A split G-quadruplex DNAzyme based magnetic graphene oxide platform for sensitive authentication of Pseudostellaria heterophylla

Abstract

Based on the different affinities of graphene oxide (GO) toward ssDNA and dsDNA, a fluorescence assay using split G-rich probes and magnetic GO (Fe₃O₄/GO) was developed for authentication of Pseudostellaria heterophylla based on the ITS sequences. In this assay, the probes mixed with Fe₃O₄/GO nanoparticles, hybridize with T-DNA for a double-stranded structure and then their G-rich sequences can fold into a G-quadruplex which combines hemin for DNAzyme. After magnetic separation, the DNAzyme supernatant can catalyze 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) oxidation with H₂O₂ and release a fluorescence signal. Herein, the Fe₃O₄/GO nanoparticles play an important role in decreasing the background signal by absorbing the excess probes and hemin. This assay allows the analysis of T-DNA in the range of 1.0 × 10⁻¹⁴ to 1.0 × 10⁻⁷ mol L⁻¹ with a low detection limit of 1.87 × 10⁻¹⁵ mol L⁻¹. In addition, this newly designed assay allows specific authentication of Pseudostellaria heterophylla from different provenances.

---

1. Introduction

Pseudostellaria heterophylla (Miq.) Pax (PH), a common traditional Chinese medicine (TCM), is widely used to treat fatigue, palpitations, night sweats, and lung and spleen diseases in children.^1,2 However, the quality of TCM is always influenced by environmental factors, agriculture, processing technique and so on.³ Nuclear ribosomal DNA internal transcribed spacer (nrDNA ITS), possessing highly conserved and variable regions, has been commonly used as a genetic marker for the identification of TCM.^4,5 For example, Yu et al. have employed PCR-based DNA sequencing technique with ITS as a biomarker for identifying PH from different geographical regions.⁶

In our previous study, we have also utilized ITS as genetic marker and applied the remarkable difference in affinity of graphene oxide (GO) toward ssDNA and dsDNA, and then established a split G-quadruplex and GO based fluorescent platform for authentication of PH.⁷ The two-dimentional GO surface can grasp the unemployed quinaldine red and provide the superiority of a low background. However, the sensitivity is unsatisfied, which can be solved via using the G-quadruplex DNAzyme as the catalytic biolabel. And G-quadruplex is a special DNA motif with a four-stranded structure constructed by guanine-rich sequences, and combine hemin to form a DNAzyme.^8–12 However, the GO nanomaterial is also reported to possess an intrinsic peroxidase-like activity,¹³ and it will provide the sensor with certain background noises. As a result, it is necessary to introduce a novel material to break these limitations.

Recently, magnetite (Fe₃O₄)/GO nanosheet has been widely applied to drug delivery, biomedical engineering and biomolecules detection.^14–17 As reported, the magnetic composites could provide sufficient surface and stability for the conjugation of biomolecules and simultaneously could be easily collected in the presence of an external magnetic field.^18–20 In this study, we have also designed two 2 : 2 split G-quadruplex probes, and employed the magnetic GO to construct an amplified and low-background senor for identifying PH from different provenances. Compared with conventional DNAzyme-based sensors, this novel self-assembled sensor provides multiple advantages: magnetic GO as an absorption platform of unemployed probes and hemin for the sake of low background signal; the separation also avoids the interference of the Fe₃O₄/GO's intrinsic peroxidase catalytic activity.

2. Experimental

2.1 Reagents

All DNA sequences were synthesized and purified by Sangon Biotechnology Co. Ltd. (Shanghai, China). The DNA stock solutions (5.0 × 10⁻⁴ mol L⁻¹) were dissolved in a 25 mM Tris–HCl buffer (pH 7.0) and stored at 4 °C until use. Hemin was purchased from Aladdin Reagents (Shanghai, China), and its stock solution (2.0 × 10⁻³ mol L⁻¹) was prepared in DMSO and stored at −20 °C. 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) was procured from Sigma Aldrich (St. Louis, MO, USA) and a stock solution (2.0 × 10⁻³ mol L⁻¹) was also prepared in DMSO. H₂O₂ was freshly prepared in a working buffer (50 mM Tris containing 150 mM NH₄Ac, 20 mM KCl, pH 7.5). The Fe₃O₄/GO nanomaterial was given by Guonan Chen's group²¹ in Fuzhou University, and its dispersion (1000 mg L⁻¹) was obtained by ultrasonic processing at the power of 150 W for 4 h. The dispersion was stored at 4 °C until use.

2.2 Sequences of oligomers

In this paper, Zheshen planted in Zherong County of Fujian Province is taken as the analyte. According to the sequencing result by Yu's group,⁶ a section of ITS sequences specific for Zheshen is chosen and labeled as T-DNA. When it comes to the design of probe, we have divided the unabridged G-rich PS2.M sequences with 2 : 2 split mode and obtained a couple of probes (probe a and b). In Scheme 1, the rose-red section of probe a and the purple section of probe b are sequences specifically complementary to T-DNA. And the green parts of both probe a and b are two GGG repeats. All the oligomers are listed in Table S1.†

Scheme 1 Principle of the split G-quadruplex DNAzyme based Fe₃O₄@GO platform for authentication of Pseudostellaria heterophylla.

2.3 Assay procedure

In a typical assay, probe a and b (1.0 × 10⁻⁸ mol L⁻¹) were initially mixed well with Fe₃O₄/GO dispersion and incubated for 60 min at room temperature. Furthermore, different concentrations of T-DNA and 100 μL working buffer (50 mM Tris containing 150 mM NH₄Ac and 20 mM KCl, pH 7.5) were added to the above solution. After 60 min, hemin (5.0 × 10⁻⁸ mol L⁻¹) and 150 μL working buffer were added simultaneously, and the mixture was allowed to incubate for another 60 min.

After magnetic separation for 60 min, 200 μL supernatant was transferred to another eppendorf tube. At last, H₂DCFDA (5.0 × 10⁻⁷ mol L⁻¹) and H₂O₂ (2.0 × 10⁻⁴ mol L⁻¹) were added to the supernatant solution, and the fluorescence signal was collected by Cary Eclipse Fluorescence Spectrophotometer. The instrument parameters were set as follows: λ_ex = 504 nm (slit 5 nm), λ_em = 510–600 nm (slit 5 nm).

3. Results and discussion

3.1 Principle of the G-quadruplex DNAzyme-based Fe₃O₄/GO sensor

As shown in Scheme 1, probe a and b in single-stranded structure would be first absorbed on the Fe₃O₄/GO surface with high affinity. Once the target T-DNA is added, the two probes could leave the GO surface and hybridize with T-DNA for double-stranded structure. In the presence of K⁺, their guanine-rich sequences get close to each other for shaping G-quadruplex structure and then combine hemin to produce G-quadruplex DNAzyme. After being exposed to an external magnetic field, the Fe₃O₄/GO composites aggregate to one side and then the supernatant is pipetted out for test. At last, a certain amount of H₂DCFDA and H₂O₂ is added and the G-quadruplex DNAzyme would catalyze the reaction effectively to give a strong fluorescence signal. It is worth mentioning that the uncombined probes and hemin can be absorbed on the Fe₃O₄/GO surface. Moreover, the process of magnetic separation can decrease the background signal by taking out the excess probes and hemin, and avoid the influence of the intrinsic peroxidase activity of the Fe₃O₄/GO nanoparticles. In the absence of T-DNA, the two probes deposited on the Fe₃O₄/GO surface could not form the G-quadruplex structure. And the hemin added subsequently is also immobilized on the GO platform. Moreover, the Fe₃O₄/GO composites could be separated from the solution under an external magnetic field.

Eventually, this supernatant solution containing nearly no DNAzyme, would release a very weak fluorescence signal after the addition of H₂DCFDA and H₂O₂.

3.2 Fluorescence analysis of T-DNA

To evaluate the feasibility for identifying target T-DNA, the performance of this split G-quadruplex DNAzyme based Fe₃O₄/GO platform is investigated. As illustrated in Fig. 1, the reaction system of H₂DCFDA and H₂O₂ (Fig. 1, curve a) releases a weak fluorescence signal. And then the signal displays an obvious increase after the addition of Fe₃O₄/GO nanoparticles (Fig. 1, curve b), which is consistent with the report that the ferromagnetic nanoparticle shows intrinsic peroxidase-like activity.²² While the probes are incubated with hemin, followed by adding H₂DCFDA and H₂O₂, the system shows a strong fluorescence signal (Fig. 1, curve c). It is attributed to the fact that the two G-rich parts could easily assemble to G-quadruplex structure in the absence of T-DNA and then produce a strong background signal. Furthermore, when the probes are incubated with Fe₃O₄/GO first, the fluorescence signal (Fig. 1, curve d) displays a little decrease compared to curve c (Fig. 1). That may be because that the split probes adsorbed by Fe₃O₄/GO surface shape less G-quadruplex DNAzyme while the hemin–Fe₃O₄/GO conjugates display superior catalytic property than the free hemin. Subsequently, as T-DNA is introduced, the signal (Fig. 1, curve e) further increases by 14.8%. Considering the strong background signal (Fig. 1, curve d), an external magnetic field is introduced to separate the redundant probes–hemin–Fe₃O₄/GO conjugates. After that, the background signal (Fig. 1, curve f) dramatically decreases, and the presence of T-DNA also induces a stronger fluorescence signal (Fig. 1, curve g) by 56.1%. The results indicate that the introduction of Fe₃O₄/GO nanoparticle can improve the sensitivity and also greatly decrease the background signal of this constructed sensor.

Fig. 1 Fluorescence emission spectra of (a) H₂DCFDA + H₂O₂; (b) Fe₃O₄/GO + H₂DCFDA + H₂O₂; (c) probes + hemin + H₂DCFDA + H₂O₂; (d) probes + Fe₃O₄/GO + hemin + H₂DCFDA + H₂O₂; (e) probes + Fe₃O₄/GO + T-DNA + hemin + H₂DCFDA + H₂O₂; (f) probes + Fe₃O₄/GO + hemin + magnetic separation + H₂DCFDA + H₂O₂; (g) probes + Fe₃O₄/GO + hemin + T-DNA + magnetic separation + H₂DCFDA + H₂O₂. The procedure of the system above is the same as that in Section 2.3 (H₂DCFDA, 5 × 10⁻⁷ mol L⁻¹; H₂O₂, 2 × 10⁻⁴ mol L⁻¹; Fe₃O₄/GO, 40 mg L⁻¹; probes, 1.0 × 10⁻⁸ mol L⁻¹; hemin, 5 × 10⁻⁸ mol L⁻¹; T-DNA, 1.0 × 10⁻¹⁰ mol L⁻¹).

3.3 Optimization of parameters

The fluorescence recovery efficiency is calculated by F/F₀, where F and F₀ denote the fluorescence intensity of this DNAzyme based Fe₃O₄/GO platform in the presence and in the absence of T-DNA respectively. To obtain the maximum F/F₀, the effect of some experimental parameters has been investigated. As shown in Fig. 2(A), both F and F₀ increase dramatically with the pH from 6.0 to 8.0 while F decreases a little as pH further increases to 8.5. And the value of F/F₀ reaches its maximum at pH 7.5 (Fig. S1A†), which is in accordance with the report that the DNAzyme-based biosensor is effective in neutral pH.²³ Therefore, pH 7.5 is selected in the following experiments.

Fig. 2 The influences of (A) pH from 6.0 to 8.5; (B) Fe₃O₄/GO from 0 to 56 mg L⁻¹; (C) hemin from 5.0 × 10⁻⁹ to 1.0 × 10⁻⁷ mol L⁻¹; F₀ (colored orange) is the fluorescence intensity in the absence of T-DNA, F (colored cyan) is the fluorescence intensity in the presence of 3.0 × 10⁻⁸ mol L⁻¹ T-DNA. The other conditions are the same as those in Section 2.3.

Furthermore, Fe₃O₄/GO nanomaterial is one of the most important parameters and a range of concentration from 0 to 56 mg L⁻¹ is investigated. Fig. 2(B) shows that both F and F₀ reduce sharply with the addition of 16 mg L⁻¹ Fe₃O₄/GO. From the result, we can speculate the addition of Fe₃O₄/GO leads to less G-quadruplex DNAzyme because of the strong adsorption of the probes on Fe₃O₄/GO surface. And then F₀ changes a little with the further increase of the Fe₃O₄/GO concentration. Moreover, both F and F/F₀ obtain the largest value in the presence of 40 mg L⁻¹ Fe₃O₄/GO (Fig. S1B†). As a result, the concentration of 40 mg L⁻¹ Fe₃O₄/GO is utilized in the following experiments.

In addition, the unreacted hemin in solution would produce a background signal and its concentration ranged from 5.0 × 10⁻⁹ to 1.0 × 10⁻⁷ mol L⁻¹ is investigated. As shown in Fig. 2(C), both F and F₀ increase gradually with the increasing hemin concentration and then F/F₀ levels off after 5.0 × 10⁻⁸ mol L⁻¹ (Fig. S1C†). As a result, 5.0 × 10⁻⁸ mol L⁻¹ hemin is employed in the following experiments.

3.4 Fluorescence analysis of T-DNA

Under the optimal conditions, the performance of this sensor has been investigated in the presence of T-DNA with different concentrations. As shown in Fig. 3(A), the fluorescence intensity shows a gradual increase along with the concentration from 0 to 1.0 × 10⁻¹¹ mol L⁻¹ [Fig. 3(A), curve a to e]. From the result in the inset of Fig. 3(A), the intensity is linearly proportional to the logarithm of the T-DNA concentration from 1.0 × 10⁻¹⁴ to 1.0 × 10⁻¹¹ mol L⁻¹ with a correlation coefficient R² of 0.9905. And the detection limit is estimated to be 1.87 × 10⁻¹⁵ mol L⁻¹ based on 3σ/slope (n = 7), which indicates that the proposed sensor is sensitive for the analysis of T-DNA. However, as T-DNA further increases to 1.0 × 10⁻¹⁰ mol L⁻¹ [Fig. 3(A), curve f], the fluorescence signal decreases a little. The decrease is unusual, it may be because the change of concentration induces the transformation of reaction mechanism and it needs further research in the following work. And then the intensity increases continuously with T-DNA from 1.0 × 10⁻¹⁰ to 1.0 × 10⁻⁷ mol L⁻¹. As shown in the inset of Fig. 3(B), the fluorescence signal is linearly proportional to the logarithm of the T-DNA concentration from 5.0 × 10⁻¹⁰ to 1.0 × 10⁻⁷ mol L⁻¹ with R² of 0.9939. As a result, the T-DNA concentration from 1.0 × 10⁻¹⁴ to 1.0 × 10⁻⁷ mol L⁻¹ is critically relevant for authentication with this G-quadruplex DNAzyme based Fe₃O₄/GO sensor.

Fig. 3 Fluorescence spectra of the split G-quadruplex DNAzyme based Fe₃O₄/GO platform with T-DNA: A: (a) 0; (b) 1.0 × 10⁻¹⁴; (c) 1.0 × 10⁻¹³; (d) 1.0 × 10⁻¹²; (e) 1.0 × 10⁻¹¹ mol L⁻¹. Inset: derived calibration curve of the fluorescence intensity at 522 nm versus T-DNA concentration. B: (f) 1.0 × 10⁻¹⁰; (g) 5.0 × 10⁻¹⁰; (h) 1.0 × 10⁻⁹; (i) 3.0 × 10⁻⁹; (j) 5.0 × 10⁻⁹; (k) 3.0 × 10⁻⁸; (l) 5.0 × 10⁻⁸ (m) 1.0 × 10⁻⁷ mol L⁻¹. The other conditions are the same as those in Section 2.3. Inset: derived calibration curve versus T-DNA concentration.

3.5 Selectivity of the G-quadruplex DNAzyme based Fe₃O₄/GO sensor

To investigate the selectivity of the split G-quadruplex DNAzyme based Fe₃O₄/GO sensor for T-DNA, we have recorded the fluorescence responses to the ITS sequences of PH from different geographical regions. Moreover, the geographical regions include Yixing city of Jiangsu province (JSYX for short), Nanjing city of Jiangsu province (JSNJ), Zhenjiang city of of Jiangsu province (JSZJ), Chuzhou city of Anhui province (AHCZ) and Xuancheng city of Anhui province (AHXC), and their sequences are obtained according to the sequencing results by Yu's group.⁶ As shown in Fig. 4, 1.0 × 10⁻¹⁰ mol L⁻¹ T-DNA could induce an obvious fluorescence enhancement compared to the blank system. Furthermore, 1.0 × 10⁻¹⁰ mol L⁻¹ JSNJ and JSZJ only cause a little increasement, while JSYX, AHCZ and AHXC bring a much weaker fluorescence enhancement than T-DNA. The above results indicate that our sensing system exhibits a high selectivity for target PH in preference to the other PH possessing three or more variation sites.

Fig. 4 Fluorescence intensity of the split G-quadruplex DNAzyme based Fe₃O₄/GO sensor in presence of T-DNA (1.0 × 10⁻¹⁰ mol L⁻¹) or other PH sequences (1.0 × 10⁻¹⁰ mol L⁻¹) from different geographical regions. The other conditions are the same as those in Section 2.3.

4. Conclusions

In summary, a split G-quadruplex DNAzyme based Fe₃O₄/GO sensor is proposed for identifying Pseudostellaria heterophylla from different geographical regions. And this sensor can detect target sequences with high sensitivity, wide dynamic range, low cost and high selectivity. Moreover, this paper not only provides a new fluorescent strategy combined G-quadruplex DNAzyme and Fe₃O₄/GO, but also opens a new application of G-quadruplex sensor in the authentication of traditional Chinese medicine.

Acknowledgements

We gratefully acknowledge the Project 81303176 supported by NSFC, the Professional technical service platform of research and development of traditional Chinese medicine and health products (14C26243501795), Youth Foundation of Fujian Health Department (2013-2-71), Project of Fujian Provincial Department of Education (type B, No. JB13092).

References

1. J. Hu, W. S. Pang, J. L. Chen, S. W. Bai, Z. Z. Zheng and X. H. Wu, BMC Complementary Altern. Med., 2013, 13, 267.
2. Y. Zhu, M. J. Qin, Y. Y. Hang and X. Q. Zheng, Journal of Plant Resources and Environment, 2007, 16, 19–22.
3. H. M. Gao, Z. M. Wang, Y. J. Li and Z. Z. Qian, Front. Med., 2011, 5(2), 195–202.
4. Z. Z. Zheng, J. Han, W. S. Pang and J. Hu, Sensors, 2013, 13, 1064–1075.
5. H. C. Hao, J. Y. Chang and F. Y. Chung, Genomic Med., Biomarkers, Health Sci., 2012, 4, 43–44.
6. Y. B. Yu, M. J. Qin, Z. T. Liang, G. D. Yu and N. H. Tan, Journal of Plant Resources and Environment, 2003, 12(4), 1–5.
7. Z. Z. Zheng, J. Hu and Z. D. He, Sensors, 2014, 14, 22971–22981.
8. B. Qiu, Z. Z. Zheng, Y. J. Lu, Z. Y. Lin, K. Y. Wong and G. N. Chen, Chem. Commun., 2011, 47, 1437–1439.
9. P. Zou, Y. Liu, H. Wang, J. Wu, F. Zhu and H. Wu, Biosens. Bioelectron., 2016, 79, 29–33.
10. J. B. Zhu, L. B. Zhang, S. J. Dong and E. K. Wang, Chem. Sci., 2015, 6, 4822–4827.
11. X. H. Mao, A. J. Simon, H. Pei, J. Y. Shi, J. Li, Q. Huang, K. W. Plaxco and C. H. Fan, Chem. Sci., 2016, 7, 1200–1204.
12. K. Nakatsuka, H. Shigeto, A. Kuroda and H. Funabashi, Biosens. Bioelectron., 2015, 74, 222–226.
13. F. L. Qu, T. Li and M. H. Yang, Biosens. Bioelectron., 2011, 26, 3927–3931.
14. S. Bi, T. T. Zhao, X. Q. Jia and P. He, Biosens. Bioelectron., 2014, 57, 110–116.
15. P. S. Wate, S. S. Banerjee, A. Jalota-Badhwar, R. R. Mascarenhas, K. R. Zope, J. Khandare and R. D. K. Misra, Nanotechnology, 2012, 23, 415101.
16. X. Y. Li, X. L. Huang, D. P. Liu, X. Wang, S. Y. Song, L. Zhou and H. J. Zhang, J. Phys. Chem. C, 2011, 115, 21567–21573.
17. C. Y. Chen, X. C. Jiang, Y. V. Kaneti and A. B. Yu, Powder Technol., 2013, 236, 157–163.
18. G. Cheng, Y. L. Liu, Z. G. Wang, J. L. Zhang, D. H. Sun and J. Z. Ni, J. Mater. Chem., 2012, 22, 21998–22004.
19. X. Y. Guan, J. M. Chang, Y. Chen and H. J. Fan, RSC. Adv., 2015, 5, 50126–50136.
20. F. Gao, H. Qu, Y. Y. Duan, J. Wang, X. Song, T. J. Ji, L. X. Cao, G. J. Nie and S. Q. Sun, RSC Adv., 2014, 4, 6657–6663.
21. G. X. Jin, L. J. Lu, X. Y. Gao, M. J. Li, B. Qiu, Z. Y. Lin, H. H. Yang and G. N. Chen, Electrochim. Acta, 2013, 89, 13–17.
22. L. Z. Gao, J. Zhuang, L. Nie, J. B. Zhang, Y. Zhang, N. Gu, T. H. Wang, J. Feng, D. L. Yang, S. Perrett and X. Y. Yan, Nat. Nanotechnol., 2007, 2, 577–583.
23. M. Liu, H. M. Zhao, S. Chen, H. T. Yu, Y. B. Zhang and X. Quan, Biosens. Bioelectron., 2011, 26, 4111–4116.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra16220g

---