Human telomerase inhibition by regioisomeric disubstituted amidoanthracene-9,10-diones

Article abstract of DOI:10.1021/jm981067o

Telomerase is an attractive target for the design of new anticancer drugs. We have previously described a series of 1,4- and 2,6-difunctionalized amidoanthracene-9,10-diones that inhibit human telomerase via stabilization of telomeric G-quadruplex structures. The present study details the preparation of three further, distinct series of regioisomeric difunctionalized amidoanthracene-9,10-diones substituted at the 1,5-, 1,8-, and 2,7-positions, respectively. Their in vitro cytotoxicity and Taq DNA polymerase and human telomerase inhibition properties are reported and compared with those of their 1,4- and 2,6-isomers. Potent telomerase inhibition ((tel)IC₅₀ values 1.3-17.3 μM) is exhibited within each isomeric series. In addition, biophysical and molecular modeling studies have been conducted to examine binding to the target G-quadruplex structure formed by the folding of telomeric DNA. These studies indicate that the isomeric diamidoanthracene-9,10-diones bind to the human telomeric G-quadruplex structure with a stoichiometry of 1:1. Plausible G-quadruplex-ligand complexes have been identified for each isomeric family, with three distinct modes of intercalative binding being proposed. The exact mode of intercalative binding is dictated by the positional placement of substituent side chains. Furthermore, in contrast to previous studies directed toward triplex DNA, it is evident that stringent control over positional attachment of substituents is not a necessity for effective telomerase inhibition.

Full text of DOI:10.1021/jm981067o

J . Med. Chem. 1998, 41, 4873-4884
4873
Hu m a n Telom er a se In h ibition by Regioisom er ic Disu bstitu ted
Am id oa n th r a cen e-9,10-d ion es
Philip J . Perry,† Anthony P. Reszka,† Alexis A. Wood,† Martin A. Read,† Sharon M. Gowan,‡
Harvinder S. Dosanjh,† J ohn O. Trent,†,§ Terence C. J enkins,†,| Lloyd R. Kelland,‡ and Stephen Neidle*,†
Cancer Research Campaign Biomolecular Structure Unit and Centre for Cancer Therapeutics, The Institute of Cancer
Research, 15 Cotswold Road, Sutton, Surrey SM2 5NG, U.K.
Received J uly 31, 1998
Telomerase is an attractive target for the design of new anticancer drugs. We have previously
described a series of 1,4- and 2,6-difunctionalized amidoanthracene-9,10-diones that inhibit
human telomerase via stabilization of telomeric G-quadruplex structures. The present study
details the preparation of three further, distinct series of regioisomeric difunctionalized
amidoanthracene-9,10-diones substituted at the 1,5-, 1,8-, and 2,7-positions, respectively. Their
in vitro cytotoxicity and Taq DNA polymerase and human telomerase inhibition properties
are reported and compared with those of their 1,4- and 2,6-isomers. Potent telomerase inhibition
(telIC50 values 1.3-17.3 µM) is exhibited within each isomeric series. In addition, biophysical
and molecular modeling studies have been conducted to examine binding to the target
G-quadruplex structure formed by the folding of telomeric DNA. These studies indicate that
the isomeric diamidoanthracene-9,10-diones bind to the human telomeric G-quadruplex
structure with a stoichiometry of 1:1. Plausible G-quadruplex-ligand complexes have been
identified for each isomeric family, with three distinct modes of intercalative binding being
proposed. The exact mode of intercalative binding is dictated by the positional placement of
substituent side chains. Furthermore, in contrast to previous studies directed toward triplex
DNA, it is evident that stringent control over positional attachment of substituents is not a
necessity for effective telomerase inhibition.
In tr od u ction
nite number of cell divisions. Telomerase activity has
been detected in 85-90% of all human cancers and may
be essential for cell immortality.9,10 Consequently,
telomerase has been proposed as a potentially highly
selective target for the development of a novel class of
antiproliferative agents.
There is increasing evidence that telomerase repre-
sents an attractive target for the rational design of new
anticancer agents in view of its central role in the
control of cellular proliferation.1,2 Human telomerase
is a specialized ribonucleoprotein with an intrinsic RNA
component that participates in the maintenance of
telomere length.3 Telomeres, the ends of eukaryotic
chromosomes, are characterized by simple G-rich re-
peating sequences of single-stranded DNA4 exemplified
by the human telomeric sequence 5′-TTAGGG.5 Telo-
meres play a vital role in maintaining the stability and
integrity of chromosomes and in protecting the ends of
chromosomes against aberrant recombination and po-
tential degradation by exonucleases.6,7 During succes-
sive rounds of cell division, the end-replication problem
results in telomere shortening and ultimately senes-
cence. As such, the loss of telomeric repeats after each
round of cell division has been likened to a ‘biological
clock’ limiting the proliferative life span of normal
somatic cells.8 Telomerase is able to utilize its endo-
genous RNA template to synthesize telomeric repeats3
and maintain stable telomere length through an indefi-
G-Rich sequences such as telomeric DNA can associ-
ate to form four-stranded inter- and intramolecular
guanine quadruplex structures.11-13 It is well-estab-
lished that the RNA template region of telomerase
requires a linear, nonfolded telomeric DNA primer in
order for telomere extension to take place. Thus, the
presence of a telomeric G-quadruplex structure results
in inhibition of the enzyme.14,15 We recently described
the human telomerase inhibition properties of a series
of 1,4- and 2,6-bis(ω-aminoalkanamido)anthracene-
9,10-diones (anthraquinones 1 and 2, Figure 1) and
proposed that their activity may be due to their ability
to bind to and stabilize G-quadruplex structures.16,17
This hypothesis was supported by evidence from UV and
NMR titration experiments.16 A porphyrin derivative,
tetrakis(N-methyl-4-pyridyl)porphine (TMPyP4), has
also been shown to bind to G-quadruplex structures18,19
and inhibit telomerase,18 possibly via an analogous
mechanism.
* To whom correspondence should be addressed. Tel: (+44)-181-
643-1675 (direct). Fax: (+44)-181-643-1675. E-mail: steve@
iris5.icr.ac.uk.
Previously, we have developed series of structurally
related 1,4- and 2,6-bis(ω-aminoalkanamido)anthracene-
9,10-diones as DNA-interactive agents. The positional
attachment of the ω-aminoalkanamido side chains has
been shown to profoundly influence their mode of DNA
binding. For example, 1,4-diamidoanthraquinones 1
have been shown to bind to duplex DNA by classical
† CRC Biomolecular Structure Unit.
‡ CRC Centre for Cancer Therapeutics.
§ Present address: Department of Medicine, Division of Hematology/
Oncology, University of Alabama at Birmingham, Birmingham, AL
35294.
| Present address: School of Chemical and Life Sciences, University
of Greenwich, Woolwich, London SE18 6PF, U.K.
10.1021/jm981067o CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/31/1998

4874 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Perry et al.
salt forms of all compounds were used for subsequent
biological and biophysical studies.
Stu d ies
Biologica l Stu d ies. Compounds 9-18a -e were
evaluated for in vitro cytotoxicity against three human
ovarian carcinoma cell lines (A2780, CH1, and SKOV-
3) using the sulforhodamine B (SRB) assay as described
previously.26 Results are presented in Table 1 as the
concentrations required to inhibit cell growth by 50%
(IC50 values). Prior to the evaluation of compounds in
a PCR-based telomerase assay, the agents were tested
for their ability to inhibit Taq polymerase. Compounds
were tested at concentrations of 10, 20, and 50 µM, and
semiquantitative estimates of Taq polymerase inhibition
are given (Table 1) for each concentration. The agents
were subsequently evaluated for their ability to inhibit
human telomerase in a modified cell-free TRAP assay
using extracts from the A2780 cell line.17 Agents were
tested at concentrations of 0.5, 1, 5, 10, 20, and 50 µM
up to the concentration that Taq polymerase inhibition
was first observed. Where appropriate, the concentra-
tions required to inhibit telomerase activity by 50%
(telIC50 values) are reported in Table 2.
Biop h ysica l Stu d ies. In a previous study16 we
demonstrated telomerase inhibition by a 2,6-diamido-
anthraquinone derivative and put forward evidence that
the mechanism of action involved binding to and result-
ant stabilization of G-quadruplex structures. This
hypothesis was supported by data from UV and NMR
titration experiments that were indicative of a binding
mode involving intercalation or base stacking. Isother-
mal titration calorimetry (ITC) has been used in the
present study to demonstrate and then quantitate the
interaction of two selected regioisomers (9e and 13d )
with the human telomeric sequence d[AG3(T2AG3)3]
under conditions shown to favor G-quadruplex forma-
tion.11,18 The resulting binding isotherms, after ap-
propriate correction, were analyzed to give the binding
enthalpy (∆H°), equilibrium binding constant (Kb), and
stoichiometry (n). The remaining thermodynamic pa-
rameters, ∆G° and T‚∆S°, were derived using standard
relationships and are presented in Table 3.
F igu r e 1. Structures of regioisomeric anthracene-9,10-diones
showing substitution at the 1,4- and 2,6-ring positions for 1
and 2, respectively.
intercalation,20 whereas their 2,6-disubstituted regio-
isomers 2 bind by a threading mode, in which the
functionalized side chains simultaneously occupy both
the DNA major and minor grooves, with intercalation
of the planar chromophore.21 Evidence supporting these
two distinct binding mechanisms has been provided by
both stopped-flow kinetics22 and molecular modeling
studies.20,21,23 Knowledge of these distinct modes of
binding has since been employed in the identification
of compounds that can preferentially bind to and
stabilize triplex DNA. For example, molecular model-
ing, DNA footprinting,24 and calorimetric25 techniques
have been used to show that 2,6-bis(aminopropionami-
do)anthraquinones can preferentially bind to and sta-
bilize triplex DNA, whereas their 1,4-regioisomers
demonstrate a propensity toward triplex disruption and
exhibit favored binding to duplex DNA.
In the present paper we describe the study of further
series of regioisomeric 1,5-, 1,8-, and 2,7-bis(aminopro-
pionamido)anthracene-9,10-diones. This has enabled us
to address the issue of how, and to what extent, the
position of side chain substituents affects telomerase
activity. Their in vitro cytotoxicity and Taq DNA
polymerase and human telomerase inhibition properties
are reported and compared with those of their 1,4- and
2,6-regioisomers. In addition, biophysical and molecular
modeling studies have been conducted in order to
examine binding to the proposed target G-quadruplex
structure formed by the folding of human telomeric
DNA.
Molecu la r Mod elin g Stu d ies. Qualitative molec-
ular modeling studies were conducted with the piperi-
dine derivatives 9a -13a to visualize how different
regioisomers might interact with and stabilize the
human telomeric G-quadruplex structure. The solution
NMR structure of the human telomeric repeat d[AG3(T2-
AG3)3] G-quadruplex11 provided an initial starting model.
An intercalation site was created between the diagonal
T2A loop and the G-quartet segment of the structure
(hereafter referred to as the 5′-AG step). Generated
structures for the isomeric piperidine derivatives 9a -
13a were minimized, and each, in turn, was docked into
the intercalation site. Different positions of the chro-
mophore were systematically evaluated for each regio-
isomer, with selections made on the basis of maximiza-
tion of π-electron overlap between the G-quartet, chro-
mophore, and adenine pair. Energy refinement of the
resulting G-quadruplex-ligand complexes using molec-
ular mechanics, dynamics, and subsequent mechanics
was carried out without the use of positional restraints.
Ch em istr y
Synthesis of the isomeric bis(aminopropionamido)-
anthracene-9,10-diones 9a -e-18a -e was accomplished
as described in Scheme 1. The free-base compounds
9-13 were only sparingly soluble in water; hence they
were converted into their acid addition salts in order to
improve their aqueous solubility.
1,8-Diaminoanthracene-9,10-dione (4) was prepared
via Gabriel synthesis from 1,8-dichloroanthracene-9,-
10-dione in quantitative yield (Scheme 2). 2,7-Diami-
noanthracene-9,10-dione (5) was obtained by Zinin
reduction of 2,7-dinitroanthracene-9,10-dione (19), pre-
pared, in turn, by nitration of anthrone (Scheme 3). The

Human Telomerase Inhibition
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4875
Sch em e 1. Synthesis of Isomeric Anthracene-9,10-dione Compoundsa
a
Reagents: (i) ClCH2CH2COCl/reflux/4 h; (ii) R2NH/NaI/EtOH/reflux/3 h; (iii) excess MeI/CHCl3/rt/24 h. 1,4-Isomers (9 and 14) and
2,6-isomers (12 and 17) were prepared as described previously (see refs 19 and 20).
Sch em e 2. Synthesis of 1,8-Diaminoanthracene-9,10-
dione 4a
lines (A2780, CH1, and SKOV-3). Acid addition salts
of the 1,4-, 1,5-, and 1,8-regioisomers (9, 10, and 11,
respectively) were generally more cytotoxic than the 2,6-
and 2,7-isomers, 12 and 13, respectively. The regio-
isomeric morpholine derivatives (9c-13c) were found
to be the least potent of the compounds studied. The
quaternary dimethiodide salts 14-18 exhibited broadly
similar cytotoxic activities. The 1,4- and 1,5-regioiso-
meric dimethiodide salts 14 and 15, respectively, were
generally less potent than their corresponding acid
addition salts, while the 1,8- and 2,6-isomers (11 and
16, and 12 and 17, respectively) were found to be
broadly similar. However, the 2,7-regioisomeric di-
methiodides 18 were more cytotoxic than their corre-
sponding acid addition salts.
a
Reagents: (i) phthalimide/AcONa/180-200 °C/1 h; (ii) concd
H2SO4/95 °C/45 min.
Sch em e 3. Synthesis of 2,7-Diaminoanthracene-9,10-
dione 5a
The majority of compounds examined were relatively
ineffective inhibitors of Taq polymerase, with little or
no inhibition of the enzyme evident at concentrations
up to 50 µM. However, three of the 1,4-disubstituted
acid addition salts (9a ,b,d ) exhibited significant inhibi-
tion at 20 µM with slight inhibition observed at 10 µM.
Inhibition of Taq polymerase by the 2,6-disubstituted
dimethiodides was also observed, with compounds show-
ing either total (17a ,c,e) or significant (17b) inhibition
at 20 µM.
The compounds were tested for their ability to inhibit
telomerase (telIC50 values) using the TRAP assay up to
the concentration at which Taq polymerase inhibition
was observed. The nucleotide ddGTP, which we have
previously shown17 to be a relatively potent inhibitor
of telomerase (telIC50 ) 8.6 µM), was used as a positive
control. A typical gel pattern arising from a TRAP
experiment is shown in Figure 2. Telomeric repeat
ladders are seen in the 0.01- and 0.04-µg protein positive
a
Reagents: (i) HNO3/0-5 °C/1-2 h/AcOH/reflux/2 h; (ii)
Na2S‚9H2O/NaOH/EtOH/reflux/6 h.
Resu lts
The amidoanthraquinones described here exhibit
moderate cytotoxicity, with IC50 values in the low-
micromolar range, against a panel of three ovarian cell

4876 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Perry et al.
Ta ble 1. In Vitro Cytotoxicity and Taq Polymerase Inhibition Data for Isomeric Anthracene-9,10-diones 9-18
IC50 (µM)a
Taq inhibitionb
compd
isomer
A2780
CH1
SKOV-3
10 µM
20 µM
50 µM
9a
9b
9c
9d
1,4
1,4
1,4
1,4
1,4
1,5
1,5
1,5
1,5
1,5
1,8
1,8
1,8
1,8
1,8
2,6
2,6
2,6
2,6
2,6
2,7
2,7
2,7
2,7
2,7
1,4
1,4
1,4
1,4
1,4
1,5
1,5
1,5
1,5
1,5
1,8
1,8
1,8
1,8
1,8
2,6
2,6
2,6
2,6
2,6
2,7
2,7
2,7
2,7
2,7
0.29
0.0025
3.7
0.016
0.2
0.26
0.019
1.95
0.04
0.26
0.42
0.36
>25
0.37
0.32
1.95
1.28
21.5
1.35
5.3
2.4
0.39
21.0
0.36
1.4
0.82
0.53
>25
0.62
0.62
2.35
1.2
>25
1.65
5.2
4.0
38
+
++
-
+
-
-
+
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
++
+++
-
+++
+++
-
++
-
+++
+++
+
9e
10a
10b
10c
10d
10e
11a
11b
11c
11d
11e
12a
12bc
12c
12d
12e
13a
13b
13c
13d
13e
14a
14b
14c
14d
14e
15a
15b
15c
15d
15e
16a
16b
16c
16d
16e
17a
17b
17c
17d
17e
18a
18b
18c
18d
18e
0.435
0.32
-
++
-
++
-
>25
0.35
0.32
0.54
0.295
-
++
+
-
-
+
+
+++
-
>25
-
0.33
0.64
1.3
-
-
-
-
5.9
49
-
-
39
>25
2.55
-
+++
-
14.0
1.8
3.15
3.15
1.8
7.5
4.4
4.6
4.3
4.5
1.65
2.0
6.3
2.1
>25
-
2.9
3.15
3.7
-
+++
+++
-
2.35
0.48
1.2
5.3
2.1
2.1
1.75
1.5
0.76
2.15
4.25
2.0
2.2
1.9
1.7
1.9
0.52
0.6
0.51
0.53
0.55
<1
0.56
1.9
-
-
2.3
-
+++
-
>100
4.15
7.2
-
-
-
-
++
-
32.5
37
8.5
11.5
>25
8.3
10.0
2.9
5.8
10.0
4.6
5.0
2.6
-
-
-
-
-
-
++
+
-
-
-
2.25
1.8
-
-
-
-
1.75
2.05
1.12
1.68
0.58
1.4
1.65
1.85
1.35
1.7
1.9
1.9
1.05
1.6
1.1
-
-
-
-
-
-
-
-
-
-
2.2
4.6
2.2
2.0
3.0
4.4
6.4
2.2
-
-
-
-
+++
++
+++
-
+++
+++
+++
-
2.25
2.1
+++
-
+++
-
++
-
++
++
0.18
0.48
0.52
0.44
0.33
4.7
2.95
4.0
-
-
1.3
1.13
-
2.0
-
a
b
Concentration required to inhibit cell growth by 50% relative to controls. Key: (+++) total, (++) significant, (+) slight, or (-) no
inhibition. c Administered as a fine suspension.
Ta ble 2. Telomerase Inhibition Data for Compounds 9a -e-18a -e
telomerase IC50 valuesa (µM) for substituents -NR2
compd
isomer
1-piperidinyl (a )
1-pyrrolidinyl (b)
4-morpholinyl (c)
NMe2 (d )
NEt2 (e)
9
10
11
12
13
14
15
16
17
18
1,4
1,5
1,8
2,6
2,7
1,4
1,5
1,8
2,6
2,7
nd
nd
nd
nd
1.8
2.0
5.0
8.8
8.2
nd
33.5
>50
>50
>50
.50
34.5
14.0
10.0
nd
nd
1.3
6.4
4.1
4.7
7.0
13.2
4.4
17.3
14.5
1.8
2.7
4.2
3.5
4.3
3.1
16.8
7.5
nd
2.3
3.7
4.5
3.1
11.1
8.6
7.8
nd
7.8
16.0
16.5
>20
a
Concentration required to inhibit telomerase activity by 50% relative to controls; nd, not determined.
control (lanes 2 and 3, respectively), while ladders were
not produced either in the absence of telomerase (lane
1) or when heat inactivation of the protein extract was
performed (lane 4). A concentration-dependent inhibi-
tion of telomerase by 13d is clearly evident in lanes 5-8
(20, 10, 5, and 1 µM, respectively). Similarly, ladders
for ddGTP are presented in lanes 9-12 at the same
range of concentrations which show it to be a less potent

Human Telomerase Inhibition
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4877
Ta ble 3. Thermodynamic Data for Binding of Compounds 9e
and 13d to G-Quadruplexa
Kb (× 104 ∆H° (kcal ∆G° c (kcal T‚∆S° d (kcal
compd nb
M-1
)
mol-1
)
mol-1
)
mol-1
)
9e
13d
1.2 7.9 ( 0.9
-5.5 ( 0.2
-6.7
-6.4
+1.2
-3.7
1.3 5.1 ( 1.1 -10.1 ( 2.0
a
Determined by ITC at 25 °C for the 22-mer sequence
d[AG3(T2AG3)3]. We estimate that all energies are within 5-10%
of calculated values as no correction was applied for the heats of
b
dissociation of the ligand dimers (see text). Number of molecules
bound per G-quadruplex; values ( e 0.2. c Free energy of binding
d
was calculated using ∆G° ) -RT ln Kb. The entropy term was
calculated using ∆G° ) ∆H° - T‚∆S°.
F igu r e 3. Calorimetric data for the titration of d[AG3(T2AG3)3]
with 9e determined at 25.00 ( 0.01 °C, showing exothermic
binding. The peaks in the upper panel represent the power
output associated with each injection of ligand into DNA as a
function of time. Integration of these peaks gives the corre-
sponding binding isotherm in the lower panel, for [ligand]:
[DNA] ratios of 0 f 3.25:1.
determined. ∆G° and T‚∆S° were derived using stan-
dard thermodynamic relationships. Figure 3 shows the
ITC data for the titration of the 22-mer human telomere
sequence with 9e. These data were subsequently cor-
rected for the heats of dilution associated with the
addition of ligand into buffer, but not for dissociation
of any dimeric species. As described previously,25 such
molecules have dimerization constants K2 in the order
of 3 × 103 M-1 at 25 °C, and we would thus expect there
to be <5% of free ligand remaining in solution at
completion of the titration for these analogous mol-
ecules. Fitting of the corrected data using a single-site
binding model allows stoichiometry (n), equilibrium
binding constant (Kb), and binding enthalpy (∆H°) to
be determined (Table 3).
F igu r e 2. Gel of a TRAP experiment showing telomerase
inhibition by 13d and ddGTP. Lane 1, lysis buffer negative
control; lane 2, positive control 0.01 µg of telomerase protein;
lane 3, positive control 0.04 µg of protein; lane 4, heat-
inactivated negative control; lanes 5-8, 13d at 20, 10, 5, and
1 µM, respectively; lanes 9-12, ddGTP at 20, 10, 5, and 1 µM,
respectively.
inhibitor than 13d . All of the compounds described,
with the exception of the morpholine derivatives 9-14c
and compound 18e, were shown to be effective inhibitors
of human telomerase (telIC50 values ranging from 1.3 to
17.3 µM). These values compare favorably with those
of previously described small-molecule inhibitors of
telomerase.18,27,28 The diminished activities exhibited
by the morpholine acid addition salts may, at least in
part, be due to their reduced basicity relative to the
other compounds described. Generally, the acid addi-
tion salts 9-13 were more active than the corresponding
quaternary dimethiodide derivatives 14-18. This may
be related to accessibility of the charge on the proto-
nated nitrogen.
Both of the regioisomers investigated (9e and 13d )
were shown to bind to the G-quadruplex structure with
a stoichiometry of 1:1 (within experimental error) and
with similar affinity. This is paralleled in the telo-
merase inhibition activities exhibited by the two com-
pounds (9e and 13d : 1.8 and 4.7 µM, respectively).
Examination of the thermodynamic data shows that
binding of both isomers is exothermic in nature. Fur-
ther, the binding enthalpies of both 9e and 13d (-5.5
and -10.1 kcal mol-1, respectively) are typical values
for DNA binding ligands.25,29
Final orientations from the averaged molecular dy-
namics structures for each regioisomer (9a -13a ) are
shown in Figure 4. Figure 4a shows a side view of the
2,6-disubstituted regioisomer 12a occupying a single
intercalation site at the 5′-AG step of the G-quadruplex
structure. Figure 4b shows the final orientation of
ITC was used to give direct measurement of the
binding enthalpy (∆H°), which then enabled equilibrium
binding constants (Kb), and stoichiometry (n) to be

4878 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Perry et al.
F igu r e 4. Comparison of binding conformations of the regioisomeric piperidine derivatives 9a -13a . (a) Side view of the 2,6-
regioisomer 12a bound in the 5′-AG step of the human telomeric G-quadruplex structure. (b) 2,6-Regioisomer 12a viewed from
above showing the orientation with respect to the adjacent G-quartet. Similar views are shown for 1,4- (c), 1,8- (d), 1,5- (e), and
2,7- (f) regioisomers (9a , 11a , 10a , and 13a , respectively).
compound 12a viewed from above, with respect to the
adjacent G-quartet. Regioisomers 9a (1,4), 11a (1,8),
10a (1,5), and 13a (2,7) are similarly shown in Figures
4c-f, respectively. Three distinct modes of intercalative
binding were established for the five different regioiso-
meric series. The 2,6-, 1,5-, and 2,7-disubstituted
regioisomers were found to adopt a favored threading
mode of intercalation in which the two substituent side
chains are positioned in opposite grooves of the G-
quadruplex structure (Figure 4b,e,f, respectively). Simi-
larly, the most favorable conformation for the 1,4-
regioisomer resulted in a threading mode of intercalation,
but with the substituent side chains positioned in
adjacent grooves (Figure 4c). In contrast, the most
favorable conformation for the 1,8-isomer corresponds
to a classical mode of intercalation. In this example,
the planar chromophore was shown to overlap two of
the four guanines of the G-quartet, with both the
substituent side chains simultaneously occupying the
same groove (Figure 4d). The remaining two guanines
of the G-quartet are stabilized by stacking with the
adenine pair (not shown). The resultant stabilization

Human Telomerase Inhibition
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4879
afforded by these compounds is due to a combination of
maximization of π-stacking interactions in the interca-
lation site, electrostatic interactions between the phos-
phate backbones and the positively charged amine side
chains, and nonbonded van der Waals interactions in
the G-quartet grooves with side chain atoms.
NMR and biophysical studies.18,19,29 G-Quadruplexes
are known to be extremely stable structures due to
favorable π-π-stacking of the G-quartets and the pres-
ence of tightly bound potassium or sodium ions. Thus,
intercalation between G-quartet planes is unfavorable
due to the disruption this would cause to the quadruplex
structure. The molecular modeling studies indicate that
a number of distinct intercalative modes of binding to
this single site within the G-quadruplex structure are
possible. These distinct binding modes are dictated by
the positional placement of substituent side chains for
each isomer, with the exact conformation adopted by
the chromophore for any one mode of binding being a
result of the maximization of π-π-interactions within
the intercalation site. Direct comparison of calculated
binding energies for individual G-quadruplex-ligand
complexes is inappropriate in the absence of detailed
simulations with explicit solvent. However, qualitative
evaluation of molecular dynamic trajectories indicates
that the diagonal T2A loop region adjacent to the
intercalation site is the most flexible region of the
structure, resulting in high mobility during molecular
dynamics simulations. This feature, combined with the
relatively large aromatic plane of the G-quartet, ac-
counts for the ability of the G-quadruplex structure to
associate with these regioisomeric anthraquinones.
In contrast to previous studies22-25 directed toward
triplex DNA in which the positional attachment of
substituent side chains was shown to profoundly influ-
ence the mode of DNA binding and resultant stabiliza-
tion, the present study indicates that a stable G-qua-
druplex-ligand complex is feasible for each regioisomeric
series. Furthermore, the finding that potent telomerase
inhibition (telIC50 values of 1.3-3.7 µM) was observed
within each isomeric series is suggestive of a role for
G-quadruplex involvement in the inhibitory activity
displayed by these compounds. This insensitivity to
positional isomerism is a consequence of the four
grooves present in the G-quadruplex structure as op-
posed to the two and three grooves in duplex and triplex
DNA, respectively. We therefore conclude that the
positional placement of substituents on the anthraquino-
ne chromophore is of less importance than the nature
of the substituents.
Discu ssion
The amidoanthraquinones described here showed only
moderate cytotoxicity against a panel of three ovarian
cell lines (Table 1) and were typically 2-3 orders of
magnitude less potent than the clinically established
quinones doxorubicin and mitoxantrone.17 If telomerase
inhibitors such as those described here are to have
application as antitumor agents, with telomerase inhi-
bition in tumor cells leading to attrition of telomere
length and consequent senescence, then prolonged
administration will most likely be necessary as a
significant number of rounds of cell division may
ultimately be required to produce such an effect. Con-
sequently, to obtain a useful comparative therapeutic
index (TI), where TI ) IC50/telIC50, conventional cyto-
toxicity should be minimized with respect to telomerase
inhibition. For the compounds examined, Tables 1 and
2 indicate that only the 1,4-isomer 14e has a TI value
greater than 1 (i.e., TI ) 1.4) with respect to the A2780
cell line used for the TRAP assay. Interestingly, and
without exception, compounds 9-18 were consistently
less cytotoxic toward SKOV-3 cells than the A2780 cell
line (Table 1). Furthermore, approximately half of the
agents evaluated were an order of magnitude less
cytotoxic to SKOV-3 cells. In contrast, cytotoxicity
levels toward CH1 cells were more comparable to those
in A2780. Therefore, compound 14e has a significantly
improved therapeutic index (TI > 8) with respect to
SKOV-3 cells and hence it would be of interest to further
investigate the telomerase inhibitory properties of this
compound in vitro using SKOV-3 cells, and in vivo with
mice bearing companion SKOV-3 xenografts. As noted
for other DNA interactive agents, the reduced levels of
cytotoxicity observed against SKOV-3 are probably due
to enhanced DNA-damage repair and tolerance mech-
anisms known to operate in this cell line compared to
the other ovarian cell lines in the panel examined.30
We have previously established a correlation between
the mode and affinity of binding to duplex DNA and
resultant biological properties for a series of analogous
diamidoanthracene-9,10-diones.20,21 The binding con-
stants to duplex DNA are comparable to those reported
here for binding to G-quadruplex DNA (Table 3).
Sheardy and co-workers recently reported,19 using Scat-
chard plot analysis, the binding constant for the inter-
action of TMPyP4 with quadruplexed T4G4 as 2.7 × 107
M-1, a value approximately 350-500 times greater than
that for the ligands examined here. This result would
appear to be somewhat surprising given that the te-
lomerase inhibition of TMPyP4 (telIC50 ) 6.5 µM) was
shown18 to be very similar to those of the compounds
described in the present study. Further, our results are
more in keeping with those reported for the binding of
ethidium bromide to quadruplexed T4G4 and porphyrin
binding to a number of G-quadruplexes, as determined
by ITC.29,31
The validity of telomerase as an anticancer target has
recently been questioned.32 However, it is now apparent
that the initial conclusions based on telomerase ‘knock-
out’ mice32 were premature, and longer-term experi-
ments33 have shown that telomerase plays an essential
role in cell proliferation. In addition, recent antisense
experiments with a 2′,5′-tetraadenylate-linked oligo-
nucleotide construct targeted toward the human telo-
merase RNA have provided further evidence validating
telomerase as an anticancer target.34 Recent evidence
also suggests that an alternative mechanism for telo-
mere maintenance may exist in certain yeast, ciliate,
and human tumor cell lines.35 Telomerase inhibitors
could well be of major use as selective anticancer drugs,
perhaps following conventional debulking therapy.
Therefore, compounds such as those described here that
inhibit telomerase indirectly by targeting telomeric
G-quadruplex structures may also have the potential
to impede non-telomerase-mediated pathways for te-
lomere maintenance. As such, the development of small
The intercalation site for the molecular modeling
studies was selected on the basis of evidence from both

4880 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Perry et al.
1.4 Hz, H-4,5), 8.71 (2H, dt, J ) 8.4 and 1.9 Hz, H-3,6), 8.83
(2H, t, J ) 1.9 Hz, H-1,8); MS (rel intensity) m/z 298 (100),
252 (75), 196 (22), 178 (23), 150 (67), 75 (34); calcd (M+)
298.0226, found 298.0240. Anal. (C14H6N2O6) C, H, N.
molecules that selectively bind to and stabilize G-
quadruplex structures, and that are noncytotoxic, may
enable the development of a new class of clinically useful
and highly selective drugs. Furthermore, the recent
finding that amidoanthracene-9,10-diones such as those
described here are nonmutagenic36 makes them of
particular appeal for potential long-term administration.
2,7-Dia m in oa n th r a cen e-9,10-d ion e, 5. To a stirred sus-
pension of 2,7-dinitroanthracene-9,10-dione (19) (9.4 g, 31.5
mmol) in ethanol (340 mL) was added a solution of sodium
sulfide nonahydrate (34.1 g, 142 mmol) and sodium hydroxide
(13.5 g, 338 mmol) in water (590 mL). The mixture was heated
at reflux for 6 h and left to stand overnight. The ethanol was
removed in vacuo and the residue cooled to 0-5 °C. The
resulting precipitate was collected by filtration, repeatedly
washed with water, and dried. Recrystallization from ethanol/
water afforded the product as an orange/red solid (7.35 g,
98%): mp 337-338 °C (lit.39 mp 330-332 °C); NMR δ (DMSO)
6.42 (4H, br s, NH2) 6.89 (2H, dd, J ) 8.5 and 1.5 Hz, H-3,6),
7.23 (2H, d, J ) 1.5 Hz, H-1,8), 7.84 (2H, d, J ) 8.5 Hz, H-4,5);
MS (rel intensity) m/z 239 (35), 176 (26), 154 (24), 136 (100),
106 (31); calcd ([M + 1]+) 239.0821, found 239.0830. Anal.
(C14H10N2O2) C, H, N.
1,5-B i s (3-c h lo r o p r o p i o n a m i d o )a n t h r a c e n e -9,10-
d ion e, 6. Gen er a l Acyla tion P r oced u r e. A suspension of
1,5-diaminoanthraquinone (3.0 g, 12.6 mmol) in 3-chloropro-
panoyl chloride (60 mL) was heated at reflux for 4 h, until
TLC indicated completion of reaction. After cooling to 0-5
°C, the mixture was filtered and the crude solid washed with
ether (3 × 50 mL). Recrystallization from DMF-EtOH (4:1
v/v) afforded chloroamide 6 (3.0 g, 57%) as yellow/brown
crystals: mp 280-281 °C; NMR δ (CDCl3) 3.04 (4H, t, J ) 6.4
Hz, COCH2), 3.95 (4H, t, J ) 6.4 Hz, CH2Cl), 7.82 (2H, t, J )
8.1 Hz, H-3,7), 8.08 (2H, dd, J ) 8.1 and 1.0 Hz, H-4,8), 9.16
(2H, dd, J ) 8.1 and 1.0 Hz, H-2,6), 12.40 (2H, s, NH); MS
(rel intensity) m/z 421 (100), 419 (74), 418 (20), 411 (25), 403
(23), 383 (52), 357 (26), 344 (25), 293 (26); calcd ([M + 1]+)
419.0565, found 419.0575. Anal. (C20H16N2O4Cl2) C, H, N, Cl.
1,8-B i s (3-c h lo r o p r o p i o n a m i d o )a n t h r a c e n e -9,10-
d ion e, 7. 1,8-Diaminoanthracene-9,10-dione (4) was treated
with 3-chloropropanoyl chloride according to the general
acylation procedure to give chloroamide 7 (3.7 g, 70%) as
orange crystals: mp 249-250 °C; NMR δ (CDCl3) 3.06 (4H, t,
J ) 6.5 Hz, COCH2), 3.97 (4H, t, J ) 6.5 Hz, CH2Cl), 7.81
(2H, t, J ) 8.5 Hz, H-3,6), 8.08 (2H, dd, J ) 8.5 and 1.0 Hz,
H-4,5), 9.18 (2H, dd, J ) 8.5 and 1.0 Hz, H-2,7), 12.18 (2H, s,
NH); MS (rel intensity) m/z 418 (21), 382 (21), 347 (13), 328
(34), 292 (25), 265 (55), 238 (90), 91 (18), 63 (46), 55 (100);
calcd ([M + 1]+) 419.0565, found 419.0550. Anal. (C20H16N2O4-
Cl2) C, H, N, Cl.
2,7-B i s (3-c h lo r o p r o p i o n a m i d o )a n t h r a c e n e -9,10-
d ion e, 8. 2,7-Diaminoanthracene-9,10-dione (5) was treated
with 3-chloropropanoyl chloride according to the general
acylation procedure to give chloroamide 8 (4.33 g, 82%) as a
yellow solid: mp 289-290 °C dec; NMR δ (DMSO) 2.92 (4H,
t, J ) 6.1 Hz, COCH2), 3.91 (4H, t, J ) 6.1 Hz, CH2Cl), 8.05
(2H, dd, J ) 8.5 and 2.0 Hz, H-3,6), 8.17 (2H, d, J ) 8.5 Hz,
H-4,5), 8.48 (2H, d, J ) 2.0 Hz, H-1,8), 10.72 (2H, s, NH); MS
(rel intensity) m/z 419 (97), 355 (78), 329 (100), 307 (77), 289
(76), 254 (73), 232 (67), 214 (27); calcd ([M + 1]+) 419.0565,
found 419.0550. Anal. (C20H16N2O4Cl2‚0.25H2O) C, H, N, Cl.
Exp er im en ta l Section
Syn th etic Ch em istr y. Melting points (mp) were recorded
on a Leica Galen III hot-stage melting point apparatus and
are uncorrected. NMR spectra were recorded at 250 MHz on
a Bruker AC250 spectrometer in either Me2SO-d6 or CDCl3
solution at 303 ( 1 K using Me4Si as internal standard.
Chemical shift assignments for the 1,5- and 1,8-isomers were
determined by 1H NMR titration experiments conducted with
Eu(fod)3 (0-2 mol equiv) and 7 in CDCl3. EI (70 eV), FAB,
and high-resolution accurate mass spectra were determined
by The School of Pharmacy (University of London, U.K.).
Elemental analyses were carried out by Medac Ltd. (Brunel
Science Center, Egham, Surrey, U.K.); results for elements
indicated by symbols were within (0.4% of theoretical values.
TLC was carried out on silica gel (Merck 60F-254) using
CHCl3-MeOH (0-20% MeOH) as eluent, with visualization
at 254 and 366 nm. Organic solutions were dried over sodium
sulfate. 1,4-Isomers (9a -e and 14a -e) and 2,6-isomers
(12a -e and 17a -e) were prepared using published proce-
dures.19,20
1,8-Dia m in oa n th r a cen e-9,10-d ion e, 4. A stirred mixture
of 1,8-dichloroanthracene-9,10-dione (41.6 g, 0.15 mol), ph-
thalimide (52.7 g, 0.385 mol), anhydrous sodium acetate (29.6
g, 0.361 mol), and nitrobenzene (77 mL) was heated to 180
°C. Quinoline (25 mL) and copper powder (300 mesh, 0.72 g)
were added, and the mixture was heated at 200 °C for 1 h.
The reaction mixture was allowed to cool and left to stand
overnight. The mixture was filtered, washed with nitroben-
zene (3 × 100 mL), ethanol (3 × 100 mL), hot water (3 × 200
mL), ethanol (2 × 100 mL), and ether (2 × 100 mL), and dried
to give the intermediate diphthalimide as a pale-yellow/orange
solid: mp > 360 °C (56.66 g, 76%). The crude solid (56.0 g)
was added to concentrated H2SO4 (400 mL) with stirring and
the mixture heated at 95 °C for 45 min. The reaction mixture
was cooled to 5 °C, and crushed ice (150 g) was slowly added.
The mixture was poured onto ice/water (1.5 L) with stirring,
and the resulting precipitate was collected by filtration,
washed with water until neutral, and dried in vacuo. Recrys-
tallization from ethanol afforded the product as red/purple
needles (27.0 g, 98%): mp 270-271 °C (lit.37 mp 269-270.5
°C); NMR δ (DMSO) 7.15 (2H, dd, J ) 8.5 and 1.4 Hz, H-2,7),
7.34 (2H, dd, J ) 7.4 and 1.4 Hz, H-4,5), 7.45 (2H, dd, J ) 8.5
and 7.4 Hz, H-3,6), 7.86 (4H, br s, NH2); MS (rel intensity)
m/z 238 (100), 210 (12), 181 (7), 154 (8), 119 (9), 91 (7), 77 (8);
calcd ([M + 1]+) 239.0821, found 239.0810. Anal. (C14H10N2O2)
C, H, N.
2,7-Din itr oa n th r a cen e-9,10-d ion e, 19. Anthrone (21.25
g, 0.109 mol) was added with stirring to a cooled solution of
fuming nitric acid (142 mL) at such a rate as to maintain a
reaction temperature of 5 °C. After completion of the addition
(ca. 1.5 h), the reaction mixture was allowed to reach ambient
temperature. The reaction mixture was added to glacial acetic
acid (430 mL) with cooling, lightly stoppered, and allowed to
stand at room temperature for 1 week. The resulting precipi-
tate was collected by filtration, washed with glacial acetic acid
(3 × 25 mL) and hexane (3 × 25 mL), and dried. The crude
solid was suspended in glacial acetic acid (4 L) and heated at
reflux until the evolution of nitrous fumes had ceased (ca. 2
h). The mixture was allowed to cool to room temperature and
left to stand for 48 h. The resulting precipitate was collected
by filtration, washed with glacial acetic acid (3 × 30 mL) and
hexane (3 × 30 mL), and dried to give a pale-yellow solid (10.34
g, 32%). Recrystallization from nitrobenzene/glacial acetic acid
(1:1 v/v) afforded a pure sample of 19: mp 290-291 °C (lit.38
mp 290-291 °C); NMR δ (DMSO) 8.48 (2H, dd, J ) 8.4 and
1,5-Bis(3-p ip er id in op r op ion a m id o)a n t h r a cen e-9,10-
d ion e, 10a . Gen er a l Am in olysis P r oced u r e. To a stirred
refluxing suspension of 1,5-bis(3-chloropropionamido)an-
thracene-9,10-dione (6) (1.00 g, 2.4 mmol) and NaI (0.3 g) in
EtOH (40 mL) was added dropwise piperidine (3.0 mL, 30
mmol) in EtOH (10 mL). The mixture was stirred at reflux
for 3 h, cooled to 0 °C, filtered, and washed with ether (50
mL). The crude solid was dissolved in hot CHCl3 (200 mL),
treated with decolorizing charcoal, and filtered. The filtrate
was washed with water, dried, and evaporated to yield an
orange solid. Recrystallization from DMF-EtOH (9:1 v/v)
afforded amide 10a (1.1 g, 89%) as orange needles: mp 214-
215 °C; NMR δ (CDCl3) 1.47 (4H, m, (CH2CH2)2CH2), 1.63 (8H,
m, N(CH2CH2)2), 2.52 (8H, t, J ) 5.0 Hz, N(CH2CH2)2), 2.72
(4H, m, COCH2CH2), 2.86 (4H, m, COCH2), 7.76 (2H, t, J )
8.0 Hz, H-3,7), 8.03 (2H, dd, J ) 8.0 and 1.0 Hz, H-4,8), 9.10

Human Telomerase Inhibition
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4881
(2H, dd, J ) 8.0 and 1.0 Hz, H-2,6), 12.31 (2H, s, NH); MS
(rel intensity) m/z 517 (100), 516 (7), 307 (43), 289 (40), 246
(100), 207 (22); calcd ([M + 1]+) 517.2815, found 517.2825.
Anal. (C30H36N4O4) C, H, N. Diacetate salt: mp 215 °C.
Dimethiodide salt 15a : mp 230 °C dec. Anal. (C32H42N4O4I2‚
H2O) C, H, N, I.
1,5-Bis(3-p yr r olid in op r op ion a m id o)a n t h r a cen e-9,10-
d ion e, 10b. Chloroamide 6 was treated with pyrrolidine
according to the general aminolysis procedure to give amide
10b (1.4 g, 80%) as yellow/orange needles: mp 194-195 °C;
NMR δ (CDCl3) 1.85 (8H, m, N(CH2CH2)2), 2.66 (8H, m, N(CH2-
CH2)2), 2.76 (4H, t, J ) 7.6 Hz, COCH2CH2), 2.96 (4H, t, J )
7.6 Hz, COCH2), 7.77 (2H, t, J ) 8.0 Hz, H-3,7), 8.04 (2H, d,
J ) 8.0 Hz, H-4,8), 9.13 (2H, d, J ) 8.0 Hz, H-2,6), 12.39 (2H,
s, NH); MS (rel intensity) m/z 489 (100), 488 (39); calcd ([M +
1]+) 489.2502, found 489.2512. Anal. (C28H32N4O4) C, H, N.
Diacetate salt: mp 135-136 °C. Dimethiodide salt 15b: mp
238 °C dec. Anal. (C30H38N4O4I2‚H2O) C, H, N, I.
1,5-Bis(3-m or p h olin op r op ion a m id o)a n th r a cen e-9,10-
d ion e, 10c. Chloroamide 6 was treated with morpholine
according to the general aminolysis procedure to give amide
10c (1.6 g, 85%) as yellow needles: mp 268 °C; NMR δ (CDCl3)
2.59 (8H, m, N(CH2CH2)2O), 2.72 (4H, t, J ) 6.0 Hz,
COCH2CH2), 2.89 (4H, t, J ) 6.0 Hz, COCH2), 3.76 (8H, t, J )
4.6 Hz, N(CH2CH2)2O), 7.79 (2H, t, J ) 8.0 Hz, H-3,7), 8.04
(2H, d, J ) 8.0 Hz, H-4,8), 9.11 (2H, d, J ) 8.0 Hz, H-2,6),
12.37 (2H, s, NH); MS (rel intensity) m/z 521 (100), 520 (30);
calcd ([M + 1]+) 521.2400, found 521.2410. Anal. (C28H32N4O6)
C, H, N. Diacetate salt: mp 266 °C. Dimethiodide salt 15c:
mp 245 °C dec. Anal. (C30H38N4O6I2) C, H, N.
1,5-Bis[3-(d im eth yla m in o)p r op ion a m id o]a n th r a cen e-
9,10-d ion e, 10d . Chloroamide 6 was treated with dimethyl-
amine (10 mL of a 5.6 M solution in EtOH) according to the
general aminolysis procedure to give amide 10d (1.30 g, 83%)
as yellow needles: mp 176-177 °C; NMR δ (CDCl3) 2.36 (12H,
s, CH3), 2.69 (4H, m, COCH2CH2), 2.84 (4H, m, COCH2), 7.77
(2H, t, J ) 8.0 Hz, H-3,7), 8.04 (2H, d, J ) 8.0 Hz, H-4,8), 9.14
(2H, d, J ) 8.0 Hz, H-2,6), 12.39 (2H, s, NH); MS (rel intensity)
m/z 437 (100), 436 (27), 307 (30), 289 (17); calcd ([M + 1]+)
437.2189, found 437.2179. Anal. (C24H28N4O4) C, H, N.
Diacetate salt: mp 142-143 °C. Dimethiodide salt 15d : mp
250 °C dec. Anal. (C26H34N4O4I2‚0.5H2O) C, H, N.
1,5-Bis[3-(d iet h yla m in o)p r op ion a m id o]a n t h r a cen e-
9,10-d ion e, 10e. Chloroamide 6 was treated with diethyl-
amine according to the general aminolysis procedure to give
amide 10e (1.28 g, 72%) as orange crystals: mp 174-175 °C;
NMR δ (CDCl3) 1.08 (12H, t, J ) 7.0 Hz, CH3), 2.65 (12H, m,
J ) 7.0 Hz, NCH2), 2.97 (4H, t, J ) 7.0 Hz, COCH2), 7.76 (2H,
t, J ) 8.0 Hz, H-3,7), 8.04 (2H, d, J ) 8.0 Hz, H-4,8), 9.13 (2H,
d, J ) 8.0 Hz, H-2,6), 12.33 (2H, s, NH); MS (rel intensity)
m/z 493 (100), 492 (36); calcd ([M + 1]+) 493.2815, found
493.2825. Anal. (C28H36N4O4‚0.5H2O) C, H, N. Diacetate
salt: mp 91 °C. Dimethiodide salt 15e: mp 235 °C dec. Anal.
(C30H42N4O4I2‚0.5H2O) C, H, N; I: calcd, 32.31; found, 33.01.
1,8-Bis(3-p ip er id in op r op ion a m id o)a n t h r a cen e-9,10-
d ion e, 11a . 1,8-Bis(3-chloropropionamido)anthracene-9,10-
dione (7) (1.50 g, 3.6 mmol) was treated with piperidine
according to the general aminolysis procedure to give amide
11a (1.64 g, 89%) as yellow needles: mp 183-184 °C; NMR δ
(CDCl3) 1.50 (4H, m, (CH2CH2)2CH2), 1.65 (8H, m, N(CH2CH2)2),
2.57 (8H, m, N(CH2CH2)2), 2.83 (4H, t, J ) 5.6 Hz, COCH2CH2),
2.88 (4H, t, J ) 5.6 Hz, COCH2), 7.77 (2H, t, J ) 8.0 Hz, H-3,6),
8.05 (2H, dd, J ) 8.0 and 1.0 Hz, H-4,5), 9.12 (2H, dd, J ) 8.0
and 1.0 Hz, H-2,7), 12.11 (2H, s, NH); MS (rel intensity) m/z
517 (31), 431 (14), 405 (9), 376 (32), 347 (14), 292 (10), 265 (8),
238 (17), 138 (100), 112 (32); calcd ([M + 1]+) 517.2815, found
517.2830. Anal. (C30H36N4O4‚1.2H2O) C, H, N. Diacetate
salt: mp 174-176 °C. Dimethiodide salt 16a : mp 244 °C dec.
Anal. (C32H42N4O4I2‚2.5H2O) C, N, I; H: calcd, 5.60; found,
5.03.
(CDCl3) 1.84 (8H, m, N(CH2CH2)2), 2.67 (8H, m, N(CH2CH2)2),
2.80 (4H, m, COCH2CH2), 2.99 (4H, m, COCH2), 7.77 (2H, t, J
) 8.0 Hz, H-3,6), 8.05 (2H, d, J ) 8.0 Hz, H-4,5), 9.14 (2H, d,
J ) 8.0 Hz, H-2,7), 12.17 (2H, s, NH); MS (rel intensity) m/z
489 (9), 417 (10), 391 (8), 362 (19), 347 (18), 292 (13), 238 (19),
155 (17), 124 (100); calcd ([M + 1]+) 489.2502, found 489.2520.
Anal. (C28H32N4O4) C, H, N. Diacetate salt: mp 179-180 °C.
Dimethiodide salt 16b: mp 228-230 °C dec. Anal. (C30H38N4-
O4I2‚2H2O) C, H, N, I.
1,8-Bis(3-m or p h olin op r op ion a m id o)a n th r a cen e-9,10-
d ion e, 11c. Chloroamide 7 was treated with morpholine
according to the general aminolysis procedure except the
mixture was heated at reflux for 24 h to give amide 11c (1.82
g, 97%) as an orange solid: mp 230 °C; NMR δ (CDCl3) 2.58
(8H, t, J ) 4.4 Hz, N(CH2CH2)2O), 2.75 (4H, t, J ) 6.6 Hz,
COCH2CH2), 2.88 (4H, t, J ) 6.6 Hz, COCH2), 3.74 (8H, t, J )
4.4 Hz, N(CH2CH2)2O), 7.76 (2H, t, J ) 7.8 Hz, H-3,6), 8.04
(2H, dd, J ) 7.8 and 1.0 Hz, H-4,5), 9.13 (2H, dd, J ) 7.8 and
1.0 Hz, H-2,7), 12.05 (2H, s, NH); MS (rel intensity) m/z 521
(100), 329 (12), 307 (45), 289 (27); calcd ([M + 1]+) 521.2400,
found 521.2420. Anal. (C28H32N4O6) C, H, N. Maleate salt:
mp 190-192 °C. Dimethiodide salt 16c: mp 232-233 °C dec.
Anal. (C30H38N4O6I2‚2H2O) C, H, N.
1,8-Bis[3-(d im eth yla m in o)p r op ion a m id o]a n th r a cen e-
9,10-d ion e, 11d . Chloroamide 7 was treated with dimethyl-
amine (10 mL of a 5.6 M solution in EtOH) according to the
general aminolysis procedure to give amide 11d (1.20 g, 76%)
as orange needles: mp 126 °C; NMR δ (CDCl3) 2.36 (12H, s,
CH3), 2.70 (4H, m, COCH2CH2), 2.80 (4H, m, COCH2), 7.75
(2H, t, J ) 8.0 Hz, H-3,6), 8.03 (2H, dd, J ) 8.0 and 1.0 Hz,
H-4,5), 9.13 (2H, dd, J ) 8.0 and 1.0 Hz, H-2,7), 12.19 (2H, s,
NH); MS (rel intensity) m/z 437 (100), 365 (15), 338 (15); calcd
([M + 1]+) 437.2189, found 437.2170. Anal. (C24H28N4O4) C,
H, N. Maleate salt: mp 188-189 °C. Dimethiodide salt 16d :
mp 263 °C dec. Anal. (C26H34N4O4I2‚H2O) C, H, N.
1,8-Bis[3-(d iet h yla m in o)p r op ion a m id o]a n t h r a cen e-
9,10-d ion e, 11e. Chloroamide 7 was treated with diethyl-
amine according to the general aminolysis procedure to give
amide 11e (1.58 g, 89%) as orange crystals: mp 175-176 °C;
NMR δ (CDCl3) 1.08 (12H, t, J ) 7.0 Hz, CH3), 2.66 (12H, m,
J ) 7.0 Hz, NCH2), 2.97 (4H, t, J ) 7.0 Hz, COCH2), 7.75 (2H,
t, J ) 8.0 Hz, H-3,6), 8.04 (2H, dd, J ) 8.0 and 1.0 Hz, H-4,5),
9.13 (2H, dd, J ) 8.0 and 1.0 Hz, H-2,7), 12.11 (2H, s, NH);
MS (rel intensity) m/z 493 (100), 307 (28), 289 (18); calcd ([M
+ 1]+) 493.2815, found 493.2800. Anal. (C28H36N4O4‚3.75H2O)
C, N; H: calcd, 7.83; found, 6.40. Maleate salt: mp 149-150
°C. Dimethiodidesalt 16e: mp 218-220°C. Anal. (C30H42N4O4I2‚
6H2O) C, N; H: calcd, 6.15; found, 4.83; I: calcd, 28.69; found,
28.24.
2,7-Bis(3-p ip er id in op r op ion a m id o)a n t h r a cen e-9,10-
d ion e, 13a . 2,7-Bis(3-chloropropionamido)anthracene-9,10-
dione (8) was treated with piperidine according to the general
aminolysis procedure to give amide 13a (1.85 g, 99%) as a pale-
yellow solid: mp 240 °C dec; NMR δ (DMSO) 1.40 (4H, m,
(CH2CH2)2CH2), 1.51 (8H, m, N(CH2CH2)2), 2.42 (8H, m,
N(CH2CH2)2), 2.56 (4H, t, J ) 5.8 Hz, COCH2CH2), 2.65 (4H,
t, J ) 5.8 Hz, COCH2), 8.04 (2H, dd, J ) 8.5 and 2.0 Hz, H-3,6),
8.15 (2H, d, J ) 8.5 Hz, H-4,5), 8.45 (2H, d, J ) 2.0 Hz, H-1,8),
10.80 (2H, s, NH); MS (rel intensity) m/z 517 (100), 329 (12),
307 (43), 289 (25), 259 (11); calcd ([M + 1]+) 517.2815, found
517.2840. Anal. (C30H36N4O4‚0.5H2O) C, H, N. Maleate
salt: mp 138-140 °C. Dimethiodide salt 18a : mp 192-193
°C dec. Anal. (C32H42N4O4I2) C, H, N.
2,7-Bis(3-p yr r olid in op r op ion a m id o)a n t h r a cen e-9,10-
d ion e, 13b. Chloroamide 8 was treated with pyrrolidine
according to the general aminolysis procedure to give amide
13b (1.68 g, 95%) as a pale-yellow solid: mp 232 °C dec; NMR
δ (DMSO) 1.69 (8H, m, N(CH2CH2)2), 2.49 (8H, m, N(CH2-
CH2)2), 2.56 (4H, t, J ) 6.5 Hz, COCH2CH2), 2.76 (4H, t, J )
6.5 Hz, COCH2), 8.04 (2H, d, J ) 8.5 Hz, H-3,6), 8.15 (2H, d,
J ) 8.5 Hz, H-4,5), 8.45 (2H, s, H-1,8), 10.65 (2H, s, NH); MS
(rel intensity) m/z 489 (100), 307 (20), 289 (12); calcd ([M +
1]+) 489.2502, found 489.2520. Anal. (C28H32N4O4‚0.5H2O) C,
1,8-Bis(3-p yr r olid in op r op ion a m id o)a n t h r a cen e-9,10-
d ion e, 11b. Chloroamide 7 was treated with pyrrolidine
according to the general aminolysis procedure to give amide
11b (1.07 g, 61%) as yellow needles: mp 184-186 °C; NMR δ

4882 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Perry et al.
H, N. Maleate salt: mp 172-174 °C dec. Dimethiodide salt
18b: mp 216-218 °C dec. Anal. (C30H38N4O4I2) C, H, N.
for 30 s, 50 °C for 30 s, and 72 °C for 1 min for 31 cycles.
Telomerase-extended PCR products in the presence or absence
of compounds were then determined either by electrophoretic
separation using 8% w/w acrylamide denaturing gels and
analysis by phosphorimaging or autoradiography or by har-
vesting on Whatman filters (25-mm glass microfiber) and
analysis by liquid scintillation counting.
3. Gr ow th In h ibition Assa y. Growth inhibition was
measured in three human ovarian carcinoma cell lines (A2780,
CH1, and SKOV-3) using the sulforhodamine B (SRB) assay
as described previously.26 Briefly, between 3000 and 6000 cells
were seeded into the wells of 96-well microtiter plates and
allowed to attach overnight. Agents (acid addition and
quaternary dimethiodide salts) were dissolved at 500 µM in
water and immediately added to wells in quadruplicate at final
concentrations of 0.05, 0.25, 1, 5, and 25 µM. Following an
incubation period of 96 h, remaining cells were fixed with ice-
cold 10% w/v trichloroacetic acid (30 min) and stained with
0.4% SRB in 1% v/v acetic acid (15 min). Mean absorbance at
540 nm for each drug concentration was expressed as a
percentage of the control untreated well absorbance, and IC50
values (concentration required to inhibit cell growth by 50%)
were determined for each agent.
4. Isoth er m a l Titr a tion Ca lor im etr y. Calorimetric
experiments were carried out using the isothermal titration
module of a Microcal MCS high-sensitivity titration calorim-
eter (Microcal Inc., Northampton, MA). Data acquisition and
analysis were performed using the Origin 4.10 software
package (supplied by the manufacturer). All experiments were
conducted at 25.00 ( 0.01 °C in aqueous buffer (10 mM Tris-
HCl, 150 mM KCl, pH 7.0). The human telomeric DNA
sequence d[AG3(T2AG3)3] was purchased from Oswel Ltd.
(Southampton, U.K.) and was used without further purifica-
tion. Binding isotherms were determined by serial 5-µL
injections of ligand solution (1.4 mM) at 400-s intervals from
a 250-µL syringe rotating at 400 rpm into DNA solution (37
µM). The raw calorimetric data were integrated and corrected
for sample concentration and cell volume affording the heat
output per injection. Control experiments were performed to
determine the heats of dilution for ligand titrated into buffer,
and the binding isotherm for each ligand-DNA interaction
was calculated by subtraction of the heat of dilution.25 The
binding enthalpy (∆H°), equilibrium binding constant (Kb), and
stoichiometry (n) for each ligand-DNA interaction were
determined from the corrected binding isotherms using a
nonlinear least-squares fitting procedure.
5. Molecu la r Mod elin g Stu d ies. The Macromodel 5.0
program40 using the AMBER* force field with continuum
solvation treatment (GB/SA model of water solvation, van der
Waals cutoff 8 Å, electrostatic cutoff 20 Å, dielectric constant
1) was used for all model building and calculations. The
solution NMR structure of the human telomeric repeat d[AG3(T2-
AG3)3] G-quadruplex11 was modeled to give an initial low-
energy starting model. Molecular mechanics energy minimi-
zation (1000-steps steepest descent with line searching and
3000-steps Polak Ribiere conjugate gradient with derivative
convergence of 0.05 kJ /Å mol) followed by dynamics (1.5-fs time
step, 40 ps at 300 K equilibrium, 100 ps at 300 K production
with time averaging of 100 sampled structures) and subse-
quent mechanics (minimization of time-averaged dynamics
structure) was used. An intercalation site was introduced
between the diagonal T2A loop and the G-quartet segment of
the structure (5′-AG step) by breaking the phosphate back-
bones and separating the structure while monitoring the
distance between the segments. The sugar-phosphate chains
were reconnected, and molecular mechanics energy minimiza-
tion (1000-steps steepest descent with line searching followed
by 1000-steps conjugate gradient) was used to relieve any
resulting steric distortion while retaining the G-quartet and
loop motifs with positional restraints.
2,7-Bis(3-m or p h olin op r op ion a m id o)a n th r a cen e-9,10-
d ion e, 13c. Chloroamide 8 was treated with morpholine
according to the general aminolysis procedure except the
mixture was heated at reflux for 5 h to give amide 13c (1.86
g, 99%) as a pale-yellow solid: mp 235 °C dec; NMR δ (DMSO)
2.43 (8H, m, N(CH2CH2)2O), 2.58-2.67 (8H, m, COCH2CH2),
3.58 (8H, m, N(CH2CH2)2O), 8.05 (2H, dd, J ) 8.5 and 2.1 Hz,
H-3,6), 8.15 (2H, d, J ) 8.5 Hz, H-4,5), 8.47 (2H, d, J ) 2.1
Hz, H-1,8), 10.71 (2H, s, NH); MS (rel intensity) m/z 521 (100),
329 (19), 307 (73), 289 (42); calcd ([M + 1]+) 521.2400, found
521.2420. Anal. (C28H32N4O6‚0.75H2O) C, N; H: calcd, 6.32;
found, 5.89. Maleate salt: mp 130-135 °C dec. Dimethiodide
salt 18c: mp 238 °C dec. Anal. (C30H38N4O6I2) C, H, N.
2,7-Bis[3-(d im eth yla m in o)p r op ion a m id o]a n th r a cen e-
9,10-d ion e, 13d . Chloroamide 8 (1.50 g, 3.6 mmol) was
treated with dimethylamine (10 mL of a 5.6 M solution in
EtOH) according to the general aminolysis procedure to give
amide 13d (1.48 g, 94%) as a pale-yellow solid: mp 202-203
°C; NMR δ (DMSO) 2.18 (12H, s, CH3), 2.55 (8H, m,
COCH2CH2), 8.05 (2H, d, J ) 9.0 Hz, H-3,6), 8.15 (2H, d, J )
9.0 Hz, H-4,5), 8.46 (2H, s, H-1,8), 10.68 (2H, s, NH); MS (rel
intensity) m/z 437 (100), 329 (11), 307 (42), 289 (22); calcd ([M
+ 1]+) 437.2189, found 437.2170. Anal. (C24H28N4O4‚1.25H2O)
C, H, N. Maleate salt: mp 167-169 °C dec. Dimethiodide
salt 18d : mp 223-224 °C. Anal. (C26H34N4O4I2‚0.75H2O) C,
H, N.
2,7-Bis[3-(d iet h yla m in o)p r op ion a m id o]a n t h r a cen e-
9,10-d ion e, 13e. Chloroamide 8 was treated with diethyl-
amine according to the general aminolysis procedure to give
amide 13e (1.56 g, 88%) as a pale-yellow solid: mp 215 °C;
NMR δ (DMSO) 0.98 (12H, t, J ) 7.1 Hz, CH3), 2.50 (12H, qt,
J ) 7.1 and 7.0 Hz, NCH2), 2.78 (4H, t, J ) 7.0 Hz, COCH2),
8.04 (2H, dd, J ) 8.5 and 2.1 Hz, H-3,6), 8.15 (2H, d, J ) 8.5
Hz, H-4,5), 8.45 (2H, d, J ) 2.1 Hz, H-1,8), 10.75 (2H, s, NH);
MS (rel intensity) m/z 493 (100), 477 (12); calcd ([M + 1]+)
493.2815, found 493.2800. Anal. (C28H36N4O4‚0.25H2O) C, H,
N. Maleate salt: mp 154-156 °C. Dimethiodide salt 18e: mp
196 °C. Anal. (C30H42N4O4I2) H, N; C: calcd, 46.40; found,
47.61.
Biologica l Stu d ies. 1. Ta q P olym er a se Assa y. Com-
pounds (acid addition and quaternary dimethiodide salts) were
tested at 10, 20, and 50 µM final concentrations in a PCR 50-
µL master mix containing 10 ng of pCI-neo mammalian
expression vector (Promega, Southampton, U.K.) and forward
(GGAGTTCCGCGTTACATAAC) and reverse (GTCTGCTC-
GAAGCATTAACC) primers (200 nmol) as described previ-
ously.17 The product of approximately 1 kb was visualized on
a 2% w/w agarose gel following amplification (30 cycles of 94
°C for 1 min, 55 °C for 1 min, and 72 °C for 2.5 min).
2. Mod ified Telom er ic Rep ea t Am p lifica tion P r otocol
(TRAP ) Assa y. The ability of agents to inhibit telomerase
in a cell-free assay was assessed with a modified TRAP assay
using extracts from exponentially growing A2780 human
ovarian carcinoma cells as described previously.17 The TRAP
assay was performed in two steps: (a) Telomerase-mediated
extension of the forward primer (TS: 5′-AATCCGTCGAGCA-
GAGTT; Oswel Ltd., Southampton, U.K.) was contained in a
40-µL reaction mix comprising TRAP buffer (20 mM Tris-HCl
(pH 8.3), 68 mM KCl, 1.5 mM MgCl2, 1 mM EGTA, 0.05% v/v
Tween 20), 0.05 µg of bovine serum albumin, 50 µM of each
deoxynucleotide triphosphate, 0.1 µg of TS primer, and 3 µCi
of [R-32P]dCTP (Amersham plc, U.K.). Protein (0.04 µg) was
then incubated with the reaction mix ( agent (acid addition
and quaternary dimethiodide salts) at final concentrations of
up to 50 µM for 20 min at 25 °C. A lysis buffer (no protein)
control, heat-inactivated protein control, and 50% protein (0.02
µg) control were included in each assay. (b) While heating at
80 °C in a PCR block of a thermal cycler (Hybaid, U.K.) for 5
min to inactivate telomerase activity, 0.1 µg of reverse CX
primer (3′-AATCCCATTCCCATTCCCATTCCC-5′) and 2 units
of Taq DNA polymerase (“red hot”; Advanced Biotechnologies)
were added. A three-step PCR was then performed: 94 °C
Models of the five regioisomeric piperidine derivatives were
built and energy-minimized using the same procedure as
described for the phosphate backbones. The minimized an-
thraquinones were, in turn, docked into the intercalation site,

Human Telomerase Inhibition
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4883
(15) Fletcher, T. M.; Sun, D.; Salazar, M.; Hurley, L. H. Effect of DNA
Secondary Structure on Human Telomerase Activity. Biochem-
istry 1998, 37, 5536-5541.
(16) Sun, D.; Thompson, B.; Cathers, B. E.; Salazar, M.; Kerwin, S.
M.; Trent, J . O.; J enkins, T. C.; Neidle, S.; Hurley, L. H.
Inhibition of Human Telomerase by a G-Quadruplex-Interactive
Compound. J . Med. Chem. 1997, 40, 2113-2116.
(17) Perry, P. J .; Gowan, S. M.; Reszka, A. P.; Polucci, P.; J enkins,
T. C.; Kelland, L. R.; Neidle, S. 1,4- and 2,6-Disubstituted
Amidoanthracene-9,10-dione Derivatives as Inhibitors of Human
Telomerase. J . Med. Chem. 1998, 41, 3253-3260.
(18) Wheelhouse, R.; Sun, D.; Han, H.; Han, F. X.; Hurley, L. H.
Cationic Porphyrins as Telomerase Inhibitors: the Interaction
of Tetra-(N-methyl-4-pyridyl)porphine with Quadruplex DNA.
J . Am. Chem. Soc. 1998, 120, 3261-3262.
(19) Anantha, N. V.; Azam, M.; Sheardy, R. D. Porphyrin Binding to
Quadruplexed T4G4. Biochemistry 1998, 37, 2709-2714.
(20) Collier, D. A.; Neidle, S. Synthesis, Molecular Modeling, DNA
Binding, and Antitumor Properties of Some Substituted Ami-
doanthraquinones. J . Med. Chem. 1988, 31, 847-857.
(21) Agbandje, M.; J enkins, T. C.; McKenna, R.; Reszka, A. P.; Neidle,
S. Anthracene-9,10-diones as Potential Anticancer Agents.
Synthesis, DNA-Binding, and Biological Studies on a Series of
2,6-Disubstituted Derivatives. J . Med. Chem. 1992, 35, 1418-
1429.
(22) Tanious, F. A.; J enkins, T. C.; Neidle, S.; Wilson, W. D.
Substituent Position Dictates the Intercalative DNA-Binding
Mode for Anthracene-9,10-dione Antitumor Drugs. Biochemistry
1992, 31, 11632-11640.
(23) Neidle, S.; J enkins, T. C. Molecular Modeling to Study DNA
Intercalation by Antitumor Drugs. Methods Enzymol. 1991, 203,
433-459.
(24) Fox, K. R.; Polucci, P.; J enkins, T. C.; Neidle, S. A Molecular
Anchor for Stabilizing Triple-Helical DNA. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 7887-7891.
and different positions of the chromophore were systematically
evaluated for each regioisomer. Selections were made on the
basis of maximization of π-electron overlap between the
G-quartet, the chromophore, and the adenine pair. Where
necessary, alterations to the conformation of the -(CH2)2- side
chain linkers were made to alleviate any intermolecular
clashes.
The resulting G-quadruplex-ligand complexes were mini-
mized using molecular mechanics (1000-steps steepest descent
with line searching and 3000-steps Polak Ribiere conjugate
gradient with derivative convergence of 0.05 kJ /Å mol) followed
by dynamics (1.5-fs time step, 100 ps at 300 K equilibrium,
100 ps at 300 K production with time averaging of 100 sampled
structures) and subsequent mechanics (minimization of time-
averaged dynamics structure). No positional restraints were
placed on any part of the model during these calculations.
Ack n ow led gm en t. This work was supported by the
Cancer Research Campaign and Institute of Cancer
Research. Paul Rogers is thanked for conducting cyto-
toxicity assays. We gratefully thank Heath Scientific for
their expert guidance and use of ITC instrumentation
for thermodynamic experiments.
Refer en ces
(1) (a) Morin, G. B. Is Telomerase a Universal Cancer Target? J .
Natl. Cancer Inst. 1995, 87, 859-861. (b) Hamilton, S. E.; Corey,
D. R. Telomerase: Anti-cancer Target or just a Fascinating
Enzyme? Chem. Biol. 1996, 3, 863-867. (c) Parkinson, E. K. Do
Telomerase Antagonists Represent a Novel Anti-cancer Strat-
egy? Br. J . Cancer 1996, 73, 1-4. (d) Raymond, E.; Sun, D.;
Chen, S.-F.; Windle, B.; von Hoff, D. D. Agents that Target
Telomerase and Telomeres. Curr. Opin. Biotechnol. 1996, 7,
583-591.
(25) Haq, I.; Ladbury, J . E.; Chowdhry, B. Z.; J enkins, T. C. Molecular
Anchoring of Duplex and Triplex DNA by Disubstituted An-
thracene-9,10-diones: Calorimetric, UV Melting and Competi-
tion Dialysis Studies. J . Am. Chem. Soc. 1996, 118, 10693-
10701.
(26) Kelland, L. R.; Abel, G.; McKeage, M. J .; J ones, M.; Goddard, P.
M.; Valenti, M.; Murrer, B. A.; Harrap, K. R. Preclinical
Antitumor Evaluation of Bis-acetato-amine-dichloro-cyclohexy-
lamine Platinum(IV): an Orally Active Platinum Drug. Cancer
Res. 1993, 53, 2581-2586.
(2) (a) Bodnar, A. G.; Ouellette, M.; Frolkis, M.; Holt, S. E.; Chiu,
C. P.; Morin, G. B.; Harley, C. B.; Shay, J . W.; Lichtsteiner, S.;
Wright, W. E. Extension of Life-span by Introduction of Telom-
erase into Normal Human Cells. Science 1998, 279, 349-352.
(b) Niida, H.; Matsumoto, T.; Satoh, H.; Shiwa, M.; Tokutake,
Y.; Furuichi, Y.; Shinkai, Y. Severe Growth Defect in Mouse
Cells Lacking the Telomerase RNA Component. Nature Genet.
1998, 19, 203-206.
(27) (a) Strahl, C.; Blackburn, E. H. Effects of Reverse Transcriptase
Inhibitors on Telomere Length and Telomerase Activity in Two
Immortalised Human Cell Lines. Mol. Cell. Biol. 1996, 16, 53-
65. (b) Yegorov, Y. E.; Chernov, D. N.; Akimov, S. S.; Bolsheva,
N. L.; Krayevsky, A. A.; Zelenin, A. V. Reverse Transcriptase
Inhibitors Suppress Telomerase Function and Induce Senescence-
like Processes in Cultured Mouse Fibroblasts. FEBS Lett. 1996,
389, 115-118. (c) Melana, S. M.; Holland, J . F.; Pogo, B. G.-T.
Inhibition of Cell Growth and Telomerase Activity of Breast
Cancer Cells In Vitro by 3′-Azido-3′deoxythymidine. Clin. Cancer
Res. 1944, 4, 693-696. (d) Fletcher, T. M.; Salazar, M.; Chen,
S.-F. Human Telomerase Inhibition by 7-Deaza-2′-deoxypurine
Nucleoside Triphosphates. Biochemistry 1996, 35, 15611-15617.
(e) Pai, R. B.; Pai, B.; Kukhanova, M.; Dutschman, G. E.; Guo,
X.; Cheng, Y.-C. Telomerase from Human Leukemia Cells:
Properties and Its Interaction with Deoxynucleoside Analogues.
Cancer Res. 1998, 58, 1909-1913.
(28) (a) Ku, W.-C.; Cheng, A.-J .; Wang, T.-C. V. Inhibition of
Telomerase Activity by PKC Inhibitors in Human Nasopharyn-
geal Cancer Cells in Culture. Biochem. Biophys. Res. Commun.
1997, 241, 730-736. (b) Yamakuchi, M.; Nakata, M.; Kawahara,
K.-I.; Kitajima, I.; Maruyama, I. New Quinolones, Ofloxacin and
Levofloxacin, Inhibit Telomerase Activity in Transitional Cell
Carcinoma Cell Lines. Cancer Lett. 1997, 119, 213-219.
(29) Guo, Q.; Lu, M.; Marky, L. A.; Kallenbach, N. R. Interaction of
the Dye Ethidium Bromide with DNA Containing Guanine
Repeats. Biochemistry 1992, 31, 2451-2455.
(3) (a) Morin, G. B. The Human Telomere Terminal Transferase
Activity is
a Ribonucleoprotein that Synthesises TTAGGG
Repeats. Cell 1989, 59, 521-529. (b) Feng, J .; Funk, W.; Wang,
S.-S.; Weinrich, S. L.; Avilion, A. A.; Chiu, C.-P.; Adams, R. R.;
Chang, E.; Allsopp, R. C.; Yu, J .; Le, S.; West, M. D.; Harley, C.
B.; Andrews, W. H.; Greider, C. W.; Villeponteau, B. The RNA
Component of Human Telomerase. Science 1995, 269, 1236-
1241.
(4) Blackburn, E. H. Structure and Function of Telomeres. Nature
1991, 350, 569-573.
(5) Meyne, J .; Ratliff, R. L.; Moyzis, R. K. Conservation of the
Human Telomere Sequence (TTAGGG)n among Vertebrates.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 7049-7053.
(6) Blackburn, E. H. Telomeres: No End in Sight. Cell 1994, 77,
621-623.
(7) Kruk, P. A.; Rampino, N. J .; Bohr, V. A. DNA Damage and
Repair in Telomeres: Relation to Aging. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 258-262.
(8) (a) Harley, C. B.; Futcher, A. B.; Greider, C. W. Telomeres
Shorten During Aging of Human Fibroblasts. Nature 1990, 345,
458-460. (b) Allsopp, R. C.; Harley, C. B. Evidence for a Critical
Telomere Length in Senescent Human Fibroblasts. Exp. Cell
Res. 1995, 219, 130-136.
(9) Kim, N. W.; Piatyszek, M. A.; Prowse, K. R.; Harley, C. B.; West,
M. D.; Ho, P. L. C.; Coviello, G. M.; Wright, W. E.; Weinrich, R.
L.; Shay, J . W. Specific Association of Human Telomerase
Activity with Immortal Cells and Cancer. Science 1994, 266,
2011-2015.
(10) Counter, C. M.; Hirte, H. W.; Bacchetti, S.; Harley, C. B.
Telomerase Activity in Human Ovarian Carcinoma. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 2900-2904.
(30) O’Neill, C. F.; Orr, R. M.; Kelland, L. R.; Harrap, K. R.
Comparison of Platinum Binding to DNA and Removal of Total
Platinum Adducts and Inter-Strand Cross-Links in three Human
Ovarian Carcinoma Cell Lines Sensitive and Resistant to
Cisplatin. Cell. Pharmacol. 1995, 2, 1-7.
(11) Wang, Y.; Patel, D. J . Solution Structure of the Human Telomeric
Repeat d[AG3(T2AG3)3] G-tetraplex. Structure 1993, 1, 263-282.
(12) Williamson, J . R. G-quartet Structures in Telomeric DNA. Annu.
Rev. Biophys. Biomol. Struct. 1994, 23, 703-730.
(13) Rhodes, D.; Giraldo, R. Telomere Structure and Function. Curr.
Opin. Struct. Biol. 1995, 5, 311-322.
(14) Zahler, A. M.; Williamson, J . R.; Cech, T. R.; Prescott, D. M.
Inhibition of Telomerase by G-quartet DNA Structures. Nature
1991, 350, 718-720.
(31) J enkins, T. C. Unpublished results.
(32) Blasco, M. A.; Lee, H.-W.; Hande, M. P.; Samper, E.; Lansdorp,
P. M.; DePinho, R. A.; Greider, C. W. Telomere Shortening and
Tumor Formation by Mouse Cells Lacking Telomerase RNA. Cell
1997, 91, 25-34.
(33) Lee, H.-W.; Blasco, M. A.; Gottlieb, G. J .; Horner, J . W.; Greider,
C. W.; DePinho, R. A. Essential Role of Mouse Telomerase in
Highly Proliferative Organs. Nature 1998, 392, 569-574.

4884 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Perry et al.
(34) Kondo, S.; Kondo, Y.; Li, G.; Silverman, R. H.; Cowell, J . K.
Targeted Therapy of Human Malignant Glioma in a Mouse
Model by 2-5A Antisense Directed Against Telomerase RNA.
Oncogene 1998, 16, 3323-3330.
Some Diphenyl and Triphenyl Derivatives of Anthracene and
Naphthalene. J . Org. Chem. 1972, 37, 1003-1011.
(38) Bo¨ernstein, E.; Schliewiensky, H.; Szczesny-Heyl, G. V. Fritzsches
Reagent: â-Dinitroanthraquinones. Chem. Ber. 1926, 59, 2813.
(39) Gubelmann, I.; Weiland, H. J .; Stallmann, O. Several New 4′-
Sulfo-ortho-benzoylbenzoic Acid Derivatives and the Corre-
sponding Anthraquinone Compounds. J . Am. Chem. Soc. 1931,
53, 1033-1036.
(35) Bryan, T. M.; Englezou, A.; Dalla-Pozza, L.; Dunham, M. A.;
Reddel, R. R. Evidence for an Alternative Mechanism for
Maintaining Telomere Length in Human Tumors and Tumor-
derived Cell Lines. Nature Med. 1997, 3, 1271-1274.
(36) Venitt, S.; Crofton-Sleigh, C.; Agbandje, M.; J enkins, T. C.;
Neidle, S. Anthracene-9,10-diones as Potential Anti-Cancer
Agents: Bacterial Mutation Studies of Amido-Substituted De-
rivatives Reveal an Unexpected Lack of Mutagenicity. J . Med.
Chem. 1998, 41, 3748-3752.
(40) Mohamadi, F.; Richards, N. G. J .; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W.
C. Macromodel - an Integrated Software System for Modeling
Organic Bioorganic Molecules using Molecular Mechanics. J .
Comput. Chem. 1990, 11, 440-467.
(37) House, H. O.; Koepsell, D. G.; Campbell, W. J . The Synthesis of
J M981067O