Author's Accepted Manuscript

Electrochemiluminescence immunosensor fortumor markers based on biological barcodemode with conductive nanospheres

Shuping Du, Zhiyong Guo, Beibei Chen, YuhongSha, Xiaohua Jiang, Xing Li, Ning Gan, Sui Wang

PII: S0956-5663(13)00659-3DOI: http://dx.doi.org/10.1016/j.bios.2013.09.041Reference: BIOS6225

To appear in: Biosensors and Bioelectronics

Received date: 22 July 2013Revised date: 19 September 2013Accepted date: 19 September 2013

Cite this article as: Shuping Du, Zhiyong Guo, Beibei Chen, Yuhong Sha,Xiaohua Jiang, Xing Li, Ning Gan, Sui Wang, Electrochemiluminescenceimmunosensor for tumor markers based on biological barcode mode withconductive nanospheres, Biosensors and Bioelectronics, http://dx.doi.org/10.1016/j.bios.2013.09.041

This is a PDF file of an unedited manuscript that has been accepted forpublication. As a service to our customers we are providing this early version ofthe manuscript. The manuscript will undergo copyediting, typesetting, andreview of the resulting galley proof before it is published in its final citable form.Please note that during the production process errors may be discovered whichcould affect the content, and all legal disclaimers that apply to the journalpertain.

www.elsevier.com/locate/bios

1

Electrochemiluminescence immunosensor for tumor markers 1

based on biological barcode mode with conductive nanospheres 2

3

Shuping Dua, Zhiyong Guoa,�, Beibei Chena, Yuhong Shaa, Xiaohua Jiangb, Xing Lia, 4

Ning Gana, Sui Wanga 5

6

a Faculty of Materials Science and Chemical Engineering, The State Key Laboratory 7

Base of Novel Functional Materials and Preparation Science, Ningbo University, 8

Ningbo 315211, PR China 9

b School of Applied Chemistry and Biological Technology, Shenzhen Polytechnic, 10

Shenzhen 518055, PR China 11

12

13

Abstract 14

A novel sandwich-type electrochemiluminescence (ECL) immunosensor was 15

developed for highly sensitive and selective determination of tumor markers based on 16

biological barcode mode. N-(4-aminobutyl)-N-ethylisoluminol (ABEI) and the second 17

antibody (Ab2) were simultaneously immobilized on conductive nanospheres to 18

construct ABEI/Ab2-CNSs probes, which could form sandwich immunocomplex by 19

Ab2 and emit ECL signals by ABEI. The gold layer coated on the surface of the 20

conductive nanospheres could extend the outer Helmholtz plane (OHP) of the ECL 21

* Corresponding author. Tel: +86 574 87600798.

E-mail address: nbuguo@163.com (Z. Guo).

2

immunosensor effectively. Benefited from it, all ABEI molecules immobilized on 22

conductive nanospheres would act as biological barcode to give in-situ ECL signals 23

without interfering with the activity of the second antibody. In such a case, the 24

sensitivity of the ECL immunosensor would be greatly improved because an antigen 25

molecule would correspond to ECL signals of thousands of ABEI molecules. Using 26

prostate specific antigen (PSA) as a model tumor marker, the ECL intensity was 27

found to increase with the logarithm of PSA concentration with a wide linear range 28

from 0.04 to 10 fg/mL. In addition, specificity, stability, reproducibility, regeneration 29

and application were satisfactory. Therefore, this developed ECL immunosensor has a 30

potential for practical detection of disease-related proteins besides tumor markers in 31

the clinical diagnostics. 32

Keywords: Electrochemiluminescence immunosensor; Tumor markers; Biological 33

barcode mode; Conductive nanospheres (CNSs) 34

35

36

1. Introduction 37

Tumor markers are biochemical substances which are produced by either the 38

tumor itself or the surrounding normal tissue as a response to tumor cells (Freedland, 39

2011; Wickström et al., 2011). Since the presence of tumor markers in serum or other 40

body fluids in response to precancerous or cancerous conditions induces an array of 41

biochemical processes, they are often used for evaluating disease process, recurrence, 42

metastasis and prognosis (Marta et al., 2013; Liu and Ma, 2013; Qu et al., 2013; 43

3

Xiang and Lu, 2012), and the low concentrations of tumor markers may be related to 44

the early stage of cancerous conditions. Therefore, it is crucial to develop diagnostic 45

tools for the detection of very low concentrations of these markers in healthy humans, 46

sub-healthy humans or patients to identify the early stage. 47

Lots of methods and strategies have been developed for the detection of tumor 48

markers, including radioimmunoassay (RIA) (Tyan et al., 2013; Su et al., 2011), 49

chemiluminescence immunoassay (CLIA) (Tian et al., 2010; Yang et al., 2010b), 50

enzyme-linked immunosorbent assay (ELISA) (Darwish et al., 2013; Chalupova et al., 51

2013; Saadi et al., 2013), chemiluminescent enzyme immunoassay (CLEIA) (Dong et 52

al., 2012; Xiao et al., 2009) and time-resolved fluorescence immunoassay (TRFIA) 53

(Hou et al., 2012; Lu et al., 2012; Niu et al., 2011), etc. However, there are still some 54

problems challenging their application: (1) the sensitivity is not high enough to 55

determine tumor markers at low level, (2) radioactive or toxic markers are needed, (3) 56

experimental procedures are complex, (4) detection time is long, and (5) instruments 57

used are expensive, and so on. 58

Electrochemiluminescence (ECL) is a valuable detection method, which has 59

been applied extensively due to its acknowledged advantages such as versatility, 60

simplified optical setup, very low background signal, and good temporal and spatial 61

control (Richter, 2004). Ru(bpy)32+ and luminol are the most widely used ECL 62

systems. Besides, N-(4-aminobutyl)-N-ethylisoluminol (ABEI), a derivative of 63

isoluminol with similar ECL mechanism as luminol, is brought to our attention 64

because of its relatively high ECL efficiency as bioassay label compared with luminol 65

4

(Tian et al., 2009; Yang et al., 2002). Based on ECL, electrochemiluminescence 66

immunoassay (ECLIA) has gained much attention in recent years due to its wide 67

dynamic range, high sensitivity, low background, environmentally friendly labels, 68

simple instrumentation and easy methodology. In many cases, a sandwich-type is the 69

commonly used mode of ECLIA. In most of the previous related research work 70

focused on labeling materials and labeling methods such as: using 71

Ru(bpy)32+-encapsulated silica nanosphere (Yang et al., 2010a; Qian et al., 2010), 72

quantum dots functionalized graphene sheet (Liu et al., 2013), dendrimer multiply 73

labeled luminol on Fe3O4 nanoparticles (Li et al., 2013a), 74

N-(aminobutyl)-N-(ethylisoluminol)-functionalized gold nanoparticles (Shen et al., 75

2011; Tian et al., 2009), Au-MSN-HRP-Ab2 composites (Wei et al., 2010) and 76

Ru-AuNPs/graphene (Li et al., 2013b) as labels, basing on energy transfer between 77

quantum dots and quantum dots (Guo et al., 2012), between Ru(bpy)32+ and quantum 78

dots (Hao et al., 2012) and between quantum dots and gold nanoparticles (Qian et al., 79

2013), using quantum dots as labels and graphene as conducting bridge (Guo et al., 80

2013), etc. However, using conductive nanospheres multi-functionalized by the 81

second antibody and luminophore as labels was seldom reported. 82

Biological barcode technology was first reported by Mirkin and his colleagues 83

(Thaxton et al., 2009; Nam et al., 2002; Oh et al., 2006), in which gold nanoparticles 84

multi-functionalized with specific probes can identify target analyte specifically and a 85

large number of oligonucleotide strands. Those oligonucleotide strands with identical 86

sequences playing a role of surrogate target to amplify the detection sensitivity 87

5

effectively are termed as barcode. Due to the efficient amplification, the detection of 88

proteins and DNA by bio-barcode assay can reach the attomolar level (Goluch et al., 89

2006; Nam et al., 2004). However, this ultrasensitive method suffers from 90

cumbersome steps greatly. 91

Herein, we present a novel sandwich-type electrochemiluminescence (ECL) 92

immunosensor for the ultrasensitive detection of tumor markers based on biological 93

barcode mode, using conductive nanospheres multi-functionalized with the second 94

antibody and luminophore ABEI. Thousands of ABEI molecules were labeled as 95

barcode, and nearly all of them could emit ECL signals efficiently with the help of 96

conductive nanospheres. Therefore, the detection sensitivity was improved greatly, 97

with a detection range of 0.04 to 10 fg/mL using prostate specific antigen (PSA) as 98

model analyte. 99

100

101

2. Experimental 102

2.1. Apparatus 103

A laboratory-built ECL detection system, as described previously (Guo and Gai, 104

2011) was used in this study. A three-electrode system, including bare or modified 105

gold electrode (� = 3 mm), platinum wire electrode and Ag/AgCl electrode as 106

working electrode, counter electrode and reference electrode, respectively was used. 107

Electrochemical impedance spectroscopy (EIS) experiment was performed with a CHI 108

660B electrochemistry workstation (Chenhua Instrument Company, Shanghai, China). 109

6

The morphology of the nanospheres used was characterized using a SU70 scanning 110

electron microscope (SEM, Hitachi, Toyko, Japan). 111

2.2. Reagents and materials 112

N-(4-aminobutyl)-N-ethylisoluminol (ABEI), bovine serum albumin (BSA), 113

tetraethoxysilane (TEOS), glutaraldehyde (GLD) and 2-aminoethanethiol were 114

purchased from Sigma-Aldrich (St. Louis, MO, USA). Chloroauric acid 115

(HAuCl4·4H2O), (3-aminopropyl)-triethoxysilane (APS), sodium citrate, 116

hydroxylammonium chloride were obtained from Shanghai Chemical Reagent Co., 117

Ltd. (Shanghai, China). Prostate specific antigen (PSA), the first antibody anti-PSA 118

(Ab1) and the second antibody anti-PSA (Ab2) were purchased from Zhengzhou 119

Biocell Biotechnology Company (Zhengzhou, China), and stored at –20 °C before use. 120

Carbonate buffer solution (CBS, pH 9.6) containing 0.015 mol/L sodium carbonate 121

and 0.035 mol/L sodium bicarbonate, and 1 mmol/L H2O2, was used as the working 122

solution for the ECL measurement. All other reagents were of analytical grade. 123

Ultra-pure water (18 M� cm), obtained from a Heal Force PW ultrapure water system 124

(Heal Force Bio-Meditech Holdings Limited, Hong Kong, China), was used in the 125

experiment throughout. 126

2.3. Synthesis of conductive nanospheres (CNSs) 127

Amino-functionalized SiO2 nanospheres were prepared as described previously 128

(Stober and Fink, 1968; Pan et al., 2012; Jiao et al., 2012) with some modifications. 129

Briefly, ethanol, water and concentrated ammonia-water were mixed to about 100 mL 130

with a volume ratio of 88:8:1, and kept stirred. Then 4.5 mL of TEOS was added and 131

7

allowed to react for 10 h to obtain SiO2 nanospheres. After centrifugal washing with 132

water, the precipitate was dispersed in anhydrous ethanol to form a 50 mg/mL 133

suspension. Finally, amino-functionalized SiO2 nanospheres were obtained by adding 134

2 mL of APS into 20 mL of the above suspension and refluxing for 4 h at 80 °C in 80 135

mL of anhydrous ethanol with stirring, the morphology of which was monitored by 136

SEM, as shown in Fig. 1A. 137

Conductive nanospheres (CNSs) were synthesized according to the literatures 138

(Hu et al., 2005; Wei et al., 2010). A 400 mL solution containing 0.0125 mg/mL gold 139

nanoparticles (AuNPs) with a diameter of 13 nm, prepared as described previously 140

(Polte et al., 2010), and 0.8 mg/mL of amino-functionalized SiO2 nanospheres was 141

stirred for 12 h, washed until no AuNPs could be found in the supernatant, and then 142

aged for 5 d at 4 °C. Subsequently, 20 mL of the obtained solution was mixed with 143

300 mL of HAuCl4/K2CO3 solution containing 0.18 mg/mL HAuCl4 and 0.25 mg/mL 144

K2CO3, followed by adding 2 mmol/L hydroxylammonium chloride aqueous solution 145

till the color of the mixture turned from colorless to reddish brown to obtain the CNSs. 146

The morphology of the obtained CNSs with a diameter of 60 – 90 nm was monitored 147

by SEM and is presented Fig. 1B. 148

149

150

2.4. Preparation of ABEI/Ab2-CNSs probes 151

The schematic diagram of the preparation process of ABEI/Ab2-CNSs probes is 152

illustrated in Fig. 2A. Firstly, a 1 mL mixture of 3 mg/mL CNSs and 7.7 mg/mL of 153

Preferred position for Fig. 1

8

2-aminoethanethiol were incubated for 10 h at room temperature and then centrifuged. 154

Then, the precipitate was dispersed and incubated in 1 mL of 0.12 mg/mL GLD for 1 155

h at room temperature and centrifuged, and the precipitate was then dispersed in 1 mL 156

mixture solution containing 0.2 mg/mL ABEI and 8 �g/mL Ab2. After incubating for 157

another 1 h at room temperature and centrifugation, ABEI/Ab2-CNSs probes were 158

finally obtained. 159

Non-conductive probes were prepared using the same procedures as described in 160

Section 2.4 using amino-functionalized SiO2 nanospheres instead of conductive 161

nanospheres (Fig. 2B). 162

2.5. Fabrication of the ECL immunosensor 163

Fig. 2C presents the fabrication protocol of the ECL immunosensor. Firstly, the 164

bare gold electrode was polished with 1.0, 0.3 and 0.05 �m Al2O3 slurries in sequence, 165

ultrasonicated, and scanned from 0 to 1.6 V in 0.5 mol/L H2SO4 until stable cyclic 166

voltammograms were obtained. Subsequently, the cleaned gold electrode was rinsed 167

and incubated in 0.1 mol/L 2-aminoethanethiol for 10 h at 4 °C. After rinsing, the 168

electrode was incubated in 2.5% GLD for 1 h at 4 °C, rinsed, and incubated in 50 169

�g/mL Ab1 for 12 h at 4 °C. Finally, the electrode obtained was rinsed and incubated 170

in 2% BSA for 1.5 h at 4 °C to block non-specific binding sites. The ECL 171

immunosensor thus obtained was ready for the immunoassay.172

2.6. ECL Detection 173

As shown in Fig. 2C, the ECL immunosensor was incubated in 50 μL of a 174

sample solution containing PSA for 40 min at 37 °C, thoroughly washed with 0.05 175

9

mol/L CBS to remove unbound PSA, and then incubated in 50 �L of ABEI/Ab2-CNSs 176

probes or non-conductive probes at 37 °C for 40 min to form the final sandwich 177

immunocomplex. Then, the electrodes were scanned using cyclic voltammetry in 0.05 178

mol/L CBS containing 1 mmol/L H2O2 from 0 to 1 V with a scan rate of 100 mV/s, 179

and the ECL signals were recorded for measurement. 180

181

182

2.6. Determination of PSA in real samples 183

Samples of human serum and saliva were collected and carefully determined 184

using this method. Samples, in which PSA could not be detected, were selected as 185

blank samples. Real samples were obtained by spiking standard PSA solution in blank 186

human serum and saliva samples, and then used for the assay as mentioned above. 187

188

189

3. Results and discussion 190

3.1. Construction and characteristics of the ECL immunosensor based on biological 191

barcode mode 192

The schematic diagram of the assay process for the proposed ECL 193

immunosensor based on biological barcode mode is depicted in Fig. 2. In case of 194

non-conductive probe (Fig. 2B), thousands of ABEI molecules and the second 195

antibody were immobilized simultaneously on the surface of amino-functionalized 196

SiO2 nanospheres. However, the ECL signal was low (curve b in Fig. 2), because the 197

Preferred position for Fig. 2

10

diameters of amino-functionalized SiO2 nanospheres were tens of nanometers at least, 198

which were much larger than the thickness of outer Helmholtz plane (OHP). OHP is 199

considered as the position of electronic exchange between solvated ions and the 200

electrode in electrode kinetics, i.e., OHP is the place where electrochemically active 201

species have to reach. Electrochemical reactions would occur when the distance 202

between electrochemically active species and the electrode is no more than that 203

between OHP and the electrode; otherwise no electrochemical reaction would occur 204

(Bard and Faulkner, 2000). Therefore, most of ABEI molecules immobilized on the 205

surface of amino-functionalized SiO2 nanospheres were beyond OHP, and they could 206

not emit ECL signals, termed as ‘ineffective ABEI’. Therefore, it is understandable 207

that the ECL signal was not so high. As to ABEI/Ab2-CNSs probe (Fig. 2A), 208

thousands of ABEI molecules and the second antibody were immobilized 209

simultaneously on the surface of conductive nanospheres. After the sandwich 210

immunocomplex was constructed, most of ABEI/Ab2-CNSs probes were close to or 211

even contacted the surface of the electrode. Thus, OHP would be extended due to the 212

existence of gold layer on the surface of conductive nanospheres, similar to that the 213

electrode surface was extended. Under such a circumstance, nearly all ABEI 214

molecules immobilized on the surface of conductive nanospheres were in the area 215

between the OHP and the electrode, and they could emit ECL signals, termed as 216

‘effective ABEI’. Therefore, at the same concentration of tumor marker PSA, the ECL 217

intensity was improved about 7 times (curve a in Fig. 2). 218

11

Compared with conventional sandwich-type ECLIA method, in which 219

luminophore is labeled directly on the second antibody, this proposed ECL 220

immunosensor based on biological barcode mode has two significant advantages: (1) 221

The sensitivity could be improved greatly. An antigen molecule corresponds to the 222

ECL signal emitted by thousands of luminophore molecules labeled as barcode on the 223

surface of conductive nanospheres. In contrast, in the conventional method, an antigen 224

molecule corresponds to the ECL signal emitted by no more than tens of luminophore 225

molecules. (2) The immunoactivity of the second antibody would not be affected 226

because luminophore molecules were labeled on the surface of conductive 227

nanospheres. In case of conventional one, at most tens of luminophore molecules 228

could be labeled to maintain the immunoactivity of the second antibody (Fung and 229

Wong, 2001; Pei et al., 2013; Zhang et al., 2008). In addition, this proposed ECL 230

immunosensor based on biological barcode mode was much easier than biological 231

barcode immunoassay, because luminophore as barcode could give real-time ECL 232

signal while the latter must employ cockamamie PCR amplification steps. 233

Electrochemical impedance spectroscopy (EIS) was employed to monitor the 234

interface properties of the electrodes in the assembly process. The impedance 235

spectrum comprises a line at low-ac modulation frequency and a semicircle at high-ac 236

modulation frequency, while the latter indicates the electron transfer resistance Ret. As 237

shown in Fig. 3A, Ret increased a little compared with the bare gold electrode (curve a) 238

when 2-aminoethanethiol and GLD was assembled (curve b). Then, Ret increased 239

remarkably when Ab1, BSA, PSA and non-conductive probe were immobilized 240

12

successively (curves c, d, e and f), due to their gradual increase of hindrance to the 241

interfacial electron transfer (Li et al., 2008). Finally, in contrast to non-conductive 242

probes, Ret decreased obviously when ABEI/Ab2-CNSs probes were assembled 243

successfully on the modified electrode (curve g) owing to the gold layer coated on the 244

surface of CNSs that improved the electrical conductivity of the electrode interface 245

and reduced Ret effectively. 246

The ECL behaviors of the immunosensor were recorded step by step in 0.05 247

mol/L pH 9.6 CBS containing 1.0 mmol/L H2O2. As shown in Fig. 3B, no ECL signal 248

was found when 2-aminoethanethiol, GLD, Ab1, BSA and PSA were immobilized on 249

the electrode successively (curves a-e), due to the absence of luminophore. When 250

non-conductive probes were connected to PSA/Ab1/GLD/2-aminoethanethiol/Au 251

electrode, ECL signal was observed owing to the ECL reaction of ABEI molecules 252

(curve f). However, the ECL intensity was not high because the nanospheres were so 253

big that most of ABEI molecules immobilized on them were far away from the 254

electrode surface and beyond the space domain of the OHP. With conductive 255

nanospheres (curve g), the ECL intensity was increased about 7 times, attributing to 256

the gold layer coated on the surface of CNSs that extend the OHP effectively, and 257

thus nearly all the ABEI molecules immobilized on CNSs participated in ECL 258

reactions. 259

260

261

3.2. Optimization of experimental conditions 262

Preferred position for Fig. 3

13

The performance of the immunosensor mainly depends on pH value of working 263

solution, concentration of H2O2, incubation time and time interval between two cyclic 264

voltammetric periods. To obtain an optimal ECL signal, the effects of the above 265

factors were investigated by detecting PSA solution at the concentration of 0.4 fg/mL. 266

The effect of pH on the ECL intensity was examined in the range of 8.94 – 9.90 267

in 0.05 mol/L CBS. As shown in Fig. 4A, the maximum ECL intensity was obtained 268

when the pH was 9.6. The ECL reaction of ABEI in alkaline solution was markedly 269

improved by the addition of H2O2 (Fig. 4B). The reason is that ABEI deprotonates in 270

alkaline solution to form an anion that can undergo electrochemical oxidation. The 271

intermediate species obtained undergoes further electro-oxidation in the presence of 272

H2O2 to produce an excited state, which produces the ECL emission finally (Richter, 273

2004; Arai et al., 1999). In addition, the presence of H2O2 not only enhanced the 274

sensitivity of the ECL reaction greatly, but also made the ECL reaction perform at a 275

relatively low potential (Yang et al., 2002). Therefore, the ECL intensity increased 276

with the increase of the H2O2 concentration and reached the maximum at 1 mmol/L. 277

The effect of incubation time was also investigated (Fig. 4C). When the incubation 278

time was longer than 40 min, the ECL intensity did not increase with the increasing 279

incubation time because the reaction was almost completed. Finally, the time interval 280

between two cyclic voltammetric periods was examined, and the results are shown in 281

Fig. 4D. The ECL intensity decreased when the time interval was 40 s or 50 s; 282

whereas, the ECL signal tended to be stable when the time interval was 60 s or more, 283

due to more effective diffusion of H2O2 in a longer time interval. Therefore, the 284

14

optimal experimental conditions of this assay were as follows: 0.05 mol/L CBS at pH 285

9.6 containing 1.0 mmol/L H2O2, 40 min incubation time and 60 s time intervals 286

between two cyclic voltammetric periods. 287

288

289

3.3. Sensitivity and linear range 290

As presented in Fig. 5A, the response ECL intensity increased with an increase 291

in the concentration of PSA (curve a�j). A good linear relationship between the ECL 292

intensity (y) and the logarithm of the analyte PSA concentrations (CPSA) in the range 293

from 0.04 to 10 fg/mL (n = 5) was observed (Fig. 5B). The regression equation was y 294

= 1480.48 + 370.29*log CPSA (fg/mL) with a correlation coefficient r of 0.9987; 295

indicating that the proposed immunosensor has an excellent analytical performance, 296

including very low detection level and wide linear ranges over two orders. To our best 297

knowledge, the sensitivity of this method for PSA is much higher than the lowest 298

detection limit of 0.6 pg/mL reported in the previous literatures (Choi et al., 2013; 299

Ahmed and Azzazy, 2013; Li et al., 2013c; Dey et al., 2012). 300

301

302

3.4. Specificity, stability, reproducibility and regeneration of the immunosensor 303

The ECL immunosensor was considered of having a high specificity for PSA, 304

which depended on the specific binding between PSA and its corresponding antibody. 305

Various species of interfering proteins including human immunoglobulin (hIgG), 306

Preferred position for Fig. 4

Preferred position for Fig. 5

15

BSA, alfa-fetoprotein (AFP) and carcino-embryonic antigen (CEA) were used to 307

further investigate the specificity of the proposed immunosensor. Results showed that 308

no ECL signal could be found when hIgG, BSA, AFP and CEA at the concentration 309

of 1 �g/mL were detected. Therefore, the specificity of the immunosensor developed 310

was acceptable. 311

The stability of the immunosensor stored in 0.01 M PBS (pH 7.4) containing 312

0.1% NaN3 at 4 °C was investigated by periodical checking of its relative activity. 313

The initial value of ECL signal for 1 fg/mL PSA was obtained when the 314

immunosensors were constructed freshly. After a storage period of a month, the ECL 315

intensity for the detection of 1 fg/mL PSA was 90.5% of the initial value. Thus, the 316

immunosensor has acceptable storage stability. 317

The reproducibility of the immunosensor was evaluated by detecting 1 fg/mL 318

PSA with five equally prepared immunosensors. The relative standard deviation (RSD) 319

of the measurements for the five immunosensors was 6.9%, indicating the excellent 320

precision and reproducibility of the immunosensor. 321

Regeneration of the immunosensor was examined by detecting 1 fg/mL PSA 322

with a same immunosensor. The immunosensor was regenerated by dipping into 0.2 323

mol/L glycine-hydrochloric acid (Gly-HCl) buffer solution (pH 2.8) for 8 min to 324

break the antibody-antigen linkage. The consecutive measurements were repeated ten 325

times, an average recovery of 91.8% and an intra-assay RSD of 8.2% were acquired. 326

The results demonstrated that the proposed immunosensor could be regenerated and 327

used for at least ten times. 328

16

3.5. Application of ECL immunosensor in human serum and saliva samples 329

Determination of the low concentrations of tumor marker in the human serum 330

and saliva is very important in the early diagnosis of cancer. In order to investigate the 331

applicability and reliability of the prepared ECL immunosensor for clinical 332

applications, recovery experiments were performed by detection of PSA spiked in 333

human serum and saliva samples. As shown in Table 1, an acceptable recovery 334

obtained was in the range of 97.3 111.3%, indicating that the developed ECL 335

immunosensor is an efficient tool for ultrasensitive determination of PSA in human 336

serum and saliva. 337

338

339

340

341

4. Conclusions 342

In this work, an ultrasensitive ECL immunosensor based on biological barcode 343

mode for the detection of very low concentrations of the tumor marker PSA was 344

constructed. ABEI/Ab2-CNSs probes was used to (1) recognize tumor markers with 345

Ab2, (2) emit ECL signal with ABEI acting as barcode, and (3) extend OHP with gold 346

layer coated to make all ABEI immobilized effective. In such a case, a tumor marker 347

antigen corresponded to the ECL signal emitted by thousands of ABEI molecules, 348

lowering the detection limit down to fg/mL level. In addition, the specificity, stability, 349

reproducibility, regeneration and application of this immunosensor were validated. 350

Preferred position for Table 1

17

Therefore, the proposed immunosensor based on biological barcode mode not only 351

built a potential detection plateau for disease-related proteins in the clinical 352

diagnostics, but also opened a new avenue to label a large number of tags in 353

immunoassay. 354

355

356

Acknowledgments 357

Financial supports from National Natural Science Foundation of China 358

(81273130, 81072336), Science and Technology Department of Zhejiang Province of 359

China (2012R405061, 2012C23101), Ningbo Science and Technology Bureau 360

(2011C50037), Key project of Shenzhen Polytechnic (2210K3070014) and 361

“Qianbaishi Candidate” fund for Higher Education of Guangdong Province are 362

gratefully acknowledged. This work was also sponsored by K.C. Wong Magna Fund 363

in Ningbo University. 364

365

366

References 367

Ahmed, H.A., Azzazy, H.M.E., 2013. Biosensors and Bioelectronics 49, 478-484. 368

Arai, K., Takahashi, K., Kusu, F., 1999. Analytical Chemistry 71, 2237-2240. 369

Bard, A.J., Faulkner, L.R., 2000. Electrochemical Methods: Fundamentals and 370

Applications, second ed. John Wiley and Sons, New York. 371

Chalupova, L., Zakovska, A., Adamcova, K., 2013. Clinical Biochemistry 46, 73-78. 372

18

Choi, J.H., Kim, H.S., Choi, J.W., Hong, J.W., Kim, Y.K., Oh, B.K., 2013. 373

Biosensors and Bioelectronics 49, 415-419. 374

Darwish, I.A., Wani, T.A., Alanazi, A.M., Hamidaddin, M.A., Zargar, S., 2013. 375

Talanta 111, 13-19. 376

Dey, A., Kaushik, A., Aryac, S.K., Bhansali, S., 2012. Journal of Materials Chemistry 377

22, 14763-14772. 378

Dong, J.X., Li, Z.F., Lei, H.T., Sun, Y.M., Ducancel, F., Xu, Z.L., Boulain, J.C., 379

Yang, J.Y., Shen, Y.D., Wang, H., 2012. Analytica Chimica Acta 736, 85-91. 380

Freedland, S.J., 2011. Cancer 117, 1123-1135. 381

Fung, Y.S., Wong, Y.Y., 2001. Analytical Chemistry 73, 5302-5309. 382

Goluch, E.D., Nam, J.M., Georganopoulou, D.G., Chiesl, T.N., Shaikha, K.A., Ryu, 383

K.S., Barron, A.E., Mirkin, C.A., Liu, C., 2006. Lab chip 6, 1293-1299. 384

Guo, Z.Y., Gai., P.P., 2011. Analytica Chimica Acta 688, 197-202. 385

Guo, Z.Y., Hao, T.T., Wang, S., Gan, N., Li, X., Wei, D.Y., 2012. Electrochemistry 386

Communications 14, 13-16. 387

Guo, Z.Y., Hao, T.T., Du, S.P., Chen, B.B., Wang, Z.B., Li, X., Wang, S., 2013. 388

Biosensors and Bioelectronics 44, 101-107. 389

Hao, T.T., Guo, Z.Y., Du, S.P., Shi, L.L., 2012. Sensors and Actuators B 171-172, 390

803-809. 391

Hou, J.Y., Liu, T.C., Lin, G.F., Li, Z.X., Zou, L.P., Li, M., Wu, Y.S., 2012. Analytica 392

Chimica Acta 734, 93-98. 393

19

Hu, Y.H., Rong, J.H., Liu Y.L., Man, S.Q., 2005. Chinese Journal of Inorganic 394

Chemistry 21, 1672-1676. 395

Jiao, Y.F., Guo, J., Shen, S., Chang, B.S., Zhang, Y.H., Jiang, X.G., Yang, W.L., 396

2012. Journal of Materials Chemistry 22, 17636-17643. 397

Li, B.L., Wang, Y.L., Wei, H., Dong, S.J., 2008. Biosensors and Bioelectronics 23, 398

965-970. 399

Li, J.P., Xu, Q., Fu, C., Zhang, Y., 2013a. Sensors and Actuators B 185, 146-153. 400

Li, M., Zhang, M., Ge, S.G., Yan, M., Yu, J.H., Huang, J.D., Liu, S., 2013b. Sensors 401

and Actuators B 181, 50-56. 402

Li, W.P., Li, M., Ge, S.G., Yan, M., Huang, J.D., Yu, J.H., 2013c. Analytica Chimica 403

Acta 767, 66-74. 404

Liu, W.Y., Zhang, Y., Ge, S.G., Song, X.R., Huang, J.D., Yan, M., Yu J.H., 2013. 405

Analytica Chimica Acta 770, 132-139. 406

Liu, Z.M., Ma, Z.F., 2013. Biosensors and Bioelectronics 46, 1-7. 407

Lu, J., Zou, L., Xiong, C., Wang, Z., Ma, H., Yang, W., Cui, T., Zhang, L., Huang, S., 408

Sun, Q., Gao, J.M., Meng, Q.H., 2012. Clinical Biochemistry 45, 1113. 409

Marta, H.V., Meritxel, G., Antoni, C., Luis, B., 2013. Clinica Chimica Acta 418, 410

22-29. 411

Nam, J.M., Park, O.J., Mirkin, C.A., 2002. Journal of the American Chemical Society 412

124, 3820-3821. 413

Nam, J.M., Stoeva, S.I., Mirkin, C.A., 2004. Journal of the American Chemical 414

Society 126, 5932-5933. 415

20

Niu, C.G., Liu, J., Qin, P.Z., Zeng, G.M., Ruan, M., He, H., 2011. Analytical 416

Biochemistry 409, 244-248. 417

Oh, B.K., Nam, J.M., Lee, S.W., Mirkin, C.A., 2006. Small 2, 103-108. 418

Pan, L.M., He, Q.J., Liu, J.N., Chen, Y., Ma, M., Zhang, L.L., Shi, J.L., 2012. Journal 419

of American Chemical Society 134, 5722-5725. 420

Pei, X.M., Zhang, B., Tang, J., Liu, B.Q., Lai, W.Q., Tang, D.P., 2013. Analytica 421

Chimica Acta 758, 1-18. 422

Polte, J., Ahner, T.T., Delissen, F., Sokolov, S., Emmerling, F., Thunemann, A.F., 423

Kraehnert, R., 2010. Journal of American Chemical Society 132, 1296-1301. 424

Qian, J., Wang, C.Q., Pan, X.H., Liu, S.Q., 2013. Analytica Chimica Acta 763, 43-49. 425

Qian, J., Zhou, Z.X., Cao, X.D., Liu, S.Q., 2010. Analytica Chimica Acta 665, 32-38. 426

Qu, S.X., Liu, J.T., Luo, J.P., Huang, Y.Q., Shi, W.T., Wang, B., Cai, X.X., 2013. 427

Analytica Chimica Acta 766, 94-99. 428

Richter, M.M., 2004. Chemical Reviews 104, 3003-3036. 429

Saadi, H., Seillier, M., Sandi, M.J., Peuget, S., Kellenberger, C., Gravis, G., Dusetti, 430

N.J., Iovanna, J.L., Rocchi, P., Amri, M., Carrier, A., 2013. Results in 431

Immunology 3, 51-56. 432

Shen, W., Tian, D.Y., Cui, H., Yang, D., Bian, Z.P., 2011. Biosensors and 433

Bioelectronics 27, 18-24. 434

Stober, W., Fink, A., 1968. Science 26, 62-69. 435

Su, B.L., Tang, D.P., Li, Q.F., Tang, J., Chen, G.N., 2011. Analytica Chimica Acta 436

692, 116-124. 437

21

Thaxton, C.S., Elghanian, R., Thomas, A.D., Stoeva, S.I., Lee, J.S., Smith, N.D., 438

Schaeffer, A.J., Klocker, H., Horninger, W., Bartsch, G., Mirkin, C.A., 2009. 439

Proceeding of National Academy of Sciences 106, 18437-18442. 440

Tian, D.Y., Duan, C.F., Wang, W., Li, N., Zhang, H., Cui, H., Lu, Y.Y., 2009. Talanta 441

78, 309-404. 442

Tian, J.N., Liu, R.J., Zhao, Y.C., Peng, Y., Hong, X., Xu, Q., Zhao, S.L., 2010. 443

Nanotechnology 21, 305101. 444

Tyan, Y.C., Jong, S.B., Yang, M.H., Wu, T.L., Chung, T.W., Liao, P.C., Lin, C.Y., 445

Huang, Y.F., 2013. Clinica Chimica Acta 420, 99-103. 446

Wei, Q., Xin, X.D., Du, B., Wu, D., Han, Y.Y., Zhao, Y.F., Cai, Y.Y., Li, R., Yang, 447

M.H., Li, H., 2010. Biosensors and Bioelectronics 26, 723-729. 448

Wickström, M., Larsson, R., Nygren, P., Gullbo, J., 2011. Cancer Science 102, 449

501-508. 450

Xiang, Y., Lu, Y., 2012. Analytical Chemistry 84, 4174-4178. 451

Xiao, Q., Li, H.F., Hu, G.M., Wang, H.R., Li, Z.J., Lin, J.M., 2009. Clinical 452

Biochemistry 42, 1461-467. 453

Yang, M., Liu, C., Qian, K., He, P., Fang, Y., 2002. Analyst 127, 1267-1271. 454

Yang, X., Yuan, R., Chai, Y.Q., Zhuo, Y., Mao, L., Yuan, S.R., 2010a. Biosensors 455

and Bioelectronics 25, 1851-1855 456

Yang, X.Y., Guo, Y.S., Wang, A.G., 2010b. Analytica Chimica Acta 666, 91-96. 457

Zhang, X.J., Ju, H.X., Wang, J., 2008. Electrochemical Sensors, Biosensors and Their 458

Biomedical Applications, first ed. Academic Press, San Diego. 459

460

22

Figure legend 461

462

Fig. 1. The SEM image of (A) SiO2 nanospheres and (B) SiO2/Au conductive 463

nanospheres (CNSs). 464

Fig. 2. The schematic diagram for (A) the preparation of the ABEI/Ab2-CNSs probe 465

with SiO2/Au conductive nanosphere, (B) the preparation of the 466

non-conductive probe with amino-functionalized SiO2 nanosphere and (C) the 467

fabrication of the ECL immunosensor using the ABEI/Ab2-CNSs probe and 468

the non-conductive probe respectively. 469

Fig. 3. EIS (A) and ECL profiles (B) of (a) bare Au electrode, (b) (a) + 470

2-aminoethanethiol + GLD, (c) (b) + Ab1, (d) (c) + BSA, (e) (d) + PSA (1 471

fg/mL), (f) (e) + non-conductive probes, and (g) (e) + ABEI/Ab2-CNSs probes. 472

Experimental conditions of EIS: 5 mmol/L Fe(CN)64-/3- solution (0.05 mol/L 473

PBS, pH 7.0); frequency range: between 0.01 and 100,000 Hz; signal 474

amplitude: 5 mV. Experimental conditions of ECL: 0.05 mol/L CBS at pH 9.6 475

containing 1.0 mmol/L H2O2; 40 min incubation time; and 60 s time intervals 476

between two cyclic voltammetric periods. 477

Fig. 4. Effect of (A) pH value; (B) H2O2 concentration; (C) incubation time; and (D) 478

time interval between two cyclic voltammetric periods on the ECL intensity. 479

Fig. 5. (A) ECL profiles of the immunosensor in the presence (a�j) of different 480

concentrations of PSA (fg/mL): (a) 0.04; (b) 0.1; (c) 0.2; (d) 0.3; (e) 0.4; (f) 1; 481

(g) 2; (h) 3; (i) 4; (j) 10. (B) Calibration curve for PSA determination. The 482

23

experimental conditions were as follows: 0.05 mol/L CBS at pH 9.6 483

containing 1.0 mmol/L H2O2, 40 min incubation time and 60 s time intervals 484

between two cyclic voltammetric periods. 485

Table 1 Recovery tests for PSA in spiked human serum and saliva samples ( sx � , n = 486

3). 487

488

24

Table 1 Recovery tests for PSA in spiked human serum and saliva samples ( sx � , n = 489

3). 490

Samples Added (fg/mL) Found (fg/mL) Recovery (%)

Serum 1 0.040 0.0445 ± 0.0046 111.3

Serum 2 0.40 0.423 ± 0.040 105.8

Serum 3 4.0 3.89 ± 0.26 97.3

Saliva 1 0.040 0.0432 ± 0.0041 108.0

Saliva 2 0.40 0.427 ± 0.038 106.8

Saliva 3 4.0 4.11 ± 0.22 102.8

491 492

493

25

� An ECL immunosensor based on biological barcode mode was proposed. 494 � Conductive nanospheres (CNSs) could extend the outer Helmholtz plane 495

effectively. 496 � ABEI/Ab2-CNSs probes were constructed by immobilizing ABEI and Ab2497

on CNSs. 498 � ABEI on CNSs give in-situ ECL signal as biological barcode without 499

interfering Ab2.500 � PSA could be detected in the concentration range from 0.04 to 10 fg/mL. 501

502

Figure 1

Figu

re 2

Figure 3

Figu

re 4

Figu

re 5

Figure 1-black and white

Figu

re 2

-bla

ck a

nd w

hite

Figure 3-black and white

Figu

re 4

-bla

ck a

nd w

hite

Figu

re 5

-bla

ck a

nd w

hite