Novel Signal-Enhancing Immunoassay for Ultrasensitive Biomarker

Feb 5, 2015 - Stephanie M. Schubert , Lisa M. Arendt , Wenhui Zhou , Shazia Baig , Stephanie R. Walter , Rachel J. Buchsbaum , Charlotte Kuperwasser ...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Novel Signal-Enhancing Immunoassay for Ultrasensitive Biomarker Detection Based on Laser-Induced Fluorescence Ji Zhang,† Shuai Wang,§ Kunping Liu,†,‡,∥ Yin Wei,‡ Xu Wang,‡ and Yixiang Duan*,‡ †

College of Chemistry, and ‡Research Center of Analytical Instrumentation, Key Laboratory of Bio-resource and Eco-environment, Ministry of Education, College of Life Sciences, Sichuan University, Chengdu 610065, P. R. China § College of Instrumentation & Electrical Engineering, Jilin University, Changchun 130061, P. R. China ∥ Faculty of Biotechnology Industry, Chengdu University, Chengdu 610106, P. R. China S Supporting Information *

ABSTRACT: An innovative signal-enhancing immunoassay for ultrasensitive biomarker detection based on laser-induced fluorescence (LIF) has been developed. A novel LIF optical system with high collection efficiency was constructed by using a parabolic mirror. Carboxyl-functionalized magnetic beads were used to immobilize antibody for achieving a conventional sandwich assay. Fluorescence from Rhodamine 6G (R6G)labeled antibody was collected by the newly designed optical system. To reduce photobleaching of R6G under laser irradiation, ethanol instead of commonly used aqueous solution was used as assay buffer in the last stage. The newly developed LIF immunoassay displayed excellent analytical performance for α-fetoprotein (AFP) detection in the concentration range from 0.005 to 1.0 ng/mL with a detection limit of 0.0016 ng/mL. The detection limit obtained in this work is about 3 orders of magnitude better than that of conventional enzyme-linked immunosorbent assay (ELISA). In addition, the proposed method exhibited excellent precision, acceptable stability, and good reproducibility. Furthermore, the proposed immunoassay was successfully applied to AFP determination in real serum specimens. Therefore, the present immunoassay was demonstrated to be a powerful tool for ultrasensitive biomarker detection.

I

such as light. A parabolic mirror focuses a parallel beam, propagating along the optical axis to the focal point.11 As a source is placed in the focal point of a parabolic mirror, spherical waves generated by the source will be reflected into a plane wave traveling as a collimated beam along the axis.12 In addition, a large solid angle subtended by the parabolic mirror guarantees most of the light from the focal point can be reflected into the collimated beam. In this novel LIF optical system, the sample was excited at the focal point of the parabolic mirror. Thus, most of the fluorescence generated from the sample was reflected and transformed to parallel waves traveling along the axis. In this way, significantly improved signal collection efficiency was achieved. Another factor that has considerable effect on the successful implementation of ultrasensitive detection based on LIF is the photophysical property associated with fluorescent dyes. In this study, Rhodamine 6G (R6G) was used to label the detection antibody, because of its broad range of absorption and emission wavelength, high fluorescence quantum yield, and pH insensitivity.13,14 However, when a laser is used as the excitation

t is well-known that organs and cells will release some specific proteins into the circulation system as a tumor develops.1 Thus, these specific proteins are used as tumor markers for diagnosing tumor and monitoring tumor progression. Enzyme-linked immunosorbent assay (ELISA)2,3 is the current gold standard for clinical biomarker detection4 due to its high specificity and simple setup. However, suffering from moderate sensitivity, ELISA is not an ideal biomarker detection tool in many cases, since biomarker levels are often very low in physiological samples especially during early disease phases.5 Therefore, developing ultrasensitive immunoassay for biomarkers of interest is extremely important. Laser-induced fluorescence (LIF) owns an excellent detection limit as well as detection sensitivity.6 It is an excellent technology to achieve ultrasensitive detection. Therefore, developing immunoassays based on LIF has attracted increasing attention and considerable research interest.7−10 Unfortunately, due to the limit of traditional optical system structure, only a small part of fluorescence emitted from a sample can be collected. Low signal collection efficiency severely limits LIF immunoassay detection sensitivity. Therefore, in this work, we wanted to construct a novel optical system based on a parabolic mirror to boost the efficiency of signal collection. A parabolic mirror is a reflective surface used to collect or project energy © XXXX American Chemical Society

Received: December 4, 2014 Accepted: February 5, 2015

A

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry source, the high intensity of the laser will result in serious photobleaching. The photobleaching dramatically decreases the fluorescence intensity of R6G, which results in lower detection sensitivity. To solve this problem, ethanol instead of commonly used aqueous solution was used as the assay buffer in the last step of the immunoassay, since R6G possesses a superior photophysical property in ethanol compared with that in water.15 To demonstrate the practicability of the immunoassay, αfetoprotein (AFP), a widely used clinical tumor marker,16−20 was used as the sample tumor antigen in this study. To improve detection specificity, carboxyl-functionalized magnetic beads were used to immobilize antibody for achieving the sandwich assay because the conventional sandwich assay requires two antibodies to bind a biomarker for generating a positive signal.21 Furthermore, due to the high loading capacity of the magnetic beads, effective preconcentration of biomarker was achieved. In this work, a novel LIF platform for ultrasensitive biomarker detection was constructed by using a parabolic mirror as a highly efficient fluorescence collection device. Subsequently, for reducing photobleaching of R6G under laser irradiation, ethanol instead of commonly used aqueous solution was used as the assay buffer. To demonstrate the practicability of the immunoassay, the method was applied for AFP detection. To improve specificity of the immunoassay, carboxyl-functionalized magnetic beads were used to immobilize the capture antibody for achieving the conventional sandwich assay. Assay results showed great improvement in detection limit and detection sensitivity. Therefore, this novel signal-enhancing immunoassay was demonstrated to be a highperformance tool for ultrasensitive biomarker detection.

Figure 1. Schematic diagram of the novel optical system.

the incidence angle of 45°. After that, the laser beam was converged by a Plano-convex lens and passed through a small hole drilled in the parabolic mirror to focus on and excite the sample. Another hole in the back of the parabolic mirror allowed a quartz capillary tube with sample to be placed. The location where the laser excited the sample was the focal point of the parabolic mirror. It is generally known that a parabolic reflector can project energy of a source at its focal point outward in a parallel beam. Therefore, fluorescence emitted from the sample was reflected into a parallel beam, propagating along the axis. A convex lens was employed to refocus the parallel beam and to transfer it into an optical fiber. To reduce spherical aberration, a best form lens, which can provide the best possible performance from a spherical singlet, was used in our experiment. Therefore, a minimum light spot produced by a spherical singlet was formed. Thus, more fluorescence can be transferred into the collimating lens and then to the optical fiber. Afterward, the fluorescence signal was captured by a single-channel spectrometer. Moreover, the best form lens was coated with an antireflection (AR) coating for 350−700 nm to provide maximum light transmission across the monitored spectral range. In addition, to reduce the impact of the laser beam on the fluorescence signal, a 550 long pass filter was placed between the best form lens and the collimating lens to filter out the laser beam influence. Last but not least, all the optical elements were mounted and adjusted carefully, thereby achieving the best performance of the optical system. Preparation of Antibody-Immobilized Magnetic Beads. The anti-AFP monoclonal antibody was immobilized on the carboxyl-functionalized magnetic beads according to standard carboxyl−amine conjugation chemistry. This method adopted EDC/NHS as coupling agents, and a two-step covalent bonding process was used.22,23 Briefly, a 180 μL suspension of carboxyl-modified magnetic beads (1 μm, 15 mg/mL) in a 1.5 mL centrifuge tube was first separated from solution in a magpearl separator. The schematic of the magnetic separation process is shown in Figure S-1, Supporting Information. The magnetic beads were washed several times with ultrapure water and then suspended to a final volume of 200 μL in ultrapure water. Then 400 μL of 153 mg/mL EDAC and 25 mg/mL NHS in ultrapure water were added to active carboxyl groups24 at room temperature for 40 min followed by triplicate rinsing with 200 μL of sodium acetate buffer solution to remove excessive EDAC and NHS. Then the magnetic beads were suspended to a final volume of 200 μL in sodium acetate buffer solution. A 180 μL solution of anti-AFP antibody (1 μg/mL)

■

EXPERIMENTAL SECTION Materials. Phosphate-buffered saline (PBS) (pH 7.0, 0.01M) was used as a washing buffer and also solvent for the protein standards. To prepare blocking solution, bovine serum albumin (BSA, Wolsen Amresco, Solon, OH) was added at 10 mg/mL to PBS. The detection and capture monoclonal antibody pair, R6G-labeled mouse antihuman AFP antibody and mouse antihuman AFP antibody, and AFP human antigen were obtained from Wanyumeilan (Beijing, China). These proteins were used without any further purification. 1-Ethyl-3(3-(dimethylamino)propyl)carbodiimide hydrochloride (EDAC) and N-hydroxysuccinimide (NHS) were purchased from Sagon Biotec (Shanghai, China). Sodium acetate buffer solution (pH 5.5, 0.05 M) was acquired by mixing a certain percentage of sodium hydroxide with acetic acid glacial. Mouse antihuman AFP antibody was diluted by sodium acetate buffer solution to expose more amino groups of the antibody, which will promote the conjugation rate. Other proteins were diluted in PBS to produce appropriate concentrations. Carboxylmodified magnetic beads (1 μm, 15 mg/mL) were supplied by Innosep Biosciences Co., Ltd. (Zhengzhou, China). Quartz capillary tubes (1.2 mm i.d., 2 mm o.d, and 40 mm long) were obtained from Lehua Quartz (China). All the chemicals were of analytical reagent grade. Construction of the Novel Optical System. A diodepumped solid-state laser (DTL-313, Laser-export Co. Ltd.; 527 nm, 490 mW) combined with a single-channel spectrometer (AvaSpec-USB2.0, Avantes) was used in the novel optical system. As shown in Figure 1, a laser beam produced by the diode-pumped solid-state laser was reflected by a mirror with B

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

■

Article

RESULTS AND DISCUSSION Superiority of the Novel Optical System. An appropriate design of the optical system plays a major role in determining analytical performance of immunoassay based on LIF. To evaluate the influence of the optical system’s structure, a comparative study of the novel optical system with a traditional device was performed. The traditional optical system was constructed following the structure of commonly used LIF systems.8,25 The structure of the traditional optical system is shown in Figure S-2, Supporting Information. A parabolic mirror was not used here, and only a collimating lens coupled to optical fiber was used to collect fluorescence emitted from the sample. A filter was placed in front of the collimating lens to reduce the impact of the laser beam on the fluorescence signal. Performance of both optical systems was verified by detecting R6G in ethanol. Samples with different concentrations of R6G were tested by both the novel optical system and the traditional optical system, respectively. The fluorescence signals observed in Figure 2 show that the novel optical system achieves a much

was then added, and the resulting suspension was shaken slightly at room temperature for 3 h. The antibody-immobilized magnetic beads were magnetically separated from the mixture and washed with PBS. After that, 500 μL of 1% BSA was added, which was followed by reacting 2 h under oscillation for blocking the excessive active sites to eliminate the risk of unspecific binding. The resulting compound was separated from the solution by triplicate rinsing with 200 μL of PBS. Finally, the compound was suspended in 180 μL of PBS and stored in a refrigerator at 4 °C for further use. Assay Protocol. The schematic of the detection procedure including the construction of the antibody-immobilized magnetic beads is shown in Scheme 1. In general terms, the Scheme 1. Detection Procedure of AFPa

a

Figure 2. Linear and sensitive dose−response of R6G for the newly design optical system and the traditional optical system with error bars (standard deviation from the mean, n = 4). Excitation was set at 527 nm, and the emission intensity was monitored at 558.90 nm. All measurements were performed in ethanol solution.

anti-AFP antibody-modified magnetic beads (20 μL) were incubated for 60 min at 37 °C with a series of samples (900 μL) containing either purified AFP antigen or clinical samples at various concentrations. After several washes with PBS, the detection antibody was added (20 μL) to react with the captured AFP for 60 min at 37 °C. The detection antibody was prepared by diluting the stock solution of R6G-labeled mouse antihuman AFP antibody with PBS to 1 μg/mL. The unbound R6G-labeled antibody was washed away by PBS. Finally, the sandwich-type immunocomplex magnetic beads were resuspended in 100 μL of ethanol. A 40 μL suspension of sandwichtype immunocomplex was introduced into a quartz capillary tube for fluorescence detection. The fluorescence signal of the immunocomplex was measured by the novel LIF platform. Clinical Sample Treatment. The undiluted serum containing AFP (30 μg/mL) was obtained from Wanyumeilan (Beijing, China). The serum was stored at −20 °C until it was used. The serum was used without any further purification but just diluted with PBS prior to assay, avoiding a level of AFP out of the calibration linear range.

stronger response than that of the traditional optical system. The sensitivity of the novel optical system, defined as the slope of the line best fitting the set of points, was improved 19.7-fold. The improved detection sensitivity with the novel optical system demonstrated the advantage of the novel LIF platform in fluorescence collection. An increase in sensitivity of 19.7-fold achieved by the novel optical system was mainly due to the use of a parabolic mirror. The parabolic mirror, as mentioned above, is a reflective surface used to collect or project energy such as light. Its shape is part of a circular paraboloid, that is, the surface generated by a parabola revolving around its axis. If a source is placed in the focal point of the parabolic mirror, spherical waves generated by the source will be reflected into a plane wave propagating as a collimated beam along the axis. In the novel optical system, the sample was excited at the focal point of the parabolic mirror. The sample emits fluorescence from the focal point, like that from a radiation source. A large solid angle subtended by the parabolic mirror at the focal point guarantees that most of the fluorescence from the focal point can be reflected into the collimated beam along the axis. The solid angle is the twodimensional angle in three-dimensional space that an object subtends at a point. It measures how large the object appears to

The construction of the antibody-immobilized magnetic beads and the detection procedure of AFP are shown.

C

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

U is the aperture angle. The aperture angle of the parabolic mirror we used here was 126.2°. The parameters of the parabolic mirror are shown in Figure 3. It is well-known that

compared with water, ethanol can double the fluorescence quantum yield of R6G from 0.45 to 0.95. On the basis of these two factors,15,26 nearly 2 orders of magnitude more photons per molecule can be obtained when using ethanol instead of water as solvent. R6G-labeled mouse antihuman AFP antibody was used to verify the difference in analytical performance between ethanol and aqueous solution. A set of solutions of R6G-labeled AFP antibody with different concentrations were prepared by using ethanol and PBS (0.01 M, pH 7.0) respectively. The assay results are shown in Figure 4. Comparing the slopes of the two

Figure 3. Schematic diagram of the parabolic mirror. Point F is the focal point of the parabolic mirror. L1 is the distance from the focal point to the opening of parabolic mirror. L2 is the diameter of the parabolic mirror. U is aperture angle. L1 = 30.5 mm, L2 = 11.1 mm, U = 126.2°.

Figure 4. Linear and sensitive dose−response of R6G-labeled AFP antibody by using ethanol or PBS (0.01 M, pH 7.0) as solvent (standard deviation from the mean, n = 4). Excitation was set at 527 nm, and the fluorescence intensity was monitored at 558.90 nm.

an observer looking from that point. To a fluorescence collection device, the solid angle determines the collection efficiency; that is, the collection efficiency is positively correlated to the solid angle. The solid angle subtended by the parabolic mirror at the focal point was 3.2π. It was calculated by ω = 4π sin 2(U /2)

lines, the fluorescence intensity of R6G-labeled AFP antibody in ethanol was about 7.4 times as strong as that in PBS. The rise in fluorescence intensity of R6G-labeled AFP antibody was smaller than that of R6G alone. This might be attributed to formation of nonfluorescent derivatives and interaction between dye molecules which are caused by the conjugation of R6G to proteins.27 However, using ethanol as solvent still dramatically raised the fluorescence intensity of R6G-labeled AFP antibody. Detection sensitivity of the immunoassay was greatly improved through this simple way. Thus, the novel LIF immunoassay’s performance in ultrasensitive detection was further improved. It is noteworthy that ethanol may eventually lead to the denaturation of protein. Furthermore, the sandwich-type immunocomplex was not particularly stable in ethanol. Thus, ethanol was only used as assay buffer in the last step of the immunoassay. At that time, the binding reaction between antibody and antigen had been finished. After the sandwichtype immunocomplex was suspended in ethanol, the fluorescence signal of the immunocomplex was measured by the novel LIF platform. Therefore, the denaturation of antibody has no effect on the detection results. Analytical Performance of the Novel LIF Immunoassay. The effectiveness of the novel LIF immunoassay was characterized by its analytical performance for detecting AFP. A series of buffer solutions containing different concentrations of AFP were used to construct a calibration curve. The intensity of the fluorescence signal departed from the calibration curve when the concentration of AFP was over 1 ng/mL. Since AFP is a 70-kDa oncofetal glycoprotein,17 the upper limit of the linearity range was 1.43 × 10−11 mol/L. Since the upper limit of

the maximum solid angle for a point was 4π. Therefore, about 80% of the fluorescence emitted from the point can be reflected into the collimated beam. With the assistance of the best form lens and the collimating lens, the parallel beam was transferred into optical fiber and then captured by the single-channel spectrometer. Therefore, most of the fluorescence from the sample was measured by this novel LIF platform. Thus, excellent signal collection efficiency was achieved by this novel LIF platform. Moreover, in this novel LIF platform, the sample was loaded in a quartz capillary. Owing to the quartz capillary’s high surface-to-volume ratio, high excitation efficiency of the sample was guaranteed. In addition, the superb optical property of the quartz capillary provided excellent light transmission for both the laser beam and the fluorescence. Dominance of Using Ethanol as Assay Buffer. Using the appropriate solvent has a significant effect on the fluorescence intensity of Rhodamine 6G under laser irradiation. According to literature reports,15 the average number of photons per molecule (nT) is given by

nT = ϕf/ϕd Φf is the fluorescence quantum yield, and Φd is the photodestruction quantum efficiency. The high intensity of the laser beam will result in serious photobleaching. Thus, when a laser is used as the excitation source, the photodestruction quantum efficiency becomes vital. The photodestruction quantum efficiency of R6G is 5.0 × 10−7 in ethanol, while it is 1.9 × 10−5 in aqueous solution. Photostability of R6G is over 30 times greater in ethanol than in water. Moreover, D

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry the linearity range was well below the dissociation constant of the antibody (Kd is about 1 × 10−9 mol/L), in principle, it is possible for the novel immunoassay to have a wider linear range. The experiments demonstrated that the linear range of the immunoassay was slightly more than 2 orders of magnitude from 0.005 ng/mL to 1 ng/mL. A wider linear range made the novel immunoassay more suitable for clinical testing. The calibration curve is shown in Figure 5. There are several data

Figure 6. Interference effects from 10 ng/mL of various substances in a 0.1 ng/mL AFP solution with error bars (standard deviation from the mean, n = 4).

Glu), recombinant human epidermal growth factor (rHu EGF), and L-leucine (L-Leu) added to solutions containing 0.1 ng/mL AFP. The maximum signal increase and suppression were observed in the presence of L-Gly and L-Glu, respectively. The former one contributed a 15.39% increase to the signal intensity while a 6.77% decrease in the signal intensity was obtained in the presence of the latter one. This result showed strong antiinterference capability of the proposed immunoassay. Though AFP was employed as a sample tumor antigen in this experiment, the novel LIF immunoassay developed in this work was not restricted to AFP detection. By immobilizing different antibodies on magnetic beads, this immunoassay can be easily adapted for detecting other biomarkers. Since the capability of this immunoassay for ultralow level biomarker sensing has been demonstrated, the method can be used to detect many ultralow level biomarkers particularly for those whose levels are lower than the detection limits of a conventional ELISA immunoassay. Prostate-specific antigen (PSA), for example, has a detection limit of 0.1 ng/mL with a conventional ELISA immunoassay.30 Actually, the concentrations of PSA for most clinical patients, especially in the early stage of prostate cancer,4 are significantly lower than 0.1 ng/mL, which can be easily detected with the proposed method. Analytical Performance for Clinical Samples. The assay result for human serum containing AFP was used to evaluate the analytical reliability as well as potential clinical applications of the novel immunoassay. Undiluted serum containing AFP (30 μg/mL) was diluted with PBS prior to assay to keep the signal response within the calibration range. A set of serums with different concentrations of AFP were assayed by the novel immunoassay following the procedure mentioned above. The assay results are shown in Figure 7. Accuracy and precision data for the novel immunoassay in detecting AFP in human serum are shown in Table 1, with relative errors from −11.33% to 28.00% and relative standard deviations from 0.77% to 11.87%. These data indicated that this immunoassay was suitable for detecting AFP in complex clinical samples such as the serum used in this experiment. In addition, magnetic beads were externally controlled by a permanent magnet because of their superparamagnetic property. Therefore, complicated sample pretreatment was eliminated since interfering substances were separated from the analyte through the magnetic separation and antibody binding process. Furthermore, effective preconcentra-

Figure 5. Calibration curve for the determination of AFP with error bars (standard deviation from the mean, n = 4). Inset: the corresponding fluorescence spectra of the immunoassay in the presence of different concentrations of AFP (from bottom to top: 0.005, 0.01, 0.1, 0.3, 0.5, and 1 ng/mL). Fluorescence emissions were recorded at 558.90 nm with an excitation wavelength of 527 nm.

points deviating from the trendline by more than a few times magnitude of the error bars. Most of them were obtained from high concentration solutions. This may partially be attributed to the fluctuation of the spectrometer or may partially be due to the slight fluorescence saturation of the photodetector. The best fit regression line was y = 7795.25 × [AFP] + 367.67 (R = 0.9921), where y is the fluorescence intensity of R6G-labeled immunocomplex, as described above. The LOD is defined as 3δb/slope, which was about 0.0016 ng/mL. Compared with the conventional ELISA with LOD of about 2 ng/mL,16 an over 3 orders of magnitude improvement in LOD was achieved by the novel LIF immunoassay. Moreover, the LOD of the proposed method was far below the threshold of AFP in human serum (10 ng/mL).1 The improvement of LOD demonstrated the capability of the novel immunoassay for sensing biomarkers at ultralow level. A systematic comparison of the proposed method with those reported in the literature 16,17,28,29 is shown in Table S-1, Supporting Information. As shown in Table S-1, the proposed immunoassay has a much better detection limit as well as detection sensitivity. The reproducibility of the proposed immunoassay was also evaluated by intra-assay and interassay coefficients of variation (CV). The CV of the intra-assay, evaluated by using one immunosensor for six replicate determinations (0.1 ng/ mL), was 2.69%. The interassay CV on six immunosensors was 6.86% (0.1 ng/mL). These results indicated acceptable reproducibility and precision of the proposed immunoassay. To investigate interference effect, some nonspecific biomolecules were used for testing the proposed immunoassay. Figure 6 summarizes the interference from the presence of a 10 ng/mL level of L-glycine (L-Gly), CD 44, L-glutamic acid (LE

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

mentioned in the text. Comparison of the proposed method with those reported in the literature. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +86-28-85418180. Fax: +8628-85418180. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge funding provided by the National Natural Science Foundation of China (no. 21275105) and National Recruitment Program of Global Experts (NRPGE). The authors particularly thank the Zhengzhou Innosep Biosciences Co., Ltd, for offering carboxyl-modified magnetic beads.

Figure 7. Assay results of AFP detection in human serum by the novel immunoassay. Error bars indicated ±1 standard deviation (n = 4).

Table 1. Results from the Novel Immunoassay for AFP in Human Seruma sample 1 2 3 4 5 a

AFP (ng/ mL) 0.01 0.03 0.05 0.15 0.30

assay result (ng/ mL) 0.0128 0.0364 0.0457 0.1480 0.2660

relative error (%) 28.00 21.33 −8.60 −1.33 −11.33

■

RSD (%) 11.87 4.78 0.77 1.35 0.81

n = 4.

tion of the target biomarker was achieved due to the high loading capacity of the magnetic beads.

■

CONCLUSIONS In summary, a novel LIF platform was constructed by using a parabolic mirror as a highly efficient fluorescence collection device. A nearly 20-fold improvement in signal intensity was achieved by such a design and device. Subsequently, another 7 times enhancement in signal intensity was achieved by using ethanol instead of aqueous solution as assay buffer. Carboxylfunctionalized magnetic beads were used as a solid support to immobilize antibody to achieve a conventional sandwich assay. Due to the high loading capacity of the magnetic beads, effective preconcentration of biomarker was achieved using this process. The proposed new LIF immunoassay showed excellent analytical performance for α-fetoprotein (AFP) detection. The detection limit for AFP was over 3 orders of magnitude better than that of the conventional enzyme-linked immunosorbent assay (ELISA). Furthermore, the proposed method was successfully applied to AFP detection in real serum specimens with high sensitivity and acceptable precision. Though AFP was employed as a single sample for sample tumor antigen in this experiment, clearly this immunoassay is not restricted to AFP detection. Through immobilizing different antibodies on carboxyl-functionalized magnetic beads, this immunoassay can also be easily adapted for detection of many other biomarkers. Therefore, such a method definitely broadens the potential applications of LIF immunoassay in ultrasensitive biomarker detection, particularly for clinical samples.

■

REFERENCES

(1) Wu, J.; Fu, Z.; Yan, F.; Ju, H. TrAC, Trends Anal. Chem. 2007, 26, 679−688. (2) Groth, S.; Recke, A.; Vafia, K.; Ludwig, R.; Hashimoto, T.; Zillikens, D.; Schmidt, E. Br. J. Dermatol. 2011, 164, 76−82. (3) Kim, W.; Lee, J. E.; Li, X. F.; Kim, S. H.; Han, B. G.; Lee, B. I.; Kim, J. K.; Choi, K.; Kim, H. J. Mult. Scler. J. 2012, 18, 578−586. (4) Liu, D.; Huang, X.; Wang, Z.; Jin, A.; Sun, X.; Zhu, L.; Wang, F.; Ma, Y.; Niu, G.; Hight Walker, A. R. ACS Nano 2013, 7, 5568−5576. (5) Stern, E.; Vacic, A.; Rajan, N. K.; Criscione, J. M.; Park, J.; Ilic, B. R.; Mooney, D. J.; Reed, M. A.; Fahmy, T. M. Nat. Nanotechnol. 2009, 5, 138−142. (6) Zare, R. N. Annu. Rev. Anal. Chem. 2012, 5, 1−14. (7) Zhang, X.; Song, C.; Chen, L.; Zhang, K.; Fu, A.; Jin, B.; Zhang, Z.; Yang, K. Biosens. Bioelectron. 2011, 26, 3958−3961. (8) Seia, M. A.; Pereira, S. V.; Fontán, C. A.; De Vito, I. E.; Messina, G. A.; Raba, J. Sens. Actuators, B 2012, 168, 297−302. (9) Yu, Q.; Zhan, X.; Liu, K.; Lv, H.; Duan, Y. Anal. Chem. 2013, 85, 4578−4585. (10) Yu, Q.; Wang, X.; Duan, Y. Anal. Chem. 2014, 86, 1518−1524. (11) Lieb, M.; Meixner, A. Opt. Express 2001, 8, 458−474. (12) Fleury, L.; Tamarat, P.; Lounis, B.; Bernard, J.; Orrit, M. Chem. Phys. Lett. 1995, 236, 87−95. (13) Das, M.; Mardyani, S.; Chan, W.; Kumacheva, E. Adv. Mater. 2006, 18, 80−83. (14) Chu, K. H.; Zhou, Y.; Fang, Y.; Wang, L. H.; Li, J. Y.; Yao, C. Dyes Pigm. 2013, 98, 339−346. (15) Soper, S. A.; Shera, E. B.; Martin, J. C.; Jett, J. H.; Hahn, J. H.; Nutter, H. L.; Keller, R. A. Anal. Chem. 1991, 63, 432−437. (16) Bi, S.; Yan, Y.; Yang, X.; Zhang, S. Chem.Eur. J. 2009, 15, 4704−4709. (17) Chang, Y. F.; Chen, R. C.; Lee, Y. J.; Chao, S. C.; Su, L. C.; Li, Y. C.; Chou, C. Biosens. Bioelectron. 2009, 24, 1610−1614. (18) Yang, W.; Sun, X.; Wang, H. Y.; Woolley, A. T. Anal. Chem. 2009, 81, 8230−8235. (19) Kobayashi, T.; Kawakita, M.; Terachi, T.; Habuchi, T.; Ogawa, O.; Kamoto, T. J. Surg. Oncol. 2006, 94, 619−623. (20) Wright, L. M.; Kreikemeier, J. T.; Fimmel, C. J. Cancer Detect. Prev. 2007, 31, 35−44. (21) Templin, M. F.; Stoll, D.; Bachmann, J.; Joos, T. O. Comb. Chem. High Throughput Screening 2004, 7, 223−229. (22) Ozsoz, M.; Erdem, A.; Kerman, K.; Ozkan, D.; Tugrul, B.; Topcuoglu, N.; Ekren, H.; Taylan, M. Anal. Chem. 2003, 75, 2181− 2187. (23) Zhang, R.; Nakajima, H.; Soh, N.; Nakano, K.; Masadome, T.; Nagata, K.; Sakamoto, K.; Imato, T. Anal. Chim. Acta 2007, 600, 105− 113.

ASSOCIATED CONTENT

S Supporting Information *

Schematic diagram of the magnetic separation process and schematic diagram of the traditional optical system as F

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (24) Sam, S.; Touahir, L.; Salvador Andresa, J.; Allongue, P.; Chazalviel, J. N.; Gouget-Laemmel, A.; Henry de Villeneuve, C.; Moraillon, A.; Ozanam, F.; Gabouze, N. Langmuir 2009, 26, 809−814. (25) Li, B.; Lai, H.; Wei, Y.; Wang, X.; Chen, Y.; Zou, M.; Duan, Y. RSC Adv. 2014, 4, 50202−50207. (26) Soper, S. A.; Nutter, H. L.; Keller, R. A.; Davis, L. M.; Shera, E. B. Photochem. Photobiol. 1993, 57, 972−977. (27) Panchuk-Voloshina, N.; Haugland, R. P.; Bishop-Stewart, J.; Bhalgat, M. K.; Millard, P. J.; Mao, F.; Leung, W.-Y.; Haugland, R. P. J. Histochem. Cytochem. 1999, 47, 1179−1188. (28) Tsai, W. C.; Lin, I. Sens. Actuators, B 2005, 106, 455−460. (29) Wang, R.; Lu, X.; Ma, W. J. Chromatogr. B 2002, 779, 157−162. (30) Giljohann, D. A.; Mirkin, C. A. Nature 2009, 462, 461−464.

G

DOI: 10.1021/ac504515g Anal. Chem. XXXX, XXX, XXX−XXX