Electrochemical Nucleic Acid Bio Sensors

Analytica Chimica Acta 469 (2002) 63–71

Review

Electrochemical nucleic acid biosensors Joseph Wang∗ Department of Chemistry and Biochemistry, College of Arts and Sciences, New Mexico State University, Box 30001-Dept. 3C, Las Cruces, NM 88003, USA Received 13 June 2001; received in revised form 5 September 2001; accepted 10 September 2001

Abstract Electrochemical devices have received considerable attention in the development of sequence-specific DNA hybridization biosensors. Such devices rely on the conversion of the DNA base-pair recognition event into a useful electrical signal. Electrochemical biosensing of DNA hybridization is not only uniquely qualified for meeting the size, cost, and power requirements of decentralized genetic testing, but offer an elegant route for interfacing—at the molecular level—the DNA-recognition and signal-transduction elements. This article reviews current directions in electrochemical DNA biosensors, and discusses recent strategies and future prospects for such electrical detection. © 2002 Elsevier Science B.V. All rights reserved. Keywords: DNA biosensors; Electrochemistry; Voltammetry; Electrode

1. Introduction Wide-scale genetic testing requires the development of easy-to-use, fast, inexpensive, miniaturized analytical devices. Traditional methods for detecting DNA hybridization, such as gel electrophoresis or membrane blots, are too slow and labor intensive. Biosensors offer a promising alternative for faster, cheaper, and simpler nucleic acid assays. DNA hybridization biosensors commonly rely on the immobilization of a single-stranded (ss) oligonucleotide probe onto a transducer surface to recognize—by hybridization— its complementary target sequence. The binding of the surface-confined probe and its complementary target strand is translated into a useful electrical signal. Transducing elements reported in the literature have included optical [1], electrochemical [2], and micro∗ Tel.: +1-505-646-2140; fax: +1-505-646-6033. E-mail address: [email protected] (J. Wang).

gravimetric [3] devices. The two major requirements for a successful operation of a DNA biosensor are high specificity (including observation of a change in a single nucleotide) and high sensitivity. Even though nucleic acids are relatively simple molecules, finding the sequence that contains the desired information, and distinguishing between perfect matches and mismatches, are very challenging tasks. Electrochemical transducers have received considerable recent attention in connection to the detection of DNA hybridization [2,4,5]. The high sensitivity of such devices, coupled to their compatibility with modern microfabrication technologies, portability, low cost (disposability), minimal power requirements, and independence of sample turbidity or optical pathway, make them excellent candidates for DNA diagnostics. In addition, electrochemistry offers innovative routes for interfacing the nucleic acid recognition system with the signal-generating element and for amplifying electrical signals. Direct electrical reading of DNA

0003-2670/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 3 - 2 6 7 0 ( 0 1 ) 0 1 3 9 9 - X

64

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

interactions thus offers great promise for developing simple, rapid, and user-friendly DNA sensing devices (in a manner analogous to miniaturized blood-glucose meters). Recent efforts have led to a host of new strategies for electrical detection of DNA hybridization [2,4,5]. Such electrochemical avenues for generating the hybridization signal are the subject of the present review.

2. Electrochemical biosensing of DNA hybridization Electrochemical detection of DNA hybridization usually involves monitoring of a current response,

resulting from the Watson–Crick base-pair recognition event, under controlled potential conditions [2,4,5]. The probe-coated electrode is commonly immersed into a solution of a target DNA whose nucleotide sequence is to be tested. When the target DNA contains a sequence which matches that of the immobilized oligonucleotide probe DNA, a hybrid duplex DNA is formed at the electrode surface (Fig. 1). Such hybridization event is commonly detected via the increased current signal of an electroactive indicator (that preferentially binds to the DNA duplex), in connection to the use of enzyme labels or redox labels, or from other hybridization-induced changes in electrochemical parameters (e.g. capacitance or conductivity).

Fig. 1. Major processes involved in electrochemical DNA biosensors based on the use of redox indicators (Ox).

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

65

In the following sections, we will focus on the major steps involved in electrochemical biosensing of DNA hybridization, namely the formation of the nucleic acid recognition layer, the actual hybridization event, and the transformation of this recognition event into an electrical signal (Fig. 1). As will be illustrated, the success of such devices requires a proper combination of synthetic-organic and surface chemistries, DNA-recognition, and electrical detection protocols. 2.1. Interfacial confinement The probe immobilization step plays a major role in the overall performance of electrochemical DNA biosensors. The achievement of high sensitivity and selectivity requires maximization of the hybridization efficiency and minimization of non-specific adsorption events, respectively. The probes are typically short oligonucleotides (25–40-mer) that are capable of hybridizing with specific and unique regions of the target nucleotide sequence. Control of the surface chemistry and coverage is essential for assuring high reactivity, orientation/accessibility, and stability of the surface-bound probe, as well as for avoiding non-specific binding/adsorption events. For example, it was demonstrated recently that the density of immobilized ss-DNA can influence the thermodynamics of hybridization, and hence, the selectivity of DNA biosensors [6]. Greater understanding of the relationship between the surface environment of biosensors and the resulting analytical performance is desired. This is particularly important as the physical environment of hybrids at solid/solution interface can differ greatly from that of hybrids formed in the bulk solution [6]. Several useful schemes for attaching nucleic acid probes onto electrode surfaces have thus been developed. The exact immobilization protocol often depends on the electrode material used for signal transduction. Common probe immobilization schemes include attachment of biotin-functionalized probes to avidin-coated surfaces [7], self-assembly of organized monolayers of thiol-functionalized probes onto gold transducers [8,9], carbodiimide covalent binding to an activated surface [10], as well as adsorptive accumulation onto carbon-paste or thick-film carbon electrodes [11]. The use of alkanethiol self-assembly methods has been particularly attractive for fabricating

Fig. 2. Schematic of the preparation of a mixed thiol-derivatized single-stranded oligonucleotide/6-mercapto-1-hexanol monolayer in a solution containing the target DNA: (A) adsorption of the ss-DNA (HS-ss-DNA); (B) formation of the mixed layer after the addition and adsorption of mercaptohexanol (from [9] with permission).

reproducible probe-modified surfaces with high hybridization activity [8,9]. For this purpose, the DNA is commonly immobilized on gold by forming mixed monolayers of thiol-derivatized single-stranded oligonucloetide and 6-mercapto-1-hexanol (Fig. 2). The thiolated probe is ‘put upright’ as a result of such coassembly with a short-chain alkanethiol. The latter, along with a hydrophilic linker (between the thiol group and DNA), is often used for minimizing non-specific adsorption effects (of unwanted nonhybridized DNA adsorbates). Such monolayer-based structures can also provide a general route for linking to the surface relevant (enzyme or redox) labels. Despite of this considerable progress, there are many fundamental questions concerning the surface orientation and accessibility, and the nature of the interfacial molecular interactions. Surface characterization techniques (e.g. XPS, reflectance IR ellipsometry) can shed useful insights into the surface coverage and organization [8,9]. 2.2. The hybridization event The development of electrochemical DNA biosensors (as well as other DNA biosensors), requires proper

66

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

attention to experimental variables affecting the hybridization event at the transducer-solution interface. These include the salt concentration, temperature, viscosity, the presence of accelerating agents, contacting time, base composition (G + C, %), and length of probe sequence. Careful control of the hybridization event is thus required. The stability of hybrids formed between strands with mismatched bases is decreased according to the number and location of the mismatches. Many DNA biosensors are not capable of selectively detecting a point mutation, as desired in numerous practical situations. Controlling the stringency of hybridization, particularly using elevated temperatures, can thus be used for discriminating among oligonucleotide hybrids (including mismatch discrimination). Control of the hybridization time can be used for tuning the linear dynamic range, with shorter time offering an extended range at the cost of lower sensitivity [12]. Detection limits ranging from the nanomolar to the picomolar concentration range can thus be achieved in connection to 5 and 60 min hybridization times. Even lower detection limits can be attained in connection to advanced amplification protocols (described in the following sections). We have demonstrated that significantly enhanced selectivity can be achieved by the use of peptide nucleic acid (PNA) probes [13]. Such DNA analogues, possess an uncharged pseudopeptide backbone (instead of the charged phosphate-sugar one of natural DNA). Because of their neutral backbone, PNA probes offer greater affinity in binding to complementary DNA, and improved distinction between closely related sequences (including the detection of single-base imperfections). This is attributed to the fact that mismatch in PNA/DNA duplexes is much more destabilizing than in a DNA/DNA duplexes (with a lowering of the Tm by 15 versus 11 ◦ C, respectively). Such mismatch discrimination is of particular importance in the detection of disease-related mutations. Proper attention should be given also to the reusability of the DNA biosensors, namely to the regeneration of the surface-bound single-stranded probe after each assay. Both thermal and chemical (sodium hydroxide, urea) regeneration schemes have been shown useful for ‘removing’ the bound target in connection with different DNA biosensor formats. Even more elegant is the use of controlled electric fields for facilitating the denaturation of the duplex

[14]. Such electronic control has been used also for differentiating among oligonucleotide hybrids. Mechanically renewed electrodes, including polishable biocomposites and graphite pencils, have also been used for regenerating a ‘fresh’ probe layer [15,16]. Alternately, one can use “one-shot” screen-printed electrodes, similar to those used for self-testing of blood glucose, and hence obviate the need for regeneration [11]. Such disposable DNA sensor strips also meet the needs of many decentralized genetic testing. 2.3. Electrochemical transduction of DNA hybridization The hybridization event is commonly detected via the increased current signal of a redox indicator (that associates with the newly formed surface hybrid), or from changes in electrochemical parameters (such as capacitance or conductivity) or in the redox activity of the nucleic acid resulting from the duplex formation. 2.3.1. Indicator-based detection Earlier devices have relied primarily on the use of electroactive hybridization indicators [4]. Such indicators are small electroactive DNA-intercalating or groove-binding substances, that posses a much higher affinity for the resulting hybrid compared to the single-stranded probe. Accordingly, the concentration of the indicator at the electrode surface increases when hybridization occurs, resulting in increased electrochemical response. Besides effective differentiation between ss- and ds-DNA, the indicator should possess a well-defined, low-potential, voltammetric response. Such properties of redox indicators are essential for attaining high sensitivity and selectivity. Both linear-scan or square-wave voltammetric modes [17] or constant-current chronopotentiometry [18] can be used to detect the association of the redox indicator with the surface duplex. Mikkelsen’s group, that pioneered the use of redox indicators, demonstrated its utility for detecting the cystic fibrosis
F508 deletion sequence associated with 70% of cystic fibrosis patients [19]. A detection limit of 1.8 fmol was demonstrated for the 4000-base DNA fragment in connection to a Co(bpy)3 3+ marker. High selectivity towards the disease sequence—but not to the normal DNA—was achieved by performing the hybridization at an elevated temperature of 43 ◦ C.

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

Such use of the electrochemical transduction mode requires that proper attention be given to the choice of the indicator and its detection scheme. Our laboratory demonstrated the use of the Co(phen)3 3+ indicator, in connection to a carbon-paste chronopotentiometric transducer and PNA probes, for detecting single-base imperfections in the p53 gene [20]. Other useful redox-active indicators include bisbenzimide dyes such as Hoecht 33258 [21] or anthracycline antibiotics such as daunomycin [22]. A daunomycin-based chronopotentiometric biosensor was combined with PCR amplification of DNA extracted from whole blood for the genetic detection of apolipoprotein E polymorphism [23]. New electroactive indicators, offering better distinction between ss- and ds-DNA have been developed for attaining higher sensitivity. Very successful has been the recent use of a threading intercalator ferrocenyl naphthalene diimide (FND) [24] that binds to the DNA duplex more tightly than usual intercalators and displays a negligible affinity to the single-stranded probe. This duplex-specific threading indicator resulted in a detection limit of 10 zmol in connection to differential pulse voltammetric monitoring of the hybridization event (Fig. 3). The oligonucleotide probe was chemisorbed onto gold electrodes through a thiol anchor. Table 1 summarizes common redox-active indicators used in electrochemical DNA hybridization biosensors. Oligonucleotides bearing electroactive reporter molecules, such as ferrocene or anthraquinone tags, have also been considered for electrical detection of surface hybridization [25,26]. Ferrocene tags are being used in a new hand-held device, the CMS eSensorTM system of Motorola Inc., that can detect up to 48 different sequences in connection to elegant surface chemistry (combining self-assembly of thiolated probes and phenylacetylene “molecular wires”) and a highly-sensitive ac voltammetric detection [27].

67

Fig. 3. Differential pulse voltammograms for the ferrocenyl naphthalene diimide indicator at the dT20 -modified electrode before (a) and after (b) hybridization with dA20 . Also shown, the chemical structure of the indicator (from [24] with permission).

It is possible also to employ metal nanoparticle labels, and to quantitate them following the hybridization and acid dissolution by a highly-sensitive electrochemical stripping protocol [28]. 2.3.2. Use of enzyme labels for detecting DNA hybridization Enzyme labels have been widely used in bioaffinity sensors, particularly in immunosensors. The use of enzyme labels to generate electrical signals offers also great promise for ultrasensitive electrochemical detection of DNA hybridization. Lumley et al. [29] demonstrated that a direct low-potential sensitive amperometric monitoring of the hybridization

Table 1 Examples of redox-active indicators used for the biosensing of DNA hybridization Indicator 3+

Co(bpy)3 Co(phen)3 3+ Hoecht 33258 Daunomycin Ferrocenyl naphthalene diimide

Detection mode transducer

Electrode

Ep,a vs. Ag/AgCl (V)

Reference

Cyclic voltammetry Chronopotentiometry Pulse voltammetry Chronopotentiometry Pulse voltammetry

Carbon paste Carbon paste Gold Screen printed Gold

0.15 0.15 0.58 0.45 0.50

[17] [18] [21] [22] [24]

68

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

event could be achieved in connection to the use of horseradish-peroxidase (HRP) labeled target and an electron-conducting redox polymer. In this system, the hybridization of enzyme labelled oligo(dA)25 target with oligo(dT)25 probe, covalently attached to electron-conducting redox hydrogel resulted in the ‘wiring’ of the enzyme to the transducer, and in a continuous hydrogen-peroxide electroreduction current. A single-base mismatch in an 18-base oligonucleotide was thus detected using a 7 ␮m-diameter carbon-fiber transducer. A HRP label has been combined by Patolsky et al. [30] with a biocatalytic precipitative accumulation of the enzyme-generating product to achieve multiple amplifications and hence, extremely low detection limits. Chronopotentiometry and faradaic impedance spectroscopy were employed for detecting the biocatalyzed deposition reaction. Applicability for the detection of mutations relevant to the Tay–Sachs genetic disorder was demonstrated. Enhanced amplification of DNA sensing processes was achieved also by using liposomes labeled with HRP in connection to faradaic impedance spectroscopic detection [31] (e.g. Fig. 4). Such use of functionalized

liposomes resulted in a dramatic signal amplification (of ca. 105 ). The same enzyme label was employed for quantitative pulse amperometric monitoring of PCR amplification [32] and for differential pulse measurements of sequences related to human cytomegalovirus DNA [33]. The coupling of enzyme-based DNA assays with an efficient magnetic removal of unwanted sample constituents has been illustrated in our laboratory [34]. Enhanced “wiring” of enzyme labels is anticipated through the emerging use of nanoparticle superstructures [35]. 2.3.3. Label-free electrochemical biosensing of DNA hybridization Increased attention has been given recently to direct label-free electrochemical detection schemes, in which the hybridization event triggers a change in an electrical signal. Such protocols greatly simplify the sensing protocol (as they eliminate the need for the indicator addition/association/detection steps) and offers an instantaneous detection of the duplex formation. Such direct, in situ detection can be accomplished by monitoring changes in the intrinsic redox activity of the

Fig. 4. Amplified electrical detection of DNA hybridization using HRP-functionalized liposomes and biocatalytic precipitation of the product of the enzymatic reaction (from [31] with permission).

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

nucleic acid target or probe or changes in the electrochemical properties of the interface. For example, it is possible to exploit changes in the intrinsic electroactivity of DNA accrued from the hybridization event [36–39]. Among the four nucleic acids bases, the guanine moiety is most easily oxidized and is most suitable for such indicator-free hybridization detection. To overcome the limitations of the probe sequences (absence of G), guanines in the probe sequence were substituted by inosine residues (pairing with C’s) and the hybridization was detected through the target DNA guanine signal [36]. A greatly amplified guanine signal, and hence, hybridization response can be obtained by using the electrocatalytic action of a Ru(bpy)3 2+ redox mediator [37]. This involves the following catalytic cycle: Ru(bpy)3 2+ → Ru(bpy)3 3+ + e−

(1)

Ru(bpy)3 3+ + G → Ru(bpy)3 2+ + G+

(2)

The presence of a guanine-containing nucleic acid target thus creates a catalytic cycle that results in a large current output. The ability of this approach to detect mutations or deletions involving guanine bases has been demonstrated [40]. This technology is currently being used in a commercial nucleic acid detection system (of Xanthon Inc.) that uses a 96-well microtiter plate format, where each well contains seven separate probe-coated indium–tin oxide electrodes (two of which are used as controls) [41]. A single microtiter plate thus allows 672 measurements (480 tests and 192 controls). In addition to anodic measurements of the target guanine, it is possible to use cathodic stripping measurements of the target adenine for sensitive detection of DNA hybridization [39]. It is possible also to exploit different rates of electron transfer through ss- and ds-DNA for probing hybridization (including mutation detection) via the perturbation in charge migration through DNA. Barton and coworkers demonstrated that such charge transport is disrupted by the presence of a single-base mismatch [42,43]. Such disruption and point mutation were detected, using a gold electrode modified with thiolated DNA, by monitoring changes in the charge transport between an electroactive methylene blue intercalator and a ferricyanide redox species. A substantially smaller electrocatalytic signal was

69

observed in the presence of the single-base mismatch. Prospects for designing electronic circuits based on manipulation of charge transport through DNA were discussed [44]. Direct, label-free, electrical detection of DNA hybridization has also been accomplished by monitoring changes in the conductivity of conducting polymer molecular interfaces, e.g. using DNA-substituted or doped polypyrrole films [45,46]. For example, Korri-Youssoufi et al. [45] has demonstrated that a 13-mer oligonucleotide-substituted polypyrrole (PPy) film displays a decrease current response during the duplex formation. Such change in the electronic properties of PPy has been attributed to bulky conformational changes along the polymer backbone due to its higher rigidity following the hybridization. New avenues for generating the hybridization signal are currently being explored in several laboratories. Siontorou et al. [47] reported on the use of self-assembled bilayer lipid membranes (BLMs) for the direct monitoring of DNA hybridization. A decrease in the ion conductivity across the lipid membrane surface, containing the single-stranded probe, was observed during the formation of the duplex. This was attributed to alterations in the ion permeation associated with structural changes in the BLM accrued by the desorption of the ds-DNA. The mechanism of interaction between oligonucleotides and BLM films was examined by Hianik et al [48]. Aoki et al. developed a novel ion-channel protocol for the indirect biosensing of DNA hybridization [49]. The system relied on the electrostatic repulsion of the diffusing ferrocyanide redox marker, accrued from the hybridization of the negatively-charged target DNA and the neutral PNA probe (Fig. 5). High specificity towards mismatch oligonucleotides was demonstrated. A related approach for amplifying DNA hybridization signals, based on the use of negatively-charged liposomes, was described by Patolsky et al. [50]. Such liposomes bind to the bound target to form a ‘giant’ negatively-charged interface that repels the anionic redox probe. The resulting barrier to the interfacial electron transfer was monitored by Faradaic impedance spectroscopy. Berggren et al. demonstrated that changes in the capacitance of a thiolated-oligonucleotide modified gold electrode, provoked by hybridization to the

70

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

Fig. 5. An ion-channel sensor based on a PNA probe immobilized on gold electrode, and detection of the hybridization based on the electrostatic repulsion of a negatively-charged redox marker (shown as an octahedron) (from [49] with permission).

complementary strand (and the corresponding displacement of solvent molecules from the surface), can be used for monitoring in high sensitivity and speed the hybridization event [51].

3. Conclusions and outlook Over the past decade, we have witnessed a tremendous progress towards the development of electrochemical nucleic acid biosensors. Such devices are of considerable interest due to their tremendous promise for obtaining sequence-specific information in a faster, simpler, and cheaper manner compared to traditional nucleic acid assays. In addition to excellent economic prospects, such devices offer innovative routes for interfacing (at the molecular level) the DNA-recognition and signal-transduction elements, i.e. an exciting opportunity for fundamental research. The realization of instant decentralized (medical, environmental, or forensic) DNA testing would require additional developmental work. Particular attention should be given to the major challenges of mismatch discrimination, signal amplification, non-specific adsorbates, as well

as integration of various processes, including sample collection, DNA extraction and amplification, with the actual hybridization detection, on a single microchip platform containing multiple functional elements and related microfluidic network. Such integration and miniaturization should lead to significant advantages in terms of cost, speed, sample/reagent consumption, simplicity, and automation. The integration of multiple biosensors in connection to DNA microarrays should lead to the simultaneous analysis of multiple nucleic acid sequences, and hence, to the generation of characteristics hybridization patterns and acquisition of expression information. Screening of DNA-protein or DNA-drug interactions would also benefit from such DNA microarrays. The rapid progress that electrical detection for DNA hybridization has made in the last decade suggests the major impact it may have in the present decade. It is envisioned that future research will lead to new electrical detection strategies, that coupled with major technological advances, will result in powerful, miniaturized, easy-to-use instruments for DNA diagnostics. Such instruments will undoubtedly accelerate the realization of large-scale genetic testing.

J. Wang / Analytica Chimica Acta 469 (2002) 63–71

Acknowledgements The author gratefully acknowledges financial support from the US Army Medical Research (Award no. DAMD17-00-1-0366) and the National Institutes of Health (Grant no. R01 14549-02).

References [1] P. Piunno, U. Krull, R. Hudson, M. Damha, H. Cohen, Anal. Chem. 67 (1995) 2635. [2] J. Wang, Chem. Eur. J. 5 (1999) 1681. [3] Y. Okahata, Y. Matsunobo, K. Ijiro, M. Mukae, A. Murkami, K. Makino, J. Am. Chem. Soc. 114 (1992) 8299. [4] S.R. Mikklesen, Electroanalysis 8 (1996) 15. [5] E. Palecek, M. Fojta, Anal. Chem. 73 (2001) 75A. [6] J. Watterson, P. Piunno, C. Wust, U. Krull, Langmuir 16 (2000) 4984. [7] R. Ebersole, J. Miller, J. Moran, M. Ward, J. Am. Chem. Soc. 112 (1990) 3239. [8] T. Herne, M. Tarlov, J. Am. Chem. Soc. 119 (1997) 8916. [9] R. Levicky, T. Herne, M. Tatlov, S. Satija, J. Am. Chem. Soc. 120 (1998) 9787. [10] K. Millan, A. Spurmanis, S.R. Mikkelsen, Electroanalysis 4 (1992) 929. [11] J. Wang, X. Cai, B. Tian, H. Shiraishi, Analyst 121 (1996) 965. [12] J. Wang, G. Rivas, C. Parrado, X. Cai, M. Flair, Talanta 44 (1997) 2003. [13] J. Wang, E. Palecek, P. Nielsen, G. Rivas, X. Cai, H. Shiraishi, N. Dontha, D. Luo, P. Farias, J. Am. Chem. Soc. 118 (1996) 7667. [14] J. Cheng, E. Sheldon, L. Wu, L. Gerrue, J. Carrino, M. Heller, M. O’Connell, Nat. Biotech. 16 (1998) 541. [15] J. Wang, J. Fernandes, L. Kubota, Anal. Chem. 70 (1998) 3699. [16] J. Wang, A. Kawde, E. Sahlin, Analyst 125 (2000) 5. [17] K. Millan, S.R. Mikkelsen, Anal. Chem. 65 (1993) 2317. [18] J. Wang, X. Cai, G. Rivas, H. Shiraishi, Anal. Chim. Acta 326 (1996) 141. [19] K.M. Millan, A. Saraulo, S.R. Mikklesen, Anal. Chem. 66 (1994) 2943. [20] J. Wang, G. Rivas, X. Cai, M. Chicharro, C. Parrado, N. Dontha, A. Begleiter, M. Mowat, E. Palecek, P.E. Nielsen, Anal. Chim. Acta 344 (1997) 111. [21] K. Hashimoto, K. Ito, Y. Ishimori, Anal. Chem. 66 (1994) 3830. [22] G. Marrazza, I. Chianella, M. Mascini, Anal. Chim. Acta 387 (1999) 297. [23] G. Marrazza, G. Chiti, M. Mascini, M. Anichini, Clin. Chem. 46 (2000) 31.

71

[24] S. Takenaka, K. Yamashita, M. Takagi, Y. Uto, H. Kondo, Anal. Chem. 72 (2000) 1334. [25] T. Ihara, M. Nakayama, M. Murata, K. Nakano, M. Maeda, Chem. Commun. (1997) 1609. [26] V. Kertez, N. Whittemore, G. Inamati, M. Manoharan, P. Cook, D. Baker, J.Q. Chambers, Electroanalysis 12 (2000) 889. [27] R.M. Umek, S.S. Lin, Y.P. Chen, B. Irvine, G. Paulluconi, V. Chan, Y. Chong, L. Cheung, J. Vielmetter, D.H. Farkas, Mol. Diag. 5 (2000) 321. [28] J. Wang, D. Xu, R. Polsky, Langmuir 17 (2001) 5739. [29] T. de Lumley, C. Campbell, A. Heller, J. Am. Chem. Soc. 118 (1996) 5504. [30] F. Patolsky, E. Katz, A. Bardea, I. Willner, Langmuir 15 (1999) 3703. [31] L. Alfonta, A.K. Singh, I. Willner, Anal. Chem. 73 (2001) 91. [32] M. Wojcienchowski, R. Sundseth, M. Moreno, R. Henkens, Clin. Chem. 45 (1999) 1690. [33] F. Azek, C. Grossiord, M. Joannes, B. Limoges, P. Brossier, Anal. Biochem. 284 (2000) 107. [34] J. Wang, D. Xu, A. Erdem, R. Polsky, M. Salazar, Talanta, in press. [35] A. Shipway, E. Katz, I. Willner, Chem. Phys. Chem. 1 (2000) 19. [36] J. Wang, G. Rivas, J. Fernandes, J.L. Paz, M. Jiang, R. Waymire, Anal. Chim. Acta 375 (1998) 197. [37] D.H. Johnston, K. Glasgow, H.H. Thorp, J. Am. Chem. Soc. 117 (1995) 8933. [38] J. Wang, A. Nasser, A. Erdem, M. Salazare, Analyst, in press. [39] E. Palecek, S. Biiova, L. Havran, R. Kizek, A. Miaulkova, F. Jelen, Talanta, in press. [40] P. Ropp, H.H. Thorp, Chem. Biol. 6 (1999) 599. [41] M.E. Napier, K. Mikulecky, K. Scaboo, A. Eckhardt, N. Baron, N. Popovich, M. Geladi, Clin. Chem. 46 (2000) A207. [42] S. Kelley, E. Boon, J.K. Barton, M. Jackson, M. Hill, Nucleic Acids Res. 27 (1999) 4830. [43] E.M. Boon, D. Ceres, T. Drummond, M. Hill, J.K. Barton, Nat. Biotech. 18 (2001) 1096. [44] P. Aich, S. Labiuk, L. Tari, L. Delbaere, W. Roesler, K. Falk, R. Steer, J. Lee, J. Mol. Biol. 294 (1999) 477. [45] H. Korri-Youssoufi, F. Garnier, P. Srivtava, P. Godillot, A. Yassar, J. Am. Chem. Soc. 119 (1997) 7388. [46] J. Wang, M. Jiang, A. Fortes, B. Mukherjee, Anal. Chim. Acta 402 (1999) 7. [47] C. Siontorou, D. Nikolelis, P. Piunno, U. Krull, Electroanalysis 9 (1997) 1067. [48] J. Hianik, M. Fajkus, B. Sivak, I. Rosenberg, P. Kois, J. Wang, Electroanalysis 12 (2000) 495. [49] H. Aoki, P. Buhlmann, Y. Umezawa, Electroanalysis 12 (2000) 1272. [50] F. Patolsky, A. Lichtenstein, I. Willner, Angew. Chem. Int. Ed. 39 (2000) 940. [51] C. Berggren, P. Stalhandske, J. Brundell, G. Johansson, Electroanalysis 11 (1999) 156.