H₂O₂-free colorimetric sensing platform of Mn-doped carbon dots with oxidase-mimetic activity for the detection of glutathione in liver disease serum

Abstract

The level of glutathione (GSH) is a crucial indicator in various pathological processes. Developing a sensitive and selective method for GSH detection is of great importance for the evaluation of the human health. In this work, Mn-doped carbon dots (Mn-CDs), which exhibit a simulated oxidase activity, were utilized to construct an H₂O₂-free colorimetric sensor for the detection of GSH. Mn-CDs catalyse the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB) to ox-TMB without the need for unstable H₂O₂, and this catalysis is inhibited by GSH, leading to a decrease in the absorbance at 652 nm. The colorimetric sensor detection of GSH demonstrated a strong linear relationship in the range of 0.5–800 µM, with a detection limit as low as 0.167 µM, a 96.5%–108.5% recovery rate, and excellent selectivity for GSH among 10 main interfering substances in human serum. Furthermore, we established a method for the visual estimation of GSH concentration based on the H₂O₂-free Mn-CDs colorimetric sensor, greatly simplifying the detection process. Compared with a commercial kit, our colorimetric sensor method provided consistent results and offered advantages in terms of cost, convenience, and potential for point-of-care testing. Importantly, we successfully applied the sensor to detect GSH levels in the serum of patients with liver disease, finding significant differences compared with healthy individuals. Overall, these results demonstrated that the H₂O₂-free Mn-CDs colorimetric sensor for detecting GSH has the characteristics of a wide linear range, excellent minimum detection limit, strong anti-interference capabilities, low cost, convenience, and stability and can be used to assess the human health.

---

1. Introduction

Glutathione (GSH) is a prevalent soluble thiol antioxidant and low molecular weight peptide within cells.¹ Its functions include the detoxification of electrophilic substances, the elimination of reactive oxygen and nitrogen species, the preservation of the fundamental thiol redox state of proteins, and the regulation of DNA synthesis, cellular growth, proliferation, and immune responses.² Hence, the detection of GSH could be utilized for the purpose of evaluating health, with a particular focus on the liver, which is the predominant source of GSH and releases nearly all synthesized GSH into the plasma.³ In recent years, various methods for detecting GSH, including surface-enhanced Raman spectroscopy, electrochemical sensing, and liquid chromatography, have been developed.^4–6 However, the employment of such techniques is contingent upon the possession of sophisticated instrumentation, the incurring of substantial detection costs, and the utilization of professional operation to ensure their judicious application. Therefore, developing a cost-effective, accurate, and user-friendly method for the assessment of GSH levels is important for the effective evaluation of the human health.

Colorimetric analysis has emerged as a cornerstone in modern analytical systems because of its cost-effectiveness, simplicity, and rapidity, and nanozyme-based colorimetric sensing has particularly achieved groundbreaking advancements. Nanozymes, regarded as one of the top 10 emerging technologies in chemistry for 2022 by the International Union of Pure and Applied Chemistry (IUPAC),⁷ exhibit significant advantages in terms of low synthesis cost, stability, and tunable activity, and they have been explored in biosensors and detection.^8–10 Although significant progress has been made in nanozyme-based GSH detection technologies, current research still faces several critical challenges. Firstly, their detection mechanisms are dependent on peroxidase-like activity and require the introduction of chemically unstable H₂O₂ as the reaction medium, which severely restricts the practical application stability of detection systems.^11–13 Secondly, most existing systems of GSH rely on noble metal-based nanomaterials, whose intricate preparation processes significantly increase the synthesis costs.^11,12,14 Thirdly, current methods of GSH exhibit narrow linear detection ranges and relatively high detection limits, failing to meet the high-sensitivity detection requirements for GSH in clinical serum samples.^13,15,16

Carbon-based nanozymes have attracted considerable attention in the field of analytical sensing because of their low toxicity, biocompatibility, and eco-friendliness. Heteroatom-doped nanozymes have been shown to exhibit increased catalytic activity due to improved electron transfer efficiency and other special properties, such as water solubility, vacancies, light absorption, magnetism, and fluorescence, and have been proven to be an effective strategy for regulating the simulated activity of nanozymes.^17–19 In the field of biosensing based on doped carbonaceous nanozymes, the peroxidase (POD)-mimicking catalytic strategy has been widely explored. However, the intrinsic limitation of this strategy lies in its strict dependence on H₂O₂ as an electron acceptor.^20,21 The instability of H₂O₂, the required exogenous addition, and its dynamic fluctuations within cells introduce significant complexity and uncertainty for practical applications, becoming a key bottleneck restricting its development.

In this study, we have synthesized manganese-doped carbon dots (Mn-CDs) nanozymes exhibiting intrinsic oxidase-mimicking activity. Mn-CDs can activate ambient dissolved oxygen (O₂) to generate superoxide anion radicals (O₂˙⁻) in situ via a single-electron transfer pathway, which then directly oxidize colourless 3,3′,5,5′-tetramethylbenzidine (TMB) to blue oxidized TMB (ox-TMB), achieving efficient enzymatic catalysis without the requirement of exogenous H₂O₂. The addition of GSH inhibits the direction of this reaction, thus allowing the visual estimation and spectrophotometric quantitative colorimetric detection of GSH (Scheme 1). In contrast to nanozymes based on precious metals, Mn-CDs developed herein were synthesized via a one-pot hydrothermal method using low-cost precursors, including citric acid, ethylenediamine, and manganese chloride, offering a significant advantage in terms of material cost. The method showed a detection range of 3 orders of magnitude and high selectivity among 10 main interferents and demonstrated satisfactory outcomes in the analysis of GSH in actual human serum samples, including those spiked with recovery tests, healthy human serum samples, and serum samples from patients diagnosed with liver disease.

Scheme 1 Schematic of the functional principle of the colorimetric detection of GSH using the oxidase-like mimetic activity of Mn-CDs.

2. Experimental section

2.1 Reagents

MnCl₂·4H₂O, citric acid, ethylenediamine (EDA), TMB, acetic acid (HAc), sodium acetate (NaAc), GSH and reduced GSH content assay kits were purchased from Macklin, China. These reagents were analytically pure. Serum was collected from healthy volunteers and patients with liver disease at the First Hospital of Lanzhou University. Deionized water was used throughout the study.

2.2 Instruments

The morphology of the material was observed via transmission electron microscopy (TEM) using a JEM-2100 transmission electron microscope (JEOL, Japan). The ultraviolet visible (UV-Vis) absorption spectra were obtained on a UV-2550 spectrophotometer (Shimadzu, Japan). Fluorescence spectra were obtained on an FS-5 fluorescence spectrometer (Edinburgh Instruments, UK). A D/Max-2400 diffractometer (Rigaku Corporation, Japan) was used for X-ray diffraction (XRD) measurements. A NEXUS 670 spectrometer (Nicolet, USA) was used for Fourier transform infrared (FTIR) spectroscopy. X-ray photoelectron spectroscopy (XPS) was performed on an Axis Supra spectrometer (Shimadzu, Japan). An EMXplus-6/1 electron paramagnetic resonance (EPR) spectrometer (Bruker, Germany) was used to detect free radicals.

2.3 Synthesis of Mn-CDs

Mn-CDs were prepared via a simple one-pot hydrothermal method. Specifically, 5 mmol citric acid and 5 mmol MnCl₂·4H₂O were completely dissolved in 20 mL of deionized water, and 10 mmol EDA was subsequently added to the mixture. After stirring for 10 min, the solution was transferred to a hydrothermal reactor. After treatment at 180 °C for 10 hours, the supernatant was collected and purified using an MD45 (1000 Da) dialysis bag for 24 hours. Then, the brown solution was centrifuged at 10 000 rpm for 5 minutes to remove solid particles and was further purified by filtration via a 0.22 µm filter membrane. After that, the collected solution was freeze-dried to obtain the Mn-CDs powder. For comparison, undoped bare CDs were also prepared using only CA and EDA.

2.4 Oxidase-mimetic catalytic activity of Mn-CDs

The oxidase-mimetic activity of Mn-CDs was evaluated by measuring the absorbance of ox-TMB at 652 nm. In short, a Mn-CDs dispersion or CDs dispersion (1 mg mL⁻¹, 200 μL) and TMB (3 mM, 100 μL) were added to 2700 μL of NaAc-HAc solution (0.2 M, pH 3.5). After a sufficient reaction time, the UV-Vis absorption spectrum was obtained using a spectrophotometer. To elucidate the catalytic mechanism of the Mn-CDs nanozyme, we investigated its steady-state kinetics and detected the free radicals produced during the reaction.

2.5 Colorimetric sensing determination of GSH with Mn-CDs

The signal for colorimetric sensing detection was based on the characteristic absorption peak of the ox-TMB substrate at 652 nm and the colour change. The analysis and detection of GSH using the colorimetric sensing system was carried out as follows: first, 200 μL of Mn-CDs solution (1 mg mL⁻¹) was added to 100 μL of TMB solution (3 mM) with 2600 μL of NaAc-HAc solution (0.2 M, pH 3.5). Next, the GSH sample was added and reacted at room temperature for 20 minutes, then the UV-Vis absorption spectrum of the solution was obtained, and the absorbance at 652 nm was recorded.

The GSH calibration curve was obtained by plotting the relative GSH concentration of ΔA at 652 nm (where ΔA = A₀–A₆₅₂, A₀ and A₆₅₂ are the absorbances of the solution in the absence and presence of GSH, respectively), and the minimum detection limit (LOD) was calculated at a signal-to-noise ratio (S/N) of 3. In the selectivity experiment, 500 μM of potentially interfering ions or substances, including K⁺, Na⁺, Ca²⁺, Mg²⁺, Zn²⁺, glycine, L-aspartate, urea, glucose and BSA, were used instead of GSH, and the other testing conditions were the same.

2.6 Real sample analysis

Twenty-one patients (ten males and eleven females, with an average age of 56.43 ± 5.84 years) with liver diseases, including seven patients with viral hepatitis, six patients with cirrhosis, and eight patients with malignant liver tumours at the First Hospital of Lanzhou University, were involved in the study. In addition, the healthy control group included 19 age- and sex-matched healthy subjects. The GSH levels in the serum samples from both liver disease patients and healthy subjects were analysed via the Mn-CDs colorimetric sensing method.

2.7 Statistical analysis

The data were subjected to statistical analysis utilizing the SPSS 25.0 software. Measurement data were described as mean ± standard deviation. The comparison of serum GSH levels between the healthy control group and the liver disease group was performed using the Mann–Whitney U test because it did not conform to the normal distribution. Asterisks denote a statistically significant difference (***p < 0.001). The receiver operating characteristic curve (ROC) was used to distinguish the levels of serum GSH in healthy and liver-diseased subjects, and the area under the curve (AUC) was calculated for the discrimination potential.

3. Results and discussion

3.1 Characterization of Mn-CDs

Mn-CDs were synthesized using a one-step hydrothermal method. The morphology and particle size of Mn-CDs were investigated via TEM. As shown in Fig. 1A, Mn-CDs appeared to be spherical particulates, and their average particle size distribution was approximately 2.12 nm. Furthermore, Mn-CDs were dispersed in a NaAc-HAc buffer solution at pH 3.5, and the optical properties of Mn-CDs were analysed via UV-Vis absorption and fluorescence spectra. The UV–Vis absorption spectrum of Mn-CDs revealed a distinct absorption band at 340 nm (Fig. S1), which was caused by the n–π* transition of the O–H and C=O/C=N groups on the surface of Mn-CDs.^22–24 Notably, the optimal excitation wavelength of fluorescence is 343 nm, with good agreement between the excitation spectrum and the corresponding chemical structure absorbed at 340 nm (Fig. S1), which was confirmed to be the cause of the photoemission.^25,26 When the 343 nm excitation wavelength was applied, a strong blue emission spectrum centred at 457 nm was obtained (Fig. S1). In addition, the bulk phase structural properties of Mn-CDs were analysed via XRD. As shown in Fig. 1B, the bare CDs have a broad peak at 2θ = 24.68°, which can be assigned to the 002 crystal plane.²⁷ In comparison, doped Mn exists in a highly dispersed state and is coordinated by various C-based and N-based functional groups, and no diffraction peaks were observed.²³ These results reveal that the amorphous structure of both materials and the carbon atoms are highly disordered.²⁸

Fig. 1 (A) TEM image and particle size distribution (inset) of Mn-CDs. (B) XRD patterns of Mn-CDs and CDs. (C) FTIR spectra of Mn-CDs and CDs. (D) XPS spectra of Mn-CDs and CDs. (E) XPS spectrum of Mn 2p in Mn-CDs. (F) Comparison of the O 1s XPS spectra of Mn-CDs and CDs.

The chemical groups on the surface of Mn-CDs were subsequently detected via FTIR. Fig. 1C shows a broad peak in the spectrum range of 3200–3600 cm⁻¹, corresponding to the stretching vibration of O–H at 3432 cm⁻¹ and N–H at 3254 cm⁻¹, demonstrating the good hydrophilicity of Mn-CDs.^27–29 The stretching vibrations of C–H, C=O/C=N in amides, C–N, and C–O–C were observed at 2923 cm⁻¹, 1657 cm⁻¹, 1386 cm⁻¹, and 1074 cm⁻¹, respectively (Fig. 1C). Additionally, the bending vibration of N–H appeared at 1551 cm⁻¹, which is consistent with a previous report.²³ Interestingly, the spectrum of Mn-CDs exhibited a very small peak at 647 cm⁻¹, which can be attributed to the stretching vibration of the Mn–O/Mn–N bond.³⁰ These results confirmed the presence of trace amounts of metal oxides on the surface of Mn-CDs.

Furthermore, the chemical composition was analysed by XPS, verifying the presence of C, O and N in both Mn-CDs and CDs (Fig. 1D). Interestingly, significant Mn 2p and Mn 3p signals were also observed in Mn-CDs (Fig. 1D), demonstrating the successful doping of Mn into the CDs. The Mn 2p peaks were decomposed into two peaks, Mn 2p_3/2 and Mn 2p_1/2, with binding energies of 641.1 and 653.3 eV, respectively, with a standardized separation value of 12.2 eV. The binding energies were further assigned to Mn II (641.1 and 653.0 eV) and Mn III (642.6 and 654.4 eV) and the corresponding vibrational peaks (Fig. 1E), confirming the presence of Mn²⁺ and Mn³⁺.^23,30

The high-resolution O 1s, N 1s, and C 1s spectra of the prepared CDs and Mn-CDs are shown in Fig. 1F and Fig. S2A and B. Evidence of Mn–O and Mn–N formation was observed at binding energies of 530.9 eV and 399.7 eV;³¹ Mn–O accounts for 26.20% of the total oxygen species, while Mn–N and pyridinic-N are included in the N 1s spectra, accounting for 64.84%, indicating the successful preparation of the material. The C=O (531.0, 531.7 eV) and C–O (532.4, 532.6 eV) peaks for the CDs and Mn-CDs are shown in Fig. 1F, and the pyridinic-N (399.5, 399.7 eV) and pyrrolic-N (400.2, 400.9 eV) peaks are shown in Fig. S2A.³² In the C 1s spectrum (Fig. S2B), the C–C/C=C, C–N, C–O and C=N/C=O peaks of the CDs are at 284.5, 285.1, 286.2, and 287.7 eV, whereas those of Mn-CDs are 284.6, 285.2, 286.4, and 288.3 eV, respectively. Notably, the corresponding XPS spectra of the C, N, and O Mn-CDs shifted by different degrees to higher binding energies, and the chemical bond percentage changed compared with that of the CDs. This phenomenon is caused by the doping of Mn, which promotes electron transfer between Mn and these species, reduces the electron density on the surface of Mn-CDs,²³ introduces new chemical bonds or changes the nature of the original chemical bonds.

3.2 Mn-CDs oxidase-like catalysed reactions and mechanisms

The simulated enzyme properties of Mn-CDs and CDs were analysed under the same conditions using TMB as a chromogenic substrate. As shown in Fig. 2A, the reaction system of “CDs + TMB + NaAc-HAc” does not have an apparent absorption peak at 652 nm, indicating that the oxidase-like activity of the CDs is not obvious. In contrast, “Mn-CDs + TMB + NaAc-HAc” displayed an obvious UV-Vis absorption signal at 652 nm, demonstrating that the reaction system successfully catalyzed and oxidized TMB to blue ox-TMB. These results confirmed that Mn-CDs had excellent oxidase-like activity and that the doping of Mn dramatically improved the enzyme activity. We further investigated the steady-state kinetics of Mn-CDs by analysing the velocity with various concentrations of TMB. A typical Michaelis–Menten curve was subsequently obtained (Fig. 2B), and a Lineweaver–Burk plot was fitted with a double reciprocal (Fig. 2C), yielding parameters K_m = 0.3576 mM and V_max = 0.3664 × 10⁻⁸ M s⁻¹. Compared with HRP (horseradish peroxidase, K_m = 0.434 mM), it has a smaller K_m value, indicating that the Mn-CDs nanozyme has a better affinity for TMB. In addition, the zeta potential measurement of Mn-CDs (1 mg ml⁻¹) showed a value of −13.5 mV (Fig. S3), suggesting that Mn-CDs carry negative charges, which accelerate the reaction via electrostatically adsorbing the positively charged TMB.

Fig. 2 (A) UV–Vis absorption spectra of Mn-CDs and CDs. (B) Plot of the Michaelis–Menten equation. (C) Plot of the Lineweaver–Burk equation. (D) EPR spectrum of Mn-CDs + DMPO in methanol. (E) Oxidase-like catalytic stability of Mn-CDs. (F) UV–Vis spectra of different systems.

To elucidate the catalytic mechanism, 5,5-dimethyl-1-pyrroline N-oxide (DMPO) was used to trap the free radicals produced in the catalytic reaction. EPR spectra assigned to O₂˙⁻ signals were observed, and the results established that the generated O₂˙⁻ played a role in the peroxidase-like activity of Mn-CDs during the catalytic process, which oxidized TMB (Fig. 2D). The catalytic mechanism originates from the ability of Mn-CDs to directly catalyse the production of O₂˙⁻ from dissolved O₂ in the solution, thereby efficiently driving the oxidation of the chromogenic substrate. While the addition of H₂O₂ might, in theory, provide an additional oxidative pathway and temporarily enhance the signal, it would do so at the cost of compromising the long-term stability and operational simplicity of the system. In contrast, the present system, by leveraging the oxidase-like activity of Mn-CDs to utilize dissolved oxygen, eliminates the reliance on the unstable and strongly oxidizing H₂O₂ that is typically required in peroxidase-like catalytic systems, avoiding the practical application instability of detection systems.^11–13 The enzymatic catalytic stability of Mn-CDs was notable, remaining at 91.76% even after 28 days (Fig. 2E).

3.3 Development of the GSH method based on H₂O₂-free Mn-CDs colorimetric sensing and the optimization of reaction conditions

GSH has strong reducibility, which can inhibit the oxidation of colourless TMB to blue ox-TMB catalysed by the oxidase-mimicking nanozyme. The underlying inhibition mechanism can be explained as follows: the oxidase-mimicking activity of Mn-CDs relies on the generation of O₂˙⁻ to oxidize TMB. When GSH is introduced, its thiol group (–SH) acts as a potent reducing agent. It competitively consumes the generated O₂˙⁻ radicals and can also reduce the already formed ox-TMB back to its colorless state. This reaction simultaneously oxidizes GSH to its dimeric form, oxidized glutathione (GSSG).^33,34 Consequently, we detected the GSH content by analyzing the colour change reflected in the UV-Vis absorption spectrum (Scheme 1). As expected, the peak strength at 652 nm was sharply reduced when GSH was added to the solution (Fig. 2F). The oxide-like simulation activity of Mn-CDs was inhibited, resulting in a weakening of the oxidase-like reaction catalysed by Mn-CDs and a decrease in the conversion efficiency of TMB to ox-TMB. This Mn-CD-mediated reaction utilizes dissolved oxygen through its conversion to O₂˙⁻ instead of the easily decomposable oxidizing agent H₂O₂, ensuring the stability of detection systems. The catalytic reaction involves the conversion of dissolved oxygen to O₂˙⁻ by Mn-CDs rather than the easily decomposable oxidizing agent H₂O₂. The addition of GSH inhibits the oxidative mimicking activity of Mn-CDs, leading to a weakened oxidase-like reaction catalysed by it, and a decreased efficiency in the transformation of TMB to ox-TMB.

After proving the feasibility of detecting GSH via H₂O₂-free Mn-CDs colorimetric sensing with oxide mimetic activity, the optimal conditions were further optimized by exploring the change in absorbance at 652 nm under different experimental conditions, including different pH values, TMB concentrations, temperatures and reaction times. Since GSH has a variety of functional groups, such as carbonyl, amine and mercaptan groups,³⁵ and TMB has limited solubility in alkaline solutions because of its synonymous structure with diamines,³⁶ the pH is the key factor affecting detection. Fig. S4A shows that the optimal pH of the detection system is 3.5, which is similar to that of other nanozymes.^37,38

The optimal conditions of substrate TMB, reaction time and temperature for the stable inhibition of TMB by GSH to complete oxidation were subsequently investigated. The results showed that the oxidation reaction of TMB is much more adequate in weakly acidic solutions than in strongly acidic and nearly neutral solutions and the optimum absorbance peaked at a TMB concentration of 3 mmol L⁻¹ (Fig. S4B), and the reaction appeared to stabilize at 20 minutes (Fig. S4C). The influence of temperature on GSH detection remained unchanged in the range of 20–25 °C (Fig. S4D), which indicates that the reaction has low sensitivity to ambient temperature and can be carried out at room temperature. All subsequent experiments for GSH detection were conducted under these optimized conditions.

3.4 Evaluation of the H₂O₂-free Mn-CDs colorimetric sensing detection platform

In accordance with the principles outlined above, a H₂O₂-free colorimetric sensing system was developed to detect GSH utilizing Mn-CDs (Fig. S5). The system's linear range, detection limit, and selectivity were subsequently assessed. The ΔA exhibited a good linear relationship with the GSH concentration in the range of 0.5–800 μM (Fig. 3A and B). The regression equation was ΔA = 0.9129c + 0.0412 (GSH, mM), and the correlation coefficient (R²) was 0.9922 (Fig. 3B). The linear range of GSH detection by the H₂O₂-free Mn-CDs colorimetric sensing system was up to three orders of magnitude, which is better than previous reports that usually obtained a linear range of 2 orders of magnitude or even below.^{12,13,16,39–44} The LOD was 0.167 µM, which is lower than previous studies,^11,14,15 suggesting that the detecting performance of the H₂O₂-free Mn-CDs colorimetric sensing platform is excellent.

Fig. 3 (A) UV–Vis absorption patterns of NaAc-HAc + TMB + Mn-CDs + GSH at various concentrations. (B) Linear relationship between the ΔA_{652 nm} value and the GSH level. (C) GSH assay selectivity against 10 possible interfering compounds (500 μM interferons and GSH were added to the system). (D) Visual colorimetric table for detection under various GSH concentrations. (E) Visual comparison of serum samples from hepatitis, cirrhosis and liver cancer patients before (left) and after (right) the addition of 500 μM GSH in the serum samples.

In order to further assess the selectivity of this H₂O₂-free Mn-CDs colorimetric sensing approach for GSH, the anti-interference experiments were conducted to examine the effects of the 10 main components in human serum, including K⁺, Na⁺, Ca²⁺, Mg²⁺, Zn²⁺, glycine, L-aspartate, urea, glucose and BSA. As illustrated in Fig. 3C, all 10 components with 500 μM concentration exhibited significantly lower colorimetric values than GSH, demonstrating that the H₂O₂-free Mn-CDs colorimetric sensing system of GSH has strong anti-interference ability and can be used to analyze GSH in human serum. It should be noted that while other endogenous reducing substances in human serum may theoretically cause potential interference to the system, their concentrations are significantly lower than that of GSH. Therefore, this study has validated the most critical potential interferents, providing compelling selectivity evidence for further exploration of the sensor's practical application in complex serum matrices. Table 1 presents the data on repeatability and precision from our experiment. By utilizing both 400 and 600 μM concentrations of GSH, the intra-assay coefficient of variation (CV) ranged from 0.92% to 2.76%, while their inter-assay CV ranged from 5.63% to 8.93%, which are below the levels of 5% and 10%, respectively. These results convincingly demonstrate that the developed H₂O₂-free Mn-CDs colorimetric sensing approach for GSH has excellent repeatability and precision.

Table 1 Reproducibility and precision of the detection of GSH based on Mn-CDs

Assay batches	Added GSH (μM)	Measured GSH (μM)	SD (μM)	CV (%)
1 (n = 10)	400	373.6	10.30	2.76
	600	575.8	5.32	0.92
2 (n = 10)	400	387	8.37	2.16
	600	582.4	9.81	1.68
3 (n = 10)	400	416.8	11.12	2.67
	600	673.4	8.84	1.31
Inter-assay (n = 3)	400	392.5	22.11	5.63
	600	610.5	54.54	8.93

---

Furthermore, we developed a visual colorimetric method based on the H₂O₂-free Mn-CDs colorimetric sensing system for the rapid detection of GSH and to investigate the potential sensing capacity of the Mn-CDs probe as a colorimetric indicator. Notably, under the otherwise optimized system parameters, a reaction time of 10 minutes was sufficient to achieve a clear visual distinction. An intuitive and semiquantitative representation of the GSH concentration is shown in Fig. 3D. The serum obtained from patients with hepatitis, liver cirrhosis, and liver cancer was analysed by the visual Mn-CDs colorimetric method, and an obvious visual difference was observed between the original serum and the serum with the added 500 μM GSH (Fig. 3E). This distinct H₂O₂-free Mn-CDs colorimetric method is significantly simplified and rapid and could be utilized as a portable detection device for in-field monitoring. It also has the advantages of low raw material cost, simple synthesis and independent detection without relying on H₂O₂. Consequently, the developed visual H₂O₂-free Mn-CDs colorimetric method for GSH has good application potential in the biomedical field.

3.5 Application in clinical serum samples analysis

Considering the vital changes in GSH in liver diseases, the Mn-CDs technology for GSH detection developed in this study holds substantial importance for the early detection of liver diseases. To evaluate the practical application of the Mn-CDs colorimetric sensing method, we carried out the parallel analysis of GSH at various concentrations in GSH standards (0 mM, 0.2 mM, 0.4 mM, 0.6 mM, and 0.8 mM) and clinical serum samples from patients with liver disease via the H₂O₂-free Mn-CDs colorimetric sensing method, and a 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) microplate assay commercial kit based on the reaction of GSH with the chromogenic substrate DTNB, which yields the stable yellow product TNB and GSSG; the absorbance at 412 nm, reflecting the GSH concentration, was used as a control. The results showed that the R² values from the H₂O₂-free Mn-CDs colorimetric sensing method and the commercial GSH kit in detecting GSH in the GSH standards and clinical serum were 0.9989 and 0.9711, respectively (Fig. 4A and B). This demonstrates that the H₂O₂-free Mn-CDs colorimetric sensing method exhibits excellent concordance with the assay commercial GSH kit and can be applied in analyzing GSH in clinical serum samples.

Fig. 4 Consistency of the GSH standards (A) and clinical serum (B) assayed using the Mn-CDs sensing system and commercial DTNB microplate assay kits. (C) Mn-CDs sensing system used to analyse the GSH levels in the serum of healthy individuals and patients with liver disease and (D) corresponding ROC curves.

Furthermore, we performed spiked recovery experiments to explore the accuracy of the H₂O₂-free Mn-CDs colorimetric sensing method in clinical samples. Three human serum samples were supplemented with a known amount of 0.2 mM GSH, and their recovery rates were 96.5%–108.5%, and their relative standard deviations (RSD) were 0.8%–3.0% (Table 2). These results suggest that the accuracy and precision of the H₂O₂-free Mn-CDs colorimetric sensing method are excellent for detecting GSH in serum samples.

Table 2 Determination of GSH in actual human serum samples

Number of serum samples	Added (mM)	Found (mM)	Recovery (%)	RSD (%)
1	0	0.124
	0.200	0.317	96.5	0.8
2	0	0.286
	0.200	0.503	108.5	3.0
3	0	0.259
	0.200	0.467	104.0	0.7

---

In addition, to validate the clinical detection efficacy of the H₂O₂-free Mn-CDs colorimetric sensing system for GSH, we analysed serum samples from 21 patients with liver disease and 19 healthy controls using the H₂O₂-free Mn-CDs colorimetric sensing method. Distinct from conventional GSH detection methods that often rely on sophisticated instruments or enzymatic reactions, the developed H₂O₂-free Mn-CD-based colorimetric sensing platform leverages its intrinsic oxidase-mimetic activity, offering significant advantages in terms of operational simplicity, low cost, and minimal sample pretreatment, presenting a promising strategy for point-of-care testing. As shown in Fig. 4C, the levels of GSH were significantly lower in the serum from patients with liver disease when compared to the healthy controls (p = 0.0002). This result is consistent with previous reports on redox homeostasis disruption during hepatic pathogenesis.^45–47 Critically, ROC curve analysis revealed that the H₂O₂-free Mn-CDs colorimetric sensing system exhibits a robust discriminative power with an area under the curve (AUC) of 0.826 for distinguishing between healthy and liver disease states (Fig. 4D). Although only a small number of clinical samples were used to analyse the clinical performance of the Mn-CDs colorimetric sensing system, our data revealed the pathological signature of serum GSH in patients with impaired liver function, highlighting the translational potential of integrating this biomarker with advanced nanosensing platforms for constructing early warning systems for hepatic disorders. The platform's high sensitivity, operational simplicity, and rapid response position it as a viable tool for convenient liver function screening and early-stage monitoring, particularly in primary care settings or resource-limited environments.

4. Conclusions

In summary, a H₂O₂-free Mn-CDs colorimetric sensing method has been proposed for the monitoring of the oxidation of GSH based on the principle that Mn-CDs can produce O₂˙⁻ in a solution containing dissolved oxygen, catalysing the oxidation of TMB into the blue ox-TMB product, and the presence of GSH can neutralize these free radicals, inhibiting the production of ox-TMB. The method has a wide linear range spanning 3 orders of magnitude, excellent minimum detection limit, strong anti-interference capabilities, and exceptional repeatability and precision. It features simple maneuverability without the need for heating reactions, demonstrates cost-effectiveness due to the absence of precious metals, shows excellent accuracy and precision for detecting GSH in serum samples, and is a potentially feasible analytical method for the evaluation of human health.

Author contributions

Lin Bo: data curation, writing – original draft. Lili Jiang: conceptualization, writing – review & editing, supervision. Zhaogui Deng: data curation, formal analysis, validation. Minmin Xu: methodology, data curation. Yifei Liu: data curation, validation. Hui Zhang: data curation, validation. Xianwei Zuo: conceptualization, writing – review & editing, supervision. Haitao Yu: conceptualization, supervision, Writing – review & editing, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Ethical statement

Human serum samples were provided by the First Hospital of Lanzhou University. We confirm that all the experimental procedures adhered to the principles outlined in the Helsinki Declaration and were approved by the Ethics Committee of Lanzhou University and the First Hospital of Lanzhou University (No. LDYYLL-2025-64).

Data availability

All data supporting this study are included in the article and its supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5an00964b.

Acknowledgements

This work was financially supported by the National Key Research and Development Program of China (No. 2024YFE0107000) and Lanzhou Science and Technology Planning Project (Grant No. 2023-2-37).

References

1. A. Meister and M. E. Anderson, Annu. Rev. Biochem., 1983, 52, 711–760.
2. L. Wang, Y. J. Ahn and R. Asmis, Redox Biol., 2019, 31, 101410.
3. M. Vairetti, L. G. Di Pasqua, M. Cagna, P. Richelmi, A. Ferrigno and C. Berardo, Antioxidants, 2021, 10, 364.
4. Á. Sánchez-Illana, F. Mayr, D. Cuesta-García, J. D. Piñeiro-Ramos, A. Cantarero, M. d. l. Guardia, M. Vento, B. Lendl, G. Quintás and J. Kuligowski, Anal. Chem., 2018, 90, 9093–9100.
5. M. Hanko, Ľ. Švorc, A. Planková and P. Mikuš, Anal. Chim. Acta, 2019, 1062, 1–27.
6. A. Tsiasioti and P. D. Tzanavaras, Microchem. J., 2023, 193, 109157.
7. L. Gao, H. Wei, S. Dong and X. Yan, Adv. Mater., 2024, 36, 2305249.
8. A. Thamilselvan and M. I. Kim, Trends Anal. Chem., 2024, 177, 117815.
9. H. Li, M. Meng, J. Cao, X. Hu, H. Wang, M. Pan and S. Wang, Trends Food Sci. Technol., 2025, 161, 105062.
10. X. Yu, W. Zhuang, W. Bai, Q. Li, H. Tan, L. Zhang, J. Zhang, Y. Yao, S. Li, H. Bai, J. Hu, X. Sun and W. Hu, Coord. Chem. Rev., 2025, 542, 216850.
11. S. Li, Y. Zhao, F. Ji, R. Zheng, X. Ji, Z. Liu and L. Wang, Colloids Surf., A, 2022, 650, 129617.
12. S. Zhu, Q. Guo, Y. Zheng, J. Yuan, D. Zuo and B. Wang, Microchim. Acta, 2024, 191, 599.
13. H. Li, H. Wu, J. Chen, Y. Su, P. Lin, W. Xiao and D. Cao, Langmuir, 2022, 38, 15559–15569.
14. S. Li, Z. Ma, Y. Cui, S. Jiao, R. Li, H. Xiao and L. Wang, Colloids Surf., A, 2024, 700, 134739.
15. D. Yin, H. Yang, S. Wang, Z. Yang, Q. Liu, X. Zhang and X. Zhang, Colloids Surf., A, 2020, 607, 125466.
16. X. Zheng, Q. Lian, L. Zhou, Y. Jiang and J. Gao, Colloids Surf., A, 2020, 606, 125397.
17. Y. Yang, X. Tan, Y. Wang, B. Shen, Y. Yang and H. Huang, Chem. Eng. J., 2023, 468, 143703.
18. X. Huang, M. Jiang, H. Zeng, J. Wu, J. Wu, X. Liu, L. Zhou and Y. Yuan, Mater. Today Commun., 2023, 36, 106775.
19. J. Wu, L. Ji, M. Kang, S. Liu, J. Liu, Z. Dong and S. Xiao, Mater. Today Commun., 2025, 47, 113155.
20. W. N. Abd-AlGhafar, R. A. Shabana, M. M. Tolba and R. El-Shaheny, Talanta, 2025, 297, 128651.
21. Z. Han, N. Wang, Y. lv, P. Liu and X. Su, Sens. Actuators, B, 2025, 440, 137955.
22. S. Sun, Q. Guan, Y. Liu, B. Wei, Y. Yang and Z. Yu, Chin. Chem. Lett., 2019, 30, 1051–1054.
23. M. Wang, H. Zhu, B. Liu, P. Hu, J. Pan and X. Niu, ACS Appl. Mater. Interfaces, 2022, 14, 44762–44771.
24. Q. Xu, R. Su, Y. Chen, S. T. Sreenivasan, N. Li, X. Zheng, J. Zhu, H. Pan, W. Li, C. Xu, Z. Xia and L. Dai, ACS Appl. Nano Mater., 2018, 1, 1886–1893.
25. B. Shi, Y. Su, L. Zhang, M. Huang, R. Liu and S. Zhao, ACS Appl. Mater. Interfaces, 2016, 8, 10717–10725.
26. M. Zhu, S. Kim, L. Mao, M. Fujitsuka, J. Zhang, X. Wang and T. Majima, J. Am. Chem. Soc., 2017, 139, 13234–13242.
27. S. K. Tammina, D. Yang, S. Koppala, C. Cheng and Y. Yang, J. Photochem. Photobiol., B, 2019, 194, 61–70.
28. M. A. Mousa, H. H. Abdelrahman, M. A. Fahmy, D. G. Ebrahim and A. H. E. Moustafa, Sci. Rep., 2023, 13, 12863.
29. V. K. Singh, P. K. Yadav, S. Chandra, D. Bano, M. Talat and S. H. Hasan, J. Mater. Chem. B, 2018, 6, 5256–5268.
30. S. A. R. Pakkath, S. S. Chetty, P. Selvarasu, A. V. Murugan, Y. Kumar, L. Periyasamy, M. Santhakumar, S. R. Sadras and K. Santhakumar, ACS Biomater. Sci. Eng., 2018, 4, 2582–2596.
31. Y. Liu, T. Sun, G. Dong, S. Zhang, K. Chu, X. Pu, H. Li and X. Liu, Ceram. Int., 2019, 45, 17467–17475.
32. Q. Ke, Y. Jin, F. Ruan, M. N. Ha, D. Li, P. Cui, Y. Cao, H. Wang, T. Wang, V. N. Nguyen, X. Han, X. Wang and P. Cui, Green Chem., 2019, 21, 4313–4318.
33. L. Li, Q. Wang and Z. Chen, Microchim. Acta, 2019, 186, 1–5.
34. S. Tang, Q. Liu, J. Hu, W. Chen, F. An, H. Xu, H. Song and Y.-W. Wang, Front. Chem., 2021, 9, 812503.
35. Q. Zhong, Y. Chen, A. Su and Y. Wang, Sens. Actuators, B, 2018, 273, 1098–1102.
36. N. K. Dega, A. B. Ganganboina, H. L. Tran, E. P. Kuncoro and R.-A. Doong, Talanta, 2021, 237, 122957.
37. M. Razavi, A. Barras, M. Ifires, A. Swaidan, M. Khoshkam, S. Szunerits, M. Kompany-Zareh and R. Boukherroub, J. Colloid Interface Sci., 2022, 613, 384–395.
38. C. Yuan, X. Qin, Y. Xu, Q. Jing, R. Shi and Y. Wang, Microchem. J., 2020, 159, 105365.
39. J. Ge, R. Cai, X. Chen, Q. Wu, L. Zhang, Y. Jiang, C. Cui, S. Wan and W. Tan, Talanta, 2018, 195, 40–45.
40. X. Lai, Y. Shen, S. Gao, Y. Chen, Y. Cui, D. Ning, X. Ji, Z. Liu and L. Wang, Biosens. Bioelectron., 2022, 213, 114446.
41. H. Li, P. Song, T. Wu, H. Zhao, Q. Liu and X. Zhu, Appl. Clay Sci., 2022, 229, 106656.
42. H. Li, Z. Wang, J. Zhao, Y. Guan and Y. Liu, ACS Sustainable Chem. Eng., 2020, 8, 16136–16142.
43. K. Sun, C. Liu, Y. Cao, J. Zhu, J. Li and Q. Huang, Anal. Chim. Acta, 2024, 1304, 342552.
44. H. Yang, X.-S. Zhang, K. Cheng, B. Liu, W. Chen, J.-X. Fan and Y.-D. Zhao, Sens. Actuators, B, 2024, 410, 135652.
45. C.-c. Lin, W.-h. Liu, Z.-h. Wang and M.-c. Yin, Eur. J. Nutr., 2010, 50, 499–506.
46. J. Wróblewska, M. Wróblewski, I. Hołyńska-Iwan, M. Modrzejewska, J. Nuszkiewicz, W. Wróblewska and A. Woźniak, Antioxidants, 2023, 12, 1325.
47. J. Yang, Y. Xiong, L. Zhou, Y. Huang, W. Chen and B. Wang, J. Med. Virol., 2019, 92, 34–44.

---