Introduction

Glutathione (GSH),the dominant nonprotein thiol in mammalian and eukaryotic cells, is synthesized endogenously from the precursor amino acids L-cysteine, L-glutamic acid, and glycine [1]. It participates in several vital biological functions, including protein and DNA synthesis, amino acid transport, enzyme activity, and maintenance of intracellular redox homeostasis, metabolism and detoxification [2]. Levels of GSH are associated with kinds of human diseases, such as cancer, HIV, AIDS, aging, and Alzheimer’s disease [3,4,5]. Due to the biological and clinical significance, it is of sustained interest in accurate and selective quantification of GSH. Numbers of methods and strategies so far have focused on determination of GSH, e.g., high-performance liquid chromatography (HPLC) [6], electrochemistry [7], chemiluminescence [8], surface-enhanced Raman scattering (SERS) [9], mass spectrometry [10], and fluorescence spectroscopy [11,12,13,14,15]. These methods, however, not only require expensive instruments and sophisticated operations but also involve complicated sample pretreatment, which limits their applications in clinical diagnostics and treatment.

Point-of-care (POC) testing allows rapid, on-site, and affordable detection of biomarker at home, which brings considerable convenience to the patient. The personal glucose meter (PGM) is currently one of the most widely used diagnostic devices in POC testing because of its portable size, time-saving, low cost and reliable quantitative results [16]. Up to now, various PGM-based methods have been fabricated for quantitative detection of a wide range of non-glucose targets based on target-responsive controlled release of glucose from gated-mesoporous silica nanoparticles (MSNs) [17,18,19,20,21,22]. For example, Tang et al. constructed an immunoassay for quantitative detection of aflatoxins based on target-induced release of glucose from the Au nanoparticles-gated MSNs by coupling a competitive-type displacement reaction mode with PGM [23]. Lu et al. established an efficient approach to quantitatively monitor telomerase activity using DNA-capped MSNs by coupling with a PGM based on target-responsive release strategy [24]. Among these methods, MSNs exhibit exceptional biocompatibility, high surface area, large pore volume, and ease of functionalization, which make it possible to load large numbers of glucose by different gates and diverse design strategies.

Ultrathin MnO2 nanosheets, one type of redox-active 2D nanomaterial, are composed of three atomic layers includes two O layers and one Mn layer. Each Mn is coordinated to six O atoms to form an edge-sharing MnO6 octahedron [25]. As a result of the presence of Mn-vacancies, single-layer MnO2 nanosheets were negatively charged [26]. At the same time, MnO2 nanosheets with strong oxidation ability can be reduced into Mn2+ by GSH, dithiothreitol and ascorbic acid [27,28,29]. These together led to the successful development of diverse assay platforms for GSH detection [30, 31]. Therefore, MnO2 nanosheets provide the possibility for fabricating portable, rapid and cost-effective PGM-based detection device for GSH detection.

We report a novel approach for rapid and selective detection of GSH using a PGM based, target-responsive release of glucose from MnO2 nanosheet-gated MSNs. Glucose was successfully loaded in the pores of the aminated MSNs, which were then coated by negatively charged MnO2 nanosheets via electrostatic interaction. The MnO2 nanosheets served as “gates” to prevent the release of the glucose. In the presence of GSH, MnO2 nanosheets can be reduced into Mn2+, leading to open the gates and induce the release of glucose from the pore. The released glucose can be monitored using an external PGM. By evaluation of the PGM signal, the concentration of GSH in the sample can be calculated.

Experimental section

Reagents and materials

Glutathione (GSH) (reduced form) was purchased from Aladdin Chemical (Shanghai, China, www.aladdin-e.com). Tetramethylammonium hydroxide, manganese chloride tetrahydrate (MnCl2‧4H2O),(3-aminopropyl) triethoxysilane (APTES), glucose, glutamic acid (Glu), glycine (Gly), aspartic acid (Asp), tyrosine (Tyr), lysine (Lys), and ascorbic acid (AA) were purchased from Sigma-Aldrich (USA, www.sigma-aldrich.com). Tetraethyl orthosilicate (TEOS), hexadecyl trimethylammonium bromide (CTAB), aqueous ammonia solution (NH3·H2O, 28%), methanol, sodium hydroxide (NaOH), sodium chloride (NaCl), potassium chloride (KCl), calcium chloride (CaCl2) and manganese chloride (MnCl2) were bought from Sinopharm Chemical Reagent Co., Ltd. (Shang hai, China, www.sino-reagent.com). Other reagents were of analytical grade and used without further purification. All solutions were prepared with Milli-Q water (resistivity > 18 MΩ‧cm) from a Millipore system.

Apparatus

Transmission electron microscopy (TEM) images were carried out on a JEM-2100 PLUS microscope (Japan, https://www.jeol.co.jp/en, Japan) at an acceleration voltage of 200 kV. Ultraviolet-visible (UV-vis) absorption spectra were measured using a Varian Cary-300 bio UV/vis spectrophotometer (USA, https://www.varian.com). The zeta potential was determined by Apparatus Mk II microelectrophoresis (England, http://www.rankbrothers.co.uk ). All glucose levels were recorded using a PGM (China, http://www.sinocare.com).

Detection of GSH

The glucose-loading MSNs capped with MnO2 nanosheets (MSNs-G@MnO2) were prepared according to the literatures [24, 32]. 50 μL of GSH at various concentrations were added into 100 μL of the glucose-loading MSNs capped with MnO2 nanosheets (MSNs-G@MnO2). The samples were shaken occasionally during the reaction at room temperature. During this process, GSH reduces MnO2 to Mn2+, which results in the destruction of the MnO2 nanosheets and causes the release of glucose. After incubation for 10 min, the concentration of released glucose was determined by a commercially available PGM.

Results and discussion

Choice of materials

MSNs have high pore volume, large surface area, controlled particles size, and excellent biocompatibility. They are promising candidates for loading a wide range of substances [24]. MnO2 nanosheets,two-dimensional layered materials, with high specific surface area and super light absorption capacity, which has attracted significant attention in developing covering-type and “turn-off-on” fluorescence sensing platforms [32]. In addition, MnO2 nanosheets can be reduced to Mn2+ by some reducing substances. Taking into account its abilities of covering and oxidation, we selected MnO2 nanosheets to construct GSH detection platform.

Detection mechanism

The detection of GSH is based on the target-responsive release of glucose from MnO2-nanosheet-gated mesoporous silica nanoparticles (MSNs-G@MnO2, Scheme 1). Noticeable amounts of glucose are firstly loaded into the pores of MSNs due to the strong preferential interaction between glucose and silica walls [33]. Then, the gates of positively charged aminated MSNs are sealed by negatively charged MnO2 nanosheets due to the electrostatic interaction. Upon the introduction of GSH, the MSN-G@MnO2 is gradually falling apart since GSH reduces MnO2 to form Mn2+ ions. This leads to the decomposition of the MnO2 nanosheets. As a result, the “gate” is destroyed and allows the pore-trapped glucose to diffuse out of MSNs for PGM detection. In this case, the acquired PGM signal is directly proportional to GSH concentration.

Scheme 1
scheme 1

Schematic illustration for quantitative detection of glutathione using an external PGM

Full size image

Characterization of MnO2-nanosheet-gated MSNs

Amino groups were introduced on the outlet of MSNs using established methods [24]. As shown from column “a” in Fig. 1a, the zeta potential of the aminated MSNs was 7.27 mV, indicating that APTES was successfully conjugated onto the MSNs to display positively charged nature. Since the MnO2 nanosheets exhibited a negative zeta potential (−11.9 mV), when aminated MSNs were coated by MnO2 nanosheets, the zeta potential became more negative (−17.6 mV) (Fig. 1a). This indicates that MnO2 nanosheets can be assembled onto aminated MSN by electrostatic interaction. Figure 1b shows typical TEM image of the aminated MSNs with uniform, nearly spherical morphology and narrow size distribution. The aminated MSNs possess well-ordered nanopores, which provided sufficient space for glucose loading. Ultrathin MnO2 nanosheets were prepared via a one-step approach [32]. As shown in Fig. 1c, MnO2 nanosheets displayed a large 2D and ultrathin plane with occasional folds and wrinkles. The formation of MnO2 nanosheets-MSNs was also confirmed by TEM (Fig. 1d), which indicated that MnO2 nanosheets can be assembled onto the surface of aminated MSNs and the MnO2 nansheets-MSNs can serve as an ideal nanocontainer for encapsulating glucose.

Fig. 1
figure 1

(a) Zeta potentials of a the aminated MSNs, b MnO2 nanosheets and c MnO2 nanosheets-coated MSNs. TEM images of (b) the aminated MSNs, (c) MnO2 nanosheets and (d) MnO2 nansheets-MSN nanoparticles

Full size image

Optimization of method

Before optimizing experimental conditions, we studied the stability of the MSNs-G@MnO2 system (Figure S1). After that, the following parameters were optimized: (a) concentration of glucose (Fig. 2a); (b) concentration of MnO2 nanosheets (Fig. 2b); (c) releasing time for glucose (Fig. 2c). Respective data and Figures are given in ESM. We found the following experimental conditions to give best results: (a) Concentration of glucose: 3 M; (b) Concentration of MnO2 nanosheets: 80 μg·mL−1; (c) Release time for glucose: 6 min.

Fig. 2
figure 2

Dependence of the PGM signal on (a) glucose concentration (b) MnO2 nanosheets concentration and (c) release time of glucose (C[GSH] = 100 μM). The error bars were derived from the standard deviation of three measurements. Error bar = SD (n = 3)

Full size image

Analytical performance of PGM-based nanoprobe toward GSH detection

Under the optimal conditions, the PGM-based nanoprobe was employed for quantifying GSH with various concentrations based on target-responsive controlled release of glucose from MSNs-G@MnO2. The detection was carried out using an external PGM after GSH reacted with MSNs-G@MnO2 for 6 min at room temperature. As shown in Fig. 3a, the PGM signal increased with the increasing GSH concentration. A linear dependence between PGM signal (mM) and GSH level (μM) was achieved in the range from 100 nM to 10 μM. The linear regression equation was Y (mM) = 1.0497 + 0.9184 × C[GSH] (μM), R2 = 0.9943. The limit of detection (LOD) was 34 nM, as calculated in terms of the rule of 3× standard deviation over the blank signal. Compared with other methods determination of GSH reported previously [7, 11, 15, 25, 30], our method gave lower or comparable detection limit (Table 1).

Fig. 3
figure 3

(a) Relationship between PGM signal and the GSH concentration. Inset shows a linear standard plot between PGM signal and the GSH concentration. (b) Selectivity of MSNs-G@MnO2 nanoparticles for GSH over other potential interferences. The error bars were derived from the standard deviation of three measurements. Error bar = SD (n = 3)

Full size image
Table 1 Comparison of this method with the newly reported approaches for GSH determination
Full size table

To investigate the selectivity of the MSNs-G@MnO2 nanoprobe for GSH, the PGM signals were recorded in the presence of potential interferences, including a wide range of electrolytes and biological species (GSH at a concentration of 10 μM and all other compounds at concentrations of 10 mM). Among them, only reducing biomolecules (AA and GSH) caused comparatively high PGM signals, while amino acids and electrolytes did not induce an obvious increase in the PGM signal (Fig. 3b). Although AA can also cause a response to this system, its concentrations (μM levels) are relatively lower than that of GSH (mM levels) in biological systems [34, 35]. In addition, other thiols, such as cysteine, homocysteine have little effect on this system [14, 25, 27, 31]. Moreover, owing to the ideal sensitivity of our strategy, the serum sample needed to be diluted before the measurement. Under such conditions, the concentration of AA would be at nM level, and would not cause interference to the assay. Thus, the specificity of MSNs-G@MnO2 nanoprobe was acceptable, which makes it a promising method for the detection of GSH in biological samples.

Application analysis in serum samples

In order to estimate the potential applicability in complex biological samples, MSNs-G@MnO2 was applied to detect GSH in human blood samples. The accuracy of the sample analysis was measured by calculating the recovery of spiking a known amount of standard GSH in three 100-fold diluted human serum samples. The total analytical results are summarized in Table 1. 2 μM and 4 μM of GSH were added into each serum samples, respectively. The recovery of GSH is ranged from 97.5% to 102.5%, and the RSD is ranged from 1.9% to 3.5%. These results indicated that the MSNs-G@MnO2 nanoparticles had the advantages of being low-cost, rapid, and portable, which can be available for GSH detection in clinical applications (Table 2).

Table 2 Recoveries results of the determination of GSH in diluted serum samples
Full size table

Conclusions

We have successfully designed a portable, sensitive sensor for quantitative detection of GSH using a PGM based on the target-induced release of glucose from MnO2-nanosheet-gated MSNs. Interestingly, this detection method is fast and convenient. In addition, this method exhibits a good linear response to the concentration range from 100 nM to 10 μM with a detection limit of 34 nM. More importantly, compared with the standard instrumental sensing methods, this PGM-based assay system is low-cost, rapid, portable and user-friendly. Thus, the strategy can be used for quantitative detection of a wide range of non-glucose targets.