBr) 2875, 2225, 1624, 1544, 1454, 1288, 1215,
1106, 824; H NMR (DMSO-d6+3 drops CF3COOH) d
3.29 (s, 3H), 3.42 (t, 2H), 3.84 (t, 2H), 7.87±8.00 (m, 2H),
8.40 (s, 1H); 13C NMR (DMSO-d6+3 drops CF3COOH) d
27.7, 58.2, 68.0, 106.8, 115.6, 118.3, 119.4, 129.0, 131.3,
134.3, 156.1; anal. calcd for C11H11N3O.1/2 H2O: C, 62.84;
H, 5.75; N, 19.97; found: C, 62.71; H, 5.70; N, 18.83.
1
5-Cyano-2-tri¯uoromethylbenzimidazole (26). 3,4-Diami-
nobenzonitrile (0.2 g, 1.5 mmol) was re¯uxed with tri-
¯uoroacetic acid (0.3 mL) for 6 h. The mixture was
neutralized with 2 N NaOH and extracted with ethyl
acetate. The ethyl acetate layer was dried (anhyd
Na2SO4) and concentrated in vacuo. Elution with (0±
10%) ethyl acetate/hexanes gave 88% of pure white
compound: mp 182±183 ꢀC; IR (KBr) 3086, 1560, 1402,
1316, 1204, 979, 824; 1H NMR (DMSO-d6+3 drops
CF3COOH) d 7.72±7.77 (1H, dd, J=1.5, 8.6 Hz), 7.88
(d, 1H, 8.42), 8.35 (s, 1H); 13C NMR (DMSO-d6+3 drops
CF3COOH) d 106.5, 106.6, 112.4, 117.3, 118.1, 119.4,
123.5, 123.8, 127.5; anal. calcd for C9H4N3F3: C, 51.20;
H, 1.91; N, 19.90; found: C, 51.45; H, 2.03; N, 19.68.
5-Cyano-2-(2-aminoethyl)benzimidazole (30). A solution
of 3,4-diaminobenzonitrile (1.0 g, 7.52 mmol) and b-ala-
nine (1.0 g, 11.3 mmol) in HCl (8 mL, 6 N) was re¯uxed
for 24 h. At this time the diamine was barely detectable
by TLC. After adjusting the pH to 7.0 with 2 N NaOH,
the compound was directly loaded on the column. Elu-
tion with (75±100%) ethyl acetate/hexanes and further
with 1±25% methanol/ethyl acetate gave 46% (0.65 g,
3.49 mmol) of pure yellow compound: mp 105±106 ꢀC;
IR (KBr) 3438, 2855, 2734, 2212, 1625, 1569, 1483,
1
1457, 1385, 1222, 1153; H NMR (CD3OD) d 3.09±3.13
5-Cyano-2-hydroxymethylbenzimidazole (27). 3,4-Diami-
nobenzonitrile (0.2 g, 1.50 mmol) was heated with gly-
colic acid (0.18 g, 2.37 mmol) in HCl (1.5 mL of 4 N) for
2 h. After neutralization with 2 M sodium carbonate the
product was extracted into ethyl acetate, dried (anhyd
Na2SO4) and concentrated in vacuo to give a white
colored solid. The compound was puri®ed by column
chromatography. Elution with (75±100%) ethyl acetate/
hexanes gave 66% (133mg, 0.76mmol) of pure com-
pound: mp 172±173 ꢀC; IR (Nujol) 3350, 2928, 2219,
(m, 4H), 7.49±7.66 (m, 2H), 7.91 (d, 1H, J=1.5 Hz); 13C
NMR d 33.13, 41.0, 106.3, 116.6, 121.0, 121.2, 127.1,
140.4, 142.6, 158.8; HRMS (EI) calcd for C10H10N4
m/z186.0905, found 186.0906.
5-Formyl-2-hydroxybenzimidazole (31). 5-Cyano-2-
hydroxybenzimidazole (0.4 g, 2.52 mmol) was dissolved
in formic acid (30 mL) and water (8 mL). Ni±Al (2.26 g)
catalyst was added to it. The mixture was heated under
nitrogen at 95 ꢀC for 10 h and passed through a Celite
bed while hot. The bed was washed with methanol and
the ®ltrate concentrated in vacuo. The pH is adjusted
with 2 N NaOH to 9.0 and extracted with ethyl acetate.
The extractions were dried (anhyd Na2SO4), con-
centrated in vacuo, and the residue puri®ed over column
chromatography using (65±90%) ethyl acetate/hexanes
to give 55% (224 mg, 1.38 mmol) of pure white product:
mp 155±156 ꢀC; IR (KBr) 3184, 1677, 1630, 1477, 1388,
1
1621, 1536, 1302, 1217, 1036, 815; H NMR (DMSO-
d6+3 drops CF3COOH) d 4.88 (s, 2H), 7.52±7.57 (dd, 1H,
J=1.5, 8.4 Hz), 7.69 (d, 1H, J=8.4 Hz), 7.95 (s, 1H); 13C
NMR (DMSO-d6+3 drops CF3COOH) d 55.8, 108.0,
115.7, 118.6, 119.5, 129.0, 131.5, 134.4; HRMS (EI) calcd
for C9H7N3O (m/z)173.0590, found 173.0590.
5-Cyano-2-[(N-benzoyl)aminomethyl]benzimidazole (28).
3,4-Diaminobenzonitrile (250 mg, 1.88 mmol) was ®nely
ground in a mortar with (0.34 g, 1.92 mmol) hippuric
acid and intimately mixed and then carefully fused for
3 h. The temperature was then raised to 160 ꢀC when
water was evolved as bubbles. After cooling the glassy
mass was dissolved in ethyl acetate and puri®ed by col-
umn chromatography. Ethyl acetate/hexanes (50±100%)
yielded (0.29 g, 1.05mmol) 86% of pale bu€ colored
compound: mp 131±132 ꢀC; IR (Nujol) 2826, 2221, 1635,
1313, 1019, 814; 1H NMR (DMSO-d6+3 drops CF3
COOH) d 4.97 (d, 2H, J=5 Hz), 7.52±7.61 (m, 3H), 7.91±
7.99 (m, 4H), 8.37 (s, 1H); 13C NMR (DMSO-d6+3 drops
CF3COOH) d 36.8, 107.7, 115.7, 118.8, 119.7, 127.8,
128.6, 128.6, 128.7, 128.8, 132.2, 132.2, 133.2, 135.0,156.1,
167.5; HRMS (EI) calcd for C16 H12 N4 O m/z 276.1011,
found 276.1012.
1
1328, 1282; H NMR (DMSO-d6+3 drops CF3COOH)
d 7.09±7.13 (d, 1H, J=8.3 Hz), 7.40 (s, 1H), 7.55±7.60
(d, 1H, J=8.3 Hz), 10.15 (s, 1H); 13C NMR (DMSO-
d6+3 drops CF3COOH) d 108.2, 108.8, 125.5, 130.1,
130.4, 135.7, 155.7, 192.2; HRMS (EI) calcd for
C8H6N2 O2 m/z 162.0429, found 162.0430.
5-Cyano-2-(2-acetamidoethyl)benzimidazole (32). 5-
Cyano-2-(2-aminoethyl)benzimidazole
(0.4 g,
2.15
mmol) was re¯uxed in dry THF (10 mL), acetic anhy-
dride (0.4 mL) and triethylamine (0.4 mL) for 4 h. The
mixture was neutralized with 2 N NaOH and extracted
with ethyl acetate to give the crude acetamide. The
acetamide was puri®ed using ¯ash column chromato-
graphy. Elution with (0±15%) methanol/ethyl acetate
gave 70% of pure white compound: mp 218±219 ꢀC; IR
(KBr) 3233, 3053, 2225, 1653, 1573, 1442, 1374, 1303,
1058, 815; 1H NMR (DMSO-d6+3 drops CF3COOH) d
1.78 (s, 3H), 3.26 (t, 2H), 3.53±3.62 (m, 2H), 7.89±7.95
(dd, 1H, J=1.4, 8.5 Hz) 8.00 (d, J=8.2 Hz, 1H), 8.45
(d, 1H, J=1.4 Hz); 13C NMR (DMSO-d6+3 drops
CF3COOH) d 22.6, 27.8, 36.7, 108.1, 115.5, 118.6,
119.3, 129.0, 131.3, 134.3, 156.2, 170.2; anal. calcd for
C12H12N4O.1/4H2O: C, 61.92; H, 5.41; N, 24.07; found:
C, 62.60; H, 5.41; N, 23.82.
5-Cyano-2-(2-methoxyethyl)benzimidazole (29). 3,4-Dia-
minobenzonitrile (0.2 g, 1.5 mmol) was re¯uxed with 2-
methoxy propionic acid (0.22 mL, 2.25 mmol) in 4 N
HCl (2.5 mL) for 7 h. After neutralization with 2 N
NaOH the mixture was extracted with ethyl acetate and
the extracts dried (anhyd Na2SO4) and concentrated in
vacuo. The product was puri®ed on column chromato-
graphy using (70±100%) ethyl acetate/hexanes to give 50%

M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
1381
Topoisomerase I-mediated DNA cleavage assays
pathlength was used in all the absorbance studies.
Absorbance versus temperature pro®les were acquired
at 260 nm in bu€er containing 10 mM sodium cacody-
late (pH 7.0), 25 mM NaCl, and 0.1 mM disodium
EDTA. The temperature was raised in 0.5 ꢀC increments
and the samples were allowed to equilibrate for 1 min at
each temperature setting prior to acquisition of the
absorbance reading (averaged over a 5 s period). For
each optically detected transition, the melting tempera-
ture (Tm) was determined as described previously.29,30
The DNA concentration was 10 mM in base pair, while
the ligand concentrations ranged from 0 to 20 mM.
Human topo I was expressed in E. coli and isolated as a
recombinant protein using a T7 expression system as
previously described.32 DNA topoisomerase I was pur-
i®ed from calf thymus gland as reported previously.33
Plasmid YepG was puri®ed using the QIAGEN Plasmid
MAXI kit. DNA end labeling was performed by
digesting YEpG with BamH I, followed by end-®lling
using Klenow polymerase. The topo I-mediated clea-
vage assay was performed by incubating topo I with
labeled DNA in the presence of di€erent concentrations
of drugs or solvent. Reaction was carried out at 23 ꢀC
for 15 min and subsequently stopped by addition of
SDS±proteinase K, followed by 37 ꢀC incubation for
1 h. DNA fragments were separated on an agarose gel
followed by autoradiography. Topo I-mediated DNA
cleavage values are reported as REC, Relative E€ective
Concentration, i.e. concentrations relative to 1, whose
value is arbitrarily assumed as 1.0. Compound 1 has
approximately 50% of the potency of Hoechst 33342 as
a topo I poison. Terbenzimidazoles tend to form dis-
tinct DNA cleavage bands in the presence of topo I.
Cleavage was calcd on the intensity of the strongest
speci®c band. The DNA cleavage pattern for camp-
tothecin is distinctly di€erent from terbenzimidazoles,
making comparisons of relative potency dicult. An
estimation of the relative potency of camptothecin was
made by comparing the doses of compound 5 and
camptothecin that induced fragmentation of approxi-
mately 10% of the labeled DNA.
Acknowledgments
Mass spectrometry was provided by the Washington
University Mass Spectrometry Resource, an NIH
Research Resource (Grant No. P41RR0954). This study
was supported by Grant CA 39662 from the National
Cancer Institute (L.F.L.), RPG CDD-98334 from the
American Cancer Society (D.S.P.), 00-64-CCR-S-0
from the New Jersey Commission on Cancer Research
(D.S.P.) and a fellowship grant from the Johnson &
Johnson Discovery Research Fund (E.J.L.).
References
1. Wang, J. C. Annu. Rev. Biochem. 1996, 65, 635±692.
2. Liu, L. F. Annu. Rev. Biochem. 1989, 58, 351±375.
3. Chen, A. Y.; Liu, L. F. Annu. Rev. Pharmacol. Toxicol.
1994, 34, 191±218.
4. Pommier, Y.; Pourquier, P.; Fan, Y.; Strumberg, D. Bio-
chim. Biophys. Acta 1998, 1400, 83±105.
MTT-microtiter plate tetrazolinium cytotoxicity assay
(RPMI 8402, CPT-K5, U937, U/CR cells)
5. Liu, L. F.; Wang, J. C. Proc. Natl. Acad. Sci. U.S.A. 1987,
84, 7024±7027.
6. Tsao, Y.-P.; Wu, H-Y.; Liu, L. F. Cell 1989, 56, 111±118.
7. Koo, H. S.; Classen, L.; Grossman, L.; Liu, L. Proc. Natl.
Acad. Sci. U.S.A. 1991, 88, 1212±1216.
8. Redinbo, M. R.; Stewart, L.; Kuhn, P.; Champoux, J. J.;
Hol, W. G. J. Science 1998, 279, 1504±1513.
9. Bjornsti, M. A.; Benedetti, P.; Viglianti, G. A.; Wang, J. C.
Cancer Res. 1989, 49, 6318.
10. Hertzberg, R. P.; Caranfa, M. J.; Hecht, S. M. Biochem-
istry 1989, 28, 4629±4638.
11. Svejstrup, J. Q.; Christainsen, K.; Gromoia, I. I.; Ander-
son, A. H.; Westergaard, O. J. Mol. Biol. 1991, 222, 669±678.
12. Chen, A. Y.; Yu, C.; Bodley, A. L.; Peng, L. F.; Liu, L. F.
Cancer Res. 1993, 53, 1332±1337.
The cytotoxicity was determined by the MTT
assay.34 36 The human lymphoblast RPMI 8402 and its
camptothecin-resistant variant cell line, CPT-K5, were
provided by Dr. Toshiwo Andoh (Aichi Cancer Center
Research Institute, Nagoya, Japan). Cells were main-
tained in suspension at 37 ꢀC and with 5% CO2 in
RPMI 1640 medium supplemented with 10% heat
inactivated fetal bovine serum, l-glutamine (2 mM),
penicillin (100 U/mL), and streptomycin (0.1 mg/mL).
The cytotoxicity assay was performed using a 96-well
microtiter plate with 2000 cells/well, in 200 ml of growth
medium. Cells were exposed continuously for 4 days to
di€erent drug concentrations, and MTT assays were
performed at the end of the fourth day. Each assay was
performed with a control that did not contain any drug.
All assays were performed at least twice in 4 replicate
wells.
13. Chen, A.; Yu, C.; Gatto, B.; Liu, L. F. Proc. Natl. Acad.
Sci. U.S.A. 1993, 90, 8131±8135.
14. Beerman, T. A.; McHugh, M.M; Sigmund, R.; Lown, J. W.;
Rao, K. E.; Bathni, Y. Biochim. Biophys. Acta 1992, 1131, 53±61.
15. Pilch, D. S.; Xu, Z.; Sun, Q.; LaVoie, E. J.; Liu, L. F.;
Breslauer, K. J. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 13565±
13570.
DNA binding anity measurements
16. Xu, Z.; Li, T.-K.; Kim, J. S.; LaVoie, E. J.; Breslauer, K.
J.; Liu, L. F.; Pilch, D. S. Biochemistry 1998, 37, 3558±3566.
17. Sun, Q.; Gatto, B.; Yu, C.; Liu, A.; Liu, L. F.; LaVoie, E.
J. J. Med. Chem. 1995, 38, 3638±3644.
18. Kim, J. S.; Yu, C.; Liu, A.; Liu, L. F.; LaVoie, E. J. J.
Med. Chem. 1997, 40, 2818±2824.
.
Poly(dA) poly(dT) was purchased from Amersham-
Pharmacia Biotech (Piscataway, NJ). The concentration
.
of poly(dA) poly(dT) was determined spectro-
photometrically using an extinction coecient at 260 nm
37
.
(E260) of 6000 L/mol base cm. All UV absorbance
experiments were conducted on an AVIV model 14DS
spectrophotometer equipped with a thermoelectrically
controlled cell holder. A quartz cuvette with a 1 cm
19. Tamura, H.; Kohchi, C.; Yamada, R.; Ikeda, T.; Koiwai,
O.; Patterson, E.; Keene, J. D.; Okada, K.; Kjeldsen, E.;
Nishikawa, K.; Andoh, A. Nuclei Acids Res. 1991, 19, 69±75.

1382
M. Rangarajan et al. / Bioorg. Med. Chem. 8 (2000) 1371±1382
20. Niwa, H.; Tsuneaki, H.; Yamada, K. A. Tett. Lett. 1981,
22, 4239±4240.
29. Marky, L. A.; Breslauer, K. J. Biopolymers 1987, 26, 1601±
1620.
21. Phillips, M. A. J. Chem. Soc., 1928, 2393.
22. Preston, P. N. Benzimidazoles. Benzimidazoles and Con-
generic Tricyclic Compounds, Part I; Preston, P. N., Ed.; John
Wiley and Sons: New York, 1981; pp 1±285.
23. Hughes, G. K.; Lions, F. J. J. Proc. Roy. Soc. N. S. Wales
1938, 71, 209±222.
24. Kim, J. S.; Sun, Q.; Yu, C.; Liu, A.; Liu, L. F.; LaVoie, E.
J. Bioorg. Med. Chem. 1998, 6, 163±172.
25. Crothers, D. M. Biopolymers 1971, 10, 2147±2160.
26. McGhee, J. D. Biopolymers 1976, 15, 1345±1375.
27. Clark, G. R.; Gray, E. J.; Neidle, S.; Li, Y.-H.; Leupin, W.
Biochemistry 1996, 35, 13745±13752.
30. Breslauer, K. J. Meth. Enzymol. 1995, 259, 221±242.
31. Rubin, E.; Pantazis, P.; Bharti, A.; Toppmeyer, D.;
Giovanella, B.; Kufe, D. J. Biol. Chem. 1994, 269, 2433±2439.
32. Liu, L. F.; Rowe, T. C.; Yang, Y. L.; Tewey, K. M.; Chen,
G. L. J. Biol. Chem. 1983, 258, 15365±15370.
33. Hsiang, Y. H.; Hertzberg, R.; Hecht, S.; Liu, L. F. J. Biol.
Chem. 1985, 260, 14873±14878.
34. Mosman, T. J. Immunol. Meth. 1983, 65, 55±63.
35. Carmichael, J.; Degra€, W. G.; Gazdar, A. F.; Minna, J.
D.; Mitchell, J. B. Cancer Res. 1987, 47, 936±942.
36. Denizot, F.; Lang, R. J. Immunol. Meth. 1986, 89, 271±
277.
28. Pilch, D. S.; Kirolos, M. A.; Breslauer, K. J. Biochemistry
1995, 34, 16107±16124.
37. Wells, R. D.; Larson, J. E.; Grant, R. C.; Shortle, B. E.;
Cantor, C. R. J. Mol. Biol. 1970, 54, 465±497.