Furthering the chemosensing of silver nanoclusters for ion detection

Abstract

Ligand-protected metal nanoclusters (NCs) having hydrodynamic diameters and fluorescent properties have reasonable potential for ion detection and bio-imaging applications by taking advantage of their smaller sizes and degradability as compared with larger nanoparticles. By controlling the pH value, here we have synthesized a fluorescent silver nanocluster, Ag–glutathione, which is found to be a versatile chemosensor operative for the detection of both manganese (Mn²⁺) and iodine (I⁻) ions. Mn²⁺ and I⁻ are further distinguished by the ratiometric absorption spectrometry method. Care was taken for sufficient spectral analyses, based upon which we have fully demonstrated the chemosensing mechanisms of Ag@glutathione (Ag@SG) NCs applied in ion detection. Having expounded this issue, we further investigated the potential application of Ag@SG NCs in bio-imaging. It is interesting to find that the as-prepared Ag@SG NCs are nontoxic and available for fluorescence imaging of MCF-7 cells, with on/off alternation of fluorescence emission in the presence of Mn²⁺ or I⁻ ions. The low cytotoxicity, good penetrability and on/off fluorescence property of Ag@SG NCs in MCF-7 cells suggest promising biological applications.

---

1. Introduction

Detection of cationic and anionic ions is important in industrial wastes and environmental processes, where specific ion determination helps monitor excessive or deficient levels in soils and water.^1,2 Overmuch ions existing in natural water and plants due to contamination of ground water, surface water and soils can enter the human body through foods and drinking water resulting in poisoning, which has been a serious concern in many areas of the globe. It is also notable that necessary trace elements (e.g. Cu²⁺, Mn²⁺, Fe³⁺, Zn²⁺, I⁻, and other ions) are essentially involved in diverse fundamental biological processes, and a small amount of these necessary trace elements plays an important role in biological activities; however, the presence of large amounts in the human body may lead to severe health issues. Thus it becomes important to develop effective methods for ion detection, not only in soils and drinking water, but also in the body and living cells.

For this purpose, many strategies have been used in the past, including chromatography, electrochemical analysis, spectrometry with artificial chromophore and fluorophore, etc.³ Among these available detection methods, fluorescence chemosensors such as conjugated fluorescent polymers have stimulated increasing research interest owing to their high sensitivity and simplicity with rapid response.⁴ Recently, with the development of nanotechnology, nanoscaled fluorescent chemosensors have become the most applied analytical tools to detect and trace ions in aqueous solution.^5–9 It has been recognized that metal nanoclusters (NCs; small nanoparticles having sizes between 1 and 10 nanometers) encapsulated in different scaffolds such as thiols and even DNA, etc., are effective for ion detection as the ligand protection provides enhanced stability frequently under a superatom cluster concept.^10–13 Increasing advances in this field enable a promising future for the design of NCs-based chemosensors having notable optical, electrical and chemical properties for bioimaging and ion detection.^14–26

However, a common view in this field is that chemosensors are usually valid for detection of either a cationic ion or an anionic ion, thus a pending question remains as to what happens if both cationic and anionic ions coexist in a complex unknown solution. Also, only rarely have fluorescent chemosensors been applied in living cells to our best knowledge. It is highly desirable to develop eco-friendly and nontoxic fluorescent chemosensors with high sensitivity for ion detection. Facile synthesis procedure and high selective capability are expected in order to enable detection of multiple ions simultaneously. In addition, the predominant sensing mechanism varies depending on the type of the sensor, and is accommodated strictly by the altering of the fluorescence property of the NCs through specific interactions with either the metallic core or the protecting ligand.^27–31 These issues are still open questions and need to be further explored.

Thanks to the biocompatible glutathione (HSG) which does not exhibit cytotoxicity itself, the Ag–glutathione complex also does not show obvious cytotoxicity with concentrations up to mmol L⁻¹.^2,32–36 Motivated by this, here we have synthesized eco-friendly silver NCs, and performed an in-depth study of ion detection utilizing the glutathione-protected silver (Ag@SG) NCs as a fluorescence chemosensor. It is interesting to find that the as-prepared Ag NCs chemosensor is effective for the detection and differentiation of Mn²⁺ and I⁻ ions. Together with an insight into the sensing mechanisms, we demonstrate feasible applications in living cells with on/off controllability in the presence of predetermined ions.

2. Experimental

2.1 Materials and synthesis

All chemicals were commercially available and used as received without further purification.

The Ag@SG NCs were synthesized according to the following procedure (Scheme 1). Firstly, an aqueous solution of AgNO₃ (84.9 mg, 0.5 mM) and glutathione (GSH; 615.0 mg, 2.0 mM) were mixed together in 25 mL water with vigorous stirring, and the solution was kept at ∼0 °C in an ice bath. Next, an aqueous solution of NaOH (0.5 M) was added into the above solution until the white precipitate completely disappeared (pH = 6). To this clear solution, freshly prepared ice-cold aqueous NaBH₄ (189.0 mg, 5.0 mL) was slowly added dropwise under vigorous stirring. Then, the reaction was allowed to proceed under constant stirring for 7 h, and the clusters were precipitated with methanol. After filtration, a brown precipitate of Ag@SG NCs was obtained, which was then washed with excess methanol.

Scheme 1 A schematic depicting the synthesis procedure of glutathione-protected silver nanoclusters in this study.

2.2 Measurements and characterization

Fluorescence spectra were collected using a Horiba Scientific Fluoromax-4 spectrophotometer equipped with a quartz cuvette of 1.0 cm path length with a xenon lamp as the excitation source. UV-visible absorption spectra were recorded with a UV-3600 spectrophotometer. Fourier transform infrared (FT-IR) spectra were obtained with an Avatar 360 FI-IR spectrometer in the range of 4000–400 cm⁻¹. The ¹H NMR spectra were measured with a Bruker Avance III 400 MHz spectrometer in D₂O solution with TMS as an internal standard. Transmission electron microscopy (TEM) images were captured with a JEOL JEM-2100F field-emission transmission electron microscope with an accelerating voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) was performed with a Thermo Scientific ESCALab 250Xi. Mass spectra were obtained with a Bruker Solarix 9.4 T high resolution mass spectrometer using an electron spray ion source.

Ion detections were conducted in a quartz cuvette (1 cm beam path). Briefly, small amounts of concentrated salt solutions were added to the Ag@SG NCs water solution, which did not give rise to obvious pH value changes. In all titration experiments, the metal cations (Fe³⁺, Co²⁺, Cd²⁺, Ca²⁺, Cr³⁺, Cu²⁺, Pb²⁺, Ni²⁺, Zn²⁺ and Mn²⁺) and anions (F⁻, Cl⁻, Br⁻, I⁻, NO₂⁻, CO₃²⁻, AcO⁻, H₂PO₄⁻, HCO₃⁻, SO₄²⁻, HPO₄²⁻, HSO₃⁻, and OH⁻) were referred to normal ionic compounds as largely used in this field.^37–41

The in vitro cytotoxicity was measured using the methyl thiazolyl tetrazolium (MTT) assay in MCF-7 cell lines. Cells in log-growth phase were seeded into a 96-well cell-culture plate at 1 × 10⁴ per well in 100 μL DMEM (Dulbecco's modified Eagle's medium). Ag@SG NCs were added to the wells (100 μL per well) of the treatment group, and the final concentration ranged from 0 to 0.5 μg mL⁻¹. The cells were incubated for 24 h at 37 °C under 5% CO₂. A combined MTT/PBS solution (100 μL, 0.5 mg mL⁻¹) was added to each well of the 96-well assay plate, and incubated for an additional 2 h. An enzyme-linked immunosorbent assay reader (μQuant, Bio-Tek, USA) was used to measure the OD570 (absorbance value) of each well referenced at 690 nm. The formula used to calculate the viability of cell growth is: viability (%) = (mean absorbance value of treatment group/mean absorbance value of control) × 100.

Prior to the bioimaging experiment, MCF-7 cells were cultured in DMEM. Cells were incubated with 0.25 μg mL⁻¹ Ag@SG NCs at 37 °C for 2 h. After washing with PBS three times to remove remaining Ag@SG NCs, the cells were then incubated with an appropriate concentration of Mn²⁺ and I⁻ solution for 30 min at room temperature. The incubated cells were washed with PBS and mounted onto a glass slide. Fluorescence images of the mounted cells were obtained using a confocal laser scanning microscope at 488 nm excitation.

3. Results and discussion

3.1 Detecting both Mn²⁺ and I⁻ ions

Fig. 1a presents a UV-visible absorption spectrum of the as-prepared Ag@SG NCs, where a broad absorption band from 350 nm to 500 nm is noted largely differing from the plasmon resonance absorption spectrum of large silver nanoparticles (i.e., >10 nm) which shows a narrow peak at about 400 nm wavelength. We checked the sizes of the obtained Ag@SG NCs by TEM imaging and found that they are spherical and well dispersed with an average size of ∼3 nm, as shown by the two inset images (more details in Fig. 1a and S1a, ESI†).

Fig. 1 (a) UV-visible absorption spectrum of Ag@SG NCs. Insets: the left is a magnified TEM image of a silver nanocluster; the right is the diameter size distribution of a typical Ag NCs sample. (b) Fluorescence excitation and emission spectra of Ag@SG NCs.

The Ag@SG NCs exhibit an intense fluorescence emission peak centered at ∼630 nm (Fig. 1b, excited at 400 nm). As a comparison, also provided in Fig. 1b is the excitation spectrum which is consistent with the aforementioned UV-visible absorption behavior. In addition, the monodispersity of the Ag@SG NCs product was checked by polyacrylamide gel electrophoresis analysis (Fig. S1b, ESI†), which indicates a decent grade of purity of the as-prepared clusters. XPS (Fig. S3a and 3b, ESI†) survey spectrum of the Ag@SG NCs confirms the presence of Ag, S, C, N, O elements, and the Ag 3d_5/2 peak at 368.2 eV is close to the Ag(0) value, which matches a previously published report;⁴² the presence of S (2p_3/2) at 162.3 eV suggests that the ligand is chemically bonded to a silver cluster in the form of thiolates,⁴³ which is in agreement with the IR spectrum (Fig. S4, ESI†) where the 2523 cm⁻¹ mode corresponding to the S–H stretching of a HSG molecule disappears, evidencing that the formation of Ag@SG NCs is stabilized by S–Ag bondings.⁴² In addition, the Auger electron peak of silver (Fig. S3c, ESI†) confirms the presence of Ag(I); in other words, Ag(0) and Ag(I) coexist in Ag@SG NCs.

Following the successful synthesis of Ag@SG NCs, we performed a study of ion detection by observing the fluorescence changes of NCs when added to aqueous metal ions. As shown in Fig. 2a, the obtained Ag@SG NCs display strong fluorescence emission at ∼630 nm; however, the fluorescence emission intensities change in the presence of different cationic metal ions. The cations Fe³⁺, Co²⁺, Cd²⁺, Ca²⁺, Cr³⁺, Cu²⁺, Pb²⁺, Ni²⁺ and Zn²⁺ were found to enhance the nascent fluorescence intensity of NCs to different extents; however, the addition of Mn²⁺ to Ag@SG NCs exclusively results in a remarkable fluorescence quenching.

Fig. 2 (a) Fluorescence responses of Ag@SG NCs (2.5 μg mL⁻¹) in H₂O to metal cations (0.1 mM). (b) Fluorescence intensity of Ag@SG NCs (2.5 μg mL⁻¹) with Mn²⁺ (1.0 mM) in the presence of various metal ions (0.1 mM) at 630 nm. Colored bars: Ag@SG NCs with metal ions stated. Gray bars: solutions of Ag@SG NCs with Mn²⁺ and 0.1 equivalent of the other ions stated. (c) Fluorescence titration profiles (λ_ex = 400 nm) of Ag@SG NCs (10 μg mL⁻¹) in the presence of increasing amounts of Mn²⁺ in H₂O. (d) The relative intensity (I/I₀) of Ag@SG NCs with Mn²⁺ increasing monitored at 630 nm with λ_ex = 400 nm. Inset: the linear detection range for 0–1.5 mM of Mn²⁺.

To further investigate the selectivity of Ag@SG NCs for Mn²⁺, we conducted a competitive experiment endeavoring to evaluate the practical effect in detecting Mn²⁺ ions in the presence of other metal ions. As shown in Fig. 2b, the coexistence of 0.1 equivalent of other metal ions in 1.0 mM Mn²⁺ ions also leads to an obvious fluorescence quenching effect. Therefore it is possible to use Ag NCs to detect Mn²⁺ ions up to millimoles even in the presence of other interference metal cations (up to 0.1 mM in this study), revealing that Ag@SG NCs have high selectivity toward Mn²⁺ among the other metal ions in water.

We have also conducted fluorescence titration experiments for the silver NCs when adding Mn²⁺ ions of varying concentrations (Fig. 2c). The fluorescence intensity at 630 nm decreases remarkably with increasing concentration of Mn²⁺ ions. Upon further increase of the Mn²⁺ concentration, fluorescence emission tends to be quenched completely. Further analysis depicted in Fig. 2d reveals that there is a good linear relationship (coefficient of determination R² = 0.97) in the low concentration range of 0 to 1.5 mM, although a linear convergence beyond 2.0 mM. From the linear fitting curve (red) it is deduced that the limit of detection for Mn²⁺ ions is ∼9.2 μM indicating decent sensitivity of this method which is comparable to that of other reported chemosensors for detecting Mn²⁺ ions.^44–47

Interesting results were noted when we further examined the selectivity of the Ag@SG NCs toward anions. Fig. 3a displays the fluorescence responses in water with F⁻, Cl⁻, Br⁻, I⁻, NO₂⁻, CO₃²⁻, AcO⁻, H₂PO₄⁻, HCO₃⁻, SO₄²⁻, HPO₄²⁻, HSO₃⁻, and OH⁻ among which only I⁻ quenches the fluorescence of Ag@SG NCs while the other anions even show minor enhancements to their fluorescence intensities. Similar to the aforementioned case for Mn²⁺, we have also performed competition experiments for I⁻ in the presence of the other anions, as shown in Fig. 3b. It is notable that the presence of other anions had no obvious interference effects compared with the experimental results when only adding I⁻ itself. In the fluorescence titration experiment of Ag@SG NCs with I⁻ (Fig. 3c), the fluorescence intensity decreased with increasing I⁻ concentration, with a good linear relationship (R² = 0.96) in the concentration range from 0.03 mM to 0.26 mM. The limit of detection for I⁻ was calculated to be ∼5.9 μM (Fig. 3d), indicating good sensitivity to the detection of I⁻ ions, which is comparable to that of previously published reports.^40,41 It is additionally worth mentioning that, when S²⁻ ions exist in the solution, Ag@SG NCs do not show good selectivity for I⁻ ions, which is probably caused by the strong affinity of silver toward S²⁻ ions.²

Fig. 3 (a) Fluorescence responses of Ag@SG NCs (2.5 μg mL⁻¹) in H₂O to anions (0.1 mM). (b) Fluorescence intensity of Ag@SG NCs (2.5 μg mL⁻¹) with I⁻ (0.1 mM) in the presence of various anions (0.1 mM) at 630 nm. Colored bars: Ag@SG NCs with anions stated. Grey bars: solutions of Ag@SG NCs with I⁻ and 1.0 equivalent of the other anions stated. (c) Fluorescence titration profiles (λ_ex = 400 nm) of Ag@SG NCs (2.5 μg mL⁻¹) in the presence of increasing amounts of I⁻. (d) The relative intensity (I/I₀) of Ag@SG NCs (2.5 μg mL⁻¹) with I⁻ concentration increasing at 630 nm (λ_ex = 400 nm). Inset: the linear detection range for 0.03–0.26 mM of I⁻.

3.2 The differentiation of Mn²⁺ and I⁻

To further validate the high selectivity of Ag@SG NCs as a chemosensor for the detection of Mn²⁺ and I⁻ in practice, the fluorescence responses of Ag@SG NCs in the presence of Mn²⁺ upon the addition of a number of anions (F⁻, Cl⁻, Br⁻, I⁻, NO₂⁻, CO₃²⁻, AcO⁻, H₂PO₄⁻, HCO₃⁻, SO₄²⁻, HPO₄²⁻, HSO₃⁻, and OH⁻) were examined. Also examined were the responses of “Ag@SG NCs + I⁻” when adding metal ions (Fe³⁺, Co²⁺, Cd²⁺, Ca²⁺, Cr³⁺, Cu²⁺, Pb²⁺, Ni²⁺ and Zn²⁺). As shown in Fig. 4a and b, the coexistence of 1.0 equivalent of other metal ions and anions did not interfere with the Mn²⁺ and I⁻ quenching effects. Besides, when Mn²⁺ and I⁻ ions were added into Ag@SG NCs simultaneously, the fluorescence was also quenched. So, is it possible to distinguish cationic Mn²⁺ from anionic I⁻?

Fig. 4 Fluorescence responses of (a) Ag NCs + I⁻ upon the addition of metal ions and (b) Ag NCs + Mn²⁺ upon the addition of anions in H₂O.

With this in mind, we have studied the ratiometric absorption behaviors for Mn²⁺ and I⁻. Fig. 5 presents results of UV-visible titration experiments of Ag@SG NCs in the presence of increasing concentrations of Mn²⁺ and I⁻ ions, respectively. It is notable that the position of the broad absorption band from 350 nm to 500 nm barely changes, but the intensity decreased gradually with increasing Mn²⁺ ion concentration (Fig. 5a). This result excludes the possibility of direct binding of Mn²⁺ to the Ag atoms due to the absorption band of Ag nanoclusters being very sensitive to the presence of adsorbed substances.⁴⁸ However, the gradual addition of I⁻ ions (Fig. 5b and c) differs distinctly from that of Mn²⁺. Besides the gradual intensity decrease of the wide absorption band from 400 nm to 500 nm, two iso-absorptive points are noted successively in different concentration ranges. The first isobestic point emerges at ∼398 nm with I⁻ concentration increasing in the lower concentration range, while more gradual addition of I⁻ ions leads to the second iso-absorptive point at ∼378 nm. In addition, the spectrum exhibits a new absorption peak at ∼420 nm in the higher I⁻ concentration range. Exclusion experiments were conducted by observing UV-visible absorption spectra of aqueous HSG in the presence of I⁻ ions (Fig. S5, ESI†); however, the addition of I⁻ to HSG solution was found to bring negligible changes. Therefore, the likely formed new complexes in the solution can be associated with the reaction of Ag@SG NCs core and I⁻ ions. These results indicated that Ag@SG NCs could be used to distinguish Mn²⁺ from I⁻ by the ratiometric absorption spectrometry method.

Fig. 5 UV-visible titration spectra of Ag@SG NCs (10 μg mL⁻¹) in the presence of increasing amounts of Mn²⁺ (a) and Ag@SG NCs (2.5 μg mL⁻¹) in the presence of increasing amounts of I⁻ (b and c).

3.3 The sensing mechanisms

The above experimental results demonstrate an interesting fact that the as-prepared Ag@SG NCs are highly sensitive for detecting and differentiating Mn²⁺ and I⁻ ions. Such a chemosensor that can detect both anions and cations is rarely reported in previous studies. In order to provide insights into the sensing mechanism, ¹H NMR experiments and TEM characterizations of Ag@SG NCs before and after the addition of Mn²⁺ and I⁻ have been conducted. As shown in Fig. 6A, upon the gradual addition of Mn²⁺ to Ag@SG NCs, all proton signals of Ag@SG NCs disappear until finally only a solvent D₂O peak is present at 4.79 ppm. With Mn²⁺ ion concentration further increasing, there was a brown material precipitated out of the solution. While in the case of I⁻ (Fig. 6B), all proton signals are almost the same indicating that the chemical environments of all the protons of –SG do not encounter obvious changes. From TEM observation, the nascent Ag@SG NCs display monodispersity (Fig. S2a and b, ESI†); however, there is an aggregation pattern in the presence of Mn²⁺ (Fig. S2c and d, ESI†). In contrast, when adding I⁻ ions (Fig. S2e and f, ESI†) the average size of particles becomes apparently larger than that of the original Ag@SG NCs. A pending question is then what determines the differences for Mn²⁺ and I⁻ ions as regards their fluorescence quenching of Ag NCs?

Fig. 6 ¹H NMR spectra in D₂O. (A) Ag@SG NCs at ∼10 μg mL⁻¹ (a), and after adding 0.5 mM of Mn²⁺ (b), 1.0 mM of Mn²⁺ (c), and 2.0 mM of Mn²⁺ (d). (B) Ag@SG NCs at ∼2.5 μg mL⁻¹ (a), and after adding 0.03 mM of I⁻ (b), 0.1 mM of I⁻ (c), and 0.3 mM of I⁻ (d). The D₂O peak at 4.79 ppm is labelled with *.

Fig. 7 provides a comparison of XPS patterns of Ag 3d core level for the Ag@SG NCs before and after reacting with Mn²⁺ and I⁻ ions. For the case when adding I⁻ ions (Fig. 7a), it is clearly seen that the peaks assigned to both Ag 3d_5/2 (368.2 eV) and Ag 3d_3/2 (374.1 eV) shift to the lower-energy side by up to 1.0 eV in the presence of I⁻ ions. This is in sharp contrast to the situation when adding Mn²⁺ ions where a minor shift is towards higher energy (Fig. 7c). The low-energy shifts in the presence of iodide ions demonstrate the complexation of Ag with electron-rich groups or elements such as iodide. This is an important piece of evidence as fluorescence quenching often could be caused by energy transfer or static quenching by forming nonfluorescent complexes. Actually many studies of the interactions of halide ions with silver or other nanoparticles have been reported,^49–51 and it is known that iodides have a stronger affinity for noble metal nanoclusters than the other counterparts.^52–54 Further evidences are determined by noting the presence of Ag_nI_x species via mass spectrometry (Fig. S6, ESI†). Combining with the ratiometric absorption behaviors for I⁻ (Fig. 5b and c), the isobestic points indicate that there are multiple reaction pathways in a solution of Ag@SG NCs with gradual addition of I⁻. It is ascertained that I⁻ ions in low concentration adsorb rapidly on the Ag@SG NCs surface owing to the high strength of the Ag–I bond or the low solubility of AgI, which leads to catalyzed etching likely via a reaction pathway analogous to that of a strong acid displacing a weak acid, as described in eqn (1):

Ag_n(SG)_x⁻ + yI⁻ + yH⁺ → Ag_n−y(AgI)_y(SG)_x−y⁻ + yHSG, or

Ag_n(SG)_x⁻ + yI⁻ + yH⁺ +yH₂O → Ag_n−y(SG)_x−y⁻ + yAgI(H₂O)⁰ + yHSG (1)

Fig. 7 XPS spectra of Ag 3d core level for the Ag@SG NCs in the presence of I⁻ (a), Ag@SG NCs itself (b) and in the presence of Mn²⁺ (c).

In this equation, the formation of Ag_n−y(AgI)_y(SG)_x−y⁻ or AgI(H₂O)⁰ species adsorbed in the Stern plane of Ag@SG NCs may play a role in the stability by truncating the van der Waals attraction at distances of closest approach relative to the electrostatic repulsion interactions.⁵⁵ Next, more I⁻ ions chemisorbed on the surface of NCs may rearrange the surface charges,¹ which will increase the weak intermolecular interactions among the newly formed iodide-interspersed silver particles and then lead to aggregation. At the same time, a small number of protection ligands means reduced sizes of the NCs, allowing for increased specific surface area and free surface energy, and hence incidental aggregation (eqn (2)):

Ag_n-y(AgI)_y(SG)_x−y⁻ + Ag_n(SG)_x⁻ + zI⁻ + γH⁺ → Ag_2nI_y+z(SG)_δ(HSG)_γ⁻, or

Ag_n(SG)_x⁻ + yAgI(H₂O)⁰ + yI⁻ → Ag_n+yI_2y(SG)_x−y(HSG)_y⁻ + yOH⁻, or

Ag_n−y(SG)_y⁻ + Ag_n−y(SG)_y⁻ + zI⁻ + γH⁺ → Ag_2n−2yI_z(SG)_δ(HSG)_γ⁻ (2)

where the values of x, y, z and n are integers (i.e., 0 to n). In brief, within the chemosensor mechanism, I⁻ ions interact with Ag@SG initially via an etching-like pathway prevailing at low I⁻ concentrations, while aggregation/reorganization dominates at high concentrations of I⁻ ions. Considering that the chemosensing mechanism for I⁻ ions could be pH-dependent, these equations indicate the tendency to form a class of Ag_xI_y(SG)_z(HSG)_r⁻ complexes, which are actually in accordance with previous proposals.^40,41 It is worth mentioning that the characteristic surface plasmon resonance band of Ag nanoparticles located around 420 nm⁻¹ was found in the higher I⁻ concentration range in Fig. 5c, which was also in accordance with the average size as typically displayed in a TEM image (Fig. S2f, ESI†).

Different from I⁻ ions, the result that Mn²⁺ induces aggregation of Ag@SG NCs in solution could simply follow a well-known AIFQ (aggregation-induced fluorescence quenching) mechanism, which profits from the interaction of hard acid Mn²⁺ ions with the multiple hard base –NH and –O moieties present in several –SG ions. What is more, Mn²⁺ is a highly paramagnetic ion, which can quench the fluorescence of metal NCs via energy transfer and weak bonding (eqn (3)):

Ag_n(SG)_x⁻ + Ag_m(SG)_y⁻ + Mn²⁺ → [Ag_n(GS)_x⋯Mn⋯(SG)_yAg_m]⁰ (3)

3.4 Cell toxicity and bioimaging

Having determined the excellent selectivity and high sensitivity of Ag@SG NCs towards Mn²⁺ and I⁻ ions in water, we endeavored to explore their potential application in living cell bioimaging. It is well known that a small quantity of Mn²⁺ is necessary for living organisms, such as having influences on bone growth and also glucose and fat metabolism, as well as an antioxidant and for prevention of cancer and anemia, etc., while excessive intakes lead to toxic symptoms and neurodegenerative disorders.⁵⁶ Iodine also plays an important role in biological activities due to its being required for the normal function of thyroid and neurologic activity.^5–8 Deficiency of iodine causes health problems such as cretinism, congenital abnormalities and goiter;^5–8 however, excessive amounts of iodine in the human body may give rise to thyrotoxicosis.^8,9 Therefore, we firstly tested the toxicity of the as-prepared silver NCs to MCF-7 cells (Fig. 8a). The cellular toxicity of Ag@SG NCs to MCF-7 cells in H₂O was determined by a MTT assay with the concentration of Ag@SG NCs ranging from 0 to 0.5 μg mL⁻¹ (Fig. 8b). Upon incubation for 24 h of 0.25 μg mL⁻¹ Ag@SG NCs (a typical concentration used for confocal imaging studies), the cellular viabilities are estimated as nearly 100%, revealing that the as-prepared Ag@SG NCs are almost not toxic for MCF-7 cells. The reason for the hypotoxicity is largely ascribed to the eco-friendly stabilization of ligand HSG and appropriate pH values that we have maintained in synthesizing the Ag@SG NCs.

Fig. 8 (a) The general operation of the process. MTT: methyl thiazolyl tetrazolium; DMEM: Dulbecco's modified Eagle's medium. (b) Cell viability values estimated by MTT proliferation test versus concentrations of Ag@SG NCs after 24 h incubation at 37 °C.

The final important result relates to the attempt to detect Mn²⁺ and I⁻ in living cells. For this study, MCF-7 cells were incubated with 0.25 μg mL⁻¹ Ag@SG NCs for 2 h firstly, and then treated separately with 1.5 mM Mn²⁺ and 0.3 mM I⁻ for 30 min. As shown in Fig. 9a, the cells only incubated with Ag@SG NCs display strong fluorescence, while the strong red fluorescence is quenched in cells when exposed to aqueous solution of Mn²⁺ (Fig. 9b) or I⁻ (Fig. 9c) ions. It is noted that the chemosensor of Ag@SG NCs is cell-permeable allowing for bioimaging in living cells and the decent fluorescence on/off property enables detection of intracellular Mn²⁺ and I⁻ ions. Besides, the shape of the MCF-7 cells further indicated no cytotoxicity effect of the Ag@SG NCs and no aggregation on the cell membrane. Our biolabeling and chemosensing effect in living cells is comparable to that of recent reports in this field, such as HSG-protected silver nanoclusters used in A549 cell imaging by Le Guevel et al.,⁵⁷ and Ag NCs were used to demonstrate cytocompatibility and show better inhibition effects through MTT assay by Li et al.⁵⁸

Fig. 9 Confocal fluorescence images of (a) only Ag@SG NCs, (b) addition of Mn²⁺ and (c) addition of I⁻ in MCF-7 cells in the fluorescence (left), bright (middle), and overlay (right) field.

4. Conclusions

In summary, we develop here an eco-friendly chemosensor, glutathione-protected silver nanoclusters, synthesized via a simple and green route. Utilizing this Ag@SG NCs chemosensor, we have realized the detection and differentiation of Mn²⁺ and I⁻ ions with high selectivity and sensitivity. The sensing mechanisms are deduced according to fluorescence and UV-visible titration, ¹H NMR, TEM images, XPS analyses etc. For the detection of Mn²⁺ ions, the silver NCs are prone to aggregate via [Ag_n(GS)_x⋯Mn⋯(SG)_yAg_m]⁰, leading to fluorescence quenching. In comparison, the detection of I⁻ ions was based on the large affinity of silver halides resulting in iodide-induced etching and aggregation/reorganization. We have successfully applied the Ag@SG NCs chemosensor to detect intracellular Mn²⁺ and I⁻ ions, which would broaden the applications of functionalized Ag NCs in biological fields.

Acknowledgements

This work is supported by the Young Professionals Programme in the Institute of Chemistry, Chinese Academy of Sciences (ICCAS-Y3297B1261). We are grateful for the financial support from CAS of China with grant no. Y31M0112C1.

Notes and references

1. F. Qu, N. B. Li and H. Q. Luo, Anal. Chem., 2012, 84, 10373–10379.
2. M. Wang, Z. Wu, J. Yang, G. Wang, H. Wang and W. Cai, Nanoscale, 2012, 4, 4087–4090.
3. R. Li, P. Xu, J. Fan, J. Di, Y. Tu and J. Yan, Anal. Chim. Acta, 2014, 827, 80–85.
4. A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97, 1515–1566.
5. R. H. Verheesen and C. M. Schweitzer, Med. Hypotheses, 2008, 71, 645–648.
6. M. Haldimann, B. Zimmerli, C. Als and H. Gerber, Clin. Chem., 1998, 44, 817–824.
7. F. Jalali, M. J. Rajabi, G. Bahrami and M. Shamsipur, Anal. Sci., 2005, 21, 1533–1535.
8. R. P. Li, P. P. Xu, J. Fan, J. W. Di, Y. F. Tu and J. L. Yan, Anal. Chim. Acta, 2014, 827, 80–85.
9. R. Hu, L. Zhang and H. B. Li, New J. Chem., 2014, 38, 2237–2240.
10. A. W. Castleman Jr and S. N. Khanna, J. Phys. Chem. C, 2009, 113, 2664–2675.
11. Z. Luo, G. U. Gamboa, J. C. Smith, A. C. Reber, J. U. Reveles, S. N. Khanna and A. W. Castleman Jr, J. Am. Chem. Soc., 2012, 134, 18973–18978.
12. Z. Luo, C. J. Grover, A. C. Reber, S. N. Khanna and A. W. Castleman, J. Am. Chem. Soc., 2013, 135, 4307–4313.
13. Z. Luo and A. W. Castleman Jr, Acc. Chem. Res., 2014, 2931–2940.
14. L. A. Peyser, A. E. Vinson, A. P. Bartko and R. M. Dickson, Science, 2001, 291, 103–106.
15. J. Zheng and R. M. Dickson, J. Am. Chem. Soc., 2002, 124, 13982–13983.
16. S. W. Chen, R. S. Ingram, M. J. Hostetler, J. J. Pietron, R. W. Murray, T. G. Schaaff, J. T. Khoury, M. M. Alvarez and R. L. Whetten, Science, 1998, 280, 2098–2101.
17. Y. Y. Yang and S. W. Chen, Nano Lett., 2003, 3, 75–79.
18. Y. Cui, Y. Wang, R. Liu, Z. Sun, Y. Wei, Y. Zhao and X. Gao, ACS Nano, 2011, 5, 8684–8689.
19. H. Xu and K. S. Suslick, ACS Nano, 2010, 4, 3209–3214.
20. J. P. Wilcoxon and B. L. Abrams, Chem. Soc. Rev., 2006, 35, 1162–1194.
21. X. Liu, Y. Wu, S. Li, Y. Zhao, C. Yuan, M. Jia, Z. Luo, H. Fu and J. Yao, RSC Adv., 2015, 5, 30610–30616.
22. L. Shang, S. J. Dong and G. U. Nienhaus, Nano Today, 2011, 6, 401–418.
23. Y. Wang, Y. Cui, R. Liu, Y. Wei, X. Jiang, H. Zhu, L. Gao, Y. Zhao, Z. Chai and X. Gao, Chem. Commun., 2013, 49, 10724–10726.
24. Y. Wang, Y. Cui, Y. Zhao, R. Liu, Z. Sun, W. Li and X. Gao, Chem. Commun., 2012, 48, 871–873.
25. R. Liu, J. Zhai, L. Liu, Y. Wang, Y. Wei, X. Jiang, L. Gao, H. Zhu, Y. Zhao, Z. Chai and X. Gao, Chem. Commun., 2014, 50, 3560–3563.
26. W. Li, R. Liu, Y. Wang, Y. Zhao and X. Gao, Small, 2013, 9, 1585–1594.
27. Y. M. Guo, Z. Wang, H. W. Shao and X. Y. Jiang, Analyst, 2012, 137, 301–304.
28. J. P. Xie, Y. G. Zheng and J. Y. Ying, Chem. Commun., 2010, 46, 961–963.
29. D. Y. Cao, J. Fan, J. R. Qiu, Y. F. Tu and J. L. Yan, Biosens. Bioelectron., 2013, 42, 47–50.
30. Y. L. Liu, K. L. Ai, X. L. Cheng, L. H. Huo and L. H. Lu, Adv. Funct. Mater., 2010, 20, 951–956.
31. X. X. Wang, P. Wu, X. D. Hou and Y. Lv, Analyst, 2013, 138, 229–233.
32. E. S. Shibu, M. A. H. Muhammed, T. Tsukuda and T. Pradeep, J. Phys. Chem. C, 2008, 112, 12168–12176.
33. A. Baksi, M. S. Bootharaju, X. Chen, H. Häkkinen and T. Pradeep, J. Phys. Chem. C, 2014, 118, 21722–21729.
34. Z. Wu, M. Wang, J. Yang, X. Zheng, W. Cai, G. Meng, H. Qian, H. Wang and R. Jin, Small, 2012, 8, 2028–2035.
35. Z. Chen, D. Lu, G. Zhang, J. Yang, C. Dong and S. Shuang, Sens. Actuators, B, 2014, 202, 631–637.
36. C. L. Zheng, Z. X. Ji, J. Zhang and S. N. Ding, Analyst, 2014, 139, 3476–3480.
37. W.-H. Ding, C. Wei, X.-J. Zheng, D.-C. Fang, W.-T. Wong and L.-P. Jin, Inorg. Chem., 2013, 52, 7320–7322.
38. W.-H. Ding, D. Wang, X.-J. Zheng, W.-J. Ding, J.-Q. Zheng, W.-H. Mu, W. Cao and L.-P. Jin, Sens. Actuators, B, 2015, 209, 359–367.
39. W.-H. Ding, W. Cao, X.-J. Zheng, W.-J. Ding, J.-P. Qiao and L.-P. Jin, Dalton Trans., 2014, 43, 6429–6435.
40. A. Kumar, R. K. Chhatra and P. S. Pandey, Org. Lett., 2010, 12, 24–27.
41. J. Zhang, X. Xu, C. Yang, F. Yang and X. Yang, Anal. Chem., 2011, 83, 3911–3917.
42. A. Baksi, M. S. Bootharaju, X. Chen, H. Hakkinen and T. Pradeep, J. Phys. Chem. C, 2014, 118, 21722–21729.
43. G. Gonella, O. Cavalleri, S. Terreni, D. Cvetko, L. Floreano, A. Morgante, M. Canepa and R. Rolandi, Surf. Sci., 2004, 566, 638–643.
44. P. S. Hariharan and S. P. Anthony, Spectrochim. Acta, Part A, 2015, 136, 1658–1665.
45. K. B. Kim, G. J. Park, H. Kim, E. J. Song, J. M. Bae and C. Kim, Inorg. Chem. Commun., 2014, 46, 237–240.
46. C. Gou, H. Wu, S. Jiang, C. Yi, J. Luo and X. Liu, Chem. Lett., 2011, 40, 1082–1084.
47. I. Kaya, M. Yildirim and M. Kamaci, Synth. Met., 2011, 161, 2036–2040.
48. T. Linnert, P. Mulvaney and A. Henglein, J. Phys. Chem., 1993, 97, 679–682.
49. J. Zhang, X. Xu, C. Yang, F. Yang and X. Yang, Anal. Chem., 2011, 83, 3911–3917.
50. J. Zhang, Y. Yuan, X. Xu, X. Wang and X. Yang, ACS Appl. Mater. Interfaces, 2011, 3, 4092–4100.
51. M. G. Espinoza, M. L. Hinks, A. M. Mendoza, D. P. Pullman and K. I. Peterson, J. Phys. Chem. C, 2012, 116, 8305–8313.
52. O. M. Magnussen, Chem. Rev., 2002, 102, 679–725.
53. J. E. Millstone, W. Wei, M. R. Jones, H. Yoo and C. A. Mirkin, Nano Lett., 2008, 8, 2526–2529.
54. M. Zhu, G. Chan, H. Qian and R. Jin, Nanoscale, 2011, 3, 1703–1707.
55. Z. Luo, J. C. Smith, T. M. Goff, J. H. Adair and A. W. Castleman Jr, Chem. Phys., 2013, 423, 73–75.
56. B. A. Pappas, D. Zhang, C. M. Davidson, T. Crowder, G. A. S. Park and T. Fortin, Neurotoxicol. Teratol., 1997, 19, 17–25.
57. X. le Guevel, C. Spies, N. Daum, G. Jung and M. Schneider, Nano Res., 2012, 5, 379–387.
58. J. Li, X. Zhong, F. Cheng, J.-R. Zhang, L.-P. Jiang and J.-J. Zhu, Anal. Chem., 2012, 84, 4140–4146.

---

Footnote

† Electronic supplementary information (ESI) available: Chemicals, properties of the silver clusters, more spectral experimental details. See DOI: 10.1039/c5ra11124b

---