Aptamer-mediated detection of thrombin using silver nanoparticle signal enhancement

Abstract

We present the first assay combining a dual aptamer sandwich format with detection by anodic stripping voltammetry with ionic silver amplification. This assay format lends itself to rapid point-of-care tests, where the use of aptamers could improve the overall stability of the assay. We have used human alpha-thrombin as a model system, and demonstrate a detection limit of 6.09 μg L⁻¹. We present the optimization of the aptamer-silver colloid attachment chemistry and the final assay format to achieve sensitive analyte detection. The use of a sandwich assay coupled with magnetic separation and ionic silver amplification, generates an assay with similar sensitivity than those reported in the literature in a format that can be used in rapid, portable testing regimes.

---

Introduction

DNA aptamers are short single-stranded oligonucleotides that can fold into complicated three dimensional structures. Some of these short DNA sequences can strongly and selectively bind with other molecules, in a similar manner to antibodies. Due to the fact that they are nucleic acid based, aptamers have many advantages over antibodies; for example, greater stability, less batch-to-batch variation due to their chemical synthesis, simplicity of modification, and reduced development and synthesis timescale. There are many examples of aptamer use as biosensor recognition elements in the literature.¹ Thrombin specific aptamers have been used extensively in the literature to demonstrate novel biosensor techniques.²

Here we describe the use of two thrombin-specific aptamers in an amplified electrochemical assay. The first, designated N15, (also known as the Bock aptamer),³ consists of 15 nucleotides and binds to thrombin with a K_d of approximately 26 nM.⁴ The second, designated N29, (also known as the Tesset aptamer), consists of 29 nucleotides and binds to thrombin more tightly with a K_d of approximately 0.5 nM.⁴ Both aptamers have similar three-dimensional structures based upon a G-quadruplex motif; however they have been found to bind to completely different regions of the thrombin molecule. N15 binds to thrombin's fibrinogen-recognition exosite and N29 to thrombin's heparin-binding exosite.⁴ Using the ability of the N15 and N19 aptamers to recognize and bind to unique regions of the thrombin molecule, it is possible to develop a “sandwich”-style assay. Several groups have used this sandwich approach to develop a range of aptamer-based biosensors for thrombin measurement. These techniques include; inductively coupled plasma-mass spectrometry,⁵ gold nanoparticle colorimetry,⁶ and on-chip detection employing magnetic beads and quantum dots.⁷ Groups have also developed sandwich-based electrochemical methods including amplification approaches using glucose dehydrogenase,⁸ pyrroquinoline quinone glucose dehydrogenase ((PQQ)GDH)⁹ and alkaline phosphatase.¹⁰

Sandwich assays have also been demonstrated using combinations of thrombin-specific aptamers and antibodies. In one electrochemically based example, thrombin-specific antibody is covalently attached to a glassy carbon electrode and acts as a capture probe; the thrombin-specific aptamer acts as the detection probe, which is intercalated by the electrochemically active compound methylene blue (MB).¹¹ An advantage of this approach is that the aptamer can be used without modification, as labelling before use is not required, due to the ability of MB to spontaneously intercalate into double-stranded DNA. Unfortunately, assays using this strategy must use a non-aptamer capture ligand to limit background MB signal from the capture phase, removing the advantages of a dual aptamer sandwich detection system.

Of many possible capture phases, magnetic particles are the most amenable for the development of point-of-care assays. Their high surface area to volume ratio, and the ability to add the magnetic particles directly to the whole sample to speed up the kinetics of the binding reaction are particular advantages. Moreover, the use of an external magnet to remove the magnetic particles from the sample following incubation greatly simplifies the assay washing steps. In one published sandwich assay, aptamer derivatised gold and magnetic nanoparticles are bound together through the presence of thrombin. Following washing, inductively coupled plasma mass spectrometry was used to measure the gold concentration, from which the original concentration of thrombin in the sample can be estimated.⁵ Although highly sensitive, the use of such large and complex instrumentation far from meets the requirements of the point-of-care biosensor market.

Among the many available detection approaches, electrochemistry linked with metal nanoparticles holds many advantages. Each 40 nm silver nanoparticle (AgNP) consists of 10⁶ silver atoms, which after dissolution form electrochemically measurable silver ions and thus in principle obtain a sensitivity enhancement factor of 10⁶.¹² What is more, electrochemical techniques work in turbid and small volume samples,¹³ do not require expensive instrumentation and can be miniaturised to portable devices. In combination with magnetic separation, the greater stability of aptamers and electrochemical readout, this format is ideal for the production of automated, inexpensive point-of-care biosensors. For a full description of the ionic-silver amplified, anodic striping voltammetric assay used herein, please see Porter et al. 2009,¹² Szymanski et al. 2010 (ref. 14) and Szymanski et al. 2011.¹⁵ A brief overview of the assay format is presented in Scheme 1.

Scheme 1 The principle of the assay. (A) Aptamer labelled silver (1) and magnetic particles (2) complex in the presence of thrombin (3). (B) A magnet is used to remove the magnetic particles and any thrombin complexed silver from the bulk reaction mixture for electrochemical analysis. (C) The silver concentration is determined by anodic stripping voltammetry and is proportional to the thrombin concentration in the sample.

Results and discussion

Conjugate optimisation

Two approaches for aptamer attachment to the silver colloid were investigated, direct attachment with 5′ thiol terminators and indirect attachment using biotinylated aptamers attached by streptavidin physisorbed to the silver nanoparticle surface. The “direction” of the sandwich, i.e. which of the N15 or N29 aptamers was used on the capture (magnetic bead) and which was used on the detection phase (silver colloid), was also investigated. Results indicate that the indirect aptamer attachment using streptavidin-coated silver nanoparticles and biotinylated aptamers produced higher signals than thiol mediated attachment (Fig. 1). Use of the N29 aptamer as the capture phase ligand and N15 for the detection phase also improves the final signal. This is most likely due to the use of the higher K_d (0.5 nM) N29 aptamer when sequestering thrombin from the sample in the first step of the assay. Using a negative control aptamer of a randomly generated sequence, we confirmed aptamer sequence specificity (data not shown). Other groups have tested the selectivity of this aptamer pair for human alpha-thrombin. We have run similar tests (data not shown) and can confirm that the system is able to differentiate human alpha thrombin from human beta and gamma thrombin, prothrombin, fibrinogen and human IgG.^5,7

Fig. 1 Conjugate optimization and assay format investigation. Vertical bars indicate final oxidation signal from assays containing 0, 100 or 1000 μg L⁻¹ thrombin when using the following capture (magnetic bead bound) and detection (silver nanoparticle bound) assay “directions”. (A) Capture aptamer N29, detection aptamer N15; biotinylated, (B) capture aptamer N29, detection aptamer N15; thiol, (C) capture aptamer N15, detection aptamer biotinylated N29, and (D) capture aptamer N15, detection aptamer N29 thiol. Note the capture aptamer is always attached via biotin.

Assay format optimization

The approach utilized in the original example of this electrochemical technique used a 96-well plate as the solid capture phase.¹² Later the capture phase was transferred onto magnetic microparticles to improve the assay kinetics and facilitate integration with microfluidics, for compatibility with testing at the point-of-care.¹⁵ Here, we demonstrate the use of the two bead format with aptamer capture and label ligands, that could be adapted to this rapid testing assay. We have optimized the relative concentrations of the magnetic (0.42, 1.25 and 3.75 mg mL⁻¹) and silver (3.3, 10.0 and 30.0 μg mL⁻¹) nanoparticles in the assay comparing the signal generated with 1000, 100 and 0 μg L⁻¹ of human alpha thrombin (Fig. 2). Results indicate that the optimum assay conditions for thrombin detection were 0.42 mg mL⁻¹ magnetic bead and 30 μg mL⁻¹ silver conjugate (final concentrations) per reaction. Fig. 2C shows that there is minimal difference between the thrombin-independent signals (background) for the particle concentration tested. These results are contrary to our expectations; we believed that the addition of a larger concentration of magnetic bead to the assay would result in more thrombin being carried forward to the silver addition step and subsequently resulting in a larger final signal. However, the lowest magnetic bead concentration gave the largest final signal. It is possible that during the electrochemical process magnetic beads (which are still present in the solution) deposit upon the electrode surface and screen or modify the plating and stripping process, which one would expect to be concentration-dependent. Using samples with a fixed concentration of silver nanoparticles and varying concentrations of magnetic bead (spanning concentrations used during assay optimisation) we were unable to demonstrate magnetic bead depended suppression of the signal (data not shown).

Fig. 2 Assay format optimisation. Final assay concentrations of the aptamer-conjugated magnetic bead (x-axis) and silver nanoparticles (y-axis) were titrated in the aptamer sandwich assay. Graphs (A), (B) and (C) show peak currents of samples containing 1000, 100 and 0 μg L⁻¹ human alpha thrombin respectively. Graphs A and C have identical z scales to enable comparison of signal to noise. Sample 1.25 mg mL⁻¹ magnetic bead and silver 3.3 μg mL⁻¹ at 100 μg L⁻¹ was lost during processing.

Assay response curves

Using a pair of thrombin-specific aptamers and ammonium thiocyanate mediated anodic stripping voltammetry we have developed a proof-of-principle assay for the detection of human alpha thrombin. In Fig. 3A, anodic striping voltammetry signals generated from assays of samples containing different levels of thrombin are presented. Fig. 3B presents the integrated oxidation signal versus thrombin concentration. Using the standard method for the calculation of the assay limit of detection (LOD) (integrated thrombin concentration from the response at 0 μg L⁻¹ + 3σ) we have achieved limits as low as 6.09 μg L⁻¹, or 164 pM. In comparison, groups using a similar two aptamer approach with magnetic beads and quantum dots reached LODs of 18 μg L⁻¹;⁷ electrochemical detection and amplification with glucose dehydrogenase 1 μM;⁸ (PQQ)GDH 10 nM;⁹ and alkaline phosphatase 0.45 nM.¹⁰ The most sensitive literature example of an aptamer sandwich detection approach for the measurement of human alpha thrombin required gold nanoparticle amplification and inductively coupled plasma-mass spectrometry detection (10 pM);⁵ however, we believe that many of these techniques are not applicable to point-of-care applications.

Fig. 3 A calibration curve and spectra for the measurement of the alpha human thrombin. (A) Voltammograms generated during the silver stripping step, correlates to the concentration of human alpha thrombin in the original sample. (B) A graph of the peak current plotted against thrombin concentration in μg L⁻¹. The fit presented is a polynomial quadratic with an R2 of 0.999; from which assay LOD was estimated to be 6.09 μg L⁻¹.

With modification to the assay format – for example, through the use of microfluidic devices – enabling more favourable kinetics, this platform will be readily transferable to the point-of-care. Here we have demonstrated an assay with performance similar to other literature examples where aptamer sandwich techniques have been used to measure human alpha thrombin. To further improve upon the detection limits we have achieved we intend to substitute the 40 nm diameter silver nanoparticles for larger particles. This will enable us to achieve higher enhancement factors than the theoretical 10⁶ fold-amplification from a single silver nanoparticle to silver ions using the silver particles employed in this study.

Experimental

Materials

The aptamers N15 (5′-Biotin-T20-GGTTGGTGTGGTTGG-3′) and N29 (5′Biotin-T20-AGTCCGTGGTAGGGCAGGTTGGGGTGACT-3′) and a negative control sequence (5′-Biotin-GGGAGACAAGAATAAACGCTCAAGCTGTTATCTATATGGTCCTTATTTTTCGACAGGAGGCTCACAACAGGC-3′) were ordered from IDT (Leuven, Belgium). They were both modified with a tail of 20 thymidines and biotin at the 5′ end. Similarly, 5′-dithiol modified aptamers with 12-thymidine tails (S–N15 and S–N29) were also ordered. Human α-thrombin, boric acid, ethylenediaminetetraacetic acid (EDTA), tris, ammonium thiocyanate, bovine serum albumin (BSA) and streptavidin were all purchased from Sigma UK. Silver colloids (40 nm diameter) were from nanoComposix and magnetic particles (Dynabeads MyOne™ 1 μm diameter, Streptavidin C1) from Invitrogen.

Preparation of magnetic conjugate

Biotinylated aptamers were conjugated to the magnetic particles according to manufacturer's instructions. To summarise the procedure, magnetic particles were washed 3 times in 10 mM Tris pH 7.5, 1 mM EDTA, 2 M NaCl. Magnetic beads were then resuspended in 4× the starting volume of 5 mM Tris pH 7.5, 0.5 mM EDTA, 2 M NaCl and 3 μM N29 aptamer and incubated for 60 minutes with gentle agitation before washing 3 times in 5 mM Tris pH 7.5, 0.5 mM EDTA 2 M NaCl. Magnetic beads were reconstituted in 4× the original starting volume in 10 mM phosphate buffer pH 7.4, 2.7 mM KCl, 0.137 M NaCl and 0.1% BSA (PBS-B) and stored at 4 °C until use.

Preparation of silver conjugate

Biotin-mediated conjugation. Silver-containing solutions were covered in aluminium foil as much as possible during conjugation to reduce light exposure. A 1 mL aliquot of silver sol was centrifuged for 10 minutes at 16 100 × g. The supernatant was removed and the pellet resuspended in 1 mL of 0.2 M H₃BO₃ pH 7.5 (Ag conjugation buffer) containing 30 μg mL⁻¹ streptavidin. After a 2 h incubation with gentle agitation the silver sol was centrifuged again and the pellet resuspended in 1 mL of 0.2 M H₃BO₃ pH 7.5 with 0.1% BSA (Ag storage buffer), and incubated at room temperature for further 30 minutes to block unbound sites on the silver nanoparticle. Silver–streptavidin conjugate prepared this way was then used to immobilise the biotinylated N15 aptamer. First, the silver–streptavidin conjugate was centrifuged and the pellet resuspended in 1 mL of Ag storage buffer supplemented with a 30 μM concentration of N15 aptamer. The mixture was then agitated for 1 h. The resulting silver–aptamer conjugate was washed twice and resuspended in a 1 mL volume of Ag storage buffer. The conjugate was stored at 4 °C until use.

Thiol-mediated conjugation. The direct immobilization of thiol-modified aptamers was based on similar protocols to those for gold nanoparticles^16,17 the silver colloid was centrifuged for 10 min at 16 100 × g, the supernatant was removed and the pellet resuspended in equal volume of Ag conjugation buffer with 3 μM TCEP-activated aptamer. The silver was incubated with thiol-modified aptamer for 24 hours at room temperature in the dark and 16 μL of 2 M NaCl was added and incubated for further 8 hours. 16 μL of 2 M NaCl was added again and after 8 hours, 20 μL of 10% BSA was added. The addition of NaCl was to weaken the electrostatic interaction between the silver nanoparticle and the oligonucleotide, which will increase aptamer surface packing density.¹⁸ The silver was then washed twice by centrifuging for 10 min at 14 000 × g and resuspending in 1 mL of Ag storage buffer.

Aptamer sandwich assay

Human alpha-thrombin was diluted to the required final concentration in PBS-B. Aptamer conjugated magnetic beads (50 μL) and thrombin samples (50 μL) were mixed and incubated with gentle agitation for 1 hour at room temperature. Using magnetic pull-down the particles were washed twice with 200 μL PBS-B, re-suspended in 100 μL aptamer conjugated silver solution and incubated for 90 minutes at room temperature in the dark with gentle agitation. Finally, the magnetic beads were washed 3 times in PBS-B and re-suspended in 1 M NH₄SCN prior to electrochemical analysis.

Electrochemistry

50 μL of the magnetic suspension in NH₄SCN solution was transferred onto the electrode surface and the silver content was measured with Anodic Stripping Voltammetry (ASV). The ASV was conducted using an electrochemical workstation (AutoLab PGSTAT 12, Windsor Scientific) with the following steps and parameters: (a) pre-treatment: 0.6 V for 15 s; (b) nucleation step: −1.6 V for 5 s; (c) deposition step: −1.2 V for 120 s; (d) stripping step: staircase sweep from −1.2 V to 0.5 V at a scan rate of 1 V s⁻¹ and a potential step of 0.01 V.

Potentials are reported with respect to the stable potential of a carbon pseudo-reference electrode in ammonium thiocyanate (1 M).¹⁹ Baseline subtraction to calculate the peak currents produced during the silver oxidation striping step were performed using General Purpose Electrochemical System version 4.9, (Eco Chemie B.V. Utrecht, The Netherlands). The sensors used in the voltammetric experiments were single-use, screen-printed carbon three-electrode cells. The sensors were provided by LIRANS, University of Bedfordshire, printed with C2000802D2 carbon ink and D60202D1 blue dielectric supplied by Gwent Electronic Materials. The sensing area of working electrode was 1.5 × 5 mm.

Conclusions

We have demonstrated that the dual aptamer based sandwich assay can be used for the detection of human alpha thrombin using a modified anodic stripping voltammetry technique. In combination with a microfluidic chip and a handheld reader the assay will be transferable to the point-of-care. In the current format assay performance is comparable to other aptamer based thrombin assays with limits of detection at 6.09 μg L⁻¹. In the future we intend to investigate methods to reduce the assay detection limit through the use of larger silver nanoparticles. Although assay specificity has been demonstrated in simple buffers, work to demonstrate assay function in complex matrixes such as human serum is still required. We are currently developing assays for other analytes relevant for point-of-care detection; and we intend to assess performance of the aptamer sandwich assay using a microfluidic platform which is currently in the final stages of development.

Acknowledgements

The authors are grateful for; the financial support provided by The National Measurement Office, Dr Jonathan Moore, Dr Adrian Horgan, Dr Sotiris Missailidis and Dr Suzanne Simmons for insightful discussions.

Notes and references

1. E. J. Cho, J. W. Lee and A. D. Ellington, Annu. Rev. Anal. Chem., 2009, 2, 241–264.
2. G. S. Baird, Am. J. Clin. Pathol., 2010, 134, 529–531.
3. L. C. Bock, L. C. Griffin, J. A. Latham, E. H. Vermaas and J. J. Toole, Nature, 1992, 355, 564–566.
4. D. M. Tasset, M. F. Kubik and W. Steiner, J. Mol. Biol., 1997, 272, 688–698.
5. Q. Zhao, X. Lu, C. G. Yuan, X. F. Li and X. C. Le, Anal. Chem., 2009, 81, 7484–7489.
6. H. Pandana, K. H. Aschenbach and R. D. Gomez, IEEE Sens. J., 2008, 8, 661–666.
7. Y. H. Tennico, D. Hutanu, M. T. Koesdjojo, C. M. Bartel and V. T. Remcho, Anal. Chem., 2010, 82, 5591–5597.
8. K. Ikebukuro, C. Kiyohara and K. Sode, Anal. Lett., 2004, 37, 2901–2909.
9. K. Ikebukuro, C. Kiyohara and K. Sode, Biosens. Bioelectron., 2005, 20, 2168–2172.
10. S. Centi, S. Tombelli, M. Minunni and M. Mascini, Anal. Chem., 2007, 79, 1466–1473.
11. Y. Kang, K. J. Feng, J. W. Chen, J. H. Jiang, G. L. Shen and R. Q. Yu, Bioelectrochemistry, 2008, 73, 76–81.
12. R. Porter, A. Kabil, C. Forstern, C. Slevin, K. Kouwenberg, M. Szymanski and B. Birch, J. Immunoassay Immunochem., 2009, 30, 428–440.
13. M. Dequaire, C. Degrand and B. Limoges, Anal. Chem., 2000, 72, 5521–5528.
14. M. Szymanski, A. P. F. Turner and R. Porter, Electroanalysis, 2010, 22, 191–198.
15. M. Szymanski, R. Porter, G. V. Dep, Y. Y. Wang and B. G. D. Haggett, Phys. Chem. Chem. Phys., 2011, 13, 5383–5387.
16. G. Mayer, Nucleic Acid and Peptide Aptamers. Methods and Protocols, Humana Press, 2009.
17. J. Liu and Y. Lu, Nat. Protoc., 2006, 1, 246–252.
18. J. J. Storhofff, R. Elghanian, C. A. Mirkin and R. L. Letsinger, Langmuir, 2002, 18, 6666–6670.
19. J. W. Dilleen, S. D. Sprules, B. J. Birch and B. G. D. Haggett, Analyst, 1998, 123, 2905–2907.

---