Tri-, tetra- and heptacyclic perylene analogues as new potential antineoplastic agents based on DNA telomerase inhibition

Article abstract of DOI:10.1016/j.bmc.2006.09.029

A recent approach in anticancer chemotherapy envisages telomerase as a potentially useful target. An attractive strategy deals with the development of compounds able to stabilize telomeric DNA in the G-quadruplex folded structure and, among them, a prominent position is found in the perylenes. With the aim to further investigate the role of drug structure, in view of possible pharmaceutical applications, we synthesized a series of compounds related to PIPER, a well-known perylene-based telomerase inhibitor. We modified the number of condensed aromatic rings and introduced different side chains to modulate drug protonation state and extent of self-aggregation. Effective telomerase inhibition was induced by heptacyclic analogues only, some showing a remarkably wide selectivity index with reference to inhibition of Taq polymerase. G-quadruplex stabilization was monitored by circular dichroism and melting experiments. Cell cytotoxicity measurements indicated a poor short-term cell killing ability for the best G-quartet binders. Besides the presence of a planar seven-condensed ring system, the introduction of a cyclic amine in the side chains critically affects the selectivity window.

Full text of DOI:10.1016/j.bmc.2006.09.029

Bioorganic & Medicinal Chemistry 15 (2007) 555–562
Tri-, tetra- and heptacyclic perylene analogues as new potential
antineoplastic agents based on DNA telomerase inhibition
Claudia Sissi,a,* Lorena Lucatello,a A. Paul Krapcho,b David J. Maloney,b
Matthew B. Boxer,b Maria V. Camarasa,c Gabriella Pezzoni,c
Ernesto Mentac and Manlio Palumboa
aDepartment of Pharmaceutical Sciences, University of Padova, Via Marzolo, 5-35131 Padova, Italy
bDepartment of Chemistry, University of Vermont, Burlington, VT 05405, USA
cCell Therapeutics Europe S.r.l., Via Ariosto 23, 20091 Bresso (MI), Italy
Received 20 June 2006; revised 12 September 2006; accepted 15 September 2006
Available online 10 October 2006
Abstract—A recent approach in anticancer chemotherapy envisages telomerase as a potentially useful target. An attractive strategy
deals with the development of compounds able to stabilize telomeric DNA in the G-quadruplex folded structure and, among them, a
prominent position is found in the perylenes. With the aim to further investigate the role of drug structure, in view of possible phar-
maceutical applications, we synthesized a series of compounds related to PIPER, a well-known perylene-based telomerase inhibitor.
We modified the number of condensed aromatic rings and introduced different side chains to modulate drug protonation state and
extent of self-aggregation. Effective telomerase inhibition was induced by heptacyclic analogues only, some showing a remarkably
wide selectivity index with reference to inhibition of Taq polymerase. G-quadruplex stabilization was monitored by circular dichro-
ism and melting experiments. Cell cytotoxicity measurements indicated a poor short-term cell killing ability for the best G-quartet
binders. Besides the presence of a planar seven-condensed ring system, the introduction of a cyclic amine in the side chains critically
affects the selectivity window.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
shorten below a critical length.2 While this occurs in
normal somatic cells, in the vast majority of cancer cells
the activation of a nuclear enzyme, called telomerase, is
observed.3,4 This enzyme is a ribonucleoprotein that
functions as a reverse transcriptase: using its RNA sub-
unit as template, it adds telomeric repeats to DNA, thus
preventing the shortening in telomere length. Therefore,
cancer cells acquire the capability to divide indefinitely
and, hence, become immortal. Impairing telomerase
activity would restore the chromosome shortening
process and thus produce cancer cell senescence and
death. Hence, the search for compounds interfering with
telomerase activity has become an attractive goal in
anticancer therapy.5,6
The search for novel chemotherapeutic approaches to
treat cancer is still an active research field, due to the al-
most unavoidable onset of serious, undesirable side ef-
fects, such as resistance and toxicity, when using the
presently available drugs. Recently, a new appealing tar-
get, the enzyme telomerase, has drawn the attention of
medicinal chemists.
Due to the inability of DNA polymerases to replicate
the end of linear DNA, a progressive shortening of the
chromosomes occurs during each replication cycle.1
However, loss of genetic information is avoided by the
occurrence of non-coding repeated sequences, called
telomeres, at the end of chromosomes. The length of
the telomeric region has multiple functions among
which the signalling to stop cell division when telomeres
Among the approaches thus far devised, blocking the
protein catalytic portion (hTERT) or its RNA domain
(hTR) gave successful results. In particular, a palmitoyl
lipid conjugate of GRN163, an antisense thio-phospho-
ramidate oligonucleotide 13 bases long, which works as
a ‘template antagonist’ for hTR, is currently undergoing
phase I clinical trials.7
Keywords: PIPER; Perylene; Telomerase; G-quadruplex.
*
Corresponding author. Tel./fax: +39 0498275711; e-mail: claudia.
sissi@unipd.it
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.09.029

556
C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
An alternative validated approach to interfere with the
telomerase-mediated DNA elongation process is to
modify the structural properties of the DNA substrate.
In fact, only the single-stranded linear form of the telo-
mere is suitable for recognition by this enzyme. Interest-
ingly, in all species, telomere sequences are rich in
guanines; in human cells, the telomeric repeat is
d(TTAGGG).8 An intrinsic property of the G-rich
sequences is their ability to fold into peculiar secondary
structures, called G-quadruplexes, stabilized by reverse
Hoogsteen hydrogen bonds connecting four guanine
residues.9,10
self-aggregation. In addition, we progressively reduced
the number of condensed aromatic rings in an attempt
to understand the role of the dimension of the planar
system in driving specific DNA structure recognition.
Spectroscopic studies (fluorescence, circular dichroism)
as well as telomerase inhibition and cell cytotoxicity as-
says were performed to correlate physico-chemical and
biochemical properties of the test derivatives.
2. Chemistry
Several G-quadruplex structures have been reported,
differing in strand orientation, glycosidic conformation
or loop geometry.11–13 In all cases, a large planar aro-
matic network is formed by the four bases. A molecule
that efficiently interacts with the planar quadruplex
structure would substantially stabilize the folded confor-
mation over the linear one, thus interfering with telome-
rase-mediated DNA elongation.14
A summary of the synthetic procedures used to prepare
the novel perylene analogues is reported in Schemes 1
and 2. Details are given in the Supplementary Material.
The number of condensed cycles has been progressively
increased from three to four and finally to seven. Side
chains exhibiting different steric hindrance as well as dif-
ferent pKa were introduced by reacting the proper anhy-
dride with the required diamine (Scheme 1). In some
instances, the terminal amine groups were converted
into permanently charged quaternary ammonium salts
by treatment with methyl iodide. The extension of the
planar system has been additionally modulated by
reduction of the carbonyl groups. This was achieved
either by reduction of the oxidized system with AlCl3/
LAH (Scheme 2, route A) or by preparation of the cyclic
amine starting from the dibromo intermediate (Scheme
2, route B). The chemistry we used to obtain this class
of compounds is rather flexible and allows us to
introduce a vast number of substituents giving ana-
logues of the leading compound, PIPER. The chemical
structures of the novel test compounds are summarized
in Figure 1.
A number of compounds, generally characterized by a
large aromatic moiety that can stack onto a G-tetrad,
were found to be active in this sense.15 Besides G-quad-
ruplex recognition, an ideal telomerase inhibitor, suit-
able for pharmacological applications, should not bind
efficiently to other structures of DNA, in particular to
the double helical nucleic acid, to avoid toxicity effects.
Interesting examples are derivatives bearing a perylene
or a porphyrin central core such as PIPER16–19 or
TMPyP4.20 Other ‘small’ molecules based on the acri-
dine or anthracenedione aromatic system produced
selective (tetraplex vs double-stranded) DNA recogni-
tion in a fashion that is dependent upon the nature
and location of charged side chains.21,22 Among them,
the trisubstituted acridine derivative BRACO 19, de-
signed using a molecular modelling approach, exhibited
a considerable efficacy in vivo.23,24
Route A
R
N
R
N
SAR studies on PIPER involving modifications of the
piperidine ring of the side chains25,26 evidenced that
the self-aggregation properties of these molecules are
largely responsible for their binding preferences to
G-quadruplexes versus double-stranded DNA.27–29 This
finding is in agreement with the NMR resolved structure
of a G-quadruplex bound to PIPER showing an external
stacking of the drug onto the G-tetrad.17
O
O
AlCl3/LAH
N(CH3)2
Route B
Br
Br
N
Starting from the available information, in the present
paper we examined a number of perylene derivatives
exhibiting different side chains to modulate their
steric hindrance, protonation state and extent of
NH2-(CH2)2-N(CH3)2
Scheme 2.
(CH2)n-NR1R2
(CH2)n-NR1R2
O
O
O
O
N
O
O
N
O
I
NH2-(CH2)n-NR1R2
CH3I
Scheme 1.

C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
557
R
N
R
N
X
X
X
X
X
X
N
R
Tricyclic derivatives
Tetracyclic derivatives
X = C=O
R = (CH2)2N(CH3)2
R = (CH2)2N(CH3)3
1
2
X = C=O
R = (CH2)2N(CH3)2
R = (CH2)3N(CH3)2
R = (CH2)2NH(CH2)2OH
R = (CH2)2N(CH3)3
R = (CH2)3N(CH3)3
R = (CH2)2N(CH2)2Et
4
5
6
7
8
9
X = CH2
R = (CH2)2N(CH3)2
3
X = CH2
R = CH3
R = (CH2)2N(CH3)2
10
11
R
N
X
X
Heptacyclic derivatives
X = C=O
X = CH2
R = (CH2)2N(CH3)2
R = (CH2)2(1-piperidino) PIPER
R = (CH2)2NH(CH2)2OH 13
12
R = (CH2)2N(CH3)3
14
R = (CH2)2(1-piperidino) 15
X
X
N
R
Figure 1. Molecular structure of test compounds.
3. Enzyme inhibition assays
A
0
4.9 9.8 19.5
39
78 156
nM
All new test derivatives were evaluated for their telome-
rase inhibition properties as determined by TRAP assay
(Fig. 2). To avoid false positive results due to drug inter-
ference with the amplification step, Taq polymerase inhi-
bition was additionally monitored. The results are
summarized in Table 1, where a selectivity index (SI,
ratio between IC50 for Taq polymerase vs telomerase
inhibition) is also included.
The need for a proper number of condensed rings for
efficient enzyme/s inhibition clearly emerged. An analy-
sis of the data for the analogues 1–3 indicated that none
of these tricycles interfered efficiently either with Taq
polymerase or telomerase.
Moving to the tetra and further to heptacyclic deriva-
tives, Taq polymerase inhibition became stronger and
stronger. The four-condensed ring system (tetracyclics
from 4 to 9) was sufficient to grant remarkable activity
although removal of the distal protonable amine in the
side chains (10), as well as reduction of the carbonyl
groups (11) markedly decreased it. However, in the
presence of these tetracyclic derivatives, telomerase
inhibition, if any, occurred at drug concentrations com-
parable to those required to inhibit Taq polymerase.
IC
Figure 2. TRAP assay performed using a cellular extract from JR8
cells in the presence of increasing concentrations of 14 (0–160 nM).
Lane A refers to the same reaction performed in the absence of cellular
extract. The arrow indicates the 36 bp internal control.

558
C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
Table 1. Biological properties of test compounds
Compound
IC50 (lM) Taq polymerase
IC50 (lM) telomerase
Selectivity indexa (SI)
Cell cytotoxicity IC50 (lM)b
H460
HT29
1
>40
no
no
no
no
—
—
3.42 0.04
P100
P100
0.16 0.03
0.42 0.03
9.3 1.6
2.21 0.28
1.14 0.21
4.40 0.62
P100
5.05 0.43
nd
2
22 2.4
>40
1.5 2.0
3
—
—
P100
nd
4
5
6.4
5.8
5.0
5.0
14
2
2
2
2
2
3.5
no
2.5
2.5
6.8
no
no
2
1.8
—
0.13 0.01
11.7 0.25
3.24 0.30
8.89 0.57
8.82 0.75
P100
15.2 0.99
0.23 0.03
9.02 1.03
1.17 0.09
P100
6
7
2
2
2
2.0
2.0
2.0
—
8
9
10
11
12
PIPER
13
14
15
>40
25
2
2
2
2
2
2
—
P100
2.30 0.20
0.9
5
0.16 0.02
0.20 0.02
0.06 0.01
0.03 0.01
0.40 0.03
5.6
25
P100
0.2
0.15
10
2.0
4.2
25
2.62 0.90
P100
28.1 5.31
5.40 0.63
no, not observed in the tested concentration range 0–40 lM and, in any case, lower than Taq polymerase inhibition nd, not determined.
a Ratio between the drug concentrations at which 50% Taq polymerase/telomerase inhibition was observed.
b Drug concentration producing 50% cancer cell death.
On the contrary, all tested heptacyclic derivatives (from
12 to 15) showed a concentration window for selective
interference with telomerase activity, with the notable
exception of derivative 13. Among heptacyclics, it is
interesting to note that the presence of a cyclic amine
in the side chains as in PIPER increased this concentra-
tion window irrespective of the reduction of the carbon-
yl groups of the planar ring system (15).
4. Self-aggregation properties
Perylene derivatives are known to self-aggregate produc-
ing a substantial quenching of their fluorescence emis-
sion.27,28 Additionally, self-aggregation is reported to
favourably increase the selective recognition of G-quad-
ruplex versus dsDNA. Thus, we monitored self-aggrega-
tion properties of our heptacyclic derivatives from 12 to
15 fluorometrically as a function of pH (Fig. 3).
The whole set of substitutions tested here showed that
the activity against telomerase and Taq is similarly
affected by the nature of the perylene side chains
although to a larger extent in the case of the DNA
amplification reaction. In particular, the most efficient
inhibition of both enzymes was observed with deriva-
tives 13 and 14. For compound 14, this can reasonably
rest in the presence of positively charged quaternary
nitrogens that facilitate interaction of the drug with
the negatively charged DNA, independently of the
polynucleotide conformation. For 13, in addition to
the basic nature of the amine, the presence of a hydroxyl
moiety that can form hydrogen bonding with DNA
bases may account for more efficient DNA
complexation.
At pH 8.0, PIPER and 15 showed essentially no fluores-
cence emission, compounds 12 and 13 presented an
intermediate behaviour, whereas compound 14 was the
most fluorescent derivative. Lowering pH to 3.0 pro-
duced an increment in fluorescence intensity for all test
compounds. In acidic solution, full protonation of the
side chains occurs, hindering self-aggregation and, in
particular, the face-to-face interactions that mainly con-
tribute to the quenching process.30 Thus, the ratio be-
tween the emission intensities at pH 3.0 and 8.0 can be
used as a parameter related to the extent of self-associa-
tion at physiologically relevant conditions. In agreement
with literature data according to which PIPER aggrega-
tion starts at pH 7.5,27 our results (Table 2) indicate
Figure 3. Emission spectra (kecc = 520 nm) of 0.5 lM PIPER or 14 in 10 mM Tris, 1 mM EDTA at pH 8.0 (dotted line) or 3.0 (solid line).

C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
Table 2. Biophysical properties of test heptacyclic compounds
559
b
Heptacyclics
Side chains
Fluorescence
quenchinga
% CD increment (295 nm)
Tm (ꢁC)
18a
(CH2)2N(CH3)2
(CH2)2Ncc
140
580
10
16.0
11.3
16.3
46.3
5.3
73.4
68.8
73.4
69.0
62.6
PIPER
20a
21
(CH2)2NH(CH2)2OH
(CH2)2N(CH3)3
(CH2)2Ncc
7
520
22a
a Ratio between the fluorescence quantum yields measured at pH 3.0 and pH 8.0, respectively.
b The melting temperature for 4GGG in the absence of drugs is 62.0 ꢁC.
c Nc, 1-piperidino.
that, at pH 8.0, PIPER and 15 are largely aggregated, 12
and 13 exhibit an intermediate extent of self-association,
whereas compound 14 is essentially not aggregated. This
latter finding is explained by the permanent charge
exhibited by the quaternary nitrogens of 14, which sub-
stantially reduce aggregation at any pH.
oligonucleotide melting profile. In our experimental con-
ditions, the G-quadruplex structure assumed by the oli-
gonucleotide showed a thermal transition at 62.0 ꢁC. In
the presence of all tested perylenes, this temperature was
incremented. As reported in Table 2, 12 and 13 proved
to be the most effective G-quadruplex stabilizers, fol-
lowed by PIPER and 14, and finally by 15. Interestingly,
the increment in Tm generally parallels that in the CD
band at 295 nm, only 14 being an outlier. Indeed, this
latter derivative is the only one presenting permanently
charged side chains, whereas all other derivatives display
acid–base equilibria which are surely affected by temper-
ature changes.
5. Circular dichroism studies
The effects of the test derivatives on G-quadruplex fold-
ing/stability were further assayed by CD spectroscopy
(Fig. 4). These studies were performed using a synthetic
DNA sequence containing four human telomeric re-
peats, 4GGG. In 100 mM KCl, this sequence assumes
an intramolecular G-quadruplex structure producing a
CD spectrum characterized by a positive and a negative
peak at 290 and 235 nm, respectively.13 All tested peryl-
enes were able to modify the oligonucleotide spectrum,
in particular inducing an increment in the positive band
alone with a shift to 295 nm. The extent of such an incre-
ment is largely modulated by the nature of perylene side
chains (Table 2). Among tested compounds, derivatives
14 and 15 emerged as the most and the least active. In
the presence of a 4GGG mutated sequence, unable to
form G-quadruplexes (MUT), no major changes in
DNA optical activity were monitored in the presence
of perylene analogues. A negligible effect was observed
also in the presence of a double helical nucleic acid.
Finally, when comparing PIPER to its analogue with re-
duced carbonyls, it emerges that partial loss of planarity
in the heptacyclic system reduces G-quadruplex recogni-
tion and this well correlated with its reduced telomerase
inhibition.
6. Cell cytotoxicity
The cytotoxic effects of test derivatives, summarized in
Table 1, were studied using human lung and colon can-
cer cell lines (H460 and HT29, respectively) following
168 h of continuous drug exposure. Almost negligible ef-
fects were observed for PIPER, 14 and 15. As far as 14,
this could be reasonably ascribed to the permanently
charged quaternary amine side chains, which likely im-
pair cellular uptake. In the other cases, the presence of
the cyclic amine groups, irrespective of the presence of
The perylene interaction with the folded form of
4GGG was further investigated by monitoring the
20000
10000
0
20000
10000
-10000
-20000
0
-4000
350
300
230
250
230
250
300
350
Wavelength [nm]
Wavelength
Figure 4. CD spectra of oligonucleotides 4GGG and MUT (4 lM) in 10 mM Tris, 1 mM EDTA and 100 mM KCl, pH 8.0, in the absence (solid line)
and in the presence (dotted line) of 14 (14 lM).

560
C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
25
20
15
10
5
carbonyl groups in the planar portion, appears to be
responsible for poor short-term cell killing ability. In
fact, the congeners 12 and 13, exhibiting non-cyclized
secondary and tertiary amine substituents, are at least
one order of magnitude more potent than PIPER and
15 as cytotoxic agents. It is to note that increasing the
selectivity index produces a decrease in cytotoxicity.
This is in line with the fact that interaction with telomer-
ic DNA is expected to produce cell death only after sev-
eral population doublings.
CD intensity at 295 nm
Tm
Interestingly, derivatives with reduced extension of the
planar ring system were shown to be lethal to cancer
cells to a notable extent. For tetracyclic derivatives, this
is in agreement with their efficient binding to dsDNA.
This is at odds with the tricyclic derivative 1, a poor
DNA-binder but exhibiting a cytotoxic IC50 in the low
micromolar range. Considering its strong structural sim-
ilarity to the well-known topoisomerase poison amona-
fide,31 this compound is likely to produce its cytotoxic
effects by interfering with enzyme(s) other than telome-
rase or polymerase, considered here.
0
50
100
150
200
250
300
350
400
450
Telomerase IC50 (nM)
Figure 5. Plot of % increment in 4GGG CD intensity at 295 nm (full
symbols) or Tm (empty symbols) as a function of telomerase IC50
determined for the test perylene derivatives.
conditions relevant to physiological processes. Indeed, it
has been suggested that the extent of aggregation can be
critical for improving the selective inhibition of telome-
rase.13,27 In fact, we observed that the selectivity index
obtained from enzyme inhibition data parallels the ex-
tent of self-aggregation in the perylene family.
7. Conclusions
Our results indicate that the dimensions of the planar
portion of DNA-directed drugs subtly modulate their
ability to recognize different DNA structural arrange-
ments. In particular, we showed that among PIPER-re-
lated ligands, a conjugated system consisting of more
than four rings is required to efficiently target G-quar-
tets. This finding is at odds with data reported on
G-quadruplex binders structurally related to anthracene
or acridine, for which good tetraplex selectivity can be
reached in the presence of just three-condensed rings,
the number and location of the side chains being crucial
for selective targeting of DNA structures.
If we assume that this parameter is preferentially
linked to G-quadruplex recognition, we would expect
also an increment in telomerase inhibition upon aggre-
gation. However, this is not the case. In fact the more
aggregated the compounds, the lower the activity on
the enzyme and additionally, the lower the interaction
with G-quadruplex (assuming the increment in CD
intensity at 295 nm and Tm as indicative of binding
efficiency).
A key role of the side chains clearly emerged also among
perylenes as they differentially modulated a large num-
ber of physico-chemical properties.
Thus, it follows that drug aggregation increases selectiv-
ity for G-quadruplex arrangements with respect to dou-
ble helical and single-stranded structures not by
enhancing affinity for G-quartets but by reducing bind-
ing efficiency to other nucleic acid conformations.
First of all, we observed that strong interactions with
G-quadruplex arrangements favour telomerase inhibi-
tion. For the oligonucleotide 4GGG, increments in opti-
cal activity at 295 nm and in melting temperature
produced by the test compounds are safely reflecting in-
creased stability of the G-quadruplex conformation. As
clearly shown in Figure 5 a good linear correlation oc-
curs between either of these two data sets and the telo-
merase inhibition in vitro. Based on these results and
on the structural similarities of these active compounds
to PIPER, it can be reasonably inferred that enzyme
inhibition is linked to stabilization of G-quadruplex
folded forms. In connection to this, derivative 14 is the
most efficient analogue. We argue that the permanent
positive charges present in its side chains strongly stabi-
lize the DNA-drug complex by ionic contacts.
Finally, the modulation of the cell killing ability ob-
served among our heptacyclic derivatives appears to be
promising for the development of effective anticancer
compounds based upon recognition of telomeric DNA
and exhibiting less severe short-term toxic effects.
8. Experimental
Test derivatives were synthesized and characterized
accordingly to procedures reported in Supplementary
Material. Synthetic oligonucleotides were purchased
from Eurogentec and used without further purification.
The sequences used in this study were 4GGG 50-AGG
GTT AGG GTT AGG GTT AGG G-30 and MUT 50-
TACAGATAGTTAGACTTAACGTTA-30. Solutions
of the test derivatives were prepared freshly just before
use.
Additionally, the substituents tested here show different
steric hindrance and different pKa values, which leads to
different degrees of drug self-association in experimental

C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
561
8.1. Fluorescence spectroscopy
Telomerase activity was assayed using a modified telo-
mere repeat amplification protocol (TRAP) assay.32
Briefly, a proper primer (TS) has been 50-labelled with
[c-32P]ATP and T4 polynucleotide kinase. After enzyme
inactivation (85 ꢁC for 5 min), a 50 ll TRAP reaction
mix (50 lM dNTPs, 0.2 lg of labelled TS, 0.1 lg of re-
turn primer ACX, 500 ng of proteic extract and 2 U
Taq polymerase) was prepared in the presence/absence
of increasing drug concentration. According to (41),
an internal control template (0.01 amol TSNT) with its
return primer (1 ng NT) was added to the reaction mix-
ture. Then, telomerase elongation step has been per-
Fluorescence spectra were recorded on a FL-20 Perkin-
Elmer spectrofluorimeter. Sample solutions at 0.5 lM
were prepared in 10 mM Tris and 1 mM EDTA and
were incubated 1 h at room temperature before scan
acquisition. Each measurement was averaged over 3
scans and recorded at 25 ꢁC using an excitation wave-
length of 500 nm.
8.2. Circular dichroism measurements
Oligonucleotide circular dichroism spectra from 230 to
350 nm were recorded using 10 mm path length cells
on a Jasco J 810 spectropolarimeter equipped with a
NESLAB temperature controller and interfaced to a
PC 100 in 10 mM Tris–HCl, 1 mM EDTA and
100 mM KCl, pH 8.0. Before data acquisition, DNA
solutions were heated at 95 ꢁC for 5 min and left to cool
at room temperature on. The reported spectrum of each
sample represents the average of 3 scans recorded with
1-nm step resolution. Observed ellipticities were convert-
ed to mean residue ellipticity [h] = deg · cm2 · dmolꢀ1
(Molar Ell.).
formed (30 min at 30 ꢁC) followed by
a PCR
amplification step (30 cycles of: 30 s at 37 ꢁC and 30 s
at 58 ꢁC). The reaction products were loaded onto a
10% polyacrylamide gel (19:1) in TBE. Gels were trans-
ferred to Whatman 3MM paper, dried under vacuum at
80 ꢁC and read using a phosphorimager apparatus
(Amersham).
Values are expressed as percent of telomerase inhibition
relative to control (no drug) lanes.
8.5. Cell cultures and toxicity assays
Thermal denaturation experiments were performed by
recording the DNA optical activity at 295 nm as a func-
tion of the temperature. Oligonucleotide solutions
(4 lM) were equilibrated at 25 ꢁC, then the signal was
recorded while increasing the temperature at 0.8 ꢁC/
min and stirring the solution to allow equilibration.
Experiments were performed in the presence or absence
of tested derivatives (4 lM final concentrations). Tm was
determined by locating the maxima/minima of the first
derivative of the curve describing the melting profile
(CD vs T). Two to four scans were repeated for each
experimental condition. When required, the melted
solution was cooled down at the same temperature
change rate.
H460 human lung cell line was maintained in defined
RPMI 1640 medium supplemented with 10% heat-inac-
tivated foetal calf serum, 2 mM of L-glutamine 50 U/ml
of penicillin G and 50 lg/ml of streptomycin, at 37 ꢁC in
humidified atmosphere and 5% of CO2. HT-29 colon
cancer cell line was grown in McCoy’s defined medium
supplemented with 10% heat-inactivated foetal calf
serum, 2 mM of L-glutamine and 10 mM Hepes.
To evaluate toxic profiles of the potential antitelomera-
sic compounds, MTT assays were performed as de-
scribed: cells were plated in 96-well plates at 10,000
cells/well and cultured overnight. Afterwards, com-
pounds were added in octuplicate and plates were incu-
bated in presence of the drug for 168 h. At the end of
this period, MTT was added to a final concentration
of 50 lg/ml, and two additional hours of incubation
were performed. After that, medium was aspirated care-
fully and 100 ll of DMSO was added per well. Soluble
formazan salts were homogenated by manual pipetting
and absorbance at 570 nm was read. Curves consisted
in 10 serial dilutions in octuplicate in each case, and
results were analysed as sigmoidal dose–response curves
using GraphPad Softwareꢂ.
8.3. Taq polymerase assay
To meet proper working conditions, compounds were
assayed with Taq polymerase reaction by using
pBR322 (2.5 ng) as a DNA template and appropriate
primer sequences (0.5 lM) to amplify the 906–1064 se-
quence of plasmid by PCR. The reaction was carried
out in a Perkin-Elmer thermocycler performing 25 cy-
cles of: 30 s at 94 ꢁC, 30 s at 65 ꢁC and 30 s at 72 ꢁC.
The reaction products were resolved on a 2% agarose
gel in TBE 1X (89 mM Tris base, 89 mM boric acid
and 2 mM Na2EDTA) and stained by ethidium
bromide.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmc.2006.09.029.
8.4. Telomerase activity assay
An aliquot of 5 · 106 JR8 cells in exponential phase of
growth was pelleted and lysed for 30 min on ice using
100 ll of 0.5% Chaps, 1 mM EGTA, 25% 2-mercap-
toethanol, 1.74% PMSF and 10% w/v glycerol. The ly-
sate was centrifuged at 13,000 rpm for 30 min at 4 ꢁC
and the supernatant collected, stored at ꢀ80 ꢁC and
used as the telomerase source.
References and notes
1. Watson, J. D. Nat. New Biol. 1972, 239, 197–201.
2. Blackburn, E. H. Cell 2001, 106, 61–73.
3. Greider, C. W.; Blackburn, E. H. Cell 1985, 43, 405–413.

562
C. Sissi et al. / Bioorg. Med. Chem. 15 (2007) 555–562
19. Han, H.; Bennett, R. J.; Hurley, L. H. Biochemistry 2000,
4. Kim, N. W.; Piatyszek, M. A.; Prowse, K. R.; Harley, C.
B.; West, M. D.; Ho, P. L.; Coviello, G. M.; Wright, W.
E.; Weinrich, S. L.; Shay, J. W. Science 1994, 266, 2011–
2015.
39, 9311–9316.
20. Shi, D. F.; Wheelhouse, R. T.; Sun, D.; Hurley, L. H.
J. Med. Chem. 2001, 44, 4509–4523.
5. Riou, J. F. Curr. Med. Chem. Anti-Cancer Agents 2004, 4,
439–443.
21. Perry, P. J.; Gowan, S. M.; Reszka, A. P.; Polucci, P.;
Jenkins, T. C.; Kelland, L. R.; Neidle, S. J. Med. Chem.
1998, 41, 3253–3260.
22. Perry, P. J.; Reszka, A. P.; Wood, A. A.; Read, M. A.;
Gowan, S. M.; Dosanjh, H. S.; Trent, J. O.; Jenkins, T. C.;
Kelland, L. R.; Neidle, S. J. Med. Chem. 1998, 41, 4873–
4884.
6. Kelland, L. R. Eur. J. Cancer 2005, 41, 971–979.
7. Akiyama, M.; Hideshima, T.; Shammas, M. A.; Hayashi,
T.; Hamasaki, M.; Tai, Y. T.; Richardson, P.; Gryaznov,
S.; Munshi, N. C.; Anderson, K. C. Cancer Res. 2003, 63,
6187–6194.
8. Meyne, J.; Ratliff, R. L.; Moyzis, R. K. Proc. Natl. Acad.
Sci. U.S.A. 1989, 86, 7049–7053.
9. Keniry, M. A. Biopolymers 2000–2001, 56, 123–146.
10. Davis, J. T. Angew. Chem., Int. Ed. Engl. 2000, 43, 668–
698.
11. Wang, Y.; Patel, D. J. Structure 1993, 1, 263–282.
12. Parkinson, G. N.; Lee, M. P.; Neidle, S. Nature 2000,
2417, 876–880.
13. Ambrus, A.; Chen, D.; Dai, J.; Bialis, T.; Jones, R. A.;
Yang, D. Nucleic Acids Res. 2006, 34, 2723–2735.
14. Cuesta, J.; Read, M. A.; Neidle, S. Mini-Rev. Med. Chem.
2003, 3, 11–21.
15. Sun, D.; Thompson, B.; Cathers, B. E.; Salazar, M.;
Kerwin, S. M.; Trent, J. O.; Jenkins, T. C.; Neidle,
S.; Hurley, L. H. J. Med. Chem. 1997, 40, 2113–
2116.
16. Kerwin, S. M.; Federoff, O. Y.; Salazaar, M.; Hurley L.
H., U.S. Patent 6156763, 2000.
23. Read, M.; Harrison, R. J.; Romagnoli, B.; Tanious, F. A.;
Gowan, S. H.; Reszka, A. P.; Wilson, W. D.; Kelland, L.
R.; Neidle, S. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
4844–4849.
24. Gowan, S. M.; Harrison, J. R.; Patterson, L.; Valenti, M.;
Read, M. A.; Neidle, S.; Lelland, L. R. Mol. Pharmacol.
2002, 61, 1154–1162.
25. Rossetti, L.; Franceschin, M.; Schirripa, S.; Bianco, A.;
Ortaggi, G.; Savino, M. Bioorg. Med. Chem. Lett. 2005,
15, 413–420.
26. Rossetti, L.; Franceschin, M.; Bianco, A.; Ortaggi, G.;
Savino, M. Bioorg. Med. Chem. Lett. 2002, 12, 2527–2533.
27. Kern, J. T.; Thomas, P. W.; Kerwin, S. M. Biochemistry
2002, 41, 11379–11389.
28. Kerwin, S. M.; Chen, G.; Kern, J. T.; Thomas, P. W.
Bioorg. Med. Chem. Lett. 2002, 12, 447–450.
29. Tuntiwechapikul, W.; Taka, T.; Bethencourt, M.; Randall
Lee, T. Bioorg. Med. Chem. 2006.
17. Fedoroff, O. Y.; Salazar, M.; Han, H.; Chemeris, V. V.;
Kerwin, S. M.; Hurley, L. H. Biochemistry 1998, 37,
12367–12374.
30. Ford, W. E. J. Photochem. 1987, 37, 189–204.
31. Hsiang, Y. H.; Jiang, J. B.; Liu, L. F. Mol. Pharmacol.
1989, 36, 371–376.
18. Han, H.; Cliff, C. L.; Hurley, L. H. Biochemistry 1999, 38,
6981–6986.
32. Kim, N. W.; Wu, F. Nucleic Acids Res. 1997, 25, 2595–
2597.