Label-free DNA electrochemical sensor based on a PNA-functionalized conductive polymer

Talanta 76 (2008) 206–210

Contents lists available at ScienceDirect

Talanta journal homepage: www.elsevier.com/locate/talanta

Label-free DNA electrochemical sensor based on a PNA-functionalized conductive polymer S. Reisberg a , L.A. Dang b , Q.A. Nguyen b , B. Piro a , V. Noel a , P.E. Nielsen c , L.A. Le b , M.C. Pham a,∗ a

Laboratoire Interfaces-Traitements-Organisation et Dynamique des Syst`emes (ITODYS), Universit´e Paris 7-Denis Diderot, associ´e au CNRS, UMR 7086, 1 rue Guy de la Brosse, Paris 75005, France b Institut de Chimie – VAST, 18 Av. Hoang Quoc Viet, Cau Giay, Hanoi, Viet Nam c Department of Cellular and Molecular Medicine, The Panum Institute, Blegdansvej 3C, Copenhagen DK-2200, Denmark

a r t i c l e

i n f o

Article history: Received 13 November 2007 Received in revised form 8 February 2008 Accepted 20 February 2008 Available online 8 March 2008 Keywords: PNA Electrochemical biosensor DNA sensor

a b s t r a c t An electrochemical hybridization biosensor based on peptide nucleic acid (PNA) probe is presented. PNA were attached covalently onto a quinone-based electroactive polymer. Changes in flexibility of the PNA probe strand upon hybridization generates electrochemical changes at the polymer–solution interface. A reagentless and direct electrochemical detection was obtained by detection of the electrochemical changes using square wave voltammetry (SWV). An increase in the peak current of quinone was observed upon hybridization of probe on the target, whereas no change is observed with non-complementary sequence. In addition, the biosensor is highly selective to effectively discriminate a single mismatch on the target sequence. The sensitivity is also presented and discussed. © 2008 Published by Elsevier B.V.

1. Introduction Several technological approaches were followed until now to develop DNA sensors. Optical systems are certainly the most frequent. Two of them are already commercialized, based on fluorescence or surface plasmon resonance spectroscopies. However, electrochemical systems have great potentialities because of their low cost, simplicity, and obvious compatibility with miniaturization [1]. Among the electrochemical methods, the most used is based on redox labels which generate a signal change upon hybridization. However, the main drawback of this technique is the need for a redox label to be added in solution, or grafted on DNA strands. To solve this problem, the redox indicator can be covalently grafted onto the electrode. Following this scheme, electrochemical methods like cyclic voltammetry (CV) [2–14], differential pulse voltammetry (DPV) [15–19] or electrochemical impedance spectroscopy (EIS) [20–28] have been successfully used. Peptide nucleic acids (PNA) are DNA mimics with a pseudopeptide backbone (see Scheme 1). PNA oligomers are able to form very stable duplex structures with complementary DNA (or PNA) oligomers [29–31]. Conversely to DNA, PNA strands are neutral, i.e. the electrostatic repulsion is absent between two hybridized

∗ Corresponding author. Tel.: +33 1 5727 7223; fax: +33 1 5727 7263. E-mail address: [email protected] (M.C. Pham). 0039-9140/$ – see front matter © 2008 Published by Elsevier B.V. doi:10.1016/j.talanta.2008.02.044

PNA or PNA–DNA strands. As a consequence, PNA–PNA as well as PNA–DNA duplexes have a higher association constant and a better thermal stability than corresponding DNA–DNA duplexes. Moreover, single-base mismatches are more selectively discriminated in a PNA–DNA context [29,32]. This would make PNA-based hybridization sensors more selective than classical DNA-based homologues. For these reasons, these molecules are of interest in many areas. The aim of this article is to focus on the peculiar property of PNA strands to change their flexibility upon hybridization. Indeed, single-stranded PNA have flexible backbones, whereas doublestranded PNA or mixed PNA–DNA duplexes have a very rigid structure. It was recently reported that PNA oligomers have even more flexible chains than equivalent DNA strands [33,34], which therefore increases the hybridization impact on the probe neighbourhood, and justifies the use of PNA as probes. Several works have already explored the strong PNA flexibility changes after DNA–PNA duplexes formation, mainly based on optical (fluorescence quenching) characterization of hybridization [35–37]. Some electrochemical sensors based on PNA probes have been reported, using redox indicators to be added in solution [38–40] or ferrocene-modified PNA probe to transduce directly the hybridization event [41,42]. In the latter case, the label-free sensor gives a “signal-off” (signal decrease) detection. In this paper, we present a label-free and direct electrochemical system based on a quinone-containing conducting polymer and demonstrate that hybridization of a complementary DNA target onto the PNA-modified electrode generates electrochemical

S. Reisberg et al. / Talanta 76 (2008) 206–210

207

Scheme 1. Comparison between DNA and PNA structures.

changes at the polymer–solution interface. These electrochemical changes are detectable using square wave voltammetry (SWV) and lead to a “signal-on” (current increase). The sensor is very selective as non-complementary or single mismatch strand do not lead to significant change. The sensor electrode can be re-used after a simple dehybridization step in pure water.

was stopped: the electrode was washed in distilled water (5 min) in order to remove non-covalently bound PNA, then in PBS (2 h at 37 ◦ C). Phosphates in PBS are able to react with NHS-activated ester. Unreacted ester groups are removed after these operations. The surface concentration of probe was estimated measuring the surface concentration of a fluorescent target strand, around 10–20 pmol cm−2 .

2. Experimental 2.2. Electrochemical methods 2.1. Chemicals N -(3-Dimethylaminopropyl)-N-ethylcarbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) were provided by Sigma. Phosphate buffer saline solution (PBS, 0.137 M NaCl; 0.0027 M KCl; 0.0081 M Na2 HPO4 ; 0.00147 M KH2 PO4 , pH 7.4) was from Sigma. Aqueous solutions were made with bi-distilled or ultrapure (Millipore) water. Juglone (5-hydroxy-1,4naphthoquinone, JUG) and 1-naphthol (1-NAP) were purchased from Fluka. 5-Hydroxy-3-thioacetic acid-1,4-naphthoquinone (JUGA) was synthesized in our laboratory. Acetonitrile (ACN) was supplied by Aldrich (HPLC grade). All other reagents used were of analytical grade. Oligonucleotides were synthesized by Eurogentec (Belgium). Peptide nucleic acids (PNA) were synthesized in the Department of Cellular and Molecular Medicine, Panum Institute, Denmark. Probe grafting was performed as follows. The poly(5-hydroxy-1, 4-naphthoquinone-co-5-hydroxy-3-thioaceticacid-1,4-naphthoquinone)-coated glassy carbon (GC) electrodes were dipped into a solution containing 0.1 ␮M of PNA probe (pPNA, presenting a free terminal amine group, see Table 1), 1.5 × 10−2 M EDC (3.10−5 mol, 1 equiv.) and 3 × 10−2 M NHS (6.10−5 mol, 2 equiv.) in distilled water at 37 ◦ C. After 20 h, the PNA immobilization reaction (amidation)

Table 1 PNA probes and DNA targets Name

Function

Sequence

pPNA tDNA-10 tDNA-15 mDNA-15 rDNA-15

Probe Target Target Mismatch Random

NH2 -TCTTCCTCTCAGCCT-H (5’) 5 GAGAGTCGGA 3 5 AGAAGGAGAGTCGGA 3 5 AGAAGGAGCGTCGGA 3 5 GATCCATGCATTCCG 3

a

Melting temperaturea 36.9 ◦ C/50.6 ◦ C 53.7 ◦ C/68.5 ◦ C 43.8 ◦ C/59.2 ◦ C –

Tm were computed from [44]. The first temperature corresponds to DNA/DNA hybrids, calculated from [44a]. The second temperature corresponds to PNA/DNA hybrids, calculated from [44b].

For electrochemical experiments, a conventional onecompartment, three-electrode cell was employed. An EG&G 263A potentiostat was used with the Echem software (Ecochemie). The working electrodes were glassy carbon disks (Toka¨ı carbon, Japan) of 0.07 cm2 area. The auxiliary electrode was a platinum grid and the reference electrode a commercial Saturated Calomel Electrode (SCE, MetrOhm). The electrochemical synthesis of poly(JUG-co-JUGA) films was carried out by electrooxidation of a mixture of 5 × 10−2 M JUG + 5 × 10−3 M JUGA + 2 × 10−3 M 1-naphthol + 0.1 M LiClO4 in acetonitrile, on GC electrodes, under dried argon atmosphere, by potential scans from 0.4 to 1 V vs. SCE during 50 cycles at 50 mV s−1 . The resulting conducting film presents a thickness of ca. 150 nm. The quinone group embedded in the polymer structure is electroactive in neutral aqueous medium in the potential range [0 V; 1 V] vs. SCE, for a surface concentration of electroactive quinones of around 2 × 10−9 mol cm−2 . The chemical structure of this polymer has been described in details elsewhere, and is simply reminded on Scheme 2 [18,19]. Hybridization was detected by recording the modification of the redox process of the quinone group, using square wave voltammetry (SWV). The following parameters were used: pulse height 50 mV, pulse width 50 ms, scan increment 2 mV, frequency 12.5 Hz, potential domain (−0.9; 0 V vs. SCE). The medium was PBS, bubbled with argon for 40 min before and during SWV measurements. The SWV scans were repeated until complete stabilization of the signal (i.e., no difference observed between two successive responses). All electrochemical experiments were conducted at 25 ◦ C. 2.3. Hybridization For hybridization experiments, 1 mL of PBS (pH 7.4) containing 100 nM of target DNA was used. The electrode (0.07 cm2 ) bearing the PNA probe strand was dipped in this target solution then heated at a temperature above the melting temperature of the

208

S. Reisberg et al. / Talanta 76 (2008) 206–210

Scheme 2. Polymer structure, hybridization reaction and transduction process.

mismatch duplex (60 ◦ C, see Table 1) during 2 h without stirring, and slowly cooled down to room temperature (at a rate of 0.5 ◦ C per min. down to 25 ◦ C during 1 h). This step is very important. Indeed, it removes mismatch duplexes and non-complementary sequences which could be adsorbed. After that, the electrode was washed again in PBS for 1 h at 25 ◦ C. In our hybridization conditions, for a surface concentration of probe of 10 pmol cm−2 , around 6 pmol cm−2 of target strands are hybridized.

3. Results and discussion Table 1 presents the PNA probe and DNA target sequences used in this study. We used as the probe strand a complementary sequence of the anti-gag gene of the HIV virus. Four target sequences were used. A 10 bases tDNA-10, shorter than the probe; a 15 bases tDNA-15 that is the full-complementary sequence of the pPNA probe; a random sequence rDNA-15 which is non specific; and a mismatch sequence mDNA-15 which presents one mismatching base. Hybridization was detected by recording the modification of the redox process of the quinone group, using square wave voltammetry (SWV) between −0.9 V and 0 V vs. SCE in phosphate buffer saline (PBS) at 25 ◦ C. In one hand, the quinone group presents a reversible electroactivity in this potential domain, which is particularly sensitive to its chemical environment, for example protons or cations concentrations. In the other hand, ODN strands are polyanionic molecules and carry a high charge density. It is therefore justified to consider that the high charge density carried on ODN can influence its environment, i.e., in our case, the polymer–solution interface and the quinone group. Actually, in previous works in the literature dealing with DNA–DNA hybridization on conducting polymers, explanations of the transduction process were based on ion-exchange hindering and charge screening [10–14,23,43]. However, our idea is that hybridization may be detected more efficiently if only the target strand is charged, and not the probe strand. Indeed, the probe strand may generate a charge screening between the target strand and the electrode. This is why the idea described in this manuscript is to use a PNA (neutral) probe, in order to have only the target strand bearing charges. In other words, binding of the target DNA onto the neutral PNA changes the surfaces charge from neutral to negative to a greater extent than with probe DNA. The

polymer structure, hybridization reaction and transduction process are illustrated on Scheme 2. Initially, hybridization of a 10 bases complementary sequence was performed, using the tDNA-10 target on a pPNA-15 probe. Measurements were made in two steps, before and after tDNA hybridization, with a target concentration of 100 nM. In order to make the result more precise, we present the differential SWV current (i) (Fig. 1, curve 1, dotted line). i is obtained by subtracting the SWV current before hybridization from the SWV current measured after hybridization. As shown, the current increase is clear. However, the maximum of i remains low (about 3 ␮A at −500 mV/SCE). Hybridization of the 15 bases complementary sequence (i.e., same length than pPNA) was performed using the tDNA-15 target strand. The results are shown in Fig. 1, curve 2 (plain line). As shown, the current is higher for the tDNA-15 than for the tDNA-10. This effect of the length on hybridization is common and has been previously shown for DNA–DNA hybridization [18,45].

Fig. 1. Differential SWV responses obtained after (1) hybridization of tDNA-10 onto pPNA; (2) hybridization of tDNA-15 onto pPNA; (3) addition of the mismatch sequence mDNA-15; (4) addition of the random sequence rDNA-15. i corresponds to the difference ih − ig ; ih is the SWV current after hybridization and ig before. Conditions detailed in the Section 2.

S. Reisberg et al. / Talanta 76 (2008) 206–210

209

Fig. 2. Differential SWV responses obtained after (1) hybridization of tDNA-15 onto pPNA; (2) dehybridization by a stringent washing procedure; (3) second hybridization of tDNA-15. Same conditions as Fig. 1. Fig. 3. Differential SWV currents, measured at −500 mV/SCE, after addition of tDNA15, for increasing concentrations from 10 nM up to 100 nM. Same conditions as Fig. 1.

These results constitute a convincing proof-of-concept for this PNA-based DNA sensor. However, the true challenge, for a DNA sensor, is to achieve a good selectivity for DNA sequence recognition. To demonstrate this, we used a random and a single mismatch target sequences. The first is a random sequence (rDNA-15) which does not present any specificity to the PNA probe. The second is much closer to the genuine target sequence, bearing only one mismatch base (mDNA-15). This is a key-experiment to demonstrate selectivity. The results obtained for these two targets are shown in Fig. 1 (curve 3, dotted line and curve 4, dash-dotted lines). As shown, incubation with rDNA-15 and mDNA-15 (followed by a washing step at 60 ◦ C) leads to current changes which are not significant, whereas hybridization with the full-complementary sequence tDNA-15 gives a clear current increase. It is also possible to denature the pPNA–tDNA hybrid, in order to regenerate the electrode surface after a first assay. To illustrate this, after a first hybridization assay, the PNA–DNA hybrid was denaturated in stringent conditions by washing in pure deionised water at 37 ◦ C under stirring during 24 h. The electrode was then reequilibrated in PBS during 2 h at the same temperature. The results are shown in Fig. 2. It appears that, after the washing step, the differential SWV current recovers a value close to the one obtained before hybridization. This means that the de-hybridization does occur. Moreover, a second hybridization on the same film leads to a current increase which is similar to the one obtained for the first assay. However, further de-hybridizations (>3) lead to a progressive decrease of the current. Finally, we measured the SWV current response as a function of the concentration of target strand, for concentrations between 10 nM and 100 nM (i.e., 10 pmol and 100 pmol in 1 mL), as shown in Fig. 3. The detection limit appears to be around 10 nM. This is a good value in comparison to most of the results published in the literature concerning DNA sensor using a direct (and reagentless) electrochemical process. All targets present in the hybridization solution do not hybridize. The lowest detectable signal is obtained for a target concentration around 10 nM, i.e., 10 pmol in 1 mL sample. We hope to decrease these values (and therefore increase sensitivity) by using lower volumes and smaller electrodes. This work is in progress, with ultra-microelectrodes and microfluidic cells.

4. Conclusion A PNA-probe modified electrode based on a conducting polymer was found to transduce the hybridization event to an electrochemical signal with current increase,“signal-on”. There is no need for a time-consuming labeling process and external indicators. The electrochemical response is obtained directly by square wave voltammetry. The sensor can discriminate a single mismatch and can be regenerated after a simple dehybridization step in pure water to be re-used. Its detection limit is around 10 nM, but does not constitute a theoretical limit and probably can be improved. The simplicity of the strategy may open the doors for application in genetic diagnosis and work is under study in this direction. Acknowledgments S. Reisberg thanks the French Ministry of Research for a Ph.D. grant, and University Paris-Diderot-Paris 7 for a Post-Doctoral position. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]

E. Bakker, Y. Qin, Anal. Chem. 75 (2006) 3965. A.B. Steel, T.M. Herne, M.J. Tarlov, Anal. Chem. 70 (1998) 4670. I.V. Yang, H.H. Thorp, Anal. Chem. 73 (2001) 5316. J. Wang, D. Xu, A. Kawde, R. Polsky, Anal. Chem. 73 (2001) 5576; J. Wang, The Analyst 130 (2005) 421. A.M. Oliveira-Brett, M. Vivan, I.R. Fernandes, J.A.P. Piedade, Talanta 56 (2002) 959. A. Liu, J.I. Anzai, Anal. Chem. 76 (2004) 2975. E. Dominguez, O. Rincon, A. Narvaez, Anal. Chem. 76 (2004) 3132. E. Katz, I. Willner, J. Wang, Electroanalysis 16 (2004) 19. M.V. Del Pozo, C. Alonso, F. Pariente, E. Lorrenzo, Biosens. Bioelectron. 20 (2005) 1549. H. Korri-Youssoufi, F. Garnier, P. Srivastava, P. Godillot, A. Yassar, J. Am. Chem. Soc. 119 (1997) 7388. A. Emge, P. Bauerle, Synth. Met. 102 (1999) 1370. N. Lassalle, E. Vieil, J.P. Correia, L.M. Abrantes, Synth. Met. 119 (2001) 407. L.A. Thompson, J. Kowalik, M. Josowicz, J. Janata, J. Am. Chem. Soc. 125 (2003) 324. J. Cha, J.I. Han, Y. Choi, D.S. Yoon, K.W. Oh, G. Lim, Biosens. Bioelectron. 18 (2003) 1241. G. Marrazza, I. Chianella, M. Mascini, Biosens. Bioelectron. 14 (1999) 43. B. Wang, L. Bouffier, M. Demeunynck, P. Mailley, A. Roget, T. Livache, P. Dumy, Bioelectrochemistry 63 (2004) 233.

210

S. Reisberg et al. / Talanta 76 (2008) 206–210

[17] D. Ozkan, A. Erdem, P. Kara, K. Kerman, B. Meric, J. Hassmann, M. Ozsoz, Anal. Chem. 74 (2002) 5931. ¨ M.C. Pham, Anal. Chem. 77 (2005) 3351. [18] S. Reisberg, B. Piro, V. Noel, [19] S. Reisberg, B. Piro, V. Noel, M.C. Pham, Bioelectrochemistry 69 (2006) 172. [20] E. Palecek, Anal. Biochem. 170 (1988) 421; E. Palecek, Talanta 56 (2002) 809. [21] K.M. Millan, S.R. Mikkelsen, Anal. Chem. 65 (1993) 2317. [22] E. Souteyrand, J.P. Cloarec, J.R. Martin, C. Wilson, I. Lawrence, S. Mikkelsen, M.F. Lawrence, J. Phys. Chem. B 101 (1997) 2980. [23] J. Wang, M. Jiang, A. Fortes, B. Mukherjee, Anal. Chim. Acta 402 (1999) 7. [24] H. Peng, C. Soeller, N. Vigar, P.A. Kilmartin, M.B. Cannell, G.A. Bowmaker, R.P. Cooney, J. Travas-Sejdic, Biosens. Bioelectron. 20 (2005) 1821. [25] E. Katz, I. Willner, Electroanalysis 15 (2003) 913. [26] C. Li, Y. Long, J.S. Lee, H. Kraatz, Chem. Commun. (2004) 574. [27] L. Alfonta, A.K. Singh, I. Willner, Anal. Chem. 73 (2001) 91. [28] J. Liu, S. Tian, P.E. Nielsen, W. Knoll, Chem. Commun. (2005) 2969. [29] P.E. Nielsen, M. Egholm, R.H. Berg, O. Buchardt, Science 254 (1991) 1497. [30] M. Egholm, O. Buchardt, L. Christensen, C. Behrens, S.M. Freier, D.A. Driver, R.H. Berg, S.K. Kim, B. Norden, P.E. Nielsen, Nature 365 (1993) 566. [31] E. Uhlmann, A. Peyman, G. Breipohl, D.W. Will, Angew. Chem. Int. Ed. Engl. 37 (1998) 2796. [32] P.E. Nielsen, M. Egholm, Peptide Nucleic Acids: Protocols and Applications, Horizon Scientific Press, Norfolk, UK, 1999.

[33] S. Sen, L. Nilsson, J. Am. Chem. Soc. 123 (2001) 7414. [34] R. Soliva, E. Sherer, F.J. Luque, C.A. Laughton, M. Orozco, J. Am. Chem. Soc. 122 (2000) 5997. [35] E. Ortiz, G. Estrada, P.M. Lizardi, Mol. Cell. Probes 12 (1998) 219. [36] B. Armitage, D. Ly, T. Koch, H. Frydenlund, H. Orum, G.B. Schster, Biochemistry 37 (1998) 9417. [37] H. Juhn, V.V. Demidov, J.M. Coull, M.J. Fiandaca, B.D. Gildea, M.D. FrankKamenetskii, Antisense Nucleic Acid Drug Dev. 11 (2001) 265. [38] J. Wang, E. Palecek, P.E. Nielsen, G. Rivas, X.H. Cai, H. Shiraishi, N. Dontha, D.B. Luo, P.A.M. Farias, J. Am. Chem. Soc. 118 (1996) 7667. [39] D. Ozkan, A. Erdem, P. Kara, K. Kerman, J.J. Gooding, P.E. Nielsen, M. Ozsoz, Electrochem. Commun. 4 (2002) 796. [40] H.X. Ju, H.T. Zhao, Front. Biosci. 10 (2005) 37. [41] H. Aoki, P. Buhlmann, Y. Umezawa, Electroanalysis 12 (2000) 1272. [42] H. Aoki, H. Tao, Analyst 132 (2007) 784. [43] V.V. Demidov, V.N. Potaman, M.D. Frank-Kamenetskii, M. Egholm, O. ¨ Buchard, S.H. Sonnichsen, P.E. Nielsen, Biochem. Pharmacol. 48 (1994) 1310. [44] (a) N. Le Novere, Bioinformatics 17 (2001) 1226; (b) U. Giesen, W. Kleider, C. Berding, A. Geiger, H. Ørum, P.E. Nielsen, Nucleic Acids Res. 26 (1998) 5004. [45] B. Piro, S. Reisberg, V. Noel, M.C. Pham, Biosens. Bioelectron. 22 (2007) 3126.