Guanine chemiluminescent biosensor capable of rapidly sensing mercury in a sample

Abstract

Using DNA aptamer and guanine chemiluminescene detection, we developed a highly sensitive biosensor for the rapid quantification and monitoring of Hg²⁺ in drinking water. The random coil structure of DNA aptamer was transformed to a hairpin-link thymine (T)–Hg²⁺–T complex, closed by the hybridization of guanine and cytosine of DNA aptamer, in the presence of Hg²⁺ in drinking water. The brightness of light, emitted from the reaction between guanine of the DNA aptamer and 3,4,5-trimethoxyl phenylglyoxal hydrate (TMPG) in the presence of tetra-n-propylammonium hydroxide (TPA) and O₂, exponentially decreased with the increase of Hg²⁺ due to the transformation of DNA aptamer. The sensitivity of the biosensor was dependent on the incubation time for forming the hairpin-link T–Hg²⁺–T complex closed by the hybridization of guanine and cytosine of the DNA aptamer. The sensitivity of the biosensor was the highest when DNA aptamer was incubated with Hg²⁺ for 20 min at room temperature (21 ± 2 °C). The sensitivity of the biosensor generated with incubation longer than 20 min was not as good as that operated with a 20 min incubation because the dehybridization of guanine and cytosine of DNA aptamer is predominant after 20 min. The limit of detection (LOD = CL₀ − 3σ) of the biosensor operated with a 20 minute incubation was as low as 2.11 nM. Also, the accuracy, precision, and recovery of the time-dependent biosensor were good within a statistically acceptable error range.

---

Introduction

Due to the virulent toxicity of Hg²⁺, a number of analytical methods using various optical detections (e.g., absorbance,^1,2 fluorescence,^3,4 chemiluminescence^5,6) have been developed for public health and environment.

It is well-known that the random coil structure of single strand DNA added in solution containing Hg²⁺ is transformed to hairpin-link structure with the interaction between thymine (T) and Hg²⁺ to form a T–Hg²⁺–T complex.^7–9 Thus, various types of DNA aptamers^7,8,10 designed using the chemical and physical interaction between T and Hg²⁺ were applied for the quantification of Hg²⁺ in human and environmental samples.

3,4,5-Trimethoxylphenyl-glyoxal (TMPG) reacts with guanine to emit dim light.^11,12 Light emitted from the reaction of TMPG and guanine in the presence of tetra-n-propylammonium hydroxide (TPA) and O₂, was enhanced.^13,14 Recently, light emitted from the reaction between guanines of DNA aptamer and TMPG was applied as a new chemiluminescence detection method of biosensor capable of rapidly quantifying analytes in a sample.^12–16 As shown in Scheme 1, dim light emitted from the reaction of TMPG and guanine of DNA aptamer can transfer energy to fluorescein (or 6-FAM) labelled with DNA aptamer to emit bright green light based on the principle of intra chemiluminescence resonance energy transfer (Intra-CRET).^13–15 The CRET in guanine chemiluminescence is similar to that in peroxyoxalate chemiluminescence reaction.^17–19 Using Intra-CRET between fluorescent dye (e.g., fluorescein, 6-FAM) and high-energy intermediate formed from the reaction between TMPG and guanine of DNA aptamer in the presence of TPA, biomarkers in human samples were quantified for the early diagnosis of prostate cancer¹⁴ and the monitoring of blood coagulation.¹³ Guanines of DNA aptamer not bound with a specific biomarker rapidly reacted with TMPG to form high-energy intermediates, whereas guanines of DNA aptamer-bound the biomarker did not react with TMPG. Thus, relative CL intensity of guanine chemiluminescence was proportionally decreased with the increase of biomarkers in a sample.^13,14

Scheme 1 Guanine chemiluminescence. (1) TMPG, (2) DNA aptamer conjugated with fluorescein, X: high-energy intermediate.

Using the complexation of T and Hg²⁺ as well as the rapid chemical reaction of guanine and TMPG in the presence of TPA, we designed appropriate DNA aptamers to develop a highly sensitive biosensor capable of rapidly quantifying Hg²⁺ in a sample.

Experimental

Chemical and materials

Three different types of T–Hg²⁺–T hairpin DNA sequences (e.g., A1, A2, A3) capable of forming a T–Hg²⁺–T complex were designed based on the previous research results reported by other research groups.^7,8,10 These T–Hg²⁺–T hairpin DNA sequences were denoted as DNA aptamer in this report even though they had not been discovered by systematic evolution of ligands by exponential enrichment (SELEX).²⁰ We purchased them from Alpha DNA.

A1: 5′-fluorescein-GGGGTTCTTCCCCTTGTTCCCC-3′

A2: 5′-fluorescein-GCGCTTCTTCCCCTTGTTGCGC-3′

A3: 5′-fluorescein-GGTTCTTCCCCTTGTTCC-3′

In order to study whether guanine–cytosine base pair formed from the hybridization of target (H1) and complementary (H2) DNAs emits light in guanine chemiluminescence reaction, H1 and H2 were purchased from Alpha DNA.

H1: 5′-fluorescein-GGGGAAAA-3′

H2: 5′-TTTTCCCC-3′

3,4,5-trimethoxyl phenylglyoxal hydrate (TMPG, 97%) was purchased from Matrix Scientific (Columbia, SC, USA). FeCl₂ (99%), FeCl₃ (99%), tetra-n-propylammonium hydroxide (TPA, 40% w/w aqueous solution), and deionized H₂O (HPLC grade) were purchased from Alfa Aesar (Ward Hill, MA, USA). N,N-Dimethylformamide (DMF) and 10× PBS were purchased from EMD (Billerica, MA, USA). Graphene oxide (GO) was purchased from Graphene Supermarket (Calverton, NY, USA). 7.0 mM Hg(NO₃)₂ stock solution was purchased from VWR.

Free single-strand DNA aptamer removal using Fe₃O₄–graphene nanoparticle

In order to remove free single-strand DNA remaining after the binding of Hg²⁺ and DNA aptamer, Fe₃O₄–graphene nanoparticles were synthesized based on our previous reports.^13,14,21 The mixture (0.5 ml) of FeCl₂ (2.5 mg) and FeCl₃ (7.5 mg) in water was added to a 1.5 ml-centrifuge tube containing graphene oxide (GO, 0.5 ml, 1 mg ml⁻¹). Ammonium hydroxide (NH₄OH, 20 μl, 30%) was also dispensed into the tube. The microcentrifuge tube was then incubated for 60 minutes at 85 °C. Fe₃O₄–graphene nanoparticles washed using a magnetic separator (Bioclone, Inc) were stored in a refrigerator. Free single-strand DNA aptamers were rapidly immobilized on the surface of Fe₃O₄–graphene nanoparticle due to the π–π static interaction between single strand DNA and graphene oxide, whereas the complex of DNA aptamer-bound Hg²⁺ remained in aqueous solution.

Determination of incubation time for the quantification of Hg²⁺

Hg²⁺ (40 nM, 100 μl) in water was mixed with mercury aptamer (1 μM, 100 μl) in PBS (pH 7.4, 137 mM NaCl, 2.7 mM KCl, 10 mM phosphate buffer) at room temperature (21 ± 2 °C). We selected PBS as buffer solution based on the previous reports to enhance the sensitivity of guanine chemiluminescence detection.¹⁵ Light emitted from the mixture with the addition of guanine chemiluminescence reagents (e.g., TMPG, TPA) was immediately measured with a luminometer (Lumat 9507, Beththold, Inc, Germany). Also, the remaining mixture in the absence of guanine chemiluminescence reagents was incubated at room temperature. Then, light emitted from the mixture incubated at 10 min intervals for up to 60 min was measured. With relative CL intensities measured after the different incubations, we determined the best incubation time to quantify trace levels of Hg²⁺ in a sample.

Quantification of Hg²⁺

As shown in Fig. 1, Hg²⁺ in a sample was rapidly quantified with the simple procedure without time consuming washing of the mixture before adding guanine chemiluminescence reagents. A certain concentration of Hg²⁺ (100 μl) in water was mixed with A1 (0.6 μM, 100 μl) in PBS buffer (pH 7.4). The mixture was incubated for 20 min at room temperature. After the incubation, the mixture (20 μl) was mixed with 20 mM TPA (10 μl) in a borosilicate test tube. Then the test tube was inserted into the luminometer. Then, light emitted in the test tube with the addition of 4 mM TMPG (100 μl) in DMF through a syringe pump of the luminometer was measured for 20 seconds. Then, Hg²⁺ concentration in the sample was determined with the linear calibration curve obtained with 6 different standards (0–400 nM).

Fig. 1 Quantification of Hg²⁺ using DNA biosensor with guanine chemiluminescence detection.

Results and discussion

Quenching effect of Hg²⁺ in guanine chemiluminescence reaction

Before the development of with guanine chemiluminescence detection capable of quantifying Hg²⁺, we studied whether Hg²⁺ is a quencher in guanine chemiluminescence reaction using single strand DNA (H1), a guanine-rich and thymine-free oligo. Fig. 2 shows that light emitted in guanine chemiluminescence reaction in the presence of 1800 nM Hg²⁺ is the same as that in the absence of Hg²⁺ within the statistically acceptable error range. However, the relative CL intensity in the presence of Hg²⁺ higher than 1800 nM Hg²⁺ was lower than that in the absence of Hg²⁺. The results indicate that Hg²⁺ higher than 1800 nM acts as a quencher of guanine chemiluminescence generated in the reaction condition of Fig. 2.

Fig. 2 Quenching effect of Hg²⁺ in guanine chemiluminescence reaction. Condition: [H1] = 1 μM in PBS (pH 7.4), [TPA] = 20 mM in H₂O, [TMPG] = 5 mM in DMF.

Determination of DNA aptamer to quantify Hg²⁺ in a sample

As shown in Fig. 3, relative CL intensities in the absence of Hg²⁺ using A1 and A2 were higher than that in the presence of only 40 nM Hg²⁺ even though the concentration of Hg²⁺ is out of the concentration range of Hg²⁺ capable of acting as a quencher in guanine chemiluminescence (see Fig. 2). Thus, these results indicate that single strand DNA bound with Hg²⁺ does not emit light or emits relatively dim light when guanine chemiluminescent reagents are added in the solution as shown in Fig. 1. Also, Fig. 3 shows that the binding rate between A1 and Hg²⁺ is faster than that between A2 and Hg²⁺. Based on these results, we select A1 to develop a highly sensitive biosensor capable of rapidly quantifying Hg²⁺ in a sample.

Fig. 3 CL emission of DNA aptamer (1 μM in PBS) in the absence and presence of Hg²⁺ (40 nM in deionized water). The mixture containing DNA aptamer and Hg²⁺ was incubated for 20 min at room temperature.

Determination of incubation time for the quantification of Hg²⁺

In order to understand the results shown in Fig. 3, first, free DNA aptamers remaining after the reaction to capture Hg²⁺ for 20 min were removed using Fe₃O₄–graphene nanoparticles we synthesized^13,14,21 and a magnetic bar based on π–π static interaction between single strand DNA and graphene oxide (see Fig. 4(a)). Then, we studied whether DNA aptamer-bound Hg²⁺ can emit light with the addition of guanine chemiluminescent reagents. As shown in Fig. 4(b), the solution containing DNA aptamer-bound Hg²⁺ does not emit light, whereas the other solution containing the mixture of free DNA aptamer and DNA aptamer-bound Hg²⁺ emits bright green light. Based on the results, plausible reaction mechanisms shown in Fig. 4(c) can be proposed. Fluorescein excited from high-energy intermediate formed from the reaction of guanine of free DNA aptamer and TMPG emit bright green light based on the principle of intra-CRET. However, guanines of DNA aptamer-bound Hg²⁺ do not react with TMPG as shown in Scheme 2. Because primary and second amines of guanine are already exhausted due to the hybridization of guanine and cytosine of DNA aptamer to stably maintain T–Hg²⁺–T complex. Thus, relative CL intensity measured after the hybridization was lower than that recorded before the hybridization as shown in Fig. S1.† Fig. 4(d) shows that the brightness of light emitted in guanine chemiluminescence reaction is dependent on the concentration of Hg²⁺.

Fig. 4 (a) No emission of DNA aptamer-bound Hg²⁺, (b) CL emissions of mixture (free DNA aptamer and DNA aptamer-bound Hg²⁺) and DNA-aptamer-bound Hg²⁺ only, (c) differences between free DNA aptamer and DNA aptamer-bound Hg²⁺ in guanine chemiluminescence, (d) CL emission in the absence and presence of Hg²⁺ in guanine chemiluminescence. [Hg²⁺] from left: 0, 5, 20, 100, 400 nM, [DNA aptamer] = 1 μM.

Scheme 2 No reaction between guanine–cytosine base pair and TMPG.

Determination of incubation time for the quantification of Hg²⁺

Fig. 5 shows that the sensitivity of biosensor capable of quantifying Hg²⁺ in a sample is dependent on the incubation time of DNA aptamer and Hg²⁺ before adding guanine chemiluminescent reagents. With the increase of incubation time to up to 20 min, relative CL intensity was decreased because of the increase of DNA aptamer-bound Hg²⁺ concentration. With longer incubation time than 20 min, however, relative CL intensity began to be enhanced. Relative CL intensity measured after a 60 min incubation was similar to that immediately measured when DNA aptamer is mixed with Hg²⁺. Fig. 5 indicates that hairpin-link T–Hg²⁺–T complex closed with the hybridization of guanine and cytosine (see Fig. 4(c)) is decomposed with the dehybridization of guanine and cytosine under this condition. Thus, it is possible that hairpin-link T–Hg²⁺–T complex closed with the hybridization of guanine and cytosine is predominantly formed until 20 min of incubation. After 20 min, however, the decomposition of hairpin-link T–Hg²⁺–T complex is dominant. Thus, we selected the 20 min incubation to develop a more sensitive biosensor capable of quantifying Hg²⁺ in a sample. Fig. S2† shows that the stability of hairpin-link T–Hg²⁺–T complex is dependent on the hybridization of T–Hg²⁺–T hairpin-DNA. The results indicate that the hairpin-link T–Hg²⁺–T complex formed with A1 is more stable than those generated with A2 and A3. Additionally, it is expected that the biosensor using A1 will be more sensitive than those using A2 and A3.

Fig. 5 Time-dependent biosensor with guanine chemiluminescence detection for the quantification of Hg²⁺.

Determination of DNA aptamer concentration

Fig. 6 shows that the sensitivity of biosensor is dependent on the concentration of A1. The best concentration of A1 for developing a more sensitive biosensor was 0.6 μM under this condition. This is because the ratio (CL₀/CL₇₀₀ = 12.7) of CL intensities measured in the absence (CL₀) and presence (CL₇₀₀) of 700 nM Hg²⁺ in 0.6 μM A1 solution are the highest. This is because (1) 0.2 μM A1 is too low to capture 700 nM Hg²⁺ within 20 min and (2) the concentration of free A1 remaining after the 20 min incubation using 1 μM A1 is higher than that using 0.6 μM A1.

Fig. 6 Effect of DNA aptamer for the quantification of Hg²⁺ in a sample. Incubation time was 20 min at room temperature.

Fig. 7 Selectivity of DNA aptamer capable of binding Hg²⁺ in a sample. Condition: [Hg²⁺] = 20 nM, [DNA aptamer] = 0.6 μM in PBS buffer (pH 7.4). The concentration of each metal spiked in tap water was 100 nM.

Selectivity of DNA aptamer

Fig. 7 shows that biosensor using DNA aptamer (e.g., A1) can selectively quantify Hg²⁺ existing in a sample containing various contaminants. Also, relative CL intensity of tap water spiked with Hg²⁺ (20 nM) was the same as that in deionized water containing Hg²⁺ (20 nM). These results indicate that Hg²⁺ in tap water containing more impurities can be rapidly quantified without any pre-treatment to remove interferences. Additionally, the results shown in Fig. S3† is a proof that the biosensor can selectively trace levels of Hg²⁺ in the presence of excess impurities.

Quantification of Hg²⁺ in tap water

As shown in Fig. 8(a), relative CL intensity was exponentially decreased with the increase of Hg²⁺ ions. Also, we obtained a wide linear calibration curve (0–400 nM) using inverse numbers of relative CL intensities measured in the absence and presence of Hg²⁺ (see Fig. 8(b)). The limit of detection (LOD = CL₀ − 3σ) of biosensor calculated using the linear calibration curve was as low as 2.11 nM (0.4 ppb). σ is the standard deviation of CL₀. Thus, we expect that the biosensor can be applied as a new tool for the quantification and monitoring of Hg²⁺ in drinking water based on the regulation of US EPA (i.e., maximum contaminant level of Hg²⁺ in drinking water is 2 ppb. http://water.epa.gov/drink/contaminants/basicinformation/mercury.cfm#four).

Fig. 8 Calibration curves for the quantification of Hg²⁺ in tap water. Condition: [Hg²⁺] = 0, 10, 20, 40, 200, and 400 nM in deionized water, [DNA aptamer] = 0.6 μM in PBS. The mixture of Hg²⁺ and DNA aptamer was incubated for 20 min at room temperature.

Table 1 shows that the accuracy, precision, and recovery of biosensor developed in this research are good within the statistically acceptable error range.

Table 1 Accuracy, precision, and recovery of biosensor capable of sensing Hg²⁺ in drinking water (n = 5)

Sample 1 (nM)	Sample 2 (nM)	Expected^a (nM)	Measured (nM)	Recovery (%)
a Expected = (sample 1 + sample 2)/2.
20	10	15	14.2 ± 0.3	94.7
40	50	45	46.6 ± 1.6	102.8
100	200	150	145.8 ± 2.7	97.2

---

Conclusions

We developed a time-dependent biosensor capable of rapidly quantifying trace levels of Hg²⁺ in drinking water using DNA aptamer and guanine chemiluminescence detection. The random coli structure of DNA aptamer in the presence of Hg²⁺ was transformed to a hairpin-link structure with the formation of T–Hg²⁺–T complex. Also, guanines of DNA aptamer hybridized with cytosine to form relatively stable hairpin hairpin-link structure. Thus, the brightness of light generated from the reaction between TMPG and guanine of DNA aptamer not bound with Hg²⁺ was exponentially decreased with the increase of Hg²⁺ concentration. The concentration of T–Hg²⁺–T complex was dependent on the incubation time of DNA aptamer and Hg²⁺ in PBS. The formation of T–Hg²⁺–T complex was predominant until 20 min of incubation. Then, T–Hg²⁺–T complex was decomposed with the dominant dehybridization of guanine and cytosine after a 20 min incubation. Thus, the sensitivity of biosensor operated with 20 min incubation was the best. In conclusion, we confirmed that the biosensor can be applied as a monitoring system capable of rapidly quantifying Hg²⁺ in drinking water based on the regulation of US EPA.

Acknowledgements

This research was performed based on the intern program (LST-2014-5) of Luminescent MD, LLC.

References

1. J. L. Duan, H. Z. Yin, R. R. Wei and W. W. Wang, Biosens. Bioelectron., 2014, 57, 139–142.
2. Y. Wang, F. Yang and X. R. Yang, Biosens. Bioelectron., 2010, 25, 1994–1998.
3. J. M. Fang, P. F. Gao, X. L. Hu and Y. F. Li, RSC Adv., 2014, 4, 37349–37352.
4. D. D. Tan, Y. He, X. J. Xing, Y. Zhao, H. W. Tang and D. W. Pang, Talanta, 2013, 113, 26–30.
5. R. Freeman, X. Q. Liu and I. Willner, J. Am. Chem. Soc., 2011, 133, 11597–11604.
6. W. H. Gao, A. Zhang, Y. S. Chen, Z. X. Chen, Y. W. Chen, F. S. Lu and Z. G. Chen, Biosens. Bioelectron., 2013, 49, 139–145.
7. L. N. Zhang, Y. H. Wang, C. Ma, P. P. Wang and M. Yan, RSC Adv., 2015, 5, 24479–24485.
8. M. Li, X. J. Zhou, W. Q. Ding, S. W. Guo and N. Q. Wu, Biosens. Bioelectron., 2013, 41, 889–893.
9. D. Li, S. P. Song and C. H. Fan, Acc. Chem. Res., 2010, 43, 631–641.
10. Q. Li, M. Michaelis, G. Wei and L. C. Ciacchi, Analyst, 2015, 140, 5243–5250.
11. M. Kai, S. Kishida and K. Sakai, Anal. Chim. Acta, 1999, 381, 155–163.
12. X. Wang, C. Lau, M. Kai and J. Z. Lu, Analyst, 2013, 138, 2691–2697.
13. S. Cho, L. Park, R. Chong, Y. T. Kim and J. H. Lee, Biosens. Bioelectron., 2014, 52, 310–316.
14. T. Cha, S. Cho, Y. T. Kim and J. H. Lee, Biosens. Bioelectron., 2014, 62, 31–37.
15. J. Khang, D. Kim, K. W. Chung and J. H. Lee, Talanta, 2016, 147, 177–183.
16. S. Cai, Z. J. Cao, C. W. Lau and J. Z. Lu, Analyst, 2015, 139, 6022–6027.
17. P. S. Park, T. H. D. Rho, Y. T. Kim, S. O. Ko, M. A. Schlautman, E. R. Carraway and J. H. Lee, Chem. Commun., 2011, 47, 5542–5544.
18. W. Choi, J. Choi and J. H. Lee, RSC Adv., 2013, 3, 22455–22460.
19. H. Cho, S. Lee and J. H. Lee, Analyst, 2012, 137, 5368–5373.
20. L. C. Bock, L. C. Griffin, J. A. Latham, E. H. Vermass and J. J. Toole, Nature, 1992, 355, 564–566.
21. H. K. Choi and J. H. Lee, Anal. Methods, 2013, 5, 6964–6968.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra17407d

‡ M. Kwon and Y. Park contributed equally in this research.

---