Anal. Chem. 2001, 73, 1889-1895
One-Step Competitive Immunoassay for Cadmium Ions: Development and Validation for Environmental Water Samples Ibrahim A. Darwish and Diane A. Blake*
Tulane University Health Sciences Center and the TulanesXavier Center for Bioenvironmental Research, New Orleans, Louisiana 70112
A rapid, simple, and reliable competitive immunoassay was developed and validated for measurement of Cd(II) in environmental water samples. This assay employed a monoclonal antibody that recognizes Cd(II)-EDTA complexes as capture reagent and a Cd(II)-EDTA conjugate of horseradish peroxidase as an enzyme label. The assay depended on a competitive binding reaction between the enzyme conjugate and Cd(II)-EDTA complexes, derived from the environmental water sample, for the binding sites of the immobilized antibody. The concentration of Cd(II) in the sample was quantified by the ability of its EDTA complexes to inhibit the binding of the enzyme conjugate to the antibody and, subsequently, color formation in the assay. The assay was specific to Cd(II), with a limit of detection of 0.3 ppb. Ca(II), Mg(II), and Fe(III), the metal ions commonly found in ambient water at relatively high concentrations, did not interfere with the assay. Mean analytical recovery of added Cd(II) was 100.29 ( 3.60. The precision of the assay was satisfactory; coefficients of variation were 3.6-10.9 and 4.81-10.21% for intraand interassay precision, respectively. The assay compared favorably with graphite furnace atomic absorption spectroscopy in its ability to accurately measure Cd(II) spiked into water samples from a Louisiana bayou. Water contamination by toxic metal ions, including cadmium, has become a major problem throughout the world.1 Cadmium is a persistent, potentially insidiously toxic metal that is ubiquitous in the general environment.2 Humans are exposed to cadmium mainly by ingestion of cadmium-contaminated food, dust, and soil and by inhalation of cadmium-containing dusts. After ingestion or inhalation of cadmium, it accumulates in the kidney, liver, lungs, and gastrointestinal tract where it can cause progressively toxic effects, including cancer and renal damage.3-5 Reduction of human exposure to cadmium is one of the main objectives of hygienists, public authorities, and the World Health * Corresponding author: (tel) 504-584-2478; (fax) 504-584-2684; (e-mail)
[email protected]. (1) Stumm, W.; Morgan, J. Aquatic Chemistry, 3rd ed.; Wiley-Interscience: New York, 1996. (2) Nriagu, J. O.; Pacyna, J. M. Nature 1988, 333, 134-39. (3) Yamada, H.; Miyahara, T.; Sasaki, Y. F. Mutat. Res. 1993, 302, 137-45. (4) Friberg, L.; Kjellstrom, T.; Nordberg, G. F. Handbook of the Toxicology of Metals; Friberg, L., Nordberg, G. F., Vouk, V. B., Eds.; Elsevier: Amsterdam, 1986; Vol 2, pp 130-43. 10.1021/ac0012905 CCC: $20.00 Published on Web 03/17/2001
© 2001 American Chemical Society
Organization. The availability of appropriate analytical techniques that could provide near real time data on cadmium levels in environmental samples could ultimately reduce the incidence of cadmium intoxication in humans. The U.S. Environmental Protection Agency (U.S. EPA) continues to investigate new technologies to improve existing analytical methodology for determination of cadmium as an environment contaminant. Many aspects of a new method must be considered to determine its potential role for detection of cadmium-contaminated water. These considerations include the following: sensitivity, simplicity, accuracy, precision, dynamic range of concentration, speed, reliability, utility, cost, and field portability.6 Graphite furnace atomic absorption spectroscopy,7 inductively coupled plasma emission spectroscopy,8 X-ray fluorescence spectroscopy,9 and stripping potentiometry10 are the EPA methods of choice for the elemental analysis of water samples. These techniques accurately measure the cadmium level in a sample; however, the analysis is expensive and sample preparation usually requires acid digestion at elevated temperatures and pressures. In addition, these instrumental methods provide no information about metal oxidation state and the sample turnaround time is relatively slow. Immunoassays offer an alternative approach, and they have significant advantages over the traditional instrument-intensive methods of metals analysis. They are remarkably quick, easily performed, reasonably portable to the contamination site, require minimum sample pretreatment, and have high throughput. Furthermore, studies have shown that the use of immunoassays can reduce analysis costs by 50% or more.11 In the environmental (5) Friberg, L.; Elinder, C. G.; Kjellstrom, T.; Nordberg, G. F. Cadmium and Health: A Toxicological and Epidemiological Appraisal; CRC Press: Boca Raton, FL, 1986. (6) Pyle, S. M.; Nocerino, J. M.; Deming, S. N.; Palasota, J. A.; Palasota, J. M.; Miller, E. L.; Hillman, D. C.; Kuharic, C. A.; Cole, W. H.; Fitzpatrick, P. M.; Watson, M. A.; Nichols, K. Environ. Sci. Technol. 1996, 30, 204-13. (7) Tsalev, D. L. Atomic Absorption Spectroscopy in Occupational and Environmental Health Practice; CRC Press: Boca Raton, FL, 1984. (8) Stoeppler, M. In Biological Monitoring of Toxic Metals; Clarkson, T. W., Friberg, L., Nordberg, G. F., Sager, P. R., Eds.; Plenum Press: New York, 1988; pp 481-97. (9) Skoog, D. A. Principals of Instrumental Analysis, 3rd ed.; Saunders: New York, 1985. (10) Bard, J. A.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; Wiley: New York, 1980. (11) Szurdoki, F.; Jaeger, L.; Harris, A.; Kido, H.; Wengatz, I.; Goodrow, M. H.; Szekacs, A.; Wortberg, M.; Zheng, J.; Stoutamire, D. W.; Sanborn, J. R.; Gilman, S. D.; Jones, A. D.; Gee, S. J.; Choudary, P. V.; Hammock, B. D. J. Environ. Sci. Health B 1996, 31, 451-8.
Analytical Chemistry, Vol. 73, No. 8, April 15, 2001 1889
field, immunoassays are now available to measure a limited number of contaminants, including industrial pollutants,12-15 pesticides,16,17 and herbicides.18-20 Most of the commercial immunoassays for environmental contaminants are directed toward halogenated or aromatic contaminants;21 however, this technique is theoretically applicable to any pollutant, including a heavy metal, if a suitable antibody can be generated. Our laboratory has previously demonstrated the ability to generate monoclonal antibodies that recognize metal ions 22,23 and described a competitive immunoassay for cadmium ions that could be performed in 4-6 h with a limit of detection of ∼7 ppb.24 This assay was performed in a two-step format; a secondary enzymelabeled anti-immunoglobulin antibody was used for detection. The present study describes the development and validation of a significantly improved, one-step immunoassay that detects cadmium ions at concentrations as low as 0.3 ppb in environmental water samples. EXPERIMENTAL SECTION Reagents. Horseradish peroxidase (HRP) enzyme (EC 1.11.1.7, type X), protease-free bovine serum albumin (BSA), and 2,4,6trinitrobenzenesulfonic acid were purchased from Sigma Chemical Co. (St. Louis, MO). 1-(4-Isothiocyanobenzyl)ethylenediamineN,N,N′,N′-tetraacetic acid (ITCBE) was purchased from Dojindo Laboratories (Gaithersburg, MD). Cadmium foil (99.999%) was obtained from Aldrich Chemical Co. (Milwaukee, WI). Atomic absorption spectroscopy standard metals (1000 ppm in 2% HNO3) were obtained from Perkin-Elmer Corp. (Norwalk, CT). 3,3′,5,5′Tetramethylbenzidine peroxidase substrate (TMB Microwell substrate) was from Kirkegaard-Perry Laboratories (Gaithersburg, MD). ELISA high-binding microwell plates were a product of Corning/Costar, Inc. (Cambridge, MA). All water was purified by filtration through a Nanopure II water purification system (Barnstead/Thermolyne, Dubuque, IA). Metal-free disposable pipet tips were a product of Oxford Labware, Inc. (St. Louis, MO). All glassware was mixed-acid washed and liberally rinsed with purified water, and all plasticware was soaked overnight in 3 M HCl and rinsed liberally with purified water before use. (12) Donnelly, J. R.; Grange, A. H.; Herron, N. R.; Nichol, G. R.; Jeter, J. L.; White, R. J.; Brumley, W. C.; Van Emon, J. J. AOAC Int. 1996, 79, 95361. (13) Chiu, Y. W.; Carlson, R. E.; Marcus, K. L.; Karu, A. E. Anal. Chem. 1995, 67, 3829-39. (14) Besarati, N. A.; Van Straaten, H. W.; Kleinjans, J. C.; Van Schooten, F. J. Mutat. Res. 2000, 468, 125-35. (15) Li, K.; Chen, R.; Zhao, B.; Liu, M.; Karu, A. E.; Roberts, V. A.; Li, Q. X. Anal. Chem. 1999, 71, 302-9. (16) Abad, A.; Moreno, M. J.; Pelegri, R., Martinez, M. I.; Saez, A. Gamon, M.; Montoya, A. J. Chromatogr., A 1999, 833, 3-12. (17) Aherne, G. W. Sci. Total Environ. 1993, 135, 73-9. (18) Bruun, L.; Koch, C.; Pedersen, B.; Jakobsen, M. H.; Aamand, J. J. Immunol. Methods 2000, 240, 133-42. (19) Lyubimov, A. V.; Garry, V. F.; Carlson, R. E.; Barr, D. B.; Baker, S. E. J. Lab. Clin. Med. 2000, 136, 116-24. (20) Yazynina, E. V.; Zherdev, A. V.; Dzantiev, B. B.; Izumrudov, V. A.; Gee, S. J.; Hammock, B. D. Anal. Chem. 1999, 71, 3538-43. (21) Van Emon, J. M.; Gerlach, C. L.; Bowman, K. J. Chromatogr., B: Biomed. Sci. Appl. 1998, 715, 211-28. (22) Blake, D. A.; Chakrabarti, P.; Khosraviani, M.; Hatcher, F. M.; Westhoff, C. M.; Goebel, P.; Wylie, D. E.; Blake, R. C., II. J. Biol. Chem. 1996, 271, 27677-85. (23) Khosraviani, M.; Blake, R. C., II.; Pavlov, A. R.; Lorbach, S. C.; Yu, H.; Delehanty, J. B.; Brechbiel, M. W.; Blake, D. A. Bioconjugute Chem. 2000, 11, 267-77. (24) Khosraviani, M.; Pavlov, A. R.; Flowers, G. C.; Blake, D. A. Environ. Sci. Technol. 1998, 32, 137-42.
1890
Analytical Chemistry, Vol. 73, No. 8, April 15, 2001
Antibody and Enzyme Conjugate. The monoclonal antibody (2A81G5) was generated by fusing SP2/0-Ag14 mouse myeloma cells with spleen cells from BALB/c mouse immunized with Cd(II)-EDTA conjugated to keyhole limpet hemocyanin. Our laboratory has previously described the isolation, purification, and characterization of this antibody.22,24 The cadmium horseradish peroxidase enzyme conjugate (Cd(II)-EDTA-HRP) was prepared by a modification of the method previously reported by our laboratory.25 Briefly, a 53.1 mM solution of ITCBE was prepared in 0.1 M sodium phosphate buffer (pH 9.5), and the concentration was verified by measuring absorbance at 280 nm (molar extinction coefficient 17 000). Cadmium foil was dissolved in warm, ultrapure HCl/HNO3 and diluted to a concentration of 100 mM with purified water. Peroxidase enzyme (5 mg) was mixed in a total volume 2 mL of a solution that contained 50 mM sodium phosphate (pH 9.5), 4 mM ITCBE, and 4 mM cadmium. The pH of the reaction mixture was rapidly adjusted to pH 9.2 by the addition of KOH, and the solution was stirred overnight at 25 °C. Unreacted ITCBE and Cd(II)-ITCBE complex were removed from the enzyme conjugate by buffer exchange using a Centricon-30 filter (Amicon, Inc., Beverly, MA) which had been treated with 100 mM EDTA solution and liberally rinsed with water before use. Protein concentration of the conjugate was determined using BCA reagent (Pierce Chemical Co., Rockford, IL), and the extent of substitution of free amino groups on the enzyme was determined by estimation of free amino groups on unreacted HRP and on HRP subjected to the conjugation procedure.26 The extent of conjugation was 84.3% of the total lysine residues. Determination of Optimum Antibody and Enzyme Conjugate Concentrations. The optimum antibody concentration for coating onto the microwell plates and the best working concentration of the enzyme conjugate were determined by checkerboard titration. Purified 2A81G5 antibody was diluted into HEPESbuffered saline (HBS: 137 mM NaCl, 3 mM KCl, and 10 mM HEPES, pH 7.4) at concentrations of 0.6, 1.25, 2.5, and 5 µg/mL and coated onto microwell plates by incubation at 37 °C for 2 h. The plates were washed with 0.05% Tween 20 in phosphatebuffered saline (PBS: 137 mM NaCl, 3 mM KCl, and 10 mM sodium phosphate, pH 7.4), and the wells were blocked with 3% BSA in HBS by incubation at 37 °C for 1 h. Cd(II)-EDTA-HRP conjugate was serially diluted in HBS through the wells of the plates and allowed to incubate in the microwells at 25 °C for 1 h. After a wash with PBS containing 0.05% Tween 20, TMB Microwell substrate (50 µL) was used for color development. The absorbance of each well was measured in a dual-wavelength mode (450-650 nm) using a Vmax Kinetic Microplate Reader (Molecular Devices, Menlo Park, CA), and the data were transformed to a fourparameter curve using SoftMax software provided with the instrument. Concentrations of antibody and enzyme conjugate that yielded a signal between 0.8 and 1.2 absorbance units were used for further testing of the competitive immunoassay procedures. Competitive Immunoassay Procedures. All procedures were carried out at 25 °C. Soluble atomic absorption grade cadmium (in HBS containing 5 mM EDTA and 1% BSA) was premixed with Cd(II)-EDTA-HRP conjugate (0.1 µg/mL, in HBS). Aliquots (50 µL) of the mixture were added to a microwell (25) Chakrabarti, P.; Hatcher, F. M.; Blake, R. C., II.; Ladd, P. A.; Blake, D. A. Anal. Biochem. 1994, 217, 70-5. (26) Habeeb, A. F. Anal. Biochem. 1966, 14, 328-36.
Figure 1. Schematic diagram of the competitive immunoassay for Cd(II). Signal was plotted versus Cd(II) concentration to generate the curves shown in Figures 4-7.
that had been previously coated with 2.5 µg/mL 2A81G5 antibody and blocked with 3% BSA. After 1-h incubation, the plates were washed and the amount of bound Cd(II)-EDTA-HRP conjugate was quantified using TMB Microwell substrate as described above. The same assay procedures could be performed in HBS amended with 50 mM EDTA and 1% BSA; however, under these conditions, the concentration of the Cd(II)-EDTA-HRP conjugate was increased to 0.5 µg/mL. Collection and Preparation of Environmental Samples. Environmental water samples were collected from Bayou Trepagnier, located ∼22 miles west of metropolitan New Orleans adjacent to the Bonnet Carre’ Spillway. Bayou Trepagnier, over the past 80 years, has received cooling water, process wastewater, and surface runoff from a petroleum refinery and manufacturing complex; its bottom sediments and soils are heavily polluted with heavy metals and with a variety of polyaromatic hydrocarbons.27,28 Bayou Trepagnier was chosen to test the Cd(II) immunoassay because its water chemistry is typical of polluted bayous found in southern Louisiana. Water samples were collected in precleaned gallon polyethylene jugs with polypropylene lids (Cole-Parmer Instrument Co., Vernon Hills, IL) and transported back to the laboratory on ice. Water was filtered through a Whatman 43 filter to remove the coarse particulates. The filtered water was then passed through a 0.45-µm syringe filter (Gelman, Ann Arbor, MI) and stored in precleaned 50 mL centrifuge tubes. These filtration steps were used according to the guidelines of the U.S. EPA for water analysis, to distinguish operationally between “sedimented” and “dissolved” heavy metals in surface and drinking waters.29 Because the assay described herein was designed to measure the “dissolved” heavy metal component, such filtration procedures seemed appropriate. A series of Cd(II)-spiked samples were (27) Koplitz, L.; Flowers, G. C.; McPherson, G.; Clymire, J.; Dowling, J.; Ramirez, S.; Washington, W. Abstracts of Papers, Metal Speciation and Contamination of Surface Waters; Jekyll Island, GA, 1995; p 38. (28) Louisiana Department of Environmental Quality. Impact Assessment of Bayou Trepagnier; Technical Report OWR/02/89/001, Baton Rouge, LA, 1989. (29) Methods and Guidance for the Analysis of Water, Version 2.; Office of Water, U. S. Environmental Protection Agency: Washington, DC, 1983.
prepared in the laboratory in the concentration range 0.63-20 ppb by diluting a Cd(II) standard (1000 ppm in 2% nitric acid) with Bayou water. Tap water from the New Orleans water system was spiked and treated by an identical procedure. The samples were conditioned by addition of a 10% volume of a concentrated buffer solution containing 1.37 M NaCl, 30 mM KCl, 500 mM EDTA, 10% BSA, and 100 mM HEPES, pH 7.4. A 100-µL aliquot of each sample was mixed with 100 µL of Cd(II)-EDTA-HRP enzyme conjugate (0.5 µg/mL in HBS), and 50 µL of the mixture was used for analysis by the immunoassay procedures described above. A standard curve for Cd(II) was obtained by using Cd(II) diluted into HBS containing 50 mM EDTA and 1% BSA by the same procedure on plates of the same series. Data Analysis. Values for IC50 were those that gave the best fit to the following equation:
A ) A0 - {(A0 - A1)[Cd(II)]/(IC50 + [Cd(II)])}
where A is the signal at a definite known concentration of soluble Cd(II), A0 is the signal in the absence of Cd(II), A1 is the signal at a saturating concentration of Cd(II), and IC50 is the Cd(II) concentration that produces a 50% inhibition of the signal. The concentrations of Cd(II) in the spiked samples were then obtained by interpolation on the standard curve. RESULTS AND DISCUSSION This study describes a new, improved format for an enzyme immunoassay that quantifies Cd(II) in environmental water samples. Figure 1 illustrates the general principles of this assay. Microwell plates are coated first with a monoclonal antibody (2A81G5) that recognizes Cd(II)-EDTA complexes. An environmental sample containing Cd(II) is mixed with a molar excess of metal-free EDTA to ensure that all the Cd(II) in the sample is present as an EDTA complex, the form recognized by the antibody. This solution is subsequently mixed with Cd(II)EDTA-HRP conjugate, and the mixture is incubated with the Analytical Chemistry, Vol. 73, No. 8, April 15, 2001
1891
Table 1. IC50 of Cd(II) Immunoassay under Different Conditionsa 2A81G5 antibody 2.5 µg/mL
5 µg/mL
Figure 2. Titration of Cd(II)-EDTA-HRP enzyme conjugate versus 2A81G5 monoclonal antibody. The purified antibody was coated onto the microwells at 5 (b), 2.5 (O), 1.25 ([), and 0.6 (]) µg of protein/ mL. Serial dilutions of the conjugate were allowed to bind to the antibody-coated plate. Signal was generated as described in the Experimental Section. Absorbance values were plotted as a function of conjugate concentration.
immobilized antibody in the microwell. During this incubation, the Cd(II)-EDTA complexes compete with Cd(II)-EDTA-HRP conjugate for binding sites of the immobilized antibody. After removal of unbound reagents, the amount of enzyme conjugate bound to the antibody is determined using a chromogenic substrate. The concentration of Cd(II) in a sample is quantified by the ability of its EDTA complex to inhibit the binding of Cd(II)-EDTA-HRP conjugate to the antibody, and color development is inversely proportional to the concentration of Cd(II) in the original sample. Choice of Antibody and Enzyme Conjugate Concentrations. Enzyme-labeled conjugate of Cd(II)-EDTA was prepared by reacting the isothiocyanato group of Cd(II)-ITCBE with the lysine -amino groups of the peroxidase enzyme. The conjugation reaction did not affect the enzyme activity or the immunoreactivity of the conjugate with the immobilized 2A81G5 antibody as shown in Figure 2. Optimal concentrations of antibody required for coating and the best working concentration of the enzyme conjugate were determined by performing competitive assays using varying concentrations of the conjugate and immobilized antibody. The most sensitive assay, using IC50 as a measure, was obtained when the antibody concentrations were 2.5 and 5 µg/ mL and the enzyme conjugate concentrations were 0.1 and 0.02 µg/mL, respectively (Table 1). Unless otherwise noted, all subsequent experiments were performed using 2.5 µg/mL antibody for coating and 0.1 µg/mL enzyme conjugate for the competition reaction. Optimization of Assay Conditions. N-(2-Hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES) was chosen as the buffer in the present work because of its pKa and negligible metalbinding capacity.30 In HBS buffer, the signal in the absence of Cd(II) was dependent upon EDTA concentration (Figure 3A). This effect is due to the inhibitory effect of EDTA on the enzyme activity of horseradish peroxidase (data not shown), not to a binding interaction between the antibody and metal-free EDTA.22,24 Unless otherwise noted, 5 mM EDTA was used for subsequent experiments. For analysis of environmental samples where the (30) Good, N. E.; Winget, G. D.; Winter, W.; Connolly, T. N.; Izawa, S.; Singh, R. M. M. Biochemistry 1966, 5, 467-77.
1892 Analytical Chemistry, Vol. 73, No. 8, April 15, 2001
Cd(II)-EDTA-HRP (µg/mL)
IC50 for Cd(II) (ppb)
0.1 0.2 0.5 1 0.02 0.05 0.1 0.2
5.91 ( 0.48b 8.43 ( 0.79 12.36 ( 1.30 18.47 ( 5.16 5.15 ( 0.46 9.23 ( 0.64 14.91 ( 0.37 17.57 ( 0.87
a Assays were performed in HBS amended with 5 mM EDTA. IC 50 is the concentration of Cd(II) that inhibited the color formation in the b competitive immunoassay by 50%. Values are mean of duplicate determinations ( standard error.
Figure 3. Effect of EDTA on the Cd(II) immunoassay. (A) Assays were performed at a fixed concentration of the immobilized antibody and Cd(II)-EDTA-HRP conjugate and varying concentrations of metal-free EDTA. (B) Assays were performed in 50 mM metal-free EDTA, a fixed concentration of immobilized antibody, and varying concentrations of Cd(II)-EDTA-HRP conjugate. Determinations were performed in duplicate and data is plotted (SD.
total metal ion concentrations may exceed 5 mM, higher concentrations of EDTA would be required to ensure that all metal ions in the sample would be converted to metal-EDTA complexes. Under these conditions, the assay can also be performed using 50 mM EDTA; however, the concentration of Cd(II)-EDTA-HRP conjugate should be increased to 0.5 µg/mL to keep the signal in the absence of Cd(II) at 0.8-1.2 absorbance units (Figure 3B). In contrast to the previously published immunoassay for Cd(II),24 this assay was independent upon pH from pH 7.0 to 7.6
Figure 4. Effect of pH on assay sensitivity. Competitive immunoassays were performed as described in the Experimental Section using atomic absorption grade Cd(II) diluted into HBS amended with 5 mM EDTA at pH 6.2 (b), 6.6 (O), 7.0 (2), 7.2 (4), 7.4 ([), 7.6 (]), 7.8 (9), and 8.2 (0).
Figure 6. Metal ion specificity of the Cd(II) immunoassay. Competitive immunoassays were performed as described in the Experimental Section using atomic absorption grade metal ions diluted into HBS amended with 5 mM EDTA and 1% BSA. Data are shown for Cd(II) (b), Fe(III) (2), Mg(II) (9), and Ca(II) ([).Determinations were performed in duplicate and values are plotted (SD. Table 2. Metal Ion Specificity of Cd(II) Immunoassay
Figure 5. Effect of BSA on assay sensitivity. Competitive immunoassays were performed as described in the Experimental Section using atomic absorption grade Cd(II) diluted into HBS amended with 5 mM EDTA at pH 7.4 containing 0 (b), 0.02 (O), 0.03 (2), 0.06 (4), 0.13 ([), 0.25 (]), 0.5, (9) and 1% (0) BSA.
(Figure 4). This relatively wide working pH range reduces the restriction that pH of the environmental samples should be carefully adjusted and controlled before analysis. According to previously reported binding data, the 2A81G5 antibody bound to Cd(II)- and Hg(II)-EDTA complexes with approximate equal affinity (equilibrium dissociation constants were 21 and 26 nM, respectively).22 Because of this high binding affinity for Hg(II)-EDTA complexes, cross-reactivity to Hg(II) is expected in the Cd(II) assay. In previous studies, addition of BSA to the assay was used to mask interference by Hg(II).24 This approach depended on two facts: Hg(II) strongly binds to free thiol groups of BSA,31 and the Hg(II)-bound BSA complex is not recognized by the 2A81G5 antibody (unpublished data). Before utilizing this approach in the present experiments, it was necessary to test the effect of BSA on the performance of Cd(II) assay. Figure 5 shows that the performance of the assay was not affected by BSA concentrations up to 1%; therefore, this concentration of BSA was added to the assay buffer for all further testing. Metal Ion Specificity. Because metals are ubiquitous in the environment and the 2A81G5 antibody has been shown to interact with a variety of different metal-EDTA complexes [dissociation constants ranging from 21 nM for Cd(II)-EDTA to 820 µM for Al(III)-EDTA],22 it was necessary to test the ability of metal(31) Alexander, M.; Berthold, H. Anal. Lett. 1998, 31, 1633-50.
metal ion
IC50 (ppb)a
cross-reactivity (%)b
Cd Mn In Co Hg Zn Cu Ni Mg Fe Ca Pb
4.82c 170.32 306.81 459.42 500.00 634.35 1028.82 1496.11 3353.9 16796.16 25630.65 ndd
100c 2.83 1.57 1.05 0.96 0.76 0.47 0.32 0.14 0.029 0.002 nd
a IC 50 is the concentration of metal-EDTA complex that inhibits the color formation in the competitive immunoassay by 50%. b Calculated as IC50 [metal ion]/IC50[Cd(II)] × 100. c Values are mean of duplicate determinations. d nd, not determined because it was more than 200 000 ppb.
EDTA complexes to cross-react in the present one-step competitive immunoassay for Cd(II). Figure 6 and Table 2 show the crossreactivity of different individual metal ions in the assay. These data were obtained using 5 mM EDTA in the assay buffer and based upon duplicate determinations. Similar results have been obtained when 50 mM EDTA was used in the assay (data not shown). The presence of Ca(II), Mg(II), and Fe(III), the metal ions most commonly present at a relatively high concentrations in environmental samples, did not interfere with the Cd(II) assay over all its entire linear working range (Figure 6). The crossreactivity of Hg(II) was reduced from 87.64% (in assay buffer without BSA, data not shown) to less than 1% when the assay buffer was supplemented with 1% BSA (Table 2). The crossreactivity exhibited by any other metal ion tested was less than 3% (Table 2). Calibration Curves and Sensitivity. Two valid calibration curves were generated using atomic absorption grade Cd(II) at a concentrations from 0.0 to 1000 ppb, prepared in HBS amended with 1% BSA and either 5 or 50 mM EDTA (Figure 7). The sensitivity of the assay was determined by identifying the limit of detection, defined as the lowest measurable concentration of Cd(II) that could be distinguishable from zero concentration (3 SD. On the basis of eight replicate measurements, the limits of Analytical Chemistry, Vol. 73, No. 8, April 15, 2001
1893
Table 4. Analytical Recovery of Cd(II) Added to Different Samples recovery (%) added Cd(II) (ppb)
Figure 7. Calibration curves for the Cd(II) immunoassay. Microwells were coated with 2.5 µg/mL purified 2A81G5 antibody. Varying Cd(II) concentrations were diluted into HBS, pH 7.4, amended with 1% BSA and 5 (b) or 50 mM (2) EDTA. The samples were mixed (1:1) with Cd(II)-EDTA-HRP conjugate, and then 50 µL of the mixture was introduced into each microwell. The microwells were further manipulated as described in the Experimental Section. The values plotted are mean (SD of eight determinations.
b
purified water
tap water
environmental watera
0.63 1.25 2.5 5 10 20
100.80 ( 8.60b 101.60 ( 2.51 102.30 ( 0.71 94.04 ( 2.25 117.50 ( 2.76 92.02 ( 0.30
95.24 ( 7.94 96.00 ( 8.00 96.80 ( 12.00 100.60 ( 12.00 102.40 ( 17.00 105.25 ( 16.85
95.79 ( 6.04 96.73 ( 4.77 90.94 ( 3.01 103.80 ( 3.17 108.20 ( 3.01 106.30 ( 1.60
average
101.38 ( 2.90
99.38 ( 12.30
100.29 ( 3.60
a Samples were collected as described in Experimental Section. Values are mean of triplicate determinations ( SD.
Table 5. Comparison of Immunoassay with Atomic Absorption Spectroscopy for Analysis of Environmental Water Samples Spiked with Cd(II)a found Cd(II), ppb
Table 3. Precision of Immunoassay for Cd(II) intraassay, n ) 8
interassay, n ) 8
Cd(II) (ppb)
SD (ppb)
CVa (%)
SD (ppb)
CV (%)
1.25 2.5 5 10 20
0.08 0.09 0.38 0.69 2.18
6.40 3.60 7.60 6.90 10.90
0.06 0.26 0.42 0.80 1.46
4.81 10.21 8.37 8.02 7.39
a
CV is the coefficient of variation.
detection were 0.2 and 0.3 ppb in assay buffers containing 5 and 50 mM EDTA, respectively. Precision and Accuracy. The intra- and interassay precisions were determined at different Cd(II) concentrations (1.25, 2.5, 5, 10, and 20 ppb). The intraassay precision was assessed by analyzing eight replicates of each sample in a single run, and the interassay precision was assessed by analyzing the same sample, as duplicates, in four separate runs. The assay gave satisfactory results over all the tested concentration levels; the coefficients of variations were 3.76-9.12 and 3.31-8.15% for intra- and interassay precision, respectively (Table 3). Accuracy of the method was tested by spike and recovery tests. Various known amounts (0.63, 1.25, 2.5, 5, 10, and 20 ppb) of atomic absorption grade Cd(II) were added to purified, tap, and environmental water samples. Each sample was subsequently analyzed, in triplicate, for Cd(II) content. The mean analytical recovery was calculated as the ratio, expressed as percentage, between the Cd(II) concentration found to the concentration added. As shown in Table 4, a quantitative recovery (101.38, 99.38, and 100.29% for purified water, tap water, and environmental water samples, respectively) of the added Cd(II) was obtained. Thus, the assay was able to accurately measure Cd(II) concentrations in purified water as well as moderately or highly complex sample matrixes. The Cd(II) spiked into the environmental water could form complexes with humic materials or other natural chelators present in the sample, but the high concentration of EDTA (50 mM) employed in the assay appeared sufficient to compete the 1894 Analytical Chemistry, Vol. 73, No. 8, April 15, 2001
b
spiked Cd(II), ppb
immunoassay
atomic absorption
0.63 1.25 2.5 5.0 10 15 20
0.76 ( 1.32 ( 0.16 2.52 ( 0.14 4.19 ( 0.14 10.22 ( 0.14 14.02 ( 0.11 20.83 ( 3.21
0.59 ( 0.07 1.30 ( 0.03 2.66 ( 0.04 5.23 ( 0.05 11.80 ( 0.15 20.08 ( 0.72 21.98 ( 0.14
0.11b
a Samples were collected as described in Experimental Section. Values are mean of triplicate determinations (SD.
Cd(II) away from these natural chelators, as evinced by the good recoveries presented in Table 4. Comparison with Atomic Absorption. In experiments with environmental water samples spiked with 0.63-20 ppb Cd(II), the values found by immunoassay correlated well with the values obtained by atomic absorption spectroscopy as shown in Table 5. Linear regression analysis of the results yielded a linear equation: Y ) 0.189 + 1.16X, r ) 0.98. CONCLUSION A new one-step immunoassay format for determination of Cd(II) in environmental water samples has been successfully developed and optimized using Cd(II)-EDTA-HRP conjugate as an enzyme label and an immobilized monoclonal antibody as a capture reagent. This format is superior to the previously reported two-step format in sensitivity, simplicity, and resistance to pH changes. Only one incubation step is required, after which the plate is ready to generate the signal. Since the assay produces a colored readout, only a colorimetric plate reader is required. The entire protocol of the present assay is very easy to perform in a 96-well plate and permits an operator to analyze a batch of 20 samples, in triplicate, and obtain the results of analysis in less than 2 h when the plate has been previously coated with 2A81G5 antibody and blocked with BSA. In addition, the assay exhibits excellent sensitivity, with the capability to determine Cd(II) in environmental water samples at concentrations as low as 0.3 ppb. This high sensitivity will enable us to validate this system for detection of Cd(II) in human serum
samples, and studies are in progress to adapt this method for assessment of human intoxication with Cd(II). The format is adaptable, in principle, to a wide variety of antibody-antigen systems. To our knowledge, this assay represents the first adaptation of the antigen-labeled immunoassay technique for the analysis of heavy metals. ACKNOWLEDGMENT Supported by grants from the Department of Energy (DEFG02-98ER62704) and the Center for Disease Control (DHHS/
CDC-R04/CCR419466-01) to the TulanesXavier Center for Bioenvironmental Research. The authors thank Dr. George C. Flowers, Department of Geology, Tulane University for collection of Bayou Trepagnier water samples and Mr. Sean Samuel, Tulane Central Instrument Facility, for atomic absorption analysis.
Received for review November 2, 2000. Accepted February 5, 2001. AC0012905
Analytical Chemistry, Vol. 73, No. 8, April 15, 2001
1895
