An ascorbic acid sensor based on protein-modified Au nanoclusters

Abstract

A simple fluorescence sensor for sensitive turn-off detection of ascorbic acid (AA) was developed by using protein-modified Au nanoclusters in aqueous media. The sensing mechanism is originated from the oxidation state change of Au nanoclusters controlled by AA. Under optimal experimental conditions, a good linear relationship between the relative fluorescence quenching intensity and the concentration of AA can be obtained in the range of 1.5 to 10 μM, with a detection limit as low as 0.2 μM. The proposed method is simple, efficient and reliable for monitoring of AA in some biological samples.

---

1. Introduction

Designing novel sensors using nanomaterials with distinct physical and chemical properties continues to be an active research topic. Quantum dots (QDs),^1,2 nano-metal oxides^3,4 and noble metal nanocrystals^5,6 are novel materials with numerous unique possibilities. Recently, noble metal nanoclusters (NCs) have opened a promising field toward the development of new fluorescent probes. NCs typically consist of several to tens of atoms, with a size regime comparable to the Fermi wavelength of the conduction electrons. In the small size regime, NCs become molecular species, and discrete states with strong fluorescence can be observed.^7,8 These NCs are highly attractive for biolabeling and bioimaging applications because they contain nontoxic metal species and thus pose little negative effects toward biosystems.

The exploration of noble metal NCs, especially Au NCs, for sensing applications has spread from metal ions, anions to small bio-molecules. In view of the structure of Au NCs, the sensing mechanisms could be summarized as the following two standpoints: on one hand, since a small amount of Au(I) was localized on the surface of the Au(0) core,⁹ any partial change of the oxidation state of the Au NCs resulted from the interactions with analytes can lead to fluorescence quenching. For example, based on the interactions of Hg²⁺ with Au(I) and CN⁻ with Au(0), fluorescent sensors for Hg²⁺ (ref. 10) and CN⁻ (ref. 11) using bovine serum albumin (BSA) stabilized Au NCs have been reported; on the other hand, a layer of surface ligands was wrapped around the NC core to protect and stabilize the Au NCs. Destruction of the bridge between the Au core and surface ligands would also quench the fluorescence of Au NCs, as in the cases of Cu²⁺ detection with glutathione (GSH) stabilized Au NCs⁶ and hydrogen peroxide detection with BSA stabilized Au NCs.¹²

Ascorbic acid (AA) is known as a vital antioxidant and plays important roles in balancing the oxidative stress of human body, which exists in body fluids at relatively high concentrations. For example, the normal concentration ranges of AA in human blood plasma and urine are 0.6–1.5 mg per 100 mL (ref. 13) and 10–60 mg in 24 h urine,¹⁴ respectively. It also has antioxidative properties and plays an important role as a vitamin in a lot of biochemical processes. AA is a medication for scurvy, drug poisoning, liver disease, allergic reactions and atherosclerosis, and helps promote healthy cell development, calcium absorption and normal tissue growth. Moreover, it is used as an anti-oxidant for the manufacture of juices and soft drinks. Therefore, the detection of AA is of great importance in pharmaceutical, clinical and food industry.¹⁵ The major analytical method for AA determination is electrochemical approaches.¹⁶ However, the interference from other similar redox potential molecules can be serious, such as uric acid and dopamine. In this vein, fluorescent sensors provide obvious advantages of simplicity, convenience and rapid implementation. It has been reported that Au(I) can be reduced to Au(0) by AA.¹⁷ That is, in the presence of AA, the oxidation state of the Au NCs will be perturbed, leading to fluorescence quenching. Thus Au NCs can be engineered into a novel fluorescent probe for AA. The proposed method was preliminarily applied to the determination of AA in human blood plasma with satisfactory results.

2. Experimental

Chemicals and materials

Bovine serum albumin (BSA) and dopamine were bought from Sigma-Aldrich. HAuCl₄·3H₂O, L-ascorbic acid and other salts were purchased from Kelong Reagent Co., Chengdu, China. All other chemicals, such as sodium hydroxide, were of analytical grade and used without further purification. All solutions were prepared with water purified with a Milli-Q purification system (Millipore, USA).

Apparatus

Fluorescence measurements were performed on an F-7000 spectrofluorometer (Hitachi, Japan) equipped with a quartz cell (1 cm × 1 cm) in the fluorescence mode. The fluorescence spectra were recorded in a wavelength range of 550–750 nm upon excitation at 270 nm. The slit width was 5 nm and 10 nm for excitation and emission, respectively. The photomultiplier tube (PMT) voltage was set at 700 V. Time scan fluorescence measurements were carried out at an excitation wavelength of 270 nm and at an emission wavelength of 621 nm. Absorption spectra from 200–350 nm were recorded on a UV-2900 spectrophotometer (Hitachi, Japan). A PHS-3E pH Meter (Shanghai Feile Co., Ltd) was used for monitoring the pH. X-ray photoelectron spectroscopy (XPS) was performed on a THETA PROBE spectrometer (Thermo Fisher Scientific Inc.) with Al Kα X-ray radiation as the X-ray source for excitation. Narrow-scan XPS spectra of the Au 4f NCs were deconvoluted by the Avantage data acquisition and processing software, using adventitious carbon to calibrate the binding energy of C1s (284.5 eV).

Synthesis of BSA-Au NCs

Fluorescent BSA modified Au NCs were synthesized in aqueous solution following a previous publication with minor modifications.⁹ In a typical synthesis procedure, all glassware used in the experiments was cleaned in a bath of freshly prepared aqua regia (HCl : HNO₃, a volume ratio of 3 : 1), and rinsed thoroughly with water prior to use. 15 mL of aqueous HAuCl₄ solution (10 mM, 37 °C) was added to BSA solution (15 mL, 50 mg mL⁻¹, 37 °C) under magnetic stirring. Then, 1.5 mL of 1 M NaOH solution was introduced and the mixture was allowed to incubate at 37 °C under vigorous stirring for 24 h. The color of the solution changed from light yellow to deep brown. The solution was then dialyzed in water for 48 h to remove unreacted HAuCl₄ and NaOH, and the final solution was stored at 4 °C in a refrigerator.

Measurement procedure

The concentration of the Au NCs in this work was approximately evaluated based on the measured BSA concentration. Pure BSA exhibits an absorbance maximum at 278 nm and this peak almost remains unchanged after the conjugation with Au NCs. According to Beer's law (C = A/εL, where C is the concentration of BSA, A is the absorbance of the solution, ε is the molar absorptivity of BSA and L is the width of the sample cuvette), the concentration 3.0 × 10⁻⁵ M was measured by spectrophotometry using a molar absorptivity of 44 000 M⁻¹ cm⁻¹ at 278 nm.¹⁸

To evaluate the quenching effect of AA on the BSA-Au NCs, an amount of 100 μL BSA-Au NCs (1 μM) was mixed with 1400 μL of varied concentrations of AA (dispersed in Tris–HCl buffer, pH = 7.4) in a 1.5 mL Eppendorf (Ep) tube. Then, the mixed solution was mixed thoroughly on a vortex mixer. The fluorescence emission spectra of BSA-Au NCs in the presence of AA were recorded under the excitation at 270 nm.

To study the usefulness of this method for AA detection in biological samples, the following metal ions and biomolecules were used for the interference experiments: Na⁺, K⁺, Ca²⁺, Mg²⁺, glucose, citric acid, SO₃²⁻, uric acid and dopamine.

To detect AA in biological samples, fresh human blood plasma was collected from the Hospital of Sichuan Agricultural University and was diluted directly 10 times with buffer before analysis. Then, 100 μL of 1 μM Au NCs and 900 μL of diluted plasma solution were mixed first, and added a standard solution of AA to evaluate the recovery. In addition, AA was measured simultaneously using a colorimetric method,¹⁹ which is a classic method for AA determination in a clinical laboratory.

3. Results and discussion

Quenching of the fluorescence of BSA-Au NCs by AA

Fluorescent Au NCs were prepared based on the method described previously with BSA as the template.⁹ The characteristics of the fluorescence spectra were consistent with those reported in the literature. Fig. 1a′ shows the emission spectrum of Au NCs in the red region around 620 nm upon excitation at 270 nm. According to Ying et al.,⁹ Au NCs are formed via in situ reducing the entrapped Au ions by the activated BSA molecules, and the BSA modified Au NCs have a common magic cluster size with 25 gold atoms. The fluorescence emission was originated from intra-band transitions of free electrons of the Au NCs. Upon addition of 10 μM AA into the 1 μM Au NC solutions, the corresponding emission of the system decreased significantly (Fig. 1b′). These results suggested that the BSA-modified Au NCs could be used as a fluorescent turn-off probe for AA in aqueous solution.

Fig. 1 Absorbance and fluorescence emission spectra of 1 μM BSA-Au NCs in the absence (a and a′) and presence (b and b′) of 10 μM AA. All solutions were prepared in pure water.

Effect of pH on the fluorescence of Au NCs

In order to achieve sensitive detection of AA, the effect of the analyte solution pH on the fluorescence of BSA-Au NCs was investigated. The response was independently tested three times at each pH and an average value was calculated. The results revealed that the fluorescence intensity of Au NCs changed slightly when the pH varied in the range of 4–10, indicating that this pH range had little effect on the fluorescence intensity of the Au NCs. For potential application of this system in biological sample analysis, we chose pH 7.4 (Tris–HCl buffer, 0.1 M) for optimizing the AA detection.

Effect of incubation time

As shown in Fig. 2, the fluorescence of BSA-Au NCs can be quenched immediately by AA. Compared with common fluorophores, such as rhodamine 6G²⁰ and the green fluorescent protein,^21–23 the Au NCs possess high resistance to photobleaching, and the fluorescence intensity of Au NCs decreased only about 2% within 5 min of continuous excitation. The fluorescence was quenched and reached equilibrium immediately after introduction of AA, indicating very fast interaction between AA and Au NCs.

Fig. 2 Time scan of fluorescence of the (a) Au NCs and (b) in the presence of AA at an excitation wavelength of 270 nm and at an emission wavelength of 621 nm.

Interference evaluation for AA detection

To evaluate the potential interference with AA detection with BSA-Au NCs, the fluorescence response of AA (5 μM) was then investigated in the presence of competing ions or biomolecules. As shown in Table 1, in the presence of 5000-fold excess of Na⁺, 2000-fold excess of K⁺, 250-fold excess of Ca²⁺, 150-fold excess of Mg²⁺, 5000-fold excess of glucose, 80-fold excess of citric acid and 30-fold excess of SO₃²⁻, the signal perturbation on AA detection (5 μM) was generally less than ±5.0%. For uric acid and dopamine, BSA-Au NCs permitted interference-free detection of AA (5 μM) up to 100- and 20-fold excess, respectively. Therefore, it is possible to use Au NCs as a fluorescent probe for AA detection in some biological fluids.

Table 1 Effect of co-existing substances on the quenched fluorescence intensity of Au NCs by 5 μM AA

Substance	Concentration/μM	Change of quenched relative fluorescence intensity/%
Na^+	25 000	−4.2
K^+	10 000	−4.4
Ca^2+	1250	+2.8
Mg^2+	750	+3.5
Glucose	25 000	−4.0
Citric acid	400	−4.7
SO3^2−	150	−3.8
Uric acid	500	−5.0
Dopamine	100	−3.4

---

Analytical figures of merit

Under optimal experimental conditions, the fluorescence of the Au NCs was highly sensitive to AA and gradually decreased as the concentration of AA increased (Fig. 3). The relative fluorescence intensity (F₀/F) versus the AA concentration was also plotted in Fig. 3, which showed a good linear correlation:

F₀/F = 0.74[AA] (μM) + 0.812,

where F₀ and F are the fluorescence intensity of the Au NCs in the absence and presence of AA, respectively. The linear correlation coefficient was 0.997. The limit of detection (LOD) is defined as LOD = (3σ/k), where σ is the standard deviation of blank measurements (n = 11) and k is the slope of the calibration curve. Here LOD was found to be 0.2 μM. As for the reproducibility in terms of relative standard deviation (RSD), it was estimated to be 1.9% (3 μM AA, and n = 11).

Fig. 3 Fluorescence quenching of the Au NCs in the presence of AA at various concentrations, and quenching efficiency plotted of the relative fluorescence intensity of Au NCs versus the concentration of AA. All solutions were prepared in 0.1 M Tris–HCl buffer at pH = 7.4. Here F₀ and F are the fluorescence intensities of the Au NCs in the absence and presence of AA, respectively. The error bars represent ±one standard deviation of three parallel measurements.

Mechanism for fluorescence quenching

X-ray photoelectron spectroscopy (XPS) was employed to investigate the oxidation state change of the Au NCs before and after interaction with AA. As shown in Fig. 4a, the Au 4f_7/2 spectrum could be deconvoluted into two distinct components centered at binding energies of 83.5 eV and 85.1 eV, which could be identified as Au(0) and Au(I), respectively. Upon interaction with AA, the peak area for Au(I) was decreased greatly (Fig. 4b). It is estimated that the content of Au(I) decreased from about 17.9% to 3% after interaction with AA (10 μM). AA can be oxidized by Au(I) (Fig. 4c),¹⁷ leading to the generation of Au(0) and thus fluorescence quenching of BSA-Au NCs (Fig. 4d).

Fig. 4 XPS spectra of Au 4f (a and b) and reaction mechanism of AA with Au(I) (c and d): (a) as-prepared 1 μM Au NCs and (b) 1 μM Au NCs + 10 μM AA mixed solution; (c) oxidation of AA by Au(I) and (d) schematic illustration of the proposed fluorescent sensor for AA detection with BSA-Au NCs.

Preliminary analytical application

The developed AA fluorescent sensor based on BSA-Au NCs was preliminarily applied to the detection of AA in fresh human blood plasma. Moreover, AA was detected using a colorimetric method to validate the sensor's accuracy. The analytical results are summarized in Table 2. The analytical results by the proposed method were in good agreement with those by the colorimetric method, and the recoveries of added AA in the diluted human blood plasma ranged from 94% to 96%. These results demonstrate the potential applicability of the BSA-Au NC based fluorescent-off probe for the quantitative detection of AA in the biological samples.

Table 2 Analytical results (mean ± s; n = 3) of AA in human blood plasma

Sample	AA found in original samples (μM)		Added AA in diluted samples^a (μM)	Recovery (%)
	Proposed method	Colorimetric method		Proposed method
a The plasma was diluted 10-fold with water before analysis.
#-1	41.4 ± 1.8	40.5 ± 0.6	2.0	96 ± 4
#-2	44.0 ± 2.2	43.2 ± 0.3	2.0	99 ± 5
#-3	43.2 ± 2.0	44.2 ± 0.4	2.0	94 ± 3

---

4. Conclusions

In conclusion, we proposed a new and simple fluorescent turn-off sensor to detect ascorbic acid with very high sensitivity using fluorescent BSA-Au NCs in aqueous media. The sensing mechanism was based on the Au(I) oxidation state change by the control of AA. This analytical procedure is notable as it can work directly in aqueous solutions and does not require any toxic noble metals or organic reagents as solvents, and it has potential for detection of AA in blood samples.

Acknowledgements

This work was partially supported by National Natural Science Foundation of China [no. 20835003].

References

1. Y. F. Chen and Z. Rosenzweig, Anal. Chem., 2002, 74, 5132–5138.
2. Z. Sheng, H. Han, D. Hu, J. Liang, Q. He, M. Jin, R. Zhou and H. Chen, Chem. Commun., 2009, 2559–2561.
3. Y. F. Zhu, J. J. Shi, Z. Y. Zhang, C. Zhang and X. R. Zhang, Anal. Chem., 2002, 74, 120–124.
4. J. J. Shi, Y. F. Zhu, X. R. Zhang, W. R. G. Baeyens and A. M. Garcia-Campana, TrAC, Trends Anal. Chem., 2004, 23, 351–360.
5. W. Chen, Z. S. Lu and C. M. Li, Anal. Chem., 2008, 80, 8485–8492.
6. W. B. Chen, X. J. Tu and X. Q. Guo, Chem. Commun., 2009, 1736–1738.
7. J. Zheng, P. R. Nicovich and R. M. Dickson, Annu. Rev. Phys. Chem., 2007, 58, 409–431.
8. Z. Jie, Z. Caiwei and R. M. Dickson, Phys. Rev. Lett., 2004, 93, 077402.
9. J. P. Xie, Y. G. Zheng and J. Y. Ying, J. Am. Chem. Soc., 2009, 131, 888–889.
10. J. P. Xie, Y. G. Zheng and J. Y. Ying, Chem. Commun., 2010, 46, 961–963.
11. Y. Liu, K. Ai, X. Cheng, L. Huo and L. Lu, Adv. Funct. Mater., 2010, 20, 951–956.
12. L. Jin, L. Shang, S. Guo, Y. Fang, D. Wen, L. Wang, J. Yin and S. Dong, Biosens. Bioelectron., 2011, 26, 1965–1969.
13. A. Cantarow and B. Schepartz, Biochemistry, Saunders, Philadephia, 2nd edn. 1957, pp. 744–778.
14. G. A. Harrison, Chemical Methods in Clinical Medicine, Churchill, London, 3rd edn. 1947, p. 292.
15. S. P. Sood, L. E. Sartori, D. P. Wittmer and W. G. Haney, Anal. Chem., 1976, 48, 796–798.
16. Z. H. Wang, J. Liu, Q. L. Liang, Y. M. Wang and G. Luo, Analyst, 2002, 127, 653–658.
17. R. W. J. Scott, O. M. Wilson, S.-K. Oh, E. A. Kenik and R. M. Crooks, J. Am. Chem. Soc., 2004, 126, 15583–15591.
18. L. Shang, Y. Wang, J. Jiang and S. Dong, Langmuir, 2007, 23, 2714–2721.
19. A. Kyaw, Clin. Chim. Acta, 1978, 86, 153–157.
20. M. Unger, E. Kartalov, C. S. Chiu, H. A. Lester and S. R. Quake, Biotechniques, 1999, 27, 1008–1014.
21. R. Heim, A. B. Cubitt and R. Y. Tsien, Nature, 1995, 373, 663–664.
22. Q. S. Hanley, V. Subramaniam, D. J. Arndt-Jovin and T. M. Jovin, Cytometry, 2001, 43, 248–260.
23. R. Y. Tsien, Annu. Rev. Biochem., 1998, 67, 509–544.

---