Development and optimization of an immunoassay for the detection of Hg(II) in lake water

Abstract In this paper, an indirect competitive enzyme‐linked immunosorbent assay (IC‐ELISA) has been developed and optimized to detect Hg(II) in tap water and lake water based on a monoclonal antibody (mAb‐A24). Some stabilizing additives (Gelatin, bovine serum albumin [BSA], polyvinyl alcohol [PVA], and polyvinyl pyrrolidone [PVP]) and surfactant (Tween‐20) have been investigated thoroughly in the optimization process. Under the optimal condition, the 50% half maximal inhibitory concentration (IC50) and limit of detection (LOD) were 1.68 and 0.079 ng/ml, respectively. These anti‐Hg mAbs also have some affinity with methyl mercury (CH3Hg) and with no cross‐reactivity with other thirteen metal ions. The developed method has shown satisfactory recovery of Hg(II), ranged between 91% and 116%, from tap water and lake water. Therefore, this immunoassay can be used to detect trace Hg(II) in environment water.

In this paper, we have developed a fast and simple enzyme-linked immunosorbent assay (ELISA) based on our newly screened monoclonal antibody (mAb) for the detection of Hg(II). Antibody sensitivity and specificity were measured before the establishment of immunosorbent assays. Effects of the protection reagents, pH and surfactant on antibody sensitivity have been evaluated. Samples spiked in tap water and lake water were tested by the developed method to evaluate the accuracy.

| Synthesis of immunogen and coating antigen
Hg was conjugated with BSA through 6-mercaptonicotinic acid (MNA) by EDC/NHS method as described previously (Liu, Xing, Yan, Kuang, & Xu, 2014). Hg-MNA-BSA could expose the Hg alone to the immune system. Briefly, 15 mg MNA, 16 mg EDC, and 23 mg NHS were dissolved in 0.5 ml DMF and stirred overnight at room temperature. Then, the solution was centrifuged at 20,000 g for 10 min, and the supernatant was added dropwise to 40 mg BSA solution (dissolved in 3 ml of 0.1 M carbonate buffer saline, pH 9.0). The mixture was kept reacted for 12 hr and dialyzed 2 days

| Hg concentration detection by ICP-MS
Before the immunization, the amount of Hg(II) in the immunogens and coating antigen was detected by ICP-MS. 0.2 ml samples were digested by concentrated HNO 3 and diluted by 2% HNO 3 to 10 ml.

| Immunization and mice screening by ELISA
The BALB/c mice, 6-8 weeks of age, were chosen and immunized five times using immunogen of Hg-MNA-BSA, at an interval of 28 days . The immunization doses were 100, 100, 100, 50, and 50 µg, respectively. After fifth immunization, the mice show the highest affinity and specificity to Hg and no cross-reactivity to MNA and other metal ion was selected for cell fusion. The selected mice were given final booster injection intraperitoneally (30 μg of immunogen directly dissolved in 200 μl of physiological saline). After 3 days, the mice were sacrificed and the spleen was taken to cell fusion according to the previously described steps.

| Preparation of monoclonal antibodies
The cell fusion experiment was performed by conventional PEG fusion technology to obtain monoclonal antibodies. At the seventh day after cell fusion, the supernatant secreted by the hybridoma cells was screened by ELISA to find the positive cells. Positive cells were subcloned three times by limited dilution method. The sensitivity, subtypes, and cross-reactivity of antibody purified from ascites fluid were detected.

| Competitive indirect ELISA (IC-ELISA)
Conventional IC-ELISA was carried out according to previous research. Hg-MNA-OVA (0.1, 0.2, 0.3, 0.5, 1, and 2 µg/ml) was diluted in CBS (0.01 M pH 9.0). 100 µl diluted coat antigen was added to the well of microtitre plates and incubated for 2 hr at 37°C. The plates were washed three times using PBST and blocked by the CBS containing 0.1% gelatin at 4°C overnight. After washing the plates twice, those plates were dried for 20 min and preserved at 4°C for use. The color development was in conformance with the standard process.
Based on a noncompetitive two-dimensional titration, OD values around 1.8 were selected for sensitivity detection. A series of Hg standard solution prepared in pH 7.4 PBS was tested according to the previously described steps. The standard curves under different combinations were obtained and IC 50 was calculated. Concentrations of coating conjugates and concentrations of antibody provided the inhibition curve with the lowest IC 50 were selected for further assay development.

| Assay optimization
The effects of pH, blocking solution, and surfactant on the competitive assay were further optimized.

| pH effect
To evaluate the effect of the pH of sample solution on the sensitivity, PBS buffers with pH 3-8 were prepared and used to dilute Hg standard solution. All other assay processes were the same as described before. All the experiments were performed three times.

| Blocking and antibody dilution buffer
The type and concentration of blocking reagents are one of the important factors to determine the stability and validity of ELISA. In this paper, Hg was recognized by antibody alone without the help of Hg chelate compound. Protein or other polymer in the blocking and antibody dilution buffer may interfere the interaction of Hg and antibody.

| Cross-reactivities (CRs)
Other metals such as CH 3 Hg, Cr(III), Cr(VI), Mg(II), Cu(II), Ni(II), Fe(II), Co(II), Mn(II), Zn(II), Au(II), As(III), Ca(III), and Cd(II) were tested by the established IC-ELISA. Metal ions with different concentration range between 1 and 1,000 ng/ml were prepared in PBS and tested. The CR was calculated based on the following formula.

| Fortification of Hg(II) in tap water and lake water
A standard dilution series of Hg(II) was spiked into tap water and lake water with concentration of 1, 2, and 5 ng/ml.

| Characterization of immunogen and coat antigen
It is well know that hapten alone could not induce immune response. scanning of Hg-MNA-BSA and Hg-MNA-OVA is shown in Figure 1.
The characteristic peaks of BSA and OVA were located at 280 nm.
Furthermore, we also synthesized CH 3 Hg-MNA-OVA as the heterologous coating and used to test their sensitivity to Hg(II) and CH 3 Hg.
Although both of them could be identified, the results indicated that antibody molecules containing an epitope with higher affinity to Hg(II). Finally, mouse 4 demonstrated the highest sensitivity to Hg(II) and used for cell fusion.

| Characterization of mAb and assay optimization
After cell fusion, subclone, and purification, mAb A24 with highest titer and sensitivity was obtained. The isotypes of obtained mAbs were IgG1 type. Before the practical application of this antibody for the detection of environmental water samples and crop, it is essential to optimize the sensitivity of ELISA and stability of kit components (Kolosova, Shim, Yang, Eremin, & Chung, 2006). Therefore, some additives were tested for their stabilizing effect and interference on antigen immobilized and antigen-antibody interaction in the microwell plates. In order to protect of immobilized antigen, sugars (sucrose), polymers (PVP, PVA, PEG, etc), gelatin, or BSA were added during the drying process and storage (Dankwardt, Müller, & Hock, 1998).
First, pH of coating solution was investigated. Electrostatic interaction plays an important role in the efficiency of antigen immobilization on the surface of binding sites of the polystyrene.
Commonly, carbonate (pH 9.6), Tris-HCl (pH 8.5), and PBS (pH 7.2) were all suitable for BSA or OVA immobilization on the polystyrene surface by passive adsorption. In the presence of detergents or other additives such as PVA and BSA, the binding sites of the polystyrene surface are occupied and antigen immobilization prevented (Gardas & Lewartowska, 1988). These additives, such as  The influence of varying the pH of the standard solution buffer to assay sensitivity was shown in Figure 3a. Hg(II) inhibition curves (1, 2, and 5 ng/ml Hg(II) were used here) were obtained at pH 3-8, and pH effects were evaluated mainly based on the sensitivity and protein stability. The absorbance values were depressed at pH 3 and pH 4. The sensitivity was decreased when pH was more than 7. The results indicating the assay have higher control OD values and are more sensitive under weak acid conditions, and Hg(II) dilution solution with pH 6 was used for the subsequent assays.
The influence of surfactant in antibody dilution buffer was shown in Figure 3b.

| Characteristics of the optimized assay
The optimized Hg(II) ELISA parameters were listed as follows: 1 μg/ml PVA, and 0.05% Tween-20. As shown in Figure 4, the IC 50 of the optimized immunoassay was 1.68 ng/ml with detection range (IC 20 -IC 80 ) of 0.22-12.67 ng/ml. The limit of detection (LOD) was 0.079 ng/ml.

| Cross-reactivities (CRs) of antibody
The newly screened antibody A24 was highly sensitive and selective for the target analyte Hg(II), and the IC 50 for CH 3 Hg was 12.8 ng/ml.
The developed ELISA had low CRs for Au(II) (3.36%) ，Cu(II) (0.84%), and negligible CRs for other tested metals (<0.84%). Therefore, the interference was negligible in tap and lake water detection.

| Detection of Hg(II) in tap water and lake water by ELISA and ICP-MS
To investigate the accuracy of the ELISA, tap water and lake water spiked with 1, 2, and 5 ng/ml Hg(II) were detected by immunoassay and ICP-MS (Sun et al., 2018). A recovery of 80%-120% was acceptable for immunoassay. The tap water and lake water have found no Hg(II) existed by ICP-MS. Recovery rates from ELISA ranged between 91% and 116%, whereas those obtained from the same samples by ICP-MS were from 100.4% to 115.5% as shown in Table 4.
The variable coefficients (CV) were from 5.6% to 16.6% for ELISA.
Other relevant works also have similar recoveries (Wang et al., 2012;Zou et al., 2017). The accurate recovery of the spiked experiment suggests this method is a simple and suitable screening tool for the detection of Hg(II) water pollution.

| CON CLUS ION
Taken together, a direct competitive ELISA for the detection of Hg (II) in tap water and lake water samples has been developed and optimized. Hg(II) could be detected without chelation compared with our previous work. ELISA kit components were optimized for better sensitivity and decreasing interference. The IC 50 for the developed assay was 1.68 ng/ml. The average recovery rates were between 91% and 116%. In our group, we have focused on application of heavy metal immunoassay for the detection of Pb(II) and Cd(II) in rice and wheat (Xu et al., 2016). Hg(II) pollution is rare in crops, and we will pay attention to Hg(II) in fish and kelp. The developed assay showed the antibody owned the potential for the detection of Hg(II) and CH 3 Hg in foods in future.

ACK N OWLED G M ENTS
We thank Tao

CO N FLI C T O F I NTE R E S T
The authors have declared that no conflict of interest exists.

E TH I C A L S TATEM ENT
All animal care and experimental protocols were ethically reviewed and approved by the Ethics Committee of Nanjing Agricultural University.