A nanobiosensor for the simple detection of small molecules using non-crosslinking aggregation of gold nanoparticles with G-quadruplexes

Surachada Chuaychob ab, Chongdee Thammakhet-Buranachai cd, Proespichaya Kanatharana cd, Panote Thavarungkul de, Chittanon Buranachai de, Masahiro Fujita *b and Mizuo Maeda *ab
aDepartment of Advanced Materials Science, Graduate School of Frontier Sciences, The University of Tokyo, Kashiwanoha 5-1-5, Kashiwa-shi, Chiba 277-8561, Japan. E-mail: mizuo@riken.jp
bBioengineering Laboratory, RIKEN Cluster for Pioneering Research, Hirosawa 2-1, Wako-shi, Saitama 351-0198, Japan. E-mail: mfujita@riken.jp
cDepartment of Chemistry, Faculty of Science, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand
dCenter of Excellence for Trace Analysis and Biosensor, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand
eDepartment of Physics, Faculty of Science, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand

Received 4th October 2019 , Accepted 4th December 2019

First published on 6th December 2019

This work demonstrates a simple and specific colorimetric sensor for a hazardous small molecule, cisplatin, using a G-quadruplex (G4) DNA as a sensing probe and non-crosslinking aggregation of gold nanoparticles (AuNPs) as a signal enhancer. AuNPs functionalized with G4 strands dispersed stably in the colloidal system. The colloidal solution colour changed in a short time from red to purple-blue after cisplatin was added to the system. The cisplatin molecule can bind at the N7 guanine base of G4, resulting in the formation of a cisplatin–DNA adduct. The adduct formed does not act as a cross-linker between the particles, but probably causes a reduction in steric repulsion, associated with unfolding of the G4 strand. This induces non-crosslinking aggregation of particles. The plasmon absorption peak shift arising from the aggregation—that is, the degree of colour change—varied with the cisplatin concentration. This sensor showed a good analytical performance with linearity from 15.0 to 30.0 μM and a detection limit of 12.9 μM. Moreover, the response of the colour change to cisplatin was found to be faster than those for other analogues such as carboplatin and oxaliplatin, indicating that this system has a high specificity for cisplatin detection.


Colorimetric sensors have gained much attention in recent years because they are suitable for simple, fast, and on-site detection.1 Gold nanoparticles (AuNPs) possess remarkable optical properties and are widely used in colorimetric detection. AuNPs exhibit localized surface plasmon resonance in the visible region with a high extinction coefficient.2 Colloidal AuNPs with a diameter of a few tens of nanometres are vivid red in colour. Because the plasmon resonance depends on interparticle plasmon coupling in addition to particle size, the particle aggregation causes a bathochromic shift from red to purple-blue.3–5 Owing to this unique optical property, target-induced aggregation of AuNPs is a critical event for colorimetric sensing.6 The aggregation is achieved primarily by a chemical interaction between the modified AuNPs, e.g., binding of a recognition element to target analytes such as sugar, heavy metals, cancerous cells, etc.7–9 Among them, DNA-functionalized AuNPs have been well studied. Pioneering work on DNA detection using DNA-functionalized AuNPs was carried out by Mirkin and coworkers.10 AuNPs functionalized with single-stranded DNA (ssDNA) show a high colloidal stability. The hybridization with a target DNA sequence induces a DNA-linked particle aggregation, in which the DNA duplex crosslinks the particles.10,11 A colorimetric change occurs from ruby red to blue due to a surface plasmon shift. On the other hand, our group reported another aggregation behaviour in response to such analytes, called non-crosslinking aggregation, which is caused by a change in the physical properties of the particle surfaces.12,13 The high colloidal stability of ssDNA–AuNPs is due to their electrostatic and steric repulsions, which are ascribed to the negatively charged and flexible DNA strands. Hybridization to their fully complementary DNA targets (in sequence as well as in chain length), however, causes the particles to aggregate. The decrease of the entropic effect by switching from flexible ssDNA to rigid double-stranded DNA (dsDNA) is likely responsible for this type of aggregation, in addition to the salt screening.14 This is in contrast to the DNA-linked particle aggregation.10,11 Interestingly, on the other hand, the particles remained dispersed after the hybridization with one base-mismatched DNA. The difference in molecular motion between the base-pairing (fully matched dsDNA) and -unpairing (dsDNA with a mismatch at one of the terminal bases) is assumed to determine the colloidal stability. This phenomenon can be exploited to detect single nucleotide mutations as a colorimetric change.15 This interfacial phenomenon is thus expected to be useful for the simple and rapid detection of single-base mutations.16

It is possible that other DNA structural changes resulting in a reduction of entropic repulsion would induce the non-crosslinking aggregation. In this work, we focused on G-quadruplex (G4) DNA as a probe; the structure of G4 DNA was first reported by Davies and co-workers in 1962.17,18 G4 adopts a four-stranded helical structure with stacks of planar guanine tetrads known as G-quartets, which are formed via Hoogsteen base-pairing, and is stabilized by a monovalent cation inserted between the quartets (Fig. 1).19In vivo, G4 structures can be found in telomeres,20 which play an important role in maintaining the genomic viability. It is now believed that the telomere shortening in each cycle of cell division is responsible for aging and cell death. In cancer cells, however, the telomere shortening is prevented by up-regulated telomerase activity, so the cells can divide indefinitely.21 Therefore, antitumor drugs and their derivatives that are capable of forming adducts with guanines, e.g., cisplatin, may be key to effectively treating cancer by disrupting the telomeres.

image file: c9ay02150g-f1.tif
Fig. 1 (A) Structure of a G-quartet. The yellow and green beads represent the sugar-phosphate residue and the monovalent cation (K+), respectively. (B) Three different G4 structures with three stacks of G-quartets are illustrated, the chair, basket, and propeller types, which correspond respectively to 21CTA, 22AG, and 35B1 (Kras) investigated in this study.

Cisplatin or cis-diamminedichloroplatinum(II) [Pt(NH3)2(Cl)2] is composed of only 11 atoms and is one of the most powerful antineoplastic drugs; it was approved by the United States Food and Drug Administration (U.S. FDA) in 1978.22,23 The cisplatin molecule can bind at N7 of guanine, one of the two purine bases.24–26 Positively charged active di-aquated cisplatin [Pt(NH3)2(H2O)2]2+ coordinates with either one (mono-adduct) or two guanine bases (di-adduct). The mono-adduct formation occurs rapidly while the di-adduct requires a much longer time to form and binds predominantly to the dsDNA target.24,25,27 The formation of these adducts deforms the helices and disrupts DNA replication, resulting in programmed cell death. A high concentration of cisplatin (800 μM), however, induces cell necrosis, giving rise to severe side effects, although a low concentration (8 μM) can cause cell apoptosis.28 The toxicity of cisplatin, which is suspected of being carcinogenic, is well known because cisplatin is non-selective,29 affecting both cancerous and non-cancerous cells, and thus its side effects have been well-documented. Environmental release of cisplatin, e.g., in the urine of patients, has also been a serious problem.30,31 For all these reasons, there is a need for a reliable monitoring system for cisplatin.

To date, several methods for detecting cisplatin by chromatography,32 spectrometry,33 fluorometry,34 and electrochemistry35 have been reported. Those approaches have, however, been limited by the complicated and/or long processes required to prepare the samples, and the expensive equipment used to analyse them. In this study, we propose a simple and rapid colorimetric sensor with high specificity for cisplatin using oligo-DNA strands with guanine-rich sequences capable of forming G4 structures and AuNPs. The formation of a cisplatin–DNA adduct will lead to unfolding of the G4 structure because cisplatin binds to the N7 atom of guanine, which stabilizes the G-quartet via Hoogsteen hydrogen bonding.36–39 The cisplatin binding might be able to induce the reduction in colloidal stability, or non-crosslinking aggregation of the particle. Thus, it is possible to realize cisplatin detection by the naked eye in a short time.



AuNPs with diameters of about 15 or 40 nm were purchased from British Biocell International (Cardiff, UK). The real sizes of AuNPs were estimated by transmission electron microscopy (TEM) (JEM-1230, JEOL, Tokyo) operated at 80 kV (Fig. S1). The mean radii of 15 and 40 nm AuNPs were 7.47 (±0.37) and 19.0 (±1.6) nm, respectively. Oligo-DNA having a sulfhydryl group at its 5′-end was purchased from Eurofins Genomics K.K. (Tokyo), and purified using an affinity column (oligonucleotide purification cartridge (OPC); Thermo Fisher Scientific, Waltham, MA) prior to use. Three different types of G4 conformation were examined—the chair- (21CTA), basket- (22AG), and propeller-type (35B1 (Kras)) conformations—which incorporated 21, 22, and 35 nucleotides (nt).40–43 These conformations are shown in Fig. 1 and their sequences are listed in Table S1. Cisplatin (pharmaceutical secondary standard grade), oxaliplatin (European Pharmacopoeia Reference Standard), silver nitrate (AgNO3), ethylenediaminetetraacetic acid (EDTA), dithiothreitol (DTT), and carboplatin (analytical grade) were purchased from Sigma-Aldrich (St. Louis, MO). Phosphoric acid (H3PO4) was purchased from Junsei Chemical (Tokyo). Dipotassium hydrogen phosphate (K2HPO4) and sodium nitrate (NaNO3) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Milli-Q water (resistivity greater than 18.2 MΩ cm) was used for all experiments.

Preparation of diaqua-cisplatin, -carboplatin, and -oxaliplatin

Cisplatin (1.0 dichloro-form equivalent) was hydrolysed to generate the active diaqua-form by reacting with 2.0 equivalents of AgNO3 in aqueous media at room temperature for 15 h in the dark.37 In this step, chloride anions dissociated from cisplatin were precipitated as silver chloride (AgCl) to prevent the back conversion of cisplatin into the dichloro-form (Scheme S1). After 15 h incubation at 25 °C, AgCl was removed by two centrifugations at 15[thin space (1/6-em)]000 rpm for 10 min at 4 °C each. The diaqua-cisplatin solution was kept in the dark at 4 °C before use. Carboplatin and oxaliplatin were dissolved in water to obtain their hydrolysed forms, and these solutions were also kept in the dark at 4 °C before use.

Preparation of G4–AuNPs

After the purification, 3 nmol of the oligo-DNAs for 35-nt G4 or 6 nmol of the oligo-DNAs for 21- and 22-nt G4s was added to an aqueous solution of AuNPs with a concentration of one optical density (OD) at 520 nm (2.3 × 10−9 and 1.5 × 10−10 M of 15 and 40 nm AuNPs, respectively), and then the solution was kept at 50 °C in the dark overnight. Next, the solvent was replaced with 10 mM phosphate buffer (K2HPO4/H3PO4) (PB) (pH 5.0) containing 1.5 mM EDTA. The potassium ion is essential for quadruplex conformation. After subsequent overnight heating at 50 °C, NaNO3 was added at a final concentration of 0.1 M to further facilitate the immobilization of DNA strands. It is should be noted that in the coexistence of potassium and sodium ions no significant conformational change of G4 occurs while some local details of the structure vary.44 The resulting solution was kept at 50 °C for another day. To remove unbound DNA, the solution was centrifuged at 15[thin space (1/6-em)]000 rpm for 30 min at 4 °C, and the precipitate was re-dispersed in the same buffer. The concentration of G4–AuNPs was adjusted to about 1.0 OD at 520 nm, unless otherwise noted.

Characterization of G4–AuNPs

The number of grafted G4 strands on the surface of AuNPs was quantified according to a commercial assay as follows. Briefly, the G4–AuNP solution was heated at 95 °C for 10 min to deform the G4 structure to the single-stranded conformation, followed by cooling down and holding at 4 °C for 10 min in EDTA solution (1.5 mM). DTT was added to a final concentration of 1.0 M to release the strands from the particles.45 After centrifugation at 15[thin space (1/6-em)]000 rpm for 30 min (4 °C), the supernatant was treated using an OliGreen ssDNA Quantitation Assay Kit (Thermo Fisher Scientific, Waltham, MA). The fluorescence of the resulting supernatant was measured with a spectrofluorometer (FP-6500; JASCO, Tokyo). The number of G4 strands was then estimated on the basis of a calibration line. The colloidal stability (dispersion/aggregation) was estimated by UV-vis spectroscopy (UV-2550; Shimadzu, Kyoto, Japan) at 25 °C. The extinction spectra were scanned from 300 to 800 nm. The size and zeta potential of G4–AuNPs were measured with a Zetasizer Nano-ZS instrument (Malvern Instruments, Malvern, UK) at 25 °C.

The size and the aggregation behaviour of G4–AuNPs were also examined by small angle X-ray scattering (SAXS). The SAXS measurements were performed at BL45XU RIKEN Structural beamline I of SPring-8, Japan. The X-ray wavelength (λ) was 0.1 nm. The camera length, calibrated using a silver behenate standard, was about 3.5 m. SAXS images were recorded with a PILATUS 3 × 2 M detector (Dectris, Baden, Switzerland) at 25 °C. The images were converted into one-dimensional intensity profiles by circular-averaging, as a function of scattering vector q = (4π[thin space (1/6-em)]sin[thin space (1/6-em)]θ)/λ, where 2θ is the scattering angle. The data were converted to the absolute intensity according to a published method.46 The scattering intensity I(q) is usually expressed by

I(q) ∝ P(q)S(q)(1)
where P(q) and S(q) are the form and structure factors of the scattering object, respectively. If the objects disperse without interfering with each other, S(q) is equal to one. For a spherical particle with radius R, the form factor is given by
image file: c9ay02150g-t1.tif(2)
where D(R) is the distribution function of R. In this study, the Gaussian distribution was applied as D(R). The mean radius of G4–AuNPs was evaluated by curve fitting with eqn (2). The structure factor S(q) is associated with the interference between the particles. We obtained the experimental S(q) by dividing I(q) of the aggregated particles by that of the dispersed particles. Curve fitting with the calculated S(q) on the basis of paracrystalline theory was performed using a similar method as in our previous study.14 A detailed description of the theoretical S(q) can be found in ref. 47 and 48. By analysing S(q), we derived structural information such as the interparticle distance in the aggregates.

Results and discussion

G4–AuNP and its feasibility as a sensing element

The numbers of grafted DNA strands immobilized on AuNPs with a diameter of 40 nm were estimated at 98, 104, and 89 for the 21-, 22- and 35-nt G4s, respectively. Compared with the case of ssDNA,13,14 the graft density for the G4 strands was quite low, likely due to the sterically bulky G4 conformation. The resulting G4–AuNPs showed a high colloidal stability because of the electrostatic and steric repulsions originating from the negative charge of the phosphate group and the conformational freedom of the DNA strands, especially the loop part of the G4 structure in this case. When cisplatin was added to the colloidal system, the stability of G4–AuNPs decreased. However, high salt concentrations were required to observe the effect of cisplatin on the changes of colloidal stability. The addition of cisplatin at a low concentration of 0.1 M NaNO3 did not result in any change, with the exception of a slight decrease in surface potential (Fig. S2), implying the formation of the DNA–cisplatin adduct. On the other hand, the addition of cisplatin at a high salt concentration of 1.0 M, under which the electrostatic repulsion is known to be negligible, reduced the colloidal stability of G4–AuNPs. The colloidal stability was also found to depend on the concentration of cisplatin, as shown in Fig. 2A. This figure shows the UV-vis spectra of 35-nt G4–AuNPs in 10 mM PB (pH 5.0) containing 1.5 mM EDTA at 1.0 M NaNO3, taken at various cisplatin concentrations. Without cisplatin, the absorption peak was observed at 529 nm, and the corresponding colloidal solution was a clear red colour (Fig. 2B), indicating that G4–AuNPs were still dispersed under the high salt concentration. Upon adding cisplatin to the colloidal solution, the plasmon peak shifted to longer wavelengths (the solution turned purple-blue) in about 10 min. The colour change was easily recognized at cisplatin concentrations above 20 μM. This proved that the addition of cisplatin induces the aggregation of 35-nt G4–AuNPs. The degree of the peak shift or colour change increased with the cisplatin concentration.
image file: c9ay02150g-f2.tif
Fig. 2 (A) UV-vis spectra of 35-nt G4–AuNPs in 10 mM PB (pH 5.0) containing 1.5 mM EDTA at 1.0 M NaNO3, with varied cisplatin concentration. All data were gathered at 10 min after adding cisplatin. (B) The corresponding 35-nt G4–AuNP solutions at various cisplatin concentrations: (a) 0, (b) 10, (c) 15, (d) 20, (e) 25, (f) 30, (g) 40, and (h) 50 μM.

We further investigated the cisplatin concentration-dependency of G4–AuNP dispersion/aggregation by SAXS. As shown in Fig. 3, the SAXS data of 35-nt G4–AuNPs varied with the cisplatin concentration. The SAXS intensity profiles under a low salt concentration of 0.1 M NaNO3 (Fig. 3A) also supported the notion that the particles remained dispersed, independent of cisplatin. The profiles were well fitted with eqn (2), as indicated by the solid curves in this figure. The fitting analysis gave a value of 18.8 nm as the mean radius. This agreed well with TEM data (19.0 nm) (Fig. S1), and was also comparable to the hydrodynamic radius Rh of unmodified AuNPs (20.7 nm) rather than that of 35-nt G4–AuNPs (24.9 nm). This clearly corresponded to the radius of core AuNPs, because the scattering by far heavier gold atoms is dominant.14 At the high salt concentration of 1.0 M NaNO3, interference peaks arising from the particle aggregation appeared in the intensity profiles (Fig. 3B). The peak at around q = 0.15 nm−1 was not distinctly visible for the case of 15 μM cisplatin, but became more distinct as the cisplatin concentration increased. Fig. 3C shows the corresponding structure factors. The structure factors were analysed by assuming that the aggregate adopts a disordered face-centred cubic structure as in our previous study.14 The centre-to-centre distance between the nearest neighbours did not seem to depend on the cisplatin concentration. The mean value was 50.7 nm, and thus the gap between the particle surfaces was estimated at 13.1 nm. The gap is due to the existence of the DNA layers, which prevent further approach of the particles. But the estimated value would be too large if the particles were covered with G4 strands, because the height of a G4 strand with three layers of G-tetrad is only about 0.7 nm, and even if the length of the hexamethylene thiol linker is taken into consideration, the layer would still be only about 1.5 nm high. In fact, the difference in Rh between G4–AuNPs and the unmodified ones, which was comparable to the DNA layer thickness, was 4.2 nm. Hence, the large gap was most likely caused by unfolding of the G4 structure.36,39

image file: c9ay02150g-f3.tif
Fig. 3 SAXS intensity data for 35-nt G4–AuNPs in 10 mM phosphate buffer (pH 5.0) at 0.1 M NaNO3 (A) and 1.0 M NaNO3 (B), and the corresponding structure factors (C). The buffer contained 1.5 mM EDTA. The data were taken at various cisplatin concentrations: 15 (red), 20 (orange), 25 (green), and 30 μM (blue). The solid lines in (A) and (C) are the fitting curves of form and structure factors, respectively.

Two plausible styles of particle aggregation could account for the present results. In the first, cisplatin molecules function as a crosslinker between the particles. Namely, the formation of an interstrand DNA–cisplatin adduct brings about the crosslinking aggregation of the particles.10,11 In previous studies, however, cisplatin basically formed the intrastrand crosslink on neighbouring nucleobases (e.g., guanine–guanine)—i.e., it underwent di-adduct formation—while interstrand crosslink adducts rarely formed.24,25 Moreover, during a short reaction time like that used here (ca. 10 min), most cisplatin molecules bind to only one guanine (i.e., mono-adduct formation).39 It thus seems extremely unlikely that cisplatin induces the crosslinking aggregation of particles. In fact, particle aggregation was not recognized after the platination of G4 without electrostatic screening, as mentioned above (Fig. S2).

In the other style of aggregation, or non-crosslinking aggregation, the reduction in steric repulsion by the deformation of the G4 structure induces the particle aggregation.12–14 This appears to contradict our previous finding that AuNPs covered with flexible ssDNA show a high colloidal stability because of the entropic effect,14 since presumably the unfolded DNA strands would be more flexible like ssDNA. The graft density of DNA might be a key factor to account for the present aggregation behaviour. Compared with the case of immobilization of the usual oligo-ssDNA strands (ca. 400 strands/particle),13,14 the number of grafted G4 strands on AuNPs was much lower, as mentioned above (90–100 strands/particle). As a result, the particles after cisplatin binding would have resembled AuNPs covered with a low density of grafted ssDNA strands. The non-crosslinking aggregation process is schematically illustrated in Fig. 4. Even though ssDNA is flexible, such a coarse DNA layer—the steric effect of which is lower than in the case of a rigid DNA structure—cannot serve as a stabilizer to prevent the particle aggregation.49 Assuming that the pitch between neighbouring nucleotides of ssDNA is 0.43 nm,50 the length of 35-nt DNA in an extended manner is ca. 15 nm. In fact, however, the DNA strands most likely lie and shrink, as supported by the finding that the difference in Rh between the DNA–cisplatin adduct AuNPs and unmodified AuNPs was only 4.0 nm because of the coarse DNA layer. Accordingly, the resulting particles might aggregate with the DNA strands being compressed, as shown in Fig. 4.

image file: c9ay02150g-f4.tif
Fig. 4 A schematic illustration of cisplatin detection by G4–AuNPs. The deformation of G4 by the formation of a cisplatin-monoadduct induces the non-crosslinking aggregation of the colloidal particles.

Effect of particle size, G4-conformation, reaction time, and salt concentration

The colour of the G4–AuNP solution changed with time after adding cisplatin. As shown in Fig. 2, the dispersed particles in the absence of cisplatin showed an absorption peak at 529 nm. The degree of colour change or progress of particle aggregation was thus evaluated by the ratio of the absorption peak at 600 nm to that at 529 nm (A600/A529). The time course of the ratio is shown in Fig. 5A. In this figure, the data for both 15 nm and 40 nm AuNPs immobilized with 35-nt G4 strands are shown. The degree of colour change depended on the size of AuNPs. The larger the particles, the greater the colour change. This is likely because a larger van der Waals attraction arose between larger AuNPs. The use of larger AuNPs was thus more advantageous for fast detection of cisplatin. After 10–20 min, the apparent colour changes were saturated. In the DLS measurements (Fig. 5B), the apparent hydrodynamic diameter (Dh) increased for a long time after adding cisplatin for any size of AuNPs, while the size of the G4–AuNPs did not change in the absence of cisplatin. The degree of colorimetric change was related to the progress of particle aggregation.
image file: c9ay02150g-f5.tif
Fig. 5 (A) Time course of the absorption peak ratio (A600/A529) after adding 50 μM cisplatin for 15 nm and 40 nm AuNPs immobilized with 35-nt G4 in 10 mM PB (pH 5.0) including 1.5 mM EDTA at 1.0 M NaNO3. (B) The corresponding change in hydrodynamic diameter is indicated with the filled symbols. The open symbols indicate data gathered in the absence of cisplatin.

Next, we investigated the effect of G4 conformation on the colloidal stability. Here, two other conformations of G4, antiparallel chair-type (21-nt G4) and antiparallel basket-type (22-nt G4), were tested in addition to 35-nt G4 (see Fig. 1).41–43 It was presumed that different topologies would respond to cisplatin differently due to G4 structural stability and accessibility to guanine N7. The absorption peak ratios (A600/A529) for 21-, 22-, and 35-nt G4–AuNPs (40 nm) are plotted as a function of cisplatin concentration in Fig. 6. The data were taken at 10 min after the addition of cisplatin. The propeller-type conformation (35-nt G4) provided the steepest slope in the concentration range of 20 μM to 40 μM cisplatin, while the chair-type (21-nt G4) and the basket-type (22-nt G4) conformations yielded lower slopes but with wider linear ranges (25 μM to 50 μM cisplatin) (Fig. 6A). When the slope is defined as detection sensitivity, the sensitivities for 35-nt, 22-nt, and 21-nt G4s are 0.038, 0.033, and 0.024 μM−1, respectively. The propeller-type 35-nt G4 exposes a planar guanine tetrad, or the so-called G-quartet, into solution (Fig. 1A). On the other hand, the chair-type 21-nt and the basket-type 22-nt G4s (Fig. 1B) protrude two loops consisting of two bases per loop and three bases per loop into solution, respectively.41–43 Such a loop probably prevents the access of cisplatin to the G-quartet. Cisplatin thus unfolds the propeller-type 35-nt G4 more rapidly than the others, resulting in the highest detection sensitivity.

image file: c9ay02150g-f6.tif
Fig. 6 Effects of G4-conformation (A), reaction time (B), and NaNO3 concentration (C) on cisplatin detection. In (A), the chair-type (21-nt), basket-type (22-nt), and propeller-type (35-nt) G4 strands were immobilized with 40 nm AuNPs. UV-vis spectra of 35-nt G4–AuNPs in 10 mM PB (pH 5.0) including 1.5 mM EDTA and 1.0 M NaNO3 were taken under different cisplatin concentrations. The reaction time was 10 min. The absorption peak ratios (A600/A529) were plotted against the cisplatin concentration, together with fitted lines. In (B), the data were evaluated for 35-nt G4–AuNPs (40 nm) in 10 mM PB (pH 5.0) including 1.5 mM EDTA and 1.0 M NaNO3. In (C), the data were evaluated for 35-nt G4–AuNPs (40 nm) in 10 mM PB (pH 5.0) including 1.5 mM EDTA at various NaNO3 concentrations. The reaction time was 10 min. In all panels, the values indicate the slopes of the fitted lines, or the sensitivity.

The effect of the reaction time, elapsed from cisplatin addition, was also studied, and the sensitivity was similarly evaluated (Fig. 6B). It can be seen that the sensitivity substantially increased with the reaction time up to 10 min, and then remained almost unchanged afterwards. Because we sought the shortest reaction time with high sensitivity, 10 min was chosen as the optimum reaction time. In addition, the effect of salt concentration on the cisplatin detection was examined. The addition of salt induces the non-crosslinking aggregation because the electrostatic repulsion is screened. It was natural that the sensitivity increased with increasing NaNO3 concentration (Fig. 6C). However, the sensitivity at 1.0–1.5 M NaNO3 was high enough to detect cisplatin.

Specificity for cisplatin

Among the platinum-based alkylating agents currently used in clinical treatment, cisplatin is the most popular; it can be used alone or in combination with other derivatives such as carboplatin and oxaliplatin, as shown in Fig. S3.51 The platinum-based cancer drugs share some structural similarities and have the same treatment mechanism. We were naturally interested in the specificity of the fabricated G4–AuNP sensor for cisplatin. Using 40 nm AuNPs functionalized with 35-nt G4, the specificity for cisplatin was investigated. The salt concentration and the reaction time were set at 1.5 M and 10 min. After cisplatin was added under these conditions, the solution colour of 35-nt G4–AuNPs changed to purple-blue. The addition of either carboplatin or oxaliplatin did not lead to any colorimetric changes in the concentration ranges investigated here. Fig. 7 shows the degree of colour change or the sensitivity for the platinum-based drugs and the corresponding image. The particle aggregation of 35-nt G4–AuNPs was induced only by cisplatin, indicating that this colloidal system has a high specificity for cisplatin. The data for the mixture of three drugs with a molar ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]1 are also given in this figure. As we expected, the sensor can single out cisplatin even in the presence of equal concentrations of possible interfering agents.
image file: c9ay02150g-f7.tif
Fig. 7 (A) Absorption peak ratio (A600/A529) for three platinum-based drugs: cisplatin, carboplatin, and oxaliplatin. UV-vis measurements of 35-nt G4–AuNPs (40 nm) in 10 mM PB (pH 5.0) including 1.5 mM EDTA and 1.5 M NaNO3 were performed by varying the drug concentration (c). The reaction time was set at 10 min. In this figure, the data for the mixture with a 1[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]1 ratio (plaid-filled) are also shown. The values indicate the slopes of the fitted lines, or the sensitivity. (B) The corresponding image at c = 30 μM.

The platinum-based drugs used here form the same types of adducts at the same sites, e.g., neighbouring G–G on DNA.51–53 Nevertheless, we found that the present colloidal system shows a high specificity for cisplatin. This must be due to the difference in the kinetics of adduct formation.52–57 Compared with the rate constants for monoadduct formations of carboplatin (1.88 × 10−5 s−1) and oxaliplatin (5.76 × 10−6 s−1), that for cisplatin (1.96 × 10−3 s−1) is much faster.55–57 It is considered that 35-nt G4–AuNPs remained dispersed for carboplatin and oxaliplatin at a short reaction time of 10 min. Therefore, if the reaction time was much longer, the particles would aggregate even for the analogues.

Linear dynamic range and limit of detection

On the basis of UV-vis absorption measurements of G4–AuNPs, the limit of detection was evaluated. From a plot of A600/A529 against wavelength (Fig. S4), the linear range was between 15.0 and 30.0 μM with a slope of 0.040 ± 0.0016 μM−1. The lower limit of detection yLOD was evaluated from the equation yLOD = yblank + 3σblank, where yblank and σblank are the mean signal value and its standard deviation for the blank sample.58 The cisplatin concentration yielding the lower detection limit was estimated at 12.9 μM (Fig. S4). This shows a little bit lower sensitivity in comparison with the conventional techniques but sufficient detection performance to identify cisplatin in urine samples because the normal level of cisplatin in urine for chemotherapy patients administered 50 mg m−2 cisplatin is between 54.3 and 321 μM.32,33


In this study, a simple, rapid and selective colorimetric sensor for a hazardous small molecule, cisplatin, using non-crosslinking aggregation of an oligo-DNA–AuNP conjugate material is demonstrated. G4 DNA structures were used as the sensing probe. The parallel propeller-type 35-nt G4 on the surface of AuNPs with a diameter of 40 nm was promising as a cisplatin sensor. The colloidal stability of AuNPs functionalized with G4 strands was high. The aggregation of G4–AuNPs was found to be induced by cisplatin binding to G4 strands. The colour of the colloidal solution changed from red to blue-purple in the presence of cisplatin, and the colour change was clearly recognized at about 10 min after the addition of cisplatin to the colloidal system. Over this short reaction period, mono-adduct formation of cisplatin is most likely to occur, or at most, intrastrand di-adduct formation. The G4 conformation changes to an ssDNA–cisplatin adduct. The steric repulsion by the resulting coarse DNA layer is insufficient to prevent the particle aggregation, so non-crosslinking aggregation occurs. Cisplatin is thus detected by the colour change of the colloidal solution, which arises from the non-crosslinking aggregation. We conjecture that the particle aggregation is mainly determined by the kinetics of cisplatin binding to the G4 strand. This is a plausible reason for the high specificity for cisplatin, compared with the specificities for analogue structures such as carboplatin and oxaliplatin.

Conflicts of interest

There are no conflicts to declare.


This work was financially supported by Grants-in-Aid for Scientific Research (S) (25220204) and (C) (16K04895) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and by funds from the Development and Promotion of Science and Technology Talented Project (DPST) jointly administered by the Ministry of Science and Technology, the Ministry of Education and the Institute for the Promotion of Teaching Science and Technology (IPST), Thailand. We thank Ms. E. Kiyohara and Mr Y. Morita, RIKEN, for their assistance with the sample preparation and instrument use. We thank Mr W. Thongyod for his assistance with the drawing of the 3D structure of DNA. The synchrotron radiation experiments were performed at BL45XU of SPring-8 with the approval of RIKEN (proposal no. 20170025 and 20180011). We also thank Dr T. Hikima for his help with the SAXS measurements. S. Chuaychob is grateful for financial support from the RIKEN International Program Associate.


  1. S. Sankoh, C. Thammakhet, A. Numnuam, W. Limbut, P. Kanatharana and P. Thavarungkul, 4-mercaptophenylboronic acid functionalized gold nanoparticles for colorimetric sialic acid detection, Biosens. Bioelectron., 2016, 85, 743–750 CrossRef CAS PubMed.
  2. X. Liu, M. Atwater, J. Wang and Q. Huo, Extinction coefficient of gold nanoparticles with different sizes and different capping ligands, Colloids Surf., B, 2007, 58, 3–7 CrossRef CAS PubMed.
  3. K. H. Su, Q. H. Wei, X. Zhang, J. J. Mock, D. R. Smith and S. Shultz, Interparticle coupling effects on plasmon resonances of nanogold particles, Nano Lett., 2003, 3, 1087–1090 CrossRef CAS.
  4. X. Huang and M. A. El-Sayed, Gold nanoparticles: Optical properties and implementations in cancer diagnosis and photothermal therapy, J. Adv. Res., 2010, 1, 13–28 CrossRef.
  5. Y. C. Yeh, B. Creran and V. M. Rotello, Gold nanoparticles: preparation, properties, and applications in bionanotechnology, Nanoscale, 2012, 4, 1871–1880 RSC.
  6. K. Saha, S. S. Agasti, C. Kim, X. Li and V. M. Rotello, Gold nanoparticles in chemical and biological sensing, Chem. Rev., 2012, 112, 2739–2779 CrossRef CAS PubMed.
  7. M. C. Daniel and D. Astruc, Gold nanoparticles: assembly, supramolecular chemistry, quantum-size-related properties, and applications toward biology, catalysis, and nanotechnology, Chem. Rev., 2004, 104, 293–346 CrossRef CAS PubMed.
  8. E. I. Laderman, E. Whitworth, E. Dumaual, M. Jones, A. Hudak, W. Hogrefe, J. Carney and J. Groen, Rapid, sensitive, and specific lateral-flow immunochromatographic point-of-care device for detection of herpes simplex virus type 2-specific immunoglobulin G antibodies in serum and whole blood, Clin. Vaccine Immunol., 2008, 15, 159–163 CrossRef CAS PubMed.
  9. L. Qin, G. Zeng, C. Lai, D. Huang, C. Zhang, P. Xu, T. Hu, X. Liu, M. Cheng, Y. Liu, L. Hu and Y. Zhou, A visual application of gold nanoparticles: Simple, reliable and sensitive detection of kanamycin based on hydrogen-bonding recognition, Sens. Actuators, B, 2017, 243, 946–954 CrossRef CAS.
  10. C. A. Mirkin, R. L. Letsinger, R. C. Mucic and J. J. Storhoff, A DNA-based method for rationally assembling nanoparticles into macroscopic materials, Nature, 1996, 382, 607–609 CrossRef CAS PubMed.
  11. N. L. Rosi and C. A. Mirkin, Nanostructure in biodiagnostics, Chem. Rev., 2005, 105, 1547–1562 CrossRef CAS PubMed.
  12. T. Mori and M. Maeda, Stability change of DNA-carrying colloidal particle induced by hybridization with target DNA, Polym. J., 2002, 34, 624–628 CrossRef CAS.
  13. K. Sato, K. Hosokawa and M. Maeda, Rapid aggregation of gold nanoparticles induced by non-cross-linking DNA hybridization, J. Am. Chem. Soc., 2003, 125, 8102–8103 CrossRef CAS PubMed.
  14. M. Fujita, Y. Katafuchi, K. Ito, N. Kanayama, T. Takarada and M. Maeda, Structural study on gold nanoparticle functionalized with DNA and its non-cross-linking aggregation, J. Colloid Interface Sci., 2012, 368, 629–635 CrossRef CAS PubMed.
  15. K. Sato, K. Hosokawa and M. Maeda, Non-cross-linking goldnanoparticle aggregation as a detection method for single-base substitutions, Nucleic Acids Res., 2005, 33, e4 CrossRef PubMed.
  16. Y. Akiyama, G. Wang, S. Shiraishi, N. Kanayama, T. Takarada and M. Maeda, Rapid naked-eye discrimination of Cytochrome P450 genetic polymorphism through non-crosslinking aggregation of DNA-functionalized golad nanoparticles, ChemistryOpen, 2016, 8, 508–512 CrossRef PubMed.
  17. M. Geller, M. N. Lipsett and D. R. Davies, Helix formation by guanylic acid, Proc. Natl. Acad. Sci. U. S. A., 1962, 48, 2013–2018 CrossRef PubMed.
  18. S. B. Zimmerman, G. H. Cohen and D. R. Davies, X-ray fiber diffraction and model-building study of polyguanylic acid and polyinosinic acid, J. Mol. Biol., 1975, 92, 181–192 CrossRef CAS.
  19. W. Humphrey, A. Dalke and K. Schulten, VMD: Visual molecular dynamics, J. Mol. Graphics, 1996, 14, 33–38 CrossRef CAS.
  20. W. E. Wright, V. M. Tesmer, K. E. Huffman, S. D. Levene and J. W. Shay, Normal human chromosomes have long G-rich telomeric overhangs at one end, Genes Dev., 1997, 11, 2801–2809 CrossRef CAS PubMed.
  21. J. W. Shay and W. E. Wright, Senescence and immortalization: role of telomeres and telomerase, Carcinogenesis, 2005, 26, 867–874 CrossRef CAS PubMed.
  22. L. Kelland, The resurgence of platinum-based cancer chemotherapy, Nat. Rev. Cancer, 2007, 7, 573–584 CrossRef CAS PubMed.
  23. S. Dasari and P. B. Tchounwou, Cisplatin in cancer therapy: molecular mechanisms of action, Eur. J. Pharmacol., 2014, 740, 364–378 CrossRef CAS PubMed.
  24. A. M. J. Fichtinger-Schepman, J. L. van der Veer, J. H. den Hartog, P. H. M. Lohman and J. Reedijk, Adducts of the antitumor drug cis-diamminedichloroplatinum(II) with DNA: formation, Identification, and Quantitation, Biochemistry, 1985, 24, 707–713 CrossRef CAS PubMed.
  25. P. M. Takahara, C. A. Frederick and S. J. Lippard, Crystal structure of the anticancer drug cisplatin bound to duplex DNA, J. Am. Chem. Soc., 1996, 118, 12309–12321 CrossRef CAS.
  26. M. H. Baik, R. A. Friesner and S. J. Lippard, Theoretical study of cisplatin binding to purine bases: why does cisplatin prefer guanine over adenine?, J. Am. Chem. Soc., 2003, 125, 14082–14092 CrossRef CAS PubMed.
  27. F. Legendre, V. Bas, J. Kozelka and J. C. Chottard, A complete kinetic study of GG versus AG plantination suggests that the doubly aquated derivatives of cisplatin are the actual DNA binding species, Chem.–Eur. J., 2000, 6, 2002–2010 CrossRef CAS.
  28. W. Liberthal, V. Triaca and J. Levine, Mechanisms of death induced by cisplatin in proximal tubular epithelial cells: Apoptosis vs. necrosis, Am. J. Physiol., 1996, 270, F700–F708 Search PubMed.
  29. L. K. Gorsic, A. L. Stark, H. E. Wheeler, S. S. Wong, H. K. Im and M. E. Dolan, EPS8 inhibition increases cisplatin sensitivity in lung cancer cells, PLoS One, 2013, 8, e82220 CrossRef.
  30. K. Lenz, G. Koellensperger, S. Hann, N. Weissenbacher, S. N. Mahnik and M. Fuerhacker, Fate of cancerostatic platinum compounds in biological wastewater treatment of hospital effluents, Chemosphere, 2007, 69, 1765–1774 CrossRef CAS.
  31. L. Curtis, A. Turner, N. Vyas and G. Sewell, Speciation and reactivity of cisplatin in river water and seawater, Environ. Sci. Technol., 2010, 44, 3345–3350 CrossRef CAS PubMed.
  32. C. M. Riley, L. A. Sternson, A. J. Repta and R. W. Siegler, High-performance liquid chromatography of platinum complexes on solvent generated anion exchangers. III. Application to the analysis of cisplatin in urine using automated column switching, J. Chromatogr., 1982, 229, 373–386 CrossRef CAS.
  33. B. Anilanmert, G. Yalçin, F. Ariöz and E. Dölen, The spectrophotometric determination of cisplatin in urine, using o-phenylenediamine as derivartizing agent, Anal. Lett., 2001, 34, 113–123 CrossRef CAS.
  34. H. Yang, H. Cui, L. Wang, L. Yan, Y. Qian, X. E. Zheng, W. Wei and J. Zhao, A label-free G-quadruplex DNA-based fluorescence method for highly sensitive, direct detection of cisplatin, Sens. Actuators, B, 2014, 202, 714–720 CrossRef CAS.
  35. J. Petrlova, D. Potesil, J. Zehnalek, B. Sures, V. Adam, L. Trnkova and R. Kizek, Cisplatin electrochemical biosensor, Electrochim. Acta, 2006, 51, 5169–5173 CrossRef CAS.
  36. H. P. Ju, Y. Z. Wang, J. You, X. M. Hou, X. G. Xi, S. X. Dou, W. Li and P. Y. Wang, Folding kinetics of single human telomeric G-quadruplex affected by cisplatin, ACS Omega, 2016, 1, 244–250 CrossRef CAS PubMed.
  37. I. O. Garnier and S. Bombard, GG sequence of DNA and the human telomeric sequence react with cis-diammine-diaquaplatinum at comparable rates, J. Inorg. Biochem., 2007, 101, 514–524 CrossRef PubMed.
  38. P. Heringova, J. Kasparkova and V. Brabec, DNA adducts of antitumor cisplatin preclude telomeric sequences from forming G quadruplexes, J. Biol. Inorg Chem., 2009, 14, 959–968 CrossRef CAS PubMed.
  39. V. Viglasky, Platination of telomeric sequences and nuclease hypersensitive elements of human c-myc and PDGF-A promoters and their ability to form G-quadruplexes, FEBS J., 2009, 276, 401–409 CrossRef CAS PubMed.
  40. A. R. de la Faverie, A. Guédin, A. Bedrat, L. A. Yatsunyk and J. L. Mergny, Thioflavin T as a fluorescence light-up probe for G4 formation, Nucleic Acids Res., 2014, 42, e65 CrossRef.
  41. S. Cogoi and L. E. Xodo, G-quadruplex formation within the promoter of the KRAS proto-oncogene and its effect on transcription, Nucleic Acids Res., 2006, 34, 2536–2549 CrossRef CAS.
  42. K. W. Lim, P. Alberti, A. Guedin, L. Lacroix, J. F. Riou, N. J. Royle, J. L. Mergny and A. T. Phan, Sequence variant (CTAGGG)n in the human telomere favors a G-quadruplex structure containing a G·C·G·C tetrad, Nucleic Acids Res., 2009, 37, 6239–6248 CrossRef CAS PubMed.
  43. Y. Wang and D. J. Patel, Solution structure of the human telomeric repeat d[AG3(T2AG3)3] G-tetraplex, Structure, 1993, 1, 263–282 CrossRef CAS.
  44. N. V. Hud, F. W. Smith, F. A. L. Anet and J. Feigon, The selectivity for K+ versus Na+ in DNA quadruplexes Is dominated by relative free energies of hydration: A thermodynamic analysis by 1H NMR, Biochemistry, 1996, 35, 15383–15390 CrossRef CAS PubMed.
  45. L. M. Demers, C. A. Mirkin, R. C. Mucic, R. A. Reynolds III, R. L. Letsinger, R. Elghanian and G. Viswanadham, A fluorescence-based method for determining the surface coverage and hybridization efficiency of thiol-capped oligonucleotides bound to gold thin films and nanoparticles, Anal. Chem., 2000, 72, 5535–5541 CrossRef CAS PubMed.
  46. D. Orthaber, A. Bergmann and O. Glatter, SAXS experiments on absolute scale with Kratky systems using water as a secondary standard, J. Appl. Crystallogr., 2000, 33, 218–225 CrossRef CAS.
  47. H. Matsuoka, H. Tanaka, T. Hashimoto and N. Ise, Elastic scattering from cubic lattice systems with paracrystalline distortion, Phys. Rev. B: Condens. Matter Mater. Phys., 1987, 36, 1754–1765 CrossRef PubMed.
  48. H. Matsuoka, H. Tanaka, N. Iizuka, T. Hashimoto and N. Ise, Elastic scattering from cubic lattice systems with paracrystalline distortion. II, Phys. Rev. B: Condens. Matter Mater. Phys., 1990, 41, 3854–3856 CrossRef PubMed.
  49. W. Zhao, W. Chiuman, J. C. F. Lam, S. A. McManus, W. Chen, Y. Cui, R. Pelton, M. A. Brook and Y. Li, DNA Aptamer folding on gold nanoparticles: From colloid chemistry to biosensors, J. Am. Chem. Soc., 2008, 130, 3610–3618 CrossRef CAS PubMed.
  50. B. Tinland, A. Pluen, J. Sturm and G. Weill, Persistence length of single-stranded DNA, Macromolecules, 1997, 30, 5763–5765 CrossRef CAS.
  51. T. C. Johnstone, G. Y. Park and S. J. Lippard, Understanding and Improving Platinum Anticancer Drugs – Phenanthriplatin, Anticancer Res., 2014, 34, 471–476 CAS.
  52. R. J. Knox, F. Friedlos, D. A. Lydall and J. J. Roberts, Mechanism of cytotoxity of anticancer platinum drugs: Evidence that cis-diamminedichloroplatinum(II) and cis-diamine-(1,1-cyclobutanedicarboxylato)platinum(II) differ only in the kinetics of their interaction with DNA, Cancer Res., 1986, 46, 1972–1979 CAS.
  53. F. A. Blommaert, H. C. M. van Dijk-Knijnenburg, F. J. Dijt, L. den Engelse, R. A. Baan, F. Berends and A. M. J. Fichtinger-Schepman, Formation of DNA adducts by the anticancer drug carboplatin: Different nucleotide sequence preferences in vitro and in cells, Biochemistry, 1995, 34, 8474–8480 CrossRef CAS PubMed.
  54. C. P. Saris, P. J. M. van de Vaart, R. C. Rietbroek and F. A. Blommaert, In vitro formation of DNA adducts by cisplatin, lobaplatin and oxaliplatin in calf thymus DNA in solution and in cultured human cells, Carcinogenesis, 1996, 17, 2763–2769 CrossRef CAS PubMed.
  55. S. S. Hah, K. M. Stivers, R. W. de Vere White and P. T. Henderson, Kinetics of carboplatin-DNA binding in genomic DNA and bladder cancer cells as determined by accelerator mass spectrometry, Chem. Res. Toxicol., 2006, 19, 622–626 Search PubMed.
  56. Y. S. Kim, S. Shin, M. Cheong and S. S. Hah, Mechanistic Insights into in vitro DNA adduction of oxaliplatin, Bull. Korean Chem. Soc., 2010, 31, 2043–2046 CrossRef CAS.
  57. D. P. Bancroft, C. A. Lepre and S. J. Lippard, Platinum-195 NMR kinetic and mechanistic studies of cis- and trans-diamminedichloroplatinum(II) binding to DNA, J. Am. Chem. Soc., 1990, 112, 6860–6871 CrossRef CAS.
  58. V. Thomsen, D. Schatzlein and D. Mercuro, Limits of detection in spectroscopy, Spectroscopy, 2003, 18, 112–114 CAS.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ay02150g

This journal is © The Royal Society of Chemistry 2020