Photoactivatable platinum(II) terpyridine derivatives for G-quadruplex DNA double anchoring

Article abstract of DOI:10.1016/j.ica.2016.02.033

Platinum(II) tolylterpyridine (ttpy) complexes, have been shown to generate monofunctional adducts with various G-quadruplex DNA forming sequences resulting from an efficient π-stacking mode on the top of the quadruplex structure. To further explore the p

Full text of DOI:10.1016/j.ica.2016.02.033

Inorganica Chimica Acta xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Photoactivatable platinum (II) terpyridine derivatives for G-quadruplex
DNA double anchoring
a
a
a
a,b,
⇑
, Marie-Paule Teulade-Fichou a,
⇑
Elodie Morel , Florent Poyer , Laurence Vaslin , Sophie Bombard
a
Institut Curie, CNRS UMR9187, INSERM U 1196, Université Paris Sud, Bât 110-112, Rue Henri Becquerel, 91405 Orsay Cedex, France
Université Paris Descartes, INSERM UMR-S-1007, 45 rue des Saints-Pères, 75270 Paris Cedex 06, France
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Platinum (II) tolylterpyridine (ttpy) complexes, have been shown to generate monofunctional adducts
with various G-quadruplex DNA forming sequences resulting from an efficient -stacking mode on the
top of the quadruplex structure. To further explore the potential of this series with regard to quadruplex
recognition, classical photocrosslinking groups (benzophenone, tetraphenylazide) have been grafted on
3
the tolylterpyridine ligand moiety, thereby generating two new derivatives Pt-ttpy-Bn and Pt-ttpy-N .
Received 1 December 2015
Received in revised form 15 February 2016
Accepted 18 February 2016
Available online xxxx
p
Evaluation of their non-covalent binding for G-quadruplex DNA has been performed by FRET-melting
and FID assays using two G-quadruplex matrices i.e. the telomeric sequence 22AG and the oncogene pro-
motor sequence c-myc, which revealed high affinity and improved selectivity as compared to the parent
compound. Subsequently the capacity of the compounds to establish one or two anchorage points (one by
platination, one by photoinduced crosslinking) with the quadruplexes has been studied by gel elec-
trophoresis with and without photoactivation. Interestingly both compounds do platinate the quadru-
plexes studied with high selectivity as the platination yield is poorly affected by the presence of
Keywords:
G-quadruplex
Platinum complexes
Photo-crosslinking
3
duplex competitor. By contrast, only the azido derivative Pt-ttpy-N was found to form a second covalent
bond within the G-quadruplexes upon photoactivation indicating a higher confinement of the crosslink-
ing moiety in this case. Finally the two compounds exhibit poor cytotoxicity in the dark on two cancer
cell lines (A2780 and HT29), whereas that of the benzophenone derivative is significantly enhanced upon
irradiation. Altogether the two new compounds represent novel prototypes usable for trapping G4 DNA
alone or eventually G4-DNA protein interactions in complex in vitro systems or in cells.
Ó 2016 Elsevier B.V. All rights reserved.
Nucleic acid secondary structures called G-quadruplex are
believed to play regulatory roles in the main functions related to
DNA processing [1]. However, although the field is booming and
quadruplex DNA is on the way to become a paradigm used to
explain most DNA processing dysfunctions, there are still a number
of essential questions unanswered. In particular, the number of
quadruplex forming sequences (QFS) in vivo in the human genome
is still a matter of controversy. This number was initially claimed
to be around 300,000 based on bioinformatics analysis but recent
in vivo studies in yeast based on quadruplex-induced genetic insta-
bility strongly suggest a drastic decrease of the number of PQS
down to 20,000 [2] whereas in vitro sequencing identified up to
it is of importance to provide new chemical biology tools for trap-
ping and mapping PQS as well as identifying their protein partners.
Photoreactive species are extensively used for the study of DNA
interactions with biological partners, and especially for trapping
proteins [5]. Moreover, since the discovery of antitumor properties
of cis-platin, numerous metal complexes have been synthesized for
their coordination potential on DNA. With the use of a compound
combining a photoreactive benzophenone and cis-platin analog,
Lippard et al. were thus able to efficiently crosslink both duplex
DNA and associated proteins, leading to identification of those pro-
teins [6]. In this line we have recently shown that compounds com-
bining a G-quadruplex stabilizing scaffold (i.e. bisquinolinum
pyridodicarboxamide [7]), and cross linking moieties led to cova-
lent trapping of G-quadruplex DNA, upon alkylation [8] and pho-
toactivation [9]. In parallel, we developed the class of platinum
7
00,000 PQS [3]. Another crucial question is whether specific pro-
tein have been evolved to process quadruplex DNA or if this occurs
via the standard machinery e.g. helicases [4]. For all these reasons
(
II) tolylterpyridine (Pt-ttpy) complexes and showed that these
square planar complexes have the ability to -stack on the top of
p
⇑
Corresponding authors at: Université Paris Descartes, INSERM UMR-S-1007, 45
rue des Saints-Pères, 75270 Paris Cedex 06, France (S. Bombard).
G-quadruplexes which favor coordination to the surrounding
nucleic bases resulting in highly specific and stable adducts [10].
http://dx.doi.org/10.1016/j.ica.2016.02.033
0
020-1693/Ó 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033

2
E. Morel et al. / Inorganica Chimica Acta xxx (2016) xxx–xxx
Therefore, based on the previous work outlined above, it was
quadruplex forming sequences 22AG and cmyc (myc22), corre-
sponding to the human telomeric sequence and the oncogene
promoter cmyc respectively (see sequences in Supporting Infor-
mation). The affinity of these platinum complexes for G-quadru-
plex and duplex DNA was first investigated by fluorescent
intercalator displacement (G4-FID) assay [16], which is based on
the competitive displacement of thiazole orange (TO) from DNA
by the compounds to be evaluated (Fig. 1). This semi-quantitative
assay allows affinity ranking for series of compounds by determi-
nation of the DC50 value i.e. the compound concentration induc-
decided to introduce photocrosslinking function on the G-quadru-
plex binding platinum (II) complex (Pt-ttpy) in the aim to develop
molecular tools with two anchoring possibilities. This double
anchorage may occur inside the quadruplex and in this case should
considerably strengthen the small molecule/quadruplex interac-
tion and eventually raise preferences for quadruplex topologies.
Alternatively these double connecting agents may allow trapping
of quadruplex/protein interactions.
We therefore envisaged the preparation of two compounds (Pt-
ttpy-Bn and Pt-ttpy-N
groups, benzophenone (Bn) and 4-azido-2,3,5,6-tetrafluorobenzoic
acid (N ), on the tolylterpyridine scaffold. These photoreactive
groups can be activated by irradiation at wavelengths close to
UVA-visible (330–365 nm) to generate highly reactive radical
intermediate species [11]. Both of these groups have been exten-
sively used for protein–protein photolabelling [12] or for the study
of protein–ligand [13] and DNA-ligand interactions [14].
3
), by grafting the two classical photoreactive
ing 50% of probe fluorescence decrease (DC50 < 0.5
characterizes high affinity binders, 0.5 < DC50 < 1 M medium to
moderate binders, 1 < DC50 < 2.5 M low affinity binders and
DC50 > 2.5 M no significant binding). For instance the non-func-
tionalized complex Pt-ttpy used here as reference shows strong
affinity for tested G-quadruplex DNA structures, oncogene cmyc
lM
l
3
l
l
and human telomeric 22AG sequences with a DC50 ꢀ 0.2
lM in
both cases. This complex was found to displace TO from duplex
The benzophenone moiety is commercially available and the
DNA with a significantly more moderate efficacy as 7-fold higher
tetrafluorobenzoate azide was prepared in
three-step synthesis [9a]. The two compounds, Pt-ttpy-Bn and
Pt-ttpy-N , were synthesized from a tolyterpyridine precursor
bearing an amino-terminated side chain on the tolyl ring (ttpy-
NH ) The preparation of this precursor was done in three steps of
high efficiency from 4-hydroxybenzaldehyde (Scheme 1).
Previously protected aminoalcohol spacer was substituted on 4-
hydroxybenzaldehyde with 96% yield. The Kröhnke reaction [15]
of this aldehyde with 2-acetylpyridine in the presence of potas-
sium hydroxide and ammonium hydroxide afforded the tolylter-
pyridine product ttpy-NHBoc with 26% in one step. The
deprotection of the amino group was performed with TFA and
a
well-described
DC50 is observed in this case (DC50 = 1.5 lM). In the same assay,
the azido derivative binds the two G-quadruplex structures with
a high affinity although the DC50 values are slightly lower: they
3
fall in the range defining high affinity binders (DC50 Pt-ttpy-N
/22AG = 0.55 M, DC50 Pt-ttpy-N /cmyc = 0.35 M). On the oppo-
site, for the benzophenone derivative the G-quadruplex affinity
is a bit more affected with DC50 values of 1.33 M and 0.61 lM
3
-
2
l
3
l
l
for 22AG and cmyc respectively, but with a clear preference for
cmyc. Most interestingly when evaluated in presence of duplex
DNA, these complexes show quite no displacement of TO.
(DC50 > 2.5 lM). Remarkably, comparison with the reference Pt-
ttpy indicates that the selectivity for G-quadruplex structure is
strongly increased when grafting a photocrosslinker moiety on
the tolylterpyridine scaffold.
led to ttpy-NH
pounds were prepared by coupling the photocrosslinking moieties
with the starting material ttpy-NH , followed by a metallation step
Scheme 2). In the first step, the carboxylic acid derivatives of ben-
2
in a quantitative yield. The bifunctional com-
2
FRET-melting experiments were also performed using G-
quadruplex DNA sequences doubly labelled with a donor–acceptor
FRET pair namely F21T and FmycT [17] (F = fluorescein, T = TAMRA,
see Supporting Information). This well-used assay is based on FRET
principle to monitor the thermal stability of DNA alone or in pres-
ence of a binding compound. As compared to standard UV-melting,
the FRET-melting enable to addition of unlabeled duplex DNA com-
petitor thereby allowing evaluation of ligand selectivity for the G-
quadruplex structure. Both compounds display strong stabilization
(
zophenone and teraphenylazide were introduced under classical
peptide-type coupling conditions using EDCI as coupling agent in
the presence of DMAP and HOBt as catalyst, affording compounds
ttpy-N
3
and ttpy-Bn with 79% and 84% yield respectively. These
, in extra
intermediates were platinated in presence of Pt(COD)Cl
2
dry methanol under argon atmosphere. After filtering the solid
from the reaction mixture and washing the platinum complexes
Pt-ttpy-N
yields.
3
and Pt-ttpy-Bn were obtained in moderate to good
of G-quadruplex structures, with
(Fig. 2) (for reference the benchmark compounds PhenDC
PDS induce of 25–30 °C in the same conditions). Fig. 2A corre-
D
T
m
values around 15–20 °C
3
and
DT
m
sponds to results obtained with the telomeric quadruplex. The
azido derivative exhibits a pattern very similar to that of the refer-
ence compound indicating a similar stabilization (first blue bar
1
. Interaction measurements
The newly prepared tolylterpyridine derivatives were evalu-
ated for their specific G-quadruplex DNA interactions using two
m
DT = 24 °C) whereas the benzophenone derivative has a signifi-
cantly lower stabilization effect (
D
T
m
= 13 °C) in consistency with
Boc
2
O
TsCl, Et
DCM
3
N
NH
2
NHBoc
NHBoc
HO
HO
TsO
THF
quant. yield
50-70 %
NHR
O
N
NHBoc
OH
O
N
Cs
2
CO
3
O
NHBoc
TsO
KOH, NH
EtOH
4
OH
DMF
0°C, 8h
6 %
6
H
O
H
O
9
N
N
3
4°C, 8h
3 %
3
R = Boc
R = H ttpy-NH
TFA
3 %
2
9
Scheme 1. Synthetic route for ttpy-NH
Nommer ttpy-NH
2
2 3
. (a) Cs CO , DMF, 60 °C, 8 h, 96%; (b) 2-acetylpyridine, KOH, NH4OH, EtOH, 34 °C, 24 h, 26%; (c) TFA, DCM, RT, 4 h, quantitative yield.
2
2
sur le schema R = H, ttpy NH .
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033

E. Morel et al. / Inorganica Chimica Acta xxx (2016) xxx–xxx
3
O
O
H
N
H
NH
2
O
N
O
O
N
O
O
O
HO
Pt(COD)Cl
2
O
Cl
MeOH extra dry
50°C, 24 h
80 %
EDCI, HOBt, DMAP
DMF
N
Pt
Cl
N
N
N
N
N
TA, 8h
N
N
8
4 %
F
F
F
F
F
N3
F
F
N3
F
H
N
H
F
F
NH2
N
O
F
N3
F
O
N
O
O
O
HO
Pt(COD)Cl2
O
Cl
MeOH extra dry
50°C, 24 h
EDCI, HOBt, DMAP
DMF
N
N
Pt
Cl
5
2 %
N
N
N
N
TA, 8h
N
N
7
9 %
Scheme 2. Coupling conditions for the Pt-ttpy-N
4 h, 52%; (c) 4-benzoylbenzoic acid, EDCI, HOBt, DMAP, DMF, RT, 8 h, 84%; (d) Pt(COD)Cl
linking agent. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3
and Pt-ttpy-Bn. (a) 4-Azidotetrafluorobenzoic acid, EDCI, HOBt, DMAP, DMF, RT, 8 h, 79%; (b) Pt(COD)Cl
2
, dry MeOH, 50 °C,
2
2
, dry MeOH, 50 °C, 24 h, 81%. In blue, toly-terpyridine moiety. In red, photo-cross-
Fig. 1. DC50 (lM) values from G4-FID assay on G-quadruplex (22AG and cmyc) and
+
duplex (ds26) DNA in 100 mM K buffer.
the FID assay. Addition of 3–10 equivalents of duplex DNA
competitor decreases the G-quadruplex binding for the three com-
pounds but to a moderate extent. However, the strong improve-
ment observed by FID assay is not reproduced here. Perfect
agreement between the two assays cannot be expected as platina-
tion may occur to various degree during both experiments (unpub-
lished results) [18].
In the case of cmyc (Fig. 2B), the three compounds show very
similar profiles (high affinity with
resistance to the competition revealing a high selectivity versus
D
T
m
ꢀ 13–15 °C) and excellent
+
Fig. 2.
buffer) and FmycT (B – 1 mM K buffer) for compounds (5 eq.) in presence of 0, 3
and 10 M of duplex DNA (ds26) as competitor.
DTm (°C) values from FRET-melting experiment on F21T (A – 10 mM K
duplex (of note due to the high stability of the cmyc quadruplex,
+
+
the assay is performed at low K ionic force thus the
m
DT values
l
cannot be compared between F21T and FmycT experiments). In
the whole, the two assays indicate that Pt-ttpy-Bn and Pt-ttpy-N
3
retained the high affinity for G-quadruplex DNA of the Pt-ttpy core.
Equally, high selectivity for G-quadruplex versus duplex DNA is
observed in all cases, especially in the case of cmyc. Nonetheless,
all compounds displayed low affinity towards duplex DNA, assess-
ing their strong structural preference for G-quadruplex DNA con-
formations. Of note and as previously shown [19], the
tolyterpyridine ligand alone shows no binding activity in both
assays thereby evidencing the crucial role of the metal that confers
cationic charge, rigidity and square planar geometry suitable for
quadruplex recognition.
2. Platination and photocrosslink of G-quadruplex DNA
To evaluate if the two compounds Pt-ttpy-Bn and Pt-ttyp-N
3
can cause double anchorage inside G-quadruplex DNA structures,
we performed platination and photocrosslinking experiments with
22AG and cmyc oligonucleotides, which were subsequently ana-
lyzed by denaturing gel electrophoresis. After incubation with
the oligonucleotides for 5 min, both compounds induce formation
of retarded bands with 22AG (Fig. 3A and C – lines c) and of accel-
erated bands with cmyc (Fig. 3B and D – lines c) which can be
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033

4
E. Morel et al. / Inorganica Chimica Acta xxx (2016) xxx–xxx
3
Fig. 3. Denaturating gel electrophoresis (15% acrylamide) of incubation of 2 eq. of compounds Pt-ttpy-Bn (A and B) and Pt-ttpy-N (C and D) with cmyc and 22AG (10 lM) in
+
the presence of K -buffer (50 mM) after 5 min incubation at 37 °C. Lane a = DNA only, b = DNA only, irradiated by 60 flashes, c = 5 min incubation of DNA and ligand, d = 5 min
incubation of DNA and ligand, followed by a 2 h post-incubation, e = 5 min incubation with ligand, followed by photoactivation and f = 5 min incubation with ligand, followed
by photoactivation and a 2 h post-incubation.
attributed to platinum adduct. Herein the adducts are formed in
yields similar to the ones already found for (photo)-alkylating
agents [8,9] varying from 11% to 19% for 22AG to 30% for cmyc,
and are significantly increased with longer incubation time (up
to 18–33% with 22AG and 40% with cmyc, Fig. 3A–D – lines d).
Of note these adducts migration characteristics i.e. retarded bands
with 22AG and accelerated with cmyc are similar to those
observed with the model compound Pt-ttpy (Fig. S1). When the
incubation is followed by photoactivation, no significant change
in the number and migration level of bands is observed in the case
of Pt-ttpy-Bn, irrespective of the incubation time (Fig. 3A and B –
lines e, f) suggesting either that photocrosslinking is not occurring
or that band migration is not modified by photoalkylation. By con-
trast, strong modification of the adducts profile is induced in the
generated from the already formed platination products. (Fig. 3D
– lines d, f). It can thus be assumed that the new bands observed
upon photoactivation correspond to double anchorage of the
3
bifunctional compound Pt-ttpy-N inside the cmyc quadruplex.
To locate the platination and photocrosslinking sites, bands
were isolated from the gels and subjected to different sequenc-
ing procedures. However partial deplatination of adducts during
the analysis procedure occurred (Fig. 3A–D – products A1 and
B1), therefore only photoalkylation sites could be identified.
As shown in our previous studies, benzophenone adducts revert
to starting material at high temperature [20], thus sequencing
0
of Pt-ttpy-Bn adducts had been done by 3 -exonuclease, whose
digestion stops at crosslinked sites [21]. Treatment of Pt-ttpy-
0
Bn photoactivated products by 3 -exonuclease showed no diges-
3
case of Pt-ttpy-N . The faster migrating bands corresponding to
tion arrests, indicating that Pt-ttpy-Bn forms no photocrosslink
platination adducts of cmyc are strongly decreased in favour of
new retarded bands (Fig. 3D – lines e, f). Similarly the migration
of the retarded bands observed upon platination of 22AG is modi-
fied (Fig. 3C – lines e, f). As irradiation alone does not affect DNA
integrity (Fig. 3A–D – lines b), the newly appearing bands induced
in these conditions account likely for photocrosslinking adducts
adducts on G-quadruplex DNA (Data not shown). Pt-ttpy-N
3
photoadducts (Fig. 3C and D – products A2, B21 and B22) have
been treated by piperidine at high temperature, which induces
0
cleavage on the 5 -side of the photocrosslinked bases. Treat-
ment of products obtained after photoactivation of Pt-ttpy-N
(Fig. 3C and D – A2, B21 and B22) highlights defined DNA
3
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033

E. Morel et al. / Inorganica Chimica Acta xxx (2016) xxx–xxx
Table 1
5
crosslinking sites (Fig. S2). Cleavage sites allowed us to deter-
Platination and photocrosslinking adducts determined for Pt-ttpy-Bn and Pt-ttpy-N
in the presence of cmyc and 22AG G-quadruplex forming sequences.
3
mine the location of photocrosslinking adducts on both 22AG
and cmyc by comparing the migration of those fragments with
the DMS-piperidine treated non-alkylated oligonucleotide.
With 22AG, the azido derivative forms covalent adducts on
guanines located within the two external 3 and 5 quartets of
the G-quadruplex structure (G8-G10-G14-G16-G22). This
crosslinking pattern is consistent with stacking of the compound
on the top of the parallel and antiparallel conformations identified
by NMR spectroscopy and X-rays for the human telomeric
Pt-ttpy-N
3
Pt-ttpy-Bn
CMYC
2AG
Platination
Photocrosslink
Platination
+
+
+
+
+
0
0
ꢁ
2
+
Photocrosslink
ꢁ
Finally the selectivity of platination of Pt-ttpy-Bn and Pt-ttpy-
3
was assessed by gel electrophoresis. To this end, formation of
platinated adducts was followed as a function of time (Fig. S3A–
D). When adding an excess of duplex DNA as a competitor, longer
incubation times are required to reach platination maxima but
these total amounts of adducts are not affected in the case of Pt-
3
sequence [22] (Fig. 4B and C). In the case of cmyc, Pt-ttpy-N also
N
binds to the external quartets of the G-quadruplex structure [23],
as shown by crosslinking of guanines situated mainly on the
0
0
5
-end (G17-G13-G8-G4) and to a lesser extent on the 3 end
(
G19-G15-G10-G6) (Fig. 4A).
As summarized in Table 1, the azido derivative is able to estab-
3
ttpy-N with both cmyc and 22AG, which confirm the high selec-
lish both platination and photocrosslinks in the two G-quadruplex
structures studied whereas the activity of the benzophenone
derivative is limited to platination. As thymines are the targets of
benzophone radical [9a], we can speculate that the benzophenone
moiety in this case adopts an unfavorable positioning with regard
to the loop thymines of both quadruplexes thereby explaining
absence of reactivity.
tivity of this compound for G-quadruplex DNA (Fig. 5). Platination
by Pt-ttpy-Bn is more affected by the presence of duplex competi-
tor, as shown by the significant decrease of adducts from 20% to
5
0% observed in this case (Fig. 5).
Encouraged by these results, we went one step ahead and eval-
uated the cytotoxicity of these new platinum complexes on two
human cancer cell lines: human ovarian cancer cell line (A2780)
and human colon carcinoma cell line (HT29). The two compounds
showed poor toxicity in the dark (IC50 > 20 lM) (Table 2) whereas
toxicity of Pt-ttpy-Bn increased significantly upon UV/Vis irradia-
tion (320–420 nm) on both cell lines, which was not observed for
Pt-ttpy-N . Whether this difference in activity results from poor
3
cellular uptake or low stability of nitrene intermediates in cellular
environment remains to be determined, in particular in the light of
the extent of platination of the cellular DNA induced by each com-
pound. Nonetheless the phototoxicity observed for Pt-ttpy-Bn
already suggests the formation of the reactive form of the ben-
zophenone moiety inducing trapping of macromolecular targets
in its vicinity.
(
(
A)
B)
In summary, we developed two platinum (II) tolylterpyridine
derivatives, equipped with photoreactive groups, Pt-ttpy-N and
3
Pt-ttpy-Bn. Interestingly these compounds retained the high bind-
ing affinity of Pt-ttpy scaffold for G-quadruplex DNA but with
(
C)
Fig. 4. Photocrosslinking sites on human telomeric repeat sequence 22AG G-
3
Fig. 5. Amount of platinated adducts reached by Pt-ttpy-Bn and Pt-ttpy-N (2 eq.)
quadruplex forming sequence ((A) parallel form and (B) antiparallel form) for Pt-
ttpy-N and (C) major photocrosslinking sites on cmyc G-quadruplex forming
3
sequence. (Of note only parallel and antiparallel forms of 22AG are shown here as
an illustration of photocrosslinking sites but do not account for all conformations in
equilibria in aqueous solution.)
on cmyc and 22AG G-quadruplex forming sequences alone (blue bars) and in the
presence of an excess of duplex DNA competitor (red bars – ds26, 10 eq.). Time
(min) needed to reach the maximum of platination is indicated on each bar. (For
interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033

6
E. Morel et al. / Inorganica Chimica Acta xxx (2016) xxx–xxx
Table 2
TAG
TG
TA
-TAMRA), were purchased at Eurogentec. Labelled
M, 1 eq.) and ligand (1 M, 5 eq.) are
3
3
2
Cytotoxicity (
l
M) in the dark and upon photoactivation with UV–Vis irradiation
oligonucleotides (0.2
l
l
2
(
5 J/cm ) of Pt-ttpy-Bn and Pt-ttpy-N
3
on A2780 and HT29 cell lines.
degassed by sonication. After a first equilibration step (25 °C,
5 min), a stepwise increase of 0.5 °C per minute to reach 95 °C
was performed. Final data were analyzed with Excel (Microsoft
Corp.) and Origin Pro 8.6 (OriginLab Corp.). The emission of
6-FAM was normalized (0–1) and T was defined as the tempera-
½
ture for which the normalized emission is 0.5.
means of 2–4 experiments.
A2780
HT29
Dark
UVA
Dark
UVA
Cellular viability (%)
Untreated cells
Cells treated by
DMSO 0.1%
100%
104%
80%
78%
98%
102%
88%
84%
m
DT values are
IC50
(
lM)
Pt-ttpy-N
Pt-ttpy-Bn
3
>20
>20
20
9.20
>20
>20
20
14.49
3.3. Gel electrophoresis
improved selectivity likely due to the steric hindrance afforded by
the photocrosslinking moieties that should block insertion into
duplex DNA. Both complexes show efficient and selective platina-
tion of the two G-quadruplex structures studied, derived from the
cmyc oncogene promoter and from the human telomeric
sequences. Upon photoactivation, double anchoring into the
quadruplex structure has been shown for the azido derivative
0
The oligonucleotides were 5 -end-labelled using a polynu-
cleotide kinase and [ 32P]-ATP (Perkin Helmer). The reaction
c
products were purified by electrophoresis on 20% acrylamide dena-
turing gel. Samples platination were prepared by folding a mixture
0
of 5 -end-radiolabeled DNA and 10
l
M of non-radiolabeled mate-
L in 50 mM KClO buffered solution.
The folding was achieved by heating the samples at 90 °C for
min, followed by slow cooling to room temperature over the
rial to a total volume of 18
l
4
3
Pt-ttpy-N resulting from platination and photoinduced alkylation.
5
The benzophenone derivative appeared unable to establish cross-
links with the two quadruplexes studied, which might be depen-
dant on the poor accessibility of thymines residues. Thus this
point remains to be explored and rationalized with a larger set of
G-quadruplex matrices of different sequence and topology.
Whether these compounds are able to perform double anchorage
of G-quadruplex DNA or can trap proteins interacting on G-
quadruplex DNA in complex environments like cells or cellular
extracts is still an open question, but already they represent molec-
ular tools with great potential for studying G-quadruplex DNA
biology.
course of 2 h to induce the formation of the quadruplex structure.
It was then incubated with 2 equivalents of platinum complexes
for 5 min or 2 h, and platinated products were separated by elec-
trophoresis on 15% denaturing gel. Half of the samples were pho-
toactivated. Bands obtained from Pt-ttpy-Bn were eluted from
gel, precipitated, treated by 3 -exonuclease at 37 °C for 30 min
and loaded on a 20% denaturing gel. Bands obtained from
3
Pt-ttpy-N were eluted from gel, precipitated and treated with
piperidine. Gels were scanned using a STORM860 (GE Healthcare).
Gel analysis was done using Image Quant Software.
0
3
. Experimental
3.4. Photoactivation experiments
Stock solution of these complexes (2 mM in DMSO, stored at
10 °C) were used for G4-FID and FRET-melting assays and ali-
Samples for irradiations were prepared by folding for G-quadru-
0
ꢁ
plex DNA. A mixture of 5 -end-radiolabeled DNA and 10
lM of
quoted to avoid freeze–thaw cycles. Stock solution of thiazole
orange (2 mM in DMSO, stored at ꢁ10 °C) was used for G4-FID
assay. Oligonucleotides purified by reversed-phase HPLC were pur-
chased from Eurogentec (Belgium). Fluorescence measurements
were performed on a FluoroMax-3 spectrophotometer.
nonradiolabeled material to a total volume of 18
l
L each in
50 mM KClO buffered solution. Folded oligonucleotides were then
4
incubated with 2 equivalents of platinum complexes for 5 min at
37 °C. Samples were irradiated in centrifuge tubes (2 mL, Eppen-
dorf). The tubes were placed into a reflector bowl filled with ice
and the samples were irradiated with a studio flash lamp (Alien-
bees B 1600 640 WS without UV filter, 3 cm distance between
flashtube and samples). Half of the samples were loaded on a
15% denaturing gel electrophoresis right after photoactivation
and the other half of samples was loaded after 2 h of post-incuba-
tion at 37 °C.
3.1. G4-FID experiments
+
Experiments were performed in 1 mL cuvettes. K 100 buffer
was used, containing 10 mM lithium cacodylate and 100 mM KCl.
0
0
G4-FID oligonucleotides, 22AG (5 -AG
3
T
2
AG
-3 ) and ds26 (5 -CA
G-3’), were purchased at Eurogentec. 0.25
folded DNA target was mixed with thiazole orange (0.5
G4-DNA and 0.75 M for ds26). Each ligand addition to the cuvette
3 2
T
AG
3
0
T
2
AG
TCGATCGA
M pre-
M for
3
-3 ), cmyc
0
0
(
5 -T
2
GAG
3
TG
GAT
3
TAG
3
3
TG TA
2
2
2
TTCGATC
2
2
l
l
3.5. Cell culture conditions
l
was followed by a 3 min equilibration time, after which the fluo-
rescence spectrum was recorded. Thiazole orange displacement
HT-29 cells (ATCC HTB38), a human colon carcinoma cell line,
were purchased from the American Type Culture Collection (ATCC)
and cultured in Dulbecco’s modified Eagle’s medium GlutaMAX
(DMEM, Life Technologies, Saint-Aubin, France) supplemented
with 10% fetal calf serum (FCS, South America origin, BioWhittaker
Lonza, Verviers, Belgium) and 1% Penicillin/Streptomycin (Life
Technologies) in humidified atmosphere under 5% CO2 in air at
(
%) was calculated as TOD(%) = 100 ꢁ [(FA/FA
is the fluorescence area of TO bound to DNA after each ligand addi-
tion and FA before any ligand addition. By plotting the TOD as a
0
) ꢂ 100], where FA
0
function of the concentration of added ligand, DC50 values are mea-
sured for 50% of TOD and are means of 2–4 experiments.
3
7 °C. A2780 cells, a human ovarian cancer cell line, were pur-
3
.2. FRET-melting experiments
chased from the European Collection of Cell Cultures (ECACC)
and cultured in RPMI-1640 medium (Life Technologies) supple-
mented with 10% fetal calf serum and 1% Penicillin/Streptomycin
+
+
Experiments were performed in 400 lL cuvettes. K 10 and K 1
buffers were used, containing 10 mM lithium cacodylate, 10 or
mM KCl, and 90 or 99 mM LiCl. FRET-melting oligonucleotides,
F21T (FAM-G [T AG -TAMRA) and FmycT (FAM-T GAG TG
3
2
in humidified atmosphere under 5% CO in air at 37 °C. Cells were
subcultured twice a week by dispersal with TrypLE Express
Enzyme (Life Technologies).
1
3
2
3
]
3
2
3
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033

E. Morel et al. / Inorganica Chimica Acta xxx (2016) xxx–xxx
7
3.6. Photocytotoxicity assays
[4] (a) O. Cohen-Fix, C. Ribeyre, J. Lopes, J.-B. Boulé, A. Piazza, A. Guédin, V.A.
Zakian, J.-L. Mergny, A. Nicolas, PLoS Genet. 5 (5) (2009) e1000475;
(
b) S.K. Bharti, J.A. Sommers, F. George, J. Kuper, F. Hamon, K. Shin-ya, M.-P.
HT-29 and A2780 cells were seeded in 24-well plates at a den-
Teulade-Fichou, C. Kisker, R.M. Brosh, J. Biol. Chem. 288 (39) (2013) 28217;
(c) O. Mendoza, N.M. Gueddouda, J.-B. Boulé, A. Bourdoncle, J.-L. Mergny,
Nucleic Acids Res. (2015).
5] (a) A. Singh, E.R. Thornton, F.H. Westheimer, J. Biol. Chem. 237 (9) (1962)
PC3006;
4
sity of 5 ꢂ 10 cells/well in 1 mL of culture medium. After 24 h of
incubation at 37 °C, compound to test, in stock solution in dimethyl
sulfoxide (Sigma–Aldrich), was diluted in culture medium in the
[
dark at the following final concentrations 5, 10 and 20
l
M or 0.1,
(b) T. Mizuhara, S. Oishi, H. Ohno, K. Shimura, M. Matsuoka, N. Fujii, Bioorg.
Med. Chem. 21 (7) (2013) 2079;
0
.5 and 1 M in duplicate. 1 mL of culture medium was replaced
l
(
c) M.M. Rowland, H.E. Bostic, D. Gong, A.E. Speers, N. Lucas, W. Cho, B.F.
Cravatt, M.D. Best, Biochemistry 50 (51) (2011) 11143.
[6] E.R. Guggenheim, D. Xu, C.X. Zhang, P.V. Chang, S.J. Lippard, ChemBioChem 10
1) (2009) 141.
with the culture medium containing the compound to test in each
well. Two plates of each compound and for each cell line were real-
ized, one for dark cytotoxicity and the other for photocytotoxicity.
After 24 h of incubation with compound at 37 °C in the dark, the
cells were washed with phosphate buffered saline (PBS, Life Tech-
nologies). Irradiation was performed in PBS with a UVA1 Sellamed
sytem, emitting in the spectral range from 340 to 420 nm as
described by Bracchitta et al. [24]. Irradiation was carried out dur-
(
[
7] (a) T. Lemarteleur, D. Gomez, R. Paterski, E. Mandine, P. Mailliet, J.-F. Riou,
Biochem. Biophys. Res. Commun. 323 (3) (2004) 802;
(b) G. Pennarun, C. Granotier, L.R. Gauthier, D. Gomez, F. Hoffschir, E. Mandine,
J.-F. Riou, J.-L. Mergny, P. Mailliet, F.D. Boussin, Oncogene 24 (18) (2005) 2917.
8] F. Doria, M. Nadai, M. Folini, M. Scalabrin, L. Germani, G. Sattin, M. Mella, M.
Palumbo, N. Zaffaroni, D. Fabris, M. Freccero, S.N. Richter, Eur. J. 19 (1) (2013)
78;
(b) M. Di Antonio, K.I.E. McLuckie, S. Balasubramanian, J. Am. Chem. Soc. 136
(16) (2014) 5860.
9] (a) D. Verga, F. Hamon, F. Poyer, S. Bombard, M.-P. Teulade-Fichou, Angew.
Chem., Int. Ed. 53 (4) (2014) 994;
[
2
ing 150 s at a final irradiance of 33.33 mW/cm to reach a delivery
2
fluence of 5 J/cm UVA. Different controls were realized as follow,
[
unirradiated and untreated cells, unirradiated and 0.1 0.2 and
0
.4% DMSO treated cells, irradiated and untreated cells and finally
(b) M. Nadai, F. Doria, L. Germani, S.N. Richter, M. Freccero, Chemistry 21 (6)
(
2015) 2330.
irradiated and DMSO treated cells.
[
10] (a) H. Bertrand, S. Bombard, D. Monchaud, E. Talbot, A. Guédin, J.-L. Mergny, R.
Grünert, P.J. Bednarski, M.-P. Teulade-Fichou, Org. Biomol. Chem. 7 (14) (2009)
2864;
M. Trajkovski, E. Morel, F. Hamon, S. Bombard, M.-P. Teulade-Fichou, J. Plavec,
Eur. J. 21 (21) (2015) 7798.
11] (a) T. Bach, Synthesis 1998 (05) (1998) 683;
After irradiation, PBS was removed and fresh medium free of
drug was added. Plates were left to incubate at 37 °C in the dark
for 3 days before evaluation of the cell viability by determination
of mitochondrial activity using the 3-[4,5-dimethylthiazol-2-yl]-
[
2
,5-diphenyl tetrazolium bromide assay (MTT, Sigma). At the time
of counting, 50 L of a MTT solution at 5 mg/mL was added to each
well. After 30 min of incubation and removal of the medium, for-
mazan crystals were taken up with 600 L of DMSO and absor-
(b) W. Sander, D. Grote, S. Kossmann, F. Neese, J. Am. Chem. Soc. 130 (13)
(
2008) 4396.
l
[
12] G.D. Prestwich, G. Dormán, J.T. Elliott, D.M. Marecak, A. Chaudhary,
Photochem. Photobiol. 65 (2) (1997) 222.
[13] (a) C. Dalhoff, M. Hüben, T. Lenz, P. Poot, E. Nordhoff, H. Köster, E. Weinhold,
ChemBioChem 11 (2) (2010) 256;
l
bance at 562 nm was measured with a Fluostar microplate
reader (BMG Labtech).
(
b) L.R. Odell, N. Chau, A. Mariana, M.E. Graham, P.J. Robinson, A. McCluskey,
ChemMedChem 4 (7) (2009) 1182.
[
14] P.L. Gilbert, D.E. Graves, J.B. Chaires, Biochemistry 30 (45) (1991) 10925.
Acknowledgements
[15] F. Kröhnke, W. Zecher, J. Curtze, D. Drechsler, K. Pfleghar, K.E. Schnalke, W.
Weis, Angew. Chem., Int. Ed. Engl. 1 (12) (1962) 626.
[
16] E. Largy, F. Hamon, M.-P. Teulade-Fichou, Anal. Bioanal. Chem. 400 (10) (2011)
419.
The authors thanks COST CM1105 and the ‘Institut National Du
Cancer’ (INCa) and the Canceropôle Île de France for supporting
this work.
3
[17] A. Decian, L. Guittat, M. Kaiser, B. Sacca, S. Amrane, A. Bourdoncle, P. Alberti,
M.-P. Teulade-Fichou, L. Lacroix, J.-L. Mergny, Methods 42 (2) (2007) 183.
[
[
18] E. Morel Ph.D. thesis, 2015.
19] H. Bertrand, D. Monchaud, A. De Cian, R. Guillot, J.-L. Mergny, M.-P. Teulade-
Fichou, Org. Biomol. Chem. 5 (16) (2007) 2555.
Appendix A. Supplementary material
[
[
20] K. Nakatani, T. Yoshida, I. Saito, J. Am. Chem. Soc. 124 (10) (2002) 2118.
21] S. Redon, S. Bombard, M.-A. Elizondo-Riojas, J.-C. Chottard, Nucleic Acids Res.
31 (6) (2003) 1605.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2016.02.033.
[22] (a) A.T. Phan, K.N. Luu, D.J. Patel, Nucleic Acids Res. 34 (19) (2006) 5715;
(
(
(
(
b) K.N. Luu, A.T. Phan, V. Kuryavyi, L. Lacroix, D.J. Patel, J. Am. Chem. Soc. 128
30) (2006) 9963;
c) S. Neidle, G. Parkinson, Nat. Rev. Drug Discovery 1 (5) (2002) 383;
d) A.T. Phan, V. Kuryavyi, K.N. Luu, D.J. Patel, Nucleic Acids Res. 35 (19) (2007)
References
[
[
1] P. Murat, S. Balasubramanian, Curr. Opin. Genet. Dev. 25 (2014) 22.
2] A. Piazza, M. Adrian, F. Samazan, B. Heddi, F. Hamon, A. Serero, J. Lopes, M.-P.
Teulade-Fichou, A.T. Phan, A. Nicolas, EMBO J. 34 (12) (2015) 1718.
3] V.S. Chambers, G. Marsico, J.M. Boutell, M. Di Antonio, G.P. Smith, S.
Balasubramanian, Nat. Biotech. (2015). Advance online publication.
6517.
[23] A. Ambrus, D. Chen, J. Dai, R.A. Jones, D. Yang, Biochemistry 44 (6) (2005) 2048.
[24] G. Bracchitta, A. Catalfo, G. De Guidi, Photochem. Photobiol. Sci. 12 (6) (2013)
967.
[
Please cite this article in press as: E. Morel et al., Inorg. Chim. Acta (2016), http://dx.doi.org/10.1016/j.ica.2016.02.033