Analyst View Article Online
PAPER
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
Cite this: DOI: 10.1039/c5an00636h
View Journal
Sensitive detection of atrazine in tap water using TELISA Zhiwei Qie,†a Jialei Bai,†a Bin Xie,b Lin Yuan,a Nan Song,a Yuan Peng,a Xianjun Fan,a Huanying Zhou,a Fengchun Chen,a Shuang Li,a Baoan Ning*a and Zhixian Gao*a A highly sensitive flow injection analysis (FIA)-based thermal enzyme-linked immunoassay, TELISA, was developed for the rapid detection of atrazine (ATZ) in tap water. ATZ and β-lactamase-labeled ATZ were employed in a competitive immunoassay using a monoclonal antibody (mAb). After the off-column liquid-phase competition, the mAb was captured on the Protein G Sepharose™ 4 Fast Flow (PGSFF) column support material. Injected β-lactamase substrate ampicillin was degraded by the column-bound ATZ–β-lactamase, generating a detectable heat signal. Several assay parameters were optimized, including substrate concentration, flow rates and regeneration conditions, as well as the mAb and ATZ-β dilution ratios and concentrations. The assay linear range was 0.73–4.83 ng mL−1 with a detection limit of 0.66 ng mL−1. An entire heat signal requires 10 min for generation, and the cycle time is less than 40 min. The results were reproducible and stable. ATZ-spiked tap water samples exhibited a recovery rate of 103%– 116%, which correlated with the UHPLC–MS/MS measurements. We attributed this significant increase in sensitivity over our previously published work to the following factors: (i) the capture of already-formed immune complexes on the column via immobilized Protein G, which eliminated chemical immobilization of the antibody; (ii) off-column preincubation allows the formation of immune complexes under nearly
Received 1st April 2015, Accepted 27th May 2015 DOI: 10.1039/c5an00636h www.rsc.org/analyst
ideal conditions; and (iii) multiple buffers can be used to, in one case, enhance immune-complex formation and in the other to maximize enzymatic activity. Furthermore, the scheme creates a universal assay platform in which sensing is performed in the off-column incubation and detection after capture in the enzyme thermistor (ET) detector, which opens up the possibility of detecting any antigen for which antibodies were available.
Introduction Atrazine (ATZ) belongs to a group of chemically related agricultural herbicides used to control broadleaf weeds. Its widespread use has resulted in extensive groundwater contamination.1 ATZ can persist in the environment, and as a result it is frequently detected in both ground and surface water.2 ATZ is regulated by the United States Environmental Protection Agency (EPA) in finished drinking water with maximum contaminant levels (MCLs) of 3.0 ng mL−1.3 Health risks associated with ATZ, such as cancer and endocrine disruption, are considered so high that the EU banned its use in
a Tianjin Key Laboratory of Risk Assessment and Control Technology for Environment and Food Safety, Tianjin Institute of Health and Environmental Medicine, Tianjin 300050, PR China. E-mail: [email protected], [email protected]; Fax: +86 022 84655403; Tel: +86 022 84655403 b Department of Pure and Applied Biochemistry, Lund University, Getingevägen 60, Box 124, S-22100 Lund, Sweden † These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2015
2004.4,5 Given the potential threat to human health, rapid, sensitive methods for routinely monitoring ATZ levels in water are urgently needed for countries where ATZ has not yet been banned. Currently, the presence of ATZ in water is monitored using chromatographic or immunological assays. Although highly sensitive, the chromatographic detection of ATZ involves tedious extraction, cleanup, and preconcentration procedures, which typically require the use of toxic reagents.6 Microtiter plate-based immunoassays are sensitive and accurate, but require extensive pipetting, washing and lengthy incubations. Furthermore, sample discoloration and/or turbidity adversely affect testing accuracy, requiring sample pretreatment procedures. Moreover, the commonly used microtiter plate-based format is not well suited for continuous monitoring.7,8 Flow injection analysis (FIA)-based thermal enzyme-linked immunoassays (TELISA) circumvent many of these drawbacks and afford a number of additional advantages.9–12 First, thermometric detection is less sensitive to sample contaminants because the thermal signal is not affected by discoloration or
Analyst
View Article Online
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
Paper
Analyst
turbidity. This eliminates sample pretreatment, which reduces costs and provides faster results. Second, the thermal transducers never come into direct contact with the sample, which greatly reduces recalibration requirements. Third, the automation of FIA-based devices is very well established. Commercially available devices can be easily integrated into the system to achieve a wide range of throughputs. Fourth, TELISA uses minimal amounts of toxic reagents, which eliminates the need for special permits as well as hazardous disposal costs. We have previously reported the development of a FIATELISA assay for detecting ATZ in corn using an enzyme thermistor (ET).13 Unfortunately, the assay possessed limited sensitivity. We suspected that this was primarily due to inefficiency and/or instability in the formation of ATZ–antibody complexes. As a result, our efforts to improve assay sensitivity have focused on improving the efficiency and stability of ATZ–antibody complex formation. Herein, we report the development of a novel immune-complex capture-based FIA-TELISA scheme for the detection of ATZ in water, which results in a significant improvement in sensitivity.
2
Experimental
2.1
Reagents
β-lactamase from Bacillus cereus and ATZ, simazine (SIM), melamine, cyanuric acid and chlorpyrifos were purchased from Sigma (St. Louis, MO, USA). Stock solutions of the herbicides (10 mg L−1) were prepared in dimethyl sulfoxide (DMSO) and stored at room temperature. The anti-ATZ mAb (subclass, mouse IgG1; immunogen, carboxylated atrazine coupled with bovine serum albumin) dissolved in PBS ( pH 7.0) was obtained in our lab.14 Ampicillin trihydrate, ampicillin Na, amoxicillin trihydrate, penicillin V potassium salt, penicillin G potassium salt and penicillin G sodium salt were purchased from Aladdin (Shanghai, China). The immunosorbent PGSFF was provided by GE Healthcare Bio-Sciences (Piscataway, NJ, USA). Protein G bound only to the Fc region of IgG, and the Fab of region IgG was available for binding antigen, which made PGSFF extremely useful for the isolation of immune complexes. Unlike protein A, protein G bound to mouse IgG1. All other chemicals used were of analytical grade. 2.2
Instruments
The FIA TELISA biosensor setup used in these studies is shown in Fig. 1. The ET instrument was purchased from Omic Bioscience AB, Sweden. The HPLC pump (UC 3281) was obtained from Union opto-electronic technology (Beijing, China). The injection valve (type 50) was purchased from Rheodyne, Cotati, USA. A standard water bath was used for the 37 °C incubations. 2.3
Preparation of ATZ–β-lactamase conjugate
The ATZ–β-lactamase conjugate was prepared as described previously with some minor modifications.7 Briefly, ATZ was activated by combining carboxylic ATZ (ATZ–COOH, MW: 283.38;
Analyst
Fig. 1
The TELISA setup used for the ATZ determinations.
9.2 mg), NHS (6.0 mg), and EDC–HCl (10.2 mg) in 200 µL DMF and stirred for 8 h under darkness at room temperature. The activated ATZ was added dropwise into an aqueous solution of β-lactamase to a final molar ratio of 30 : 1, followed by stirring at 4 °C for 12 h. The mixture was then dialyzed (8000 Da cutoff ) against PBS ( pH 7.0). The ATZ–β-lactamase conjugate was characterized, and the molar ratio was determined using a MALDI micro MX mass spectrometer (Waters/Micromass, Manchester, UK) fitted with a nitrogen laser. The instrument was operated in a positive-ion linear mode, and the MALDI plate was spotted using the dried-droplet technique with sinapinic acid as the matrix. 2.4
Assay protocol and sample preparation
The pre-incubation sample contained a certain amount of ATZ mixed with 0.8 mg mL−1 ATZ–β-lactamase and 2.8 mg mL−1 mAb to give a final dilution ratio of 100 and 8000, respectively, in a total volume of 500 μl. The competitive reaction was carried out at 37 °C for 30 min in a running buffer (PBS, 20 mmol L−1 phosphate buffer ( pH 7.0) and 2% DMSO (v/v)). The pre-incubated mixture (370 µL) was then injected into an ET mounted with a column containing the PGSFF support material equilibrated with running buffer, and the operating temperature was set at 37 °C. After washing, 370 µL of ampicillin trihydrate substrate (4 mmol L−1 dissolved in running buffer) was injected. The column was regenerated by injecting 12 mL of regeneration buffer (100 mmol L−1 of Gly–HCl ( pH 2.3), 1% DMSO (v/v), 0.1% Triton X-100 (v/v) and 500 mmol L−1 NaCl). The sample and substrate injections employed a flow rate of 400 µL min−1, and the regeneration flow rate was 1200 µL min−1. The reactor was held in a 20% ethanol solution at 4 °C to prevent bacterial growth when not in use. ATZ calibration curves were prepared using tap water. Tap water pH was adjusted to 7.0, filtered with a 0.22 µm PTFE filter (Phenomenex, Torrance, CA) and spiked with 50 µL of standard solution with increasing ATZ concentrations (0.1, 0.2, and 0.4 mg L−1 in DMSO). Blank samples were prepared in the same way with no spiked ATZ. The comparison of blank samples with samples using distilled water as the matrix to evaluate its effect showed no statistically significant difference.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Analyst
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
2.5
Paper
Validation
Water sample measurements were validated using UHPLC-MS/ MS. Tap water samples (5 mL) were spiked with 25 µL ATZ at various concentrations (0.1, 0.2, and 0.4 mg L−1 dissolved in methanol). Blank samples contained no ATZ. After shaking for 3 min, 5 mL chloroform was added. The mixtures were extracted with ultrasonication for 10 min and allowed to stand for 30 min. The organic layer was the collected. The chloroform was then dried via mild N2 drying, and 70% methanol–water (v/v) (5 mL) was subsequently added. The solid was completely dissolved by continuous shaking. The samples were quantified by using an Agilent 1260 ultra high performance liquid chromatograph (UHPLC) fitted with a Poroshell 120 reversed-phase C18 analytical column (2.1 mm × 50 mm, 2.7 μm particle size) (Agilent, MA, USA). The column temperature was set at 45 °C, the mobile phase A was 0.1% formic acid in acetonitrile (LC-MS grade) and mobile phase B was a mix of ultrapure water acetonitrile mix (LC-MS grade) (9 : 1, v/v) with an additional 0.1% (v/v) of formic acid with a flow rate of 0.2 mL min−1. The UHPLC and an Agilent 6410B triple–quadrupole mass spectrometer (UHPLC/MS/MS) were interfaced with an electrospray ionization source. All the compounds were ionized in the positive ion mode and analyzed by the UHPLC/MS/MS with the following parameters: capillary voltage, 2 kV; nebulizer pressure, 45 psig; drying gas, 8 L min−1; source temperature 30 °C; and gas temperature, 350 °C. The cone and desolvation gas flows were 75 and 750 L h−1, respectively. High-purity nitrogen (99.99%, Sifang, Tianjin, China) was used as a collision gas and the gas flow was set at 0.2 mL min−1 for a typical pressure of 4.2 × 10−3 mbar. Both the analytical device was controlled and the data were processed by MassHunter Workstation software B.03.01. The analysis was performed in triplicate.
ATZ was determined through the measurement of the β-lactamase activity in the reactor. ATZ–mAb and ATZ–β– lactamase at dilution ratios of 1 : 8000 and 1 : 100 were added into the sample solution, followed by incubation at 37 °C for 30 min. Glycerin (0.5%) was added to weaken unspecific combinations.
3
The most important result of this study was the significant increase in assay sensitivity that was achieved by reconfiguring the standard FIA-TELISA. This reconfiguration involved moving the antibody–antigen complex formation from an oncolumn to off-column liquid-phase incubation step, which speeds up and enhances complex formations, as well as allows control over incubation times. The immune complexes were then captured by Protein G immobilized on the column support material and subsequently detected, which had a number of advantages. First, the mAb was in the liquid phase instead of being immobilized, which allowed for more stable binding, as well as eliminating the need for a stop-flow step, which improved signal stability. Second, mAb–antigen binding efficiency was higher due to increased diffusion and longer incubation times. Third, optimized buffers could be employed for the binding and column detection phases. The ‘antibody’ sensing phase was particularly sensitive when detecting small antigens such as ATZ. The addition of a pre-incubation step added complexity to the assay, but given the considerable increase sensitivity it afforded, this was an acceptable trade-off. 3.1
2.6
Data analysis
B B1 ; B0 B1
ATZ–β-lactamase conjugate characterization
The ATZ–β-lactamase conjugate was prepared as described in Experimental. The conjugate was analyzed using a MALDI
The experimental signals were normalized using eqn (1) Normalized response ¼
Results and discussion
ð1Þ
where B was the signal determined in an increasing concentration of ATZ, B∞ was the background signal obtained in an excess concentration of ATZ, and B0 was the signal measured in the absence of ATZ. The normalized response was plotted as a function of ATZ concentration, and the experimental data of the calibration curve were fitted to eqn (2). Inhibition ¼ A2 þ
A1 A2 ΛP : X 1þ X0
ð2Þ
Cross-reactivity investigations were conducted by measuring the cross-reactivity rate (CR). According to eqn (3), the CR was calculated as the percentage of the IC50 of the competitive inhibition curves for ATZ over that for interfering compounds under optimized conditions. %CR ¼
IC50 ðATRÞ 100: IC50 of cross‐reacting compound
This journal is © The Royal Society of Chemistry 2015
ð3Þ
Fig. 2 Mass spectrograms of the ATZ–β-lactamase conjugate and the β-lactamase.
Analyst
View Article Online
Paper
Analyst
micro MX mass spectrometer (Fig. 2). Based on this data, the molecular weights of the β-lactamase and the ATZ– β-lactamase conjugate were determined to be 28 678 and 31 783 Daltons, respectively. This corresponds to an average ATZ : β-lactamase coupling ratio of approximately 11 : 1. This confirmed the successful preparation of the ATZ–β-lactamase conjugate.
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
3.2
Assay optimization
The running buffer designed for these studies comprised PBS, 20 mmol L−1 phosphate buffer ( pH 7.0) and 2% DMSO (v/v). PBS was chosen because it was well suited for the analysis of biological interactions. A concentration of 2% DMSO was added to enhance ATZ solubility and reduce non-specific binding. This DMSO concentration did not have any noticeable effect on the ATZ–mAb or mAb–Protein G interaction. The reconfiguration of the FIA-TELISA assay required the optimization of several assay parameters, including the ATZ–β-lactamase : ATZ–mAb ratio, choice of substrates, substrate concentration, and flow rates, as well as buffers for mAb–ATZ binding, regeneration and running buffers. The β-lactamase used in these studies, penicillinase from Bacillus cereus, belonged to the B metallo-β-lactamase class and had a broad substrate spectrum capable of catalyzing the hydrolysis of virtually any β-lactam antibiotic.15 The β-lactamase was extremely active, making it ideal for thermal FIA detection. The concentration of the ATZ–β-lactamase conjugate was varied between 2 and 32 µg mL−1 and the concentration of the ATZ–mAb conjugate was varied between 0.08 and 1.2 µg mL−1 and they were sequentially varied in order to determine the optimal concentration and ratio, as measured by assay sensitivity. The optimal concentrations of ATZ–β-lactamase and ATZ–mAb were determined to be 8 and 0.35 µg mL−1, respectively. To determine which β-lactamase substrate resulted in the highest sensitivity, several substrates were tested, including ampicillin trihydrate, sodium ampicillin, amoxicillin trihydrate, potassium penicillin V, potassium penicillin G and sodium penicillin G. Using the optimized conjugate ratios
Fig. 3
Analyst
mentioned above, the ampicillin class of substrates resulted in higher sensitivity than that of the penicillin. Further investigation revealed that 4 mmol L−1 ampicillin trihydrate resulted in the strongest signal inhibition, which suggests that the substrate under that concentration was most sensitive to ATZ– β-lactamase variation. Consequently, 4 mmol L−1 ampicillin trihydrate was used to prepare the calibration curves. The effective dissociation of the antibody complex from the PGSFF enables the reuse of the column, which is essential for the practical application of this biosensor technology. The strength of the binding interaction between Protein G and the Fc domain of the antibodies was well established and varies among species.16 The combinations of Gly–HCl and DMSO had been shown to be highly efficient dissociation buffers and had been widely used for the regeneration of columns, which enabled the repeated use of the immobilized antibodies in biosensor applications.17 Using this as a starting point, we had developed a regeneration buffer comprised of 0.2 mol L−1 Gly– HCl ( pH 2.3), 1% DMSO (v/v), 0.1% Triton X-100 (v/v) and 500 mmol L−1 NaCl, which efficiently dissociated the Protein G–antibody interaction. All the solutions were strictly filtered before use to eliminate interference induced by dissolved solids. 3.3
Assay protocol and calibration curves
Based on these results, an assay protocol was developed, and the course of the signal along with the assay time is shown in Fig. 3. ATZ present in the sample competed with ATZ–β-lactamase for binding sites on mAb during the incubation step. The incubated mixture was then injected into the system. The ATZ–mAb and ATZ–β-lactamase–mAb complexes formed during incubation were captured by the immobilized Protein G. A running buffer washed the column, and then 4 mM ampicillin trihydrate was injected. The resulting 10 min thermal signal was detected and used to quantify the amount of bound ATZ–β-lactamase present. The signal decreased as the amount of ATZ increased. The column was regenerated by injecting a regeneration buffer. The flow rates for the capture and detection phases were 400 μl min−1 and 1200 μl min−1, respectively,
Schematic of the assay protocol (a) and course along with assay time (b).
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
Analyst
Paper
in the regeneration phase. The flow rate was accelerated during regeneration to reduce the assay cycle time to 40 min. After regeneration, the baseline was regained by running buffer for 15 min, followed by next assay cycle. The reproducibility of the biosensor response in terms of the coefficient of variation for repeated sample detection in the presence of 6.25 μg mL−1 ATZ was not more than 8% (n = 6). The ability of the TELISA biosensor to quantitatively detect ATZ in tap water was studied. A calibration curve was prepared using a series of ATZ-spiked tap water samples ranging from 0.19 ng mL−1 to 25 ng mL−1. The ATZ competed with the ATZ– β-lactamase conjugate for binding sites on the mAb. The percentage inhibition was proportional to ATZ concentration 1 0 C 97:178 C 2:473 C. Each concentration was A X 1þ 1:777 analyzed three times and the data plotted (Fig. 4). The linear range was from 0.73 to 4.83 ng mL−1. The detection limit was 0.66 ng mL−1, which is the concentration on calibration corresponding to three times the standard deviation of reproducibility as inhibition strength. We noted that the lifetime of the column was shorter than that observed in our previous study, which were 30 to 40 determinations vs. 130 to 150 determinations. The main cause, we assume, is the considerably harsher regeneration condition, which could result in more severe activity loss of PGSFF. Compared with our previously published work, a highly sensitive and universal assay platform was constructed by introducing a preincubation step and an Fc-region-specific immunosorbent. Separating the sensing phase from that of the detection is helpful in the completion of the competitive reaction, which was thought to be crucial to sensitivity. The immunosorbent captured immune complexes by combining with the Fc region of IgG to make possible the assay target change in the off-column incubation without switching columns, which provided the basis for universal detection. The sensing phase, an off-column preincubation in a 37 °C water bath, similar to that of traditional ELISA, was not conB B B% I ¼ 97:552 @
Fig. 4
ATZ calibration curve using a log scale.
This journal is © The Royal Society of Chemistry 2015
ditioned by the detection phase. Thus, the reaction could be carried out under nearly ideal conditions, such as optimized dilution ratio of the mAb, buffer additive, and full immunecomplex formation time. Moreover, optimal antibody activity was also maintained through eliminating chemical immobilization, whereas, in previous work, on-column sensing took place simultaneously with the detection phase. The analyte ATZ competed with the ATZ–β-lactamase conjugate for binding sites on the immobilized mAb in a flowing buffer, which evidently restricted optimization. 3.4
Assay cross-reactivity and validation
Cross-reactivity was measured as a percentage of the ratio between the ATZ IC50 value over the value of the interfering compound. The results are shown in Table 1. The assay was highly specific for ATZ. Immunoassays sometimes suffered from nonspecific binding, which could result in false positives. To determine the validity of the TELISA capture assay results, tap water samples were spiked with various known amounts of ATZ and the samples were analyzed by UHPLC/MS/MS. The recovery and precision of these measurements were determined (Table 2). The tap water samples were checked for the presence of ATZ prior to these measurements and no detectable levels of herbicide were observed. Sample pretreatment included only filtration and pH adjustment. Excellent recoveries in the
Table 1 Cross-reactivity of some s-triazines and their analogs with the ATZ biosensor
IC50 (ng mL−1)
Cross-reactivity (%)
ATZ
1.81
100
SIM
4.80
37.7
Melamine
NCa
NC
Tricyanic acid
NC
NC
Chlorpyrifos
NC
NC
Compound
a
Structural formula
No competition.
Analyst
View Article Online
Paper
Analyst
Table 2 ATZ analysis in tap water samples using the biosensor developed (Mean ± SD, n = 3)
Spiked concentration (ng mL−1)
TELISA (Mean ± SDa) (ng mL−1)
Recovery ratio (%)
UHPLC/MS/ MS (Mean ± SD) (ng mL−1)
Recovery ratio (%)
0.50 1.00 2.00
0.58 ± 0.02 1.16 ± 0.06 2.07 ± 0.10
116.72 116.29 103.56
0.49 ± 0.08 1.08 ± 0.04 2.07 ± 0.03
97.87 108.32 103.65
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
a
Standard deviation.
range of 103%–116% were obtained with CV values of less than 2.0%. This confirms that the captured FIA-TELISA biosensor accurately measured the amount of ATZ in tap water, requiring minimal sample pretreatment.
4 Conclusions We described a novel capture FIA-TELISA assay that was capable of specifically determining ATZ levels in tap water. The assay was sensitive and highly reproducible. The linear range was from 0.73 to 4.83 ng mL−1 with a detection limit of 0.66 ng mL−1. The method required minimal sample pretreatment and the FIA-based scheme simplified automation. It took 10 min to generate the entire heat peak after sample injection and the assay circle is no more than 40 min, making rapid detection possible. The TELISA determinations showed excellent correlation with the UHPLC/MS/MS measurements. The minimal cross-reactivity indicated that the assay was highly specific for ATZ. The flow rates for different phases were carefully optimized to maintain time effectiveness and assay accuracy. The sensitivity was more than adequate for the determination of ATZ levels in water required by health authorities. The Protein G capture column was less stable than the antibody column used in our previous work. However, the stability was more than adequate for routine use with 30 to 40 determinations being able to be performed for each column. Approaches for extending column life were under investigation. The scheme that created a universal assay platform by separating the sensing phase from that of the detection phase is equally important. This provided flexibility because the assay target could be changed in the off-column incubation, eliminating the need to switch columns. In addition, the offcolumn incubation allowed for the conditions of immune complex formation to be optimized, which was particularly important when detecting small molecules with antibodies. Conversely, the detection could focus on optimizing the enzymatic reaction and signal detection. Future studies will focus on the determination of other commonly used pesticides.
Acknowledgements The authors gratefully acknowledge the financial support by the National Natural Science Foundation of China
Analyst
(No. 21177159 and 21207161) and the National Science and Technology Supporting Program (2012BAK08B06, 2012BAJ25B03-02).
References 1 N. Lavignac, K. R. Brain and C. J. Allender, Concentration dependent atrazine–atrazine complex formation promotes selectivity in atrazine imprinted polymers, Biosens. Bioelectron., 2006, 22, 138–144. 2 G. G. Ying, R. S. Kookana and M. Mallavarpu, Release behavior of triazine residues in stabilised contaminated soils, Environ. Pollut., 2005, 134, 71–77. 3 G. A. Smith, B. V. Pepich and D. J. Munch, Preservation and analytical procedures for the analysis of chloro-striazines and their chlorodegradate products in drinking waters using direct injection liquid chromatography tandem mass spectrometry, J. Chromatogr. A, 2008, 1202, 138–144. 4 N. Sathiakumar and E. Delzell, A review of epidemiologic studies of triazene herbicides and cancer, Crit. Rev. Toxicol., 1997, 27, 599–613. 5 F. Ackerman, The economics of atrazine, Int. J. Occupational Environ. Health, 2007, 13, 437–445. 6 N. S. L. Colla, C. E. Domini, J. E. Marcovecchio and S. E. Botté, Latest approaches on green chemistry preconcentration methods for trace metal determination in seawater – A review, J. Environ. Manage., 2015, 151, 44–55. 7 N. Sai, Y. Chen, N. Liu, G. G. Yu, P. Su, Y. Feng, Z. Zhou, X. Liu, H. Zhou, Z. Gao and B. A. Ning, A sensitive immunoassay based on direct hapten coated format and biotin– streptavidin system for the detection of chloramphenicol, Talanta, 2010, 82, 1113–1121. 8 M. C. Hennion and D. Barcelo, Strengths and limitations of immunoassays for effective and efficient use for pesticide analysis in water samples: A review, Anal. Chim. Acta, 1998, 362, 3–34. 9 M. Yakovleva, S. Bhand and B. Danielsson, The enzyme thermistor-A realistic biosensor concept. A critical review, Anal. Chim. Acta, 2013, 766, 1–12. 10 B. Xie and B. Danielsson, in Handbook of Biosensors and Biochips, ed. R. S. Marks, D. C. Cullen, I. Karube, C. R. Lowe and H. H. Weetall, 1st edn, 2007, ch. 38. 11 B. Xie, K. Ramanathan and B. Danielsson, Principles of Enzyme Thermistor Systems: Applications to Biomedical and Other Measurements, Adv. Biochem. Eng. Biotechnol., 1999, 64, 1–33. 12 B. Mattiasson, C. Borrebaeck, B. Sanfridson and K. Mosbach, Thermometric enzyme linked immunosorbent assay: TELISA, Biochim. Biophys. Acta, 1977, 483, 221–227. 13 Z. W. Qie, B. A. Ning, M. Liu, J. L. Bai, Y. Peng, N. Song, Z. Q. Lv, Y. Wang, S. M. Sun, X. Su, Y. H. Zhang and Z. X. Gao, Fast detection of atrazine in corn using thermometric biosensors, Analyst, 2013, 138, 5151–5156.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Analyst
16 K. StöbelL, A. Schönberg and C. Staak, A new non-species dependent ELISA for detection of antibodies to Borrelia burgdorferisl in zoo animals, Int. J. Med. Microbiol., 2002, 291, 88–99. 17 V. B. Kandimalla, N. S. Neeta, N. G. Karanth, M. S. Thakur, K. R. Roshini, B. E. A. Rani, A. Pasha and N. G. K. Karanth, Regeneration of ethyl parathion antibodies for repeated use in immunosensor: a study on dissociation of antigens from antibodies, Biosens. Bioelectron., 2004, 20, 903–906.
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
14 Z. Q. Lv, C. H. Wang, T. T. Wang, C. C. Chen, Y. Wang, B. A. Ning, M. Liu, J. Q. Liu, J. L. Bai, Y. Peng and Z. X. Gao, Detection of atrazine residue in food sample by a monoclonal antibody-based enzyme-linked immunosorbent assay, Biomed. Environ. Sci., 2013, 26, 398–402. 15 S. Zhou, Y. F. Zhao, M. Mecklenburg, D. J. Yang and B. Xie, A novel thermometric biosensor for fast surveillance of β-lactamase activity in milk, Biosens. Bioelectron., 2013, 49, 99–104.
Paper
This journal is © The Royal Society of Chemistry 2015
Analyst
PAPER
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
Cite this: DOI: 10.1039/c5an00636h
View Journal
Sensitive detection of atrazine in tap water using TELISA Zhiwei Qie,†a Jialei Bai,†a Bin Xie,b Lin Yuan,a Nan Song,a Yuan Peng,a Xianjun Fan,a Huanying Zhou,a Fengchun Chen,a Shuang Li,a Baoan Ning*a and Zhixian Gao*a A highly sensitive flow injection analysis (FIA)-based thermal enzyme-linked immunoassay, TELISA, was developed for the rapid detection of atrazine (ATZ) in tap water. ATZ and β-lactamase-labeled ATZ were employed in a competitive immunoassay using a monoclonal antibody (mAb). After the off-column liquid-phase competition, the mAb was captured on the Protein G Sepharose™ 4 Fast Flow (PGSFF) column support material. Injected β-lactamase substrate ampicillin was degraded by the column-bound ATZ–β-lactamase, generating a detectable heat signal. Several assay parameters were optimized, including substrate concentration, flow rates and regeneration conditions, as well as the mAb and ATZ-β dilution ratios and concentrations. The assay linear range was 0.73–4.83 ng mL−1 with a detection limit of 0.66 ng mL−1. An entire heat signal requires 10 min for generation, and the cycle time is less than 40 min. The results were reproducible and stable. ATZ-spiked tap water samples exhibited a recovery rate of 103%– 116%, which correlated with the UHPLC–MS/MS measurements. We attributed this significant increase in sensitivity over our previously published work to the following factors: (i) the capture of already-formed immune complexes on the column via immobilized Protein G, which eliminated chemical immobilization of the antibody; (ii) off-column preincubation allows the formation of immune complexes under nearly
Received 1st April 2015, Accepted 27th May 2015 DOI: 10.1039/c5an00636h www.rsc.org/analyst
ideal conditions; and (iii) multiple buffers can be used to, in one case, enhance immune-complex formation and in the other to maximize enzymatic activity. Furthermore, the scheme creates a universal assay platform in which sensing is performed in the off-column incubation and detection after capture in the enzyme thermistor (ET) detector, which opens up the possibility of detecting any antigen for which antibodies were available.
Introduction Atrazine (ATZ) belongs to a group of chemically related agricultural herbicides used to control broadleaf weeds. Its widespread use has resulted in extensive groundwater contamination.1 ATZ can persist in the environment, and as a result it is frequently detected in both ground and surface water.2 ATZ is regulated by the United States Environmental Protection Agency (EPA) in finished drinking water with maximum contaminant levels (MCLs) of 3.0 ng mL−1.3 Health risks associated with ATZ, such as cancer and endocrine disruption, are considered so high that the EU banned its use in
a Tianjin Key Laboratory of Risk Assessment and Control Technology for Environment and Food Safety, Tianjin Institute of Health and Environmental Medicine, Tianjin 300050, PR China. E-mail: [email protected], [email protected]; Fax: +86 022 84655403; Tel: +86 022 84655403 b Department of Pure and Applied Biochemistry, Lund University, Getingevägen 60, Box 124, S-22100 Lund, Sweden † These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2015
2004.4,5 Given the potential threat to human health, rapid, sensitive methods for routinely monitoring ATZ levels in water are urgently needed for countries where ATZ has not yet been banned. Currently, the presence of ATZ in water is monitored using chromatographic or immunological assays. Although highly sensitive, the chromatographic detection of ATZ involves tedious extraction, cleanup, and preconcentration procedures, which typically require the use of toxic reagents.6 Microtiter plate-based immunoassays are sensitive and accurate, but require extensive pipetting, washing and lengthy incubations. Furthermore, sample discoloration and/or turbidity adversely affect testing accuracy, requiring sample pretreatment procedures. Moreover, the commonly used microtiter plate-based format is not well suited for continuous monitoring.7,8 Flow injection analysis (FIA)-based thermal enzyme-linked immunoassays (TELISA) circumvent many of these drawbacks and afford a number of additional advantages.9–12 First, thermometric detection is less sensitive to sample contaminants because the thermal signal is not affected by discoloration or
Analyst
View Article Online
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
Paper
Analyst
turbidity. This eliminates sample pretreatment, which reduces costs and provides faster results. Second, the thermal transducers never come into direct contact with the sample, which greatly reduces recalibration requirements. Third, the automation of FIA-based devices is very well established. Commercially available devices can be easily integrated into the system to achieve a wide range of throughputs. Fourth, TELISA uses minimal amounts of toxic reagents, which eliminates the need for special permits as well as hazardous disposal costs. We have previously reported the development of a FIATELISA assay for detecting ATZ in corn using an enzyme thermistor (ET).13 Unfortunately, the assay possessed limited sensitivity. We suspected that this was primarily due to inefficiency and/or instability in the formation of ATZ–antibody complexes. As a result, our efforts to improve assay sensitivity have focused on improving the efficiency and stability of ATZ–antibody complex formation. Herein, we report the development of a novel immune-complex capture-based FIA-TELISA scheme for the detection of ATZ in water, which results in a significant improvement in sensitivity.
2
Experimental
2.1
Reagents
β-lactamase from Bacillus cereus and ATZ, simazine (SIM), melamine, cyanuric acid and chlorpyrifos were purchased from Sigma (St. Louis, MO, USA). Stock solutions of the herbicides (10 mg L−1) were prepared in dimethyl sulfoxide (DMSO) and stored at room temperature. The anti-ATZ mAb (subclass, mouse IgG1; immunogen, carboxylated atrazine coupled with bovine serum albumin) dissolved in PBS ( pH 7.0) was obtained in our lab.14 Ampicillin trihydrate, ampicillin Na, amoxicillin trihydrate, penicillin V potassium salt, penicillin G potassium salt and penicillin G sodium salt were purchased from Aladdin (Shanghai, China). The immunosorbent PGSFF was provided by GE Healthcare Bio-Sciences (Piscataway, NJ, USA). Protein G bound only to the Fc region of IgG, and the Fab of region IgG was available for binding antigen, which made PGSFF extremely useful for the isolation of immune complexes. Unlike protein A, protein G bound to mouse IgG1. All other chemicals used were of analytical grade. 2.2
Instruments
The FIA TELISA biosensor setup used in these studies is shown in Fig. 1. The ET instrument was purchased from Omic Bioscience AB, Sweden. The HPLC pump (UC 3281) was obtained from Union opto-electronic technology (Beijing, China). The injection valve (type 50) was purchased from Rheodyne, Cotati, USA. A standard water bath was used for the 37 °C incubations. 2.3
Preparation of ATZ–β-lactamase conjugate
The ATZ–β-lactamase conjugate was prepared as described previously with some minor modifications.7 Briefly, ATZ was activated by combining carboxylic ATZ (ATZ–COOH, MW: 283.38;
Analyst
Fig. 1
The TELISA setup used for the ATZ determinations.
9.2 mg), NHS (6.0 mg), and EDC–HCl (10.2 mg) in 200 µL DMF and stirred for 8 h under darkness at room temperature. The activated ATZ was added dropwise into an aqueous solution of β-lactamase to a final molar ratio of 30 : 1, followed by stirring at 4 °C for 12 h. The mixture was then dialyzed (8000 Da cutoff ) against PBS ( pH 7.0). The ATZ–β-lactamase conjugate was characterized, and the molar ratio was determined using a MALDI micro MX mass spectrometer (Waters/Micromass, Manchester, UK) fitted with a nitrogen laser. The instrument was operated in a positive-ion linear mode, and the MALDI plate was spotted using the dried-droplet technique with sinapinic acid as the matrix. 2.4
Assay protocol and sample preparation
The pre-incubation sample contained a certain amount of ATZ mixed with 0.8 mg mL−1 ATZ–β-lactamase and 2.8 mg mL−1 mAb to give a final dilution ratio of 100 and 8000, respectively, in a total volume of 500 μl. The competitive reaction was carried out at 37 °C for 30 min in a running buffer (PBS, 20 mmol L−1 phosphate buffer ( pH 7.0) and 2% DMSO (v/v)). The pre-incubated mixture (370 µL) was then injected into an ET mounted with a column containing the PGSFF support material equilibrated with running buffer, and the operating temperature was set at 37 °C. After washing, 370 µL of ampicillin trihydrate substrate (4 mmol L−1 dissolved in running buffer) was injected. The column was regenerated by injecting 12 mL of regeneration buffer (100 mmol L−1 of Gly–HCl ( pH 2.3), 1% DMSO (v/v), 0.1% Triton X-100 (v/v) and 500 mmol L−1 NaCl). The sample and substrate injections employed a flow rate of 400 µL min−1, and the regeneration flow rate was 1200 µL min−1. The reactor was held in a 20% ethanol solution at 4 °C to prevent bacterial growth when not in use. ATZ calibration curves were prepared using tap water. Tap water pH was adjusted to 7.0, filtered with a 0.22 µm PTFE filter (Phenomenex, Torrance, CA) and spiked with 50 µL of standard solution with increasing ATZ concentrations (0.1, 0.2, and 0.4 mg L−1 in DMSO). Blank samples were prepared in the same way with no spiked ATZ. The comparison of blank samples with samples using distilled water as the matrix to evaluate its effect showed no statistically significant difference.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Analyst
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
2.5
Paper
Validation
Water sample measurements were validated using UHPLC-MS/ MS. Tap water samples (5 mL) were spiked with 25 µL ATZ at various concentrations (0.1, 0.2, and 0.4 mg L−1 dissolved in methanol). Blank samples contained no ATZ. After shaking for 3 min, 5 mL chloroform was added. The mixtures were extracted with ultrasonication for 10 min and allowed to stand for 30 min. The organic layer was the collected. The chloroform was then dried via mild N2 drying, and 70% methanol–water (v/v) (5 mL) was subsequently added. The solid was completely dissolved by continuous shaking. The samples were quantified by using an Agilent 1260 ultra high performance liquid chromatograph (UHPLC) fitted with a Poroshell 120 reversed-phase C18 analytical column (2.1 mm × 50 mm, 2.7 μm particle size) (Agilent, MA, USA). The column temperature was set at 45 °C, the mobile phase A was 0.1% formic acid in acetonitrile (LC-MS grade) and mobile phase B was a mix of ultrapure water acetonitrile mix (LC-MS grade) (9 : 1, v/v) with an additional 0.1% (v/v) of formic acid with a flow rate of 0.2 mL min−1. The UHPLC and an Agilent 6410B triple–quadrupole mass spectrometer (UHPLC/MS/MS) were interfaced with an electrospray ionization source. All the compounds were ionized in the positive ion mode and analyzed by the UHPLC/MS/MS with the following parameters: capillary voltage, 2 kV; nebulizer pressure, 45 psig; drying gas, 8 L min−1; source temperature 30 °C; and gas temperature, 350 °C. The cone and desolvation gas flows were 75 and 750 L h−1, respectively. High-purity nitrogen (99.99%, Sifang, Tianjin, China) was used as a collision gas and the gas flow was set at 0.2 mL min−1 for a typical pressure of 4.2 × 10−3 mbar. Both the analytical device was controlled and the data were processed by MassHunter Workstation software B.03.01. The analysis was performed in triplicate.
ATZ was determined through the measurement of the β-lactamase activity in the reactor. ATZ–mAb and ATZ–β– lactamase at dilution ratios of 1 : 8000 and 1 : 100 were added into the sample solution, followed by incubation at 37 °C for 30 min. Glycerin (0.5%) was added to weaken unspecific combinations.
3
The most important result of this study was the significant increase in assay sensitivity that was achieved by reconfiguring the standard FIA-TELISA. This reconfiguration involved moving the antibody–antigen complex formation from an oncolumn to off-column liquid-phase incubation step, which speeds up and enhances complex formations, as well as allows control over incubation times. The immune complexes were then captured by Protein G immobilized on the column support material and subsequently detected, which had a number of advantages. First, the mAb was in the liquid phase instead of being immobilized, which allowed for more stable binding, as well as eliminating the need for a stop-flow step, which improved signal stability. Second, mAb–antigen binding efficiency was higher due to increased diffusion and longer incubation times. Third, optimized buffers could be employed for the binding and column detection phases. The ‘antibody’ sensing phase was particularly sensitive when detecting small antigens such as ATZ. The addition of a pre-incubation step added complexity to the assay, but given the considerable increase sensitivity it afforded, this was an acceptable trade-off. 3.1
2.6
Data analysis
B B1 ; B0 B1
ATZ–β-lactamase conjugate characterization
The ATZ–β-lactamase conjugate was prepared as described in Experimental. The conjugate was analyzed using a MALDI
The experimental signals were normalized using eqn (1) Normalized response ¼
Results and discussion
ð1Þ
where B was the signal determined in an increasing concentration of ATZ, B∞ was the background signal obtained in an excess concentration of ATZ, and B0 was the signal measured in the absence of ATZ. The normalized response was plotted as a function of ATZ concentration, and the experimental data of the calibration curve were fitted to eqn (2). Inhibition ¼ A2 þ
A1 A2 ΛP : X 1þ X0
ð2Þ
Cross-reactivity investigations were conducted by measuring the cross-reactivity rate (CR). According to eqn (3), the CR was calculated as the percentage of the IC50 of the competitive inhibition curves for ATZ over that for interfering compounds under optimized conditions. %CR ¼
IC50 ðATRÞ 100: IC50 of cross‐reacting compound
This journal is © The Royal Society of Chemistry 2015
ð3Þ
Fig. 2 Mass spectrograms of the ATZ–β-lactamase conjugate and the β-lactamase.
Analyst
View Article Online
Paper
Analyst
micro MX mass spectrometer (Fig. 2). Based on this data, the molecular weights of the β-lactamase and the ATZ– β-lactamase conjugate were determined to be 28 678 and 31 783 Daltons, respectively. This corresponds to an average ATZ : β-lactamase coupling ratio of approximately 11 : 1. This confirmed the successful preparation of the ATZ–β-lactamase conjugate.
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
3.2
Assay optimization
The running buffer designed for these studies comprised PBS, 20 mmol L−1 phosphate buffer ( pH 7.0) and 2% DMSO (v/v). PBS was chosen because it was well suited for the analysis of biological interactions. A concentration of 2% DMSO was added to enhance ATZ solubility and reduce non-specific binding. This DMSO concentration did not have any noticeable effect on the ATZ–mAb or mAb–Protein G interaction. The reconfiguration of the FIA-TELISA assay required the optimization of several assay parameters, including the ATZ–β-lactamase : ATZ–mAb ratio, choice of substrates, substrate concentration, and flow rates, as well as buffers for mAb–ATZ binding, regeneration and running buffers. The β-lactamase used in these studies, penicillinase from Bacillus cereus, belonged to the B metallo-β-lactamase class and had a broad substrate spectrum capable of catalyzing the hydrolysis of virtually any β-lactam antibiotic.15 The β-lactamase was extremely active, making it ideal for thermal FIA detection. The concentration of the ATZ–β-lactamase conjugate was varied between 2 and 32 µg mL−1 and the concentration of the ATZ–mAb conjugate was varied between 0.08 and 1.2 µg mL−1 and they were sequentially varied in order to determine the optimal concentration and ratio, as measured by assay sensitivity. The optimal concentrations of ATZ–β-lactamase and ATZ–mAb were determined to be 8 and 0.35 µg mL−1, respectively. To determine which β-lactamase substrate resulted in the highest sensitivity, several substrates were tested, including ampicillin trihydrate, sodium ampicillin, amoxicillin trihydrate, potassium penicillin V, potassium penicillin G and sodium penicillin G. Using the optimized conjugate ratios
Fig. 3
Analyst
mentioned above, the ampicillin class of substrates resulted in higher sensitivity than that of the penicillin. Further investigation revealed that 4 mmol L−1 ampicillin trihydrate resulted in the strongest signal inhibition, which suggests that the substrate under that concentration was most sensitive to ATZ– β-lactamase variation. Consequently, 4 mmol L−1 ampicillin trihydrate was used to prepare the calibration curves. The effective dissociation of the antibody complex from the PGSFF enables the reuse of the column, which is essential for the practical application of this biosensor technology. The strength of the binding interaction between Protein G and the Fc domain of the antibodies was well established and varies among species.16 The combinations of Gly–HCl and DMSO had been shown to be highly efficient dissociation buffers and had been widely used for the regeneration of columns, which enabled the repeated use of the immobilized antibodies in biosensor applications.17 Using this as a starting point, we had developed a regeneration buffer comprised of 0.2 mol L−1 Gly– HCl ( pH 2.3), 1% DMSO (v/v), 0.1% Triton X-100 (v/v) and 500 mmol L−1 NaCl, which efficiently dissociated the Protein G–antibody interaction. All the solutions were strictly filtered before use to eliminate interference induced by dissolved solids. 3.3
Assay protocol and calibration curves
Based on these results, an assay protocol was developed, and the course of the signal along with the assay time is shown in Fig. 3. ATZ present in the sample competed with ATZ–β-lactamase for binding sites on mAb during the incubation step. The incubated mixture was then injected into the system. The ATZ–mAb and ATZ–β-lactamase–mAb complexes formed during incubation were captured by the immobilized Protein G. A running buffer washed the column, and then 4 mM ampicillin trihydrate was injected. The resulting 10 min thermal signal was detected and used to quantify the amount of bound ATZ–β-lactamase present. The signal decreased as the amount of ATZ increased. The column was regenerated by injecting a regeneration buffer. The flow rates for the capture and detection phases were 400 μl min−1 and 1200 μl min−1, respectively,
Schematic of the assay protocol (a) and course along with assay time (b).
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
Analyst
Paper
in the regeneration phase. The flow rate was accelerated during regeneration to reduce the assay cycle time to 40 min. After regeneration, the baseline was regained by running buffer for 15 min, followed by next assay cycle. The reproducibility of the biosensor response in terms of the coefficient of variation for repeated sample detection in the presence of 6.25 μg mL−1 ATZ was not more than 8% (n = 6). The ability of the TELISA biosensor to quantitatively detect ATZ in tap water was studied. A calibration curve was prepared using a series of ATZ-spiked tap water samples ranging from 0.19 ng mL−1 to 25 ng mL−1. The ATZ competed with the ATZ– β-lactamase conjugate for binding sites on the mAb. The percentage inhibition was proportional to ATZ concentration 1 0 C 97:178 C 2:473 C. Each concentration was A X 1þ 1:777 analyzed three times and the data plotted (Fig. 4). The linear range was from 0.73 to 4.83 ng mL−1. The detection limit was 0.66 ng mL−1, which is the concentration on calibration corresponding to three times the standard deviation of reproducibility as inhibition strength. We noted that the lifetime of the column was shorter than that observed in our previous study, which were 30 to 40 determinations vs. 130 to 150 determinations. The main cause, we assume, is the considerably harsher regeneration condition, which could result in more severe activity loss of PGSFF. Compared with our previously published work, a highly sensitive and universal assay platform was constructed by introducing a preincubation step and an Fc-region-specific immunosorbent. Separating the sensing phase from that of the detection is helpful in the completion of the competitive reaction, which was thought to be crucial to sensitivity. The immunosorbent captured immune complexes by combining with the Fc region of IgG to make possible the assay target change in the off-column incubation without switching columns, which provided the basis for universal detection. The sensing phase, an off-column preincubation in a 37 °C water bath, similar to that of traditional ELISA, was not conB B B% I ¼ 97:552 @
Fig. 4
ATZ calibration curve using a log scale.
This journal is © The Royal Society of Chemistry 2015
ditioned by the detection phase. Thus, the reaction could be carried out under nearly ideal conditions, such as optimized dilution ratio of the mAb, buffer additive, and full immunecomplex formation time. Moreover, optimal antibody activity was also maintained through eliminating chemical immobilization, whereas, in previous work, on-column sensing took place simultaneously with the detection phase. The analyte ATZ competed with the ATZ–β-lactamase conjugate for binding sites on the immobilized mAb in a flowing buffer, which evidently restricted optimization. 3.4
Assay cross-reactivity and validation
Cross-reactivity was measured as a percentage of the ratio between the ATZ IC50 value over the value of the interfering compound. The results are shown in Table 1. The assay was highly specific for ATZ. Immunoassays sometimes suffered from nonspecific binding, which could result in false positives. To determine the validity of the TELISA capture assay results, tap water samples were spiked with various known amounts of ATZ and the samples were analyzed by UHPLC/MS/MS. The recovery and precision of these measurements were determined (Table 2). The tap water samples were checked for the presence of ATZ prior to these measurements and no detectable levels of herbicide were observed. Sample pretreatment included only filtration and pH adjustment. Excellent recoveries in the
Table 1 Cross-reactivity of some s-triazines and their analogs with the ATZ biosensor
IC50 (ng mL−1)
Cross-reactivity (%)
ATZ
1.81
100
SIM
4.80
37.7
Melamine
NCa
NC
Tricyanic acid
NC
NC
Chlorpyrifos
NC
NC
Compound
a
Structural formula
No competition.
Analyst
View Article Online
Paper
Analyst
Table 2 ATZ analysis in tap water samples using the biosensor developed (Mean ± SD, n = 3)
Spiked concentration (ng mL−1)
TELISA (Mean ± SDa) (ng mL−1)
Recovery ratio (%)
UHPLC/MS/ MS (Mean ± SD) (ng mL−1)
Recovery ratio (%)
0.50 1.00 2.00
0.58 ± 0.02 1.16 ± 0.06 2.07 ± 0.10
116.72 116.29 103.56
0.49 ± 0.08 1.08 ± 0.04 2.07 ± 0.03
97.87 108.32 103.65
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
a
Standard deviation.
range of 103%–116% were obtained with CV values of less than 2.0%. This confirms that the captured FIA-TELISA biosensor accurately measured the amount of ATZ in tap water, requiring minimal sample pretreatment.
4 Conclusions We described a novel capture FIA-TELISA assay that was capable of specifically determining ATZ levels in tap water. The assay was sensitive and highly reproducible. The linear range was from 0.73 to 4.83 ng mL−1 with a detection limit of 0.66 ng mL−1. The method required minimal sample pretreatment and the FIA-based scheme simplified automation. It took 10 min to generate the entire heat peak after sample injection and the assay circle is no more than 40 min, making rapid detection possible. The TELISA determinations showed excellent correlation with the UHPLC/MS/MS measurements. The minimal cross-reactivity indicated that the assay was highly specific for ATZ. The flow rates for different phases were carefully optimized to maintain time effectiveness and assay accuracy. The sensitivity was more than adequate for the determination of ATZ levels in water required by health authorities. The Protein G capture column was less stable than the antibody column used in our previous work. However, the stability was more than adequate for routine use with 30 to 40 determinations being able to be performed for each column. Approaches for extending column life were under investigation. The scheme that created a universal assay platform by separating the sensing phase from that of the detection phase is equally important. This provided flexibility because the assay target could be changed in the off-column incubation, eliminating the need to switch columns. In addition, the offcolumn incubation allowed for the conditions of immune complex formation to be optimized, which was particularly important when detecting small molecules with antibodies. Conversely, the detection could focus on optimizing the enzymatic reaction and signal detection. Future studies will focus on the determination of other commonly used pesticides.
Acknowledgements The authors gratefully acknowledge the financial support by the National Natural Science Foundation of China
Analyst
(No. 21177159 and 21207161) and the National Science and Technology Supporting Program (2012BAK08B06, 2012BAJ25B03-02).
References 1 N. Lavignac, K. R. Brain and C. J. Allender, Concentration dependent atrazine–atrazine complex formation promotes selectivity in atrazine imprinted polymers, Biosens. Bioelectron., 2006, 22, 138–144. 2 G. G. Ying, R. S. Kookana and M. Mallavarpu, Release behavior of triazine residues in stabilised contaminated soils, Environ. Pollut., 2005, 134, 71–77. 3 G. A. Smith, B. V. Pepich and D. J. Munch, Preservation and analytical procedures for the analysis of chloro-striazines and their chlorodegradate products in drinking waters using direct injection liquid chromatography tandem mass spectrometry, J. Chromatogr. A, 2008, 1202, 138–144. 4 N. Sathiakumar and E. Delzell, A review of epidemiologic studies of triazene herbicides and cancer, Crit. Rev. Toxicol., 1997, 27, 599–613. 5 F. Ackerman, The economics of atrazine, Int. J. Occupational Environ. Health, 2007, 13, 437–445. 6 N. S. L. Colla, C. E. Domini, J. E. Marcovecchio and S. E. Botté, Latest approaches on green chemistry preconcentration methods for trace metal determination in seawater – A review, J. Environ. Manage., 2015, 151, 44–55. 7 N. Sai, Y. Chen, N. Liu, G. G. Yu, P. Su, Y. Feng, Z. Zhou, X. Liu, H. Zhou, Z. Gao and B. A. Ning, A sensitive immunoassay based on direct hapten coated format and biotin– streptavidin system for the detection of chloramphenicol, Talanta, 2010, 82, 1113–1121. 8 M. C. Hennion and D. Barcelo, Strengths and limitations of immunoassays for effective and efficient use for pesticide analysis in water samples: A review, Anal. Chim. Acta, 1998, 362, 3–34. 9 M. Yakovleva, S. Bhand and B. Danielsson, The enzyme thermistor-A realistic biosensor concept. A critical review, Anal. Chim. Acta, 2013, 766, 1–12. 10 B. Xie and B. Danielsson, in Handbook of Biosensors and Biochips, ed. R. S. Marks, D. C. Cullen, I. Karube, C. R. Lowe and H. H. Weetall, 1st edn, 2007, ch. 38. 11 B. Xie, K. Ramanathan and B. Danielsson, Principles of Enzyme Thermistor Systems: Applications to Biomedical and Other Measurements, Adv. Biochem. Eng. Biotechnol., 1999, 64, 1–33. 12 B. Mattiasson, C. Borrebaeck, B. Sanfridson and K. Mosbach, Thermometric enzyme linked immunosorbent assay: TELISA, Biochim. Biophys. Acta, 1977, 483, 221–227. 13 Z. W. Qie, B. A. Ning, M. Liu, J. L. Bai, Y. Peng, N. Song, Z. Q. Lv, Y. Wang, S. M. Sun, X. Su, Y. H. Zhang and Z. X. Gao, Fast detection of atrazine in corn using thermometric biosensors, Analyst, 2013, 138, 5151–5156.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Analyst
16 K. StöbelL, A. Schönberg and C. Staak, A new non-species dependent ELISA for detection of antibodies to Borrelia burgdorferisl in zoo animals, Int. J. Med. Microbiol., 2002, 291, 88–99. 17 V. B. Kandimalla, N. S. Neeta, N. G. Karanth, M. S. Thakur, K. R. Roshini, B. E. A. Rani, A. Pasha and N. G. K. Karanth, Regeneration of ethyl parathion antibodies for repeated use in immunosensor: a study on dissociation of antigens from antibodies, Biosens. Bioelectron., 2004, 20, 903–906.
Published on 27 May 2015. Downloaded by University of Iowa on 11/06/2015 08:36:21.
14 Z. Q. Lv, C. H. Wang, T. T. Wang, C. C. Chen, Y. Wang, B. A. Ning, M. Liu, J. Q. Liu, J. L. Bai, Y. Peng and Z. X. Gao, Detection of atrazine residue in food sample by a monoclonal antibody-based enzyme-linked immunosorbent assay, Biomed. Environ. Sci., 2013, 26, 398–402. 15 S. Zhou, Y. F. Zhao, M. Mecklenburg, D. J. Yang and B. Xie, A novel thermometric biosensor for fast surveillance of β-lactamase activity in milk, Biosens. Bioelectron., 2013, 49, 99–104.
Paper
This journal is © The Royal Society of Chemistry 2015
Analyst
