1
Electrophoresis 2014, 00, 1–7
Tingzhen Yang1 Marina Vdovenko2 Xue Jin1 Ivan Yu. Sakharov2 ∗ Shulin Zhao1 1 Key
Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources, College of Chemistry and Pharmacy, Guangxi Normal University, Guilin, China 2 Department of Chemistry, Lomonosov Moscow State University, Moscow, Russia
Received December 14, 2013 Revised March 20, 2014 Accepted March 23, 2014
Research Article
Highly sensitive microfluidic competitive enzyme immunoassay based on chemiluminescence resonance energy transfer for the detection of neuron-specific enolase A microfluidic competitive enzyme immunoassay based on chemiluminescence resonance energy transfer (CRET) was developed for highly sensitive detection of neuron-specific enolase (NSE). The CRET system consisted of horseradish peroxidase (HRP)/luminol as a light donor and fluorescein isothiocyanate as an acceptor. When fluorescein isothiocyanatelabeled antibody binds with HRP-labeled antigen to form immunocomplex, the donor and acceptor are brought close each other and CRET occurs in the immunocomplex. In the MCE, the immunocomplex and excess HRP–NSE were separated, and the chemiluminescense intensity of immunocomplex was used to estimate NSE concentration. The calibration curve showed a linearity in the range of NSE concentrations from 9.0 to 950 pM with a correlation coefficient of 0.9964. Based on a S/N of 3, the detection limit for NSE determination was estimated to be 4.5 pM, which is two-order magnitude lower than that of without CRET detection. This assay was applied for NSE quantification in human serum. The obtained results demonstrated that the proposed immunoassay may serve as an alternative tool for clinical analysis of NSE. Keywords: Chemiluminescence / Enzyme immunoassays / Microchip electrophoresis / Neuron-specific enolase / Resonance energy transfer DOI 10.1002/elps.201300630
1 Introduction Increased concentration of neuron-specific enolase (NSE) in human serum may be attributed to cerebral injury due to physical damage or ischemia caused by infarction or cerebral hemorrhage, coupled with increased permeability of the blood–brain barrier [1], and the serum NSE concentration has been reported to correlate with the extent of damage and neurological outcome [2]. Additionally, a secondary elevation of serum NSE concentration may be a marker of delayed neuronal injury resulting from cerebral vasospasm, and serum NSE concentration reflects the metabolic and secretion activity of tumors [3]. As the ␥ -subunit of enolase, NSE presents primarily in the cytoplasm of neurons and neuroendocrine cells. The NSE levels are 5–12 ng/mL in serum and 20 ng/mL
Correspondence: Professor Shulin Zhao, Key Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources, College of Chemistry and Pharmacy, Guangxi Normal University, Guilin 541004, China E-mail: [email protected] Fax: +86-773-5832294
in cerebrospinal fluid in normal human beings [4]. Therefore, the detection of serum NSE concentration may be a useful tool for the tumors diagnosis. Several assays procedures based on electrochemical immunoassay [5–7], gel electrophoresis [8], and fluoroimmunoassay [9] have been reported to be effective for NSE detection during the past two decades. Although each method has its advantages, many reported techniques involve still some tedious and time-consuming procedures. Thus, it is highly desirable to develop simple, sensitive, and selective assay for the determination of NSE in human serum. MCE is a modern trace analysis technique. Since Manz and co-workers presented this technique in 1993 [10], MCE has been rapidly developed, and widely used in chemical, biochemical, and clinic analysis [11]. Microfluidic immunoassay also attracts a great attention in analytical chemists. The advantages of MCE include shortened assay time, lower sample/reagent consumption, and the possibility of being fully automated [12, 13]. However, since the microfluidic channels are extremely narrow, and minimum amounts of
∗
Abbreviations: CL, chemiluminescence; CRET, chemiluminescence resonance energy transfer; HRP, horseradish peroxidase; NSE, neuron-specific enolase; PIP, Para-iodophenol C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Additional corresponding author: Professor Ivan Yu. Sakharov, E-mail: [email protected]
Colour Online: See the article online to view Fig. 6 and Scheme 1 in colour.
www.electrophoresis-journal.com
2
T. Yang et al.
detectable labels are used, a highly sensitive detection system is required for this assay. Although LIF detection as a detection system is already widely used in MCE to improve its sensitivity [14–17], a conventional LIF detector is sophisticated and functions at a limited number of wavelengths. Furthermore, a derivatization of analytes with a fluorophore is often necessary. Chemiluminescence (CL)-based detection offers advantages such as simplicity in instrumentation, high sensitivity, and wide linear range [18, 19]. It has been widely used in immunoassays [20–22]. CL resonance energy transfer (CRET) involves nonradioactive (dipole–dipole) energy transfer between a CL donor and fluorophore acceptor when are in close proximity (normally ⬍10 nm) [23, 24]. Since no external light source is used for excitation in CRET, nonspecific signals caused by external light excitation often observed in fluorescence measurements can be minimized. Therefore, using CRET system can enhance the detection sensitivity. Recently, Zhang and co-workers developed a CRET-based assay for the detection of adenosine triphosphate present in cancer cells [25]. Willner and co-workers applied horseradish peroxidase (HRP) mimicking DNAzyme-stimulated CRET to quantum dots for the detection of DNA, metal ions, and aptamer substrate complexes [26, 27]. Lee and co-workers reported a graphene-based CRET immunosensing platform for homogeneous detection of C-reactive protein in human serum samples [28]. These studies indicated that the CRET system was simple and useful to detect biomolecules with low detection limit. In the present work, taking the advantages of approaches mentioned above, we designed a novel analytical strategy based on microfluidic enzyme immunoassay with CRETbased detection for the determination of NSE. The antigen (NSE) was labeled with HRP, which catalyzed the oxidation of luminol with the production of light (donor), and the antibody was labeled with FITC (acceptor). When the immunocomplex is formed, CRET occurs in immunocomplex. Based on this principle, a highly sensitive microfluidic enzyme immunoassay for NSE determination was developed, and applied for the quantification of NSE in human serum.
2 Materials and methods
Electrophoresis 2014, 00, 1–7
Figure 1. The diagram of the layout of the glass/PDMS microchip. S: sample reservoir; B: buffer reservoir; SW: sample waste reservoir; BW: buffer waste reservoir; R: the oxidizer reagent reservoir.
Solutions of the antigen and antibodies were prepared by dissolving the reagents in 30 mM PBS, pH 7.4. The electrophoretic buffer was 30 mM borate solution (pH 9.0) containing 0.1 mM luminol and 30 mM SDS. The oxidizer solution was 35 mM NaHCO3 buffer solution (pH 8.8) containing 100 mM H2 O2 and 0.5 mM PIP. All solutions were filtered through 0.22 m membrane filters before use. CL spectra were measured with a LS-55 luminescence spectrometer (Perkin-Elmer, USA); UV-visible spectra were measured with a TU-1901 UV-visible spectrophotometer (Beijing Purkinje General Instrument, China). The MCE–CL detection was performed using a laboratory built system described previously [29]. A home-made glass/PDMS microchip was used in this work, and its schematic layout is illustrated in Fig. 1. The procedure of microchip fabrication was as described previously [19]. The channel between reservoir S and SW was used for sampling, the channel between B and BW was used for the separation and the channel between R and BW was used for the oxidizer introduction. The width of microchannels is 65 m (except oxidizer introduction channel is 250 m). The depth of all microchannels is 25 m, and the double “T” size is 60 m. 2.2 Immunological reaction A 20 L volume of standard NSE or sample solution were mixed with 20 L of 0.25 M FITC-labeled antibody (Ab*) and 20 L of 0.50 M HRP-labeled antigen (Ag*) in a 0.5 mL centrifuge tube. This solution was diluted up to 100 L, and the mixture solution was then incubated for 35 min at 37°C. The resulting solution was analyzed by MCE–CL.
2.1 Reagents and apparatus 2.3 MCE–CL assay All reagents used in this work were of analytical grade. NSE (Ag), FITC-labeled antibody (Ab*), and HRP-labeled antigen (Ag*) were purchased from Sangon Biotechnology (Shanghai, China). Luminol was purchased from Aldrich (Milwaukee, WI, USA). SDS was provided by Shanghai Reagents (Shanghai, China). Para-iodophenol (PIP), hydrogen peroxide (H2 O2 ), and sodium hydrogen carbonate (NaHCO3 ) were obtained from Taopu Chemicals (Shanghai, China). Water was purified by employing a Milli-Q plus 185 equip from Millipore (Bedford, MA, USA), and used throughout the work. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The microchannels were rinsed sequentially with 0.1 M NaOH, water, and electrophoretic buffer for 5 min each. Prior to the MCE separation, the reservoirs B, S, SW, and BW were filled with the electrophoretic buffer, reservoir R was filled with the oxidizer solution, and vacuum was applied to the reservoir BW in order to fill the separation channel with the electrophoretic buffer. Then, the electrophoretic buffer in reservoir S was replaced by sample solution. For loading the sample solution, a set of electrical potentials was applied to five reservoirs: reservoir S at 550 V, reservoir B at 200 V, www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
reservoir BW at 300 V, reservoir SW at grounded, and reservoir R floating. The sample solution was transported from reservoir S to SW in a pinched mode. After 25 s, potentials were switched to reservoir B, S, SW, and R at 2700, 1700, 1700, and 550 V, respectively, while reservoir BW was grounded for separation and detection.
3 Results and discussion 3.1 Design of CRET system Design of an effective CRET system is critical for developing a useful microfluidic enzyme immunoassay with CRET-based detection. Luminol-PIP-H2 O2 reaction is one of the most sensitive CL reactions, and widely used in bioanalysis [30]. Since HRP effectively catalyzes this CL reaction [31,32], this enzyme was used to label NSE (antigen). FITC is a highly fluorescent compound, and a widely used fluorescent label in immunoassay [33], and the CL emission spectrum of luminol was largely overlapped with the FITC absorption spectrum [34]. Therefore, FITC was selected to label the anti-NSE antibody. Based on the principle of CRET, a novel homogeneous microfluidic enzyme immunoassay for NSE detection was developed. The competitive immunoassay is illustrated in Scheme 1. Free NSE and HRP-labeled NSE compete for interaction with FITC-labeled anti-NSE antibody. Two complexes, that is, NSE–FITC–anti-NSE and HRP–NSE–FITC–anti-NSE are formed. However, CRET occurs only in the HRP–NSE– FITC–anti-NSE immunocomplex, where the donor and acceptor are brought close to each other, which results in CL intensity markedly increase compared to that of HRP–NSE.
Microfluidics and Miniaturization
3
Therefore, the concentration of NSE is inversely proportional to the CRET intensity observed. In the MCE experiments, the HRP–NSE–FITC–anti-NSE immunocomplex and free HRP–NSE were quickly separated, and the CL intensity of the immunocomplex was used to determine NSE concentration. Because the concentration of NSE is inversely proportional to that of the HRP–NSE–FITC– anti-NSE immunocomplex, the concentration of NSE in the sample can be determined by measuring the CL intensity of HRP–NSE–FITC–anti-NSE immunocomplex after being separated from the HRP–NSE with high sensitivity.
3.2 Optimization of conditions for immunoreaction High concentration of FITC-labeled antibody and HRPlabeled antigen gave a high CL intensity for the immunocomplex when NSE concentration was kept unchanged. However, the FITC-labeled antibody concentration higher than 50 nM had no benefits for the increase in the CL intensity of immunocomplex. In order to ensure the FITC-labeled antibody and HRP-labeled antigen to have an enough immunocapacity with NSE for competitive immunoreactions, the concentration of HRP-labeled antigen should be in excess. Considering the detection sensitivity and reagent consumption as a whole, 50 nM FITC-labeled antibody and 100 nM HRP-labeled antigen were chosen for the immunoreactions. The dependence of immunoreaction time on the efficiency of the production of the immunocomplex was studied. The HRP–NSE and FITC–anti-NSE antibody were mixed and incubated for 20–45 min at 37°C, and then were injected into the MCE channel. As shown in Fig. 2, the CL signal increased
Scheme 1. Schematic illustration of CRET-based competitive immunoassay for NSE.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
4
T. Yang et al.
Figure 2. The effect of the incubation time on the CL intensity of the immunocomplex of FITC–anti-NSE and HRP–NSE. A mixture of 0.05 M FITC–anti-NSE and 0.1 M HRP–NSE was incubated at 37°C, and then analyzed at different time intervals from 20 to 45 min. Electrophoresis buffer was 30 mM borate solution (pH 9.0) containing 30 mM SDS and 0.1 mM luminol. The oxidizer solution was 35 mM NaHCO3 buffer (pH 8.8) containing 0.1 M H2 O2 and 0.5 mM PIP.
rapidly with increasing the incubation time up to 35 min, and then remained constant. The results indicated that the binding of HRP–NSE to FITC–anti-NSE reached equilibrium for 35 min.
3.3 Optimization of conditions for CL reaction To improve the sensitivity of CL detection following the CE separation, the experimental conditions such as pH of oxidizer solution, concentrations of H2 O2 , and PIP in postcolumn oxidizer solution, luminol, and SDS in the electrophoresis buffer solution were optimized. Since the volume of the solution in the separation channel of MCE is very small compared with the volume of the postcolumn oxidizer solution, the acidity of CL reaction is mainly dependent on the oxidizer solution. The effect of pH of oxidizer solution (35 mM NaHCO3 ) on the CL intensity was investigated by varying pH from 8.0 to 9.5. It was found that the maximum CL was obtained at pH 8.8. H2 O2 as an oxidizer of luminol plays an important role in the CL reaction. With the increase of H2 O2 concentration from 0.06 to 0.10 M, the CL intensity also increased, and further increasing H2 O2 concentration result in the decrease of CL intensity. Therefore, a 0.1 M H2 O2 solution was used in further work. The effect of luminol concentration was investigated. With the increase of lumilol concentration from 0.05 to 0.10 mM, the CL intensity increased. Further increasing the luminol concentration resulted in the decrease of CL
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Electrophoresis 2014, 00, 1–7
Figure 3. Effect of PIP concentration on CL intensity produced in enhanced CL reaction. Thirty micromolar borate solution (pH 9.0) containing 30 mM SDS and 0.1 mM luminol, and 35 mM NaHCO3 (pH 8.8) containing 0.1 M H2 O2 and 0.1–0.6 mM PIP were used as electrophoresis buffer and the oxidizer solution, respectively.
intensity. Thus, the optimal luminol concentration was 0.1 mM. It is well known that HRP-catalyzed luminol-H2 O2 CL reaction may be significantly enhanced at the introduction of enhancers to luminol solution. The most popular enhancer is PIP [35]. Therefore, the effect of PIP concentration in the range of 0.1–0.6 mM on the CL intensity was also investigated, and the results were shown in Fig. 3. As can be seen, the optimum PIP concentration was 0.5 mM. Under favorable conditions, PIP increased CL intensity approximately seven times.
3.4 Optimization of the electrophoresis conditions The conditions for CE separation including pH values of the electrophoretic buffer and SDS concentration were optimized, and the separations were characterized by resolution (Rs) values (Rs = 2(T2 − T1 )/(W2 + W1 ), where T1 and T2 are the migration time of peak-1 and peak-2, W1 and W2 are the bottom width of peak-1 and peak-2). The obtained results indicated that with the increase of pH value from 8.5 to 9.0, the Rs of the immunocomplex and HRP–NSE increased. Further increasing the pH resulted in the decrease of Rs value (Fig. 4). In addition, the concentration of SDS in electrophoretic buffer also affected the separation efficiency. When the concentration of SDS was 30 mM, a maximal R value for immunocomplex and HRP–NSE was obtained. Upon MCE performance, the immunocomplex and HRP–NSE are separated, and electropherogram is presented in Fig. 5A. After the introduction of free NSE to the reaction mixture, a ration of CL intensity (peaks height)
www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
Microfluidics and Miniaturization
5
Figure 4. The effect of pH value of the electrophoretic buffer on resolution of the immunocomplex of FITC–anti-NSE with HRP– NSE and HRP–NSE. Electrophoretic buffer was 30 mM borate buffer (pH 9.0) containing 0.1 mM luminol and 30 mM SDS at different pH value. The oxidizer solution was 35 mM NaHCO3 buffer containing 0.1 M H2 O2 (pH 8.8) and 0.5 mM PIP.
corresponding to the concentrations of the immunocomplex and HRP–NSE is changed (Fig. 5B). These results demonstrated the competitive binding of NSE to the specific antibody. It is worth noting that the immunocomplex was well separated from free HRP–NSE making the proposed assay selective and useful for clinical analysis.
3.5 Analytical figures of merit The method was evaluated in terms of response linearity, LOD, and reproducibility. Under the selected MCE–CL conditions, seven standard NSE solutions at various concentrations were analyzed. Figure 6 showed the plot of relative CL intensity versus the concentration of NSE. Linear regression analysis of data yielded the following equation: ⌬ I = 7.35 [NSE] + 1.25, where ⌬ I is the relative CL intensity (V), and [NSE] is the concentration of NSE (pM). The calibration curve showed excellent linearity in the range from 9 to 950 pM with a correlation coefficient of 0.9964. Based on a S/N of 3, the detection limit for NSE was estimated to be 4.8 pM. Assay reproducibility was estimated by analyzing a standard solution of NSE of 450 pM (n = 9). The results showed that the RSDs of the migration time and peak height of the peak 1 were 2.8 and 3.5%, whereas for peak 2 the migration time and peak height were 2.6 and 3.9%, respectively.
3.6 Quantification of NSE in human serum Human serum samples taken from four healthy volunteers and three lung cancer patients were analyzed to demonstrate the feasibility of proposed MCE–CL method for the determination of NSE. The electropherograms obtained upon the
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 5. Electropherograms from separating the noncompetitive and competitive immunoreaction solutions. (A) without NSE; (B) with 0.45 nM NSE. The concentration of FITC–anti-NSE and HRP–NSE are 0.05 and 0.1 M, respectively. MCE conditions: electrophoretic buffer is 30 mM phosphate buffer (pH 9.0) containing 30 mM SDS and 0.1 mM luminol; the oxidizer solution was 35 mM NaHCO3 buffer containing 0.1 M H2 O2 (pH 8.8) and 0.5 mM PIP. Peak identification: (1) HRP–NSE–FITC–anti-NSE immunocomplex; (2) HRP–NSE.
analysis of the studied samples are shown in Fig. 7. The results are summarized in Table 1. The NSE level in the serum samples from healthy subjects was found be in the range 6.6 × 10−11 –10.1 × 10−11 M. The NSE level in the samples of lung cancer patients was found to be higher and lay in the range 96.2 × 10−11 –130.4 × 10−11 M. RSDs for the determination of NSE were in the ranges of 2.7–5.1%. Recovery of NSE from these samples was studied. NSE was spiked to each sample and analyzed. Recoveries were found to be in the range of 96.7–102.6%. The results suggested a potential application of the proposed assay in express primary diagnosis of diseases such as lung cancer patients.
www.electrophoresis-journal.com
6
T. Yang et al.
Electrophoresis 2014, 00, 1–7 Table 1. Assay results and recovery of NSE in human serum samples
Sample
RSD Found (10−11 M) (%, n = 3)
Normal 1 6.6 Normal 2 8.3 Normal 3 10.1 Normal 4 7.9 Patient 1 96.2 Patient 2 130.4 Patient 3 117.3
2.7 3.5 3.1 3.3 4.1 5.1 4.7
Added Total found Recovery (10−11 M) (10−11 M) (%) 10 10 10 10 100 100 100
16.32 18.16 20.36 17.81 192.9 231.6 214.4
97.2 98.6 102.6 99.1 96.7 101.2 97.1
4 Concluding remarks Figure 6. The plot of relative CL intensity versus the concentration of NSE. MCE conditions were as in Fig. 5.
A novel microfluidic competitive enzyme immunoassay based on CRET has been developed for the detection of NSE. The CRET system consisted of HRP/luminol as a light donor and FITC as an acceptor. When FITC-labeled antibody binds to the conjugate of the antigen (NSE) and HRP, the donor and acceptor are brought close each other and CRET occurs. In the MCE, the immunocomplex and HRP–NSE conjugate were separated. By using the CRET detection, the method allows the detection of 4.8 × 10−12 M for NSE, which is two orders magnitude lower than that of without CRET detection. To the best of our knowledge, the proposed CRET-based immunoassay for NSE is one of the most sensitive immunoassay in a microfluidic format. Although microfluidic competitive enzyme immunoassay based on CRET is demonstrated only for NSE in this work, assays for other compounds can be developed similarly by using corresponding antigens and respective antibodies. This work was supported by the National Natural Science Foundations of China (NSFC, Grant nos. 21175030, 21311120056), and the Russian Foundation for Basic Research (13-04-91164-GFEN_a). The authors have declared no conflict of interest.
5 References [1] Barton, A. C., Davis, F., Higso, S. P. J., Anal. Chem. 2008, 80, 9411–9416. [2] Martens, P., Raabe, A., Johnsson, P., Stroke 1998, 29, 2363–2366. [3] Baudin, E., Gigliotti, A., Ducreux, M., Ropers, J., Comoy, E., Sabourin, J. C., Bidart, J. M., Cailleux, A. F., Bonacci, R, Ruffie P., Schlumberger, M., Br. J. Cancer 1998, 78, 1102–1107. Figure 7. Electropherograms of serum samples. (A) normal serum sample; (B) patient serum sample (diluted ten times). The conditions were as Fig. 6. Peak identification: (1) HRP–anti-NSE– FITC–NSE immunocomplex; (2) HRP–anti-NSE. MCE conditions were as in Fig. 5.
[4] Marangos, P. J., Schmechel, D. E., Ann. Rev. Neurosci.1987, 10, 269–295. [5] Barton, A. C., Seamus, F. D., Higson, P. J., Anal. Chem. 2008, 80, 9411–9416. [6] Zhong, Z. Y., Peng, N., Qing, Y., Shan, J. L., Li, M. X., Guan, W., Dai, N., Gu, X. Q., Wang, D., Electrochimica Acta 2011, 56, 5624–5629.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
Microfluidics and Miniaturization
7
[7] Han, J., Zhuo, Y., Chai, Y. Q., Yuan, Y. L., Yuan, R., Biosens. Bioelectron. 2012, 31, 399–405.
[21] Zhao, L. X., Sun L., Chu, X. G., Trends Anal. Chem. 2009, 28, 404–414.
[8] Inga, A., Monika, B., Markus, O., Sigrid, P., Otto, W., Hans, A. K., Olaf, G., Thomas, W., Lancet 1996, 348, 846–849.
[22] Marquette, C. A., Blum, L. J., Bioanalysis 2009, 1, 1259–1269.
[9] Cao, Z. J., Li, H., Laua C., Zhang, Y. H., Anal. Chim. Acta 2011, 698, 44–50. [10] Harrison, D. J., Fluri, K., Seiler, K., Fan, Z., Effenhauser, C. S., Manz, A., Science 1993, 261, 895–897. [11] Reyes, D. R., Iossifidis, D., Manz, A., Anal. Chem. 2002, 74, 26223–26236. [12] Morozov, V. N., Groves, S., Turell, M. J., Bailey, C., J. Am. Chem. Soc. 2007, 129, 12628–12629. [13] Bange, A., Halsall, H. B., Heineman, W. R., Biosens. Bioelectron. 2005, 20, 2488–2503. [14] Koutny, L. B., Schmalzing, D., Taylor, T. A., Fuchs, M., Anal. Chem. 1996, 68, 18–22. [15] Chiem, N., Harrison, D. J., Anal. Chem. 1997, 69, 373–378. [16] Wang, J., Ibanez, A., Chatrathi, M. P., J. Am. Chem. Soc. 2003, 125, 8444–8445. [17] Dishinger, J. F., Kennedy, R. T., Anal. Chem. 2007, 79, 947–954. [18] Roda, A., Guardigli, M., Anal. Bioanal. Chem. 2012, 402, 69–76. [19] Vdovenko, M. M., Demiyanova, A. S., Chemleva, T. A., Sakharov, I. Yu., Talanta 2012, 94, 223–226. [20] Sakharov, I. Yu., Demiyanova, A. S., Gribas, A. V., Uskova, N. A., Efremov, E. E., Vdovenko, M. M., Talanta 2013, 115, 414–417.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[23] Sapsford, K. E. Berti, L., Medintz, I. L., Angew. Chem. Int. Ed. 2006, 45, 4562–4588. [24] Huang, X. Y., Li, L., Qian, H. F., Dong, C. Q., Ren, J., Angew. Chem. Int. Ed. 2006, 45, 5140–5143. [25] Zhang, S., Yan, Y., Bi, S., Anal. Chem. 2009, 81, 8695–8701. [26] Freeman, R., Liu, X., Willner, I., J. Am. Chem. Soc. 2011, 133, 11597–11604. [27] Liu, X., Freeman, R., Golub, E., Willner, I., ACS Nano 2011, 5, 7648–7655. [28] Lee, J. S., Joung, H. A., Kim, M. G., Park, C. B., ACS Nano, 2012, 6, 2978–2983. [29] Zhao, S., Li, X., Liu, Y. M., Anal. Chem. 2009, 81, 3873–3878. ´ A. M., Aaron, J. J., Anal. [30] Lara, F. J., Garc´ıa-Campana, Chim. Acta 2010, 679, 17–30. [31] Thorpe, G. H. G., Kricka, L. J., Moseley, S. R., Whitehead, T. P., Clin. Chem. 1985, 31, 1335–1341. [32] Whitehead, T. P., Thorpe, G. H. G., Carter, T.J.N., Groucutt, C., Kricka, L. J., Nature 1983, 305, 159–160. [33] Shi, M. Zhao, S., Huang, Y., Liu, Y. M., Ye, F., J. Chromatogr. B 2011, 879, 2840–2846. [34] Qin, G., Zhao, S., Huang, Y., Jiang, J., Ye, F., Anal. Chem. 2012, 84, 2708–2712. [35] Wang, J., Ren, J., Electrophoresis 2005, 26, 2402–2408.
www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
Tingzhen Yang1 Marina Vdovenko2 Xue Jin1 Ivan Yu. Sakharov2 ∗ Shulin Zhao1 1 Key
Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources, College of Chemistry and Pharmacy, Guangxi Normal University, Guilin, China 2 Department of Chemistry, Lomonosov Moscow State University, Moscow, Russia
Received December 14, 2013 Revised March 20, 2014 Accepted March 23, 2014
Research Article
Highly sensitive microfluidic competitive enzyme immunoassay based on chemiluminescence resonance energy transfer for the detection of neuron-specific enolase A microfluidic competitive enzyme immunoassay based on chemiluminescence resonance energy transfer (CRET) was developed for highly sensitive detection of neuron-specific enolase (NSE). The CRET system consisted of horseradish peroxidase (HRP)/luminol as a light donor and fluorescein isothiocyanate as an acceptor. When fluorescein isothiocyanatelabeled antibody binds with HRP-labeled antigen to form immunocomplex, the donor and acceptor are brought close each other and CRET occurs in the immunocomplex. In the MCE, the immunocomplex and excess HRP–NSE were separated, and the chemiluminescense intensity of immunocomplex was used to estimate NSE concentration. The calibration curve showed a linearity in the range of NSE concentrations from 9.0 to 950 pM with a correlation coefficient of 0.9964. Based on a S/N of 3, the detection limit for NSE determination was estimated to be 4.5 pM, which is two-order magnitude lower than that of without CRET detection. This assay was applied for NSE quantification in human serum. The obtained results demonstrated that the proposed immunoassay may serve as an alternative tool for clinical analysis of NSE. Keywords: Chemiluminescence / Enzyme immunoassays / Microchip electrophoresis / Neuron-specific enolase / Resonance energy transfer DOI 10.1002/elps.201300630
1 Introduction Increased concentration of neuron-specific enolase (NSE) in human serum may be attributed to cerebral injury due to physical damage or ischemia caused by infarction or cerebral hemorrhage, coupled with increased permeability of the blood–brain barrier [1], and the serum NSE concentration has been reported to correlate with the extent of damage and neurological outcome [2]. Additionally, a secondary elevation of serum NSE concentration may be a marker of delayed neuronal injury resulting from cerebral vasospasm, and serum NSE concentration reflects the metabolic and secretion activity of tumors [3]. As the ␥ -subunit of enolase, NSE presents primarily in the cytoplasm of neurons and neuroendocrine cells. The NSE levels are 5–12 ng/mL in serum and 20 ng/mL
Correspondence: Professor Shulin Zhao, Key Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources, College of Chemistry and Pharmacy, Guangxi Normal University, Guilin 541004, China E-mail: [email protected] Fax: +86-773-5832294
in cerebrospinal fluid in normal human beings [4]. Therefore, the detection of serum NSE concentration may be a useful tool for the tumors diagnosis. Several assays procedures based on electrochemical immunoassay [5–7], gel electrophoresis [8], and fluoroimmunoassay [9] have been reported to be effective for NSE detection during the past two decades. Although each method has its advantages, many reported techniques involve still some tedious and time-consuming procedures. Thus, it is highly desirable to develop simple, sensitive, and selective assay for the determination of NSE in human serum. MCE is a modern trace analysis technique. Since Manz and co-workers presented this technique in 1993 [10], MCE has been rapidly developed, and widely used in chemical, biochemical, and clinic analysis [11]. Microfluidic immunoassay also attracts a great attention in analytical chemists. The advantages of MCE include shortened assay time, lower sample/reagent consumption, and the possibility of being fully automated [12, 13]. However, since the microfluidic channels are extremely narrow, and minimum amounts of
∗
Abbreviations: CL, chemiluminescence; CRET, chemiluminescence resonance energy transfer; HRP, horseradish peroxidase; NSE, neuron-specific enolase; PIP, Para-iodophenol C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Additional corresponding author: Professor Ivan Yu. Sakharov, E-mail: [email protected]
Colour Online: See the article online to view Fig. 6 and Scheme 1 in colour.
www.electrophoresis-journal.com
2
T. Yang et al.
detectable labels are used, a highly sensitive detection system is required for this assay. Although LIF detection as a detection system is already widely used in MCE to improve its sensitivity [14–17], a conventional LIF detector is sophisticated and functions at a limited number of wavelengths. Furthermore, a derivatization of analytes with a fluorophore is often necessary. Chemiluminescence (CL)-based detection offers advantages such as simplicity in instrumentation, high sensitivity, and wide linear range [18, 19]. It has been widely used in immunoassays [20–22]. CL resonance energy transfer (CRET) involves nonradioactive (dipole–dipole) energy transfer between a CL donor and fluorophore acceptor when are in close proximity (normally ⬍10 nm) [23, 24]. Since no external light source is used for excitation in CRET, nonspecific signals caused by external light excitation often observed in fluorescence measurements can be minimized. Therefore, using CRET system can enhance the detection sensitivity. Recently, Zhang and co-workers developed a CRET-based assay for the detection of adenosine triphosphate present in cancer cells [25]. Willner and co-workers applied horseradish peroxidase (HRP) mimicking DNAzyme-stimulated CRET to quantum dots for the detection of DNA, metal ions, and aptamer substrate complexes [26, 27]. Lee and co-workers reported a graphene-based CRET immunosensing platform for homogeneous detection of C-reactive protein in human serum samples [28]. These studies indicated that the CRET system was simple and useful to detect biomolecules with low detection limit. In the present work, taking the advantages of approaches mentioned above, we designed a novel analytical strategy based on microfluidic enzyme immunoassay with CRETbased detection for the determination of NSE. The antigen (NSE) was labeled with HRP, which catalyzed the oxidation of luminol with the production of light (donor), and the antibody was labeled with FITC (acceptor). When the immunocomplex is formed, CRET occurs in immunocomplex. Based on this principle, a highly sensitive microfluidic enzyme immunoassay for NSE determination was developed, and applied for the quantification of NSE in human serum.
2 Materials and methods
Electrophoresis 2014, 00, 1–7
Figure 1. The diagram of the layout of the glass/PDMS microchip. S: sample reservoir; B: buffer reservoir; SW: sample waste reservoir; BW: buffer waste reservoir; R: the oxidizer reagent reservoir.
Solutions of the antigen and antibodies were prepared by dissolving the reagents in 30 mM PBS, pH 7.4. The electrophoretic buffer was 30 mM borate solution (pH 9.0) containing 0.1 mM luminol and 30 mM SDS. The oxidizer solution was 35 mM NaHCO3 buffer solution (pH 8.8) containing 100 mM H2 O2 and 0.5 mM PIP. All solutions were filtered through 0.22 m membrane filters before use. CL spectra were measured with a LS-55 luminescence spectrometer (Perkin-Elmer, USA); UV-visible spectra were measured with a TU-1901 UV-visible spectrophotometer (Beijing Purkinje General Instrument, China). The MCE–CL detection was performed using a laboratory built system described previously [29]. A home-made glass/PDMS microchip was used in this work, and its schematic layout is illustrated in Fig. 1. The procedure of microchip fabrication was as described previously [19]. The channel between reservoir S and SW was used for sampling, the channel between B and BW was used for the separation and the channel between R and BW was used for the oxidizer introduction. The width of microchannels is 65 m (except oxidizer introduction channel is 250 m). The depth of all microchannels is 25 m, and the double “T” size is 60 m. 2.2 Immunological reaction A 20 L volume of standard NSE or sample solution were mixed with 20 L of 0.25 M FITC-labeled antibody (Ab*) and 20 L of 0.50 M HRP-labeled antigen (Ag*) in a 0.5 mL centrifuge tube. This solution was diluted up to 100 L, and the mixture solution was then incubated for 35 min at 37°C. The resulting solution was analyzed by MCE–CL.
2.1 Reagents and apparatus 2.3 MCE–CL assay All reagents used in this work were of analytical grade. NSE (Ag), FITC-labeled antibody (Ab*), and HRP-labeled antigen (Ag*) were purchased from Sangon Biotechnology (Shanghai, China). Luminol was purchased from Aldrich (Milwaukee, WI, USA). SDS was provided by Shanghai Reagents (Shanghai, China). Para-iodophenol (PIP), hydrogen peroxide (H2 O2 ), and sodium hydrogen carbonate (NaHCO3 ) were obtained from Taopu Chemicals (Shanghai, China). Water was purified by employing a Milli-Q plus 185 equip from Millipore (Bedford, MA, USA), and used throughout the work. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The microchannels were rinsed sequentially with 0.1 M NaOH, water, and electrophoretic buffer for 5 min each. Prior to the MCE separation, the reservoirs B, S, SW, and BW were filled with the electrophoretic buffer, reservoir R was filled with the oxidizer solution, and vacuum was applied to the reservoir BW in order to fill the separation channel with the electrophoretic buffer. Then, the electrophoretic buffer in reservoir S was replaced by sample solution. For loading the sample solution, a set of electrical potentials was applied to five reservoirs: reservoir S at 550 V, reservoir B at 200 V, www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
reservoir BW at 300 V, reservoir SW at grounded, and reservoir R floating. The sample solution was transported from reservoir S to SW in a pinched mode. After 25 s, potentials were switched to reservoir B, S, SW, and R at 2700, 1700, 1700, and 550 V, respectively, while reservoir BW was grounded for separation and detection.
3 Results and discussion 3.1 Design of CRET system Design of an effective CRET system is critical for developing a useful microfluidic enzyme immunoassay with CRET-based detection. Luminol-PIP-H2 O2 reaction is one of the most sensitive CL reactions, and widely used in bioanalysis [30]. Since HRP effectively catalyzes this CL reaction [31,32], this enzyme was used to label NSE (antigen). FITC is a highly fluorescent compound, and a widely used fluorescent label in immunoassay [33], and the CL emission spectrum of luminol was largely overlapped with the FITC absorption spectrum [34]. Therefore, FITC was selected to label the anti-NSE antibody. Based on the principle of CRET, a novel homogeneous microfluidic enzyme immunoassay for NSE detection was developed. The competitive immunoassay is illustrated in Scheme 1. Free NSE and HRP-labeled NSE compete for interaction with FITC-labeled anti-NSE antibody. Two complexes, that is, NSE–FITC–anti-NSE and HRP–NSE–FITC–anti-NSE are formed. However, CRET occurs only in the HRP–NSE– FITC–anti-NSE immunocomplex, where the donor and acceptor are brought close to each other, which results in CL intensity markedly increase compared to that of HRP–NSE.
Microfluidics and Miniaturization
3
Therefore, the concentration of NSE is inversely proportional to the CRET intensity observed. In the MCE experiments, the HRP–NSE–FITC–anti-NSE immunocomplex and free HRP–NSE were quickly separated, and the CL intensity of the immunocomplex was used to determine NSE concentration. Because the concentration of NSE is inversely proportional to that of the HRP–NSE–FITC– anti-NSE immunocomplex, the concentration of NSE in the sample can be determined by measuring the CL intensity of HRP–NSE–FITC–anti-NSE immunocomplex after being separated from the HRP–NSE with high sensitivity.
3.2 Optimization of conditions for immunoreaction High concentration of FITC-labeled antibody and HRPlabeled antigen gave a high CL intensity for the immunocomplex when NSE concentration was kept unchanged. However, the FITC-labeled antibody concentration higher than 50 nM had no benefits for the increase in the CL intensity of immunocomplex. In order to ensure the FITC-labeled antibody and HRP-labeled antigen to have an enough immunocapacity with NSE for competitive immunoreactions, the concentration of HRP-labeled antigen should be in excess. Considering the detection sensitivity and reagent consumption as a whole, 50 nM FITC-labeled antibody and 100 nM HRP-labeled antigen were chosen for the immunoreactions. The dependence of immunoreaction time on the efficiency of the production of the immunocomplex was studied. The HRP–NSE and FITC–anti-NSE antibody were mixed and incubated for 20–45 min at 37°C, and then were injected into the MCE channel. As shown in Fig. 2, the CL signal increased
Scheme 1. Schematic illustration of CRET-based competitive immunoassay for NSE.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
4
T. Yang et al.
Figure 2. The effect of the incubation time on the CL intensity of the immunocomplex of FITC–anti-NSE and HRP–NSE. A mixture of 0.05 M FITC–anti-NSE and 0.1 M HRP–NSE was incubated at 37°C, and then analyzed at different time intervals from 20 to 45 min. Electrophoresis buffer was 30 mM borate solution (pH 9.0) containing 30 mM SDS and 0.1 mM luminol. The oxidizer solution was 35 mM NaHCO3 buffer (pH 8.8) containing 0.1 M H2 O2 and 0.5 mM PIP.
rapidly with increasing the incubation time up to 35 min, and then remained constant. The results indicated that the binding of HRP–NSE to FITC–anti-NSE reached equilibrium for 35 min.
3.3 Optimization of conditions for CL reaction To improve the sensitivity of CL detection following the CE separation, the experimental conditions such as pH of oxidizer solution, concentrations of H2 O2 , and PIP in postcolumn oxidizer solution, luminol, and SDS in the electrophoresis buffer solution were optimized. Since the volume of the solution in the separation channel of MCE is very small compared with the volume of the postcolumn oxidizer solution, the acidity of CL reaction is mainly dependent on the oxidizer solution. The effect of pH of oxidizer solution (35 mM NaHCO3 ) on the CL intensity was investigated by varying pH from 8.0 to 9.5. It was found that the maximum CL was obtained at pH 8.8. H2 O2 as an oxidizer of luminol plays an important role in the CL reaction. With the increase of H2 O2 concentration from 0.06 to 0.10 M, the CL intensity also increased, and further increasing H2 O2 concentration result in the decrease of CL intensity. Therefore, a 0.1 M H2 O2 solution was used in further work. The effect of luminol concentration was investigated. With the increase of lumilol concentration from 0.05 to 0.10 mM, the CL intensity increased. Further increasing the luminol concentration resulted in the decrease of CL
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Electrophoresis 2014, 00, 1–7
Figure 3. Effect of PIP concentration on CL intensity produced in enhanced CL reaction. Thirty micromolar borate solution (pH 9.0) containing 30 mM SDS and 0.1 mM luminol, and 35 mM NaHCO3 (pH 8.8) containing 0.1 M H2 O2 and 0.1–0.6 mM PIP were used as electrophoresis buffer and the oxidizer solution, respectively.
intensity. Thus, the optimal luminol concentration was 0.1 mM. It is well known that HRP-catalyzed luminol-H2 O2 CL reaction may be significantly enhanced at the introduction of enhancers to luminol solution. The most popular enhancer is PIP [35]. Therefore, the effect of PIP concentration in the range of 0.1–0.6 mM on the CL intensity was also investigated, and the results were shown in Fig. 3. As can be seen, the optimum PIP concentration was 0.5 mM. Under favorable conditions, PIP increased CL intensity approximately seven times.
3.4 Optimization of the electrophoresis conditions The conditions for CE separation including pH values of the electrophoretic buffer and SDS concentration were optimized, and the separations were characterized by resolution (Rs) values (Rs = 2(T2 − T1 )/(W2 + W1 ), where T1 and T2 are the migration time of peak-1 and peak-2, W1 and W2 are the bottom width of peak-1 and peak-2). The obtained results indicated that with the increase of pH value from 8.5 to 9.0, the Rs of the immunocomplex and HRP–NSE increased. Further increasing the pH resulted in the decrease of Rs value (Fig. 4). In addition, the concentration of SDS in electrophoretic buffer also affected the separation efficiency. When the concentration of SDS was 30 mM, a maximal R value for immunocomplex and HRP–NSE was obtained. Upon MCE performance, the immunocomplex and HRP–NSE are separated, and electropherogram is presented in Fig. 5A. After the introduction of free NSE to the reaction mixture, a ration of CL intensity (peaks height)
www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
Microfluidics and Miniaturization
5
Figure 4. The effect of pH value of the electrophoretic buffer on resolution of the immunocomplex of FITC–anti-NSE with HRP– NSE and HRP–NSE. Electrophoretic buffer was 30 mM borate buffer (pH 9.0) containing 0.1 mM luminol and 30 mM SDS at different pH value. The oxidizer solution was 35 mM NaHCO3 buffer containing 0.1 M H2 O2 (pH 8.8) and 0.5 mM PIP.
corresponding to the concentrations of the immunocomplex and HRP–NSE is changed (Fig. 5B). These results demonstrated the competitive binding of NSE to the specific antibody. It is worth noting that the immunocomplex was well separated from free HRP–NSE making the proposed assay selective and useful for clinical analysis.
3.5 Analytical figures of merit The method was evaluated in terms of response linearity, LOD, and reproducibility. Under the selected MCE–CL conditions, seven standard NSE solutions at various concentrations were analyzed. Figure 6 showed the plot of relative CL intensity versus the concentration of NSE. Linear regression analysis of data yielded the following equation: ⌬ I = 7.35 [NSE] + 1.25, where ⌬ I is the relative CL intensity (V), and [NSE] is the concentration of NSE (pM). The calibration curve showed excellent linearity in the range from 9 to 950 pM with a correlation coefficient of 0.9964. Based on a S/N of 3, the detection limit for NSE was estimated to be 4.8 pM. Assay reproducibility was estimated by analyzing a standard solution of NSE of 450 pM (n = 9). The results showed that the RSDs of the migration time and peak height of the peak 1 were 2.8 and 3.5%, whereas for peak 2 the migration time and peak height were 2.6 and 3.9%, respectively.
3.6 Quantification of NSE in human serum Human serum samples taken from four healthy volunteers and three lung cancer patients were analyzed to demonstrate the feasibility of proposed MCE–CL method for the determination of NSE. The electropherograms obtained upon the
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 5. Electropherograms from separating the noncompetitive and competitive immunoreaction solutions. (A) without NSE; (B) with 0.45 nM NSE. The concentration of FITC–anti-NSE and HRP–NSE are 0.05 and 0.1 M, respectively. MCE conditions: electrophoretic buffer is 30 mM phosphate buffer (pH 9.0) containing 30 mM SDS and 0.1 mM luminol; the oxidizer solution was 35 mM NaHCO3 buffer containing 0.1 M H2 O2 (pH 8.8) and 0.5 mM PIP. Peak identification: (1) HRP–NSE–FITC–anti-NSE immunocomplex; (2) HRP–NSE.
analysis of the studied samples are shown in Fig. 7. The results are summarized in Table 1. The NSE level in the serum samples from healthy subjects was found be in the range 6.6 × 10−11 –10.1 × 10−11 M. The NSE level in the samples of lung cancer patients was found to be higher and lay in the range 96.2 × 10−11 –130.4 × 10−11 M. RSDs for the determination of NSE were in the ranges of 2.7–5.1%. Recovery of NSE from these samples was studied. NSE was spiked to each sample and analyzed. Recoveries were found to be in the range of 96.7–102.6%. The results suggested a potential application of the proposed assay in express primary diagnosis of diseases such as lung cancer patients.
www.electrophoresis-journal.com
6
T. Yang et al.
Electrophoresis 2014, 00, 1–7 Table 1. Assay results and recovery of NSE in human serum samples
Sample
RSD Found (10−11 M) (%, n = 3)
Normal 1 6.6 Normal 2 8.3 Normal 3 10.1 Normal 4 7.9 Patient 1 96.2 Patient 2 130.4 Patient 3 117.3
2.7 3.5 3.1 3.3 4.1 5.1 4.7
Added Total found Recovery (10−11 M) (10−11 M) (%) 10 10 10 10 100 100 100
16.32 18.16 20.36 17.81 192.9 231.6 214.4
97.2 98.6 102.6 99.1 96.7 101.2 97.1
4 Concluding remarks Figure 6. The plot of relative CL intensity versus the concentration of NSE. MCE conditions were as in Fig. 5.
A novel microfluidic competitive enzyme immunoassay based on CRET has been developed for the detection of NSE. The CRET system consisted of HRP/luminol as a light donor and FITC as an acceptor. When FITC-labeled antibody binds to the conjugate of the antigen (NSE) and HRP, the donor and acceptor are brought close each other and CRET occurs. In the MCE, the immunocomplex and HRP–NSE conjugate were separated. By using the CRET detection, the method allows the detection of 4.8 × 10−12 M for NSE, which is two orders magnitude lower than that of without CRET detection. To the best of our knowledge, the proposed CRET-based immunoassay for NSE is one of the most sensitive immunoassay in a microfluidic format. Although microfluidic competitive enzyme immunoassay based on CRET is demonstrated only for NSE in this work, assays for other compounds can be developed similarly by using corresponding antigens and respective antibodies. This work was supported by the National Natural Science Foundations of China (NSFC, Grant nos. 21175030, 21311120056), and the Russian Foundation for Basic Research (13-04-91164-GFEN_a). The authors have declared no conflict of interest.
5 References [1] Barton, A. C., Davis, F., Higso, S. P. J., Anal. Chem. 2008, 80, 9411–9416. [2] Martens, P., Raabe, A., Johnsson, P., Stroke 1998, 29, 2363–2366. [3] Baudin, E., Gigliotti, A., Ducreux, M., Ropers, J., Comoy, E., Sabourin, J. C., Bidart, J. M., Cailleux, A. F., Bonacci, R, Ruffie P., Schlumberger, M., Br. J. Cancer 1998, 78, 1102–1107. Figure 7. Electropherograms of serum samples. (A) normal serum sample; (B) patient serum sample (diluted ten times). The conditions were as Fig. 6. Peak identification: (1) HRP–anti-NSE– FITC–NSE immunocomplex; (2) HRP–anti-NSE. MCE conditions were as in Fig. 5.
[4] Marangos, P. J., Schmechel, D. E., Ann. Rev. Neurosci.1987, 10, 269–295. [5] Barton, A. C., Seamus, F. D., Higson, P. J., Anal. Chem. 2008, 80, 9411–9416. [6] Zhong, Z. Y., Peng, N., Qing, Y., Shan, J. L., Li, M. X., Guan, W., Dai, N., Gu, X. Q., Wang, D., Electrochimica Acta 2011, 56, 5624–5629.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
Electrophoresis 2014, 00, 1–7
Microfluidics and Miniaturization
7
[7] Han, J., Zhuo, Y., Chai, Y. Q., Yuan, Y. L., Yuan, R., Biosens. Bioelectron. 2012, 31, 399–405.
[21] Zhao, L. X., Sun L., Chu, X. G., Trends Anal. Chem. 2009, 28, 404–414.
[8] Inga, A., Monika, B., Markus, O., Sigrid, P., Otto, W., Hans, A. K., Olaf, G., Thomas, W., Lancet 1996, 348, 846–849.
[22] Marquette, C. A., Blum, L. J., Bioanalysis 2009, 1, 1259–1269.
[9] Cao, Z. J., Li, H., Laua C., Zhang, Y. H., Anal. Chim. Acta 2011, 698, 44–50. [10] Harrison, D. J., Fluri, K., Seiler, K., Fan, Z., Effenhauser, C. S., Manz, A., Science 1993, 261, 895–897. [11] Reyes, D. R., Iossifidis, D., Manz, A., Anal. Chem. 2002, 74, 26223–26236. [12] Morozov, V. N., Groves, S., Turell, M. J., Bailey, C., J. Am. Chem. Soc. 2007, 129, 12628–12629. [13] Bange, A., Halsall, H. B., Heineman, W. R., Biosens. Bioelectron. 2005, 20, 2488–2503. [14] Koutny, L. B., Schmalzing, D., Taylor, T. A., Fuchs, M., Anal. Chem. 1996, 68, 18–22. [15] Chiem, N., Harrison, D. J., Anal. Chem. 1997, 69, 373–378. [16] Wang, J., Ibanez, A., Chatrathi, M. P., J. Am. Chem. Soc. 2003, 125, 8444–8445. [17] Dishinger, J. F., Kennedy, R. T., Anal. Chem. 2007, 79, 947–954. [18] Roda, A., Guardigli, M., Anal. Bioanal. Chem. 2012, 402, 69–76. [19] Vdovenko, M. M., Demiyanova, A. S., Chemleva, T. A., Sakharov, I. Yu., Talanta 2012, 94, 223–226. [20] Sakharov, I. Yu., Demiyanova, A. S., Gribas, A. V., Uskova, N. A., Efremov, E. E., Vdovenko, M. M., Talanta 2013, 115, 414–417.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[23] Sapsford, K. E. Berti, L., Medintz, I. L., Angew. Chem. Int. Ed. 2006, 45, 4562–4588. [24] Huang, X. Y., Li, L., Qian, H. F., Dong, C. Q., Ren, J., Angew. Chem. Int. Ed. 2006, 45, 5140–5143. [25] Zhang, S., Yan, Y., Bi, S., Anal. Chem. 2009, 81, 8695–8701. [26] Freeman, R., Liu, X., Willner, I., J. Am. Chem. Soc. 2011, 133, 11597–11604. [27] Liu, X., Freeman, R., Golub, E., Willner, I., ACS Nano 2011, 5, 7648–7655. [28] Lee, J. S., Joung, H. A., Kim, M. G., Park, C. B., ACS Nano, 2012, 6, 2978–2983. [29] Zhao, S., Li, X., Liu, Y. M., Anal. Chem. 2009, 81, 3873–3878. ´ A. M., Aaron, J. J., Anal. [30] Lara, F. J., Garc´ıa-Campana, Chim. Acta 2010, 679, 17–30. [31] Thorpe, G. H. G., Kricka, L. J., Moseley, S. R., Whitehead, T. P., Clin. Chem. 1985, 31, 1335–1341. [32] Whitehead, T. P., Thorpe, G. H. G., Carter, T.J.N., Groucutt, C., Kricka, L. J., Nature 1983, 305, 159–160. [33] Shi, M. Zhao, S., Huang, Y., Liu, Y. M., Ye, F., J. Chromatogr. B 2011, 879, 2840–2846. [34] Qin, G., Zhao, S., Huang, Y., Jiang, J., Ye, F., Anal. Chem. 2012, 84, 2708–2712. [35] Wang, J., Ren, J., Electrophoresis 2005, 26, 2402–2408.
www.electrophoresis-journal.com
