Gold Nanoparticles Doped Conducting PolymerNanorod Electrodes: Ferrocene CatalyzedAptamer-Based Thrombin Immunosensor

Md. Aminur Rahman,†,‡ Jung Ik Son,† Mi-Sook Won,§ and Yoon-Bo Shim*,†

Department of Chemistry, Pusan National University, and Korea Basic Science Institute, Pusan 609-735, South Korea

Au nanoparticles-doped conducting polymer nanorodselectrodes (AuNPs/CPNEs) were prepared by coating Aunanorods (AuNRs) with a conducting polymer layer. TheAuNRs were prepared through an electroless depositionmethod using the polycarbonate membrane (pore diam-eter, 50 nm, pore density, 6 × 108 pores/cm2) as atemplate. The AuNPs/CPNEs combining catalytic ac-tivity of ferrocene to ascorbic acid were used for thefabrication of an ultrasensitive aptamer sensor forthrombin detection. The AuNPs/3D-CPNEs were char-acterized employing cyclic voltammetry (CV), X-rayphotoelectron spectroscopy (XPS), scanning electronmicroscopy (SEM), and atomic force microscopy (AFM).Sandwiched immunoasay for r-human thrombin withNH2-functionalized-thrombin binding aptamer (Apt)immobilized on AuNPs/3D-CPNEs was studied throughthe electrocatalytic oxidation of ascorbic acid by theferrocene moiety that was bound with an antithrombinantibody and attached with the Apt/3D-CPNEs probethrough target binding. Various experimental param-eters affecting thrombin detection were optimized, andthe performance of the thrombin aptamer sensor wasexamined. The Apt/AuNPs/3D-CPNEs based throm-bin sensor exhibited a wide dynamic range of 5-2000ng L-1 and a low detection limit of 5 ng L-1 (0.14 pM).The selectivity and the stability of the proposed throm-bin aptamer sensor were excellent, and it was testedin a real human serum sample for the detection ofspiked concentrations of thrombin.

The three-dimensional (3D) nanostructured materials arepromising for diverse applications.1-4 Of these applications, anultrasensitive detection could be achieved using the 3D nanoelec-trode ensemble in electrochemical biosensing.3,4 The ultrasensi-tivity is due to the high signal-to-noise ratio, which can be easily

obtained in the case of nanoelectrodes enasemble. It has beenreported that using of the 3D nanoelectrodes ensemble isadvantageous in electrochemical sensing with high efficiencybecause of the geometry of an exposed structure.5 The 3Dnanoelectrodes ensemble can be fabricated using the templatemethod in which metals were grown within the pores of ananoporous membrane through electroless deposition.6,7 Gold hasbeen commonly used for fabrication of the 3D nanoelectrodesensemble. However, the application of 3D Au nanoelectrodeensembles in biomolecular sensing is limited3-5 due to the factthat only self-assembling thiolated biomolecules can be covalentlyimmobilized through the formation of a Au-thiol bond.4 Thus,the modification of the surface of 3D gold nanoelectrodes isneeded for directly immobilizing biomolecules. For this, 3Dnanoelectrodes coated with conducting polymers having functionalgroups are promising materials in biosensing. Previously, we havereported the conducting polymer layer composed of nanoparticlesas an enzyme immobilizing substrate for the fabrications ofphosphate8 and in vivo glutamate9 biosensors. However, theapplication of Au nanoparticles-deposited conducting polymercoated nanoelectrodes ensemble (AuNPs/3D-CPNEs) in biomo-lecular sensing has not yet been reported. For the application of3D conducting polymer electrodes to the biosensor probe, wefabricated a thrombin aptamer sensor with the 3D polymerelectrode by immobilizing a catalyst for the signal enhancement.

Thrombin is a protein that has many effects in coagulationcascade. It is a serine protease that converts soluble fibrinogento insoluble strands of fibrin as well as catalyzing many coagula-tion-related reactions. Because of its clinical importance, manyaptamer sensors have been reported using optical10-12 andelectrochemical methods.13-16 Thrombin detection using an

* Corresponding author. E-mail: ybshim@pusan.ac.kr. Phone: (+82) 51 5102244. Fax: (+82) 51 514 2430.

† Department of Chemistry.‡ Present address: Department of Applied Chemistry, Konkuk University.§ Korea Basic Science Institute.

(1) Sapp, S. A.; Mitchell, D. T.; Martin, C. R. Chem. Mater. 1999, 11, 1183–1185.

(2) Long, J. W.; Dunn, B.; Rolison, D. R.; White, H. S. Chem. Rev. 2004, 104,4463–4492.

(3) Gasparac, R.; Taft, B. J.; Lapierre-Devlin, M. A.; Lazareck, A. D.; Xu, J. M.;Kelley, S. O. J. Am. Chem. Soc. 2004, 126, 12270–12271.

(4) Roberts, M. A.; Kelley, S. O. J. Am. Chem. Soc. 2007, 129, 11356–11357.

(5) Lapierre-Devlin, M. A.; Asher, C. A.; Taft, B. J.; Gasparac, R.; Roberts, M. A.;Kelley, S. O. Nano Lett. 2005, 5, 1051–1055.

(6) Menon, V. P.; Martin, C. R. Anal. Chem. 1995, 67, 1920–1928.(7) Martin, C. R. Science 1994, 266, 1961–1966.(8) Rahman, M. A.; Park, D.-S.; Chang, S.-C.; McNeil, C. J.; Shim, Y.-B. Biosens.

Bioelectron. 2006, 21, 1116–1124.(9) Rahman, M. A.; Kwon, N.-H.; Won, M.-S.; Choe, E. S.; Shim, Y.-B. Anal.

Chem. 2005, 77, 4854–4860.(10) Pavlov, V.; Xiao, Y.; Shlyahovsky, B.; Willner, I. J. Am. Chem. Soc. 2004,

126, 11768–11769.(11) Ho, H.; Leclerc, M. J. Am. Chem. Soc. 2004, 126, 1384–1387.(12) Wang, X.; Li, F.; Su, Y.; Sun, X.; Li, X.; Schluesener, H. J.; Tang, F.; Xu, S.

Anal. Chem. 2004, 76, 5605–5610.(13) Hansen, J. A.; Wang, J.; Kawde, A.; Xiang, Y.; Gothelf, K. V.; Collins, G.

J. Am. Chem. Soc. 2006, 128, 2228–2229.(14) Baldrich, E.; Acero, J. L.; Reekmans, G.; Laureyn, W.; O’Sullivan, C. K.

Anal. Chem. 2005, 77, 4774–4784.

Anal. Chem. 2009, 81, 6604–6611

10.1021/ac900285v CCC: $40.75 2009 American Chemical Society6604 Analytical Chemistry, Vol. 81, No. 16, August 15, 2009Published on Web 07/14/2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

aptamer based probe has several advantages that make it an idealbiosensing element.17 Aptamers are synthetic DNA/RNA basedreceptors originated from an in vitro selection process, so-calledsystematic evolution of ligands by exponential enrichment(SELEX).18-20 Aptamers are chemically stable in contrast toantibodies or enzymes. They often undergo conformationalchanges upon target binding, which offers high flexibility in designof novel biosensors with high sensitivity and selectivity. The firstin vitro selected aptamer is thrombin-binding aptamer (TBA, 5′-GGTTGGTGTGGTTGG-3′), which has been targeted toward aprotein, thrombin21 at certain conditions that TBA folds into aquadruplex structure.22

There have been numerous reports on the aptamer basedthrombin biosensors.23-30 The use of aptamer functionalizedAuNPs as a catalytic label for the amplified optical detection ofthrombin was reported.29 The detection limit of thrombin wasdetermined to be 2 nM. Although, the sensitivity and the accuracyof optical aptamer sensors are comparable to electrochemicalaptamer sensors, electrochemical aptamer sensors reveal certainadvantages when compared to optical aptamer sensors. Becauseof their high sensitivity and selectivity, simple instrumentation,low production cost, and the fact that they are fast, accurate,compact, portable, and inexpensive, electrochemical methods havereceived particular attention. Many of the electrochemical methodsare based on a change in the electrochemical response of anelectroactive label. For example, the possibility to couple catalyticor biocatalytic labeling enables amplified detection of thrombin,and thus enhances the sensitivity of sensing processes. Anelectrochemical thrombin aptamer sensor was developed bytethering a redox-active methylene blue (MB) label to the aptamernucleic acid.24 In the absence of a target, the immobilized aptameris thought to relatively unfolded, thereby allowing the tetheredMB to collide with the electrode and transfer an electron. In thepresence of a target, electron transfer is inhibited, presumablydue to a binding-induced conformational change in the aptamerthat significantly increases the electron-tunneling distance. Al-though the method is elegant and simple, the current responsewas not directly proportional to the thrombin concentration butproportional to the logarithm concentration of thrombin. Though

the sensor could be regenerated, the aptamer itself was labeledwith MB, which could affect the binding affinity between the targetand aptamer to a certain degree. The detection limit of thrombinwas determined to be 20 nM. Another amperometric aptamersensor based on a redox-active indicator that intercalate intodouble-stranded DNA has been reported.25 A nucleic acid inhairpin configuration that includes the thrombin recognitionsequence was linked to a gold electrode, and methylene blue wasintercalated in the duplex stem of the probe hairpin structure.The detection limit was reported to be 1.1 nM. On the other hand,ferrocene as a label has been the subject of intense investigation,as it is a molecule that exhibits excellent reversibility of its redoxreaction.31 Previously, an electrochemical aptamer sensor basedon bifunctionalized aptamer with a terminal electroactive ferrocenegroup as the reporter and the thiol function as the anchor on agold electrode surface was reported.32 In the presence of throm-bin, the conformational changes result in the folded quadruplexstructure, bringing the ferrocene label to the electrode surface,thus enhancing the electron transfer. The detection limit wasreported to be 0.5 nM. However, no catalytic property wasexamined with this ferrocene labeled thrombin aptamer sensor.It had been shown that ferrocene can electrocatalyzed theoxidation of ascorbic acid.33 Thus, we tried to develop a highlysensitive thrombin-aptamer sensor based on gold nanoparticles(AuNPs) deposited 3D nanoelectrode ensemble (AuNPs/3D-CPNEs) immobilized ferrocene that electrocatalyzed the oxidationof ascorbic acid in the sample solution as a substrate.

In the present study, 3D-CPNEs were fabricated using apolycarbonate membrane (pore diameter, 50 nm, pore density, 6× 108 pores/cm2) as a template and were characterized usingcyclic voltammetry (CV), X-ray photoelectron spectroscopy(XPS), scanning electron microscopy (SEM), and atomic forcemicroscopy (AFM). A sandwiched immunoassay of humanthrombin at the NH2-functionalized-thrombin binding aptamer(Apt)-immobilized on the AuNPs deposited 3D-CPNEs wasexamined through the electrocatalytic oxidation of ascorbic acidby the ferrocene moiety that was bound with an antithrombinantibody through avidin-biotin interaction. Various experi-mental parameters affecting the thrombin detection wereoptimized, and the detection limit was determined. Theselectivity and the stability of this thrombin aptamer sensorwere also discussed.

EXPERIMENTAL SECTIONReagents. N-hydroxysuccinimide (NHS), biotin, streptavidin,

ascorbic acid, 1-ethyl-3(3-(dimethylamino)-propyl) carbodiimide(EDC), N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid(HEPES), ferricyanide, sodium bicarbonate, sodium chloride,hydrogen tetrachloroaurate (HAuCl4), glycerol, antihuman throm-bin, immunoglobulin G (IgG), human serum albumin, andbovine serum albumin (produced in mouse) were purchasedfrom Sigma Co. Track-etch polycarbonate membrane filterswere obtained from Sterlitech Corporation. The thickness ofthe membranes were 6 µm with a nominal pore diameter of 80

(15) Centi, S.; Tombelli, S.; Minunni, M.; Mascini, M. Anal. Chem. 2007, 79,1466–1473.

(16) Willner, I.; Zayats, M. Angew. Chem., Int. Ed. 2007, 46, 6408–6418.(17) Wang, J.; Wang, L.; Liu, X.; Liang, Z.; Song, S.; Li, W.; Li, G.; Fan, C. Adv.

Mater. 2007, 19, 3943–3946.(18) Hermann, T.; Patel, D. J. Science 2000, 287, 820–825.(19) Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818–822.(20) Tuerk, C.; Gold, L. Science 1990, 249, 505–510.(21) Bock, L. C.; Griffin, L. C.; Latham, J. A.; Vermass., E. H.; Toole, J. J. Nature

1992, 355, 564–566.(22) Marathias, V. M.; Bolton, P. H. Biochemistry 1999, 38, 4355–4364.(23) Wang, J. J. Am. Chem. Soc. 2006, 128, 2228–2229.(24) Xiao, Y.; Lubin, A. A.; Heeger, A. A.; Plaxco, K. W. Angew. Chem., Int. Ed.

2005, 44, 5456–5459.(25) Bang, G. S.; Cho, S.; Kim, B. G. Biosens. Bioelectron. 2005, 21, 863–870.(26) Dittmer, W. U.; Reuter, A.; Simmel, F. C. Angew. Chem., Int. Ed. 2004,

43, 3550–3553.(27) Li, J. W.; Fang, X. H.; Tan, W. H. Biochem. Biophys. Res. Commun. 2002,

292, 31–40.(28) Hamaguchi, N.; Ellington, A. D.; Stanton, M. Anal. Biochem. 2001, 294,

126–131.(29) Pavlov, V.; Xiao, Y.; Shlyahovsky, B.; Willner, I. J. Am. Chem. Soc. 2004,

126, 11768–11769.(30) Bock, L. C.; Griffin, L. C.; Latham, J. A.; Vwemass, E. H.; Toole, J. J. Nature

1992, 355, 564–566.

(31) Van Staveren, D. R.; Metzler-Nolte, N. Chem. Rev. 2004, 104, 5931–5986.(32) Radi, A.-E.; Sanchez, J. L. A.; Baldrich, E.; O’Sullivan, C. K. J. Am. Chem.

Soc. 2006, 128, 117–124.(33) Lertanantawong, B.; O’Mullane, A. P.; Zhang, J.; Surareungchai, W.;

Somasundrum, M.; Bond, A. M. Anal. Chem. 2008, 80, 6515–6525.

6605Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

nm and a pore density of 6.0 × 10-8 pores/cm-2. Tetrabuty-lammonium perchlorate (TBAP, electrochemical grade) wasreceived from Fluka, purified, and then dried under vacuumat 10-5 Torr. Acetonitrile (AN, 99.8%, anhydrous, sealed underN2 gas) and dichloromethane (99.8%, anhydrous, sealed underN2 gas), sodium sulfite (98%), formaldehyde (37% wt %),potassium chloride, ferrocene carboxylic acid, and ammoniumhydroxide (NH3 content 28-30%) were obtained from Sigma-Aldrich Co. Tin(II) chloride and trifluoroacetic acid wereobtained from Aldrich Co. Human R-thrombin (specific activity,3 725 Units/mg; concentration, 9.7 mg/mL) was obtained fromHaematologic Technologies Inc. The thrombin aptamers wereobtained from Bioneer Co. (South Korea), which have thefollowing sequences: 5′-NH2-GGT TGG TGT GGT TGG-3′ and5′-SH-GGT TGG TGT GGT TGG-3′.

A terthiophene monomer bearing a carboxylic acid group,5,2′;5′,2′′-terthiophene-3-carboxylic acid (TTCA) was synthesizedaccording to a previous report.34 The streptavidin-ferroceneconjugate was synthesized by the EDC-NHS method.32 Briefly,10 µM of ferrocenecarboxylic acid was reacted for 2 h with amixed solution of 10 mM EDC and 10 mM NHS in HEPES. Tothis solution, 0.1 mg/mL of streptavidin was added and incubatedfor 2 h. Then, ferrocene was conjugated with streptavidin throughthe covalent bond formation between the amine groups ofstreptavidin and carboxylic acid groups of ferrocenecarboxylicacid. Biotinylated antithrombin was synthesized by the cross-linking method with glutaraldehyde. Gold nanoparticles (AuNP)were prepared according to a previously reported procedure.35

From the high-resolution transmission electron microscope (HR-TEM), the particle size of AuNPs were about 4.0 nm. All otherchemicals were of extra pure analytical grade and used withoutfurther purification. All aqueous solutions were prepared in doublydistilled water, which was obtained from a Milli-Q water purifyingsystem (18 MΩ cm).

Instruments. Cyclic voltammograms were recorded using aKosentech model KST-P2 (South Korea) and an EG & G 273Apotentiostat/galvanostat. XPS experiments were performed usinga VG Scientific ESCALAB 250 XPS spectrometer with a mono-chromated Al KR source with charge compensation at the KoreaBasic Science Institute (KBSI), Busan. SEM images were obtainedusing a Cambridge Stereoscan 240. Atomic force microscopic(AFM) experiments were carried out on a multimode AFM systemfrom Digital Instrument Inc. Commercial Si3N4 tips (125 mmlength, 300 kHz resonance frequency, 5-10 nm radius) wereattached to a triangular cantilever made of the same material.The force constant was 200-1000 N/m.

Preparation of 3D Gold Nanorods. The 2D gold nanodiskswere prepared using the electroless plating procedure reportedpreviously6-8 with slight modification. Briefly, after the membranewas wet for 2.5 h in methanol, the polycarbonate templatemembrane was sensitized with Sn2+ by immersing it 30 min intoa solution containing 0.026 M SnCl2 and 0.07 M trifluoroaceticacid in 1:1 methanol/water as the solvent. After the membranewas rinsed with methanol for 30 min, the sensitized membranewas immersed for 10 min into a solution of 0.029 M AgNO3 inaqueous ammonia. After the membrane was washed with

methanol to remove excess AgNO3, it was immersed into theAu plating bath (10 mL), which included 3.75 g/L HAuCl4, 40g/L Na2SO3, 5 g/L EDTA, 30 g/L K2HPO4, and 0.6 mL offormaldehyde. The pH of the gold plating solution wascontrolled at about 10. Formaldehyde was used as a reducingagent that reduces the Au(I) to metallic Au. The gold electro-less deposition was carried out for 24 h at 4 °C. The membranewas rinsed with water after plating and then immersed in 25%HNO3 for 12 h to remove the surface bound chemicals. Themembrane was then thoroughly rinsed with water and air-dried.To prepare 3D gold nanorods, the surface gold from one faceof the membrane was removed and chemical etching wascarried out on the same side of the membrane, which wascarried out by using a mixture of 50:50 dichloromethane andethanol.36 Finally, 3D gold nanorods was heated at 150 °C for10 min to improve sealing the membrane around 3D goldnanorods.

Preparation of 3D Conducting Polymer Nanoelectrodes.The unetched side of the membrane was adhered to a strip ofadhesive conducting copper tape. A piece of insulating tape (NittoCo., Japan) was punched to be made a 0.03 cm2 hole and pastedonto the etched side of the membrane so that the exposedcopper tape was covered completely on both sides to preventexposure to the analyte solution. The resulting 3D goldnanoelectrodes were used to prepare 3D conducting polymernanoeletrodes (3D-CPNEs) by electrochemical deposition ofa conducting polymer film on 3D gold nanoelectrodes. Thedeposition was performed through electropolymerization ofthe 0.1 µM TTCA monomer in 0.1 M TBAP/AN by cyclingthe potential between 0 and 1500 mV three times at the scanrate of 100 mV/s. After electropolymerization, the 3D-CPNEswere washed with AN to remove the excess monomer.

Fabrications of the Aptamer Probe and the IndirectSandwich Catalytic Aptamer Sensor. The 3D-CPNEs wereimmersed for 6 h in a 0.01 M HEPES buffer solution (pH 7.0)containing 10.0 mM of EDC and 10 mM of NHS to activate thecarboxylic acid groups of the 3D-CPNEs. Then, the EDC treated3D-CPNEs were washed with an HEPES buffer solution andsubsequently incubated for 2 h in a 2 µM amine group-modifiedthrombin binding aptamer (NH2-TBA) in the 0.01 M HEPESbuffer (pH ) 7.0) at 4 °C. The NH2-TBA was immobilized ontothe 3D-CPNEs through the formation of covalent bondsbetween carboxylic acid groups of the conducting polymer (CP)and amine groups of the aptamer. The aptamer immobilized3D-CPNEs were then rinsed thoroughly with the HEPES bufferto remove the weakly adsorbed aptamers and subsequentlyincubated for 1 h in a 0.1% BSA solution for minimizing thenonspecific binding. The 3D-CPNEs/NH2-TBA probes werethen incubated for 30 min in an HEPES solution containingthrombin at various concentrations. The thrombin bindingprobe was incubated for 30 min in an HEPES solutioncontaining biotinylated antithrombin antibody after beingrinsed. The resulting assembly was then immersed in a HEPESsolution containing steptavidin-conjugated ferrocene for 30 minand subsequently washed with HEPES buffer solution. By thisstep, ferrocene was attached with antithrombin throughstreptavidin-biotin interactions. In the presence of a substrate,(34) Lee, T. Y.; Shim, Y.-B. Anal. Chem. 2001, 73, 5629–5632.

(35) Shiddiky, M. J. A.; Rahman, M. A.; Shim, Y.-B. Anal. Chem. 2007, 79,6886–6890. (36) Krishnamoorthy, K.; Zoski, C. G. Anal. Chem. 2005, 77, 5068–5071.

6606 Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

ascorbic acid, the attached ferrocene electrochemically cata-lyzed the oxidation of ascorbic acid, thus amplified the currentresponse. Since, thrombin was sandwiched between thrombinaptamer and antithrombin antibody attached with ferrocene,the catalytic current response was proportional to the thrombinconcentration. The schematic presentation of the fabricationof an aptamer sensor probe and electrochemical catalyticdetection of thrombin are presented in Figure 1.

RESULTS AND DISCUSSIONCharacterization of 3D Conducting Polymer Nanoelec-

trodes. The 3D-CPNEs were prepared through electropolymer-ization of the 0.1 µM TTCA monomer on 3D gold nanorods in a0.1 M TBAP/AN solution using a potential cycling method. Atthe first anodic scan, the CV exhibited one oxidation peak ataround 1.3 V where the monomer oxidized and formed thepolymer immediately on the Au nanorods (AuNRs) (figure notshown). A polymer reduction peak was observed at around 1.1 Vin the reverse cathodic scan due to the reduction of immediatelyformed polymer at 1.3 V. The peak currents at 1.3 and 1.1 Vincreased as the number of potential cycles increased, indicatingthe formation and the growth of the PolyTTCA film on 3D Aunanorods (3D-CPNEs). The resulted 3D-CPNEs were electro-chemically characterized by recording CV in a blank phosphatebuffer solution (figure not shown). The Au reduction peakobtained for 3D AuNRs at about 0.75 V vs Ag/AgCl significantlydecreased, indicating that the 3D gold nanorods were coated withthe conducting polymer layer.

The morphology of 3D-CPNEs were characterized using SEM.Parts a and b of Figure 2 show the SEM images obtained for 3D-AuNRs and 3D-CPNEs, respectively. The shining extent of 3D-AuNRs decreased for 3D-CPNEs, indicating the formation of athin conducting polymer layer on 3D-AuNRs. To confirm theformation of 3D-CPNEs, SEM images were obtained after dis-solving the polycarbonate membrane by dichloromethane and areshown in the insets of Figure 2a,b. From the SEM images, thediameters of the 3D-AuNRs and 3D-CPNEs were determined tobe 30 and 60 nm, respectively.

The 3D-CPNEs were also characterized using the XPS method.Figure 3 shows the survey spectra obtained for (a) 3D-AuNRsand (b) 3D-CPNEs. The survey spectrum for 3D-AuNRs showstwo sharp peaks at 83.50 and 87.3 eV. These peaks correspondedto Au 4f7 and Au 4f5, respectively. In addition to Au 4f peaks, Au

4d5, Au 4d3, and Au 4p3 peaks were also observed at 334.4, 353.3,and 546.4 eV, respectively. The survey spectrum of 3D-AuNRsalso shows C 1s and O 1s peaks. The C 1s and O 1s peaks camefrom the polycarbonate membrane. The C 1s peak observed at284.8 eV corresponded to C-H or C-C bonds, whereas the O 1s

Figure 1. The schematic representation of the fabrication of the Apt/3D-CPNEs-based thrombin aptamer sensor.

Figure 2. (a) SEM images obtained for the (a) 3D-AuNRs and (b)3D-CPNEs.

Figure 3. XPS analysis for the (a) 3D-AuNRs and (b) 3D-CPNEs.

6607Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

peaks observed at 532.3 and 533.2 eV corresponded to the CdOand C-O, respectively. On the other hand, the survey spectrumof 3D-CPNEs also shows C 1s, O 1s, and Au 4f peaks. In addition,the S 2p peak at 164.4 eV was observed, while it was not observedin the 3D-AuNRs survey spectrum due to the presence ofpolyterthiophene film on the surface. The intensities of C 1s andO 1s peaks increased whereas the intensity of Au peaks decreased,indicating that the conducting polymer was successfully coatedon the 3D-AuNRs.

Redox Properties of the Conjugated Ferrocene on the 3D-CPNEs. Figure 4a shows CVs recorded for the final aptamerimmobilized 3D-CPNEs probe (Apt/3D-CPNEs) in the presenceof antithrombin-conjugated ferrocene as an indicator attached withthe aptamer probe through the target binding (wine line) as shownin Figure 1. A redox peak was observed at approximately +0.1/+0.32 V vs. Ag/AgCl, which was not observed in the absence ofconjugated ferrocene (green line). These results indicate that theredox peak came solely from conjugated ferrocene itself. The peakcurrent was directly proportional to the scan rate between 20 and200 mV/s, indicating that the redox reaction was involved in asurface-confined process.37 The formal potential of the redoxreaction was determined to be +0.21 V. The peak separationbetween anodic and cathodic peak potential was determined tobe +0.22 V, which indicates a qusai-reversible electron transferprocess of ferrocene. In order to enhance the electron transferprocess of the conjugated ferrocene, the 3D-CPNEs were modifiedby electrochemically depositing gold nanoparticles (AuNPs/3D-

CPNEs). AuNPs are widely used in biotechnology38 due to theirunique physical and chemical properties.39 In our previous works,we have used AuNPs for increasing the surface conductivities ofconducting polymer and dendrimer bonded conducting polymerlayers.40,41 The CV recorded (dark cyan line) at the AuNPsdeposited 3D-CPNEs (Apt/AuNPs/3D-CPNEs) with conjugatedferrocene also showed a pair of redox peaks at +0.04/+0.32 V vsAg/AgCl. The formal potential of the redox peak was determinedto be +0.18 V, which was lower than the obtained one withoutAuNPs. In this case, the peak current was also proportional tothe scan rate between 20 and 200 mV/s, indicating that the redoxreaction was also involved in a surface-confined process. Althoughthe formal potential of the conjugated ferrocene lowered by only0.03 V, the redox peak currents were significantly increased. Theredox peak currents of the conjugated ferrocene obtained for theApt/AuNPs/3D-CPNEs were three or four times higher than thatobtained for Apt/3D-CPNEs. This result revealed that AuNPsacted as efficient electron transfer promoters for the redoxreaction of the conjugated ferrocene.

The electrocatalytic behavior of the ferrocene-conjugated Apt/AuNPs/3D-CPNEs probe toward the ascorbic acid oxidation wasstudied. Figure 4b shows CVs recorded for ferrocene-conjugatedApt/3D-CPNEs without (blue line) or with AuNPs (green line)on the oxidation of ascorbic acid in a 0.01 M HEPES buffersolution (pH ) 7.0). For comparison, a CV on the oxidation ofascorbic acid at a bare gold electrode (red line) was also recorded.The small oxidation peak of ascorbic acid was observed at about+0.55 V at the bare gold electrode, while it was shifted to a lesspositive potential from +0.55 V to +0.3 V at the ferrocene-conjugated Apt/3D-CPNEs modified electrode without AuNPs.The anodic peak current of ascorbic acid was found to be 5 timeshigher than that obtained for a bare gold electrode. These resultsrevealed that conjugated ferrocene at the Apt/3D-CPNEs probehad a catalytic activity toward the oxidation of ascorbic acid. Onthe other hand, the oxidation potential of ascorbic acid slightlylowered for the ferrocene-conjugated Apt/3D-CPNEs (+0.27 V)with AuNPs. However, when the Apt/3D-CPNEs electrode wasmodified using AuNPs, the anodic peak current of ascorbic acidoxidation was found to be 2 times higher than that obtainedwithout using AuNPs. The higher oxidation current was observeddue to the fact that AuNPs increased the surface conductivitiesof 3D-CPNEs modified electrode.

Optimization of Experimental Parameters for ThrombinDetection. The experimental parameters for the detection ofthrombin with the ferrocene-conjugated Apt/AuNPs/3D-CPNEsprobe were optimized in terms of aptamer concentration, ferroceneconcentration in the conjugation, ascorbic acid concentration, pH,and temperature, where the thrombin concentration was keptconstant. The effect of aptamer concentration on the electrocata-lytic oxidation of ascorbic acid was studied between 0.2 and 5µM (Figure 5a). The catalytic oxidation current of ascorbic acidgradually increased from 0.2 to 2.0 µM of aptamer concentrationimmobilized. Over 2.0 µM concentration of aptamer, the catalytic

(37) Murray, R. W. Chemically Modified Electrodes. In Electroanalytical Chem-istry, Vol. 13; Bard, A. J., Ed.; Marcel Dekker: New York, 1984; pp 191-368.

(38) Alivisatos, A. P. Science 1996, 271, 933–937.(39) Mirkin, C. R.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996,

382, 607–609.(40) Rahman, M. A.; Noh, H. B.; Shim, Y.-B. Anal. Chem. 2008, 80, 8020–8027.(41) Singh, K.; Rahman, M. A.; Son, J. I.; Kim, K. C.; Shim, Y.-B. Biosens.

Bioelectron. 2008, 23, 1595–1601.

Figure 4. (a) CVs recorded for Apt/3D-CPNEs (wine line) and Apt/AuNPs/3D-CPNEs (dark cyan line) with conjugated ferrocene andfor Apt/AuNPs/3D-CPNEs (green line) without conjugated ferrocenein a 0.01 M HEPRS buffer solution. (b) The electrocatalytic oxidationof ascorbic acid with a bare gold electrode (red line), Apt/3D-CPNEs(blue line), and Apt/AuNPs/3D-CPNEs (green line) with ferrocene-conjugated in a 0.01 M HEPES buffer solution (pH ) 7.0).

6608 Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

oxidation current did not increase significantly. Thus, the aptamerconcentration was used as 2.0 µM in the subsequent experiments.The effect of ferrocene concentration for the conjugation was alsoexamined between 3 and 30 µM (Figure 5b). The catalyticoxidation current gradually increased from 3 to 10 µM. At higherconcentrations over 10 µM, the catalytic current did not increasesignificantly. Thus, the optimum concentration of ferrocene forthe conjugation was selected as 10 µM. The effect of ascorbicacid concentration was studied between 1 and 20 mM (Figure5c). The catalytic oxidation current gradually increased from 1 to10 mM. Over 10 mM concentration of ascorbic acid, the catalyticcurrent was not found to be significantly increased due to thesaturation effect. Thus, the optimum concentration of ascorbic acidwas determined as 10 mM. The effect of pH on the ascorbic acidoxidation was studied over the pH range of 4.5-8.5 in a HEPESbuffer solution containing 10 mM ascorbic acid (Figure 5d). Theoxidation current gradually increased from pH 4.5 to 7.5 and thendecreased at pH values higher than 7.5. The first (pK1) andsecond (pK2) values of ascorbic acid are 4.15 and 11.50.42 Thedecrease in current response over pH 7.5 may be due to theincrease amount of dissociated ascorbic acid at higher pH.The maximum catalytic oxidation current was observed at apH of 7.5. Thus, the optimum pH was chosen as 7.5. The effectof temperature on the detection of thrombin was studiedbetween 10 and 80 °C (figure not shown). The catalyticresponse was found to be gradually increased as the temper-ature was increased from 10 to 35 °C. The catalytic responsewas not found to be significantly changed between 35 and 45°C. However, the current response rapidly decreased from 45to 80 °C due to the deactivation of thrombin. On the basis ofthe temperature-response profile, the optimum temperaturewas selected as 35 °C.

Interference Effect. Common proteins, such as immunoglobinG (IgG) and human serum albumin (HSA), can interfere with thethrombin detection. In order to assess the possibility of interfer-ence, the catalytic oxidation current response was measured forIgG and HSA (figure not shown). IgG and HSA did not interferein thrombin detection. This was due to the fact that the thrombinaptamer and antithrombin antibody cannot interact with IgG andHSA, thus, the conjugated ferrocene cannot interact on theimmunosensor and no catalytic current response was observed.In addition to IgG and HSA, the ferrocene-conjugated Apt/AuNPs/3D-CPNEs sensor did not respond to bovine serum albumin(BSA). Thus, the response of the present aptamer sensor was veryselective for human R-thrombin.

Calibration Plot. The electrocatalytic current responses weremeasured for ferrocene-conjugated Apt/3D-CPNEs and Apt/AuNPs/3D-CPNEs based sensors by varying the thrombin con-centration. Figure 6 shows the electrocatalytic responses andcalibration plots obtained for Apt/3D-CPNEs (Figure 6a,b) andApt/AuNPs/3D-CPNEs (Figure 6c,d) based aptamer sensors forthrombin detection. The inset in part c shows the electrocatalyticresponses at 0, 5.0, 10, and 50 ng L-1 of thrombin concentrations.At the thrombin concentration of 0 ng L-1 (no thrombinpresent), no electrocatalytic response was observed. Theferrocene conjugated antithombin antibody might not beattached with the aptamer probe without thrombin, thus theelectrocatalytic oxidation of ascorbic acid by the conjugatedferrocene was not observed. On the other hand, electrocatalyticresponses were obtained at 5.0, 10, and 50 ng L-1 thrombinconcentrations. At lower concentrations than 5.0 ng L-1 (1.0and 2.0 ng L-1), we did not get any electrocatalytic responses.CVs recorded at these concentrations were similar to thatrecorded for 0 ng L-1, indicating that the aptamer probe wasnot able to detect thrombin at lower concentrations than 5.0ng L-1. Under the optimized conditions, the electrocatalyticresponse of thrombin detection was linear in the ranges of

(42) Kurtum, G.; Vogel, W.; Andrussow, K. Pure Appl. Chem. 1960, 1, 187-536 (IUPAC Technical Reports and Recommendation, Dissociation Constantof Organic Acids in Aqueous Solutions).

Figure 5. The effects of the concentration of (a) ascorbic acid, (b) ferrocene in the conjugation solution, (c) ascorbic acid and (d) the pH onthe electrocatalytic response for thrombin detection.

6609Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

10-1000 and 5-2000 ng L-1 for Apt/3D-CPNEs and Apt/AuNPs/3D-CPNEs based aptamer sensors, respectively. Theselinear dependencies yielded regression equations of Ip (µA)) 9.15 + 0.02[C] (ng L-1) and Ip (µA) ) 33.8 + 0.05[C] (ngL-1) with the correlation coefficients of 0.998 and 0.999,respectively. The reproducibility expressed in terms of therelative standard deviation (RSD) was about 4.3% and 5.7% at athrombin concentration of 80 ng L-1. The detection limits ofthrombin were determined to be 10.0 (0.28 pM) and 5 ng L-1

(0.14 pM) for Apt/3D-CPNEs and Apt/AuNPs/3D-CPNEs,respectively. These values of detection limits were much lowerthan the detection limits obtained from the electrochemicalthrombin detections based on a binding induced conformationalchange where ferrocene (0.5 nM)32 and methylene blue (∼20nM)24 were used as labels. The detection limits in the presentstudy were also 2-4 times lower than the most sensitivethrombin aptasensor (0.5 pM).23 The sensitivity of the thrombinaptamer sensor was found to be 2 times enhanced with AuNPs.The detection limit of thrombin was also determined for 2D-Au nanodisk (2D-AuNDs) and 3D-AuNRs electrodes, where athiolated TBA (5′-SH-GGT TGG TGT GGT TGG-3′) was used.The results are shown in Table 1. It can be seen that the Apt/AuNPs/3D-CPNEs exhibited the highest sensitivity for theaptamer sensor-based thrombin detection.

Stability of the Apt/AuNPs/3D-CPNEs Based ThrombinAptamer Sensor. The stability of the present thrombin aptamersensor was determined by measuring the response once a dayfor 1 month. After each measurement, the aptamer sensor was

stored in a phosphate buffer solution at 4 °C. From theseexperiments, no significant decrease in thrombin detection wasobserved for 3 weeks. For 3 weeks, the aptamer sensor retainedmore than 95% of its initial response. After 3 weeks, the responsewas gradually decreased due to the gradual decrease in theinteraction between aptamer and thrombin. These results indicatethat the thrombin aptamer sensor exhibited not only highsensitivity but also long-term stability.

Real Sample Analysis. The practical applicability of theproposed aptamer sensor was investigated by detecting thrombinin a human real serum sample. The Apt/AuNPs/3D-CPNEs basedthrombin aptamer sensor did not detect thrombin in the serumsample. This was due to the fact that healthy human serum sampledoes not contain thrombin.15 However, to examine the applicabilityof this aptamer sensor in serum sample, we performed spike andrecovery experiments. The serum sample was spiked with 5, 10,and 20 ng L-1 thrombin. Figure 7 shows the electrocatalyticresponses obtained in thrombin spiked serum samples. Afterelectrochemical measurement, the calibration method was usedto determine thrombin concentration. The current response ofthis aptamer sensor in a serum sample was slightly lower (about3%) than that obtained in a blank buffer solution. The thrombinconcentration recovery was between 95% and 98%, which clearlyindicates the potentiality of this aptamer sensor for thrombindetection in real biological samples.

CONCLUSIONSFerrocene-catalyzed ascorbic acid oxidation based an ultra-

sensitive thrombin aptamer sensor was fabricated by covalentlyimmobilizing a thrombin binding aptamer onto 3D-CPNEs. 3D-CPNEs were fabricated by coating a conducting polymer layeron the surface of 3D-AuNRs, which were prepared through theelectroless deposition method. The sensitivity of the thrombindetection was further enhanced by depositing AuNPs on to the

Figure 6. Electrocatalytic responses and calibration plots forthrombin detection with (a, b) Apt/3D-CPNEs and (c,d) Apt/AuNPs/3D-CPNEs. The inset in part c shows electrocatalytic responses at0, 5, 10, and 50 ng L-1 of thrombin concentration.

Figure 7. Electrocatalytic responses obtained in spiked (5, 10, and20 ng L-1 thrombin) serum samples.

Table 1. Comparison of the Hydrodynamic Range andthe Detection Limit of Thrombin with Other AptamerBased Nanostructured Sensor Probes

nanostructuredprobe

dynamic range(ng L-1)

detection limit(ng L-1)

Apt/ 2D-AuNDs 10-1000 10Apt/3D-AuNRs 5-2000 5Apt/3D-CPNEs 10-1000 10Apt/AuNPs/3D-CPNEs 5-2000 5

6610 Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v

3D-CPNEs before aptamer immobilization (Apt/AuNPs/3D-CPNEs). The aptamer sensor was successfully characterized usingSEM, XPS, and electrochemical techniques. SEM results con-firmed the formation of 3D-AuNRs and 3D-CPNEs, and thediameters of the 3D-AuNRs and 3D-CPNEs were determined tobe 30 and 60 nm, respectively. XPS results further confirmed thatthe conducting polymer was successfully coated on the 3D-AuNRs.The redox and catalytic properties of the conjugated ferrocene atApt/AuNPs/3D-CPNEs and Apt/3D-CPNEs were characterizedusing a cyclic voltammetric technique. The electrocatalytic oxida-tion of ascorbic acid by the labeled-ferrocene increased thesensitivity of the thrombin aptamer sensor. The Apt/AuNPs/3D-CPNEs based aptamer sensor exhibited a wide linear range(5-2000 ng L-1) and a very low detection limit (0.14 pM). TheRSD value was determined to be 5.7% at the thrombinconcentration of 80 ng L-1. The common proteins, such as IgG,

HSA, and BSA, did not interfere with the thrombin detection.The present aptasensor also exhibited a long-term stability of3 weeks. The sensitivity of the Apt/AuNPs/3D-CPNEs sensorwas 10 times higher than the most sensitive thrombin aptamersensor reported until now and was successfully tested in a realhuman serum sample for the detection of spiked amounts ofthrombin.

ACKNOWLEDGMENTThe work is supported by Korea Ministry of Environment as

“The Eco-technopia 21 project” (Grant No. 091-082-078).

Received for review February 6, 2009. Accepted July 4,2009.

AC900285V

6611Analytical Chemistry, Vol. 81, No. 16, August 15, 2009

Dow

nloa

ded

by P

USA

N N

AT

UN

IV C

EN

TR

AL

LIB

on

Aug

ust 1

4, 2

009

Publ

ishe

d on

Jul

y 14

, 200

9 on

http

://pu

bs.a

cs.o

rg |

doi:

10.1

021/

ac90

0285

v