Fabrication of Electrochemical DNA Sensors on Gold-Modified

Feb 24, 2014 - Fabrication of single Pt@Au nanowire electrodes for monitoring hydrogen peroxide released from living cells. Yong Liu , Yaoyao Zhang , ...
0 downloads 7 Views 1MB Size
Letter pubs.acs.org/ac

Fabrication of Electrochemical DNA Sensors on Gold-Modified Recessed Platinum Nanoelectrodes S. Ehsan Salamifar and Rebecca Y. Lai* Department of Chemistry, University of Nebraska−Lincoln, Lincoln, NE 68588-0304, United States S Supporting Information *

ABSTRACT: We report the use of gold-modified recessed platinum (Pt) nanoelectrodes in the fabrication of linear and stem-loop probe-based electrochemical DNA (E-DNA) sensors. Pt nanoelectrodes with a radius less than 10 nm were reproducibly fabricated using an optimized laser pulling technique. Prior to sensor fabrication, the nanoelectrode was electrochemically etched to create a recessed nanopore, followed by electrodeposition of gold into the nanopore using either cyclic voltammetry or constant potential amperometry. Both techniques enabled controlled deposition of gold into the nanopores, resulting in a nanostructured gold electrode with a well-defined surface area. In addition, we systematically determined the optimal experimental condition for DNA probe immobilization and target interrogation. The electron transfer rate constants of methylene blue, as determined using alternating current voltammetry, were found to be much higher than those obtained from E-DNA sensors fabricated on conventional macroscale electrodes. While this unique phenomenon requires further investigation, our results clearly show that these gold-modified nanoelectrodes can be used as substrates for this class of electrochemical biosensors.

F

sensors, with nanoelectrodes can potentially be used for realtime sensing inside living cells.16,17 Fabrication of an E-DNA sensor on a Pt nanoelectrode involves six defined steps (Figure 1). Details of the fabrication procedure are included in the Supporting Information. In brief,

abrication of nanoelectrodes for analytical applications has been the subject of numerous studies in the past decade.1 To date, various methods have been employed to produce nanoelectrodes of different shapes. But independent of the geometry, a nanoelectrode is often classified as a voltammetric electrode with at least one dimension below 100 nm.2 The unique properties of nanoelectrodes such as high masstransport rate, fast electrochemical responses, rapid potential switching, small RC constants, and the ability to perform measurements in highly resistive solutions, have resulted in their application in a wide range of research areas, including fundamental electrochemical research, neurobiology and single cell studies, as well as in scanning electrochemical microscopy (SECM).2−6 More recently, the use of miniaturized sensors, specifically for the detection of nucleic acids and diagnostic proteins, has attracted substantial attention.7,8 Many previously reported electrochemical nucleic acid sensing platforms can, in fact, benefit from miniaturization.9 The electrochemical DNA (E-DNA) sensor developed by Fan et al. is one of the sensors that can potentially be used with miniaturized electrodes for in vivo studies, owing to its “reagentless” nature and its compatibility with realistically complex sample matrices such as blood serum.10,11 However, most E-DNA sensors were fabricated on conventional macroscale electrodes; sensor performance on a nanoscale electrode has yet to be explored.10−13 Thus, the focus of this study is not only on the fabrication of gold-plated recessed Pt nanoelectrodes, but also on the use of these electrodes in the fabrication of E-DNA sensors.14,15 The combination of this versatile sensing platform, which also includes the electrochemical aptamer-based (E-AB) and peptide-based (E-PB) © 2014 American Chemical Society

Figure 1. Shown are the steps used in the fabrication of an E-DNA sensor on a Pt nanoelectrode. Step 1: fabrication of the electrode using the laser puller. Step 2: electrode polishing using the beveller. Step 3: electrochemical etching of the electrode. Step 4: deposition of gold into the nanopore. Step 5: immobilization of DNA probes and passivating diluents onto the gold surface. Step 6: hybridization of the DNA probes with the full-complement target DNA and sensor regeneration. Received: November 25, 2013 Accepted: February 24, 2014 Published: February 24, 2014 2849

dx.doi.org/10.1021/ac403816h | Anal. Chem. 2014, 86, 2849−2852

Analytical Chemistry

Letter

we first used an optimized pulling technique to fabricate the Pt nanoelectrode (step 1). It was then mechanically polished using the beveller to expose the Pt wire imbedded in the quartz capillary (step 2). Next, part of the Pt wire was electrochemically etched to create a nanopore, forming a recessed electrode (step 3). Gold was subsequently deposited into the nanopore using either cyclic voltammetry (CV) or constant potential amperometry (CP) (step 4). The electrodeposited gold was used as the substrate for the E-DNA sensor. To fabricate the EDNA sensor, we immobilized a thiol- and methylene blue (MB)-modified DNA probe onto the electrode surface. The sensor was further passivated with 6-mecapto-1-hexanol (C6OH) (step 5). Both linear (LP) and stem-loop (SLP) DNA probes were used in this study. Last, the as-fabricated E-DNA sensor was interrogated with the full-complement target DNA (step 6). While the ability to use these electrodes for sensor applications is important, one of the focuses of this study is the fabrication of the nanoelectrode itself. Electrochemical characterization is crucial in determining the active area of the nanoelectrode. To calculate the radius of a polished disk-shaped electrode (a), we used the following equation: iss = 4nFDCa

despite having a sharper end and being longer in length, are significantly more flexible and less fragile, thereby allowing them to be used for in vivo studies with minimal risk of breakage. Since gold is the most commonly used substrate for E-DNA sensor fabrication, modification of the Pt nanoelectrode with gold is thus necessary. Fabrication of gold microelectrodes can be accomplished using a coiled heater; however, fabrication of a gold nanoelectrode using the laser puller is extremely challenging due to the large difference between the melting point of gold (1064 °C) and quartz (1710 °C). In this case, modification of the Pt nanoelectrode was accomplished using the method reported by Jena et al., but the main difference is that by using the optimal parameters described here, the resultant electrode is long, tapered, and surrounded by a robust layer of quartz.19 In brief, prior to gold deposition, the nanoelectrode was electrochemically etched by applying an AC waveform with a 4 V amplitude at 80 Hz in an aqueous solution containing 10% hydrochloric acid and 1.5 M calcium chloride.20 This 10 s etching process resulted in a recessed nanoelectrode with a nanometer-sized pore. CVs of ferrocenemethanol (FcM) recorded during various stages of the etching process are shown in Figure S2 of the Supporting Information. Etching time longer than 10 s did not produce a substantial change in FcM current (data not shown). Prior to E-DNA sensor fabrication, we deposited gold into the recessed Pt nanoelectrode using CV. Figure S3A of the Supporting Information shows the CVs recorded using a Pt nanoelectrode in a solution containing 1.2 mg mL−1 gold chloride at a scan rate of 20 mV s−1. These voltammograms were collected to monitor the amount of gold deposited during each cycle. To further verify successful deposition of gold, we recorded a CV of the electrode in 0.05 M H2SO4; the presence of gold was verified by its signature oxide formation and reduction peaks (Figure S3B of the Supporting Information). Since the reduction charge per gold unit area is 390 μC, the microscopic surface area was calculated to be 750 μm2.21 Furthermore, CVs of the electrode recorded in 4 mM FcM during each step of the fabrication process were also used to assess the electrode size (Figure S4 and S5 of the Supporting Information). Using iss from the CVs and eq 1, the radius (a) of the Pt electrode before electrochemical etching was found to be 32 nm. A large decrease in iss was evident after etching, presumably because of the limited diffusion of FcM into the nanopore. As expected, iss increased by ∼3500 times after the deposition of gold. This large increase in iss could be attributed to the deposition of porous and dendritic gold structures into the nanopore. To roughly estimate the size of the electrodeposited gold structure using iss, we used eq 2, a formula that is applicable for this system, with the assumption that the gold structure assumes a hemispherical geometry.

(1)

where iss is diffusion-limited steady-state current of the redox mediator, n is the number of electrons transferred per molecule, F is Faraday’s constant, and D and C are the diffusion coefficient and bulk concentration of the redox molecule, respectively. Using the procedure described in the Supporting Information, we successfully fabricated a sharp, long, and tapered Pt electrode with a radius less than 1 nm, as calculated using the iss value obtained from the voltammogram (Figure 2).

Figure 2. (A) SEM image of a polished Pt nanoelectrode. (B) A CV recorded using a Pt nanoelectrode at a scan rate of 20 mV s−1 in 4 mM FcM in 0.1 M KCl.

To our best knowledge, this is the smallest reported size for an individual nanoelectrode fabricated using a laser pulling technique.18 Due to the uniform cylindrical shape of the Pt wire imbedded inside the tapered portion of the quartz capillary, the same electrode can be used for multiple experiments without exhibiting a significant change in the electrode area, even after successive polishing steps. However, our attempts to use scanning electron microscopy (SEM) and transmission electron microscopy (TEM) to further characterize the electrode, especially when the electrode radius was less than 10 nm, was unsuccessful due to challenges described elsewhere.2 SEM images of larger electrodes, often a result of excess polishing, can be used to confirm the predicted geometry of these electrodes. As can be seen in Figure S1 of the Supporting Information, the Pt wire was properly sealed and located at the center of the quartz capillary. These electrodes,

iss = 2πnFDCa

(2)

where all parameters are exactly the same as in eq 1. The surface area of the deposited gold, as determined using the radius (a) obtained from eq 2, was 350 μm2. The difference in surface area obtained from integration of the gold oxide reduction peak and using eq 2 can be attributed to the morphology of electrodeposited gold. The material deposited using CV could be porous and dendrite-like, rather than having a hemispherical shape with a uniform outer layer, an assumption that eq 2 is based on. Furthermore, depending on the actual structure and pore size of the deposited material, 2850

dx.doi.org/10.1021/ac403816h | Anal. Chem. 2014, 86, 2849−2852

Analytical Chemistry

Letter

Figure 3. AC voltammograms of the (A) LP and (B) SLP E-DNA sensors before hybridization, after hybridization with a 2 μM full-complement target, and after regeneration with 4 M GHCl.

despite the high porosity, the deposited gold structure is wellsuited for use as the E-DNA sensor substrate. Two types of E-DNA sensors, including the LP and SLP designs, were used in this study (Figure S8 of the Supporting Information). The signaling mechanism of these sensors has been reported in the literature and is thus not the main focus of the current study.23 For both LP and SLP sensors, in the absence of the target, electron transfer between MB and the electrode is efficient, giving rise to the high MB current. Upon hybridization to the target, the change in probe conformation and/or reduction in probe flexibility limits MB access to the electrode, which is signified by a large decrease in MB current.23 Figure 3 shows the AC voltammograms of both sensors before hybridization, after hybridization with 2 μM fullcomplement target, and after sensor regeneration with 4 M guanidine hydrochloride (GHCl). In the absence of the target, a well-defined AC voltammetric peak was observed at approximately −0.3 V (vs Ag/AgCl) for both sensors; this potential is consistent with the reduction potential of MB under similar experimental conditions. The lack of change in the MB current with time further confirmed successful immobilization of DNA probes onto the gold surface. Due to the low MB current observed in CV, the electron transfer kinetics were analyzed using ACV. On the basis of the method developed by Wooster and Creager,24,25 the heterogeneous electron transfer rate (ks) of MB for the LP sensor was ∼600 s−1 (Figure S9A of the Supporting Information). For the SLP sensor, the value was even higher (∼2200 s−1) (Figure S9B of the Supporting Information). While the fitting is not optimal because of the lack of current information at extremely low and high frequencies, these values are substantially higher than those obtained from the same sensors fabricated on macroscale gold electrodes without nanostructures (∼50 s−1 for LP; ∼280 s−1 for SLP) (Figure S9, panels C and D, of the Supporting Information). Overall, these results suggest more facile electron transfer kinetics between the MB label on the DNA probe and the underlying nanostructured gold electrode. However, it could also be due to the inadequate surface passivation by C6OH. Further investigation is necessary to elucidate the reasons behind this observation. Despite the differences in electron transfer kinetics, both sensors responded well to the target. A large decrease in MB current was observed in the presence of a 2 μM fullcomplement target, indicating successful hybridization (Figure 3). The signal suppression was 98% and 81% for the LP and

access of FcM to all parts of the electrode could be obstructed. The use of two different methods (i.e., gold oxide formation/ reduction and FcM electrochemistry) to determine surface area is advantageous; the ratio of the two values can be used to estimate the extent of porosity in the deposited material. Further attempts were made to improve the morphology of gold deposited using CV. For example, we evaluated the effect of scan rate and the number of scans on the deposited material; these attempts, however, were unsuccessful, presumably attributed to poor adhesion of the deposited material onto the electrode surface. Fabrication of an E-DNA sensor on this material was unsuccessful; the MB current observed upon DNA probe immobilization was extremely low (data not shown). Thus, we shifted our focus toward the use of CP for gold deposition. Deposition of gold using CP is more common, and more importantly, it has been shown to produce gold structures suitable for use as an E-DNA sensor substrate.13 Figure S6A of the Supporting Information shows the current−time curve recorded while the electrode was held at 0.0 V (vs Ag/AgCl) for 300 s. The increase in current was minimal at the beginning of the process, indicating a slow deposition process. However, a sharp increase in the slope of the current−time curve was observed around 25 s, indicating a much faster deposition rate. This drastic increase in deposition kinetics is likely to occur after the nanopore has been completely filled and the growth extends beyond the quartz-covered region. It is worth noting that a similar current−time profile was reported in a previous study that focused on the fabrication of platinized nanoelectrodes.22 The gold structure resulting from CP deposition was characterized using CV in 0.5 M H2SO4 (Figure S6B of the Supporting Information). The presence of gold was verified by the presence of the signature gold redox peaks. The CVs of FcM recorded before, after electrochemical etching, and after gold deposition are shown in Figure S7 of the Supporting Information. Using the aforementioned method, the surface area was calculated to be 1000 μm2 and 0.9 μm2 using the charge under the gold oxide reduction peak and eq 2, respectively. The difference between the two values is much larger than that obtained from the CV system; this difference, again, could be attributed to the high porosity and restricted access of FcM to parts of the electrode. Although attempts to obtain an optical image of these electrode materials were unsuccessful, the materials deposited using CV and CP are presumed to be different in structure and porosity. However, 2851

dx.doi.org/10.1021/ac403816h | Anal. Chem. 2014, 86, 2849−2852

Analytical Chemistry

Letter

(6) Sun, P.; Mirkin, M. V. Anal. Chem. 2006, 78, 6526−6534. (7) Sassolas, A.; Leca-Bouvier, B. D.; Blum, L. J. Chem. Rev. 2008, 108, 109−139. (8) Zou, Z.; Kai, J.; Rust, M. J.; Han, J.; Ahn, C. H. Sens. Actuators, A 2007, 136, 518−526. (9) Roth, K. M.; Peyvan, K.; Schwarzkopf, K. R.; Ghindili, A. Electroanalysis 2006, 18, 1982−1988. (10) Fan, C.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. 2003, 100, 9134−9137. (11) Ricci, F.; Lai, R. Y.; Plaxco, K. W. Chem. Commun. 2007, 36, 3768−3770. (12) Caňete, S. J. P.; Lai, R. Y. Chem. Commun. 2010, 46, 3941− 3943. (13) Yang, W.; Gerasimov, J. Y.; Lai, R. Y. Chem. Commun. 2009, 20, 2902−2904. (14) Munge, B.; Liu, G. D.; Collins, G.; Wang, J. Anal. Chem. 2005, 77, 4662−4666. (15) Hahm, J.; Lieber, C. M. Nano Lett. 2004, 4, 51−54. (16) Zhao, S.; Yang, W.; Lai, R. Y. Biosens. Bioelectron. 2011, 26, 2442−2447. (17) Gerasimov, J. Y.; Lai, R. Y. Chem. Commun. 2010, 46, 395−397. (18) Li, Y.; Bergman, D.; Zhang, B. Anal. Chem. 2009, 81, 5496− 5502. (19) Jena, B. K.; Percival, S. J.; Zhang, B. Anal. Chem. 2010, 82, 6737−6743. (20) Zhang, B.; Zhang, Y.; White, H. S. Anal. Chem. 2004, 76, 6229− 6238. (21) McEwen, G. D.; Chen, F.; Zhou, A. Anal. Chim. Acta 2009, 643, 26−37. (22) Wang, Y.; Noël, J. M.; Velmurugan, J.; Nogala, W.; Mirkin, M. V.; Lu, C.; Guille Collignon, M.; Lemaître, F.; Amatore, C. Proc. Natl. Acad Sci. U.S.A. 2012, 109, 11534−11539. (23) Yang, W.; Lai, R. Y. Langmuir 2011, 27, 14669−14677. (24) Creager, S. E.; Wooster, T. T. Anal. Chem. 1998, 70, 4257− 4263. (25) Sumner, J. J.; Weber, K. S.; Hockett, L. A.; Creager, S. E. J. Phys. Chem. B 2000, 104, 7449−7459.

SLP sensors, respectively. This degree of signal attenuation is similar to that obtained from sensors fabricated on conventional gold electrodes.23 The slight difference could be due to the difference in surface probe coverage. Post-hybridization electron transfer kinetics, however, was not analyzed because of the low MB current. Sensor regeneration was attempted by incubating the hybridized sensors in 4 M GHCl for 4 s. For the LP sensor, sensor regeneration was successful; ∼73% of the initial signal was recovered. However, this was not the case for the SLP sensor. The same behavior was observed for a number of SLP sensors fabricated on these electrodes. It is possible that the structures on the electrode obstruct the DNA probes from resuming the original stem-loop conformation even after removal of the target DNA. Since the LP sensor is more compatible with these electrodes, we performed a calibration experiment to determine its limit of detection. As shown in Figure S10 of the Supporting Information, the sensor responded well to low concentrations of the target; a limit of detection of 20 nM was determined for this sensor. In conclusion, we have demonstrated, for the first time, the use of a gold-modified recessed Pt nanoelectrode in the fabrication of E-DNA sensors. These electrodes are highly compatible with the reagentless and reusable electrochemical sensing platform, which also includes E-AB and E-PB sensors. This combined sensing approach can potentially be used for real-time detection of target analytes inside living cells.



ASSOCIATED CONTENT

S Supporting Information *

Experimental conditions; nanoelectrode fabrication procedures; E-DNA sensor fabrication procedures; SEM images of nanoelectrode; CVs of nanoelectrode in FcM after each step of the fabrication process; CVs of FcM at different concentrations; electrochemical deposition of gold using CV and CP; CVs of gold-modified electrodes in H2SO4; LP and SLP E-DNA sensor constructs; electron transfer kinetics analysis for LP and SLP E-DNA sensors fabricated on nanoelectrodes and macroscale electrodes; ACVs of LP sensor in presence of different concentrations of target. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the Army Research Office (Grant W911NF-09-2-0039), National Science Foundation (Grant CHE-0955439), and Nebraska EPSCoR (Grant EPS-1004094). The authors would also like to thank Anita J. Zaitouna for the help with data analysis and Daniel Schmidt for the SEM images.



REFERENCES

(1) Yeh, J. I.; Shi, H. Wiley Interdiscip. Rev.: Nanomed. Nanobiotechnol. 2010, 2, 176−188. (2) Cox, J. T.; Zhang, B. Annu. Rev. Anal. Chem. 2012, 5, 253−272. (3) Bard, A. J.; Faulker, L. R. Electrochemical Methods, 2nd ed.; Wiley: New York, 2001. (4) Ugo, P.; Moretto, L. M.; Vezza, F. ChemPhysChem. 2002, 3, 917− 925. (5) Murray, R. W. Chem. Rev. 2008, 108, 2688−2720. 2852

dx.doi.org/10.1021/ac403816h | Anal. Chem. 2014, 86, 2849−2852