TECHNICAL NOTE pubs.acs.org/ac
Sensitive Chemiluminescence Immunoassay by Capillary Electrophoresis with Gold Nanoparticles Yan-ming Liu,*,† Lin Mei,† Li-juan Liu,† Long-fei Peng,† Yong-hong Chen,‡ and Shu-wei Ren‡ † ‡
College of Chemistry and Chemical Engineering, Xinyang Normal University, Xinyang 464000, China, and Xinyang Central Hospital, Xinyang 464000, China ABSTRACT: This technical note describes a new chemiluminescence immunoassay hyphenated to capillary electrophoresis (CE-based CL-IA) with gold nanoparticles (AuNPs) technique for biological molecules determination. AuNPs were used as a protein label reagent in the light of its excellent catalytic effect to the CL reaction of luminol and hydrogen peroxide. AuNPs conjugate with antibody (Ab) to form tagged antibody (Ab*), and then Ab* link to antigen (Ag) to produce an Ab*-Ag complex by a noncompetitive immunoreaction. The mixture of the excess Ab* and the Ab*-Ag complex was baseline separated and detected within 5 min under the optimized conditions. This new protocol was evaluated with human immunoglobulin G (IgG) as the target molecule. The calibration curve of IgG was in the range of 0.008-5 μg/mL with a correlation coefficient of 0.995. The detection limit (S/N = 3) of IgG was 1.14 10-3 μg/mL (7.1 pmol/L, 0.39 amol). The proposed AuNPs enhanced CEbased CL-IA method was successfully applied for the quantification of IgG in human sera from patients. It proves that the present method could be developed into a new and sensitive biochemical analysis technique.
C
hemiluminescence immunoassay (CL-IA) has been well established for the quantitation of low concentration analytes in complex biosamples and has been widely used in the research of clinical diagnosis.1,2 CL-IA quantitation is generally realized by indirectly measuring the intensity of the CL label since most of the biomolecules (such as tumor marker) have no native CL emission. Various CL labels have been investigated and applied in CL-IA to produce and amplify signal readout.3,4 To date, the most widely used labels in CL-IA are enzymes5,6 and luminol (including isoluminol and its derivatives).7 Enzymes are frequently used labels for antigen (Ag) or antibody (Ab) molecules to visualize the binding event.8 However, enzyme labels require suitable substrates. The synthesis of the labels and the labeling procedure are time-consuming and labor intensive. The labeling also requires demanding reaction conditions, such as low temperature and a light-tight environment. Luminol is one of the most efficient CL reagents and is coupled to ligands via reactions with the amino group. However, the resulting conjugates have much lower CL efficiency than that of parent compounds and the biomolecules activity could decrease more or less. Nanotechnology, one of the critical technologies of the 21st century, is multidisciplinary and interdisciplinary and covers diverse fields including chemistry, physics, materials science, engineering, biology, and even medicine. In past years, an emerging approach for CL-IA is the employment of metal nanoparticles labels.9-11 The use of metal nanoparticles as CL labels in bioassays has exponentially increased due to their many attractive features, e.g., less fastidious working conditions, more integrated functions, and higher stability than the enzyme and luminol labels.12-15 r 2011 American Chemical Society
Over the past decade, functionalized gold nanoparticles (AuNPs) have attracted great interest in biomolecular detection and clinical diagnostic applications because of their facile synthesis and surface modification and strong catalytic properties, making functionalized gold nanoparticles an excellent candidate for bioconjugation.16-18 It has been found that AuNPs can greatly enhance the CL intensity of the luminol/H2O2 system due to the catalysis of AuNPs on the radical generation and electron-transfer processes.19 The AuNPs have been used in CL-IA as a label reagent. Fan et al.20 and Li et al.21 have developed CL-IA using AuNPs labels based on the conversion of AuNPs to Au3þ by dissolution after immunoreactions. The resulting Au3þ reacted with luminol and H2O2 to produce enhanced CL signals. However, the dissolution of AuNPs needed extremely severe conditions (highly concentrated HNO3-HCl or poisonous HBr-Br2), which led to a high CL background and denaturation of the proteins. Duan et al.22 established a microplate compatible CLIA protocol for human IgG based on the CL reaction between luminol and AgNO3 with catalysis by the AuNPs. Although this protocol avoided the undesired procedure of dissolving AuNPs, the CL detection was influenced easily by the sediment from the mixture of basic luminol and AgNO3. More recently, Yang et al.23 reported the sandwich CL-IA based on the luminol/Ab labeled AuNPs conjugates and magnetic beads for the detection of Received: September 13, 2010 Accepted: December 30, 2010 Published: January 10, 2011 1137
dx.doi.org/10.1021/ac103166n | Anal. Chem. 2011, 83, 1137–1143
Analytical Chemistry carcinoembryonic antigen in serum. However, the requirements of complicated labeling and washing procedure were painful. Sensitive measurement of bioactive molecules, such as some important proteins or tumor markers, is a critical need in many aspects of modern biochemical and clinical analysis research, especially in early cancer detection, while maintaining minimum sample consumption and operational simplicity.24,25 Capillary electrophoresis (CE) is an established microseparation technique, which provides advantages in terms of high separation efficiency, short analysis time, low cost, and simplicity.26,27 CEbased CL-IA that combines the high separation efficiency of CE, the high sensitivity of CL, and the high specificity of IA has been proven to be a promising technique for complex biological compound assays and diagnostic clinical analysis.28-30 To the best of our knowledge hitherto, the CE-based CL-IA with AuNPs for measuring biomolecules has not been reported. In this work, a novel and sensitive CE-based CL-IA method has been developed using AuNPs as a protein label reagent for the first time. On the basis of the noncompetitive immunoreactions, the formed Ab*-Ag complex and unbound Ab* can be efficiently separated by CE and sensitively detected by CL. The conditions for separation and detection were investigated. The applicability of the proposed technique was illustrated in the quantitative analysis of IgG in human sera samples.
’ EXPERIMENTAL SECTION Chemicals. Human IgG and goat-antihuman IgG were obtained from Solarbio Science & Technology Co. Ltd. (Beijing, China). Hydrogen tetrachloroaurate(III) trihydrate (HAuCl4 3 3H2O, Au 49.5% min) and polyvinylpyrolidone (PVP, Mr = 1 300 000) were purchased from Alfa Aesar (A Johnson Matthey Company, Ward Hill, MA). Sodium dodecyl sulfate (SDS) and bovine serum albumin (BSA) were from Siobio Biotechnology Inc. (Shanghai, China). Luminol was purchased from Yacoo Chemical Factory (Suzhou, China). All reagents used were of analytical grade. The pure water (18.2 MΩ cm) used was processed with an Ultrapure Water System (Kangning Water Treatment Solution Provider, China). All solutions were stored in the refrigerator at 4 °C and filtered through 0.22 μm cellulose acetate membrane filters (Shanghai Xingya Purification Material Factory, China) before use. Apparatus. The basic design of the CE-CL system has been previously described.31 Briefly, a high-voltage supply (0-30 kV, Shanghai Institute of Applied Physics, Chinese Academy of Sciences, China) was used to drive the electrophoresis. A 50 cm 75 μm i.d. uncoated fused silica capillary (Hebei Yongnian Ruifeng Chromatographic Apparatus, China) was used for the separation. On one end of the separation capillary, 5 cm of the polyimide coating was burned off. After etching with 40% HF for 2 h, this end of the capillary was inserted into a 20 cm 530 μm i. d. reaction capillary. The outlet of the separation capillary was located at the detection window, which was made by burning 1 cm of the polyimide of the reaction capillary and placed in front of the photomultiplier tube (PMT, CR 120, Binsong Photonics, Beijing, China). The pair of 40 cm 320 μm i.d. capillaries were used to introduce postcolumn CL reagents to a tee reservoir from two vials located 20 cm above it. The CL reagents were delivered by gravity through the reagent-introducing capillaries, mixed, and flowed coaxially along the separation capillary. The CL reaction took place at the separation capillary outlet when analyte eluted out from the separation capillary. The ground electrode was also put into one joint of the tee. CL emission was collected with a
TECHNICAL NOTE
Figure 1. The TEM images of bare AuNPs (A) and Ab-conjugated AuNPs (B).
BPCL ultraweak luminescence analyzer (Institute of Biophysics, Chinese Academy of Science, Beijing, China) and then recorded and processed with a computer using BPCL software. Transmission electron microscopy (TEM) images were obtained using a JEM-2100F transmission electron microscope (JEOL Ltd., Tokyo, Japan) operating at 200 kV. The samples for the TEM measurements were prepared by placing a drop of AuNPs colloid on a carbon-coated copper grid. Absorption spectra were recorded by a UVmini-1240 UV-vis spectrophotometer (Shimadzu Corporation, Kyoto, Japan) using a quartz cell with the path length of 1.0 cm. The CL spectra were measured by a Cary Eclipse fluorescence spectrophotometer (Varian Inc., Walnut Creek) with the excitation light source being turned off. Preparation of Gold Nanoparticles. AuNPs were synthesized by sodium citrate reduction of HAuCl4 in water.32 Briefly, 0.5 mL of 2% sodium citrate was rapidly added to 50 mL of boiling 0.01% HAuCl4 solution under vigorous stirring, and then the solution changed color from pale-yellow to wine-red. Boiling continued for an additional 10-15 min, and then the heating source was removed. The colloids were stirred for another 15 min and cooled to room temperature. After cooling, the synthesized AuNPs was filtered through a 0.22 μm cellulose membrane and stored at 4 °C. TEM was used to characterize the prepared AuNPs (Figure 1A). TEM images showed that the average diameter of AuNPs used for labeling was about 36 nm. The concentration of the AuNPs was calculated to be about 8.5 10-10 mol/L. Preparation of AuNPs-Antibody Conjugate. The AuNPslabeled goat-antihuman IgG was prepared by following the reported procedure with some modifications.33 The pH of the colloidal AuNPs was adjusted to 8.5 with 0.1 mol/L K2CO3. Typically, 10 μL of 440 μg/mL goat-antihuman IgG (10% more than the minimum amount determined by a flocculation assay) was added to 0.5 mL of AuNPs under agitation, followed by incubation at room temperature for 30 min. Afterward, 5% BSA was added to a final concentration of 1% with stirring for 5 min. Then the conjugate was centrifuged at 12 000 rpm for 30 min. The red sediment was resuspended in 0.5 mL of 0.04 mol/L pH 7.4 PBS (containing 1% BSA and 0.05% Tween 20) three times with the supernatant being removed each time. The obtained AuNPs labeled goat-antihuman IgG can be used directly or stored at 4 °C for months. Preparation of Immune Sample. The IA protocol was of noncompetitive format, and the immunoreaction was conducted as follows: Ab þ AuNPs f Ab ð1Þ Ag þ AbðexcessÞ f Ab þ Ab- Ag 1138
ð2Þ
dx.doi.org/10.1021/ac103166n |Anal. Chem. 2011, 83, 1137–1143
Analytical Chemistry
TECHNICAL NOTE
Figure 2. UV-vis absorption spectrum (A) and CL spectrum (B). Curves: (a) bare AuNPs; (b) Ab-conjugated AuNPs; (c) AuNPs dilution with pure water (with the same volume as Ab solution); (d) luminol þ H2O2; (e) bare AuNPs þ luminol þ H2O2; (f) Ab-conjugated AuNPs þ luminol þ H2O2; (g) AuNPs dilution with pure water (with the same volume as Ab solution) þ luminol þ H2O2.
where Ag is the human IgG with a limited amount and Ab* is the AuNPs labeled goat-antihuman IgG with an excessive fixed amount in eq 2. A volume of 10 μL of human IgG standard or serum sample was mixed with 10 μL of AuNPs labeled goat-antihuman IgG in a 200 μL microcentrifuge tube and then diluted with 0.01 mol/L pH 7.4 PBS to 50 μL. After incubation at 37 °C for 1 h, the mixture was diluted 10-fold by 0.01 mol/L pH 7.4 tris-HCl (containing 0.2% SDS) and analyzed by CE as described below. CE Procedure. The new capillary was preconditioned sequentially by flushing with 2.0 mol/L NaOH-CH3OH, 1.0 mol/L NaOH, water, and electrophoretic buffer for 20 min before the first use. At the beginning of each day, the separation capillary was reconditioned by flushing with 0.1 mol/L NaOH, water, and electrophoretic buffer for 3-5 min successively. After three consecutive injections, the separation capillary was flushed with 0.1 mol/L NaOH, water, and electrophoretic buffer for 2 min, respectively. The electrophoretic buffer was 10 mmol/L Na2B4O7 with 0.20 mmol/L luminol and 1.0% PVP at pH 11.0. The postcolumn CL reagents, one containing 40 mmol/L H2O2 and another containing 20 mmol/L NaHCO3, were, respectively, introduced from two vials. The voltage of PMT for collecting the CL signal was set at -800 V. The sample was introduced by electrokinetic injection at 10 kV for 5 s. CE separation was performed at 20 kV. The peak areas of the Ab*-Ag complex were used for quantification in this work. Preparation of Human Serum Samples. Human serum samples from healthy people were provided by Xinyang Normal University (Xinyang, China), and samples from patients were provided by Xinyang Central Hospital (Xinyang, China). The fresh blood samples were immediately centrifuged for 10 min at 1200 rpm to remove erythrocytes. Subsequently, the supernatant was diluted 40 000-fold by using 0.01 mol/L pH 7.4 PBS as the diluent buffer and analyzed.
’ RESULTS AND DISCUSSION Characterization of the Ab-Conjugated AuNPs. TEM was employed to characterize the bare AuNPs and Ab-conjugated AuNPs (Figure 1). There are differences between the TEM images of AuNPs and Ab-conjugated AuNPs. As seen in Figure 1B, Abconjugated AuNPs were surrounded by a transparent ring, while the rings were absent in bare AuNPs images (Figure 1A),
indicating that Ab has been tagged onto the surface of AuNPs. TEM images revealed that the average diameters of the Abconjugated AuNPs was about 40 nm. UV-vis absorption spectrum was also performed to characterize the formation of Ab-conjugated AuNPs as shown in Figure 2A. The absorbance of the bare AuNPs decreased with the formation of Ab-conjugated AuNPs. Additional evidence of the Ab linkage on the AuNPs surface can be provided by a CL spectrum (Figure 2B). The CL intensity of the bare AuNPs was found to decrease with the formation of Ab-conjugated AuNPs. Mechanism Discussion. It is well-known that the reaction of luminol with H2O2 in alkaline solution in the absence of a catalyst produces weak CL. Cui’s group found that the presence of AuNPs can enhance the CL of the luminol-H2O2 system and the catalytic mechanism was discussed in detail.19 The study of Hayashi et al.34 and Kricka et al.35 revealed that the electron-transfer processes to produce enhancement of the CL in catalytic reactions take place on the surface of AuNPs. The maximum CL intensity of luminolH2O2 and some analyte-sensitized luminol-H2O2 maxima have been reported at 425 nm in the literature.36 In this work, series experiments were carried out to verify whether the 3-aminophthalate ion was the CL emitter for the catalytic effect of the bare AuNPs and Ab-conjugated AuNPs on the luminolH2O2 CL reaction. The CL spectra of luminol-H2O2 (curve d), luminol-H2O2-bare AuNPs (curve e), luminol-H2O2-Abconjugated AuNPs (curve f), and luminol-H2O2-bare AuNPs dilution with pure water (curve g) were shown in Figure 2B, respectively. The results showed that the maximum CL intensity appeared at 425 nm for the above four mixtures, and a much lower CL peak was observed in the presence of Ab-conjugated AuNPs. These results indicated that the luminophor of the Ab-AuNPs conjugate enhancing luminol-H2O2 reaction is still 3-aminophthalate ions, the oxidation product of luminol. Determination of the Minimum Amount of Ab for AuNPs Labeling. Generally, AuNPs exist as a monodispersed colloid phase at low ionic strength due to the charge repulsion effect and tend to aggregate due to charge neutralization at high ionic strength.37 The conjugation of macromolecules (such as proteins) with AuNPs not only affords stabilization of the system but also introduces biocompatible functionalities onto AuNPs for further biological interactions. Functionalized AuNPs were conjugated with protein molecules through electrostatic attraction between the surface-terminated negatively charged carboxylate groups of 1139
dx.doi.org/10.1021/ac103166n |Anal. Chem. 2011, 83, 1137–1143
Analytical Chemistry citrate and the positively charged amino groups of the protein.38 The stability of the resulting AuNPs in biological media is highly dependent upon the protein concentration. The minimum amount of protein for AuNPs labeling was optimized by employing a flocculation assay as described in the literature.39 In our experiment, 10 μL of goat-antihuman IgG solution with concentrations ranging from 150 to 500 μg/mL were added to 500 μL of AuNPs under agitation and followed by incubation at room temperature for 30 min. Then 50 μL of 10% NaCl was added to the mixture. In the undersaturation condition, not all AuNPs interact specifically with Ab. After addition of electrolytes, AuNPs become unstable and they flocculate rapidly. This can be reflected from the color change of the colloid (red to gray). At the critical concentration (covering the Au surface with roughly a full monolayer of Ab), the colloid was pink.39 In the saturation condition, no color change was observed. The color changes upon the addition of 50 μL of 10% NaCl in this flocculation assay were explored. Results showed that the colloids were gray, pink, and red when the concentration of goat-antihuman IgG were lower than 300, 350, and higher than 400 μg/mL, respectively. So the minimum Ab amount for AuNPs labeling was 400 μg/mL in this work. Further, the concentration of the goat-antihuman IgG used for labeling was 440 μg/mL, which was 10% more than the minimum amount in order to ensure that the Ab amount was enough for the stability of the AuNPs-Ab conjugate. Choice of Immune Sample Solvent. The protein adsorption on capillary inner wall occurrs especially for high sample concentration. It often leads to tailing, peak broadening, and thus reduced resolution and reproducibility. So it is necessary to dilute the immune sample and minimize protein adsorption. First, the immune sample was diluted 10-fold by pure water. However, no sample peaks appeared (Figure 3A). Subsequently 0.01 mol/ L pH 7.4 PBS was chosen to dilute the sample also 10-fold and the result as shown in Figure 3B. The peaks of free Ab* and the Ab*-Ag immunocomplex were clearly seen but not well separated. The highest CL intensity and the best resolution were obtained (Figure 3C) when 0.01 mol/L pH 7.4 tris-HCl (containing 0.2% SDS) was selected as the sample solvent. Tris-HCl was widely used as a buffer for Ag and/or Ab dilution in a commercial EIA Kit. SDS plays an important role in minimizing protein adsorption in CE. In previous papers, CE-based CL-IA was conducted most favorably at high pH.28,31 In this study, the best experimental result was obtained at pH 11.0. Upon high voltage, protein molecules having negative charges migrated against the electroosmotic flow (EOF) and entered the SDS zone. This process caused the protein molecules to migrate slowly and stack at the boundary between the SDS and sample zones.40 As a result, detection sensitivity and separation efficiency were improved. Orthogonal Design for the Electrophoretic Buffer. In order to gain desired experimental results, orthogonal experimental design (OAD) was utilized for the optimization of the electrophoretic buffer. In this study, the electrophoretic buffer include three factors: pH, concentration of Na2B4O7, and concentration of PVP. Since pH and concentration of Na2B4O7 greatly influence the ionization of the silanols on the capillary wall and ionic strength of the buffer, changing the buffer pH and concentration are the most direct strategy for optimizing separation. In order to decrease the adsorption of protein on the capillary inner wall, PVP was used as a capillary inner wall dynamic coating reagent for protein separation. So an OAD L9 (34) in triplicate was used to evaluate effects of aforesaid three factors. Each factor was evaluated in three levels. Experiments were carried out with
TECHNICAL NOTE
Figure 3. Effect of the different immune sample solvents. (A) pure water; (B) 0.01 mol/L pH 7.4 PBS; (C) 0.01 mol/L pH 7.4 tris-HCl (containing 0.2% SDS). Peak 1, immunecomplex; peak 2, free Ab*; peak X, unknown compounds. Conditions: electrophoretic buffer, 10 mmol/ L Na2B4O7 with 0.20 mmol/L luminol and 1.0% PVP at pH 11.0; the postcolumn CL reagents, 40 mmol/L H2O2 and 20 mmol/L pH 12.0 CBS; PMT voltage, -800 V; electrokinetic injection with 10 kV for 5 s; separation voltage, 20 kV; separation capillary, 50 cm 75 μm i.d.
Table 1. Assignment of the Factors and Levels of the Electrophoretic Buffer by Using a L9 (34) Matrix buffer concentration
PVP concentration
no.
buffer pH
(mmol/L)
(%)
1 2
10 11
5 5
0.6 0.8
3
12
5
1.0
4
10
10
0.8
5
11
10
1.0
6
12
10
0.6
7
10
15
1.0
8
11
15
0.6
9
12
15
0.8
the pH at 10, 11, or 12, buffer concentration at 5, 10, or 15 mmol/ L, and PVP concentration at 0.6, 0.8, or 1.0%, respectively. The assignment of the factors and the levels in this OAD were listed in Table 1. According to Table 1, the electrophoregrams of the Ab*-Ag complex and the free Ab* were shown in Figure 4. Obviously, the optimal electrophoretic buffer was composed of 10 mmol/L Na2B4O7 and 1.0% PVP at pH 11.0. Under this condition, the Ab*-Ag complex and free Ab* were baseline revolved within 5 min. Optimization of the CL Detection Conditions. Some important experimental parameters influencing the CL reaction were optimized, including the particle size of the AuNPs, the concentrations of luminol, H2O2, and NaHCO3 (carbonate buffer solution, CBS), and the pH of CBS. The size effects of AuNPs on the CL intensity of luminol-H2O2 were investigated in the range from 16 to 68 nm in diameter, and the most intensive CL signal was obtained with 36 nm AuNPs. Figure 5A shows the CL intensity against the luminol concentration. It can be seen that the optimal luminol concentration is 0.20 mmol/L. The effect of the H2O2 concentration on the CL intensity showed that the optimum H2O2 concentration is 40 mmol/L (Figure 5B). Also, investigations on the concentrations and pH of CBS indicated that the optimum concentration 1140
dx.doi.org/10.1021/ac103166n |Anal. Chem. 2011, 83, 1137–1143
Analytical Chemistry
TECHNICAL NOTE
Figure 4. Electropherograms for OAD for the electrophoretic buffer. Peak a, immunecomplex; peak b, free Ab*; peak X, unknown compounds. Conditions: electrophoretic buffer, 10 mmol/L Na2B4O7 with 0.20 mmol/L luminol and 1.0% PVP at pH 11.0; the postcolumn CL reagents, 40 mmol/L H2O2 and 20 mmol/L pH 12.0 CBS; PMT voltage, -800 V; electrokinetic injection with 10 kV for 5 s; separation voltage, 20 kV; separation capillary, 50 cm 75 μm i.d.
and pH are 20 mmol/L and 12.0, respectively, as shown in Figure 5C,D. Optimization of the Separation Voltage. The effect of the separation voltage was studied from 16 to 24 kV. As shown in Figure 6, the CL intensity increased and the resolution decreased with the increase in the separation voltage. When the separation voltage was 20 kV, the resolution was calculated to be about 1.6. Therefore, 20 kV was set. Detection of Human IgG by CE-Based CL-IA. The detection of human IgG by the proposed method was performed, and the electropherograms were shown in Figure 7. Figure 7A is the electropherogram of AuNPs labeled goat-antihuman IgG (Ab*). After the addition of the different concentration of human IgG and incubation, the peaks of the immunocomplex appeared
(shown in Figure 7B-D). It can be seen that the peak areas of the immunocomplex increased with the increase in concentration of human IgG and the peaks became relatively broad. We speculate that the broadening of the peaks may be caused by two factors. First, the concentration (or peak area) of the immunocomplex increased with the increase in concentration of human IgG accordingly. Second, it could be an indication of the heterogeneity in the size distribution of the Ab-conjugated AuNPs. Such size differences can cause a slightly different migration time, which results in the width of the CE peaks.41 Linearity, Detection Limit, and Precision. Under the optimized conditions, we examined the linearity, detection limit, and precision of the present method for human IgG analysis. For the linearity, a series of concentration (from 0.008 to 5 μg/mL) of 1141
dx.doi.org/10.1021/ac103166n |Anal. Chem. 2011, 83, 1137–1143
Analytical Chemistry
TECHNICAL NOTE
Figure 5. Effect of luminol (A), H2O2 (B), and CBS concentrations (C) and CBS pH (D) on CL intensity. Conditions: electrophoretic buffer, 10 mmol/L Na2B4O7 with 0.20 mmol/L luminol and 1.0% PVP at pH 11.0; the postcolumn CL reagents, 40 mmol/L H2O2 and 20 mmol/L pH 12.0 CBS; PMT voltage, -800 V; electrokinetic injection with 10 kV for 5 s; separation voltage, 20 kV; separation capillary, 50 cm 75 μm i.d.
Figure 6. Effect of separation voltage on CL intensity and resolution. Conditions: electrophoretic buffer, 10 mmol/L Na2B4O7 with 0.20 mmol/L luminol and 1.0% PVP at pH 11.0; the postcolumn CL reagents, 40 mmol/L H2O2 and 20 mmol/L pH 12.0 CBS; PMT voltage, -800 V; electrokinetic injection with 10 kV for 5 s; separation voltage, 20 kV; separation capillary, 50 cm 75 μm i.d.
human IgG standards were analyzed. The linearity was obtained by plotting the peak areas of Ab*-Ag complex versus the concentrations of the human IgG. The correlation coefficient acquired was 0.995 and the detection limit (S/N = 3) was 1.14 10-3 μg/mL (7.1 pmol/L, 0.39 amol). The precisions (measured by relative standard deviation, RSDs, n = 5) were studied by assaying 0.5 μg/ mL human IgG within a day (intraday) and in 4 days (interday). The RSDs of the migration time were 2.0% for intraday and 3.1% for interday, respectively. The RSDs of the peak area were 3.0% for intraday and 4.6% for interday, respectively. Applications in the Human Serum Analysis. For verification of this method, the contents of human IgG in sera of nine
Figure 7. Electropherograms of Ab* and the immunocomplex. The concentration of AuNPs labeled goat-antihuman IgG were constant. Parts A, B, C, and D correspond to 0, 0.1, 0.5, and 3.0 μg/mL human IgG, respectively; (E) blank. Peak 1, immunocomplex; peak 2, free Ab*; peak X, unknown compounds. Conditions: electrophoretic buffer, 10 mmol/L Na2B4O7 with 0.20 mmol/L luminol and 1.0% PVP at pH 11.0; the postcolumn CL reagents, 40 mmol/L H2O2 and 20 mmol/L pH 12.0 CBS; PMT voltage, -800 V; electrokinetic injection with 10 kV for 5 s; separation voltage, 20 kV; separation capillary, 50 cm 75 μm i.d.
patients from Xinyang Central Hospital were determined by the proposed method (X) and immunoturbidimetry (ITM, Y) employed in the Xinyang Central Hospital as control. The results of the nine samples determined by the two methods were listed in 1142
dx.doi.org/10.1021/ac103166n |Anal. Chem. 2011, 83, 1137–1143
Analytical Chemistry
TECHNICAL NOTE
Table 2. Analytical Results of Nine Human Sera Samples (n = 3) human IgG content (g/L) no. of human sera samples
this method (RSD, %)
ITM
1
3.04 (2.0)
3.48
2
3.28 (2.4)
3.90
3
5.30 (4.2)
4.35
4
10.56 (6.6)
9.45
5
11.49 (8.4)
10.43
6
10.38 (6.0)
10.98
7
11.21 (7.3)
12.07
8 9
17.09 (7.0) 42.56 (2.0)
17.84 41.10
Table 2. Two methods coincided well with a regression equation of Y = 0.9636X þ 0.3194 and a correlation coefficient of 0.997. In order to further examine the reliability of the proposed method, the recovery of human IgG from the spiked human serum was tested. Three different IgG spiked level (0.04, 1.0, and 4.0 μg/mL) were added to the human serum. Recoveries are in the range of 85.0-107.0%. These results proved that this method for the analysis of IgG in human serum was reliable and has potential application in the analysis of real samples.
’ CONCLUSIONS A novel CE-based CL-IA using AuNPs as a label technique has been developed and evaluated. Ultrasensitive detection of picomole/liter (attomoles) levels of IgG was achieved. This method has many outstanding features including high efficiency, high selectivity, and low cost. The simple sample preparation process can avoid the multiple procedures of immunoreactions and washing steps and subsequently reduce the errors and deviations caused inevitably by those steps. The excellent derivatization and catalysis ability of AuNPs coupled with the CE-based CL-IA approach could afford a new technique for the ultrasensitive determination of bioactive molecule (e.g., tumor markers) in early disease diagnosis.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected]. Fax: þ86-376-6390622.
’ ACKNOWLEDGMENT This work was supported by the National Natural Science Foundation of China (Grant 21075106), Henan Innovation Project for University Research Talents (Grant 2005126), and Natural Science Foundation of Henan Province of China (Grant 092300410122). ’ REFERENCES (1) Weeks, I. Chemiluminescence Immunoassay; Elsevier: Amsterdam, The Netherlands, 1992. (2) Zhao, L. X.; Sun, L.; Chu, X. G. TrAC, Trends Anal. Chem. 2009, 28, 404–415. (3) Zhuang, H. S.; Huang, J. L.; Chen, G. N. Anal. Chim. Acta 2004, 512, 347–353. (4) Qi, H. L.; Zhang, Y.; Peng, Y. G.; Zhang, C. X. Talanta 2008, 75, 684–690.
(5) Bi, S.; Zhou, H.; Zhang, S. S. Biosens. Bioelectron. 2009, 24, 2961– 2966. (6) Wang, X.; Zhang, Q. Y.; Li, Z. J.; Ying, X. T.; Lin, J. M. Clin. Chim. Acta 2008, 393, 90–94. (7) Aoyagi, S.; Iwata, T.; Miyasaka, T.; Sakai, K. Anal. Chim. Acta 2001, 436, 103–108. (8) Li, X. M.; Yang, X. Y.; Zhang, S. S. TrAC, Trends Anal. Chem. 2008, 27, 543–553. (9) Liu, C. H.; Li, Z. P.; Du, B. A.; Duan, X. R.; Wang, Y. C. Anal. Chem. 2006, 78, 3738–3744. (10) Wang, Z. P.; Hu, J. Q.; Jin, Y.; Yao, X.; Li, J. H. Clin. Chem. 2006, 52, 1958–1961. (11) Poznyak, S. K.; Talapin, D. V.; Shevchenko, E. V.; Weller, H. Nano Lett. 2004, 4, 693–698. (12) Kim, J. Y.; Piao, Y. Z.; Hyeon, T. Chem. Soc. Rev. 2009, 38, 372– 390. (13) De Dios, A. S.; Díaz-García, M. E. Anal. Chim. Acta 2010, 666, 1–22. (14) Zhang, S. B.; Wu, Z. S.; Guo, M. M.; Shen, G. L.; Yu, R. Q. Talanta 2007, 71, 1530–1535. (15) Sanchez-Martínez, M. L.; Aguilar-Caballos, M. P.; G omezHens, A. Anal. Chim. Acta 2009, 636, 58–62. (16) Huang, X. H.; El-Sayed, M. A. J. Adv. Res. 2010, 1, 13–28. (17) Daniel, M. C.; Astruc, D. Chem. Rev. 2004, 104, 293–346. (18) Misra, T. K.; Huang, K. P.; Liu, C. Y. J. Colloid Interface Sci. 2010, 344, 137–143. (19) Zhang, Z. F.; Cui, H.; Lai, C. Z.; Liu, L. J. Anal. Chem. 2005, 77, 3324–3329. (20) Fan, A. P.; Lau, C. W.; Lu, J. Z. Anal. Chem. 2005, 77, 3238– 3242. (21) Li, Z. P.; Wang, Y. C.; Liu, C. H.; Li, Y. K. Anal. Chim. Acta 2005, 551, 85–91. (22) Duan, C. F.; Yu, Y. Q.; Cui, H. Analyst 2008, 133, 1250–1255. (23) Yang, X. Y.; Guo, Y. S.; Wang, A. G. Anal. Chim. Acta 2010, 666, 91–96. (24) Weston, A. D.; Hood, L. J. Proteome Res. 2004, 3, 179–196. (25) Nock, S.; Wilson, D. S. Angew. Chem., Int. Ed. 2003, 42, 494– 500. (26) García-Campa~ na, A. M.; Lara, F. J.; Gamiz-Gracia, L.; HuertasPerez, J. F. TrAC, Trends Anal. Chem. 2009, 28, 973–986. (27) Mastichiadis, C.; Niotis, A. E.; Petrou, P. S.; Kakabakos, S. E.; Misiakos, K. TrAC, Trends Anal. Chem. 2008, 27, 771–784. (28) Liu, Y. M.; Zheng, Y. L.; Cao, J. T.; Chen, Y. H.; Li, F. R. J. Sep. Sci. 2008, 31, 1151–1155. (29) Wang, J. H.; Huang, W. H.; Liu, Y. M.; Cheng, J. K.; Yang, J. Anal. Chem. 2004, 76, 5393–5398. (30) Wang, J. N.; Ren, J. C. Electrophoresis 2005, 26, 2402–2408. (31) Liu, Y. M.; Mu, H. B.; Zheng, Y. L.; Wang, C. Q.; Chen, Y. H.; Li, F. R.; Wang, J. H.; Cheng, J. K. J. Chromatogr., B 2007, 855, 280–285. (32) Frens, G. Nat. Phys. Sci. 1973, 241, 20–22. (33) Zhang, C.; Zhang, Z. Y.; Yu, B. B.; Shi, J. J.; Zhang, X. R. Anal. Chem. 2002, 74, 96–99. (34) Hayashi, T.; Tanaka, K.; Haruta, M. J. Catal. 1998, 178, 566–575. (35) Kricka, L. J.; Voyta, J. C.; Bronstein, I. Methods Enzymol. 2000, 305, 370–390. (36) Liu, H. Y.; Zhang, L.; Zhou, J. M.; Hao, Y. H.; He, P. G.; Fang, Y. Z. Anal. Chim. Acta 2005, 541, 123–127. (37) Zhao, W. T.; Lee, T. M. H.; Leung, S. S. Y.; Hsing, I. M. Langmuir 2007, 23, 7143–7147. (38) Wangoo, N.; Bhasin, K. K.; Mehta, S. K.; Suri, C. R. J. Colloid Interface Sci. 2008, 323, 247–254. (39) Genevieve, M.; Vieu, C.; Carles, R.; Zwick, A.; Briere, G.; Salome, L.; Trevisiol, E. Microelectron. Eng. 2007, 84, 1710–1713. (40) Huang, Y. F.; Hsieh, M. M.; Tseng, W. L.; Chang, H. T. J. Proteome Res. 2006, 5, 429–436. (41) Vicente, G.; Colon, L. A. Anal. Chem. 2008, 80, 1988–1994.
1143
dx.doi.org/10.1021/ac103166n |Anal. Chem. 2011, 83, 1137–1143
