Massive Immuno Multiresidue Screening of Water Pollutants

Article pubs.acs.org/ac

Massive Immuno Multiresidue Screening of Water Pollutants Paulina Dobosz,† Sergi Morais,† Emilio Bonet,‡ Rosa Puchades,† and Á ngel Maquieira*,† †

Instituto Interuniversitario de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Departamento de Química, Universitat Politècnica de València, Camino de Vera s/n, 46071 Valencia, Valencia, Spain ‡ Gamaser, S.L., Ronda Isaac Peral 4, 46980 Paterna, Valencia, Spain S Supporting Information *

ABSTRACT: An immuno multiresidue screening assay in microarray format for the determination of complex chemical mixtures at the microgram per liter level, using antibody-functionalized gold nanoparticles, is presented. The analytical method relies on the use of a cocktail of nanogold-labeled specific antibodies, acting as recognition and detection species. The concept of multireside screening is proved by developing a multiplex assay on a compact disk support for the determination of 2-(2,4,5trichlorophenoxy)propionic acid, 3-phenoxybenozic acid, 4-nitrophenol, alachlor, atrazine, azoxystrobin, chlorpyrifos, diazinon, diuron, endosulfan, fenthion, forchlorfenuron, imidacloprid, malathion, pentachlorophenol, pyraclostrobin, sulfasalazine, and triclosan, achieving detection limits of 0.07, 0.24, 10.9, 0.21, 0.14, 0.11, 0.11, 102, 0.36, 1.8, 1.7, 0.06, 0.08, 5.8, 1.0, 0.39, 0.003, and 12 μg/L, respectively. Due to the selectivity of the antibody-functionalized nanoparticles, the developed screening methodology allows the simultaneous determination of mixtures of water pollutants in a 10-plex configuration. The analytical performances were compared with those of reference chromatographic methods by the analysis of spiked water samples, the sensitivity and recovery results being in good agreement. The presented screening approach directly quantifies the concentration of complex chemical mixtures without sample treatment or preconcentration steps in a total time of 35 min.

M

crucial, since the overall risk may often be governed by only a few key analytes.6 Therefore, the development of multiresidue screening methods for the detection of complex mixtures of chemicals is of great interest. Usually this type of analysis is performed by liquid or gas chromatography coupled with mass spectrometry detection;7 however, for each group of compounds with different chemical features, a specific analytical method is required, making the monitoring process complex, time-consuming, and expensive. The optimal solution would be a method able to analyze the group of compounds in a single assay, allowing for appropriate information and timely decisionmaking. An ideal situation that could fulfill this requirement is the use of rapid, in situ, low-cost methods working as early warning systems, showing trends and variations of the pollutants’ concentrations in real time. Immunochemical methods have been proved as excellent screening techniques for the detection of environmental pollutants, as they are sensitive and selective and offer a rapid response. Different multianalyte immunoassay approaches have been developed; for example, Desmet et al.8 developed an

onitoring of water quality involves trace determination of chemical pollutant residues with the ultimate goal of achieving a good water status and management. For this, sensitive analysis is carried out by means of official methods. However, for many years, screening techniques developed in different formats and configurations have supported the environmental impact assessments since they are sensitive, selective, rapid, and cost-effective.1,2 Besides, they show additional advantages as they are prone to online monitoring and in situ analysis, which makes if possible for them to act as early warning techniques, for immediate assessment of problematic environments. In the European Union (EU), there are more than 100000 registered chemicals, 30000−70000 of which are of daily use.3 Although most of these compounds are present at low concentrations, the effect of complex chemical mixtures is one of the major concerns in human health and water quality evaluation.4 To address this issue, the European Commission established the European Water Framework Directive to achieve the best possible chemical water status.5 The assessment of the impact of pollutants in aquatic systems is a challenging task requiring improved analytical tools to detect the occurrence of single compounds and chemical mixtures. The identification of the copresence of chemical substances is © XXXX American Chemical Society

Received: June 5, 2015 Accepted: September 7, 2015

A

DOI: 10.1021/acs.analchem.5b02354 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry ELISA for the determination of five pollutants using a microarray format integrated within a classical 96-well plate. However, for massive in situ sample screening, the simplification of the operational steps is required. Other approaches based on a disposable dipstick have been developed in a triplex assay for the detection of pesticide residues;9 however, the semiquantitative nature of the method and the limited multiplex capability are the main bottlenecks. Also, advanced flow immunosensors were developed with fluorescence detection for triazine herbicides.10 Recently, the possibility of using suspension arrays of encoded microcarriers has emerged as an alternative screening tool. Thus, Li et al. developed an approach that determines simultaneously pesticide and veterinary drug residues in a heptaplex format in 2 h.11 The sensitivity and the multiplex capability, however, have to be improved during the fabrication and preparation of the suspension array. Compact disk technology has been demonstrated to have unique advantages as an analytical tool. In this sense, the standard optical disks are ideal platforms for the development of high-throughput, multiplex, cheap, and easy-to-use microanalytical devices.12 Besides, the use of compact disks in combination with a portable disk player detector addresses the current environmental concerns regarding water quality in situ evaluation. In previous works,13,14 we proved that digital versatile disk technology was very promising for multiplexing analysis. Here, we present a multiresidue warning approach that enables the massive and selective screening for chemical pollutants in different n-plex configurations, creating a particular chemical footprint. The approach uses specific antibody-functionalized gold nanoparticles for direct detection of chemical mixtures. This tracing mode is very advantageous because it allows for the use of simple detectors such as photometric ones or more sophisticated detectors such as plasmonics, reaching high sensitivity in both cases. The presented approach enables the arrangement of different multiplex configurations within the same compact disk as the sensing surface; thus, compounds showing structural resemblance are analyzed in parallel, expanding the potential of immunoassays toward simultaneous screening of dozens of analytes. To the best of our knowledge, this is the first development of a multiresidue screening system based on specific antibodyfunctionalized gold nanoparticles as sensing and detection species. The immuno multiresidue approach relies on compact disk technology that captures the key events leading to water contamination by creation of characteristic signatures to assess the environmental impact of complex chemical mixtures. In this work the chemical signatures are represented in Tetris layout. As a proof of concept, the simultaneous and parallel analysis of 18 water pollutants of different chemical families is addressed.

Riedel-de Haen (Seelze, Germany), imidacloprid (IMD) was from Bayer AG (Leverkusen, Germany), and chlorpyrifos (CLP), fenthion (FTN), diazinon (DZN), malathion (MLT), and pentachlorophenol (PCP) standards were from Dr. Ehrenstorfer (Augsburg, Germany). The previously characterized monoclonal and polyclonal antibodies15−32 against the set of target analytes were used for a multiresidue determination approach. Ovalbumin (OVA) was used for the preparation of the coating conjugates via activation of the carboxylate group of haptens with the active ester method, followed by formation of amide bonds with the free amino groups of the carrier protein.33 The coating conjugates were purified by gel filtration chromatography using PD-10 desalting columns (GE Healthcare Europe GmbH, Barcelona, Spain). The preparation and characterization of antibodyfunctionalized gold nanoparticles were performed as described.34 Microimmunoassay Protocol. The assay consists of the direct detection of the immunoreaction event by the use of antibody-functionalized nanoparticles. A scheme of the multiresidue assay is depicted in Figure 1. The polycarbonate surface

Figure 1. Scheme of the immuno multiresidue screening assay.

of standard DVDs (CD Rohling-up GmbH, Saarbrücken, Germany) was printed in microarray format (20 arrays per disk of 7 × 7 spots) by dispensing 50 nL of the coating conjugates using a noncontact printing device (AD 1500 BioDot, Inc., Irvine, CA). The spots were 500 μm in diameter with a center to center distance of 1.0 mm, achieving an array density of 4.0 spot/mm2. Within each microarray, spots for the analytes (four replicates) and positive and negative controls are included. A BSA/KLH solution (10 and 1.0 mg/L, respectively) and GAR− Au (1/800; GAR = goat antirabbit) were used as positive controls for the immunoreaction and amplification steps, respectively. As a negative control, ovalbumin (5.0 mg/L) was used. After 16 h at 4 °C, proteins were physically adsorbed onto the polycarbonate side of the DVD, and the disk was thoroughly washed with PBS-T, rinsed with deionized water, and dried by centrifugation at 800 rpm. The single-target competitive assay consisted of dispensing 20 μL of the antibody-functionalized gold nanoparticles in PBS-T, with or without analyte over the array. After 25 min of incubation at room temperature, the disk was washed with PBS-T and deionized water. The immunoreaction was developed by dispensing 1.0 mL of silver enhancer solution and distributing it evenly across the entire disk (⌀ = 12 cm) using a dummy plastic surface. After 10 min, the dummy cover was removed and the reaction stopped by washing the disk with water. For multiresidue n-plex assay, 20 μL of the gold-labeled antibody cocktail solution in PBS-T, with or without the mixture of the standards, was dispensed onto the array. Next, the immuno-

■

EXPERIMENTAL SECTION Chemicals. Gold nanoparticles (5 nm), keyhole limpet hemocyanin (KLH), bovine serum albumin (BSA), silver enhancer solutions (A and B), dimethyl sulfoxide (DMSO), Tris base, and standards for 2-(2,4,5-trichlorophenoxy)propionic acid (TPA), 3-phenoxybenzoic acid (PBA), alachlor (ALA), atrazine (ATZ), diuron (DIU), forchlorfenuron (FCF), and triclosan (TCS) were from Sigma-Aldrich (Madrid, Spain). 4-Nitrophenol (4NP) was from Merck (Darmstadt, Germany), azoxystrobin (AZB) was from Syngenta AG (Basel, Switzerland), endosulfan (END) and pyraclostrobin (PYS) were from B


Article

Analytical Chemistry Table 1. Color-Coded Shared Reactivity of Antibody-Functionalized Nanoparticlesa

a

Signal intensity (au): −, 10501.

analytical procedures are given in detail in the Supporting Information.

assay was performed as described before. The disk was read using a DVD drive, and the data were analyzed as previously described.14 Briefly, during the DVD reading, the laser (λ = 650 nm) hits the immunoreaction product, which modifies the reflection properties of the DVD surface, attenuating the reflected signal intensity that reaches the photodiode of the pickup. The analog signals from the drive’s photodetector are extracted before they are digitized during the reading of the digital content and brought into a data acquisition program to further convert them into an image. Digital images are processed for quantitative detection of the immunoreaction products, their optical density being inversely proportional to the concentration of the analytes. Analysis of Water Samples. The disk was segmented into 20 arrays, and a total of 15 surface water samples from the local Turia and Júcar Rivers were analyzed simultaneously with the developed method. Also, each sample was aliquoted and stored at 4 °C prior to use. The aliquots were spiked with a mixture of targeted compounds at different levels, covering the different analytical working ranges, and determined directly without previous extraction. Before the analysis, the samples were first conditioned by mixing 9 parts by volume of drinking water with 1 part by volume of 10-fold concentrated PBS-T, pH 7.4. To this mixture was added the antibody-functionalized gold nanoparticle cocktail solution. For the calibration, a stock mixture of standards was prepared accordingly, and diluted to cover the working range of each assay. In the disk, 5 out of the 20 arrays were used for the calibration. Then the immunoassay was performed as described for the multiresidue n-plex assay. Each sample was analyzed in 3 replicates on 3 disks (36 replicate spots measured per analyte). Finally, the disks were read by the DVD drive and the results quantified. For comparison purposes, the spiked samples were analyzed in parallel by high-performance liquid chromatography/tandem mass spectrometry (LC/MS2),35 and by solid-phase extraction gas chromatography/mass spectrometry (SPE-GC/MS).36 The

■

RESULTS AND DISCUSSION Assay Development. The setup of an immuno multiresidue screening method on a compact disk requires the integration of the larger number of single analytes on the assay. In this work, 18 immunoassays are integrated in a disk. To reach this goal, the first task was the selection of the optimal immunoreagents among a set of antibodies and their respectively coating conjugates, tested in a direct format. The chemical structures of the targeted compounds and haptens and the optimal concentrations of the immunoreagents are shown in Tables S1 and S2, respectively. Purified antibodies (polyclonal and monoclonal) were labeled with 5 nm colloidal gold and tested by checkerboard titration in a competitive assay. The optical density of the colloidal gold solutions ranged from 0.01 to 0.4, whereas the coating conjugate concentrations varied from 0.1 to 50 mg/L. Under the optimized conditions, the sensitivity, limit of detection, and working range of each of the 18 single assays were established. As can be seen, different sensitivities are reached (Table S3). Particularly, it is worth mentioning the extremely good sensitivity (low IC50) for atrazine, azoxystrobin, chlorpyrifos, forchlorfenuron, imidacloprid, and sulfasalazine (SSZ) with a limit of detection close to or below 0.1 μg/L, which allows the determination of these analytes under the EU directive regarding drinking water.37 In contrast, 4-nitrophenol, diazinon, endosulfan, malathion, and triclosan show IC50 values above the tens or hundreds of micrograms per liter. The rest of the tested analytes show sensitivities in the low microgram per liter range. Nevertheless, all the compounds were included in the study because this research aims to demonstrate the capabilities of the multianalyte residue screening method. For commercial applications, optimal immunoreagents should be used or obtained. C


Article

Analytical Chemistry

OVA−p6, the one used to obtain the highest sensitive assay in ELISA format for forchlorfenuron.24 Both diuron and forchlorfenuron belong to the family of phenylurea pesticides; however, in forchlorfenuron, the chloropyridine ring makes the difference between this compound and other urea derivatives. Therefore, three different haptens functionalized at different positions were tested (Table S5). The only hapten that was not recognized by the diuron antibody was the hapten s5, which contains the spacer arm ((5methoxy-5-oxopentyl)mercapto group) at the C-2 position of the pyridine ring. In addition, the ureido group, a chemical moiety that usually plays a fundamental role in the molecular recognition of this kind of compound, remains unchanged, and the titer of the forchlorfenuron antibody was high when tested against the OVA−s5 conjugate. The shared reactivity values for haptens m6 and p6 with the diuron antibody were 26% and 34%; thus, they could not be used as coating conjugates for the multiplex determination of both analytes. The assay with the highest sensitivity was obtained using hapten s5 (0.16 μg/L), whereas, for m6 and p6, the results were very similar, giving IC50 values of 0.32 and 0.27 μg/L, respectively. In general, it is known, but tough to understand, that the replacement of the hapten in the coating conjugate allows elimination of the shared reactivity of the antibody; thus, more analytes could be integrated in a multiplex assay, though in some cases the assay loses sensitivity. This means that the multiresidue analysis can be tailored to one’s needs, by adding, replacing, or eliminating the analytes depending on the particular purpose, as long as there are no interferences between other target molecules. Therefore, though it is possible to avoid nonspecific recognition, both the sensitivity and selectivity of the assay limit the development of a multiresidue analysis by the immunological approach, shared reactivity studies being key when designing multiplex immunoassays. Antibody-functionalized nanoparticles against structurally similar compounds require solving key aspects, mainly to integrate the maximum number of selected immunoreagents for multiresidue assay using a heterogeneous pool of both polyclonal and monoclonal antibodies, performing at the best sensitivity (low microgram per liter range) and selectivity. The low solubility of the targets in water must also be considered. Considering these points, we set up a 10-plex assay, including the simultaneous determination of 3-phenoxybenzoic acid, alachlor, atrazine, azoxystrobin, chlorpyrifos, diuron, forchlorfenuron, pyraclostrobin, sulfasalazine, and triclosan. All of them are still used and are frequently present in the aquatic environment. The specific recognition of each antibody only by its own coating conjugate is presented in Figure 2. As is shown, each antibody-functionalized nanoparticle is specific toward its own coating conjugate, showing no reactivity or unspecific recognition with other protein−hapten conjugates. This result is important since a lack of selectivity might produce false positives, limiting its use as a screening methodology. Though the number of possible combinations of a 10-plex assay using 18 antibody-functionalized nanoparticles is huge, the selectivity was the main limitation to integrate a larger number of immunoassays. The 10-plex system was not the only possible multiplex configuration. For example, when pentachlorphenol has to be analyzed, five other targets (alachlor, atrazine, azoxystrobin, forchlorfenuron, and pyraclostrobin) can be included, forming a hexaplex immunoassay. Similarly, a heptaplex can be constructed from such analytes as 3phenoxybenzoic acid, 4-nitrophenol, atrazine, azoxystrobin,

As far as the selectivity is concerned, reactivity studies were performed. The polyclonal nature of most of the used antibodies (4NP, ALA, ATZ, CLP, DZN, END, FTN, IMD, MLT, PBA, PCP, SSZ, TCS, and TPA) and the chemical structures of some haptens that share similar arm spacers made it necessary to study the reactivity of nanogold antibodies against the pool of coating conjugates. For that, 18 different coating conjugates at the optimal concentrations were checkerboard titrated with the pool of specific nanogold-labeled antibodies. The results are shown in Table 1. Overall, 3 of the 18 tested antibody-functionalized nanoparticles (AZB, FCF, and PYS) showed high selectivity, exclusively recognizing their own protein−hapten conjugates. Alachlor, atrazine, and sulfasalazine antibodies were quite specific, showing only shared reactivity for a few other protein−hapten conjugates. Antibodies such as those raised toward pentachlorophenol, diazinon, fenthion, or 4-nitrophenol were nonspecific, presenting a high degree of shared reactivity with other coating haptens. Concretely, the nanogold antibody for 4-nitrophenol (4NP) specifically recognized its own coating conjugate (100%) but also reacted nonspecifically with other haptens to different extents (2-(2,4,5-trichlorophenoxy)propionic acid, 3phenoxybenzoic acid, alachlor, atrazine, chlorpyrifos, endosulfan, fenthion, pentachlorophenol, and triclosan), limiting the sensitivity of each particular immunoassay. This is probably due to the fact that particular polyclonal sera comprise antibodies that recognize a specific moiety of the chemical structure of the target analyte, but an important amount of them react with the carrier protein and the spacer arm. For instance, the spacer arm was common for diazinon, fenthion, and malathion haptens. Thus, coating conjugates share the thiophosphate group, and the aliphatic chain through the carrier protein is attached, so false-positive results when testing in heterologous formats might occur. Though the number of possible n-plex assays is huge, obviously as the number of analytes to be implemented increases, the chances to set up a selective multiresidue assay decreases as nonspecific signals might show up. Therefore, we proposed the hypothesis that using a set of haptens with different arm spacers could be a strategy to palliate the shared reactivity issue. The hypothesis was evaluated by testing goldlabeled antibodies for 4NP and TPA against a panel of four haptens for chlorpyrifos. The chemical structures of the haptens and the reactivity are shown in Table S4. Both 4NP and TPA antibodies recognize the coating conjugates for chlorpyrifos at different extensions. Antibodies for 4NP did not recognize hapten C2, whereas antibodies for TPA showed a nonspecific signal for all tested coating conjugates. Although the antichlorpyrifos antibody showed good titers for all studied protein−hapten conjugates, the best sensitivity was achieved with hapten C4. The aromatic ring was preserved in haptens C2, C3, and C4; therefore, the antichlorpyrifos antibody showed high titers, compared to hapten C1, where the titer was low probably due to the derivatization of the aromatic ring. In conclusion, by changing hapten C4 for hapten C2, the shared reactivity of 4-nitrophenol for chlorpyrifos disappeared, so both haptens could be determined simultaneously. However, the sensitivity of the assay for chlorpyrifos using C2 was very low (mg/L) in comparison with that of the assay using C4 (μg/L). The hypothesis was also tested by studying the influence of three forchlorfenuron haptens with the diuron antibody. This antibody shared reactivity with the protein−hapten conjugate D


Article


requirements for particular multiresidue screening and for onsite monitoring. In particular, the proposed method can be very useful to detect emerging pollutants such as prescription drugs, personal care products, and chemicals used in agriculture and industry. Multiresidue Microimmunoassay. The calibration curves obtained for the simultaneous determination of the 10 integrated immunoassays are shown in Figure 3. The standard curves for the 10 analytes were fitted using the four-parameter logistic equation and are the mean of 20 curves performed on different days and on different disks. The analytical parameters of the 10-plex assay are shown in Table S6. As can be seen, the signals of the assay in the absence of analytes were between 12000 and 16000 au, showing a clear decrease of the signal in the competitive format. The relative standard deviation values along the whole calibration curve were below 10%. It is also worth mentioning that the mixture of antibody-functionalized analytes gave a sensitivity similar to that corresponding to single assays under the same working conditions. This allows combinations of antibodies in different plex configurations to be employed. Besides, the analytical performances of the multiplexed immunoassay are comparable to those obtained with ELISA using the same antibodies, reaching detection limits at the microgram per liter level. The on-time stability of the antibody-functionalized gold nanoparticles was also studied to determine their utility as tracers for direct multi-immunoassay. For that, three batches of each labeled antibody were prepared and stored at room

Figure 2. Images showing the selectivity of antibody-functionalized nanoparticles: 1, alachlor; 2, atrazine; 3, azoxystrobin; 4, chlorpyrifos; 5, diuron; 6, forchlorfenuron; 7, 3-phenoxybenzoic acid; 8, pyraclostrobin; 9, sulfasalazine; 10, triclosan. The positive controls are the spots framed in black squares, and the negative control is the circle in the center of the array.

forchlorfenuron, imidacloprid, and pyraclostrobin. A specific application of the developed multiresidue approach is the massive screening of pollutants present in natural waters. Moreover, the arrangement of the microarray matrixes on the disk could be designed in such a way that all 18 analytes would be analyzed in parallel but not simultaneously within the same array. Also, the methodology has the advantage that it can determine groups of analytes “a la carte”, without the need to turn to an indiscriminate analysis of targets. The presented approach shows great flexibility and the ability to analyze a mixture of compounds of different families, such as insecticides, herbicides, antibiotics, and bacteriocides, which fulfills the

Figure 3. Images of the 10-plex microimmunoassay after the disk is read. The array layout is the same as in Figure 2. The panels show the spot intensity. The concentrations of the mixture of analytes in panels 1, 2, and 3 are 0, 1.0, and 16.0 μg/L, respectively. The plots are the competition curves for the 10-plex assay. E


Article


at 100 μg/L, a slight signal decrease was found for the ALA, ATZ, DIU, PBA, SSZ, and TCS systems ( 130%), whereas only 1 was underestimated (R < 80%). Besides, no false positives were detected when blank samples were analyzed. Most of the pollutants detected are potentially used as chemicals in agriculture and industry, and residues can be found in natural waters. Also, some of them belong to families of compounds that have a direct link to the watch list of the EU. In summary, the proposed methodology might be able to identify emerging and little-known pollutants gathering the appropriate haptens and antibodies.

temperature, and the activity was measured for a period of 24 weeks. The nanogold conjugates maintained the activity for at least 13 weeks (Student’s t test, 0.05 threshold level). Regarding the variability between batches, the signal change was less than 7%, showing good quality assurance in antibody-functionalized nanoparticle production. Considering that the conjugates were stored at room temperature, three months is a good mark showing that the lifetime would be easily enlarged by varying the storage conditions. Regarding the assay quality, two positive controls were included in the array. The first one comprises a mixture of KLH/BSA, and the second was an antirabbit IgG−gold standard solution (see the layout in Figure 2). The role of the KLH/BSA solution was to control the first step of the assay and relied on the ability of the antibodies to specifically recognize the immunogenic carrier proteins. The second positive control aims to give information about the amplification step, this signal being used as an inter- and intradisk internal calibrator. The optimum concentration of the control solutions, based on the signal intensity and reproducibility, was 1.0 mg/L and a 1/800 dilution ratio for KLH/BSA and antirabbit IgG−gold, respectively. To determine the intra- and interdisk relative standard deviation of positive controls, 5 disks were tested, each one with 20 arrays and 3 replicates per array, a total of 300 spots being averaged. The intradisk RSD varied from 5% to 12%, whereas the interdisk RSD ranged from 3% to 7%, indicating their suitability. Note also that the first positive control (BSA/ KLH) differs in intensity between antibody-functionalized nanoparticles, and this is due to the different titers against the carrier protein. A negative control (OVA) was also integrated in the array to corroborate the selectivity of the assays. The negative control was similar in signal to the disk surface background. Recognition of different haptens by a single antibody is not the only problem that has to be faced during multiplex optimization. To fully characterize all systems in terms of their selectivity, it is crucial to calculate cross-reactivities (CRs) between analytes and structurally related compounds. This study was carried out with 5 main similar compounds used as competitors for each of the 10 selected analytes. Table S7 shows the cross-reactivity expressed as a percentage of the IC50 for each analyte. The majority of the analytes show only a small percentage of cross-reactivity toward some structurally similar compounds. The most problematic appeared to be the diuron antibody, which cross-reacts with the studied arylurea herbicides. The binding of the different cross-reactants is probably related to the structure of the respective immunogens. However, this fact could be used as an advantage to detect the presence of the compounds from the same family using only one class-specific antibody. This feature can be exploited to define signatures by the application of statistical tools that can help increase the selectivity of the multiplex assay. On the other hand, the most specific were monoclonal antibodies against AZB and PYS, while the forchlorfenuron antibody presented certain cross-reactivity with thidiazuron. Also, cross-reactivity studies were performed using the analytes corresponding to the 10-plex assay. For that, different concentrations of each analyte (0.1, 1.0, 10, and 100 μg/L) were tested against the other nine to check any possible cross-reactivity between them. First, no cross-reactivity was observed when the concentration of the analytes was within the working ranges. However, when azoxystrobin, forchlorfenuron, and pyraclostrobin were present

■

CONCLUSIONS Antibody-functionalized gold nanoparticles are suitable immunoreagents for developing direct assays for the simultaneous determination of complex pesticide and drug residue mixtures in a multiplex configuration. The developed screening analytical approach presents the capacity to quantify analytes of different natures with high sensitivity and selectivity, serving as a warning system for water quality monitoring. As each antibody can be functionalized individually, there is a possibility of “on-demand” adaptation of the immunoassay for the detection of emerging or little-known pollutants, depending on the particular interests of the end users, the nature of the water pollution, or the legislation requirements, by preparing the appropriate cocktail solutions and working in different multiplexed configurations. The shared reactivity is the limiting factor to design a highplex multiresidue screening assay. According to our results, the use of monoclonal antibodies and different protein−hapten conjugates can be a practical solution to achieve a high-plex immuno multiresidue assay. Though the immuno multiresidue approach cannot compete with standard chromatographic methods in terms of multiresidue capacity, the present methodology may be a very useful alternative for the rapid screening not only of the targeted compounds but also of structurally related compounds. Another advantage is the possibility to determine analytes with specific chemical characteristics that are nonviable by the current reference techniques in a single assay, in a short time and in situ, using the same detection system. Moreover, the presented method can be successfully applied in a mixed assay format, such as coating conjugate and antibody-coated types together, as there is no need to use F


Article


(4) European Commission. The combination effects of chemicals chemical mixtures. Communication from the Commission to the Council, COM/2012/0252 final, Brussels, May 31, 2012. (5) European Commission. Directive 2013/39/EU of the European Parliament and the Council of 12 August 2013 amending Directives 2000/60/EC and 2008/105/EC as regards priority substances in the field of water policy. Off. J. Eur. Union 2013, L226/1−L226/17. (6) Kortenkamp, A.; Faust, M. Int. J. Androl. 2010, 33, 463−472. (7) Huerta, B.; Jakimska, A.; Llorca, M.; Ruhí, A.; Margoutidis, G.; Acuña, V.; Sabater, S.; Rodriguez-Mozaz, S.; Barcelò, D. Talanta 2015, 132, 373−381. (8) Desmet, C.; Blum, L. J.; Marquette, C. A. Anal. Chem. 2012, 84, 10267−10276. (9) Wang, L.; Wang, Y.; Fang, Q. Microchim. Acta 2009, 164, 113− 118. (10) Jornet, D.; Gonzalez-Martínez, M. A.; Maquieira, A.; Puchades, R. Anal. Chem. 2007, 79, 9331−9339. (11) Liu, N.; Gao, Z.; Ma, H.; Su, P.; Ma, X.; Li, X.; Ou, G. Biosens. Bioelectron. 2013, 41, 710−716. (12) Morais, S.; Tortajada-Genaro, L.; Maquieira, Á . Expert Rev. Mol. Diagn. 2014, 14, 773−775. (13) Morais, S.; Tortajada-Genaro, L. A.; Arnandis-Chover, T.; Puchades, R.; Maquieira, A. Anal. Chem. 2009, 81, 5646−5654. (14) Morais, S.; Tamarit-López, J.; Puchades, R.; Maquieira, Á . Environ. Sci. Technol. 2010, 44, 9024−9029. (15) Oubina, A.; Ballesteros, B.; Galve, R.; Barcelo, D.; Marco, M. Anal. Chim. Acta 1999, 387, 255−266. (16) Gascón, J.; Oubiña, A.; Ballesteros, B.; Barceló, D.; Camps, F.; Marco, M.-P.; González-Martínez, M. A.; Morais, S.; Puchades, R.; Maquieira, A. Anal. Chim. Acta 1997, 347, 149. (17) Casino, P.; Morais, S.; Puchades, R.; Maquieira, A. Environ. Sci. Technol. 2001, 35, 4111−4119. (18) Parra, J.; Mercader, J. V.; Agulló, C.; Abad-Somovilla, A.; AbadFuentes, A. Anal. Chim. Acta 2012, 715, 105−112. (19) Brun, E. M.; Garcés-García, M.; Puchades, R.; Maquieira, A. J. Agric. Food Chem. 2005, 53, 9352. (20) Brun, E. M.; Garcés-García, M.; Escuín, E.; Morais, S.; Puchades, R.; Maquieira, Á . Environ. Sci. Technol. 2004, 38, 1115. (21) Schneider, P.; Goodrow, M. H.; Gee, S. J.; Hammock, B. D. J. Agric. Food Chem. 1994, 42, 413. (22) Manclús, J. J.; Abad, A.; Lebedev, M. Y.; Mojarrad, F.; Micková, B.; Mercader, J. V.; Primo, J.; Miranda, M. A.; Montoya, A. J. Agric. Food Chem. 2004, 52, 2776−2784. (23) Brun, E. M.; Garcés-García, M.; Puchades, R.; Maquieira, A. J. Immunol. Methods 2004, 295, 21−35. (24) Suárez-Pantaleón, C.; Mercader, J. V.; Agulló, C.; AbadSomovilla, A.; Abad-Fuentes, A. J. Agric. Food Chem. 2008, 56, 11122− 11131. (25) Lee, J. K.; Ahn, K. C.; Park, O. S.; Kang, S. Y.; Hammock, B. D. J. Agric. Food Chem. 2001, 49, 2159−2167. (26) Brun, E. M.; Garcés-García, M.; Bañuls, M. J.; Gabaldón, J. A.; Puchades, R.; Maquieira, A. Environ. Sci. Technol. 2005, 39, 2786− 2794. (27) Noguera, P.; Maquieira, A.; Puchades, R.; Brunet, E.; Carramolino, M. M.; Rodríguez-Ubis, J. C. J. Environ. Monit. 2002, 4, 442−448. (28) Shan, G.; Wengatz, I.; Stoutamire, D. W.; Gee, S. J.; Hammock, B. D. Chem. Res. Toxicol. 1999, 12, 1033−1041. (29) Pastor-Navarro, N.; Gallego-Iglesias, E.; Maquieira, A.; Puchades, R. Anal. Chim. Acta 2007, 583, 377−383. (30) Mercader, J. V.; Suárez-Pantaleón, C.; Agulló, C.; AbadSomovilla, A.; Abad-Fuentes, A. J. Agric. Food Chem. 2008, 56, 7682. (31) Morais, S.; Casino, P.; Marín, M. L.; Puchades, R.; Maquieira, A. Anal. Bioanal. Chem. 2002, 374, 262−268. (32) Brun, E. M.; Bonet, E.; Puchades, R.; Maquieira, A. Environ. Sci. Technol. 2008, 42, 1665−1672. (33) Langone, J. J.; Vanvunakis, H. Methods Enzymol. 1982, 84, 628− 640.

labeled secondary detection antibodies. All these features make our development highly versatile and scalable to different scenarios. The ability of the herein described immuno multiresidue methodology to determine priority water contaminants with a good analytical outcome together with the speed of analysis and the small sample and reagent volume requirement makes this system a good approach for a fast and reliable field test for qualitative and quantitative screening of water samples.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b02354. HPLC/MS−GC/MS determination and tables giving the chemical structures of the analytes and haptens used in this work, optimal coating conjugate concentrations and gold-labeled antibody dilutions, LOD, IC50, and WR for the single assays, reactivity for gold-labeled 4NP, TPA, and CLP antibodies, chemical hapten structures for forchlorfenuron, LOD, IC50, WR, slope, and r2 for the 10-plex assay, cross-reactivity values for structurally related compounds in the 10-plex immunoassay, and results obtained for the immuno multiresidue and chromatographic analysis of spiked water samples (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Spanish Ministry of Economy and Competitiveness (Project CTQ2013-45875-R) and by the Generalitat Valenciana (Grants PROMETEO II 2014/040 and ACOMP 2012/158). P.D. acknowledges a fellowship (Grant Grisoliá 2011/010) funded by the Generalitat Valenciana for her Ph.D. studies. Dr. A. Montoya, Dr. A. Abad-Fuentes, Dr. Shirley Gee, and Prof. Bruce D. Hammock kindly provided some of the coating conjugates and antibodies used in this work.

■

REFERENCES

(1) Ginebreda, A.; Kuzmanovic, M.; Guasch, H.; de Alda, M. L.; López-Doval, J. C.; Muñoz, I.; Ricart, M.; Romaní, A. M.; Sabater, S.; Barceló, D. Sci. Total Environ. 2014, 468−469, 715−723. (2) Altenburger, R.; Ait-Aissa, S.; Antczak, P.; Backhaus, T.; Barceló, D.; Seiler, T.-B.; Brion, F.; Busch, W.; Chipman, K.; de Alda, M. L.; de Aragão Umbuzeiro, G.; Escher, B. I.; Falciani, F.; Faust, M.; Focks, A.; Hilscherova, K.; Hollender, J.; Hollert, H.; Jäger, F.; Jahnke, A.; Kortenkamp, A.; Krauss, M.; Lemkine, G. F.; Munthe, J.; Neumann, S.; Schymanski, E. L.; Scrimshaw, M.; Segner, H.; Slobodnik, J.; Smedes, F.; Kughathas, S.; Teodorovic, I.; Tindall, A. J.; Tollefsen, K. E.; Walz, K.-H.; Williams, T. D.; Van den Brink, P. J.; van Gils, J.; Vrana, B.; Zhang, X.; Brack, W. Sci. Total Environ. 2015, 512−513, 540−551. (3) Schwarzenbach, R. P.; Escher, B. I.; Fenner, K.; Hofstetter, T. B.; Johnson, C. A.; von Gunten, U.; Wehrli, B. Science 2006, 313, 1072− 1077. G


Article

Analytical Chemistry (34) Dobosz, P.; Morais, S.; Puchades, R.; Maquieira, A. Biosens. Bioelectron. 2015, 69, 294−300. (35) Kaufmann, A.; Roth, S.; Ryser, B.; Widmer, M. J. AOAC Int. 2001, 85, 853−860. (36) Sabik, H.; Jeannot, R.; Rondeau, B. J. Chromatogr., A 2000, 885, 217−236. (37) European Commission. Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption, 1998.

H


Massive Immuno Multiresidue Screening of Water Pollutants

Recommend Documents