Introduction

Glutathione (GSH) is an important low molecular weight thiol compound in biosystems. Its structure is γ-l-glutamyl-l-cysteine-l-glycine. The thiol group in the cysteine (Cys) residue of glutathione is its bioactive site [1,2,3], with the high nucleophilic nature, enabling GSH to detoxify and scavenge free radicals under physiological conditions [4]. The GSH, and its disulfide (GSSG), is one of the most important redox systems in eukaryotic cells [5, 6]. The level of GSH directly affects the redox ability of organisms [6, 7]. This system plays an important role in maintaining the homeostasis of the cells [8]: First, GSH participates in the signal transduction of programmed cell death–related processes, which is called apoptosis [8]; second, it helps vitamin E, ascorbic acid, and other antioxidants to regenerate [9]. Besides, it is also usually used to evaluate redox state and detoxification ability of organisms [10]. What’s more, it can also be used to treat oxidation and free-radical–mediated injury [4]. Thus, GSH plays an important role in several diseases, such as liver damage, cervical, mammary, thymic, and colon cancers [11, 12]. In order to understand the influence of GSH in these diseases, the precise measurement of GSH concentrations in cells is greatly significant. Therefore, the development of a quantitative GSH method has been a heated topic.

Up to now, many methods have been established for quantitative detection of glutathione. The most commonly used ones are spectrophotometric and spectrofluorimetric methods [13, 14]. However, most of the developed chromophores and fluorophores exhibit poor selectivity between GSH and other thiol compounds in cells, which may cause interference to the test results [10, 15,16,17,18,19]. Fortunately, matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) has appeared as an appealing technology for detection of biological molecules in complex samples due to its inherent advantages, such as fast speed, simple operation, high throughput, and salt tolerance [20,21,22,23]. Although MALDI-TOF MS often emerges as a promising method for analysis of large biomolecules, it has poor performance toward detection of low molecular weight biomolecules like GSH due to the serious interference of matrix signal peaks in the low mass area (≤ 600 Da) [24]. Also, in order to overcome the weaknesses of poor reproducibility of the MALDI MS as a quantitative analysis tool, an ionic liquid is also often used [25, 26]. To selectively detect GSH, several nanomaterials including MALDI probes and matrices have been investigated. For example, Chiang et al. developed a gold nanoparticle as a new matrix for surface-assisted laser desorption/ionization mass spectrometry (SALDI MS) detection of GSH [27]. The gold nanoparticles were used to capture GSH and the quantitative detection of GSH could be readily achieved through using N-2-mercaptopropio-nylglycine (MPG)–modified gold nanoparticles (MPG-AuNPs) as the internal standard. Similarly, Oh et al. used maleimide-presenting AuNPs (Mal-AuNPs) for selective extraction of GSH and used for quantitative detection of GSH by SALDI MS [28]. The isotope-labeled GSH was used as internal standard. Recently, Wang et al. prepared an Au nanoparticle–modified porous silicon chip for selective detection of GSH by SALDI MS [29].

In this work, without using noble metal nanomaterials, we developed a novel tag-based strategy for selective and quantitative analyses of GSH in solutions through traditional MALDI-TOF MS (Scheme 1). Herein, four quinones, which have been reported to react with GSH with high efficiency via the thiol-ene Michael addition [30, 31], were firstly screened for tagging of GSH in solution. Among them, QLD appeared to be the most promising tag for the analysis of GSH by MALDI-TOF MS. Furthermore, the quantitative analysis of GSH in solution and cell lysates was also successfully demonstrated by using glutamic acid-cysteine-alanine (ECA) as an internal standard.

Scheme 1
scheme 1

Schematic illustration of the QLD-based tag strategy for quantitative assay of GSH by MALDI-TOF MS

Full size image

Materials and Experiments

Chemicals and Materials

GSH, α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (DHB), acetonitrile (ACN), 2,4,6-trihydroxyacetophenone (THAP), aminoacridine (9-AA), 2,7-dimethoxynaphthalene (DMAN), N-(1-naphthyl) ethylenediamine dihydrochloride (NEDC), trifluoroacetic acid (TFA), and ammonium biphosphate (ADP) were obtained from Sigma-Aldrich (Shanghai, China). 1,4-Benzoquinone (BQ), 1,4-naphthonequinone (NQ), and 1,4-anthraquinone (AQ) were purchased from J&K Scientific Ltd. (Beijing, China). Glutamic acid-cysteine-alanine (ECA) was synthesized by ChinaPeptides Co., Ltd. (Shanghai, China). QLD was purchased from WuXi AppTec Co., Ltd. (Shanghai, China). The colorimetric kit was purchased from Shanghai Institute of Biological Products Co., Ltd. (Shanghai China). All aqueous solutions were prepared using Milli-Q water by Milli-Q system. All commercial solvents and reagents were obtained in the highest available purity and used without further purification.

Sample and Matrix Preparation

The GSH and ECA were dissolved in phosphate buffer (PBS) solution at 2 mM. BQ, NQ, AQ, and QLD were dissolved in ACN/water 1/1 (v/v) as stock tag solutions (2 mM), respectively. The matrix DHB was prepared in 30% ACN/2 mM ADP aqueous solution at 20 mg mL−1. The CHCA and DHB were prepared in 50% ACN/0.1% TFA (v/v) (TA50) aqueous solution at concentrations of 5 and 20 mg mL−1, respectively. THAP and NEDC were dissolved in ACN/water 1/1 (v/v) ratio at 10 mg mL−1. 9-AA was dissolved in methyl alcohol at 10 mg mL−1. DMAN was dissolved in chloroform at 10 mg mL−1. One microliter of analyte and 1 μL of matrix solution were mixed on a MALDI steel target (MTP 384 ground steel target, Bruker Daltonics) and dried at room temperature before further MALDI-TOF MS analysis.

Determination of Reaction Efficiency by LC-MS Analysis

The peak areas of quinones at their retention time were obtained with a reverse-phase high-performance liquid chromatography (RP-HPLC) combined with electrospray ionization mass spectrometry (ESI-MS). A Zorbax Stablebond Analytical 300SB-C18 column (2.1 × 50 mm, 1.8 μm, Agilent Technologies) was used for separation. For elution, 50% mobile phase I (0.1% formic acid aqueous solution) and 50% mobile phase II (0.1% formic acid acetonitrile solution) were used in 0–5 min. The flow rate was 0.7 mL/min, and the detection wavelength was 190.0–400.0 nm. Quadrupole time-of-flight ESI-MS (microTOF-Q II, Bruker Daltonics Inc., Germany) was employed in positive ion mode. The peak area ratios of the pure quinones and quinones after reacting with GSH in different concentration ratios (quinones: GSH = 1:1, 2:1, and 4:1) were measured by LC-MS and used for calculated reaction efficiency (RE). The result was obtained as the average value of three independent measurements.

Quantitative Assay of GSH

Unless otherwise specified, GSH and ECA were separately mixed and reacted with tag solution at 1/1 (v/v) ratio for 60 min at room temperature. For quantitative analysis of GSH, the labeled ECA was pre-prepared as an internal standard with concentrations of 10 μM, 100 μM, and 1000 μM. After diluting the GSH solution to three different ranges of concentrations with 4~40 μM, 40~400 μM, and 400~4000 μM, the labeled ECA solutions were added to the labeled GSH solutions, respectively. The sample was directly analyzed after addition of MALDI matrix solution.

HeLa Cell Preparation

HeLa cells were grown in Dulbecco’s modified Eagle’s media (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS) at 37 °C in a humidified 5% CO2 incubator. After trypsin treatment, HeLa cells were centrifuged at 1500 rpm for 5 min to remove supernatant, followed by washing twice with PBS. Cell density was adjusted to 5 × 106 cells/mL. The cells were vortexed and cooled in ice for 15 min.

Quantitative Assay of GSH in HeLa Cell Lysate

The supernatant of cell lysate was collected after centrifugation. And then, a certain amount of supernatant was mixed with QLD solution at 1/1 (v/v) ratio for reaction 60 min at room temperature. Next, QLD-ECA at concentrations of 10, 100, and 1000 μM was added into corresponding mixture at QLD concentrations of 40, 400, and 4000 μM, respectively. The sample was directly analyzed after addition of MALDI matrix solution.

MALDI-TOF MS Analysis

MALDI-TOF MS experiments were performed on an Autoflex speed TOF/TOF mass spectrometry (Bruker, Germany) with a pulsed Nd:YAG laser at wavelength 355 nm. Each spectrum was obtained by 500 laser shots and a delayed extraction time of 150 ns in positive reflector and negative reflector ion modes. The laser energy was kept about 10% above the threshold of corresponding matrix to obtain the best signal quality with low noise. For the evaluation of the limits of detection (LODs), laser energy was adjusted from 0 to 100% above the threshold energy of corresponding matrix ions by variable laser attenuator. The LOD was defined as the lowest quantity of analyte detected with a S/N of [M + H]+ ≥ 3 based on MALDI-TOF MS evaluations of serial 10-fold dilutions of each analyte. All experiments were carried out at least three times to ensure the reproducibility of the spectral profiles. Flex Analysis 3.3080 was used to work up the data.

UV-Vis Analysis

UV-Vis absorbance was obtained using a UV-2550 spectrophotometer (SHIMADZU, Kyoto, Japan) from 300 to 400 nm at a wavelength step of 2 nm.

Results and Discussion

Primary Test of the Quinone-Based Tag Molecules

Ideally, a useful tag molecule for MALDI-TOF MS analysis should have three key characteristics: first, the tag molecule can react with target analyte with very high efficiency under mild conditions; second, the tag-analyte conjugate should have a relative high molecular weight to avoid interference by matrices in the low molecular area; third, the tag-analyte conjugate should process high ionization efficiency without fragmentation upon MALDI-TOF MS analysis [32, 33]. In the present work, four quinones, BQ, NQ, AQ, and QLD, were tested as tag molecules for quantification of GSH by MALDI-TOF MS analysis (Figure 1). It is reported that the quinones were able to react with GSH with very high efficiency through thiol-Michael addition reaction [34,34,36]. To test the performance of these tag molecules, the mass spectra of the equal-volume mixture of tag (2 mM) and GSH (2 mM) solution were first recorded using DHB as the matrix. As shown in Figure 2, all the tag-GSH conjugates could be clearly observed in the mass spectra without fragment ions. These results indicated that all the tag molecules can readily react with GSH. Moreover, the ion signal intensities corresponding to NQ-GSH and QLD-GSH were similar, both of which were higher than those of BQ-GSH and AQ-GSH. The results can be explained by the higher UV absorption intensities of NQ-GSH and QLD-GSH as compared with those of BQ-GSH and AQ-GSH (Figure S1). Meanwhile, the interference peaks of the usage of NQ tag between m/z 400 and 500 were more severe than that of the QLD tag. Therefore, QLD was selected for the following experiments. Meanwhile, the interference peaks of the usage of NQ tag between m/z 400 and 500 were more severe than that of the QLD tag. Moreover, the reaction efficiency of GSH with QLD was higher than those with AQ and NQ, and can reach > 99% at the molar ratio of QLD to GSH at 4:1 (Figure S2), which is appropriate for the precise quantitative analysis. Therefore, QLD was selected for the following experiments.

Figure 1
figure 1

The reaction of GSH with quinone-based tag molecules

Full size image
Figure 2
figure 2

MALDI mass spectra of equal-volume mixture of (a) BQ (2 mM) and GSH (2 mM), (b) NQ (2 mM) and GSH (2 mM), (c) AQ (2 mM) and GSH (2 mM), and (d) QLD (2 mM) and GSH (2 mM). All the mixtures were incubated at room temperature for 60 min before MALDI-TOF MS analysis using 20 mg mL−1 DHB as a matrix. BQ-GSH (m/z 415.1), NQ-GSH (m/z 465.1), AQ-GSH (m/z 515.1), QLD-GSH (m/z 466.1)

Full size image

Optimization of Sample Preparation Conditions

With QLD in hand, we thus tried to find an optimal condition for tagging GSH. Firstly, the kinetic of the reaction of QLD with GSH was monitored by UV-Vis spectroscopy, and the results are shown in Figure S3a. The reaction proceeded smoothly at room temperature and reached the maximum at about 60 min. In addition, the reaction between QLD and GSH was assessed to be independent on temperature (Figure S3b). Therefore, 60 min was determined as a promising reaction time for the GSH and QLD at room temperature before MALDI-TOF MS analysis.

It is well documented that the sample preparation conditions, including the selection of matrices, additives, and sample spotting methods, play critical roles in MALDI-TOF MS analysis [37]. Hence, six different matrices, including three MALDI matrices suitable for positive ion detection, i.e., THAP, DHB, and CHCA; and three matrices suitable for negative ion detection, i.e., 9-AA, DMAN, and NEDC, were screened for analysis of tagged GSH. As shown in Figure S4 and Figure S5, among all the matrices, DHB exhibited the strongest ion signal against the lowest interferences, compared with other tested matrices. Although a clean ion signal corresponding to the QLD-GSH was also obtained in the negative ion mode, its signal intensity was not as high as that in the positive ion mode. To further improve the MALDI-TOF MS analysis, the use of matrix additive was also investigated (Figure S6). It can be observed that ADP additive is able to increase the signal intensity of the QLD-GSH compared with those with TFA as an additive and those without any additive. And, as can be seen from Figure S6, the standard deviation (SD) of the experimental results stays the most minor with ADP as an additive, so we suggest it can improve the reproducibility of the experimental results [37]. With the optimized condition in hand, the LOD of GSH by using QLD tag was assessed to be 10 fmol μL−1 (Figure S7), which was improved by four orders of magnitude compared with that without using QLD (Figure S8). To further verify the reproducibility, the spectra of QLD-GSH were collected in different distributed regions of a sample spot, in different spots of a sample, or in different samples. As shown in Figure S9, the relative standard deviations (RSDs) of the peak intensity of QLD-GSH were 4.41%, 5.11%, and 10.36% for shot-to-shot, spot-to-spot, and sample-to-sample experiments, respectively. In a word, the proposed QLD-based method with high sensitivity and reproducibility is promising for GSH detection in biological samples, e.g., nanomoles per microliter level in cancer cells [16,17,18], while the traditional MALDI detection of GSH does not meet the sensitivity and reproducibility requirements.

Quantitative Analysis of GSH

In order to carry quantitative detection and avoid isotopic labeling, ECA (A: pI = 6.02) with one amino acid difference from the GSH (ECG, G: pI = 5.97) was synthesized as an internal standard (Figure 3a). Mass spectra of QLD-tagged ECA and GSH showed that their ionization abilities were very close at same concentrations (Figure 3b–d). The similar ionization ability of GSH and ECA can ensure the accuracy and reproducibility of results in quantitative experiments. Next, the quantitative analysis of GSH was demonstrated by using ECA as the internal standard (Figure 4). As shown in Figure 4a, when the ECA was set at a constant concentration of 1000 μM, the signal intensity of QLD-GSH got lower with the decrease of GSH concentration and a linear standard curve could be obtained by plotting the relative intensities of QLD-GSH and QLD-ECA (I(GSH)/I(ECA)) versus GSH concentrations of 400~4000 μM (Figure 4b). Similarly, the linear curves at GSH concentration ranges of 40~400 (Figure 4c) and 4~40 (Figure 4d) μM were also obtained by using 100 and 10 μM ECA as the internal standards, respectively. The linear correlation coefficients (R2) of all the three standard curves were higher than 0.998, indicating the excellent linearity over three orders of magnitude. Moreover, the SDs of the peak ratios measured at different GSH concentrations (4~4000 μM) range from 0.003 to 0.024, with an average value of 0.011.

Figure 3
figure 3

(a) Structural comparison between GSH and ECA. MALDI mass spectra of the equal-molar mixture of QLD-GSH and QLD-ECA at a concentration of (b) 1000, (c) 100, and (d) 10 μM, using 20 mg mL−1 DHB as a matrix. QLD-GSH (m/z 466.1). QLD-ECA (m/z 480.1)

Full size image
Figure 4
figure 4

MALDI-TOF MS quantitative detection of GSH using ECA as the internal standard. (a) Typical mass spectra of equal-molar mixture of QLD and GSH at 400, 800, 2000, and 3200 μM using 1000 μM ECA as the internal standard. Linear standard curves for GSH at concentration ranges of (b) 400 to 4000 μM, (c) 4 to 40 μM, and (d) 40 to 400 μM. QLD-GSH (m/z 466.1). QLD-ECA (m/z 480.1)

Full size image

To get the access of the recovery rate, we further detected the GSH at nine concentrations within the range of 10~3000 μM. The results are listed in Table 1. All of the recovery rates corresponding to these nine GSH concentrations were more than 97% and the SDs were no more than 8%. The results suggested the accuracy and reliability of the proposed method.

Table 1 Recoveries of GSH at Nine Added Concentrations
Full size table

Qualitative and Quantitative Detection of GSH in Cell Lysates

Encouraged by the above results, we further applied the present method for determination of the endogenous GSH level in the cancer cell lysates. The endogenous GSH level in HeLa cell lysates assessed by this method was 59.02 ± 1.37 μM, which is close to 52.08 ± 18.04 μM obtained by using the commercial colorimetric kit (Figure 5). Nevertheless, it should be mentioned that our proposed method has more minor SD and better reproducibility than the commercial colorimetric method. Moreover, the proposed MALDI-TOF MS method, with low sample consumption as well as high-throughput features, can avoid the interference from the sample autofluorescence and other factors presented in the spectroscopy assay to a large extent.

Figure 5
figure 5

MALDI-TOF MS quantitation of GSH in HeLa cell lysates using 100 μM ECA as the internal standard. (a) Mass spectra of HeLa cell lysates using 20 mg mL−1 DHB as matrix. (b) Calibration curve of GSH in HeLa cell lysates. (c) Quantitative comparison of the GSH concentration detected by the proposed MALDI-TOF MS method and commercial colorimetric method. The data were obtained from three independent experiments. QLD-GSH (m/z 466.1). QLD-ECA (m/z 480.1)

Full size image

Conclusions

In this work, we have designed a simple, cost-/time-efficient, and sensitive tag-based strategy for quantitative analysis of GSH via MALDI-TOF MS. A UV-absorptive aromatic QLD tag was selected for efficient tagging of GSH through a rapid thiol-ene click reaction. Furthermore, an excellent linear relationship with more than three orders of magnitude dynamic range was readily achieved by combination use of a non-isotope ECA as an internal standard. Finally, the quantitative analysis of GSH was also successfully achieved in HeLa cell lysates. Therefore, based on the simple preparation process and the ability of accurate detection, our proposed QLD-based tag strategy may show tremendous applications in fast quantification of GSH in complex biological samples by MALDI-TOF MS.