Electrochemical DNA sandwich assay with a lipase label for attomole detection ofDNA

Abstract

A fast and sensitive electrochemical lipase-based sandwich hybridization assay for detection of attomole levels of DNA has been developed. A combination of magnetic beads, used for pre-concentration and bioseparation of the analyte with a lipase catalyst label allowed detection ofDNA with a limit of 20 amol.

---

Sensitive real time DNA assaying is crucial in cancer and bacterial disease diagnostics and treatment,^1,2 and it is of major significance for identification of biological pathogens and analysis of genetically modified organisms.^2,3 A constant demand for cost-effective approaches for rapid assays of specific DNA sequences at low concentrations inspired a search for new technologies and methodologies alternative to traditional optical detection methods^4,5 and radioisotope labeling.⁶SPR,⁷gravimetry,⁸ and electrochemical methods^9–11 are now actively used for DNA sensing. Electrochemical approaches for DNA detection are most promising, since they have the potential to provide rapid detection with lower costs, smaller equipment size and lower power requirements.¹¹ Recent trends of using nanoparticles/silver enhancement^12,13 and quantum dot methodologies^14,15 or multiplexing redox labels per each individual hybridized DNA^9,16 allow for further improvements of the DNA assay sensitivity. When combined with enzymatic signal amplification strategies, electrochemical methods have reported detection limits as low as femtomole DNA sensing. The use of peroxidaseenzyme as a label and the redoxpolymer as a “wire” allowed detection in the order of 10⁵ copies of DNA hybridized in a sandwich construction.¹⁷ Willner et al. used peroxidase¹⁸ and alkaline phosphatase¹⁹ as indicators of a biorecognition event which enabled 10 fmol ml⁻¹DNA detection.¹⁹ Further development of this method allowed attomole detection limits to be reached,²⁰ and efforts are now aimed at novel strategies for ultrasensitive detection.

Here we report a principally new methodology for DNA detection: a lipase-based electrochemical assay for attomole (10⁻¹⁸ mol)²¹detection ofDNA sequences. Lipases are inexpensive, commercially available enzymes and find wide applications in biotechnology, bioorganic synthesis, oleochemistry, and the paper and food industries.^22,23 Compared to other bond-cleaving enzymes, e.g. proteases, lipases are not prone to self-digestion.²³ In the current study we combine a magnetic bead (MB) sandwich hybridization capture assay used for pre-concentration and bioseparation,^13,24 with a lipase-based amplification and electrochemical readout system as shown in Fig. 1. The target sequence binds a lipase–DNA conjugate to the MBs. The captured lipases are applied to a self-assembled monolayer (SAM) of alkanethiols on a goldelectrode containing an internal estergroup and a terminal ferrocene (Fc) redox label for the electrochemical read-out of the DNA assay. The lipase present at the SAM surface will catalytically cleave the esters releasing Fc and thereby lead to a decrease in the electrochemical signal from the ferroceneredox label.

Fig. 1 (A) Stepwise modification of MB with biotinylated capture (blue), target (red) and biotinylated reporter (green) DNAs in a sandwich DNA hybridization assay and further labelling with lipase through the streptavidin linker. (B) Structure of 9-mercaptonon-1′-yl 4-ferroceneamidobutanoate used in the assay. (C) Principles of the assay: hybridization-activated enzymatic cleavage of the redox-active electrode film by the lipase-labelled DNA sandwich on magnetic beads releases the redox active probe from the electrode surface, thus decreasing the measured electrochemical signal.

For assay development, 9′-mercaptonon-1′-yl 4-ferroceneamidobutanoate was synthesised (ESI†) and the efficient cleavage of the Fc-alkanethiol ester bond in solution by the recombinant lipase from Thermomyces lanuginosus and lipase A from Candida antarctica was proven (verified by HPLC analysis, Fig. S2, ESI†). Then the Fc-alkanethiol ester was immobilized on a goldelectrodevia the thiol group to form the monolayer (Fig. 1C). The redox activity of this Fc-alkanethiol ester SAM and of SAMs produced in a series of dilution experiments of the Fc-alkanethiol ester SAMs by a number of simple alkanethiols was electrochemically investigated to determine the optimum conditions for “on-surface” cleavage of the Fc-alkanethiol ester by the lipases. For the successful monolayer and lipase combinations, further optimization of pH and buffer conditions was performed.

The magnitude of the Fc signal showed a tendency to decrease in phosphate buffer solutions (Fig. S3†), but was restored when the electrodes were transferred to H₂SO₄ solutions, indicative of a complex, environment-dependent structural rearrangement of the SAMs (Fig. S4†). In 1 M H₂SO₄ solution, the Fcredox peaks were centred at 555 ± 2 mV and characterized by the electron transfer (ET) rate constant k_s of 7.8 ± 0.3 s⁻¹.²⁵ The ester bond was only accessible for lipase hydrolysis in diluted Fc-alkanethiol ester SAMs, formed during 2 h modification and, alternatively, diluted by 1-hexanethiol, as verified by experiments with lipase and lipase-modified MBs (Fig. 2). This is remarkable, as previous reports have suggested that the ester in a SAM formed by a structurally related Fc-alkanethiol ester, (9-(5′-ferrocenylpentanoyloxy)nonyl disulfide), was inaccessible for lipase cleavage.²⁶ Exposure of the Fc SAM to enzymes inactive in ester bond cleavage (peroxidase) did not produce any decrease in the Fc electrochemical signal.

Fig. 2 (A) Representative cyclic voltammograms of the Fc-alkanethiol ester-modified microarray goldelectrode (solid line) before and (dashed lines) after treatment with lipase from Candida antarctica, treatment times varying from 1 to 24 h, scan rate being 0.1 V s⁻¹. (B) Representative cyclic voltammograms of the Fc-alkanethiol ester-modified microarray goldelectrode before (solid line) and after (dashed line) treatment with MB reacted with 0.1 fmol of target DNA, scan rate is 0.1 V s⁻¹. (C) Normalized Fc signal, expressed as the ratio between the Fcredox peak surface area in CVs before and after reaction with hybridized MB labelled with lipase from Candida antarctica, calibrated versus concentration of the target DNA.

In our case a stable and quantitatively reproducible signal from the Fc-terminated SAMs was obtained with microelectronically fabricated gold microelectrodes in 1 M H₂SO₄ solution, and it is a result of the highly reproducible electrode surface characteristics.²⁷ On average, 0.2 ± 0.01 pmol of Fcgroups (or 0.8 × 10⁻¹⁰ mol cm⁻²) were involved in the ET reaction as estimated by the Fc peak area integration. Upon reaction with lipase, the signal from Fc decreased correspondingly (Fig. 2A). However, it never dropped to zero, which is consistent with some restrictions in the lipase catalysis, such as restricted catalytic accessibility to some of the ester bonds in the Fc alkanethiol ester in the SAMs.

The sandwich hybridization assay was performed on the surface of streptavidin-coated MBs (the best results were obtained with 1 μm size MBs), saturated with a biotinylated capture DNAprobe (5′-biotin-(CH₂)₆-TTT TT T TTT TAA GTC GAA CGA GCT TCC-3′). The target DNA, composed of two 18 nts regions complementary both to the capture and reporter probes, and linked by six arbitrary nucleotides (5′-AACTCA CCA GTT CGC CAC TGA CGT GGA AGC TCG TTC GAC TTA-3′), was added to the suspension of the capture DNA-modified MB. Next, the reporter probe (5′-GTG GCG AAC TGG TGA GTT TTT TTT TTT-TEG-biotin-3′), containing a biotin modification, was added (Fig. 1C, Table S1, ESI†). The targeted DNA sequence was captured in the hybridization sandwich architecture on the MBs, and finally the biotinylated reporter probe was labelled with a biotin-conjugated lipase through a streptavidin linker. Both the streptavidin linker and lipase–biotin conjugate were added stepwise in excess, providing in a first step full saturation of all DNAbiotins with streptavidin (10-fold excess) and further in a second step with biotinylated lipase (4-fold excess). Of the tested lipases, only lipase from Candida antarctica displayed catalytic activity after conjugation with biotin, and it was therefore used for DNAlabelling.

Through the whole hybridization assay the MBs were used for pre-concentration, washing away excess material and bioseparation of the reactants, while retaining the MB by the use of a magnet. The completed lipase–target DNA–MBs were collected and allowed to react on goldelectrodes modified with the Fc-alkanethiol ester SAM. The 1 h reaction time satisfied conditions of a complete reaction within all DNA concentration ranges tested. The MB-captured lipases caused cleavage of a fraction of the ester bond in the Fc-alkanethiol ester SAM, resulting in diffusion of a fraction of the Fcgroups from the electrode surface. As a result the electrochemical signal from the Fc consistently decreased (Fig. 2B) and was calibrated versus concentration of the added DNA target within the 0.02–40 fmol range (Fig. 2C). The DNA detection limit was 20 amol, which is comparable to the best reliable results reported in the literature using alternative electrochemical DNA assays.^{13,14,16,17,20}

A relatively fast concentration saturation of the Fc signal decrease, after 1 h reaction time, was observed in all experiments and, similarly to the reaction between the Fc-alkanethiol ester SAM and lipase in solution (Fig. 2A), the signal never dropped to zero. No interference was observed from 20 nts long non-complementary DNA sequences added. In the assay with non-complementary DNA, from 1 to 4% increase in the Fc signal, correlating with a background signal from unmodified magnetic beads, was observed in the first scan. Upon consecutive CV scanning in 1 M H₂SO_4, the signal gradually stabilised at the original level.

To verify that the decrease of the Fcredox signal originates only from the specific lipase digestion of the ester bonds and not from some restructuring of these SAMs, we have devised an experiment to detect the formation of surface hydroxylgroups left after lipase hydrolysis. This was done by reaction of the SAM surface (before and after MB–lipase treatment) with a phosphoramidite-biotin modifier. Phosphoramidites are commonly used in DNA synthesis and react with hydroxylgroups. Before treating the SAMs with MB–lipase no hydroxylgroups are present at the surface, whereas lipase cleavage results in the formation of hydroxylgroups at the surface (See Fig. 1). Subsequently, the phosphoramidite-biotin modifier-treated SAMs were subjected to reaction with streptavidin-conjugated fluorescent quantum dotnanocrystals (Fig. 3A). Fluorescence microscopy of these electrodes verified the absence of free hydroxyls at the SAM that was not lipase-treated (Fig. 3B), whereas the lipase-treated SAM displayed strong fluorescence after treatment with the streptavidin-conjugated fluorescent quantum dotnanocrystals (Fig. 3C), confirming the formation of free hydroxyls by the lipase cleavage reaction.

Fig. 3 (A) Scheme for a stepwise modification of the lipase-digested SAMs with fluorescent quantum dotnanoparticles conjugated to streptavidin. (B, C) Fluorescence microscope images of the Fc-alkanethiol ester modified gold microarray electrodes, (B) no reaction with lipase and (C) after reaction with lipase-labelled hybridized MB. Both surfaces in B and C were treated with the biotin-phosphoramidite linker and subsequently with streptavidin-modified quantum dotnanocrystals (Qdot® Streptavidin conjugate Q10121MP, see ESI for details†) to visualize hydroxylgroups at the surface left after lipase hydrolysis. Scale bar is 25 μm.

In conclusion, we have demonstrated a new lipase-based DNA hybridization sandwich assay for monitoring attomole levels of DNA. A lipaseenzyme was used as a catalyst label providing a decrease in the electrochemical signal at minute DNA concentrations. A combination of magnetic beads used for pre-concentration and bioseparation of the analyte with the use of this hydrolytic enzyme allows ultrasensitive detection ofDNAs at the electrode surface due to accumulation of the catalysis product (ester bond cleavage and removal of the redox label from the zone of electrochemical reaction). The achieved 20 amol detection limit approaches the hitherto best obtained results of other electrochemical DNA detection schemes,^{13,14,16,17,20} while microelectronically produced electrodes²⁷ provided high reproducibility of the detected signals. Along with this, the developed electrochemical lipase and MB-based sandwich assay represents a new approach for sensitive DNA hybridization detection and can also be used to increase the sensitivity of sandwich immunoassays.²⁴ Future studies will investigate the potential of this approach for selective and highly sensitive bacterial DNA detection.

The work was supported by The Danish Food Industry Agency (DFFE) and the Danish National Research Foundation. E.E.F thanks the Carlsberg Foundation for financial support.

Notes and references

1. A. Rasooly and J. Jacobson, Biosens. Bioelectron., 2006, 21, 1851.
2. O. Lazcka, F. J. Del Campo and F. X. Muñoz, Biosens. Bioelectron., 2007, 22, 1205.
3. I. Palchetti and M. Mascini, Analyst, 2008, 133, 846.
4. S. Tyagi and F. R. Kramer, Nat. Biotechnol., 1996, 14, 303.
5. E. E. Ferapontova, C. P. Mountford, J. Crain, A. H. Buck, P. Dickinson, J. S. Beattie, P. Ghazal, J. G. Terry, A. J. Walton and A. R. Mount, Biosens. Bioelectron., 2008, 24, 422.
6. G. H. Séller and M. M. Manak, DNA probes, Stockton Press, New York, 2nd edn, 1993.
7. K. A. Peterlinz, R. M. Georgiadis, T. M. Herne and M. J. Tarlov, J. Am. Chem. Soc., 1997, 119, 3401.
8. Y. Okahata, Y. Matsunobo, K. Ijiro, M. Mukae, A. Murkami and K. Makino, J. Am. Chem. Soc., 1992, 114, 8299.
9. C. J. Yu, Y. J. Wan, H. Yowanto, J. Li, C. Tao, M. D. James, C. L. Tan, G. F. Blackburn and T. J. Meade, J. Am. Chem. Soc., 2001, 123, 11155.
10. C. Fan, K. W. Plaxco and A. J. Heeger, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 9134.
11. T. G. Drummond, M. G. Hill and J. K. Barton, Nat. Biotechnol., 2003, 21, 1192.
12. S. J. Park, T. A. Taton and C. A. Mirkin, Science, 2002, 295, 1503.
13. J. Wang, D. Xu and R. Polsky, J. Am. Chem. Soc., 2002, 124, 4208.
14. J. A. Hansen, R. Mukhopadhyay, J. Ø. Hansen and K. V. Gothelf, J. Am. Chem. Soc., 2006, 128, 3860.
15. E. Katz, I. Willner and J. Wang, Electroanalysis, 2004, 16, 19.
16. F. Patolsky, Y. Wiezmann and I. Willner, J. Am. Chem. Soc., 2002, 124, 770.
17. T. de Lumley-Woodyear, D. J. Caruana, C. N. Campell and A. Heller, Anal. Chem., 1999, 71, 394.
18. F. Patolsky, M. Zayatz, E. Katz, A. Bardea and I. Willner, Anal. Chem., 1999, 71, 3171.
19. F. Patolsky, A. Lichtenstein and I. Willner, Nat. Biotechnol., 2001, 19, 253.
20. F. Azek, C. Cgrossiord, M. Joannes, B. Limoges and P. Brossier, Anal. Biochem., 2000, 284, 107.
21. Results are reported in moles, not mol l⁻¹, since pre-concentration ability of magnetic beads allowed amol amounts of DNA to be collected both from 50 μl and 5 ml solutions, with a lowest DNA concentration thus limited by 4 fM.
22. R. D. Schmid and R. Verger, Angew. Chem., Int. Ed., 1998, 37, 1608.
23. S. Shipovskov, Biotechnol. Prog., 2008, 24, 1262.
24. A. Ambrosi, M. T. Castañeda, A. J. Killard, M. R. Smyth, S. Alegret and A. Merkoçi, Anal. Chem., 2007, 79, 5232.
25. E. Laviron, J. Electroanal. Chem., 1979, 101, 19.
26. G. Valincius, I. Ignatjev, G. Niaura, M. Kažemėkaitė, Z. Talaikytė, V. Razumas and A. Svendsen, Anal. Chem., 2005, 77, 2632.
27. E. Ferapontova, J. Terry, A. J. Walton, C. P. Mountford, J. Crain, A. H. Buck, P. Dickinson, C. J. Campbell, J. S. Beattie, P. Ghazal and A. R. Mount, Electrochem. Commun., 2007, 9, 303.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, synthetic scheme, electrochemical characterization of SAMs. See DOI: 10.1039/b924627d

---