Conjugating a groove-binding motif to an Ir(iii) complex for the enhancement of G-quadruplex probe behavior

Article abstract of DOI:10.1039/c6sc00001k

In this study, the reported G-quadruplex groove binder benzo[d,e]isoquinoline was linked to a cyclometallated Ir(iii) complex to generate a highly selective DNA probe 1 that retains the favorable photophysical properties of the parent complex. The linked complex 1 showed advantages of both parent complex 2 and groove binder 3. Similar to 3, the conjugated complex 1 exhibits a superior affinity and selectivity for G-quadruplex DNA over other conformations of DNA or proteins, with the fold enhancement ratio obviously improved compared with parent complex 2. The molecular modelling revealed a groove-binding mode between complex 1 and G-quadruplex. Meanwhile 1 also possesses the prominent advantages of transition metal complex probes such as a large Stokes shift and long lifetime phosphorescence, which could be recognized in strong fluorescence media through time-resolved emission spectroscopy (TRES). We then employed 1 to develop a detection assay for AGR2, a potential cancer biomarker, as a "proof-of-principle" demonstration of the application of a linked complex for DNA-based detection in diluted fetal bovine serum. We anticipate that this conjugation method may be further employed in the development of DNA probes and have applications in label-free DNA-based diagnostic platforms.

Full text of DOI:10.1039/c6sc00001k

View Article Online
View Journal
Chemical
Science
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Wang, Z. Mao,
T. Kang, C. Wong, J. Mergny, C. Leung and M. Dik-Lung, Chem. Sci., 2016, DOI: 10.1039/C6SC00001K.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/chemicalscience

View Article Online
DOI: 10.1039/C6SC00001K
Chemical Science
Page 1 ofJ7ournal Name
Dynamic Article Links ►
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Conjugating a groove-binding motif to Ir(III) complex for the
enhancement of G-quadruplex probe behavior†
Modi Wang,‡a Zhifeng Mao,‡a Tian-Shu Kang,b Chun-Yuen Wong,c Jean-Louis Mergny,*de Chung-Hang
Leung*b and Dik-Lung Ma*a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
In this study, the reported G-quadruplex groove binder benzo[d,e]isoquinoline was linked to a
cyclometallated Ir(III) complex to generate a highly selective DNA probe 1 that retains the favorable
photophysical properties of the parent complex. The linked complex 1 showed advantages of both parent
10 complex 2 and groove binder 3. Similar with 3, the conjugated complex 1 exhibits superior affinity and
selectivity for G-quadruplex DNA over other conformations of DNA or protein, with the fold
enhancement ratio obviously improved comparing with the parent complex 2. The molecular modelling
revealed a groove-binding mode between complex 1 and G-quadruplex. Meanwhile 1 also possesses the
prominent advantage of transition metal complex probe such as large Stokes shift and long lifetime
15 phosphorescence, which could be recognized in strong fluorescence media through time-resolved
emission spectra (TRES) measurements. We then employed 1 to develop a detection assay for AGR2, a
potential cancer biomarker, as a “proof-of-principle” demonstration of the application of a linked
complex for DNA-based detection in diluted fetal bovine serum. We anticipate that this conjugating
method may be further employed in the development of DNA probes and the application in label-free
20 DNA-based diagnostic platform.
quadruplex DNA structure.27-33 Transition metal complexes have
Introduction
several useful qualities in this application. Their ligands can be
varied in order to tune their photophysical properties and
50 interactions with target biomolecules. Their phosphorescence
The anterior gradient homolog 2 (AGR2) protein plays an
important role in the proliferation and migration of tumor cells,1
and the detection of AGR2 is very useful in cancer diagnosis.
25 Common lengthy duration of measurement, which requires
several days for sample translation and analysis, can increase the
mental burden of patients during testing,2 prompting a need for
alternative detection methods. Oligonucleotide-based detection
has received wide attention in the recent literature.3-21 It depends
30 on the ability of an oligonucleotide to change conformation upon
exposure to the target. The new conformation can be converted
into a measurable signal by appropriate signal transducers. One
recent example is an oligonucleotide-based detection assay for
AGR2 using fluorescently-labeled DNA.2, 22 In this assay, AGR2
35 induces a structural change of the DNA oligonucleotide, thus
altering the distance between the fluorophore and quencher.2, 22
This approach has the disadvantage that fluorescently-labeled
oligonucleotides are relatively expensive, thus more economical
alternative may possibly be found in a label-free strategy that
40 utilizes unmodified DNA and selective DNA probes. Several
organic dyes have been reported as DNA probes, such as
thioflavin T (ThT),23, 24 crystal violet (CV)25 and thiazole orange
(TO),26 some of which can recognize G-quadruplex (G4) DNA
via the end stacking of G-quartet.
generally has
a long emission lifetime, which can be
distinguished from highly fluorescent media through the use of
time-resolved emission spectra (TRES) measurements.
Additionally, the large Stokes shifts of metal complexes can
55 reduce self-quenching.34, 35 Our group has previously reported a
series of cyclometallated Ir(III) complexes as selective G-
quadruplex probes and demonstrated the optimization of their
photophysical and G-quadruplex-binding properties via changing
the auxiliary N^N or C^N ligands.36 However, as structural
60 changes affect both G-quadruplex-binding affinity and
photophysical properties, it can be sometimes difficult to improve
the G-quadruplex recognition abilities of a metal complex
without adversely influencing its photophysical characteristics.
In this study, we sought to design and synthesize a
65 benzo[d,e]isoquinoline-linked
Ir(III)
complex
1
by
functionalizing a parent luminescent Ir(III) complex 2 with a
known G-quadruplex groove binder 3 (Fig. 1).37, 38 The linked
complex 1 can therefore be considered to be comprised of a
signaling unit, the Ir(III) complex 2, linked to the recognition unit,
70 the benzo[d,e]isoquinoline motif 3. In our design, the recognition
unit associates with G-quadruplex DNA, altering the environment
of the probe. The signalling unit is sensitive to this change, and a
corresponding change in luminescent response can be measured.
45
More recently, transition metal complexes have been
explored as selective probes for DNA, in particular the G-
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

View Article Online
DOI: 10.1039/C6SC00001K
Chemical Science
Page 2 of 7
(a)
(b)
Scheme 1. Synthesis of linked Ir(III) complex 1. Reagents and
conditions: a) Boc2O, CH2Cl2, 0 to 25 oC, 3 h, 76%; b) methyl 4-
o
(bromomethyl)benzoate, K2CO3, CH3CN, 0 to 25 C, overnight, 91%;
c) TFA, CH2Cl2, 25 oC, overnight, 95%; d) 1,3-dibromopropane,
o
o
K2CO3, DMF, 40 to 50 C, overnight, 76%; e) K2CO3, DMF, 80 C, 6
h, 72%; f) LiOH, H2O, MeOH, reflux, 2 h, 99%; g) 2-chloro-N-(1,10-
phenanthrolin-5-yl)acetamide, K2CO3, DMF, 65 oC, overnight, 75%;
h) dichloro-bridged [Ir(ppy)2Cl]2, MeOH:CH2Cl2
=
1:1, 25 oC,
overnight, 70%. DMF = N,N-dimethylformamide, Boc2O = di-tert-
butyl dicarbonate, TFA = trifluoroacetic acid.
Fig. 1. (a) Chemical structure of Ir(III) complex 1‒2 and the G-
quadruplex loop binder 3. (b) Schematic diagram delineates the AGR2
sensing platform utilizing the DNA binder linked Ir(III) complex 1.
25 fluorescence resonance energy transfer (FRET) melting assay
indicated that 5 μM of 2 did not stabilize F21T G-quadruplex
DNA (Fig. S2). This suggests that complex 2 binds weakly to the
G-quadruplex motif, which could account for its low luminescent
enhancement towards G-quadruplex DNA.
Thus, our desired outcome for the complex 1 was to (1) localize
the binding area of the metal complex to G-quadruplex DNA,
thus increasing selectivity; and (2) strengthen the binding affinity
of the Ir(III) complex for G-quadruplex DNA while maintaining
its desirable photophysical properties. And we finally applied
Ir(III) complex 1 to construct a AGR2 detection assay to
demonstrate the proof-of-concept application of the Ir(III)
complex functionalized with a G-quadruplex binder.
30
As all known G-quadruplex structures are characterized by
grooves which are structurally and chemically very different from
the minor groove of the dsDNA, we decided to link a G-
quadruplex groove-binding motif to the parent complex 2. In
2008, Ma and co-workers reported a drug-like G-quadruplex
5
Results and Discussion
35 binding ligand found through high throughput virtual screening
(VS).39 In the following year Randazzo and co-workers
incorporated the VS and NMR experiments to identify a G-
quadruplex groove binder from 6,000 compounds.37 Following
this study, in 2012 our group identified the carbamide40 motif
40 from a natrual product and natural product–like compound
database of over 20,000 compounds through VS. To our
knowledge, there are few other examples of utilizing a structure-
based virtual screening approach to screen a large number of
compounds for G-quadruplex groove binders. As these scaffords
45 were screened from a large database, we decided to choose the
10 To demonstrate the additional G-quadruplex binding affinity
contributed by 3, we chose the cyclometallated complex 2 as our
parent complex, as 2 shows minimal selectivity for G-quadruplex
DNA. As a G-quadruplex-based detection platform generally
exploits the switching of DNA from ssDNA or dsDNA to G-
15 quadruplex,
the
high
G-quadruplex/ssDNA
or
G-
quadruplex/dsDNA luminescence enhancement ratios of G-
quadruplex probe can increase the sensitivity of the detection
platform. Also real biological samples usually contain matrix
ssDNA or dsDNA as common interferences, thus high
20 enhancement ratio is crucial in real applications. Complex 2
displayed ca. 1.3 to 1.5 luminescence enhancement ratio for
ckit87up, ckit1 and Pu27 G-quadruplexes over double-stranded
DNA, while ca. 1.5 luminescence enhancement ratio over single-
stranded DNA (ssDNA) was observed (Fig. S1a). Moreover, a
recognition
unit
from
these
two
studies.
The
benzo[d,e]isoquinoline scaffold 3 reported by Randazzo’s group
was chosen as the G-quadruplex recognition motif because it
could be easily attached to the N^N donor ligand using simple
2
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C6SC00001K
Page 3 of 7
Chemical Science
(Fig. S4b). Furthermore, the presence of 10 μM of competitor
DNA did not significantly affect the melting temperature
enhancement of G-quadruplex DNA in the presence of 1 (Fig.
S4c). The result indicates that complex 1 exhibited a stronger
30 affinity towards G-quadruplex than 2, which is attributed to the
attachment of the G-quadruplex-binding motif 3.
In emission titration experiments, 1 displayed ca. 3.5
luminescence enhancement ratio for ckit87up, ckit1 and Pu27 G-
quadruplexes over ssDNA, while ca. 3.2 luminescence
35 enhancement ratio over dsDNA was observed (Fig. S1a). In
emission titration experiments,
luminescence enhancement ratio for ckit87up, ckit1 and Pu27 G-
quadruplexes over ssDNA, while ca. 3.2 luminescence
1
displayed
a
ca. 3.5
a
enhancement ratio over dsDNA was observed (Fig. S1a). This is
40 comparable to organic G-quadruplex probes, which exhibited a
high fold-enrichment for G-quadruplex DNA and moderate
enhancement for ssDNA, dsDNA23, 41 Meanwhile, modified ThT
analogues have been reported by Kuwahara’s group with high G-
quadruplex/ssDNA or G-quadruplex/dsDNA enhancement
45 ratios.24
Consistent with other Ir(III) complexes, 1 displayed a ca.
240 nm Stokes shift which is 3.7-fold larger than that of ThT, and
a microsecond lifetime (Table S2) which in principle could
distinguish the long lifetime luminescence of 1 from the high
50 autofluorescence of the surrounding sample matrix environment
by TRES. This is particularly important in the application of the
sensing probe in real samples. In order to demonstrate that we
could identify the luminescence of 1 in high autofluorescence
samples, rhodamine B (RhB), coumarin (Cm) and perylene (PY)
55 were introduced into 1 as sources of autofluorescence. In the
steady-state emission spectra, PY displayed an emission peak
located at the 450 nm region while the emission of 1 was
observed clearly at 590 nm. Meanwhile, the strong peak of Cm is
located at 420 nm with a tail that extends up to 650 nm. In this
60 scenario, the peak of 1 is significantly influenced by the peak tail.
Meanwhile, the RhB‒1 mixture displayed only one peak,
presumably due to the similar peak maximum of RhB and 1. For
TRES measurements, the time gate is defined as the time after the
complete fluorescence decay of the three organic dyes. Upon
65 TRES measurement, no emission peak corresponding to Cm and
PY was observed and the emission peak of 1 had become
dominant. For the RhB‒1 mixture, the peak intensity was reduced
upon TRES measurement. We believe the reduced peak should
attribute to 1 only, since the fluorescence of RhB was completely
70 decayed before TRES measurement (Fig. 2). The results indicate
an advantage to our long lifetime luminescent complex 1, namely
that the luminescence of 1 could be identified by TRES
Fig. 2. Steady-state photoluminescence and TRES of 1 in the
presence of (a, b) perylene, (c, d) coumarin and (e‒g) rhodamine B
fluorescent media.
synthetic protocol. The selective G-quadruplex-binding of 3 was
validated by FRET melting (Fig. S3).
The Ir(III) complex 1 was synthesized by reaction of a
precursor complex with a phenanthroline N^N ligand derivatized
with 3, as depicted in Scheme 1. The first step was the protection
of piperazine 1a with Boc2O to give 1b, followed by treatment
with methyl 4-(bromomethyl)benzoate to give intermediate 1c.
After deprotection of the Boc group with trifluoroacetic acid, the
resulting compound 1d was reacted with compound 1f, which
10 was obtained by the alkylation of naphthalimide 1e with 1,3-
dibromopropane, to afford derivative 1g in 72% yield. After
hydrolysis of the methyl ester with LiOH, the resulting acid 1h
was immediately reacted without further isolation with 2-chloro-
N-(1,10-phenanthrolin-5-yl)acetamide to yield the 1,10-
15 phenanthroline derivative 1i. Finally 1i was reacted with half an
equivalent of the organometallated dimer [Ir(ppy)2Cl]2, followed
by anion exchange with NH4PF6, giving the Ir(III) complex 1 in
70% yield. The structures of the compounds were confirmed by
NMR spectroscopy, elemental analysis and mass spectrometry
20 (see the Supporting Information).
5
measurement and potentially be applied in
autofluorescence sample matrix.
a
strong
75
The superior luminescence and binding selectivity of 1
compared to the parent complex 2 demonstrates that the
attachment of the G-quadruplex-binding benzo[d,e]isoquinoline
motif successfully enhanced the interaction between the metal
complex with G-quadruplex DNA. To examine the binding area
FRET-melting assays revealed that complex 1 selectively
stabilizes G-quadruplex DNA. The melting temperature of F21T
G-quadruplex DNA was increased by 13 °C upon addition of 5
μM of 1 (Fig. S4a), while no significant increase of melting
25 temperature was observed for dsDNA at the same concentration
80 of complex 1, we investigated the luminescence response of
complex 1 towards G-quadruplexes containing different sizes of
3-side loops (Fig. S5). The G-quadruplex structure in this
experiment has been investigated in the literature.42 The
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

View Article Online
DOI: 10.1039/C6SC00001K
Chemical Science
Page 4 of 7
Fig. 4. Luminescence enhancement of the system with different numbers
of hybridized bases in the presence of 100 nM AGR2.
while the binder linked 1 showed no significant enhancement
(Fig. S1b). This may be due to the larger three-dimensional
nature of the conjugated complex 1, which may reduce non-
45 specific binding of the complex towards proteins.
Fig. 3. Side view of the interactions of 1 with G-quadruplex structure in
hypothetical molecular model. The G-quadruplex is depicted as a ribbon
representation (green), while
1 is depicted as a space-filling
representation showing carbon (beige), oxygen (red) and nitrogen
(blue).
Given the encouraging selectivity of complex 1 towards G-
quadruplex DNA compared to other types of DNA or proteins,
we employed 1 to construct a G-quadruplex-based AGR2 sensing
platform to demonstrate the proof-of-concept application of a
50 linked Ir(III) complex for DNA-based sensing. The AGR2
sensing platform is depicted schematically in Fig. 1b. Probe A
containing a G-quadruplex-forming sequence and the AGR2
aptamer sequence was partially hybridized to its antisense
sequence (Probe B) to form a duplex substrate. The addition of
55 AGR2 will cause the dissociation of the duplex substrate due to
the strong binding of AGR2 to its aptamer sequence. The
liberated G-quadruplex sequence will then fold into a G-
quadruplex structure upon the addition of potassium ions, and
will be recognized by 1 with a strong luminescence response. In
60 this design, Probe A, which contains both the G-quadruplex and
AGR2 aptamer sequences, was hybridized by the antisense DNA,
Probe B. This Probe B, which hybridized only a part of the
aptamer and G-quadruplex, resists the folding of G-quadruplex in
the presence of potassium ions. Meanwhile the length of the
65 Probe B strand is important for the sensitivity. An excessively
long Probe B may cause the probe DNA duplex to become
extraordinary stable and resist the dissociation by aptamer‒target
binding, or prevent the G-quadruplex sequence from folding to
the G-quadruplex after the probe DNA duplex binds to the AGR2.
70 Meanwhile, an excessively short Probe B may not fully prevent
the folding of Probe A, resulting in a background signal increase.
As a result, the shortest ckit1 G-quadruplex was selected for
AGR2 sensing along the three highest fold enrichment G-
quadruplexes.
luminescence enhancement of complex 1 increased with loop size
from 2 nucleotides (nt) and reached a maximum enhancement at
6 nt. This suggests that the G-quadruplex loop is involved in the
binding interaction with 1, which is consistent with the previous
observation that motif 3 mainly recognizes the 3-side of G-
quadruplex grooves.37 Finally, 1 showed a weaker luminescence
response towards an intermolecular G-quadruplex (5-TG4T-3)
compared to an intramolecular ckit1 G-quadruplex, which we
attributed to the lack of loops in an intermolecular G-quadruplex
5
10 for recognition by 1. Possibly however, all grooves in the
tetramolecular complex are identical, and different from the other
G-quadruplex (Fig. S6).
The G-quadruplex binding of 1 was further examined
utilizing the Molsoft ICM method (ICM-Pro 3.6-1d molecular
15 docking software).43 Firstly, the geometry of 1 was optimized
using density functional theory (DFT) calculations. We initially
studied the binding of complex 1 to an intramolecular (3+1) G-
quadruplex with a long central loop (PDB: 2LOD)44 to study the
loop binding behaviour of 1. In the low energy binding pose
20 between 1 and G-quadruplex, 1 is predicted to interact with the
A3‒T4 loop as well as the G1‒G3, G7‒G8 groove (Fig. 3). While
the binding unit 3 was also predicted to inteact with the bases of
G17 and C19 in the 3-side loop. Since no salt-bridge interaction
or hydrogen bonding was predicted in this model, we anticipate
25 that the interaction between G-quadruplex and 1 is mainly
attributed to the hydrophobic interaction between the groove and
the binding motif 3. To further investigate the G-quadruplex-
binding mode of 1, we calculated the predicted binding pose of 1
towards ckit1 (PDB: 4WO3)45 and human telomeric G-
30 quadruplex (PDB: 1KF1).46 The lowest-energy binding poses
revealed that 1 binds to the loop region of both ckit1 and human
telomeric G-quadruplex structures (Fig. S7). Complex 1 is
predicted to interact with the T12‒A13 loop as well as the G14‒
G16 groove of human telomeric G-quadruplex. Additionally, 1
35 was predicted to interact with C9 and A19 in the loop as well as
the G6‒G8 groove of the ckit1 G-quadruplex. The molecular
docking results further validate the groove-binding behaviour of 1
that is conferred by the recognition unit 3.
75
To demonstrate the feasibility of our sensing platform, 80
nM of AGR2 was introduced into a solution containing 0.5 μM
hybridized duplex substrate. After incubation at 37 °C for 45 min,
KCl was added to facilitate the formation of the G-quadruplex,
followed by the addition of 0.5 μM 1. Encouragingly, we
80 observed that the system showed an obvious luminescence
enhancement in the presence of AGR2 (Fig. S8). To improve the
performance of the sensing platform, we further optimized
several experimental parameters that are relevant for this assay.
We found that the relative luminescence enhancement of the
85 system was highly dependent on the concentration of complex 1,
with a maximal response obtained at 0.5 μM of complex.
Furthermore, we also optimized the DNA and potassium
Interestingly, 1 showed reduced luminescent enhancement
40 for AGR2 protein compared to the parent complex 2. Complex 2
showed a ca. 1.6-fold enhancement in the presence of AGR2,
4
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C6SC00001K
Page 5 of 7
Chemical Science
Fig. 6. (a) Relative luminescence intensity of the system (path B) in the
presence of 50 nM AGR2 or 250 nM other proteins. (b) Luminescence
response of the system (path B) in the presence of increasing
concentrations of AGR2 in 2.5 % (v/v) FBS.
Fig. 5. Linear plot of the change in luminescence intensity at λ = 585
nm vs. AGR2 concentration using the sensing mechanism path B.
45 the duplex substrate induced the formation of a positive band at
about 260 nm and a slight negative peak at about 235 nm, which
are characteristic signals for parallel G-quadruplex DNA. Taken
together, these data suggest that the luminescence enhancement
of the system originated from the specific interaction of 1 with
50 the G-quadruplex motif, which is in turn a consequence of the
dissociation of the duplex DNA substrate caused by the specific
binding of AGR2 (Fig. S12).
To further improve the sensitivity of the AGR2 detection
assay, we set out to reduce the background signal of the assay. A
55 significant contribution to the background signal is due to
undesirable binding of 1 to ssDNA. Thus we introduced enzymes
capable of DNA digestion. We employed exonuclease III (ExoIII)
concentration for better sentitivity (Fig. S9).
After optimization, the length of Probe B was also found
to be important for the sensitivity of the assay. If Probe B is too
long, the duplex substrate may be too stable and resist
dissociation upon the addition of AGR2, thus impacting the
detection limit. On the other hand, an excessively short Probe B
sequence may permit partial dissociation of the duplex substrate
even in the absence of AGR2, thus raising the background signal.
We found that the fold enhancement of the system was maximum
5
10 when the length of Probe B was 18 bases (Fig. 4).
After optimization of the experimental parameters, we
investigated the luminescence response of the system to different
concentrations of AGR2. Encouragingly, the luminescence
intensity was increased as more AGR2 was added (Fig. S10). The
15 limit of detection of the assay (path A) was 10 nM at a signal-to-
noise ratio of 3, and a linear detection range of target AGR2 from
10 to 60 nM with a maximum 2.7-fold enhancement was
observed.
and exonuclease
I (ExoI) to catalyze the digestion of
mononucleotides from the 3-hydroxyl end of dsDNA and ssDNA,
60 respectively.47, 48 The assay is performed in a similar fashion as
previously described (path A in Fig. 1), but with the additional
presence of ExoIII and ExoI (path B). Antisense DNA released
after the dissocation of the duplex substrate, as well as the intact
duplex DNA, will be digested by ExoI and ExoIII, respectively.
65 In contrast, the folded G-quadruplex structure and the AGR2-
bound aptamer sequence resists digestion by ExoIII and ExoI. As
a consequence, the background signal caused by undesirable
binding of complex 1 to ssDNA and dsDNA will be eliminated.
Using the modified method (path B), the maximum fold-
70 enhancement was increased to 3.5-fold from (2.6-fold), while the
detection limit for AGR2 was lowered to 1 nM with a linear
range of detection from 1 to 28 nM (Fig. 5 and S13).
Additionally, to evaluate the selectivity of this assay, we
investigated the response of the system to human neutrophil
75 elastase (HNE), immunoglobulin G (IgG), insulin and tumor
necrosis factor alpha (TNF-α). AGR2 protein induced the highest
luminescence enhancement of the system among the various
substances under study (Fig. 6a), indicating the high selectivity of
this aptamer-based assay for the detection of AGR2. We attribute
80 the high selectivity of this assay to the high specificity of the
AGR2 aptamer for its cognate target.
If 1 was added to AGR2 only, no enhanced luminescence
20 was observed (Fig. S11a), indicating that 1 did not directly
interact with AGR2. Thus, we envisage that the luminescence
enhancement of 1 was due to the specific binding of AGR2 to its
aptamer, which promotes the dissociation of the duplex substrate
and the formation of the G-quadruplex motif in the presence of
25 K+ ions that is recognized by 1. To further verify the mechanism
of this assay, we designed mutant probe DNA sequences that are
unable to form a G-quadruplex structure and are also unable to
bind AGR2, due to the lack of key guanine residues and specific
mutations in the aptamer sequence. No significant change in the
30 luminescence signal of 1 was displayed when mutant Probe A
and Probe B were incubated with AGR2 (Fig. S11b). Another
control experiment using mutant Probe Amutant2 that bears an
intact AGR2 aptamer segment but cannot form the ckit1 G-
quadruplex was carried out. No significant enhancement was
35 observed even upon addition of 80 nM AGR2 (Fig. S11c). This
demonstrates that the signal enhancement originates from the
formation of the ckit1 G-quadruplex, rather than the AGR2-
aptamer complex. This result suggests that the formation of the
ckit1 G-quadruplex motif was important for the luminescent
40 enhancement of the system. Circular dichroism (CD)
spectroscopy was further performed to demonstrate the expected
DNA conformational change in this assay. Firstly, the addition of
AGR2 into buffer alone did not produce a significant change in
CD signal (data not shown). However, the addition of AGR2 to
We next examined the suitability of the assay to detect
AGR2 in a biological matrix. We performed the detection assay
in a sensing system containing 2.5% (v/v) fetal bovine serum. We
85 observed that the 1/DNA system showed enhanced luminescence
intensity in diluted fetal bovine serum as the concentration of
AGR2 was increased (Fig. 6b). This result demonstrates the
possibility of this approach for the quantification of AGR2 in
serum for cancer diagnosis.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

View Article Online
DOI: 10.1039/C6SC00001K
Chemical Science
Page 6 of 7
couplings. The following abbreviations are used for convenience
in reporting the multiplicity of NMR resonances: s, singlet; d,
doublet; t, triplet; q, quartet; m, multiplet; br, broad. All NMR
data was acquired and processed using standard Bruker software
60 (Topspin).
Conclusions
In conclusion, we have successfully linked
a known G-
quadruplex groove binder, benzo[d,e]isoquinoline motif, to a
Ir(III) complex, thus generating a highly selective G-quadruplex
probe 1 which showed advantages of both parent complex 2 and
groove binder 3. Similar with 3, the conjugated complex 1
exhibits superior affinity and selectivity for G-quadruplex DNA
over other conformations of DNA or protein, with the fold
enhancement ratio obviously improved comparing with the parent
10 complex 2. The molecular modelling revealed a groove-binding
mode between complex 1 and G-quadruplex. Meanwhile it also
possesses the prominent advantage of transition metal complex
probe such as large Stokes shift and long lifetime
phosphorescence. We successfully employed time-resolved
15 emission spectra (TRES) measurements to demonstrate the
detectability of long lifetime luminescence of 1 in the strong
fluorescence media. We then employed 1 to develop a G-
quadruplex-based sensing system for the detection of AGR2, a
potential serum biomarker for cancer, as a “proof-of-principle”
20 concept. A detection limit of 1 nM for AGR2 was achieved by
using this label-free method with superior selectivity over a
variety of other proteins, and the assay could function effectively
for AGR2 detection in diluted fetal bovine serum. We anticipate
that this conjugating method may be further employed for the
25 development of G-quadruplex probes as well as the detection of
specific biomarkers associated with various human diseases.
5
FRET melting assay. The ability of the 1 to stabilize G-
quadruplex DNA was investigated using
resonance energy transfer (FRET) melting assay. The labelled G-
quadruplex-forming oligonucleotide F21T (5′-FAM-
65 d(G3[T2AG3]3)-TAMRA-3′; donor fluorophore FAM: 6-
carboxyfluorescein; acceptor fluorophore TAMRA: 6-
a fluorescence
carboxytetramethylrhodamine) was diluted to 200 nM in a
potassium cacodylate buffer (100 mM KCl, pH 7.0), and then
heated to 95 °C in the presence of the indicated concentrations of
70 1. The labeled duplex-forming oligonucleotide F10T (5′-FAM-
dTATAGCTA-HEG-TATAGCTATAT-TAMRA-3′)
(HEG
linker: [(-CH2-CH2-O-)6]) was treated in the same manner, except
that the buffer was changed to 10 mM lithium cacodylate (pH
7.4). Fluorescence readings were taken at intervals of 0.5 °C over
75 the range of 25 to 95 °C.
Acknowledgements
This work is supported by Hong Kong Baptist University
(FRG2/14-15/004), the Health and Medical Research Fund
80 (HMRF/14130522), the Research Grants Council (HKBU/201811,
HKBU/204612 and HKBU/201913), the French Agence
Nationale de la Recherche/Research Grants Council Joint
Research Scheme (A-HKBU201/12; Oligoswitch ANR-12-IS07-
0001), State Key Laboratory of Environmental and Biological
85 Analysis Research Grant (SKLP-14-15-P001), National Natural
Science Foundation of China (21575121), Guangdong Province
Natural Science Foundation (2015A030313816), Hong Kong
Baptist University Century Club Sponsorship Scheme 2015,
Interdisciplinary Research Matching Scheme (RC-IRMS/14-
90 15/06), the State Key Laboratory of Synthetic Chemistry, the
Science and Technology Development Fund, Macao SAR
(103/2012/A3 and 098/2014/A2), the University of Macau
(MYRG091(Y3-L2)-ICMS12-LCH, MYRG2015-00137-ICMS-
QRCM and MRG023/LCH/2013/ICMS) and Conseil régional
95 d'Aquitaine, France.
Experimental section
Materials. Immunoglobulin G, insulin, coumarin, perylene,
rhodamine B and other reagents, unless specified, were purchased
30 from Sigma Aldrich (St. Louis, MO) and used as received.
Iridium chloride hydrate (IrCl3.xH2O) was purchased from
Precious Metals Online (Australia). All oligonucleotides were
synthesized by Techdragon Inc. (Hong Kong, China). Fetal
bovine serum 10270 (GIBCO® , origin: South America, EU
35 approved origin) was purchased from Life Technologies (Grand
Island, NY, USA). Human tumor necrosis factor-alpha was
purchased from Sangon Biotech (Shanghai, China). Human
neutrophil elastase was purchased from Innovative Research
(Novi, MI,USA).
40 General experimental. Mass spectrometry was performed at the
Mass Spectroscopy Unit at the Department of Chemistry, Hong
Kong Baptist University, Hong Kong (China). Deuterated
solvents for NMR purposes were obtained from Armar and used
as received. Circular dichroism (CD) spectra were collected on a
45 JASCO-815 spectrometer. Steady state emission spectra were
recorded by QM-4 Photon Technology International while TRES
was measured by a Horiba Fluorolog TCSPC spectrophotometer.
1H and 13C NMR were recorded on a Bruker Avance 400
Notes and references
a Department of Chemistry, Hong Kong Baptist University, Kowloon Tong,
Hong Kong, China. Email: edmondma@hkbu.edu.hk
bState Key Laboratory of Quality Research in Chinese Medicine, Institute
100 of Chinese Medical Sciences, University of Macau, Macao, China. Email:
duncanleung@umac.mo
cDepartment of Biology and Chemistry, City University of Hong Kong,
Kowloon Tong, Hong Kong, China
dUniversity of Bordeaux, ARNA laboratory, Bordeaux, France. Email:
105 jean-louis.mergny@inserm.fr
spectrometer operating at 400 MHz (1H) and 100 MHz (13C). H
1
50 and 13C chemical shifts were referenced internally to solvent shift
(acetone-d6: 1H  2.05, 13C  29.8; CD3Cl: 1H C76.8).
Chemical shifts (are quoted in ppm, the downfield direction
being defined as positive. Uncertainties in chemical shifts are
typically ±0.01 ppm for 1H and ±0.05 for 13C. Coupling constants
55 are typically ± 0.1 Hz for 1H-1H and ±0.5 Hz for 1H-13C
eINSERM, U869, IECB, Pessac, France.
‡ These authors contributed equally to this work.
† Electronic Supplementary Information (ESI) available: compound
characterisation and supplementary data. See DOI: 10.1039/b000000x/
110
1. Y. Zhang, T. Z. Ali, H. Zhou, D. R. D'Souza, Y. Lu, J. Jaffe, Z. Liu,
A. Passaniti and A. W. Hamburger, Cancer Res., 2010, 70, 240-248.
6
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C6SC00001K
Page 7 of 7
Chemical Science
2. T. Shi, Y. Gao, S. I. Quek, T. L. Fillmore, C. D. Nicora, D. Su, R.
29. N. H. Abd Karim, O. Mendoza, A. Shivalingam, A. J. Thompson, S.
Ghosh, M. K. Kuimova and R. Vilar, RSC Adv., 2014, 4, 3355-3363.
30. J. E. Reed, A. A. Arnal, S. Neidle and R. Vilar, J. Am. Chem. Soc.,
2006, 128, 5992-5993.
Zhao, J. Kagan, S. Srivastava, K. D. Rodland, T. Liu, R. D. Smith, D.
W. Chan, D. G. Camp, A. Y. Liu and W.-J. Qian, J. Prot. Res., 2014,
13, 875-882.
60
5 3. Y. Imaizumi, Y. Kasahara, H. Fujita, S. Kitadume, H. Ozaki, T.
Endoh, M. Kuwahara and N. Sugimoto, J. Am. Chem. Soc., 2013,
135, 9412-9419.
31. K. Suntharalingam, A. J. P. White and R. Vilar, Inorg. Chem., 2009,
48, 9427-9435.
32. N. H. Campbell, N. H. A. Karim, G. N. Parkinson, M. Gunaratnam,
V. Petrucci, A. K. Todd, R. Vilar and S. Neidle, J. Med. Chem., 2012,
55, 209-222.
4. A. Shoji, M. Kuwahara, H. Ozaki and H. Sawai, J. Am. Chem. Soc.,
2007, 129, 1456-1464.
65
10 5. Y. Tao, M. Li, J. Ren and X. Qu, Chemical Society Reviews, 2015,
DOI: 10.1039/C5CS00607D.
33. A. Arola-Arnal, J. Benet-Buchholz, S. Neidle and R. Vilar, Inorg.
Chem., 2008, 47, 11910-11919.
6. L. Wu and X. Qu, Chemical Society Reviews, 2015, 44, 2963-2997.
7. D. Li, S. Song and C. Fan, Acc. Chem. Res., 2010, 43, 631-641.
34. J. Liu, Y. Liu, Q. Liu, C. Li, L. Sun and F. Li, J. Am. Chem. Soc.,
2011, 133, 15276-15279.
8. H. Pei, X. Zuo, D. Zhu, Q. Huang and C. Fan, Acc. Chem. Res., 2014, 70 35. C. Li, M. Yu, Y. Sun, Y. Wu, C. Huang and F. Li, J. Am. Chem. Soc.,
15
20
25
30
35
47, 550-559.
2011, 133, 11231-11239.
9. C. Fan, K. W. Plaxco and A. J. Heeger, Proc. Natl. Acad. Sci. U. S.
A., 2003, 100, 9134-9137.
36. D.-L. Ma, D. S.-H. Chan and C.-H. Leung, Acc. Chem. Res., 2014, 47,
3614-3631.
10. A. A. Lubin and K. W. Plaxco, Acc. Chem. Res., 2010, 43, 496-505.
11. X. Liu, C.-H. Lu and I. Willner, Acc. Chem. Res., 2014, 47, 1673-
1680.
37. S. Cosconati, L. Marinelli, R. Trotta, A. Virno, L. Mayol, E.
Novellino, A. J. Olson and A. Randazzo, J. Am. Chem. Soc., 2009,
131, 16336-16337.
75
12. X. Liu, F. Wang, R. Aizen, O. Yehezkeli and I. Willner, J. Am. Chem.
Soc., 2013, 135, 11832-11839.
38. B. Pagano, J. Amato, N. Iaccarino, C. Cingolani, P. Zizza, A.
Biroccio, E. Novellino and A. Randazzo, ChemMedChem, 2015, 10,
640-649.
13. S. De Tito, F. Morvan, A. Meyer, J.-J. Vasseur, A. Cummaro, L.
Petraccone, B. Pagano, E. Novellino, A. Randazzo, C. Giancola and
D. Montesarchio, Bioconjug. Chem., 2013, 24, 1917-1927.
14. D. Musumeci, J. Amato, A. Randazzo, E. Novellino, C. Giancola, D.
Montesarchio and B. Pagano, Anal. Chem., 2014, 86, 4126-4130.
15. B. Pagano, L. Margarucci, P. Zizza, J. Amato, N. Iaccarino, C.
Cassiano, E. Salvati, E. Novellino, A. Biroccio, A. Casapullo and A.
Randazzo, Chem. Commun., 2015, 51, 2964-2967.
80 39. D.-L. Ma, T.-S. Lai, F.-Y. Chan, W.-H. Chung, R. Abagyan, Y.-C.
Leung and K.-Y. Wong, ChemMedChem, 2008, 3, 881-884.
40. D.-L. Ma, D. S.-H. Chan, W.-C. Fu, H.-Z. He, H. Yang, S.-C. Yan
and C.-H. Leung, PLoS ONE, 2012, 7, e43278.
41. A. Renaud de la Faverie, A. Guédin, A. Bedrat, L. A. Yatsunyk and
85
J.-L. Mergny, Nucleic Acids Res., 2014, 42, e65.
42. A. Guédin, J. Gros, P. Alberti and J.-L. Mergny, Nucleic Acids Res.,
2010, 38, 7858-7868.
16. X. Liu, R. Freeman, E. Golub and I. Willner, ACS Nano, 2011, 5,
7648-7655.
43. M. Totrov and R. Abagyan, Proteins: Structure, Function, and
Bioinformatics, 1997, 29, 215-220.
17. C.-H. Lu, B. Willner and I. Willner, ACS Nano, 2013, 7, 8320-8332.
18. N. Enkin, F. Wang, E. Sharon, H. B. Albada and I. Willner, ACS 90 44. M. Marušič, P. Šket, L. Bauer, V. Viglasky and J. Plavec, Nucleic
Nano, 2014, 8, 11666-11673.
Acids Res., 2012, 40, 6946-6956.
19. Y. Xiao, X. Qu, K. W. Plaxco and A. J. Heeger, J. Am. Chem. Soc.,
2007, 129, 11896-11897.
45. D. Wei, J. Husby and S. Neidle, Nucleic Acids Res., 2015, 43, 629-
644.
20. C. Zhao, L. Wu, J. Ren, Y. Xu and X. Qu, J. Am. Chem. Soc., 2013,
135, 18786-18789.
46. G. N. Parkinson, M. P. H. Lee and S. Neidle, Nature, 2002, 417, 876-
95
880.
40 21. Y. Kasahara, Y. Irisawa, H. Fujita, A. Yahara, H. Ozaki, S. Obika
and M. Kuwahara, Anal. Chem., 2013, 85, 4961-4967.
47. C. Song and M. Zhao, Anal. Chem., 2009, 81, 1383-1388.
48. X. Su, X. Zhu, C. Zhang, X. Xiao and M. Zhao, Anal. Chem., 2012,
84, 5059-5065.
22. J. Wu, C. Wang, X. Li, Y. Song, W. Wang, C. Li, J. Hu, Z. Zhu, J. Li,
W. Zhang, Z. Lu and C. J. Yang, PLoS ONE, 2012, 7, e46393.
23. J. Mohanty, N. Barooah, V. Dhamodharan, S. Harikrishna, P. I.
100
45
Pradeepkumar and A. C. Bhasikuttan, J. Am. Chem. Soc., 2013, 135,
367-376.
24. Y. Kataoka, H. Fujita, Y. Kasahara, T. Yoshihara, S. Tobita and M.
Kuwahara, Anal. Chem., 2014, 86, 12078-12084.
25. D.-M. Kong, J.-H. Guo, W. Yang, Y.-E. Ma and H.-X. Shen, Biosens.
Bioelectron., 2009, 25, 88-93.
50
26. I. Lubitz, D. Zikich and A. Kotlyar, Biochemistry, 2010, 49, 3567-
3574.
27. S. Shi, J. Zhao, X. Geng, T. Yao, H. Huang, T. Liu, L. Zheng, Z. Li,
D. Yang and L. Ji, Dalton Trans., 2010, 39, 2490-2493.
55 28. J.-L. Yao, X. Gao, W. Sun, S. Shi and T.-M. Yao, Dalton Trans.,
2013, 42, 5661-5672.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7