Introduction

Environmental pollution by heavy metals is a growing problem worldwide, especially in developing countries. At least 20 metals are known to be toxic and half of these, including mercury, lead, copper, arsenic, nickel, cadmium, selenium, silver, and zinc, are heavy metals [1]. Mercury is a toxic metal that exists naturally in the environment, and its pollution can threaten the health of humans and wildlife [2–9]. Monitoring of mercury levels in water, soil, paint, and other media is important for health and safety. Current analytical methods for detection of mercury include atomic sbsorption spectrometry (AAS), inductively coupled plasma atomic emission spectrometry (ICPAES), inductively coupled plasma mass spectrometry, and atomic fluorescence spectrometry, which are sensitive and reliable [10, 11]. However, these methods are expensive and require highly specialized apparatus and trained operators, which makes them unsuitable for onsite detection. By contrast, immunoassays using specific monoclonal antibodies are rapid, inexpensive, simple, and highly sensitive [12]. Because metal ions are too small to elicit an immune response, they are usually chelated with isothiocyanate derivatives of bifunctional chelating agents, and conjugated to a carrier protein via the isothiocyanate [13–16]. This process is usually conducted by adjusting the pH to 9.0, but at this pH mercury would precipitation. This makes it difficult to produce a mercury antigen using isothiocyanate derivatives of chelators. Amino derivatives of chelators are much cheaper than the isothiocyanate derivatives, and can be linked to carrier proteins at pH 7.0 using glutaraldehyde. In the present study, a complete antigen and specific monoclonal antibody (mAb) for mercury were prepared using an amino derivative of EDTA (1-(4-aminobenzyl)ethylenediamine-N,N,N′,N′-tetraacetic acid (aminobenzyl-EDTA)). ELISA for the detection of mercury based on specific monoclonal antibodies was developed.

Materials and Methods

Reagents

Aminobenzyl-EDTA was obtained from Dojindo Laboratories (Gaithersburg, MD). Mercury(II) nitrate monohydrate (99.99%), goat anti-mouse horseradish peroxidase conjugate (GAM-HRP) hypoxanthine-aminopterin-thymidine medium (HAT), and hypoxanthine-thymidine medium (HT), bovine serum albumin (BSA), keyhole limpet hemocyanin (KLH), Freund’s complete and incomplete adjuvants, 3,3′,5,5′-tetramethylbenzidine (TMB), mouse monoconal antibody isotyping kit, and atomic absorption standard metals (1,000 μg/mL) were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO). Fetal bovine serum (FBS) was obtained from Tianjin Haoyang Co. (Tianjin, China). ELISA 96-well plates and tissue culture plates were purchased from Corning Co. (Cambridge, MA). Water used for reagent preparation was purified by a Millipore (MUL-TYPE9000) water purification system. A BIO-TEK ELX800 plate reader was purchased from Bio-Tek Instruments Inc. (Winooski, VT). Sp2/0 cells were preserved by our laboratory. BALB/c mice were obtained from the Experimental Animal Center of Bethune Medical College (Changchun, China). All glassware and plasticware was treated with HNO3 (H2O/HNO3, 1:1, v/v) overnight and rinsed with ultrapure water. Amicon Centricon 30 concentrator tubes (Millipore Co., Bedford, MA) were treated with 100 mmol/L EDTA and rinsed with ultrapure water before use.

Antigen Preparation and Animal Immunization

Mercury ions were conjugated to the carrier protein (BSA or KLH) via the bifunctional chelating reagent [17]. The amino group of the bifunctional EDTA derivative was linked to the carrier protein using glutaraldehyde.

Mercury(II) nitrate monohydrate (5.14 mg, 15 μmol) was dissolved in 100 μL of ultrapure water, and mixed with a solution of aminobenzyl-EDTA (3.98 mg, 10 μmol) in dimethyl sulfoxide (50 μL). The pH of the mixed solution was adjusted to 5.5 with 1 mol/L KOH. Then 1 mL of glutaraldehyde (mass fraction, 1%) was added dropwise to the solution and the pH was adjusted to 7.0. After reaction at room temperature for 2 h, the solution was added dropwise to 1 mL of carrier protein solution (KLH or BSA, 10 mg/mL, pH 7.0, in 0.1 mol/L HEPES) with stirring. The pH was adjusted to 7.0, and the mixture was left to react for 18 h at 37.5°C. Then the conjugate (KLH-aminobenzyl-EDTA-Hg or BSA-aminobenzyl-EDTA-Hg) was purified by removing low-molecular-weight reagents via centrifugation (Centricon 30), and washed thoroughly six times with HEPES buffer (0.1 mol/L, pH 7.0). BSA-aminobenzyl-EDTA was prepared as described above without the addition of Hg2+. All the conjugates were adjusted to a final concentration of 1 mg/mL using metal-free HEPES buffer (0.1 mol/L, pH 7.0).

Four 8-week-old female BALB/c mice were injected intraperitoneally with KLH-aminobenzyl-EDTA-Hg (100 μg/mouse) at 2-weekly intervals [18]. The immunogen was emulsified in complete Freund’s adjuvant for the first injection and in incomplete Freund’s adjuvant for the subsequent injections. One week after the sixth immunization, tail vein blood was collected for polyclonal antibody titer testing by indirect ELISA, with unimmunized mice as the control. Ninety-six-well polystyrene plates were coated with 1.25 μg/mL BSA-aminobenzyl-EDTA-Hg or BSA-aminobenzyl-EDTA (100 μL/well), washed three times with PBS containing 0.05% Tween 20, and blocked with 3% BSA in PBS. After a washing step, the immobilized antigens were incubated with 10,000 times diluted sera of the immunized mice. PBS-diluted serum from a nonimmunized mouse was used as the control. After incubation for 1 h at 37°C, the 96-well polystyrene plates were washed again and incubated sequentially with GAM-HRP and TMB microplate substrate as described previously [19]. The absorbance of each plate was measured by a microplate reader at 450 nm. The data are expressed as the mean ± SD of triplicate measurements.

Hybridoma Production and Screening

The mouse that showed the highest antibody reactivity to BSA-aminobenzyl-EDTA-Hg and the lowest serum antibody reactivity to BSA-aminobenzyl-EDTA-Hg was injected with 100 μg of KLH-aminobenzyl-EDTA diluted in 100 μL of HEPES buffer (0.1 mol/L, pH 7.0).

Myeloma cells (Sp2/0) with high viability and rapid growth were prepared before cell fusion. Three days after the last injection the mouse was killed. Then the spleen cells were harvested aseptically, washed in RPMI 1640 medium, and fused with Sp2/0 cells at a ratio of 10:1 (spleen/myeloma) using 50% polyethylene glycol (PGE 2000). The cell pellets were suspended in HAT selective medium (RPMI 1640 + 20% FBS + HAT), and then added dropwise to 96-well cell culture plates (100 μL/well) that had been coated with mouse peritoneal cavity cells as feeder cells 1 day prior. The plates were incubated at 37°C in an atmosphere with 5% CO2 for 7 days. The HAT medium was changed every other day and switched to HT medium when most of the non-fused cells were eliminated [13].

When the hybridoma cells had grown to 20–30% confluence, the clone supernatant was screened for the ability to bind to immobilized conjugates (BSA-aminobenzyl-EDTA-Hg or BSA-aminobenzyl-EDTA) by the indirect ELISA method as described above.

The clone that showed high titer to BSA-aminobenzyl-EDTA-Hg and low titer to BSA-aminobenzyl-EDTA was selected to test whether the culture supernatant could bind with chelated mercury or aminobenzyl-EDTA by competitive ELISA. Aminobenzyl-EDTA at concentrations of 0, 1, 2, 4, 6, 8, 10, 12, 14, 16 mmol/L (50 μL/well) and culture supernatant (50 μL/well) were added to the 96-well plates previously coated with BSA-aminobenzyl-EDTA-Hg conjugate. The subsequent steps were the same as in the indirect ELISA method described above. The results were used to identify the antibodies that were not bound primarily to metal-free aminobenzyl-EDTA, and select an optimal concentration of aminobenzyl-EDTA for subsequent anti-mercury antibody screening.

For anti-mercury antibody screening, a mercury solution at concentrations of 0.001, 0.01, 0.1, 1, 10, 100, 1,000 μg/L was mixed with aminobenzyl-EDTA at the optimum concentration. These mixtures were used to inhibit antibody binding to BSA-aminobenzyl-EDTA-Hg immobilized in the 96-well plate with the protocol described above. This screen yielded positive hybridoma cell lines that did not bind to aminobenzyl-EDTA and Hg2+, but specifically bound to aminobenzyl-EDTA-Hg2+. These cell lines were subsequently subcloned by limiting dilution. The stable clones were expanded and preserved in liquid nitrogen.

Isotyping, Production and Purification of mAb

The immunoglobulin subclasses of the antibodies were determined using a mouse monoclonal antibody isotyping kit according to the manufacturer’s recommendations. Only the cell line secreting IgG was expanded for ascites production.

Female 10-week-old BALB/c mice, pre-treated with incomplete Freund’s adjuvant, were intraperitoneally injected with hybridoma cells. Ascitic fluid was collected and clarified by centrifugation at 3,000×g for 10 min, and then initially purified using saturated ammonium sulfate solution [20]. Samples were further purified through Hi Trap Protein G HP. The protein concentration of the purified antibody was determined by BCA protein assay.

Determination of Affinity Constants for BSA-aminobenzyl-EDTA-Hg

The affinity constant (K aff) of the monoclonal antibody was determined by indirect ELISA according to Beatty et al. [21]. Ninety-six-well plates were coated with BSA-aminobenzyl-EDTA-Hg at concentrations of 2.0, 1.0, 0.5 μg/mL, and then incubated with serial concentrations of purified mAb. The indirect ELISA was performed as described above. Three sigmoid curves were constructed of OD450 against concentration of the antibody added to the wells.

Competitive Indirect ELISA Optimization

Optimum concentrations of mAb and the coating antigen were determined by indirect ELISA. The optimum incubation time of mAb with aminobenzyl-EDTA-Hg (30, 45, 60, and 90 min), concentration of GAM-HRP (2,500, 5,000, and 10,000 times dilution), and time of color development (5, 10, 15, and 20 min) were determined by indirect ELISA. To evaluate the effect of the pH on assay performance, HBS buffer solutions (137 mM NaCl, 3 mM KCl, and 10 mM phosphate) with pH values of 4, 5, 6, 7, 8, and 9 were prepared. Then the optimum conditions determined above were used for all other assay conditions. Standard curves (data not shown) were constructed using the mean of triplicate results to evaluate the effect of the pH on assay performance.

Serially diluted mercury standard solutions (0.001, 0.01, 0.1, 1, 10, 100, 1,000, and 10,000 μg/L) were mixed with equal volumes of HBS containing 2 mmol/L aminobenzyl-EDTA. The mixtures were incubated with equal volumes of diluted mAb. The subsequent steps were the same as those in the competitive indirect ELISA (CI-ELISA) described above, and with the determined optimal assay conditions. In the CI-ELISA, analytes that do not react with the mAb have absorbances close to 100% that of the zero analyte control, and analytes that reacted with the mAb will decrease the absorbance. The standard inhibition curve of the CI-ELISA was constructed by plotting the absorbance against the logarithm of the analyte concentration. These curves were fitted to a four-parameter logistic equation, y = (A–D)/[1 + (x/C)B + D], where x is the concentration of analyte, y is the absorbance at 450 nm (OD450), A is the maximum absorbance with no analyte, B is the curve slope at the inflection point, C is the half maximal inhibitory concentration (IC50), and D is the minimum absorbance with an infinite concentration of analyte [22].

CR with Other Metals

CR of the mAb with other metals (Cu2+, Zn2+, Cd2+, Cr3+, Pb2+, Co2+, Ni2+, Mg2+, Ag+, Ca2+, Mn2+, Mg2+, and Fe3+) was determined by CI-ELISA. The 96-well plates were coated with BSA-aminobenzyl-EDTA-Hg and blocked with 1% BSA. Each metal was diluted to the concentration range of 10−7 to 1 mmol/L in HBS containing the metal-free aminobenzyl-EDTA (2 mmol/L). Then the metal solutions (50 μL/well) and an equal volume of culture supernatant were added to the 96-well plates. The subsequent steps were same as for the indirect ELISA described above. The CR was obtained from the IC50, which was defined as the concentration of inhibitor r to inhibit color development by 50% compared with control wells containing no competitor, of the mercury standard (IC50 mercury) and of the other metal (IC50 other metal) from the same plate as follows: CR% = (IC50 mercury/IC50 other metal) × 100.

Measurement of Hg2+ in Environmental Water Samples

Water samples were collected from the Songhua River (Jilin, China), Yitong River (Changchun, China), and Xinlicheng Lake (Changchun,China), and the mercury ion concentrations were measured by CI-ELISA. The water samples were adjusted to pH 7.0, and then they were mixed with equal volumes of HBS containing 2 mmol/L aminobenzyl-EDTA. The mixtures were incubated with equal volumes of diluted mAbs. The subsequent steps were the same as those used in the CI-ELISA standard curve construction. The mercury contents of the water samples were also measured by GFAAS. All the samples were prepared and analyzed in triplicate.

Results

Antigen Preparation and Animal Response

In this study, aminobenzyl-EDTA was used to prepare mercury antigens. The complete immunogen (KLH-aminobenzyl-EDTA-Hg) was used to immunize BALB/c mice. One week after the sixth immunization, blood from each mouse was collected from the tail vein, and the sera were tested for antibody responses by indirect ELISA. As shown in Table 1, all the mice showed much higher antibody reactivity to immobilized BSA-aminobenzyl-EDTA-Hg than immobilized BSA-aminobenzyl-EDTA. Among the mice, mouse 1 showed the highest antibody reactivity to immobilized BSA-aminobenzyl-EDTA-Hg and the lowest reactivity to immobilized BSA-aminobenzyl-EDTA.

Table 1 Serum reactivity of BALB/c mice injected with KLH-aminobenzyl-EDTA-Hg a
Full size table

Hybridoma Production and Screening

The mouse 1 was injected with KLH-aminobenzyl-EDTA (100 μg in 100 μL of HEPES buffer). Three days later, the mouse was killed and the spleen cells were harvested for fusion with Sp2/0 cells. After fusion, indirect ELISA was used to screen the culture supernatants of 517 hybridoma clones for their ability to bind to immobilized BSA-aminobenzyl-EDTA-Hg and BSA-aminobenzyl-EDTA. The supernatant from Sp2/0 was used as a negative control. Two strains of hybridoma cell lines (3D9 and 5F7) showed high binding to BSA-aminobenzyl-EDTA-Hg and poor binding to BSA-aminobenzyl-EDTA.

The supernatants from the 3D9 and 5F7 cell lines were tested for binding with chelated mercury and aminobenzyl-EDTA by CI-ELISA. Aminobenzyl-EDTA concentrations >6 mmol/L did not inhibit reaction of the antibodies with the immobilized BSA-aminobenzyl-EDTA-Hg (Fig. 1). Consequently, 2 mmol/L aminobenzyl-EDTA was selected for subsequent experiments.

Fig. 1
figure 1

CI-ELISA with 3D9 and 5F7. a Inhibition by aminobenzyl-EDTA, and b inhibition when Hg2+ was serially diluted into 2 mmol/L EDTA. Each point represents the mean of triplicate determinations

Full size image

When Hg2+ was serially diluted with 2 mmol/L aminobenzyl-EDTA and incubated with the immobilized BSA-aminobenzyl-EDTA-Hg in the presence of hybridoma culture supernatant from the 3D9 and 5F7 cell lines, the ability of mercury to inhibit the antibody binding to BSA-aminobenzyl-EDTA-Hg depended on the mercury concentration (Fig. 1).

The two positive hybridoma cell lines (3D9 and 5F7) were subcloned by limiting dilution and expanded to produce ascites.

Subclass Determination and Production of Monoclonal Antibody

The immunoglobulin subclasses of the antibodies were determined using a mouse monoclonal antibody isotyping kit. The antibody isotyping secretions by 3D9 and 5F7 were IgM and IgG1, respectively. Only the cell line (5F7) secreting IgG was expanded for ascitic fluid production. The mAb produced from the ascitic fluid in mice (2.80 mg/mL) was purified and stored at −20°C for further characterization studies.

Determination of K aff for BSA-aminobenzyl-EDTA-Hg

The K aff was determined by indirect ELISA using serial dilutions of the purified antibody and three concentrations of immobilized BSA-aminobenzyl-EDTA-Hg (Fig. 2). K aff was calculated using K aff = (n–1)/2(n[Ab′]–[Ab]), where [Ab] and [Ab′] are the antibody concentrations at OD50 (half maximal optical density) and OD50′ of the 96-well plate coated with [Ag] and [Ag′], with [Ag′] = [Ag]/n. The K aff of mAb 5F7 was 4.31 × 109 L/mol.

Fig. 2
figure 2

Affinity constant curve of mAb 5F7 against aminobenzyl-EDTA-Hg. Each point represents the mean of triplicate determinations

Full size image

CI-ELISA Optimization

The optimal concentrations of mAb and coating antigen were determined to be 70 ng/mL and 1 μg/mL, respectively, by ELISA. The indirect ELISA showed that the optimal incubation time of mAb with the mixture of aminobenzyl-EDTA was 60 min, the optimal dilution of GAM-HRP was 5,000 times, and the optimal time of color development was 10 min. The pH range suitable for the CI-ELISA was 5.5–7.0 and the optimal pH was 7.0.

Standard curves of the CI-ELISA were constructed at the optimal assay conditions. As shown in Fig. 3, the IC50 of the optimized assay was 2.955 ± 0.0041 μg/L, and the detection range (IC20–80) was 0.087–790.4 μg/L. The lower limit of detection, which was defined as two standard deviations (SDs) above the lowest detectable level, was estimated as 0.042 μg/L.

Fig. 3
figure 3

Standard inhibition curve of the CI-ELISA

Full size image

CR with Other Metals

The mAb 5F7 was highly selective for Hg2+. The CRs with Cd2+, Pb2+ were 1.87% and 0.381%, respectively, and with other metals were less than 0.2% (Table 2). These results indicate that the mAb 5F7 has high specificity for aminobenzyl-EDTA-Hg.

Table 2 Cross-reactivity (CR) of other metals with mAb (5F7)
Full size table

Measurement of the Hg2+ in Environmental Water Samples

The mercury ion concentrations in water samples from the Songhua River (Jilin, China), Yitong River (Changchun, China), and Xinlicheng Lake (China) were measured by CI-ELISA and GFAAS. The concentrations obtained by CI-ELISA correlated well with those obtained by GFAAS, and the mean recovery was 88.82% to 104.64% (Table 3).

Table 3 Hg2+ concentrations in three environmental water samples obtained by CI-ELISA and GFAAS
Full size table

Discussion

The molecular masses of metal ions are too low and their structures are too simple to induce an immune response. Consequently, preparation of a complete immunogen is critical for antibody-based immunoassays of metals. In this study, an amino derivative of EDTA (aminobenzyl-EDTA) was used to prepare mercury antigens. Mercury ions were chelated with aminobenzyl-EDTA and the resulting chelator-metal complexes were conjugated to a carrier protein (KLH or BSA) to produce complete antigens for mercury (KLH-aminobenzyl-EDTA-Hg and BSA-aminobenzyl-EDTA-Hg). After the sixth immunization, all the mice showed much higher antibody reactivity to immobilized BSA-aminobenzyl-EDTA-Hg than immobilized BSA-aminobenzyl-EDTA (Table 1), immunogens prepared from aminobenzyl-EDTA could stimulate production of specific antibodies against chelated mercury.

Metal antigens have been prepared from isothiocyanate derivatives of chelators in other studies, but required a pH adjustment to 9.0 [13–16]. If applying the reported isothiocyanate derivatives of chelators to prepare antigen for mercury, the mole of mercury can’t surpass that of chelator because the mercury that not chelated by chelator would precipitate when the pH was adjusted to 9.0. In the case, there would be some metal-free aminobenzyl-EDTA sites of the antigen, which will induce production of nonspecific polyclonal antibody against metal-free chelator but not against the chelated mercury. In this study, aminobenzyl-EDTA, which is much cheaper alternative to the isothiocyanate derivatives, was linked to carrier protein at pH 7.0. This avoided the potential precipitation of mercury with pH adjustment to 9.0. The concentration of mercury(II) (5.14 mg, 10 μmol) was 1.5 times that of aminobenzyl-EDTA (3.98 mg, 10 μmol), which guarantees that all the aminobenzyl-EDTA sites of the antigen will be occupied by mercury. This ensures that the immunogen (KLH-aminobenzyl-EDTA-Hg) will induce production of specific polyclonal antibody against chelated mercury but not against the metal-free chelator. Immunoassays offer an alternative approach, and have significant advantages over, traditional instrument-intensive methods of metal analysis. They are remarkably quick, simple, transportable, require minimum sample pretreatment, and have high throughput. Furthermore, studies have shown that the use of immunoassays can reduce the analysis cost by 50% or more [23, 24]. Our results revealed that the mAb 5F7 prepared from aminobenzyl-EDTA displayed high affinity and specificity for the aminobenzyl-EDTA-Hg complex. Subsequently, the mAb was used to establish a CI-ELISA for mercury ions. The lower limit of detection of the assay was 0.042 μg/L, which is much lower than the World Health Organization threshold for mercury in drinking water (0.5 μg/L). The assay was successfully applied to the measurement of mercury ion concentrations in three environmental water samples. The concentrations of mercury in the water samples determined by CI-ELISA correlated well with those obtained by GFAAS, and the mean recovery was 88.82% to 104.64% (Table 3). This indicates the method could be used for monitoring of mercury in water. Furthermore, the time for preparing sample and aminobenzyl-EDTA complex of the immunoassay were shorter than the traditional methods. The immunoassay is a promising analytical tool for rapid and sensitive determination of mercury ion concentrations in agricultural products and the environment.