Modified Rare Earth Semiconductor Oxide As a New Nucleotide Probe

J. Phys. Chem. B 2006, 110, 25633-25637

25633

Modified Rare Earth Semiconductor Oxide As a New Nucleotide Probe S. Shrestha,† C. E Mills,‡ J. Lewington,‡ and S. C. Tsang*,† The Surface and Catalysis Research Centre, School of Chemistry, UniVersity of Reading, Reading, RG6 6AD, United Kingdom, and Smiths Detection, 459 Park AVenue, Bushey, Watford, Herts WD23 2BW, United Kingdom ReceiVed: July 23, 2006; In Final Form: October 3, 2006

Recent rapid developments in biological analysis, medical diagnosis, pharmaceutical industry, and environmental control fuel the urgent need for recognition of particular DNA sequences from samples. Currently, DNA detection techniques use radiochemical, enzymatic, fluorescent, or electrochemiluminescent methods; however, these techniques require costly labeled DNA and highly skilled and cumbersome procedure, which prohibit any in-situ monitoring. Here, we report that hybridization of surface-immobilized single-stranded oligonucleotide on praseodymium oxide (evaluated as a biosensor surface for the first time) with complimentary strands in solution provokes a significant shift of electrical impedance curve. This shift is attributed to a change in electrical characteristics through modification of surface charge of the underlying modified praseodymium oxide upon hybridization with the complementary oligonucelotide strand. On the other hand, using a noncomplementary single strand in solution does not create an equivalent change in the impedance value. This result clearly suggests that a new and simple electrochemical technique based on the change in electrical properties of the modified praseodymium oxide semiconductor surface upon recognition and transduction of a biological event without using labeled species is revealed.

1. Introduction The transduction of biorecognition event into a physically measurable value presents a great interest because of the wide applications in food, clinical biological, and environmental fields. There are many techniques available for the detection purposes, but the main challenge is to develop the most suitable transducer for the particular type of biological interaction. For example, amperometric transducers are particularly well adapted for coupling with redox enzyme systems because recognition provokes consumption or production of redox active species that are readily detected.1 However, it is rather difficult to deal with biological systems as antibody/antigen or DNA hybridization events since they are based on complex molecular affinity. Currently, for the DNA detection, common techniques including radiochemical, enzymatic, fluorescent, or electrochemiluminescent methods have been developed. However, all these techniques require expensive labeled DNA and a rather cumbersome and highly skilled procedure. On the other hand, potential advantages of using a simple electrochemical impedance spectroscopy (EIS) method to study these biomolecular reactions are particularly noted. However, as far as we are aware, there is a very limited research that has actually taken place in this area. The key problem limiting the development is the intrinsic poor sensitivity of the transducer for biosensing processes,2 despite the fact that various transducer materials including glass, silicon, gold, platinum, and carbon paste electrodes have been explored.3 In addition, the problems of instability of the output signal and poor signal transformation have yet to be solved.4 A recent extensive screening exercise * Author to whom correspondence should be addressed. Tel: 44(0)1189316346; fax: 44(0) 1189316632; e-mail: [email protected]. † University of Reading. ‡ Smiths Detection.

to search promising transducer materials for various biomolecule attachments has been initiated.5 Our interest in this work is to develop rare earth oxide, especially praseodymium oxide, as a perspective semiconductive sensing material because of its excellent potentiality as an alternative material to silicon. It gives many unique properties such as high dielectric constant (κ ∼ 26-30), large band gap (3.9 eV), and high electron affinity (0.962 ( 0.024 eV),6 and it is relatively easy to present as a thin film.7 More importantly, this semiconductor oxide material intrinsically possesses a relatively high electrical conductivity at low temperature.7,8 For example, the electron-hopping mechanism between mixed valence states in Pr6O11 lattice accounts for its high electrical conductivity.8,9 The higher sensitivity/selectivity of the praseodymium oxide compared to tin oxide based sensors for detection of ethanol vapor has recently been demonstrated.10 The major objective of this research was to carry out a feasibility study of using Pr6O11 as new sensing material for the development of a novel electrochemical DNA-sensor. Physical adsorption of a single-stranded oligonucleotide on the oxide surface was first carried out. Electrochemical impedance spectroscopic (EIS) study was then employed to detect the shift in impedance because of subsequent hybridization with the complementary base pairs from solution using a two-probe impedance spectroscopy.11 This short paper reports the initial findings. 2. Experimental 1. Materials. Synthetic oligonucleotide primers, probe oligonucleotides, oligo-1 (AAC-GAT-CGA-GCT-GCA-A), complementary oligonucleotides, oligo-2 (TTG-CAG-CTC-GAT-CGTT), and noncomplementary oligonucleotides, oligo-3 (CGTACC-AAG-ATG-AAC-G), were purchased from Eurogentec S. A., Belgium and were used after diluting them to a concentration

10.1021/jp0646729 CCC: $33.50 © 2006 American Chemical Society Published on Web 11/23/2006

25634 J. Phys. Chem. B, Vol. 110, No. 51, 2006

Shrestha et al.

Figure 1. (a) A designed nylon cell for the in-situ AC impedance measurement of a praseodymium oxide deposited on ITO. (b) Top view of the cell lid.

of 100 µM in sterilized deionized water. Deionized water was purified using a Milli-Q purification system (Millipore, resistance of 18 MΩcm). Sodium chloride (biomolecular grade) was obtained from Sigma Biochemicals and was used as buffer for AC impedance measurements. Tin-doped indium oxide (ITO) on glass electrodes was obtained from Image Optics Components Ltd, United Kingdom. Praseodymium oxide (Pr6O11) material was deposited on ITO-electrodes by applying a negative sweeping voltage (cathodic electrodeposition) to the aqueous solution containing Pr(NO3)3 and H2O2 using cyclic voltametry, followed by annealing the film at 500 °C for an hour prior to adsorption of oligonucleotide.7 Detailed microscopic characterization of the praseodymium oxide on ITO surface was conducted before the physical adsorption of oligonucelotide. Scanning electron microscopy (SEM) was conducted using a Cambridge Stereoscan 360. Atomic force microscopy (AFM) was carried out using an Explorer AFM in noncontact mode. 2. Electrochemical Measurements. AC impedance measurement was performed using an impedance analyzer (Solartron SI 1260/gain-phase analyzer). The impedance measurement was taken by sweeping the frequency between 106 to 0.5 Hz at voltage of 25 mV and constant temperature of 43 °C employing the two-probe system. Figure 1a shows the 1.5 cm × 2.5 cm × 2.5 cm Nylon cell designed for the AC impedance measurement. At the bottom of the cell, there was a circular groove of 1-cm diameter to place a small magnetic bar as stirrer while performing the measurement. There were two cut slots of 1 cm apart where the platinum electrode (1.5 cm2) and coated ITO electrode were located. The working electrode was immersed in 2-cm depth of 1 mM sodium chloride buffer (9 mL) solution as an electrolyte after the careful exclusion of any entrapped air bubbles from the solution. Nylon screws on the lid were used to fix the ITO slide in position. Sticky copper tape was used at the end of ITO slide to maintain an electrical contact. Then, the coated ITO and platinum electrodes were connected to the impedance analyzer via two coaxial cables. The nylon cell was fitted with a lid, which was screwed at four corners with holes for wiring up the platinum and ITO electrodes. Silicon rubber

septa were used to seal any holes/gaps around the electrical cell and contacts. A lining of thin rubber gasket was also placed round the edge at the bottom of the cell lid for the isolation of the cell from the environment as shown in Figure 1b. Most part of the cell was immersed in a water bath kept at constant temperature of 43 °C. Buffer solution inside the cell and water in the exterior water bath were stirred continuously using a magnetic stirrer to achieve an isothermal environment. N2 was purged into the buffer solution for at least 15 min for degassing purpose before any measurement. Impedance measurements of oxide-deposited ITO were taken at a constant temperature of 43 °C using the Nylon cell, as previously described above. Four microliters (giving the concentration of 0.044 µM in the cell) of oligo-1 was injected into the buffer at time zero. Three consecutive impedance measurements were taken each at 15-min intervals after the injection. In general, no further change in the impedance value was recorded for the last two measurements, which implied that a sufficient time for the adsorption of oligo-1 onto the oxide surface was ensured. After the measurements, the same quantity of the oligonucleotides was added and impedance measurements were repeated intermittently until reaching the surface saturation at 68 µL (0.748 µM). The oxide-deposited ITO presaturated with oligo-1 as a probe oligonucleotide was then treated with the complementary (oligo2) and noncomplementary (oligo-3) oligonucleotides. Fifteen microliters (0.165 µM) of oligo-2 or oligo-3 was injected into the buffer at time zero, and three consecutive impedance measurements were taken at 15 min after the injection. After the measurements, the same quantity of the oligonucleotides was added and impedance measurements were repeated intermittently until a total amount of oligo-2 or oligo-3 reached 300 µL (3.30 µM). 3. Results and Discussion Topographical Features of Pr6O11 on the ITO. We have shown earlier that an ultrathin layer of praseodymium oxide in the form of Pr6O11 with a high internal surface area can be deposited on tin-doped indium oxide surface (ITO) by applying

Modified Rare Earth Semiconductor Oxide

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25635

Figure 2. SEM images showing the topographical features of 300-500 nm Pr6O11 particles deposited on the ITO surface (left) top view and (right) side view after the electrochemical and heat treatments.

Figure 3. AFM image showing the 3-D view of Pr6O11 particles on ITO surface.

a negative sweeping voltage (cathodic electrodeposition) to the aqueous solution containing Pr(NO3)3 and H2O2 using cyclic voltammetry, followed by annealing the film at 500 °C for 1 h. AC impedance measurements revealed that the deposited film shows a much higher electrical conductivity than the pure ITO.7 In this work, we carried out close examination of the deposited praseodymium oxide film on ITO before the material was used as the host for the oligonucleotide immobilization. As seen from the SEM and AFM images of Figures 2 and 3, this film is assembled with a monolayer coverage of 300-500 nm spherical praseodymium oxide particles packed closely on the ITO surface. Impedance Measurement upon Adsorption of Probe Oligonucleotide. Figure 4 shows the typical semicircle complex impedance spectrum (with real and imaginary components) of Pr6O11 deposited ITO indicative of its semiconducting properties. Upon addition of probe oligonucleotide (oligo-1), the shift of semicircle to the right (increase in impedance values at all frequencies) is clearly observed. Figure 5 compiles the detailed percent change in the real impedance values (collected at 0.5 Hz frequency) of the Pr6O11 deposited ITO according to the volume of probe oligonucleotide added. Again, a progressive but significant increase in impedance values upon each addition of oligo-1 until reaching a saturation plateau at about 68 µL is clearly observed. The increase in impedance is attributed to the induced charge separation of the underlying Pr6O11 material upon its adsorption of probe oligonucleotide, which may be related to the high affinity of the phosphate backbone of the oligonucleotide to the surface resulting in localization of electron carriers in semiconducting oxide.12-14 In addition, the counterions from solution might have associated with the oligo (in the Debye Hu¨kel approximation) enhancing the charge separation of the probe oligonucleotide adsorbed on the oxide

Figure 4. A typical right shift of the impedance spectra after addition of 68 µL probe oligonucleotide (oligo-1) to Pr6O11/ITO (obtained using a sweeping frequency from106 Hz to 0.5 Hz with an applied voltage of 25 mV).

Figure 5. The change in real impedance (collected at 0.5 Hz) according to the increase in volume of oligo-1.

surface.15,16 It is also envisaged that the surface of the Pr6O11 on ITO was approaching a total coverage of the probe oligonucleotide when 68 µL was added to the Nylon cell. Without adding oligonucleotide (adding water instead), no enhancement in impedance value was observed. Table 1summarizes the percentage changes in the impedance values for four repeated measurements with each time using a fresh electrode and new oligo-1 solution (68 µL). The average change in the impedance was found to be +17.34% (see Table 1). On the other hand, we recorded a slight increase in

25636 J. Phys. Chem. B, Vol. 110, No. 51, 2006

Shrestha et al.

TABLE 1: Percent Change in the Impedance Value at 43 °C, 0.5 Hz, and Applied Voltage of 25 mV on Frequent Addition of Probe Oligonucleotide to the Pr6O11/ITO Electrode

trial

% increase in impedance after addition of 68 µL oligo-1 (data collected at 43 ˚C, 0.5 Hz, 25 mV)

1 2 3 4

16.59 18.72 18.72 15.62

Figure 7. Percent change in impedance (collected at 0.5 Hz) of Pr6O11/ ITO presaturated with adsorbed probe oligo-1 upon a subsequent addition of complementary oligo-2 as compared to the noncomplementary oligo-3.

Figure 6. A left shift of the impedance spectra of the preadsorbed probe oligo-1 on Pr6O11/ITO upon addition of 300 µL complementary oligo-2 (obtained using a sweeping frequency from106 Hz to 0.5 Hz with an applied voltage of 25 mV).

impedance in our control experiment (no oligonucleotide) because of a gradual evaporation of solution during the experiment, but the value would not be higher than +1.42%. Impedance Measurement upon Interaction of the SurfaceAdsorbed Probe Oligonucleotide with its Complementary and Noncomplementary Oligonucleotides. Figure 6 shows the complex impedance spectrum of Pr6O11 preadsorbed with probe oligonucleotide (oligo-1). It is very interesting to note the shifts in semicircle to the left (decrease in impedance values at all frequencies) upon addition of complementary oligonucleotide (oligo-2). It is attributed to the change of the relief of charge separation at the underneath of the semiconductive Pr6O11 oxide, previously inflexed by the surface adsorption of probe oligonucleotides, through hybridization with its complimentary base pairs (double-stranded). It has previously been observed that the overall charge density on the DNA duplex is significantly lower than the single-stranded entity as the charge is effectively neutralized by the counterions located between the duplex strands.12 Besides, the DNA structure may facilitate a better electron transfer, particularly through the π-stacks of the double helix leading to its semiconductive nature for a longer range charge transport.17 Consequently, we show that the formation of double-stranded nucleic acids at the vicinity of the semiconducting oxide reduces the overall impedance of the oxide. Figure 7 compiles the detailed percent change in the real impedance values of the preadsorbed probe oligonucleotide on Pr6O11 according to the volume of complementary oligo-2 added (incremented by 15 µL for each point). The progressive decrease in impedance values (within 10% percent of measurement error at each point) upon each addition of oligo-2 until reaching a saturation plateau at about 300 µL is clearly observed. As seen, there is no equivalent effect on impedance upon addition of

Figure 8. Percent change in impedance (collected at 0.5 Hz) versus time of Pr6O11/ITO presaturated with adsorbed probe oligo-1 upon additions of complementary oligo-2 of 100 µL, 200 µL, and 300 µL.

the noncomplementary oligo-3, but some minor degree of fluctuation in impedance signal is detected (this reproducible fluctuation could be due to the instantaneously perturbation on surface polarity by a sudden addition of the noncomplementary oligo to the probe oligo, but the slight drop in impedance would gradually be reverted). A significantly larger quantity of oligo-2 (300 µL or 3.30 µM) than the maximum quantity of oligo-1 present on the surface (68 µL or 0.748 µM) is required to reach the saturation point (molar ratio of 4.4). In addition, we observed a much longer equilibrium time for the oligo-2 to reach a steady impedance signal than oligo-1 which gave almost immediate response upon its addition to the fresh oxide surface. Figure 8 clearly shows the long equilibrium times (>20 min) required to achieve the steady signals with the additions of 100 µL, 200 µL, and 300 µL. These observations clearly suggest that it takes a relatively long time for the hybridization of the oligo-2 with the surface-adsorbed oligo-1, and the interactions appear to be in a dynamic state deviating from the ideal 1:1 molar ratio. A simple equivalent circuit model (see Scheme 1) has been built to represent the electrical interfaces between the Pr6O11 on ITO with and without oligonucleotide adsorption and its

Modified Rare Earth Semiconductor Oxide

J. Phys. Chem. B, Vol. 110, No. 51, 2006 25637

SCHEME 1: The Envisaged Electrical Interfaces between Cu (Connecting Tape), ITO, Deposited Pr6O11, and the Outermost Biolayers and the Corresponding Equivalent Circuit Modela

a C3 and R3 represent the interfacial capacitance and resistance of ITO, and C2 and R2 represent the interfacial capacitance and resistance of praseodymium oxide which are critically affected by oligonucleotides.

TABLE 2: Simulated Resistance (RA, R1, R2, and R3) and Capacitance Values (C1, C2, and C3)a sequential treatments oligo-1 oligo-2b

RA

R1

(6.78 × Ω) +10.38% -4.48%

(1.3 × 10 Ω) -4.92% +1.12%

103

C1 3

R2

(2.13 × 10 F) -33.95% -11.73% -5

C2

(27.82 × 10 Ω) -69.74% -16.94% 5

R3

(5.48 × 10 F) -26.11% +30.27% -5

C3

(1.81 × 10 Ω) +4.41% -6.72% 6

(0.43 × 10-5 F) +4.42% -3.81%

a On the basis of the equivalent circuit model in Scheme 1 using the entire impedance frequency range and the change of these values (in percent) after the sequential treatments with oligonucleotides. b Change in resistance and capacitor values are referenced to the preadsorbed oligo-1 sample. The significant changes in resistances and capacitors of greater than 15% are bolded.

biointeraction with complementary oligonucleotide. As seen from Table 2, there has been a significant change in the capacitor and resistance values especially the C1, C2, and R2 (>15% change) when the praseodymium oxide film is modified with the probe oligonucloetide. This confirms that the electrical properties of the underlying praseodymium oxide film can be greatly altered through surface adsorption of oligonucleotide indicative of the ultrasensitivity of the film toward biomolecule as a useful biosensor surface. Also, a major change in capacitance (C2) and resistance (R2) was recorded as compared to the adsorbed oligo-1 when the complementary oligo-2 was added. This again reflects the difference in the charge separation between surface-adsorbed single-stranded oligonucleotide and DNA duplex. 4. Conclusions It is shown, for the first time, that the semiconducting praseodymium oxide particles as a thin film can be developed as a direct electrochemical nucleic acid sensor without the use of labeled oligonucletides at slightly elevated temperatures above ambient conditions, which can provide a suitable surface for the detection of oligonucelotides. We believe that this new method of utilizing the unique electrical properties of praseodymium oxide combined with surface modification with singlestranded probe oligonucleotide would greatly simplify the detection of complementary oligonucelotides in solution by using electrochemical means. In this work, a measurable shift in electrochemical impedance spectrum because of change in the interfacial electrical properties of the adsorbed singlestranded nucleic acid and its complementary partner can be

clearly observed. The change in resistance and capacitor values of the modified oxide causing the shift is quantified by our simulation work using equivalent circuit models. References and Notes (1) Guan, J.-G.; Miao, Y.-Q.; Zhang, Q.-J. J. Biosci. Bioeng. 2004, 97 (4), 219-226. (2) Sadik, O. A.; Xu, H.; Gheorghiu, E.; Andreescu, D.; Balut, C.; Gheorghiu, M.; Bratu, D. Anal. Chem. 2002, 74, 3142-3150. (3) Zhang, S.; Wright, G.; Yang, Y. Biosens. Bioelectron. 2000, 15, 273. (4) Opekar, F.; Stulik, K. Crit. ReV. Anal. Chem. 2002, 32 (3), 253259. (5) Evans, J. Chem. World 2006, 3 (7), 42-46. (6) Aspinall, H. C.; Gaskell, J.; Williams, P. A.; Jones, A. C.; Chalker, P. R.; Marshall, P. A.; Bickley, J. F.; Smith, L. M.; Critchlow, G. W. Chem. Vap. Deposition 2003, 9 (5), 235-238. (7) Shrestha, S.; Marken, F.; Elliot, J.; Yeung, C. M. Y.; Mills, C. E.; Tsang, S. C. J. Electrochem. Soc. 2006, 153 (7), C517-C520. (8) Honig, J. M. J. Chem. Educ. 1966, 43, 76-82. (9) Chandrasekar, G. V.; Mehrotra, P. N.; SubbaRao, G. V.; SubbaRao, E. C.; Rao, C. N. R. Trans. Faraday Soc. 1966, 63, 1295-1301. (10) Tsang, S. C.; Bulpitt, C. Sens. Actuators, B 1998, 52, 226-235. (11) Weimer, U.; Go¨pel, W., A. C. Sens. Actuators, B 1995, 26-27, 13-18. (12) Nooney, M. G.; Campbell, A.; Murrell, T. S.; Lin, X.-F.; Hossner, L. R.; Chusuei, C. C.; Goodman, D. W. Langmuir 1998, 14, 2750-2755. (13) Gao, W.; Dickinson, L.; Grozinger, C.; Morin, F. G.; Reven, L. Langmuir 1996, 12, 6429-6435. (14) Cao, G.; Hong, H.-G.; Mallouk, T. E. Acc. Chem. Res. 1992, 25, 420-427. (15) Manning, G. S. Q. ReV. Biophys. 1978, 11 (2), 179-246. (16) Record, M. T.; Anderson, C. F.; Lohman, T. M. Q. ReV. Biophys. 1978, 11, 103-178. (17) Kelley, S. O.; Jackson, N. M.; Hill, M. G.; Barton, J. K. Angew. Chem., Int. Ed. 1999, 38 (7), 941-945.