Toxicon 96 (2015) 89e95
Contents lists available at ScienceDirect
Toxicon journal homepage: www.elsevier.com/locate/toxicon
Horseradish peroxidase and antibody labeled gold nanoparticle probe for amplified immunoassay of ciguatoxin in fish samples based on capillary electrophoresis with electrochemical detection Zhaoxiang Zhang*, Ying Liu, Chaoying Zhang, Wenxiu Luan State Key Laboratory Base of Eco-chemical Engineering, College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 November 2014 Received in revised form 24 January 2015 Accepted 28 January 2015 Available online 29 January 2015
This paper describes a new amplified immunoassay with horseradish peroxidase (HRP) and antibody (Ab) labeled gold nanoparticles (AuNPs) probe hyphenated to capillary electrophoresis (CE) with electrochemical (EC) detection for ultrasensitive determination of ciguatoxin CTX1B. AuNPs were conjugated with HRP and Ab, and then incubated with limited amount of CTX1B to produce immunocomplex. The immunoreactive sample was injected into capillary for CE separation and EC detection. Enhanced sensitivity was obtained by adopting the AuNPs as carriers of HRP and Ab at high HRP/Ab molar ratio. The calibration curve of CTX1B was in the range of 0.06e90 ng/mL. The detection limit was 0.045 ng/mL, which is 38-fold lower than that of HPLC-MS method for CTX1B analysis. The proposed method was successfully applied for the quantification of CTX1B in contamined fish samples by simultaneously labeling Ab and HRP on AuNPs. The amplified IA with HRP and Ab labeled AuNPs probe hyphenated to CE and EC detection provides a sensitive analytical approach for the determination of trace ciguatoxin in complex samples. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Capillary electrophoresis Electrochemical detection Gold nanoparticles Immunoassay CTX1B
1. Introduction Ciguatera is a global disease caused by ingestion of fishes contaminated with the potent polyether toxins known as ciguatoxins (CTXs). These toxins are produced by the marine dinoflagellate Gambierdiscus toxicus (Yasumoto and Murata, 1993). Herbivorous fish accumulate these toxins in their musculature and viscera after ingesting dinoflagellates. The spread of ciguatera strongly impacts public health and considered a worldwide health problem. The symptoms of ciguatera begin with gastrointestinal problems, such as nausea, vomiting, diarrhea, and abdominal pain within 12 h of eating a toxic fish. CTX1B is the major toxin involved in ciguatera fish poisoning in the Pacific region, contributing to 90% of the toxicity of ciguateric carnivorous fish and posing a health risk at levels above 0.1 ppb (Dickey and Plakas, 2010). A major problem in avoiding the disease is that fish contaminated with CTXs look,
* Corresponding author. E-mail address: [email protected] (Z. Zhang). http://dx.doi.org/10.1016/j.toxicon.2015.01.015 0041-0101/© 2015 Elsevier Ltd. All rights reserved.
smell, and taste normal. In addition, neither cooking nor freezing deactivates the heat-stable ciguatoxins, so an unwitting diner could easily consume a fatal meal if measures are not taken beforehand to test the fishes. The traditional methods for the determination of CTXs involve testing lipid extracts by mouse bioassay (Lewis and Sellin, 1993), high performance liquid chromatography-mass spectrometry (HPLC-MS) (Lewis et al., 1999; Otero et al., 2010; Yogi et al., 2011) and enzyme-linked immunosorbent assay (ELISA) (Oguri et al., 2003; Tsumuraya et al., 2010, 2012), etc. Among the methods, immunoassay remains the most desirable method for accurate, sensitive, routine, and portable use. CE based immunoassay that combines the high separation power of CE and the high ligand specificity of immunoassay has proved to be a powerful technique for the separation and analysis of toxins (Zhang et al., 2012). By CE based immunoassay, the assay is faster because the immunoreaction occurs in solution. However, research on ciguatera has been severely hindered by the lack of labeled antigen or antibody. Therefore, it is necessary to develop a reliable and specific CE based immunoassay for detecting of CTXs in contaminated fish.
90
Z. Zhang et al. / Toxicon 96 (2015) 89e95
Gold nanoparticles (AuNPs) have attracted great interest in biomolecular and toxin detection due to their facile synthesis and surface modification, making functionalized AuNPs an excellent candidate for bioconjugation (Lai et al., 2011; Yin et al., 2012). In particular, the large specific surface area and favorable biocompatibility of AuNPs can provide a suitable and promising platform for immobilizing enzyme moleculars for signal amplification in enzymelinkage reactions (Liu et al., 2010). All these made AuNPs been extensively used for the extraction and enrichment of analytes in complex matrices (Liu et al., 2008). Tseng's group employed AuNPs to extract and enrichment proteins (Lin et al., 2008) aminothiols (Shen et al., 2009; Chang and Tseng, 2010), indoleamines (Li et al., 2009) and melamine (Chang et al., 2010) from complex matrix followed by CE separation that demonstrated the AuNPs have capability to extracting a variety of molecules due to their high absorption capacity and facile functionalization. Ambrosi et al. applied AuNPs as multienzyme carriers to enzyme based immunoassay for the detection of biomarker (Ambrosi et al., 2010). The use of AuNPs allows the attachment of a multiple enzyme molecule which can generate an amplified optical signal. Lin et al. designed a triple signal amplification strategy for ultrasensitive immunosensing of cancer biomarker by using microbead carried AuNPs as tracing tag to label signal Ab and AuNPs induced silver deposition for anodic stripping analysis (Lin et al., 2012). Liu et al. employed membrane-based biosensing interface through the use of functional AuNPs in combination with in situ atom transfer radical polymerization reaction for detection of cell membrane binding proteins with high degree signal amplification (Liu and Cheng, 2012). Miao et al. reported an ultrasensitive sensing strategy for Agþ assay based on the combination of AuNPs and smart enzyme cleavage mediated signal amplification with the detection limit as low as 40 pg/mL (Miao et al., 2013). Cui et al. combined an electrochemical immunosensor using AuNPs/carbon nanotubes hybrids with HRP-functionalized AuNPs for the sensitive detection of human IgG with a detection limit of 40 pg/mL (Cui et al., 2008). Recently, CE-based chemiluminescence immunoassay by using AuNPs as protein label reagent was reported for biological molecules determination (Liu et al., 2011; Jiang et al., 2013). AuNPs were conjugated with Ab or HRP labeled Ab, and then link to Ag to produce immunocomplex by noncompetitive immunoreaction. In a previous paper, we developed a CE-based EC immunoassay enhanced by AuNPs for the simultaneous determination of four shellfish toxins (Zhang et al., 2013). The AuNPs not only modified the mobilities of analytes to improve resolution but also used as multianalytes carrier to generate amplified EC signals. A dual amplification technique combining field-amplified sample injection and gold nanoparticles as multienzyme carriers was developed for detection of E. coli in scallop samples (Zhang et al., 2012). The assay resulted in the improved sensitivity of 1400 fold when compared with traditional CE using 10 kV electrokinetic injection for 10 s. In this work, we proposed the combination of HRP and Ab labeled AuNPs probe with CE and EC detection for sensitive determination of CTX1B. The use of AuNPs as carriers allows the simultaneous attachment of multiple HRP and Ab which can generate an amplified EC signal. On the basis of the noncompetitive immunoreactions, the formed immunocomplex, unbound HRP-AuAb and excess HRP can be efficiently separated by CE and sensitively detected by EC detection. The amplified sensitivity was enhanced by using bioconjugates featuring HRP labels and Ab linked to AuNPs at high HRP/Ab ratio. To the best of our knowledge hitherto, it is not found the report about the quantitative determination of CTX1B by EC immunoassay by simultaneously labeling Ab and HRP on AuNPs.
2. Material and methods 2.1. Instrumentation Experiments were carried out using a laboratory-built CE based EC immunoassay system as described previously (Zhang and Zhang, 2012). Briefly, CE separation was performed in a model MPI-A CE setup (Remax Electronics Inc., Xi'an, China), equipped with a highvoltage power supplier (0e30 kV) for driving the electrophoresis and an EC potentiostat (0e2.5 V) for detection. Fused-silica capillaries (75 mm i.d., 375 mm o.d., Yongnian Chromatogr. Components Ltd., Hebei, China) with 30 cm and 5 cm were used as separation and reaction capillaries, respectively. The reaction capillary was coaxial along the separation capillary and working electrode. The buffer reservoir at the high voltage end was enclosed in a plexiglass box fitted with an interlock for operator safety. A commercial UV visible spectrophotometer (Tianpu Analytical instrument Co., Ltd, Shanghai, China) was used to measure the absorbance of the AuNPs and immunocomplex. An H7100 transmission electron microscopy (TEM) (Hitachi High-Technologies Corp., Tokyo, Japan) operating at 75 keV was used to collect TEM images. 2.2. Chemicals and reagents CTX1B (Ag) and anti-CTX3C Ab were purchased from Abraxis (USA). HRP (MW ¼ 44,000) and lyophilized 99% bovine serum albumin (BSA) were from Sigma. Hydrogen tetrachloroaurate (III) trihydrate (HAuCl4$3H2O, 99.9%), trisodium citrate, polyvinylpyrolidone (PVP, Mr ¼ 1,300,000), OAP and H2O2 were obtained from Shanghai Reagent Company (China). All buffer reagents and other chemicals were of analytical grade and supplied by local standard reagent suppliers, unless otherwise stated. All solutions were prepared in doubly distilled water. BrittoneRobinson (BR) buffer of various pH were prepared by dissolving appropriate amount of H3BO3, H3PO4 and HAc, and then adjusting the pH with concentrated NaOH. The CE running buffer was 10.0 mM BR buffer with 1.0% PVP and 1.0 mM H2O2 at pH 5.0. 2.3. Preparation and characterization of gold nanoparticles AuNPs were synthesized by sodium citrate reduction of HAuCl4 in water (Jin et al., 2003; Liu and Lu, 2006). Precisely, 25.0 mL of 38.8 mM sodium citrate was rapidly added to 250.0 mL of boiling 1.0 mM HAuCl4 solution under vigorous stirring. The solution changed color from pale yellow to deep red and then refluxed for another 30 min. After cooling, the synthesized AuNPs was filtered through a 0.22 mm cellulose membrane and stored at 4 C. The size of AuNPs was verified by TEM, and their concentration was estimated by UVevis spectroscopy. The concentration of the AuNPs was calculated to be about 5.6 nM. 2.4. Preparation of HRP and Ab labeled AuNPs probe The HRP and Ab labeled AuNPs probe was prepared by following a published procedure (Cui et al., 2008; Liu et al., 2011). 3.5 mL of 5.0 mg/mL HRP and 1.5 mL of 5.0 mg/mL Ab (which corresponding to the molar ratio of HRP/Ab is about 9/1) were added in 1.0 mL of the AuNPs solution (containing 0.04% trisodium citrate, 0.26 mM K2CO3, and 0.02% sodium azide) under agitation, followed by gently mixing and incubation at room temperature for 2 h. Then 100 mL of 1% BSA solution was added with stirring for 30 min at room temperature for blocking. The conjugate was centrifuged at 15,000 rpm for 20 min at 4 C. The oiled drop was washed with BR buffer (containing 1% BSA and 0.05% Tween 20) and resuspended in
Z. Zhang et al. / Toxicon 96 (2015) 89e95
100 mL of 1% BSA. The obtained HRP-Au-Ab probes can be used directly or stored at 4 C for months. 2.5. Preparation of fish samples Remnants of fish implicated in ciguatera fish poisoning incidents were obtained from Dongshan Island (Fujian, China). Fish samples were extracted by following the reported procedure (Yogi et al., 2011) with some modifications. In brief, 200 g of each flesh sample were extracted twice by homogenizing for 5 min in 600 mL of acetone in a blender and then centrifuged for 10 min at 3600 rpm. The combined supernatant evaporated to near dryness with a rotary evaporator under reduced pressure. The residue was dissolved in 50 mL of methanol/water (9:1) and then defatted with hexane (100 mL). The upper hexane phase was removed and discarded and the lower methanol/water phase was filtrated though a 0.45 mm filter. Store the extract at 4 C for future use. 2.6. Immunoassay based on HRP-Au-Ab probe hyphenated to CE with EC detection The schematic diagram of the immunoassay based on HRP-AuAb probe hyphenated to EC detection is shown in Fig. 1. AuNPs were conjugated with HRP and Ab to immobilize the HRP and Ab on the surface of AuNPs at high HRP/Ab ratio (Eq (1) and Fig. 1a), and incubated with Ag to form immunocomplex by noncompetitive format (Eq (2) and Fig. 1b). HRP þ Ab þ Au / Ab-Au-HRP
(1)
Ag þ Ab-Au-HRP (excess) / Ab-Au-HRP þ Ag-Ab-Au-HRP
(2)
where Ag is the CTX1B, and Ab-Au-HRP is the HRP and Ab labeled AuNPs probe with an excessive fixed amount in Eq (2). A volume of 10 mL of CTX1B standard or fish sample was mixed with 10 mL of Ab-Au-HRP probe in a 200 mL microcentrifuge tube and then diluted with 10.0 mM BR buffer to 50 mL. After incubation at 37 C for 40 min, the mixture was analyzed by CE separation and EC detection. The capillaries were flushed daily in the order of 1.0 M NaOHeCH3OH (15 min), H2O (1 min) and running buffer for 10 min successively. Between two runs, the capillaries were conditioned with running buffer for 6 min. After the EC signal reached a constant value, the incubated solution was introduced by electrokinetic injection at 10 kV for 10 s (Fig. 1c). CE separation was performed at 15 kV. The unbound HRP, HRP-Au-Ab probe and the formed immunocomplex were separated according to the velocity
Fig. 1. Schematic illustration of the developed noncompetitive immunoassay with CE separation and EC detection by AuNPs as carrier and amplification.
91
difference in the separation capillary (Fig. 1d) and successively entered the reaction capillary and catalyze the reaction between OAP and H2O2 to 2-aminophenoxazine-3-one (AP) (Fig. 1e). The AP was then EC detected at the Pt working electrode with the detection potential of 0.35 V (Zhang et al., 2012). The peak areas of the immunocomplex were used for quantification in this work. 3. Results and discussion 3.1. Characterization of the HRP-Au-Ab probe HRP and Ab were covalently linked to AuNPs to form HRP-Au-Ab probe through the interactions between AuNPs and mercapto or primary amine groups of biomolecules. This biomaterial was made to replace the conventional HRPeAb complex. Because a single AuNP can efficiently couple to multiple HRP and Ab around its surface, so the introduction of AuNPs can provide EC signal amplification, and improve the performance and sensitivity of the CE EC detection method. Fig. 2 shows TEM images of AuNPs before and after conjugation with HRP and Ab. Size analysis showed that the average size of AuNPs before conjugation was about 13 nm, whereas the average size of HRP-Au-Ab probe was about 33 nm. The 20 nm increase in size for the bioconjugate AuNPs indicated that HRP and Ab have been tagged onto the surface of AuNPs. UVevis absorption spectrum was also performed to characterize the formation of HPR and Ab conjugated AuNPs probe as shown in Fig. 3. The absorbance of the bare AuNPs decreased with the formation of HRP-Au-Ab probe. Moreover, 280 nm and 400 nm absorption peaks were observed compared to that of pure AuNPs. These results provided the evidence for the HPR and Ab loading on the surface of AuNPs. 3.2. Determination of the amount of HPR and Ab adsorbed onto the AuNPs AuNPs were used as carriers of multiple HRP and Ab for multilabel amplification to enhance sensitivity. The amount of HRP and Ab molecules adsorbed onto the AuNPs was determined (Bradford, 1976). After the HRP and Ab adsorption process, the HRP-Au-Ab conjugations were precipitated by centrifugation. The sum concentrations of HRP and Ab in solution before adsorption and in the supernatant after adsorption were determined. The difference in the amount of HRP and Ab before and after adsorption was calculated and represented the amount of HRP and Ab adsorbed onto the AuNPs surfaces. The sum amount of HRP and Ab in the prepared stock HRP-Au-Ab dispersion was estimated to be 14.38 mg. In this work, the EC signals achieved from the catalytic reaction of the carried HRP to the H2O2/OAP system. The more HRP labeled on the AuNPs surface, the higher the sensitivity will be. To determine the ratio of HRP and Ab in HRP-Au-Ab probe, the HRP-Au-Ab dispersion was reacted with HRP substrate OAP and H2O2. The reaction produced AP with characteristic UVevis absorbance peak at 435 nm. This was compared to a standard curve constructed with underivatized HRP, after subtracting the background absorbance of an equivalent dispersion of underivatized AuNPs. The amount of HRP in the stock HRP-Au-Ab dispersion was determined to be 9.97 mg. So the amount of Ab in the HRP-Au-Ab solution was 4.41 mg. From these data, the molar ratio of HRP and Ab in the HRP-Au-Ab probe was about 9/1, which was corresponded to the ratio of the initial amount. For determination of the number of HRP and Ab molecule adsorbed onto the AuNPs surfaces, various concentrations of HRP and Ab with fixing the molar ratio of HRP/Ab of 9/1 were mixed with 5.6 nM AuNPs solution to conjugation. After incubation, the conjugated sample was injected into capillary for CE separation and
92
Z. Zhang et al. / Toxicon 96 (2015) 89e95
Fig. 2. The TEM images of bare AuNPs (A) and HRP-AuNPs-Ab probe (B).
Fig. 3. UVevis absorption spectra of HRP-Au-Ab (a), AuNPs (b), HRP (c) and Ab (d).
UVevis absorbance detection. As shown in Fig. 4, the peak areas of HRP-Au-Ab probe increase with increasing of the concentration of Ab and HRP. When the concentrations of Ab and HRP were 3.7 and 8.62 mg/L, respectively, the peak area appeared to reach its maximum value that corresponded to the saturated adsorption capacity for AuNPs. The sum amount of HRP and Ab adsorbed onto the AuNP surface is calculated (Xie et al., 2010) of 39, and that corresponded to a number of HRP and Ab molecules of 35 and 4 for each AuNP, respectively. 3.3. Optimization of conditions for immunocomplex formation In order to determine the effect of incubation time on immunocomplex formation, the HRP-Au-Ab probe and CTX1B mixture was incubated at 37 C for 10, 20, 30, 40, 50, and 60 min before analysis by CE-EC detection. The EC detection signal of HRP-Au-AbCTX1B immunocomplex increased rapidly with incubation time up to 40 min. Then the peak value of immunocomplex did not change significantly at incubation times greater than 40 min, suggesting that the reaction had reached equilibrium by this time. Therefore, 40 min was selected as the incubation time. 3.4. Optimization of CE conditions After incubation, the immunoreactive sample was directly subjected to CE separation. The unbound HRP, HRP-Au-Ab probe and the formed immunocomplex were separated in the separation capillary. The parameters affecting the separation including
Fig. 4. Determination of the number of HRP and Ab molecule adsorbed onto the AuNPs surfaces. AuNPs concentration: 5.6 nM. The signal and error bars represent averages based on three measurements.
separation voltage, concentration and pH of running buffer, concentration of PVP were optimized. The effect of separation voltage was examined in the range of 10e20 kV. The results showed that the peak height and separation efficiency of immunocomplex was first increased with the increase in the separation voltage up to 15 kV. When further increased separation voltage the detector noise began to increase and the separation efficiency decreased since Joule heating becomes larger. Thus, 15 kV was used for the CE separation. Different running buffer solutions including phosphate, borate, acetate, citrate, tartrate, and BR buffer etc. with various pH and concentrations were examined for separation and detection of the immunoreactive samples. The results obtained show that BR buffer solution was more suitable because the peak heights and resolution excel than those of other buffer solutions. As a small molecular glycoprotein, HRP would be positive-charged and absorbed to the capillary wall when pH is below 6.5. So PVP was selected as a dynamic coating reagent added in buffer solution to decrease the adsorption of protein on the capillary inner wall (Liu et al., 2011). The effects of BR and PVP concentration on the separation of immunoreactive sample were studied. BR and PVP concentration ranging from 2 to 30 mM and 0.2%e3.0% (v/v) were tested, respectively. It was found that the peak height and separation efficiency (theoretical plate number, N) was first increase with increase of BR concentration up to 10 mM, and then decrease with
Z. Zhang et al. / Toxicon 96 (2015) 89e95
further increase of BR concentration due to Joule heating. While for PVP concentrations test, the peak shapes and resolution were increased with increasing PVP concentrations to 1.0%, and then further increases in the PVP concentrations, the peak shapes and resolution kept almost constant. Therefore, 10 mM BR and 1.0% PVP were used for further experiment.
93
Changing pH of buffer solution is the most direct strategy for optimizing separation conditions since pH greatly influences the ionization of silanols on capillary wall. Moreover, the pH change may induce conformational changes in the formation of the immunocomplex and influence the EC detection since the isoelectric point of HRP is 6.5 and the optimal condition of HRP catalysis activity is about pH 4 (Wang et al., 2004). Therefore, an acidic running buffer was chosen for CE separation. It was found that peak shapes and resolution improve with the increase of running buffer pH from 3.0 to 5.0. However, peak shapes become worse and resolution decrease with further increasing running buffer pH from 5.5 to 7.0. According to the experimental results described above, the best conditions for the separation of immunoreactive sample were as following: 15 kV separation voltage and an electrophoretic buffer composing of 10 mM BR and 1.0% PVP at pH 5.0. Under this optimal condition, the immunoreactive sample was well separated within 10 min with the peaks sharp and symmetrical.
3.5. Detection of CTX1B by AuNPs amplified immunoassay with CE separation and EC detection The detection of CTX1B by the proposed method with noncompetitive format, and the electropherograms were shown in Fig. 5. Fig. 5A is the electropherogram of unbound HRP and HRP-AuAb probe. After the addition of different concentration of CTX1B and incubation, the peaks of the HRP-Au-Ab-CTX1B immunocomplex appeared (shown in Fig. 5BeD). It can be seen that the peak areas of immunocomplex increased with the increase in concentration of CTX1B. Under the optimized conditions, the proposed AuNPs amplified immunoassay with CE separation and EC detection was evaluated in terms of linearity, limit of detection (LOD), and precision for CTX1B analysis. For the linearity, a series of concentration (from 0.06 to 120 ng/mL) of CTX1B standards were analyzed (Fig. 6). The linearity was obtained by plotting the peak areas of HRP-Au-AbCTX1B immunocomplex versus the concentrations of the CTX1B. The results indicate that the peak areas of immunocomplex were directly proportional to the concentrations of CTX1B in the range 0.06e90 ng/mL. Linear regression analysis of the results yielded the equation of y ¼ 25.85x þ 1.78, where y is peak areas of immunocomplex (nC) and x is the concentration of CTX1B (ng/mL). The
Fig. 5. Electropherograms of unbound HRP, HRP-Au-Ab probe and the formed HRP-AuAb-CTX1B immunocomplex. The concentrations of HRP-Au-Ab probe were constant. Parts A, B, C, and D correspond to 0, 5.0, 20.0, and 50.0 ng/mL CTX1B, respectively. Peak 1, unbound HRP; peak 2, HRP-Au-Ab probe; peak 3, HRP-Au-Ab-CTX1B immunocomplex. Conditions: electrophoretic buffer, 10 mM BR buffer with 1.0 mM H2O2 and 1.0% PVP at pH 5.0; substrate OAP concentration: 1.0 mM; electrokinetic injection with 10 kV for 10 s; separation voltage, 15 kV; detection potential, 0.35 V; CE separation capillary, 50 cm 75 mm id; enzymic catalytic reaction capillary, 5 cm 75 mm id.
Fig. 6. Calibration curves of CTX1B by AuNPs amplified immunoassay with CE separation and EC detection. The main figure is the curve with the CTX1B concentration of 0.06e120 ng/mL and the inset is the linear portion of the curve in the main figure to show the linear range. Error bars, standard deviation over three replicates.
94
Z. Zhang et al. / Toxicon 96 (2015) 89e95
amounts of CTX1B in fish samples were calculated from the regression equation. The correlation coefficient acquired was 0.997 and the LOD (S/N ¼ 3) was 0.045 ng/mL. The repeatability (measured by relative standard deviation, RSD, n ¼ 7) was studied by assaying 20 ng/mL CTX1B standard solution within a day (intraday) and in 4 days (interday). The RSDs of the migration time were 2.6% for intraday and 3.3% for interday, respectively. The RSDs of the peak area were 3.8% for intraday and 5.2% for interday, respectively.
Table 2 Comparison of methods for the analysis of CTXs. Method
LOD
Refs
HPLC/MS/MS gradient reversed-phase HPLC/MS/MS Ultraperformance LC/MS ELISA AuNPs amplified immunoassay with CE separation and EC detection
0.25 pg/g 0.04e0.2 ppb 1.68 ng/mL 5 ng/mL 0.045 ng/mL
Yogi et al., 2011 Lewis et al., 1999 Otero et al., 2010 Oguri et al., 2003 This work
3.6. Evaluation of specificity Table 3 Recoveries for the detection of CTX1B in fish samples.
To investigate cross-reactivity with other CTX congeners, we used CTX3C as a model congener and tested the formation of the immunocomplex of CTX1B-Ab-Au-HRP in the presence of Ab-AuHRP, CTX1B and CTX3C, respectively. The concentrations of CTX1B and CTX3C were varied from 0.3 to 200 ng/mL. The results show that the signal of the immunocomplex of CTX1B-Ab-Au-HRP increases with the increasing concentration of CTX1B. The immunocomplex can be detectable even for the CTX1B with the concentration as low as 0.3 ng/mL. The CTX3C was also tested in parallel as a negative control. Even in the presence of 200 ng/mL of CTX3C, there is no significant immunocomplex of CTX3C-Ab-AuHRP observed. The results clearly indicate that the developed AuNPs amplified immunoassay with CE separation and EC detection is specifically applied for the detection of CTX1B, and the crossreaction over the other CTX congeners by the antiCTX1B antibody is negligible.
evaluated by repeatedly analyzing each fish sample five times within a working day. RSDs for the CTX1B determination were between 2.6 and 4.9%.These results proved that the proposed method for the analysis of CTX1B in fish sample was reliable and has potential application in the analysis of real samples.
3.7. Detection of CTX1B in fish samples
4. Conclusions
The contents of CTX1B in eight contaminated fish samples were determined by the proposed method (Y) and HPLC-MS (X) and summarized in Table 1. The results exhibit the two methods coincided well with a regression equation of Y ¼ 0.997X þ 0.543 and a correlation coefficient of 0.992. The HPLC-MS method was performed according to the procedures reported in the literature (Lewis et al., 1999; Otero et al., 2010; Yogi et al., 2011). The LOD of CTX1B using HPLC-MS was obtained at 1.72 ng/mL. The sensitivity of the proposed AuNPs amplified immunoassay with CE separation and EC detection was almost 38 times greater than that of HPLC-MS for CTX1B analysis. The comparison of the LOD between the proposed method and the assays reported for the detection of CTXs was listed in Table 2. The enhanced sensitivity of the proposed AuNPs amplified immunoassay was mainly due to adopting AuNPs as carriers at high HRP/Ab molar ratio which extremely amplified the detection signal. Recoveries of CTX1B from fish sample matrix were also studied by adding CTX1B standard solution into fish samples, and the spiked samples were then analyzed again. Recoveries were found to be in the range of 89.4e107.5% (Table 3). The assay precision was
1 2 3 4 5 6 7 8
CTX1B content (mg/Kg) This method (RSD, %)
HPLC-MS
6.2 128.5 86.1 112.6 43.8 76.5 93.2 56.7
Contents lists available at ScienceDirect
Toxicon journal homepage: www.elsevier.com/locate/toxicon
Horseradish peroxidase and antibody labeled gold nanoparticle probe for amplified immunoassay of ciguatoxin in fish samples based on capillary electrophoresis with electrochemical detection Zhaoxiang Zhang*, Ying Liu, Chaoying Zhang, Wenxiu Luan State Key Laboratory Base of Eco-chemical Engineering, College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 November 2014 Received in revised form 24 January 2015 Accepted 28 January 2015 Available online 29 January 2015
This paper describes a new amplified immunoassay with horseradish peroxidase (HRP) and antibody (Ab) labeled gold nanoparticles (AuNPs) probe hyphenated to capillary electrophoresis (CE) with electrochemical (EC) detection for ultrasensitive determination of ciguatoxin CTX1B. AuNPs were conjugated with HRP and Ab, and then incubated with limited amount of CTX1B to produce immunocomplex. The immunoreactive sample was injected into capillary for CE separation and EC detection. Enhanced sensitivity was obtained by adopting the AuNPs as carriers of HRP and Ab at high HRP/Ab molar ratio. The calibration curve of CTX1B was in the range of 0.06e90 ng/mL. The detection limit was 0.045 ng/mL, which is 38-fold lower than that of HPLC-MS method for CTX1B analysis. The proposed method was successfully applied for the quantification of CTX1B in contamined fish samples by simultaneously labeling Ab and HRP on AuNPs. The amplified IA with HRP and Ab labeled AuNPs probe hyphenated to CE and EC detection provides a sensitive analytical approach for the determination of trace ciguatoxin in complex samples. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Capillary electrophoresis Electrochemical detection Gold nanoparticles Immunoassay CTX1B
1. Introduction Ciguatera is a global disease caused by ingestion of fishes contaminated with the potent polyether toxins known as ciguatoxins (CTXs). These toxins are produced by the marine dinoflagellate Gambierdiscus toxicus (Yasumoto and Murata, 1993). Herbivorous fish accumulate these toxins in their musculature and viscera after ingesting dinoflagellates. The spread of ciguatera strongly impacts public health and considered a worldwide health problem. The symptoms of ciguatera begin with gastrointestinal problems, such as nausea, vomiting, diarrhea, and abdominal pain within 12 h of eating a toxic fish. CTX1B is the major toxin involved in ciguatera fish poisoning in the Pacific region, contributing to 90% of the toxicity of ciguateric carnivorous fish and posing a health risk at levels above 0.1 ppb (Dickey and Plakas, 2010). A major problem in avoiding the disease is that fish contaminated with CTXs look,
* Corresponding author. E-mail address: [email protected] (Z. Zhang). http://dx.doi.org/10.1016/j.toxicon.2015.01.015 0041-0101/© 2015 Elsevier Ltd. All rights reserved.
smell, and taste normal. In addition, neither cooking nor freezing deactivates the heat-stable ciguatoxins, so an unwitting diner could easily consume a fatal meal if measures are not taken beforehand to test the fishes. The traditional methods for the determination of CTXs involve testing lipid extracts by mouse bioassay (Lewis and Sellin, 1993), high performance liquid chromatography-mass spectrometry (HPLC-MS) (Lewis et al., 1999; Otero et al., 2010; Yogi et al., 2011) and enzyme-linked immunosorbent assay (ELISA) (Oguri et al., 2003; Tsumuraya et al., 2010, 2012), etc. Among the methods, immunoassay remains the most desirable method for accurate, sensitive, routine, and portable use. CE based immunoassay that combines the high separation power of CE and the high ligand specificity of immunoassay has proved to be a powerful technique for the separation and analysis of toxins (Zhang et al., 2012). By CE based immunoassay, the assay is faster because the immunoreaction occurs in solution. However, research on ciguatera has been severely hindered by the lack of labeled antigen or antibody. Therefore, it is necessary to develop a reliable and specific CE based immunoassay for detecting of CTXs in contaminated fish.
90
Z. Zhang et al. / Toxicon 96 (2015) 89e95
Gold nanoparticles (AuNPs) have attracted great interest in biomolecular and toxin detection due to their facile synthesis and surface modification, making functionalized AuNPs an excellent candidate for bioconjugation (Lai et al., 2011; Yin et al., 2012). In particular, the large specific surface area and favorable biocompatibility of AuNPs can provide a suitable and promising platform for immobilizing enzyme moleculars for signal amplification in enzymelinkage reactions (Liu et al., 2010). All these made AuNPs been extensively used for the extraction and enrichment of analytes in complex matrices (Liu et al., 2008). Tseng's group employed AuNPs to extract and enrichment proteins (Lin et al., 2008) aminothiols (Shen et al., 2009; Chang and Tseng, 2010), indoleamines (Li et al., 2009) and melamine (Chang et al., 2010) from complex matrix followed by CE separation that demonstrated the AuNPs have capability to extracting a variety of molecules due to their high absorption capacity and facile functionalization. Ambrosi et al. applied AuNPs as multienzyme carriers to enzyme based immunoassay for the detection of biomarker (Ambrosi et al., 2010). The use of AuNPs allows the attachment of a multiple enzyme molecule which can generate an amplified optical signal. Lin et al. designed a triple signal amplification strategy for ultrasensitive immunosensing of cancer biomarker by using microbead carried AuNPs as tracing tag to label signal Ab and AuNPs induced silver deposition for anodic stripping analysis (Lin et al., 2012). Liu et al. employed membrane-based biosensing interface through the use of functional AuNPs in combination with in situ atom transfer radical polymerization reaction for detection of cell membrane binding proteins with high degree signal amplification (Liu and Cheng, 2012). Miao et al. reported an ultrasensitive sensing strategy for Agþ assay based on the combination of AuNPs and smart enzyme cleavage mediated signal amplification with the detection limit as low as 40 pg/mL (Miao et al., 2013). Cui et al. combined an electrochemical immunosensor using AuNPs/carbon nanotubes hybrids with HRP-functionalized AuNPs for the sensitive detection of human IgG with a detection limit of 40 pg/mL (Cui et al., 2008). Recently, CE-based chemiluminescence immunoassay by using AuNPs as protein label reagent was reported for biological molecules determination (Liu et al., 2011; Jiang et al., 2013). AuNPs were conjugated with Ab or HRP labeled Ab, and then link to Ag to produce immunocomplex by noncompetitive immunoreaction. In a previous paper, we developed a CE-based EC immunoassay enhanced by AuNPs for the simultaneous determination of four shellfish toxins (Zhang et al., 2013). The AuNPs not only modified the mobilities of analytes to improve resolution but also used as multianalytes carrier to generate amplified EC signals. A dual amplification technique combining field-amplified sample injection and gold nanoparticles as multienzyme carriers was developed for detection of E. coli in scallop samples (Zhang et al., 2012). The assay resulted in the improved sensitivity of 1400 fold when compared with traditional CE using 10 kV electrokinetic injection for 10 s. In this work, we proposed the combination of HRP and Ab labeled AuNPs probe with CE and EC detection for sensitive determination of CTX1B. The use of AuNPs as carriers allows the simultaneous attachment of multiple HRP and Ab which can generate an amplified EC signal. On the basis of the noncompetitive immunoreactions, the formed immunocomplex, unbound HRP-AuAb and excess HRP can be efficiently separated by CE and sensitively detected by EC detection. The amplified sensitivity was enhanced by using bioconjugates featuring HRP labels and Ab linked to AuNPs at high HRP/Ab ratio. To the best of our knowledge hitherto, it is not found the report about the quantitative determination of CTX1B by EC immunoassay by simultaneously labeling Ab and HRP on AuNPs.
2. Material and methods 2.1. Instrumentation Experiments were carried out using a laboratory-built CE based EC immunoassay system as described previously (Zhang and Zhang, 2012). Briefly, CE separation was performed in a model MPI-A CE setup (Remax Electronics Inc., Xi'an, China), equipped with a highvoltage power supplier (0e30 kV) for driving the electrophoresis and an EC potentiostat (0e2.5 V) for detection. Fused-silica capillaries (75 mm i.d., 375 mm o.d., Yongnian Chromatogr. Components Ltd., Hebei, China) with 30 cm and 5 cm were used as separation and reaction capillaries, respectively. The reaction capillary was coaxial along the separation capillary and working electrode. The buffer reservoir at the high voltage end was enclosed in a plexiglass box fitted with an interlock for operator safety. A commercial UV visible spectrophotometer (Tianpu Analytical instrument Co., Ltd, Shanghai, China) was used to measure the absorbance of the AuNPs and immunocomplex. An H7100 transmission electron microscopy (TEM) (Hitachi High-Technologies Corp., Tokyo, Japan) operating at 75 keV was used to collect TEM images. 2.2. Chemicals and reagents CTX1B (Ag) and anti-CTX3C Ab were purchased from Abraxis (USA). HRP (MW ¼ 44,000) and lyophilized 99% bovine serum albumin (BSA) were from Sigma. Hydrogen tetrachloroaurate (III) trihydrate (HAuCl4$3H2O, 99.9%), trisodium citrate, polyvinylpyrolidone (PVP, Mr ¼ 1,300,000), OAP and H2O2 were obtained from Shanghai Reagent Company (China). All buffer reagents and other chemicals were of analytical grade and supplied by local standard reagent suppliers, unless otherwise stated. All solutions were prepared in doubly distilled water. BrittoneRobinson (BR) buffer of various pH were prepared by dissolving appropriate amount of H3BO3, H3PO4 and HAc, and then adjusting the pH with concentrated NaOH. The CE running buffer was 10.0 mM BR buffer with 1.0% PVP and 1.0 mM H2O2 at pH 5.0. 2.3. Preparation and characterization of gold nanoparticles AuNPs were synthesized by sodium citrate reduction of HAuCl4 in water (Jin et al., 2003; Liu and Lu, 2006). Precisely, 25.0 mL of 38.8 mM sodium citrate was rapidly added to 250.0 mL of boiling 1.0 mM HAuCl4 solution under vigorous stirring. The solution changed color from pale yellow to deep red and then refluxed for another 30 min. After cooling, the synthesized AuNPs was filtered through a 0.22 mm cellulose membrane and stored at 4 C. The size of AuNPs was verified by TEM, and their concentration was estimated by UVevis spectroscopy. The concentration of the AuNPs was calculated to be about 5.6 nM. 2.4. Preparation of HRP and Ab labeled AuNPs probe The HRP and Ab labeled AuNPs probe was prepared by following a published procedure (Cui et al., 2008; Liu et al., 2011). 3.5 mL of 5.0 mg/mL HRP and 1.5 mL of 5.0 mg/mL Ab (which corresponding to the molar ratio of HRP/Ab is about 9/1) were added in 1.0 mL of the AuNPs solution (containing 0.04% trisodium citrate, 0.26 mM K2CO3, and 0.02% sodium azide) under agitation, followed by gently mixing and incubation at room temperature for 2 h. Then 100 mL of 1% BSA solution was added with stirring for 30 min at room temperature for blocking. The conjugate was centrifuged at 15,000 rpm for 20 min at 4 C. The oiled drop was washed with BR buffer (containing 1% BSA and 0.05% Tween 20) and resuspended in
Z. Zhang et al. / Toxicon 96 (2015) 89e95
100 mL of 1% BSA. The obtained HRP-Au-Ab probes can be used directly or stored at 4 C for months. 2.5. Preparation of fish samples Remnants of fish implicated in ciguatera fish poisoning incidents were obtained from Dongshan Island (Fujian, China). Fish samples were extracted by following the reported procedure (Yogi et al., 2011) with some modifications. In brief, 200 g of each flesh sample were extracted twice by homogenizing for 5 min in 600 mL of acetone in a blender and then centrifuged for 10 min at 3600 rpm. The combined supernatant evaporated to near dryness with a rotary evaporator under reduced pressure. The residue was dissolved in 50 mL of methanol/water (9:1) and then defatted with hexane (100 mL). The upper hexane phase was removed and discarded and the lower methanol/water phase was filtrated though a 0.45 mm filter. Store the extract at 4 C for future use. 2.6. Immunoassay based on HRP-Au-Ab probe hyphenated to CE with EC detection The schematic diagram of the immunoassay based on HRP-AuAb probe hyphenated to EC detection is shown in Fig. 1. AuNPs were conjugated with HRP and Ab to immobilize the HRP and Ab on the surface of AuNPs at high HRP/Ab ratio (Eq (1) and Fig. 1a), and incubated with Ag to form immunocomplex by noncompetitive format (Eq (2) and Fig. 1b). HRP þ Ab þ Au / Ab-Au-HRP
(1)
Ag þ Ab-Au-HRP (excess) / Ab-Au-HRP þ Ag-Ab-Au-HRP
(2)
where Ag is the CTX1B, and Ab-Au-HRP is the HRP and Ab labeled AuNPs probe with an excessive fixed amount in Eq (2). A volume of 10 mL of CTX1B standard or fish sample was mixed with 10 mL of Ab-Au-HRP probe in a 200 mL microcentrifuge tube and then diluted with 10.0 mM BR buffer to 50 mL. After incubation at 37 C for 40 min, the mixture was analyzed by CE separation and EC detection. The capillaries were flushed daily in the order of 1.0 M NaOHeCH3OH (15 min), H2O (1 min) and running buffer for 10 min successively. Between two runs, the capillaries were conditioned with running buffer for 6 min. After the EC signal reached a constant value, the incubated solution was introduced by electrokinetic injection at 10 kV for 10 s (Fig. 1c). CE separation was performed at 15 kV. The unbound HRP, HRP-Au-Ab probe and the formed immunocomplex were separated according to the velocity
Fig. 1. Schematic illustration of the developed noncompetitive immunoassay with CE separation and EC detection by AuNPs as carrier and amplification.
91
difference in the separation capillary (Fig. 1d) and successively entered the reaction capillary and catalyze the reaction between OAP and H2O2 to 2-aminophenoxazine-3-one (AP) (Fig. 1e). The AP was then EC detected at the Pt working electrode with the detection potential of 0.35 V (Zhang et al., 2012). The peak areas of the immunocomplex were used for quantification in this work. 3. Results and discussion 3.1. Characterization of the HRP-Au-Ab probe HRP and Ab were covalently linked to AuNPs to form HRP-Au-Ab probe through the interactions between AuNPs and mercapto or primary amine groups of biomolecules. This biomaterial was made to replace the conventional HRPeAb complex. Because a single AuNP can efficiently couple to multiple HRP and Ab around its surface, so the introduction of AuNPs can provide EC signal amplification, and improve the performance and sensitivity of the CE EC detection method. Fig. 2 shows TEM images of AuNPs before and after conjugation with HRP and Ab. Size analysis showed that the average size of AuNPs before conjugation was about 13 nm, whereas the average size of HRP-Au-Ab probe was about 33 nm. The 20 nm increase in size for the bioconjugate AuNPs indicated that HRP and Ab have been tagged onto the surface of AuNPs. UVevis absorption spectrum was also performed to characterize the formation of HPR and Ab conjugated AuNPs probe as shown in Fig. 3. The absorbance of the bare AuNPs decreased with the formation of HRP-Au-Ab probe. Moreover, 280 nm and 400 nm absorption peaks were observed compared to that of pure AuNPs. These results provided the evidence for the HPR and Ab loading on the surface of AuNPs. 3.2. Determination of the amount of HPR and Ab adsorbed onto the AuNPs AuNPs were used as carriers of multiple HRP and Ab for multilabel amplification to enhance sensitivity. The amount of HRP and Ab molecules adsorbed onto the AuNPs was determined (Bradford, 1976). After the HRP and Ab adsorption process, the HRP-Au-Ab conjugations were precipitated by centrifugation. The sum concentrations of HRP and Ab in solution before adsorption and in the supernatant after adsorption were determined. The difference in the amount of HRP and Ab before and after adsorption was calculated and represented the amount of HRP and Ab adsorbed onto the AuNPs surfaces. The sum amount of HRP and Ab in the prepared stock HRP-Au-Ab dispersion was estimated to be 14.38 mg. In this work, the EC signals achieved from the catalytic reaction of the carried HRP to the H2O2/OAP system. The more HRP labeled on the AuNPs surface, the higher the sensitivity will be. To determine the ratio of HRP and Ab in HRP-Au-Ab probe, the HRP-Au-Ab dispersion was reacted with HRP substrate OAP and H2O2. The reaction produced AP with characteristic UVevis absorbance peak at 435 nm. This was compared to a standard curve constructed with underivatized HRP, after subtracting the background absorbance of an equivalent dispersion of underivatized AuNPs. The amount of HRP in the stock HRP-Au-Ab dispersion was determined to be 9.97 mg. So the amount of Ab in the HRP-Au-Ab solution was 4.41 mg. From these data, the molar ratio of HRP and Ab in the HRP-Au-Ab probe was about 9/1, which was corresponded to the ratio of the initial amount. For determination of the number of HRP and Ab molecule adsorbed onto the AuNPs surfaces, various concentrations of HRP and Ab with fixing the molar ratio of HRP/Ab of 9/1 were mixed with 5.6 nM AuNPs solution to conjugation. After incubation, the conjugated sample was injected into capillary for CE separation and
92
Z. Zhang et al. / Toxicon 96 (2015) 89e95
Fig. 2. The TEM images of bare AuNPs (A) and HRP-AuNPs-Ab probe (B).
Fig. 3. UVevis absorption spectra of HRP-Au-Ab (a), AuNPs (b), HRP (c) and Ab (d).
UVevis absorbance detection. As shown in Fig. 4, the peak areas of HRP-Au-Ab probe increase with increasing of the concentration of Ab and HRP. When the concentrations of Ab and HRP were 3.7 and 8.62 mg/L, respectively, the peak area appeared to reach its maximum value that corresponded to the saturated adsorption capacity for AuNPs. The sum amount of HRP and Ab adsorbed onto the AuNP surface is calculated (Xie et al., 2010) of 39, and that corresponded to a number of HRP and Ab molecules of 35 and 4 for each AuNP, respectively. 3.3. Optimization of conditions for immunocomplex formation In order to determine the effect of incubation time on immunocomplex formation, the HRP-Au-Ab probe and CTX1B mixture was incubated at 37 C for 10, 20, 30, 40, 50, and 60 min before analysis by CE-EC detection. The EC detection signal of HRP-Au-AbCTX1B immunocomplex increased rapidly with incubation time up to 40 min. Then the peak value of immunocomplex did not change significantly at incubation times greater than 40 min, suggesting that the reaction had reached equilibrium by this time. Therefore, 40 min was selected as the incubation time. 3.4. Optimization of CE conditions After incubation, the immunoreactive sample was directly subjected to CE separation. The unbound HRP, HRP-Au-Ab probe and the formed immunocomplex were separated in the separation capillary. The parameters affecting the separation including
Fig. 4. Determination of the number of HRP and Ab molecule adsorbed onto the AuNPs surfaces. AuNPs concentration: 5.6 nM. The signal and error bars represent averages based on three measurements.
separation voltage, concentration and pH of running buffer, concentration of PVP were optimized. The effect of separation voltage was examined in the range of 10e20 kV. The results showed that the peak height and separation efficiency of immunocomplex was first increased with the increase in the separation voltage up to 15 kV. When further increased separation voltage the detector noise began to increase and the separation efficiency decreased since Joule heating becomes larger. Thus, 15 kV was used for the CE separation. Different running buffer solutions including phosphate, borate, acetate, citrate, tartrate, and BR buffer etc. with various pH and concentrations were examined for separation and detection of the immunoreactive samples. The results obtained show that BR buffer solution was more suitable because the peak heights and resolution excel than those of other buffer solutions. As a small molecular glycoprotein, HRP would be positive-charged and absorbed to the capillary wall when pH is below 6.5. So PVP was selected as a dynamic coating reagent added in buffer solution to decrease the adsorption of protein on the capillary inner wall (Liu et al., 2011). The effects of BR and PVP concentration on the separation of immunoreactive sample were studied. BR and PVP concentration ranging from 2 to 30 mM and 0.2%e3.0% (v/v) were tested, respectively. It was found that the peak height and separation efficiency (theoretical plate number, N) was first increase with increase of BR concentration up to 10 mM, and then decrease with
Z. Zhang et al. / Toxicon 96 (2015) 89e95
further increase of BR concentration due to Joule heating. While for PVP concentrations test, the peak shapes and resolution were increased with increasing PVP concentrations to 1.0%, and then further increases in the PVP concentrations, the peak shapes and resolution kept almost constant. Therefore, 10 mM BR and 1.0% PVP were used for further experiment.
93
Changing pH of buffer solution is the most direct strategy for optimizing separation conditions since pH greatly influences the ionization of silanols on capillary wall. Moreover, the pH change may induce conformational changes in the formation of the immunocomplex and influence the EC detection since the isoelectric point of HRP is 6.5 and the optimal condition of HRP catalysis activity is about pH 4 (Wang et al., 2004). Therefore, an acidic running buffer was chosen for CE separation. It was found that peak shapes and resolution improve with the increase of running buffer pH from 3.0 to 5.0. However, peak shapes become worse and resolution decrease with further increasing running buffer pH from 5.5 to 7.0. According to the experimental results described above, the best conditions for the separation of immunoreactive sample were as following: 15 kV separation voltage and an electrophoretic buffer composing of 10 mM BR and 1.0% PVP at pH 5.0. Under this optimal condition, the immunoreactive sample was well separated within 10 min with the peaks sharp and symmetrical.
3.5. Detection of CTX1B by AuNPs amplified immunoassay with CE separation and EC detection The detection of CTX1B by the proposed method with noncompetitive format, and the electropherograms were shown in Fig. 5. Fig. 5A is the electropherogram of unbound HRP and HRP-AuAb probe. After the addition of different concentration of CTX1B and incubation, the peaks of the HRP-Au-Ab-CTX1B immunocomplex appeared (shown in Fig. 5BeD). It can be seen that the peak areas of immunocomplex increased with the increase in concentration of CTX1B. Under the optimized conditions, the proposed AuNPs amplified immunoassay with CE separation and EC detection was evaluated in terms of linearity, limit of detection (LOD), and precision for CTX1B analysis. For the linearity, a series of concentration (from 0.06 to 120 ng/mL) of CTX1B standards were analyzed (Fig. 6). The linearity was obtained by plotting the peak areas of HRP-Au-AbCTX1B immunocomplex versus the concentrations of the CTX1B. The results indicate that the peak areas of immunocomplex were directly proportional to the concentrations of CTX1B in the range 0.06e90 ng/mL. Linear regression analysis of the results yielded the equation of y ¼ 25.85x þ 1.78, where y is peak areas of immunocomplex (nC) and x is the concentration of CTX1B (ng/mL). The
Fig. 5. Electropherograms of unbound HRP, HRP-Au-Ab probe and the formed HRP-AuAb-CTX1B immunocomplex. The concentrations of HRP-Au-Ab probe were constant. Parts A, B, C, and D correspond to 0, 5.0, 20.0, and 50.0 ng/mL CTX1B, respectively. Peak 1, unbound HRP; peak 2, HRP-Au-Ab probe; peak 3, HRP-Au-Ab-CTX1B immunocomplex. Conditions: electrophoretic buffer, 10 mM BR buffer with 1.0 mM H2O2 and 1.0% PVP at pH 5.0; substrate OAP concentration: 1.0 mM; electrokinetic injection with 10 kV for 10 s; separation voltage, 15 kV; detection potential, 0.35 V; CE separation capillary, 50 cm 75 mm id; enzymic catalytic reaction capillary, 5 cm 75 mm id.
Fig. 6. Calibration curves of CTX1B by AuNPs amplified immunoassay with CE separation and EC detection. The main figure is the curve with the CTX1B concentration of 0.06e120 ng/mL and the inset is the linear portion of the curve in the main figure to show the linear range. Error bars, standard deviation over three replicates.
94
Z. Zhang et al. / Toxicon 96 (2015) 89e95
amounts of CTX1B in fish samples were calculated from the regression equation. The correlation coefficient acquired was 0.997 and the LOD (S/N ¼ 3) was 0.045 ng/mL. The repeatability (measured by relative standard deviation, RSD, n ¼ 7) was studied by assaying 20 ng/mL CTX1B standard solution within a day (intraday) and in 4 days (interday). The RSDs of the migration time were 2.6% for intraday and 3.3% for interday, respectively. The RSDs of the peak area were 3.8% for intraday and 5.2% for interday, respectively.
Table 2 Comparison of methods for the analysis of CTXs. Method
LOD
Refs
HPLC/MS/MS gradient reversed-phase HPLC/MS/MS Ultraperformance LC/MS ELISA AuNPs amplified immunoassay with CE separation and EC detection
0.25 pg/g 0.04e0.2 ppb 1.68 ng/mL 5 ng/mL 0.045 ng/mL
Yogi et al., 2011 Lewis et al., 1999 Otero et al., 2010 Oguri et al., 2003 This work
3.6. Evaluation of specificity Table 3 Recoveries for the detection of CTX1B in fish samples.
To investigate cross-reactivity with other CTX congeners, we used CTX3C as a model congener and tested the formation of the immunocomplex of CTX1B-Ab-Au-HRP in the presence of Ab-AuHRP, CTX1B and CTX3C, respectively. The concentrations of CTX1B and CTX3C were varied from 0.3 to 200 ng/mL. The results show that the signal of the immunocomplex of CTX1B-Ab-Au-HRP increases with the increasing concentration of CTX1B. The immunocomplex can be detectable even for the CTX1B with the concentration as low as 0.3 ng/mL. The CTX3C was also tested in parallel as a negative control. Even in the presence of 200 ng/mL of CTX3C, there is no significant immunocomplex of CTX3C-Ab-AuHRP observed. The results clearly indicate that the developed AuNPs amplified immunoassay with CE separation and EC detection is specifically applied for the detection of CTX1B, and the crossreaction over the other CTX congeners by the antiCTX1B antibody is negligible.
evaluated by repeatedly analyzing each fish sample five times within a working day. RSDs for the CTX1B determination were between 2.6 and 4.9%.These results proved that the proposed method for the analysis of CTX1B in fish sample was reliable and has potential application in the analysis of real samples.
3.7. Detection of CTX1B in fish samples
4. Conclusions
The contents of CTX1B in eight contaminated fish samples were determined by the proposed method (Y) and HPLC-MS (X) and summarized in Table 1. The results exhibit the two methods coincided well with a regression equation of Y ¼ 0.997X þ 0.543 and a correlation coefficient of 0.992. The HPLC-MS method was performed according to the procedures reported in the literature (Lewis et al., 1999; Otero et al., 2010; Yogi et al., 2011). The LOD of CTX1B using HPLC-MS was obtained at 1.72 ng/mL. The sensitivity of the proposed AuNPs amplified immunoassay with CE separation and EC detection was almost 38 times greater than that of HPLC-MS for CTX1B analysis. The comparison of the LOD between the proposed method and the assays reported for the detection of CTXs was listed in Table 2. The enhanced sensitivity of the proposed AuNPs amplified immunoassay was mainly due to adopting AuNPs as carriers at high HRP/Ab molar ratio which extremely amplified the detection signal. Recoveries of CTX1B from fish sample matrix were also studied by adding CTX1B standard solution into fish samples, and the spiked samples were then analyzed again. Recoveries were found to be in the range of 89.4e107.5% (Table 3). The assay precision was
1 2 3 4 5 6 7 8
CTX1B content (mg/Kg) This method (RSD, %)
HPLC-MS
6.2 128.5 86.1 112.6 43.8 76.5 93.2 56.7
