Anal. Chem. 2009, 81, 4005–4009
Monoclonal Antibody to Trivalent Chromium Chelate Complex and Its Application to Measurement of the Total Chromium Concentration Kazuhiro Sasaki,* Shinichi Oguma, Yukie Namiki, and Naoya Ohmura Environmental Science Research Laboratory, Central Research Institute of Electric Power Industry, 1646 Abiko, Abiko-Shi, Chiba, Japan 270-1194 Isothiocyanobenzyl group-appended ethylenediamine tetraacetic acid (EDTA) was used to covalently couple Cr(III) · EDTA to keyhole limpet hemocyanin for use as an immunogen. An obtained monoclonal antibody (RD3G4) bound to Cr(III) · EDTA with an equilibrium dissociation constant (Kd) of 9.7 nM, which was 100-fold tighter than the Kds for the other tested EDTA-metal complex. In particular, there was an over 2000-fold affinity difference between Cr(III) · EDTA and Fe(III) · EDTA, although the ion radius of trivalent chromium (0.76 Å) was quite close to that of ferric ion (0.79 Å). Hexavalent chromium could be detected by the antibody after being reduced into trivalent form. An immunoassay format showed an IC50 of 87 nM for hexavalent chromium, with a detection limit of 30 nM (1.6 µg/L). Therefore, the addition of reducing agents to the mixture of tri- and hexavalent chromium allows determination of the total chromium concentration by the immunoassay. Hexavalent chromium could be isolated from trivalent chromium by an anion-exchange column, and thus, the concentration of hexavalent chromium in tri- and hexa- mixture can also be estimated by the immunoassay.
currently regulated under the level of 50∼100 µg/L MCL (maximum contaminant level) for total chromium.3,4 The most frequently used methods in environmental analysis of chromium today are atomic absorption spectroscopy (AAS), inductively coupled plasma atomic emission spectrometry (ICPAES), inductively coupled plasma mass spectroscopy (ICP-MS), and colorimetry with 1,5-diphenylcarbazide.5 To determine the total amount of chromium, AAS, ICP-AES, and ICP-MS methods are sensitive and accurate but they are time-consuming and require sophisticated equipment, generally in a laboratory setting. To determine the concentration of hexavalent chromium, the diphenylcarbazide colorimetric method is simple and fast, but certain substances react with the 1,5-diphenylcarbazide reagent to form a colored product which is absorbed at 520-540 nm that may obscure or interfere with the quantitation of the chromate peak.6 Immunoassays offer a simple, fast, and cost-effective alternative, if suitable antibodies and pretreatment protocols are available or can be developed.7 For example, in order to produce a rapid test kit for determination of the cadmium concentration in foods and agricultural samples, we have developed immunochromatography and pretreatment systems.8-12 Although studies about the generation and properties of monoclonal and recombinant antibodies
Chromium is a transition metal most commonly found in the environment in two distinct forms, trivalent (Cr3+) and hexavalent (Cr6+). The toxicities of the two forms of chromium are vastly different. Trivalent chromium is generally nontoxic, while hexavalent chromium is both toxic and carcinogenic and is considered an environmental pollutant.1 The solubility and negative charge of its more common forms, chromate and dichromate (CrO42- and Cr2O72-), lead to limited adsorption in aquifer minerals and result in the high mobility of Cr6+ in aquifers.2 The contamination of groundwater and soils is a result of its industrial activities, including metal plating, pigment production, and plywood.1 For drinking water, chromium is
(3) U.S. EPA. List of Drinking Water Contaminants & MCLs, EPA, 816-F-03016, http://www.epa.gov/safewater/contaminants/index.html#listmcl. (4) California Department of Public Health. Chromium-6 in Drinking Water: Regulation Update, http://ww2.cdph.ca.gov/CERTLIC/DRINKINGWATER/ Pages/Chromium6.aspx. (5) Ohashi, K. The Newest Analysis Technology for Environmental Measurements; Sakai, T., Oguma, K., Motomizu, S., Eds.; CMC Publishing: Tokyo, Japan, 2005; pp 76-82. (6) California Environmental Protection Agency. Method 425, 1997. (7) Vanderlaan, M.; Stanker, L. H.; Watkins, B. E.; Roberts, D. W. Immunoassays for Trace Chemical Analysis: Monitoring Toxic Chemicals in Humans, Food, and the Environment; American Chemical Society: Washington, DC, 1991. (8) Tawarada, K.; Sasaki, K.; Ohmura, N.; Matsumoto, N.; Saiki, H. Bunseki Kagaku 2003, 52, 583. (9) Sasaki, K.; Tawarada, K.; Okuyama, N.; Kayama, F.; Abe, K.; Okuhata, H.; Maruyama, Y.; Arakane, T.; Miyasaka, H.; Fujikawa, T.; Ohmura, N. Bunseki Kagaku 2007, 56, 29–36. (10) Sasaki, K.; Tawarada, K.; Arakane, T.; Okuyama, A.; Maruyama, Y.; Okuhata, H.; Kayama, F.; Abe, K.; Miyasaka, H.; Fujikawa, T.; Thomas, R. G.; Ohmura, N. Bunseki Kagaku 2008, 57, 105–112. (11) Sasaki, K.; Tawarada, K.; Ohmura, N. Commun. Soil Sci. Plan. 2009, 40, 345-351. (12) Abe, K.; Sakurai, Y.; Okuyama, A.; Sasaki, K.; Tawarada, K. J. Sci. Food Agric. 2009, 89, 1097-1100.
* To whom correspondence should be addressed. E-mail: k-sasaki@ criepi.denken.or.jp. Phone: +81-471-82-1181. Fax: +81-471-83-3347. (1) Shanker, A. K. Mode of Action and Toxicity of Trace Elements. In Trace Elements: Nutritional Benefits, Environmental Contamination, and Health Implications; Prasad, M. N. V., Ed.; John Wiley & Sons: Hoboken, NJ, 2008; pp 537-542. (2) Blowes, D. Science 2002, 295, 2024–2025. 10.1021/ac900419c CCC: $40.75 2009 American Chemical Society Published on Web 04/16/2009
Analytical Chemistry, Vol. 81, No. 10, May 15, 2009
4005
directed toward chelated metal ions have been published,13-21 antibodies directed toward an epitope that includes some form of ionic chromium have not been reported until we produced an antibody Gb3G12 binding to Cr(III) · EDTA with an equilibrium dissociation constant (Kd) of 0.6 µM.22 After the reducing step and the formation of the EDTA complex, 0.1 mg/L of Cr6+ could be detected by the antibody.22 This paper describes the generation and binding properties of a novel monoclonal antibody that exhibited a 62-fold higher affinity for Cr(III) · EDTA than Gb3G12. Intriguingly, it decreased the affinity for Fe(III) · EDTA by over 2000-fold compared with Cr(III) · EDTA, although the difference in the ion radius between trivalent chromium and ferric ion is 0.03 Å. By reduction of hexavalent chromium to its trivalent form, the total chromium concentration of the mixture of tri- and hexavalent chromium can be determined with a detection limit of 1.6 µg of Cr/L using an immunoassay format employing the novel antibody. MATERIALS AND METHODS Materials. BALB/cA Jcl inbred mice were purchased from Clea Japan, Inc. (Tokyo, Japan). 1-(4-Isothiocyanobenzyl)ethylenediamine-N,N,N′,N′-tetraacetic acid (isothiocyanobenzyl-EDTA) was obtained from Dojindo (Kumamoto, Japan). Keyhole limpet hemocyanin (KLH) and Ovalbumin (OVA) were obtained from Sigma-Aldrich (A2512 and H7017; St. Louis, MO). Cr · EDTAprotein conjugates were prepared as described by Darwish and Blake.23 1-(4-Isothiocyanobenzyl) ethylenediamine-N,N,N′,N′-tetraacetic acid (isothiocyanobenzyl-EDTA) was obtained from Dojindo (Kumamoto, Japan). Cy-5 conjugated F(ab′)2 fragment of goat anti mouse IgG (no. 286402) was obtained from Jackson ImmunoResearch (West Grove, PA). The myeloma cell line (NS0) was purchased from The Institute of Physical and Chemical Research Cell Bank (Tsukuba, Japan). Preparation of Protein-Chelate Conjugates. The preparation of protein-chelate complexes was based on Chakrabarti’s report.16 Briefly, 2 mg of protein (KLH or OVA) were dissolved in 3 mL of 100 mM boric acid (pH 9.0) and then mixed with 1 mg of isothiocyanobenzyl-EDTA overnight. During this time, isothiocyanobenzyl-EDTA bound with the amino group of the (13) Reardan, D. T.; Meares, C. F.; Goodwin, D. A.; McTigue, M.; David, G. S.; Stone, M. R.; Leung, J. P.; Bartholomew, R. M.; Frincke, J. M. Nature 1985, 316, 265–268. (14) Gillette, R. W.; Singleton, J.; Janowicz, A.; Gilman, S. C. J. Immunol. Methods 1989, 124, 277. (15) Love, R. A.; Villafranca, J. E.; Aust, R. M.; Nakamura, K. K.; Jue, R. A.; Major, J. G.; Radhakrishnan, R., Jr.; Butler, W. F. J. Mol. Biol. 1993, 32, 10950. (16) Chakrabarti, P.; Hatcher, F. M.; Blake, R. C., II.; Ladd, P. A.; Blake, D. A. Anal. Biochem. 1994, 217, 70–75. (17) 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, 27677–27685. (18) Khosraviani, M.; Blake, R. C., II.; Pavlov, A. R.; Lorbach, S. C.; Yu, H.; Delehanty, J. B.; Brechbiel, M. W.; Blake, D. A. Bioconjugate Chem. 2000, 11, 267–277. (19) Johnson, D. K.; Combs, S. M.; Parsen, J. D.; Jolley, M. E. Environ. Sci. Technol. 2002, 36, 1042–1047. (20) Corneillie, T. M.; Whetstone, P. A.; Fisher, A. J.; Meares, C. F. J. Am. Chem. Soc. 2003, 125, 3436. (21) Blake, R. C., II.; Pavlov, A. R.; Kohsraviani, M.; Ensley, H. E.; Kiefer, G. E.; Yu, H.; Li, X.; Blake, D. A. Bioconjugate Chem. 2004, 15, 1125. (22) Sasaki, K.; Tawarada, K.; Ohmura, N. Environmental Science Research Laboratory Rep. No. V08007; CRIEPI: Tokyo, Japan, 2009. (23) Darwish, I. A.; Blake, D. A. Anal. Chem. 2001, 73, 1889–1895.
4006
Analytical Chemistry, Vol. 81, No. 10, May 15, 2009
protein. Before the conjugation with EDTA, the pH of the solution was adjusted to 7.0 using a desalting column (10DG; Bio-Rad Laboratories; Hercules, CA) to prevent metal precipitation. To make Cr(III) · EDTA-protein conjugate, 80 µL of 10 µM CrCl3 solution was added to 4 mL of 1 mg/mL EDTA-protein solution. Immunization of Mice and Hybridoma Production. Three mice were injected intraperitoneally at 2 week intervals with Cr(III) · EDTA-KLH conjugate emulsified in adjuvant (TiterMax Gold; CytRx; Norcross, GA). The fourth injection was given 7 days after the third injection. Three days after the fourth injection, the reactivity of sera toward 10 µM Cr(III) · EDTA and 10 µM metal free EDTA were analyzed with a KinExA system (see below). Four days after the fourth injection, spleen cells were fused with myeloma cells using polyethylene glycol 1500 (no. 783641; Roche Diagnostics GmbH; Penzberg, Germany). The fused cells were cultured in six 96 well flat-bottomed microculture plates for 14 days. Positive clones were screened by the methods reported previously.24 Hybridoma cells were cloned by colony formation using methylcellulose semisolid medium (ClonaCell-HY, Stemcell Technologies Inc., Vancouver, Canada). Ascitic fluids were produced in mice by injecting hybridoma cells. Antibodies were purified using a Protein A-Superose column kit (Bio-Rad Laboratories; Hercules, CA). The purity of the antibody was confirmed by HPLC (high-performance liquid chromatography) and demonstrated a single peak at a retention time (data not shown). Solid Phase Preparation. Antigen coated plastic beads were prepared as follows. A total of 0.4 g of polymethylmethacrylate (PMMA) beads (Sapidyne Instruments, Inc., Boise, ID) were suspended in 1 mL of Cr(III) · EDTA-OVA conjugate solution (1 mg of protein/mL) and mixed gently overnight. The overlying solution was removed, and the PMMA beads were subsequently blocked against nonspecific binding with BSA solution (1 mg/ mL) by gentle mixing for >2 h. Immunoassay Procedure for KinExA. Antibody assays were performed with KinExA 3000 (Sapidyne Instruments; Boise, ID). The KinExA 3000 consists of a single flow cell located at the focal point of a filter fluorometer. Solid phase material (OVA-Cd · EDTA coated PMMA particles in the present case) was suspended and flowed into the flow cell, where it was trapped against a screen. Assay samples consisted of RD3G4 monoclonal antibody (final concentration of 0.5 nM) mixed with a solution containing metal ions and EDTA in 20 mM Tris (pH 7.0). An equilibrium measurement began with a flow of running buffer, which was 20 mM Tris (pH 7.0) at 0.25 mL/min for 30 s, to establish a baseline. Next, the sample was flowed through the solid phase for 50 s at a rate of 0.6 mL/min. During this period, free antibody present in the sample (if any) accumulated on the solid phase Cr(III) · EDTA. The solid phase was then washed with running buffer for 30 s, followed by a flow of 2 nM Cy-5 labeled anti mouse IgG for 96 s, followed by an additional 30 s of the running buffer, all at 0.25 mL/min. Finally, residual bulk fluorescence (i.e., not attached to the solid phase) was removed by a 90 s flow of running buffer at 1.5 mL/min. The signal difference between the fluorescent signal after the final wash and the initial baseline was calculated and used as a measure of the free antibody in the sample. Kd values were calculated directly from the (24) Sasaki, K.; Glass, T. R.; Ohmura, N. Anal. Chem. 2005, 77, 1933.
Figure 1. The responsivities of sera from mice and the supernatants from hybridoma cell culture. Responses relative to the control were compared on the basis of the results from the KinExA assay with 10 µM Cr(III) · EDTA or 10 µM metal free EDTA. Each control response is the signal obtained from each sample without antigen. (A) Serum reactivity of mice injected with Cr(III) · EDTA-KLH and (B) representative results of hybridoma screening.
measured free antibody, using the KinExA Pro software supplied with the instrument, in the manner described previously.25,26 Reduction of Hexavalent Chromium and Immunoassay. Potassium dichromate (K2Cr2O7), which is a hexavalent chromium, was used for immunoassay with RD3G4 antibody. Mixing 200 µL of potassium dichromate solution and 400 µL of 0.3 M sodium metabisulfite resulted in trivalent chromium in 0.2 M sodium metabisulfite. To form Cr(III) · EDTA, 400 µL of the reduced samples were mixed with 400 µL of 0.25 M Tris buffer (pH 8.0) containing 25 µM EDTA, and then, in order to facilitate the EDTA complex formation, the mixture was incubated at 50 °C for 10 min. To eliminate reductive activity, 400 µL of 1% BSA dissolved in 0.25 M Tris buffer (pH 8.0) was added and mixed. A total of 1000 µL of the mixture was mixed with 500 µL of 1.5 nM RD3G4 antibody dissolved in PBS buffer, and then it was applied to the KinExA assay described above. Column Treatment for Hexavalent Chromium. Potassium dichromate was isolated using a column packed with strong base anion-exchange resin (DIAION NSA100, Mitsubishi Chemical, Tokyo Japan). The resin was washed with 0.1 M HCl repeatedly, and then the resin slurry was transferred to a glass column (12.5 mm diameter × 20 mm height). Prior to passage of the sample, the column was washed with 20 mL of DW. Sample solutions containing potassium dichromate and other metals (if any) passed through the column at a flow rate of 2 mL/min. Following passage of the sample, the column was washed with 5 mL of DW. The adsorbed dichromate was then reduced to 2Cr3+ and eluted from the column using 15 mL of 0.2 M sodium metabisulfite. The solutions collected through the column were stored at 4 °C until they were analyzed using inductively coupled plasma mass spectrometry (ICP-MS, Thermo Elemental X7, Thermo Fisher Scientific, Yokohama, Japan). RESULTS AND DISSCUSSION Immune Response and Hybridoma Screening. Sera from two out of three mice immunized with the Cr(III) · EDTA-KLH conjugate showed reactivity toward Cr(III) · EDTA but not toward (25) Darling, R. J.; Brault, P. A. Assay Drug Dev. Technol. 2004, 2, 647. (26) Ohmura, N.; Lackie, S. J.; Saiki, H. Anal. Chem. 2001, 73, 3392–3399.
metal free EDTA (Figure 1A). The most preferentially responsive of these mice (no. 3) was used for the preparation of hybridoma cells. Screen of the hybridomas by KinExA assay for their ability to bind to Cr(III) · EDTA-OVA yielded 34 positive clones. The reactivity of these clones toward Cr(III) · EDTA and metal free EDTA was analyzed and compared. Some clones exhibited only a small difference in reactivity, other clones exhibited a more potent reactivity toward Cr(III) · EDTA, with a variety of potencies (Figure 1B). A single colony was isolated from the clone which showed the largest difference in reactivity (3G), and this strain was identified as RD3G4. Binding Specificity. The binding specificity of RD3G4 monoclonal antibody was investigated by conducting equilibrium binding measurements on a KinExA immunoassay instrument. Nine different metal-EDTA complexes, metal free EDTA and purified antibody were used for the quantification of the free, unbound antibody present in reaction mixtures of antibody, antigen, and antibody-antigen complexes. Figure 2 shows the results, which are also summarized in Table 1. RD3G4 exhibited especially high affinity binding to Cr(III) · EDTA, with a Kd of 9.7 × 10-9 M, but it also bound to Fe(III) · EDTA, at a 2000-fold weaker Kd, despite an ionic radius for Fe3+ that was only 0.03 Å larger than Cr3+. The second highest affinity of the tested complex was exhibited by Ni(II) · EDTA and the decrease for Ni(II) · EDTA versus Cr(III) · EDTA was over 90-fold (Table 1). The ionic radii of Cr3+, Fe3+, Ni2+ are 0.76, 0.79, and 0.83 Å, respectively. Therefore, in terms of these metals, the affinity order was the inverse of the size of the radii of these metal ions. This could be a unique feature of this particular antibody, which is markedly different from other antibodies reported previously. We made another anti-Cr(III) · EDTA monoclonal antibody, Gb3G12,22 which was produced by a hybridoma from a different mouse that yielded RD3G4. The antibody Gb3G12 showed only a 4-fold lower affinity for Fe(III) · EDTA and a 120-fold lower affinity for Ni(II) · EDTA than Cr(III) · EDTA. RD3G4 and Gb3G12 have a relatively close affinity for Cr(III) · EDTA and Ni(II) · EDTA. However, the pattern between Cr(III) · EDTA and Fe(III) · EDTA is completely different. Like Gb3G12, to the best of our knowledge, most mAbs for metal-chelate complexes display a relatively Analytical Chemistry, Vol. 81, No. 10, May 15, 2009
4007
Figure 2. Response curves for RD3G4 antibody reacting with metal-EDTA complexes (“free” means metal free EDTA). Solid lines are calculated from the best fit Kd for each data set. Kd values for the curves shown, approximately equal to the midpoint for the curves shown, are summarized in Table 1. Error bars represent the standard error of the mean of triplicate determination. Table 1. Binding of RD3G4 Antibody to EDTA Complexes Cr(III) · EDTA Ni(II) · EDTA Cu(II) · EDTA Zn(II) · EDTA Fe(III) · EDTA Cd(II) · EDTA Mn(II) · EDTA Ca(II) · EDTA Mg(II) · EDTA free EDTAc
Kda (µM)
CRb (%)
0.0097 0.90 3.0 3.7 21 40 640 1100 1400 2700
100 1.1 0.3 0.26 0.046 0.024 0.002 0.000 88 0.000 69 0.000 36
a Kd is the equilibrium dissociation constant. b CR is the cross reactivity calculated as 100[Kd of Cr(III) · EDTA]/(Kd of the EDTA complex). c Free EDTA means metal free EDTA.
similar affinity between metal-chelate complexes of similar ionic radius. For example, there is a 0.7 Å difference in ionic radius between Cd2+ and Hg2+. Of the EDTA complexes, antibody E5 showed the highest affinity for Cd(II) · EDTA and only a 2-fold lower affinity for Hg(II) · EDTA, while all the other metals had a greater than 100-fold decrease in the affinity for the metal-chelate complex.27 Another antibody, A4, bound most tightly to Hg(II) · EDTA complexes, and it showed a 4-fold lower affinity for the Cd(II) · EDTA and more than 100-fold lower affinity for the other metal EDTA complexes.27 The affinities of these two antibodies correlated well with the extent of the structural difference for these metal-EDTA complexes.27 In a similar case, antibody 2D12.5 bound to NBD complex of the lanthanides.20 When the radii of the tested trivalent lanthanide ions varied in small increments across a series from 1.03 Å (Lu3+) to 1.21 (La3+), the stability of the antibody-antigen complex changed consecutively.20 Reduction of Hexavalent Chromium and Immunoassay. As shown in Figure 2, the RD3G4 antibody can be used for the measurement of trivalent chromium with a detection limit of (27) Jones, R. M.; Yu, H.; Delehanty, J. B.; Blake, D. A. Bioconjugate Chem. 2002, 13, 408–415.
4008
Analytical Chemistry, Vol. 81, No. 10, May 15, 2009
Figure 3. Response curve for the Cr(VI) immunoassay. The X-axis represents the chromium concentration before the reduction. Error bars represent the standard error of the mean of triplicate determination.
approximately 3 nM (0.16 µg/L). Hexavalent chromium cannot form the EDTA complex because it normally occurs as chromate (CrO42-) or dichromate (Cr2O72-) oxyanions.1 Therefore, only trivalent chromium can be detected by RD3G4 with EDTA in a mixture of trivalent and hexavalent chromium. The method for reducing hexavalent chromium to trivalent chromium is generally known and is performed by using chemical reducing agents such as sulfur dioxide (SO2), sodium bisulfite (NaHSO3), or sodium metabisulfite (Na2S2O5). Figure 3 shows the results of the immunoassay conducted with RD3G4 and potassium dichromate after reducing with sodium metabisulfite. Signals dependent on the concentration of potassium dichromate were observed with a detection limit of 30 nM (1.6 µg of Cr/L). The reason why the detection limit of hexavalent chromium was 10 times larger than that of trivalent chromium is that the original solution had to be diluted by adding reducing and neutralizing agents. After the addition of sodium metabisulfite to the mixture of tri- and hexavalent chromium, all chromium was in the trivalent
form, which can form the EDTA complex, and therefore, the total chromium concentration can be determined by the assay format used in Figure 3. Isolation and Reduction of Hexavalent Chromium. From the sample containing both tri- and hexavalent chromium, determination of only the hexavalent chromium concentration using the RD3G4 antibody requires the isolation of hexavalent chromium from trivalent chromium. Hexavalent chromium (anion) and trivalent chromium (cation) were expected to be separable by means of an ion-exchange column. Potassium dichromate solution and chromium(III) chloride solution were individually passed through a column packed with a strong base anion-exchange resin, and the solution and the effluent were then analyzed by ICP-MS. While the complete uptake of 1 µM potassium dichromate solution did occur, Cr3+ was not retained on the column (data not shown). With the use of the reducing agent 0.2 M sodium metabisulfite, the dichromate on the resin was reduced to Cr3+, which was predicted to be eluted from the column. The percentage of Cr recovery was calculated from the amounts of potassium dichromate applied to the column and the chromium
Table 2. Recovery of the Chromium from Column Cr2O72- concn (µM)
recoverya (%)
1.0 2.0 1.0b
108.0 ± 5.4 95.8 ± 4.6 106.6 ± 22.8
a These values were calculated from ICP-MS data. b The solution contained 20 µM of Cr3+, Cu2+, and Mn2+ and 40 µM of Mg2+.
found in the eluate (Table 2). The recovery of Cr was quantitative under the examined condition. Coexisting cations, i.e., Cr3+, Cu2+, Mn2+, and Mg2+, had no influence on the recovery. Because the concentration of the sodium metabisulfite used for the elution corresponded to those applied for the immunoassay shown in Figure 3, the detection limit of hexavalent chromium would be expected to also be 30 nM (1.6 µg of Cr/L), which was the concentration in the column eluate. Received for review February 25, 2009. Accepted April 1, 2009. AC900419C
Analytical Chemistry, Vol. 81, No. 10, May 15, 2009
4009
