immunochromatographic electrochemical biosensor for rapid and sensitive detection of prostate-specific antigen

Available online at www.sciencedirect.com

Biosensors and Bioelectronics 23 (2008) 1659–1665

A nanoparticle label/immunochromatographic electrochemical biosensor for rapid and sensitive detection of prostate-specific antigen Ying-Ying Lin a,b , Jun Wang a , Guodong Liu a , Hong Wu a , C.M. Wai b , Yuehe Lin a,∗ a

b

Pacific Northwest National Laboratory, Richland, WA 99352, United States Department of Chemistry, University of Idaho, Moscow, ID 83843, United States

Received 15 October 2007; received in revised form 13 January 2008; accepted 29 January 2008 Available online 21 February 2008

Abstract We present a nanoparticle (NP) label/immunochromatographic electrochemical biosensor (IEB) for rapid and sensitive detection of prostatespecific antigen (PSA) in human serum. This IEB integrates the immunochromatographic strip with the electrochemical detector for transducing quantitative signals. The NP label, made of CdSe@ZnS, serves as a signal-amplifier vehicle. A sandwich immunoreaction was performed on the immunochromatographic strip. The captured NP labels in the test zone were determined by highly sensitive stripping voltammetric measurement of the dissolved metallic component (cadmium) with a disposable-screen-printed electrode, which is embedded underneath the membrane of the test zone. Several experimental parameters (e.g., immunoreaction time, the amount of anti-PSA-NP conjugations applied) and electrochemical detection conditions (e.g., preconcentration potential and time) were optimized using this biosensor for PSA detection. The analytical performance of this biosensor was evaluated with serum PSA samples according to the “figure-of-merits” (e.g., dynamic range, reproducibility, and detection limit). The results were validated with enzyme-linked immunosorbent assay (ELISA) and showed high consistency. It is found that this biosensor is very sensitive with the detection limit of 0.02 ng mL−1 PSA and is quite reproducible (with a relative standard deviation (R.S.D.) of 6.4%). This method is rapid, clinically practical, and less expensive than other diagnostic tools for PSA; therefore, this IEB coupled with a portable electrochemical analyzer shows great promise for simple, sensitive, quantitative point-of-care testing of disease-related protein biomarkers. © 2008 Published by Elsevier B.V. Keywords: Nanoparticles; Electrochemical immunosensor; Test strip; Prostate-specific antigen

1. Introduction Prostate cancer (PCa) has become one of the most frequently diagnosed cancers and the third leading cause of cancer morbidity and mortality among males in the United States (Jemal et al., 2006). Therefore, early, definitive, and sensitive diagnosis of PCa is needed to initiate therapy to avoid the worsening development of the disease. It has been shown that serum prostate-specific antigen (PSA) is the most reliable tumor marker to detect PCa at the early stage and to monitor the recurrence of the disease after treatment (Benson et al., 1992; Bradford et al., 2006; Brawer, 1999; Stephan et al., 2006). Currently, most PSA testings take place at dedicated centralized laboratories using large, automated analyzers, requiring sample transportation, increased waiting time and increased administration and medi-

∗

Corresponding author. Tel.: +1 509 376 0529; fax: +1 509 376 5106. E-mail address: [email protected] (Y. Lin).

0956-5663/$ – see front matter © 2008 Published by Elsevier B.V. doi:10.1016/j.bios.2008.01.037

cal costs (Acevedo et al., 2002; Healy et al., 2007). Near-patient or point-of-care testing (POCT) is highly needed to reduce the number of clinic visits, decrease costs to the patient and the healthcare system, increase patient satisfaction and improve clinical outcome. Recent advances in biosensor development based on nanomaterials and nanostructures as integral components have brought POCT for PSA closer to reality. The immunosensors based on various labels, e.g., enzyme, DNA, nanoparticle, carbon nanotubes, have been developed for PSA diagnosis (Healy et al., 2007). For example, Lind and Kubista (2005) reported DNA–antibody conjugate-based immuno-PCR for sensitivity detection of PSA. Nam et al. (2003) have developed a novel nanoparticle-based bio-barcode for ultrasensitive detection of PSA. Electric detection using an antibody modified microcantilever and antibody coated silicon nanowire field-effect sensor have been reported for diagnosis of PSA (Wee et al., 2005; Zheng et al., 2005). The optical immunosensor based on a fluorescence label or Raman reporter, label-free surface plasma

1660

Y.-Y. Lin et al. / Biosensors and Bioelectronics 23 (2008) 1659–1665

resonance (SPR), and the electrochemical immunosensor based on the nanomaterial label (e.g., nanoparticles and carbon nanotubes), have also been reported recently (Grubisha et al., 2003; Huang et al., 2005; D’Orazio, 2003; Sarkar et al., 2002; Yu et al., 2006; Wang et al., 2008). In general, these approaches are quite sensitive, for example, the detection limit of immuno-PCR method can be very low (0.2 pg mL−1 PSA). However, most of them need a long assay time, e.g., 60–150 min and sophisticated instruments. The single-step lateral-flow immunochromatographic assay that combines chromatography with immunoassay has attracted great interest for developing simple, rapid, and sensitive diagnostic tools. The well-known principle of the assay is based on the migration of samples and reagents along antibody-coated membrane strips where the corresponding affinity interaction takes place, and the analyte can be detected in just a few minutes. This technique minimizes analysis time (avoiding a long immunoreaction time and multiple washing steps) and provides an easy, rapid, and less-expensive immunoassay of biomarkers. It has been successfully used for in-field biomarker detection and clinical diagnosis of biospecimens (Cuzzubbo et al., 2001; Jin et al., 2005). The first stage of this method is based on visible judgment (visual or colorimetric detection) to qualify the analyte; dyes or gold nanoparticles (NPs) are used, and this enables rapid and qualitative analysis (Jin et al., 2005; Nagatani et al., 2006; Zhang et al., 2006; Fernandez-Sanchez et al., 2005). However, these approaches could not be used for accurate quantitative diagnosis. An immunochromatographic strip in connection with a fluorescence detector has been reported to quantify an analyte in aqueous samples (Kim et al., 2003). This approach offers a greater sensitivity and dynamic range as well as a better quantitative capability than those based on just visual judgment. However, these approaches suffered from optical interference (e.g., photobleaching), the liquid effect in the chromatographic test-strip or lacking of direct quantitative data. Electrochemical immunoassays and immunosensors combined with immunochromatographic test strip are ideally suited for decentralized point-of-care testing or field detection of bioagents due to its high sensitivity, miniaturization, low cost, and less power requirement. We have reported immunochromatographic electrochemical biosensor (IEB) for detection of human chorionic gonadotronphin (HCG) based on metal ion labels (Lu et al., 2005). Due to the small number of metal ion labels per antibody (8–10 metal ions per antibody), the sensitivity of this approach is not good enough for certain applications where the biomarker concentration at biological samples is very low. Fernandez-Sanchez et al. (2004) integrated a lateral flow immunoassay format with impedance detection using electrochemical transducer coated with pH-sensitive polymer layer to complete the detection of free and total PSA. Although the lowest detection level is 3 ng mL−1 , it requires a relatively long analysis time (e.g., 30 min). Recently, nanoparticle (NP) based electrochemical biosensors and bioassays have shown great promise for detection of trace biomolecules because of versatile amplification approaches (Bao et al., 2006; Dequairem et al., 2000; Georganopoulou et al., 2005; Huhtinen et al., 2004; Jain, 2005; Liu et al., 2006; Liu and Lin, 2007; Liu et al., 2007; Nam et

al., 2003; Wang et al., 2006). However, nanoparticle label-based electrochemical immunoassay integrated with test strip has less been reported for assay of biomarkers. In this report, a NP label/immunochromatographic/electrochemical biosensor (IEB) for rapid and sensitive detection of the PSA in serum samples is presented. Quantum dots (QDs) NPs made of CdSe@ZnS are used to label anti-PSA antibodies for amplifying signal output. Since one quantum dot nanoparticle contains over ten thousands of metal ions, the voltammetric signal using QD labels is greatly enhanced compared with metal ion label. A covalent binding approach to immobilize the second antibody in the test zone has also been developed in this study. This IEB, integrating chromatographic separation, immunoassay, and electrochemical detection techniques with advanced nanotechnologies, offers a good option for rapid and sensitive detection of biomarkers. 2. Materials and methods 2.1. Chemicals and materials Diaminoheptane, phosphate buffer saline, bovine serum album (BSA), Tween-20, and glutaraldehyde were purchased from Sigma. Mouse monoclonal PSA antibody pairs (MOT40081A and MO-T40081 B), standard serum PSA, and the control (serum without PSA) were obtained from Anogen (Mississauga, Canada). Human PSA ELISA kit (Cat. No. 1500) was purchased from Alpha Diagnostic International (San Antonio, TX, USA). Human-serum samples were obtained from Golden West Biologicals (Temecula, CA, USA). Polyester backing materials, glass fibers, and absorbent materials were purchased from Millipore (Bedford, MA, USA). Nitrocellulose membrane (AE 98) was purchased from Whatman Inc. (Florham Park, NJ, USA). A CdSe@ZnS QD and antibody conjugation kit was purchased from Molecular Probes Inc. (Eugene, OR, USA). All chemicals were used without further purification. All solutions were prepared with ultra-pure water (18.3 M cm, Nanopure, Barnstead, USA) or autoclaved water in this study. 2.2. The preparation of QD–anti-PSA conjugates The anti-PSA–QD conjugates were prepared by following the QD conjugation kit protocol. The details are described in the supplementary data or can be found in our previous report (Wu et al., 2007; Wang et al., 2008). 2.3. Preparation of immunochromatographic electrochemical biosensor Fig. 1 is a schematic diagram of the immunochromatographic/electrochemical biosensor. This IEB integrates the immunochromatographic test strip with an electrochemical detector. The test strip (A) is housed in a chamber consisting of a cover (B) and a bottom (C). The test strip has four discrete zones. The first zone is the sample loading zone, which consists of a sample application pad. The sample application pad was made from glass fiber. The pad was cut into 0.6 cm × 2 cm

Y.-Y. Lin et al. / Biosensors and Bioelectronics 23 (2008) 1659–1665

1661

Fig. 1. Schematic diagram of the IEB. (A) A test strip; (B) a cover; (C) a bottom and (D) a SPE.

and stored in a desiccator at room temperature before use. The second zone is a contact zone with a glass-fiber pad loaded with anti-PSA–QD conjugates by physical adsorption. The third zone is a test zone where the nitrocellulose membrane was covalently immobilized with the second anti-PSA. The size of the test-zone membrane is 0.6 cm in width and 0.2 cm in length. An absorbent pad placed at the end of the IEB forces the sample solution to migrate across the strip. The screen printed electrode (SPE) (D) was embedded underneath the test zone. To accommodate multiple functional components within the integral device of the IEB, all of the parts of the device were manually mounted on the adhesive surface of the backing material (typically an inert plastic, e.g., polyester).

(w/v) BSA and 0.1% Tween-20 in 0.01 M PBS buffer) for 2 h, followed by washing with 0.01 M PBS buffer containing 0.1% Tween-20 for three times, 5 min each time. The membrane was then dried in a nitrogen box for 1 h and stored at 4 ◦ C in a dry state.

2.3.1. The immobilization of anti-PSA antibodies onto the surface of the test zone membrane The membrane was first modified with space arms by incubation with 2.5% diaminoheptane in deionized water (pH ∼ 12) for 1 h with constant agitation (Masson et al., 1993). Then the membrane was washed with deionized water and 0.01 M PBS buffer for 12 h on a shaking stage at room temperature for removing unbound chemicals. The diaminoheptane-modified nitrocellulose membrane was then treated with 1% glutaraldehyde in 0.01 M PBS buffer for 4.5 h with constant shaking at room temperature, followed by washing with 0.01 M PBS for 3 h to remove unbound glutaraldehyde. The washing buffer for the above two washing steps need to be changed per hour. The membrane was then dried in a nitrogen box for 1 h and stored at 4 ◦ C in a dry state. The aldehyde-modified nitrocellulose membrane was cut into pieces (0.2 cm × 0.6 cm) with a homemade punch. These membranes were directly immersed into a 1 mg mL−1 anti-PSA solution followed by drying at 4 ◦ C. Then the anti-PSA-modified membranes were washed with PBS buffer containing 0.1% Tween-20 and were subsequently blocked with 1% BSA in PBS buffer followed by drying and storing at 4 ◦ C before use.

2.4. Lateral flow immunoassay

2.3.2. Blocking of the lateral flow membranes of the strip The nitrocellulose membrane used for immunoreaction and flowing was treated with blocking reagent (a solution of 3%

2.3.3. Preparation of the anti-PSA–QD conjugate on glass fiber pad The prepared anti-PSA–QD conjugate solution was diluted 10 times with PBS buffer containing 0.5% (w/v) BSA. A desired volume of the diluted anti-PSA–QD conjugate solution was applied onto the end part of a glass fiber pad and allowed to dry and store in a desiccator at 4 ◦ C.

The performance of the IEB relied on non-competitive assay formats, where the primarily anti-PSA QD conjugate was physically absorbed on the end of the glassy fiber and the second anti-PSA was immobilized on the test zone. 100 ␮L of sample solutions which diluted from PSA serum standard with PBS buffer were added to the sample pad and allowed to flow through the whole strip. The PSA in sample can conjugate with anti-PSA QD and be trapped by the test zone. As for the extra buffer or sample can be removed by the capillary force which provided by the absorbent pad. After a fixed immunoreaction time (e.g., 7 min), the reaction was stopped by drawing two lines in front and back of test zone with a hydrophobic magic pen or peeling off the test zone by a tweezer. The total assay time can be completed in less than 10 min. 2.5. Electrochemical detection All electrochemical experiments were carried out with an electrochemical analyzer CHI 660A (CH Instruments, Austin, TX, USA) or a portable electrochemical analyzer (Andcare Biosciences, Inc.) connected to a personal computer. Disposable SPEs consisting of a carbon working electrode, a carbon counter electrode, and an Ag/AgCl reference electrode were purchased from Alderon Biosciences, Inc. (Durham, NC, USA) for

1662

Y.-Y. Lin et al. / Biosensors and Bioelectronics 23 (2008) 1659–1665

electrochemical measurements. A sensor connector (Alderon Biosciences, Inc.) is used to connect SPE to the CHI electrochemical analyzer. All the potentials are referred to Ag/AgCl reference electrode. Ten microliters of 1 M HCl was applied to the test-zone membrane of the strip and incubated for 2.0 min, and then a 50 ␮L solution of the acetate buffer with 10 ppm of Hg was added. The SWV experiments were performed under the following conditions—step 1: Pretreatment: 65 s 0.6 V. Step 2: Preconcentration: constant potential at −1.4 V with a different time period, e.g., 80 s. The scanning potential is from −1.0 to −0.5 V, with increments of 4 mV, amplitude of 25 mV, and a frequency of 15 Hz (Lu et al., 2005; Wang et al., 2008). All the SWVs were conducted with baseline correction using the CHI 660a software. 2.6. Enzyme-linked immuno-sorbent assay ELISA experiments were conducted with microplates according to the ELISA kit protocol (Alpha Diagnostic International, San Antonio, TX, USA) and the ELISA measurements were carried out with microplate reader. 3. Results and discussion 3.1. Principle of the PSA assay In this paper, QDs (commercially available with a core/shell structure consisting of CdSe as the core and ZnS as the shell) were chosen as labeling materials to tag PSA antibodies, and electrochemical stripping analysis was selected as a detection method for this QD label. During the assay, a liquid sample solution (100 ␮L) was first applied to the sample loading zone (Fig. 1). The fluid migrated toward the other end of the strip because of capillary action which was driven by the absorbent pad. The immunoreactions took place during fluid migration. As the fluid reached the contact zone, the PSA in the fluid reacted with the QD-labeled anti-PSA that was pre-loaded in this zone, forming PSA–anti-PSA–QD complexes. The complexes continued to move along the strip and entered the test zone where the covalently bound second anti-PSA antibody captured the antigen (with different epitopes) in the complexes, forming sandwich-type anti-PSA–PSA–anti-PSA–QD complexes. These complexes remained in the test zone while the rest of the liquid sample migrated into the absorbent zone. After a suitable reaction period, the formation of the QD-attached sandwich complexes within the test zone was completed. The conjugated QDs in the complexes were then dissolved under acid conditions, and cadmium ions (Cd2+ ) were released and quantified by the electrochemical technique. This method is quite suitable for quantitative analysis because electrochemical signals are proportional to the PSA concentration in samples due to the amount of attached QDs, depending on the PSA concentration in the samples. The prepared anti-PSA–QD conjugates have been characterized by TEM and electrochemistry (see Fig. S1 (A and B), Supplementary data). The degree of dissolution of the QDs with 1 M HCl was studied with different dissolution time from 30 s

Fig. 2. SWV responses of IEB with (a) 8 ng mL−1 PSA and (b) 0 ng mL−1 PSA in the serum. The curves were conducted with baseline correction using the CHI 660A software. The immunoreaction time was 20 min; 1 ␮L of anti-PSA–QD conjugate was loaded on the glass fiber; the preconcentration time was 120 s.

to 3 min. It was found that electrochemical signals from Cd or Zn increased with the increase of the dissolution time and it almost reached a plateau after 2 min. Therefore, 2 min of dissolution time was selected throughout the experiments (data not shown). 3.2. Optimization of experimental parameters Fig. 2 presents the typical SWVs of PSA assay with nanoparticle-based IEB in the presence and absence of PSA. It can be seen that a well-defined cadmium striping peak was observed in the presence of 8 ng mL−1 PSA (curve a). The cadmium peak would be attributed to the oxidation of cadmium, which came from the QD tag of immuno-complexes (anti-PSA–PSA–anti-PSA–QD) formed in the test zone during the immunoreaction. A small signal was also observed in the serum with no PSA (control, curve b). Such a response would be resulted from the nonspecific adsorption of anti-PSA–QD conjugate on the test zones. To minimize such nonspecific adsorption, increase the sensitivity and achieve rapid analysis, we optimized some experimental parameters such as immunoassay reaction time, the amount of anti-PSA–QD conjugates applied and electrochemical preconcentration time. The electrochemical immunoreaction time was first examined with standard PSA samples. In this study, we chose different immunoreaction time such as 7, 10, 15, 20, and 25 min. To avoid the effect of matrices, both PSA samples and the controls (absence of PSA) were diluted 10 times from standard solutions in serum with PBS buffer, and 100 ␮L solutions of PSA samples and control were applied for the optimization. Fig. 3a shows the histogram of the difference of electrochemical signal between the PSA samples and the controls versus different immunoreaction time. As shown in this figure, the electrochemical signal difference between the samples and the controls was much smaller when the immunoreaction time was 7 min, which was due to less completion of sandwich immunoreaction in

Y.-Y. Lin et al. / Biosensors and Bioelectronics 23 (2008) 1659–1665

Fig. 3. (a) Histogram of electrochemical signal difference between the sample and control vs. the immunoreaction time. The difference is obtained by subtracting the signal of the sample with the signal of the control. 2 ␮L of anti-PSA–QD conjugates were loaded onto the glass fiber, and the secondary PSA antibody was immobilized onto the test zone with 1.0 mg mL−1 PSA antibodies. 100 ␮L of 8 ng mL−1 PSA sample was applied to the IEB; a 100 ␮L control was applied to another IEB with the same loading of anti-PSA–QD conjugates. SWV conditions: pretreatment: 65 s at 0.6 V. Preconcentration: constant potential at −1.4 V for 80 s. (b) Histogram of electrochemical response from the sample and control vs. the loading amount of the anti-PSA–QD conjugates. Different amounts of anti-PSA conjugated QD were loaded onto the glass fiber. The immunoreaction time was 10 min while the preconcentration time was 2 min. Other conditions were the same as those in (a). (c) Histogram of electrochemical response from sample and control vs. preconcentration time. 1.0 ng mL−1 PSA and control (absence of PSA) were applied with other optimal conditions: 2 ␮L of anti-PSA conjugated QD were loaded onto the glass fiber with an immunoreaction time of 10 min. Preconcentration was conducted at a constant potential of −1.4 V with different time period.

1663

test zone. The signal difference achieved the largest with the 10 min reaction time. It indicates that the sandwich immunoreactions in test zone are completed and nonspecific binding is small for 10 min reactions. When the reaction time is longer than 10 min, the electrochemical signal differences become smaller with increased reaction time, which may be attributed to the fact that the prolonged reaction time can increase non-specific interactions. Although 10 min reaction time generates the biggest signal difference, there are still a little nonspecific binding from the controls. To eliminate the nonspecific adsorption of antiPSA–QD conjugates and achieve the highest sensitivity of the IEB, the amount of anti-PSA–QD conjugates applied on the contact zone was optimized. The blocking reagent BSA was used to treat the membrane of the strip for reducing the nonspecific adsorption. However, the physical adsorption of anti-PSA conjugates on the strip also depends on its concentration. Serious nonspecific absorption may result when the amount of the anti-PSA–QD conjugates is too high. The optimization experiments were achieved with different amounts of anti-PSA–QD conjugates applied to the glass fiber pad of the contact zone. Fig. 3b shows the histogram of the electrochemical signal from the PSA samples and the controls versus different amounts of anti-PSA–QD conjugates. It can be seen from this figure that the electrochemical signal from PSA samples increases with the increase of the amount of anti-PSA–QD conjugates, which is attributed to the fact that the amount of formed anti-PSA–PSA–anti-PSA–QD sandwich immuno-complexes in the test zone is proportional to the amount of anti-PSA–QD conjugates adsorbed on the contact zone. The electrochemical signals from PSA samples start to level off after 3 ␮L of conjugate solution, which is ascribed to the saturation of the sandwich immunoreactions with 3 ␮L or more conjugates. It also can be seen from Fig. 3b that the signal from the control (absence of PSA) also increases with the increase of the amount of anti-PSA–QD conjugates, which is caused by nonspecific adsorption of anti-PSA–QD conjugates on the test strip. However, the signal of the control with 2 ␮L anti-PSA–QD conjugates is much smaller than that with 3 or more ␮L conjugates. Therefore, a loading amount of 2 ␮L conjugate was chosen for the assay, and the background signal can be further eliminated by reducing the electrochemical preconcentration time. Electrochemical preconcentration time of the stripping analysis has some influence on the assay because the background signal may increase with the long preconcentration time. The trade-off between the highest sensitivity and the least background signal from nonspecific interactions was studied. With the optimal conditions (2 ␮L anti-PSA–QD conjugates, reaction time of 10 min), different preconcentration times for preconcentration of Cd2+ onto the electrode surface were examined with a 1 ng mL−1 PSA sample and the control. Fig. 3c shows the histogram of the electrochemical signal from the PSA sample and the control versus a different preconcentration time. As shown in this figure, the electrochemical signals increase with the increase of the preconcentration time, and the background signal from the control appears with 105 s preconcentration. Thereafter, the

1664

Y.-Y. Lin et al. / Biosensors and Bioelectronics 23 (2008) 1659–1665

Fig. 4. Typical SWV responses at IEB while increasing the PSA concentration (from bottom to top: 0, 0.05, 0.1 ng mL−1 ) under optimal conditions. The immunoreaction time was 10 min; 2 ␮L of anti-PSA–QD conjugate was loaded on the glass fiber; the preconcentration time was 80 s.

background signals increase with the increase of the preconcentration time. However, there is no signal from the control when the preconcentration is less than 80 s. It can be seen from this figure that the background signal from the control was completely eliminated with a short preconcentration time, e.g., 80 s. At the same time, the sensitivity of the assay at IEB is acceptable. Therefore, an 80 s preconcentration time was chosen for the assay. 3.3. The analytical performance and validation of the nanoparticle label-based IEB Under optimal experimental conditions, the analytical performance of the IEB for the assay of PSA was examined with standard serum samples (0 to 4 ng/mL). All the PSA standards were prepared by 20-fold dilution of stock serum samples and control with PBS buffer. Fig. 4 shows SWVs of the IEB with 0, 0.05 and 0.1 ng/mL PSA in serum. As shown in this figure, a negligible signal was observed in the control experiment (in the absence of PSA), whereas the well-defined peaks (from cadmium oxidation) were observed at such low concentrations, which are attributed to the the enhanced sensitivity by the QD labels. It is found that the peak current increases with the increase of the concentration of PSA. The linear range for PSA is from 0.05–4 ng/mL with R2 = 0.995. The method is reproducible with a relative standard deviation of 6.4 %. The detection limit of the IEB for PSA is estimated to be 0.02 ng/mL based on S/N = 3, which is much lower than current clinical recommendations for a qualitative test of a PSA marker with a cutoff point at 4.0 ng/mL (Gann et al., 1995). More details about the performance of this method have been shown in the previous publication (Liu et al., 2007). With a 20-fold dilution, the detection range is useful for detection of PSA in serum sample with PSA concentration in the range of 1–80 ng/mL. The dilution of serum samples could be helpful for reducing the interference of the matrices.

The nanoparticle label-based IEB was validated with a human serum sample (Golden West Biologicals, Inc.) and commercial ELISA PSA kit (Alpha Diagnostic International). ELISA analysis of PSA was performed with the manufacture instruction of PSA ELISA kit and a microplate reader was used to record the absorbance of tested samples. The 20-fold diluted serum samples with PBS buffer were spiked with final concentrations of 0.5 ng/mL PSA. The serum sample without addition of standard PSA served as a control. These samples were detected with IEB and ELISA, respectively. For the control sample (without PSA), PSA was not detectable with both IEB and ELISA. However, for the 0.5 ng/mL PSA spiked sample, the detectable value is 0.53 ± 0.13 ng/mL with IEB and 0.56 ± 0.04 ng/mL with ELISA, respectively. The recoveries for these spiked samples obtained from the IEB and ELISA are 105% and 111%, respectively. The results from IEB are consistent with those from ELISA. It indicates that the IEB data for PSA measurements are accurate and reliable. Therefore, nanoparticle label-based IEB demonstrates promise for sensitive detection of disease biomarkers and open up a new avenue for various clinical applications. 4. Conclusion We have successfully demonstrated an approach using NP label-based IEB for rapid and sensitive detection of PSA in human serum. The rapidity of this method is derived from the advantages of the test strip (e.g., fast immunoreaction and separation), and the high sensitivity is due to the NP-generated signal amplification and inherent high sensitivity of the electrochemical techniques. Therefore, this IEB, combining the advantages of an immunochromatographic strip technique and the power of an electrochemical detection technique with signal amplification of nanotechnology, provides an option for screening PCa. The capability of accurately detecting spiked human serum samples demonstrates the potential for clinical diagnosis of PCa. This approach will lead to a hand-held device for faster, more clinically accurate, and less expensive detection of PSA than currently available immunoassays, test-strips and kits. In general, integrating the IEB with advanced nanotechnology (NP labels) provides an opportunity to develop a simple, highly sensitive, and portable diagnostic device for rapid and relatively low-cost point-of-care testing of biomarkers in biological fluids. Acknowledgements The work was supported by a laboratory-directed research and development program at Pacific Northwest National laboratory (PNNL). The research described in this paper was performed at the Environmental Molecular Sciences Laboratory, a national scientific user facility sponsored by the DOE’s Office of Biological and Environmental Research and located at PNNL. PNNL is operated by Battelle for DOE under Contract DE-AC0576RL01830.

Y.-Y. Lin et al. / Biosensors and Bioelectronics 23 (2008) 1659–1665

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bios.2008.01.037. References Acevedo, B., Perera, Y., Ruiz, M., Rojas, G., Benitez, J., Ayala, M., Gavilondo, J., 2002. Clin. Chim. Acta 317, 55–63. Bao, Y.P., Wei, T.F., Lefebvre, P.A., An, H., He, L.X., Kunkel, G.T., Muller, U.R., 2006. Anal. Chem. 78, 2055–2059. Benson, M.C., Whang, I.S., Olsson, C.A., Mcmahon, D.J., Cooner, W.H., 1992. J. Urol. 147, 817–821. Bradford, T.J., Tomlins, S.A., Wang, X.J., Chinnaiyan, A.M., 2006. Urol. OncolSemin. Ori. Inves. 24, 538–551. Brawer, M.K., 1999. CA Cancer J. Clin. 49, 264–281. Cuzzubbo, A.J., Endy, T.P., Nisalak, A., Kalayanarooj, S., Vaughn, D.W., Ogata, S.A., Clements, D.E., Devine, P.L., 2001. Clin. Diag. Lab. Immunol. 8, 1150–1155. D’Orazio, P., 2003. Clin. Chim. Acta 334, 41–69. Dequairem, M., Degrand, C., Limoges, B., 2000. Anal. Chem. 72, 5521–5528. Fernandez-Sanchez, C., McNeil, C.J., Rawson, K., Nilsson, O., 2004. Anal. Chem. 76, 5649–5656. Fernandez-Sanchez, C., McNeil, C.J., Rawson, K., Nilsson, O., Leung, H.Y., Gnanapragasam, V., 2005. J. Immunol. Methods 307, 1–12. Gann, P.H., Hennekens, C.H., Stampfer, M.J., 1995. JAMA 273, 289–294. Georganopoulou, D.G., Chang, L., Nam, J., Thaxton, C.S., Mufson, E.J., Klein, W.L., Mirkin, C.A., 2005. Proc. Natl. Acad. Sci. U.S.A. 102, 2273–2276. Grubisha, D.S., Lipert, R.J., Park, H.Y., Driskell, J., Porter, M.D., 2003. Anal. Chem. 75, 5936–5943. Healy, D.A., Hayes, C.J., Leonard, P., McKenna, L., O’Kennedy, R., 2007. Trends Biotechnol. 25, 125–131. Huang, L., Reekmans, G., Saerens, D., Friedt, J.M., Frederix, F., Francis, L., Muyldermans, S., Campitelli, A., Van Hoof, C., 2005. Biosens. Bioelectron. 21, 483–490.

1665

Huhtinen, P., Soukka, T., Lovgren, T., Harma, H., 2004. J. Immunol. Methods 294, 111–122. Jain, K.K., 2005. Clin. Chim. Acta 358, 37–54. Jemal, A., Siegel, R., Ward, E., Murray, T., Xu, J.Q., Smigal, C., Thun, M.J., 2006. CA Cancer J. Clin. 56, 106–130. Jin, Y., Jang, J.W., Han, C.H., Lee, M.H., 2005. J. Agric. Food Chem. 53, 7639–7643. Kim, Y.M., Oh, S.W., Jeong, S.Y., Pyo, D.J., Choi, E.Y., 2003. Environ. Sci. Technol. 37, 1899–1904. Lind, K., Kubista, M., 2005. J. Immunol. Methods 304, 107–116. Liu, G.D., Wang, J., Wu, H., Lin, Y., 2006. Anal. Chem. 78, 7417–7423. Liu, G., Lin, Y., 2007. Talanta 74, 308–317. Liu, G., Lin, Y.Y., Wang, J., Wu, H., Wai, C.M., Lin, Y., 2007. Anal. Chem. 79, 7644–7653. Lu, F., Wang, K.H., Lin, Y., 2005. Analyst 130, 1513–1517. Masson, M., Lauritzen, E., Holm, A., 1993. Electrophoresis 14, 860–865. Nagatani, N., Tanaka, R., Yuhi, T., Endo, T., Kerman, K., Takamura, Y., Tamiya, E., 2006. Sci. Tech. Adv. Mater. 7, 270–275. Nam, J., Thaxon, J., Mirkin, C.A., 2003. Science 301, 1884–1886. Sarkar, P., Pal, P.S., Ghosh, D., Setford, S.J., Tothill, I.E., 2002. Int. J. Pharmaceut. 238, 1–9. Stephan, C., Klaas, M., Muller, C., Schnorr, D., Loening, S., Jung, K., 2006. Clin. Chem. 52, 59–64. Wang, J., Liu, G., Engelhard, M.H., Lin, Y., 2006. Anal. Chem. 78, 6974–6979. Wang, J., Liu, G.D., Wu, H., Lin, Y., 2008. Small 4, 82–86. Wu, H., Liu, G., Wang, J., Lin, Y., 2007. Electrochem. Commun. 9, 1573–1577. Wee, K.W., Kang, G.Y., Park, J., Kang, J.Y., Yoon, D.S., Park, J.H., Kim, T.S., 2005. Biosens. Bioelectron. 20, 1932–1938. Yu, X., Munge, B., Patel, V., Jensen, G., Bhirde, A., Gong, J.D., Kim, S.N., Gillespie, J., Gutkind, J.S., Papadimitrakopoulos, F., Rusling, J.F., 2006. J. Am. Chem. Soc. 128, 11199–11205. Zhang, G.P., Wang, X.N., Yang, J.F., Yang, Y.Y., Xing, G.X., Li, Q.M., Zhao, D., Chai, S.J., Guo, J.Q., 2006. J. Immunol. Methods 312, 27–33. Zheng, G.F., Patolsky, F., Cui, Y., Wang, W.U., Lieber, C.M., 2005. Nat. Biotechnol. 23, 1294–1301.