A new strategy for the development of monoclonal antibodies for the determination of human procalcitonin in serum samples.

Procalcitonin (PCT)-a diagnostic serum parameter for bacterial infection and sepsis-is of great interest in the field of biosensors for point-of-care testing. Its detection needs specific biological recognition elements, such as antibodies. Herein, we describe the development and characterization of rat monoclonal antibodies (mAbs) for PCT, and their application in enzyme-linked immunosorbent assays (ELISAs) for the determination of PCT in patient serum samples. From about 50 mAbs, two mAbs, CALCA 2F3 and CALCA 4A6, were selected as a pair with high affinity for PCT in sandwich immunoassays. Both mAbs could be used either as capture or as detection mAb. They were Protein G-purified and biotinylated when used as detection mAb. The setup of two sandwich ELISAs with standards of human recombinant (hr) PCT, using either CALCA 2F3 (assay A) or CALCA 4A6 (assay B) as capture mAbs and the biotinylated mAbs CALCA 4A6 or CALCA 2F3, respectively, as detection mAbs, led to highly specific determinations of PCT without cross-reactivity to calcitonin and katacalcin. Test midpoints (IC(50)) of both assays were determined for hrPCT standards in 4% (w/v) human serum albumin and found with 2.5 (assay A) and 2.7 µg L(-1) (assay B). With both sandwich ELISAs a collection of eight patient serum samples have been determined in comparison to the determination by the Elecsys BRAHMS PCT assay. Good correlations between our prototype ELISAs and the BRAHMS assay could be demonstrated (R (2): assay A, 0.996 and assay B, 0.990). The use of these newly developed anti-PCT mAbs should find broad applications in immunosensors for point-of-care diagnostics of sepsis and systemic inflammation processes.

Multi-parameter determination of TNFα, PCT and CRP for point-of-care testing.

A proof of concept for the simultaneous multi-parameter determination of three inflammation and sepsis parameters-TNFα, PCT and CRP-using a compact optical immunosensor is demonstrated. Harmonized assay conditions revealed standard curves with test midpoints (IC(50)) of 380 µg L(-1) for TNFα, 2300 µg L(-1) for PCT, and 2645 µg L(-1) for CRP.

Detection of quorum sensing molecules in Burkholderia cepacia culture supernatants with enzyme-linked immunosorbent assays.

The Burkholderia cepacia complex (Bcc) employs a quorum sensing (QS) mechanism which is a cell density-dependent bacterial communication system to regulate certain gene expressions. As with many other Gram-negative bacteria, Burkholderia cepacia species use (N-acyl-)homoserine lactones (AHLs or HSLs) as signalling molecules. Because of the essential role of QS in bacterial behavior, the aim of this study was to demonstrate the applicability of our in-house-developed enzyme-linked immunosorbent assays (ELISAs) for the detection of bacterial activities via HSLs in B. cepacia strain LA3 culture supernatants. For this purpose the previously developed monoclonal antibodies (mAbs) HSL1/2-2C10 and HSL1/2-4H5 were exploited. N-3-Oxo-decanoyl-L-homoserine lactone (3-oxo-C10-HSL) was used as main analyte throughout all experiments. With the bacterial culture medium (named ABC medium) a matrix effect in both ELISAs was visible (slight increase in optical density, shift in test midpoints (IC(50)) and working ranges). For example, ELISA with mAb HSL1/2-2C10 and enzyme tracer HSL3-HRP (HSL derivative conjugated to horseradish peroxidase) had an IC(50) of 120 µg L(-1) for 3-oxo-C10-HSL in phosphate-buffered saline versus 372 µg L(-1) in ABC medium. A significant increase of HSLs in B. cepacia strain LA3 culture supernatants after 12 h to 48 h of growth was observed. Although the analytical result of these immunoassays cannot distinguish HSLs from homoserines (HSs), the appearance of these compounds can be easily followed. Hydrolysis and spiking experiments were carried out with these biological samples. According to our knowledge, these are the first immunoassays for the detection of quorum sensing molecules in biological culture supernatants. This study provides a cost-effective, fast, and sensitive analytical method for detection of HSLs/HSs in biological samples without complex sample preparation and will offer a quick idea about B. cepacia activities. The low sample amount requirement (less than 1 mL) constitutes a tremendous advantage for many analytical questions with biological samples.

Development and characterization of rat monoclonal antibodies for N-acylated homoserine lactones.

Quorum sensing (QS) is a communication mechanism between bacteria using diffusible chemical signaling molecules, which are called autoinducers (AI). By detecting the concentration of quorum sensing molecules through binding to a specific receptor protein, bacteria regulate their gene expressions when the concentration of autoinducers and thus the cell density reaches a threshold level. Many Gram-negative bacteria use acylated homoserine lactones (HSLs) as autoinducers. Because of the broad biological functions of HSLs, interest in detection and analysis of HSLs is increasing with a view to their medical, biotechnological, and agricultural applications. In this study, an anti-HSL antibody-based immunochemical detection method has been developed. Four structurally distinct HSL haptens, named HSL1, HSL2, HSL3, and HSL4, have been designed for antibody and assay development. New rat anti-HSL monoclonal antibodies (mAbs) have been produced in-house and characterized with enzyme-linked immunosorbent assays (ELISA), both in the coating antigen and in the enzyme tracer format. Eight mAbs (HSL1-1A5, HSL1-8E1, HSL1/2-2C10, HSL1/2-4H5, HSL4-4C9, HSL4-5E12, HSL4-5H3, and HSL4-6D3) will be presented in this paper. We demonstrate that the anti-HSL mAbs have distinguished sensitivity and selectivity toward HSLs depending upon their chemical structures. The optimized assays are capable of detecting HSLs in the microgram per liter (low micromolar to nanomolar) range. The best IC(50) (test midpoint) was 134 ± 30 µg L(-1) (n = 54) for N-(3-oxodecanoyl)-L-homoserine lactone (3-oxo-C10-HSL) using mAb HSL1/2-2C10 and HSL1-HRP in the enzyme tracer format. In the coating antigen format, the most selective mAb for N-octanoyl-L-homoserine lactone (C8-HSL) was mAb HSL4-4C9. Additionally, anti-HSL mAbs showed higher sensitivity against hydrolyzed HSLs, namely homoserines. These compounds might also occur under certain biological conditions. This study marks the beginning of new ways for quick and cost-effective HSL detection, requiring small sample amounts (less than 1 mL) and little to no sample preparation.

Analysis of DDT isomers with enzyme-linked immunosorbent assay and optical immunosensor based on rat monoclonal antibodies as biological recognition elements.

New rat monoclonal antibodies (mAbs) for DDT [1,1,1 -trichloro-2,2-bis (4-chlorophenyl) ethane], namely DDT 7C12, DDT 1C1, and DDT 1B2, were developed, characterized, and applied in ELISA both in coating antigen and in enzyme-tracer format. The latter used horseradish peroxidase (HRP) or glucose oxidase as enzymes. The lowest concentration of p,p'-DDT was determined with mAb DDT 7C12 and DDT-hapten HRP, with a test midpoint (IC50) of 0.5 +/- 0.2 microg/L (n=10) in 40 mM PBS (phosphate-buffered saline). The mouse anti-rat immunoglobulin lambda-light chain mAb LA1B12 was used as capture mAb. The best IC50 for o,p'-DDT in 40 mM PBS was 1.0 +/- 0.3 microg/L (n=12) and was obtained with mAb DDT 1C1 and DDT-hapten HRP, whereas mAb DDT 1B2 was very selective for p,p'-DDT with an IC50 of 4.2 + 1.6 microg/L (in 40 mM PBS, n=9). An optical immunosensor was optimized and applied for the analysis of DDT (or DDT equivalents). This immunosensor consists of a bench-top optical readout device and disposable sensor chips, which include the fluidic system. Evanescent field excitation and emission of the fluorophore Oyster-645 was used. An IC50 for p,p'-DDT [in 5% (v/v) isopropanol in 40 mM PBS] of 4 microg/L was obtained using DDT 7C12-Oyster-645. ELISA and immunosensor were used for the analysis of p,p'-DDT in unspiked and spiked surface water samples. Within the working ranges of these immunotechniques, recoveries ranged from 80 to 120%.

Enzyme-linked immunosorbent assays for the sensitive analysis of 2,4-dinitroaniline and 2,6-dinitroaniline in water and soil.

The development and characterization of one rat monoclonal antibody (mAb) for 2,4-dinitroaniline and of two rat mAbs for 2,6-dinitroaniline are described. With the immunization of rats with 2,4,6-trinitrophenyl-glycylglycine-keyhole limpet hemocyanine (KLH) conjugate one mAb (PK 5H6) has been developed and formatted into a competitive enzyme-linked immunosorbent assay (ELISA). This assay no. 1 is very sensitive for 2,4-dinitroaniline with a test midpoint of 0.24 +/- 0.06 mug L(-1) (n = 19) in 40 mM phosphate-buffered saline (PBS). A second hapten, 3-(4-amino-2,6-dinitrophenyl)propionic acid, which was also conjugated to KLH and used for the immunization of rats, led to two sensitive ELISAs for 2,6-dinitroaniline in 40 mM PBS with test midpoints of 0.61 +/- 0.08 mug L(-1) (n = 15; mAb DNT4 3C6; assay no. 2) and 0.94 +/- 0.29 mug L(-1) (n = 17; mAb DNT4 1A7, assay no. 3). Selectivities of all mAbs were checked with more than 20 compounds, including nitroaromatic compounds, 2,6-dinitroaniline pesticides, and other substituted derivatives of aniline. As very noticeable cross-reactivities, all mAbs recognize 2-chloro-4,6-dinitroaniline, 4-chloro-2,6-dinitroaniline and 2-bromo-4,6-dinitroaniline, the last of these being a major metabolite of the azo dye Disperse Blue 79. As first demonstrations of applications, two ELISAs (assays no. 1 and 2) were used for the analysis of 2,4- or 2,6-dinitroaniline in spiked water and soil samples. Recovery data were determined and the majority of these data were in the range of 90-120%. These assays can contribute to a very cost-effective and environmentally friendly immunochemical surveillance monitoring of environmental samples for contaminations with these compounds. To the best of the authors' knowledge, these are the first antibodies described for 2,4-dinitroaniline and for 2,6-dinitroaniline.

Development and characterization of new rat monoclonal antibodies for procalcitonin.

The development of selective and sensitive biological recognition elements, e.g., antibodies, for the detection of relevant blood markers is a great challenge in the field of biosensors. In this context, five new rat monoclonal antibodies (mAbs) for procalcitonin (PCT), a marker for bacterial infection and sepsis, were developed and characterized. One mAb, PROC1 3G3, was used as capture antibody. Four mAbs, PROC4 6C6, PROC4 6B2, PROC4 1G3, and PROC4 1D6, were used as detection mAbs, either as Protein G-purified or as biotinylated mAbs. A surface plasmon resonance (SPR) biosensor was used to characterize the antigen-antibody biomolecular interactions. The capture mAb (PROC1 3G3) has an equilibrium dissociation constant (K (D)) of 3.42 x 10(-8) M. All four detection mAbs (PROC4 6C6, PROC4 6B2, PROC4 1G3, and PROC4 1D6) are of high affinity (K (A) = 2.81-6.11 x 10(8) M(-1); K (D) = 1.64-3.56 x 10(-9) M) and have moderate dissociation rate constants (k (d) = 1.70-2.40 x 10(-3) s(-1)). Four different sandwich enzyme-linked immunosorbent assays (ELISAs) with standards of human recombinant (hr) PCT, using PROC1 3G3 as capture mAb and PROC4 mAbs as detection mAbs, respectively, led to highly specific determinations of PCT without cross-reactivities to calcitonin and katacalcin. The lower limits of quantification (LLOQ) for hrPCT (in 40 mM phosphate-buffered saline (PBS), pH 7.6) with these assays ranged from 2.3 to 12.8 microg L(-1). In addition, sandwich ELISAs were set up with biotinylated PROC4 mAbs, and with hrPCT in 4% human serum albumin (diluted 1:10 in 40 mM PBS, including 1:5 (v/v) LowCross Buffer(R)). The LLOQs of these sandwich assays ranged from 4.1 to 6.0 microg L(-1) and were thus much closer together for the different assays. With the latter assay setup (PROC1 3G3 as capture mAb, PROC4 6C6-biotin as detection mAb) a first collection of five serum samples was determined (healthy volunteers, unspiked, and spiked). Recovery rates for the spiked samples ranged from 98.3 to 115.7%. The newly developed anti-PCT mAbs should find broad applications in immunosensors for point-of-care diagnostics of sepsis and systemic inflammation processes.

Development and application of a sol-gel immunosorbent-based method for the determination of isoproturon in surface water.

An immunosorbent was fabricated by encapsulation of monoclonal anti-isoproturon antibodies in sol-gel matrix. The immunosorbent-based loading, rinsing and eluting processes were optimized. Based on these optimizations, the sol-gel immunosorbent (SG-IS) selectively extracted isoproturon from an artificial mixture of 68 pesticides. In addition to this high selectivity, the SG-IS proved to be reusable. The SG-IS was combined with liquid chromatography-tandem mass spectrometry (LC-MS-MS) to determine isoproturon in surface water, and the linear range was up to 2.2 microg/l with correlation coefficient higher than 0.99 and relative standard deviation (RSD) lower than 5% (n=8). The limit of quantitation (LOQ) for 25-ml surface water sample was 5 ng/l.

On-line immunoaffinity column-liquid chromatography-tandem mass spectrometry method for trace analysis of diuron in wastewater treatment plant effluent sample.

An on-line immunoaffinity column with liquid chromatography/tandem mass spectrometry (IAC-LC-MS/MS) method for the determination of diuron in water matrices was described. This method used a sol-gel immunoaffinity column (20 mm x 4 mm I.D.) for on-line sample cleanup and enrichment, a monolithic analytical column (100 mm x 4.6 mm I.D.) for separation, and a triple quadrupole mass spectrometer for quantitation. The major challenges for the on-line set-up were discussed. The optimized on-line protocol was emphasized by the fact that low limit of quantitation (LOQ) of 1.0 ng/L was achieved with only 2.5-mL sample. In addition, a satisfactory accuracy ( approximately 90% of recovery) and precision (<6% of relative standard deviation) at 50 ng/L concentration were also obtained. Due to the ability of the sol-gel immunoaffinity column to eliminate matrix effect, the on-line IAC-LC-MS/MS analysis method can reliably determine diuron in wastewater treatment plant effluent sample.

A new, versatile field immunosensor for environmental pollutants: development and proof of principle with TNT, diuron, and atrazine.

This paper presents a new, versatile, portable miniaturized flow-injection immunosensor which is designed for field analysis. The temperature-controlled field prototype can run for 6h without external power supply. The bio-recognition element is an analyte-specific antibody immobilized on a gold surface of pyramidal structures inside an exchangeable single-use chip, which hosts also the enzyme-tracer and the sample reservoirs. The competition between the enzyme-tracer and the analyte for the antigen-binding sites of the antibodies yields in the final step a chemiluminescence signal that is inversely proportional to the concentration of analyte in the given range of detection. A proof of principle is shown for nitroaromatics and pesticides. The detection limits (DL; IC20) reached with the field prototype in the laboratory was below 0.1 microg l(-1) for 2,4,6-trinitrotoluene (TNT), and about 0.2 microg l(-1) for diuron and atrazine, respectively. Important aspects in this development were the design of the competition between analyte and enzyme-tracer, the unspecific signal due to unspecific binding and/or luminescence background signal, and the flow pattern inside the chip.

Enzyme-linked immunosorbent assays based on rabbit polyclonal and rat monoclonal antibodies against isoproturon.

This work describes the production and characterization of rabbit polyclonal antisera (pAb) and rat monoclonal antibodies (mAb) against isoproturon. Coating antigen and enzyme-tracer formats were developed. Standard curves for isoproturon were conducted either in 40 mM phosphate buffered saline (PBS) or in Milli-Q water. PAb 352 together with the best enzyme tracer revealed in the optimized ELISA (enzyme tracer format) a test midpoint of 1.06 +/- 0.34 microg/L (n = 19, standard set up in Milli-Q water) with a detection limit of about 0.1 microg/L. The comparable ELISA with mAb IOC 7E1 had test midpoints of 0.07 +/- 0.04 microg/L (n = 7, standards in Milli-Q water) and 0.11 +/- 0.08 microg/L (n = 33; standards in 40 mM PBS). The limits of detection were about 0.003 and 0.01 microg/L in Milli-Q water and PBS, respectively. Noticeable cross reactivities (CRs) were seen with the major metabolites, namely 4-isopropylaniline, 4-isopropylphenylurea, and 1-(4-isopropylphenyl)-3-methylurea. With pAb 352, these CRs were 5%, 7%, and 31%, respectively, and with mAb IOC 7E1, they were 3%, 5%, and ca. 19%, respectively. All arylurea herbicides had only minor CRs, which ranged from no CR (e.g., chlorosulfuron) to a maximum of 3.3% (chlortoluron). Influences of organic solvents (methanol, ethanol, acetonitrile, and acetone) were evaluated. Both pAb- and mAb-based immunoassays showed the highest tolerance for methanol, up to 5%. Ethanol and acetonitrile could not be used above 2% without an influence on the assays. The same was true for acetone, although tested only in the mAb-based assay. Water samples of different origins and matrices were spiked and analyzed with these pAb and mAb ELISAs. The results demonstrated that these immunoassays are useful screening tools.

Extending the working range of immunoanalysis by exploitation of two monoclonal antibodies.

A newly established rat monoclonal antibody (mAb) for isoproturon, namely, IOC 10G7, is described. This mAb shows a standard curve for isoproturon in phosphate-buffered saline with a test midpoint of 5.5 +/- 1.8 microg/L (n = 15). In combination with the formerly developed mAb IOC 7E1, IOC 10G7 can be exploited to extend the working range for the analysis of isoproturon. Both antibodies were formatted into a competitive enzyme-linked immunosorbent assay (ELISA), using the same enzyme-tracer. MAb IOC 7E1 and mAb IOC 10G7 have different affinities for the target compound, but the signal-response curves of the single antibodies overlap. Cross-reactivity (CR) patterns of both antibodies are comparable, showing the highest CR for the metabolite 1-(4-isopropylphenyl)-3-methylurea (IOC 10G7, 9%; IOC 7E1, 19%). The system described here includes the combined, but individual, usage of both assays on one microtiter plate, as well as the strategy for mixing the two antibodies for the utilization in one assay. When standards are performed in Milli-Q water, the working range for isoproturon with the individual ELISAs using mAb IOC 7E1 is from 0.01 to 1 microg/L (test midpoint = 0.11 +/- 0.03 microg/L; n = 17) and with IOC 10G7, it is 1-100 microg/L (test midpoint = 10.3 +/- 1.6 microg/L; n = 32). The working range with mixed antibodies is usually on the order of 0.03-30 microg/L (test midpoint = 0.5 +/- 0.2 microg/L; n = 17). These strategies (mAbs individually and mixed) cover a range of 4 and 3 orders of magnitude, respectively. As a demonstration, water samples of different origins and an extract of mixed sediment were analyzed. The advantages of these strategies are discussed.

A SPR-based immunosensor for the detection of isoproturon.

The proof of principle of a reusable surface plasmon resonance (SPR)-based immunosensor for the monitoring of isoproturon (IPU), a selective and systemic herbicide, is presented. The detecting rat monoclonal anti-isoproturon antibody (mAb IOC 7E1) was reversibly immobilized through the use of a capture mouse anti-rat (kappa-chain) monoclonal antibody (mAb TIB 172), which was covalently immobilized on the sensor chip surface. Such strategy features a controlled binding of the captured detecting antibody as well as facilitates the surface regeneration. The capture of the anti-IPU mAb by the antibody (TIB 172) coated sensor surface could be carried out up to 120 times (immobilization/regeneration cycles) without any evidence of activity loss. With a high test midpoint and a low associated SPR signal, the direct detection format was shown to be unsuitable for the routine analysis of isoproturon. However, the limit of detection (LOD) could be easily enhanced by using a strategy based on a surface competition assay, which improved all immunosensor parameters. Moreover, the sensitivity and working range of the indirect format were found to be dependent on the surface density of the anti-IPU mAb IOC 7E1. As expected for competitive formats, the lowest surface coverage (0.5 ng/mm(2)) allowed a lower detection of the herbicide isoproturon with a calculated LOD of 0.1 microg/l, an IC(50) (50% inhibition) of 5.3+/-0.6 microg/l, and a working range (20-80% inhibition) of 1.3-16.3 microg/l.

How can immunochemical methods contribute to the implementation of the Water Framework Directive?

Immunochemical methods (in particular immunoassays) have been applied to spring and surface water samples, respectively, which were set-up as reference materials (RM) within two proficiency testing campaigns. For the first set of proficiency tests (PTs) described here (which were actually the second round of PTs organized, spring 2005), three ELISAs (enzyme-linked immunosorbent assays) were employed in the enzyme tracer format for isoproturon, diuron, and atrazine, respectively. Results were evaluated in comparison with conventional reference methods (LC, GC). Based on their Z-score laboratory performances, the results for isoproturon and diuron were satisfactory, both for fortified spring water and for the blind solution. The results for atrazine were strongly influenced by other triazines present and needed detailed interpretation. For the second set of PTs described here (which were actually the third round of PTs organized, spring 2006), two ELISAs in the coating antigen format were used for isoproturon and diuron, and the result was included with the results obtained by conventional methods during the PTs. The results (the Z-scores) for isoproturon were again classified as satisfactory, in both fortified surface water and blind solution. The results for diuron in ELISA showed an influence of the water matrix, while the analysis of the blind solution was satisfactory. In addition, an ELISA in the enzyme tracer format was applied to analyze isoproturon, diuron, and atrazine in surface water samples, which had been set-up and spiked during a field trial (tank experiment) at the Maas River at Eijsden, The Netherlands. The immunoassay results were compared with those from an in-house on-line SPE LC/MS-MS used as reference. Although the immunochemical results were sometimes higher than those determined in the reference analysis, the general concentration trends in the samples were similar. The contribution of immunochemical methods to the implementation of the European Water Framework Directive is also discussed.

Comparison of two express immunotechniques with polyelectrolyte carriers, ELISA and FIIAA, for the analysis of atrazine.

Conventional immunoassays on microtiterplates are very useful analytical tools in environmental analysis, but the long assay times, usually in the range of hours, are a drawback. To overcome this disadvantage, the development of fast (express) assay formats is described, which use polyelectrolytes as carriers. Two semi-homogeneous immunochemical methods, namely the polyelectrolyte-ELISA (enzyme-linked immunosorbent assay) and the express-FIIAA (flow injection immunoaffinity analysis) for the analysis of the herbicide atrazine were set-up. Using polyclonal antisera for atrazine, the following results were obtained. Standard curves for atrazine showed a linear range from 3 to 100mugl(-1) in polyelectrolyte-ELISA and 0.3-100mugl(-1) in express-FIIAA. The test midpoints in polyelectrolyte-ELISA and express-FIIAA were 12 and 5mugl(-1), respectively. The duration time of the immunochemical reaction was in both assays 15min, but the total assay time differed (30min (polyelectrolyte-ELISA) and 18min (express-FIIAA)). A significant difference between the formats could be observed in the number of samples that can be determined per day. The polyelectrolyte-ELISA can handle samples in parallel on a microtiterplate (usually 20/plate), whereas in the express-FIIAA the samples are automatically analysed one after another. This first demonstration of these techniques shows the potential of these methods, but also their limitations.

Development of new rat monoclonal antibodies with different selectivities and sensitivities for 2,4,6-trinitrotoluene (TNT) and other nitroaromatic compounds.

Five new rat monoclonal antibodies (mAbs) for 2,4,6-trinitrotoluene (TNT) and other nitroaromatic compounds, including, especially, the metabolite 2-amino-4,6-dinitrotoluene (2-ADNT), are described. Five heterogeneous, competitive enzyme-linked immunosorbent assays (ELISAs) were developed. Assay 1 uses mAb DNT4 3F6 as recognition element and gives a standard curve for TNT in 40 mmol L(-1) phosphate buffered saline (PBS) with a test midpoint (IC50) of 0.26+/-0.08 microg L(-1) (n=20). Assay 2 (mAb DNT4 4G4) has an IC50 of 0.35+/-0.07 microg L(-1) (n=18), assay 3 (mAb DNT4 1A3) has an IC50 of 0.73+/-0.14 microg L(-1) (n=15), and assay 4 (mAb DNT4 1A7) has an IC50 of 2.32+/-0.70 microg L(-1) (n=15). Assay 5 (mAb DNT2 4B4) is very selective for 2-ADNT and has an IC50 of 8.5+/-1.7 microg L(-1) (n=15) in PBS. These antibodies for nitroaromatic compounds differ not only in their sensitivity but also in their selectivity. Major cross-reactants are 1,3,5-trinitrobenzene, 2-ADNT, 4-amino-2,6-dinitrotoluene (4-ADNT), 2,4-dinitroaniline, 3,5-dinitroaniline, and 2,6-dinitroaniline. Although assay 5 is not highly sensitive, the mAb DNT2 4B4 in this assay is highly selective for 2-ADNT. Of all the compounds tested, only 2,4-dinitroaniline and 3,5-dinitroaniline had relevant cross reactivities, 18% and about 26%, respectively. Two ELISAs, using mAbs DNT4 3F6 and DNT2 4B4, were used to analyze different concentrations of TNT and 2-ADNT, respectively, in three different surface water matrices (river and lake water). Both assays were affected by the matrix, but usually performed well (recovery within the range 70-120%). In addition, these ELISAs were used to analyze mixtures of TNT, 2-ADNT, and 4-ADNT, at three different concentrations, in the same water matrices. A different recognition pattern was clearly visible with both assays and depended on the cross reactivities of the corresponding mAb.