Methyl aniline, N,N-dimethyl aniline, zinc chloride, 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDC), N,N-dimethylformamide (DMF), and N-hydroxysuccinimide (NHS) were obtained from Aladdin Reagent Company. All other reagents used in the study met or exceeded analytical reagent grade standards. Female Balb/c mice were provided by the Guangdong Medical Laboratory Animal Center (SCXK (Yue) 2018-0002) and were subsequently housed and maintained at the South China Agricultural University Animal Center (SYXK (Yue) 2019-0136).

As shown in Figure 1, Step 1: Add 7 g (6.5 mmol) of N-methylaniline to a 250 mL single-necked flask. Dissolve it completely in 100 mL of anhydrous methanol. Then, add 17 g (19.7 mmol) of methyl acrylate and 3 mL of triethylamine. React in an 85°C oil bath overnight to obtain compound 1 (8.8 g, 70%).

Step 2: Add 100 mL of anhydrous ethanol to dissolve 8.8 g (4.5 mmol) of compound 1 in a 250 mL single-necked flask. Then add 4.6 mL (4.5 mmol) of benzaldehyde, 5.7 mL (4.5 mmol) of N,N-dimethylaniline, and 20 g (13.5 mmol) of zinc chloride. React at 100°C in an oil bath overnight. Concentrate the solution, adjust the pH to 8 with 10% aqueous sodium hydroxide, add an appropriate amount of ethyl acetate, and then pump-filter to obtain compound 2 (2 g, 10%).

Step 3: Dissolve 2 g (4.96 mmol) of compound 2 in 10 mL of methanol in a 100 mL single-necked flask. Add 10 mL of 10% aqueous sodium hydroxide solution and react at 70°C in an oil bath overnight. Concentrate the solution, adjust the pH to 8 with dilute hydrochloric acid, extract with an appropriate amount of ethyl acetate, and purify the residue by column chromatography (PE:EA = 15:1; 10:1; 1:1; EA) to obtain compound 3.

Step 4: Dissolve compound 3 in 20 mL of acetonitrile in a 100 mL single-necked vial. Slowly add 10 mL of an acetonitrile solution with DDQ while stirring. The color of the reaction solution will immediately darken. Stir the reaction for 0.5 h. Perform column chromatography purification to obtain a small amount of the target product MG-H1.

As shown in Figure 2, Dissolve 1.5 g (10 mmol) of p-carboxybenzaldehyde in 60 mL of anhydrous ethanol in a 250 mL flask. Add 4 g (3 mmol) of zinc chloride and 4 mL (3.15 mmol) of N,N-diethylaniline, and reflux overnight. Prepare a 3:2 mixture of acetonitrile and methanol, then add DDQ to completely dissolve the compound. Weigh compound 4 into a 100 mL single-necked vial, add the mixture, evaporate the solvent, wash off the DDQ with ether, and purify the residue by column chromatography to obtain the green product MG-H2.

The three-dimensional (3D) energy-minimized structure of the MG hapten was constructed using Sybyl 8.1 (provided by Professor Xin’an Huang from Guangzhou University of Chinese Medicine). Initially, a sketch was created and then geometrically optimized using the standard Tripos force field, incorporating non-bonded interactions and Gasteiger-Hückel charges with an 8 Å cutoff, a termination gradient of 0.005 kcal/(molÅ), and dielectric constants based on previous research. The surface electrostatic potential of the obtained structure was generated using the MOLCAD surface program in Sybyl 8.1 with Gasteiger-Hückel charges.

Immunogens and coating antigens were synthesized using previously established methods. Immunogens were prepared by conjugating haptens with carrier proteins, such as lactoglobulin (LF). Coating antigens were prepared by conjugating all haptens with bovine serum albumin (BSA). The structure of the final conjugates was verified using ultraviolet–visible (UV–vis) spectral data.

Female BALB/c mice aged 6–8 weeks were subcutaneously immunized with the immunogen, with four immunizations at 3-week intervals. A mouse exhibiting the highest titer against each immunogen was selected as the spleen cell donor for hybridoma production. Spleen cells from the immunized mouse were fused with SP2/0 myeloma cells, which had been revived 10 days prior to fusion, using 50% PEG 1500 as the fusion agent. The hybridoma cells were distributed in a 96-well cell culture plate and cultured in a medium containing 80% Dulbecco’s Modified Eagle’s Medium (DMEM), 20% Fetal Bovine Serum (FBS), and Hypoxanthine-Aminopterin-Thymidine (HAT). After 2 weeks, the medium was replaced with Hypoxanthine-Thymidine (HT). Cells were incubated in a cell culture incubator (37°C, 5% CO2). After 7 days, antibody titer and binding characteristics in the culture supernatant were tested using the same method as for serum. Cells with high affinity and specificity were subcloned using limited dilution until a stable monoclonal cell line secreting antibodies was obtained.

Animal experiments strictly adhere to the guidelines of Chinese laws and regulations, the studies involving animals were reviewed and approved by the Animal Ethics Committee of South China Agricultural University. The animal immunization and hybridoma preparation process was carried out according to the previously established protocol. Male BALB/c mice aged 10–12 weeks were injected with liquid paraffin. After 1 week, hybridoma cells were injected into the mice. Two weeks later, ascitic fluid was collected from mice with noticeably enlarged abdomens. Proteins were precipitated using ammonium sulfate and then purified using Protein A columns. Protein content was determined by ultraviolet spectrophotometry using the following formula: Protein Concentration/(mg/mL) = [1.45 A₂₈₀–0.74 A₂₆₀] × Dilution Factor.

The IC-ELISA method was used to optimize working conditions, including antigen coating, antibody working concentration, and organic solvent tolerance. After optimization, a competitive IC-ELISA method was employed to establish a standard curve and determine the specificity of mAbs against MG analogs. The cross-reactivity rate was calculated using the formula:

The design of haptens is paramount in antibody preparation. Recent hapten designs targeting MG have primarily focused on introducing reactive groups onto the benzene ring without incorporating dimethylamino groups. However, the prevailing trend in recent years has neglected the presence of dimethylamino groups, despite the fact that numerous monoclonal antibodies (mAbs) produced using these designs exhibit significant cross-reactivity with crystal violet, suggesting that the dimethylamino group could be a crucial factor contributing to this cross-reactivity. To validate this hypothesis, we conducted a systematic review of the latest advancements in the field and performed in-depth analyses utilizing computational modeling techniques. As depicted in Figure 2, due to the strong electron-donating property of nitrogen atoms, the structure of MG is highly negatively charged, thereby facilitating the formation of hydrogen bonds. In contrast, crystal violet possesses similar dimethylamino antigenic epitopes, which may undermine the selectivity of the prepared antibodies toward malachite green. Introducing spacer arms onto the benzene ring results in a more positively charged overall structure, disrupting the original high negative charge state. Notably, the charge associated with the central double bonds resembles that of the parent drug, and the incorporation of spacer arms alongside dimethylamino groups has minimal impact on the overall charge distribution. Consequently, we propose a novel hapten design (Hapten 1), which not only incorporates dimethylamino groups to enhance antigen selectivity but also retains the central double bonds to mimic the immunogenicity of the parent drug. Furthermore, to further enhance antibody coating efficiency, we devised another hapten (Hapten 2) featuring exposed dimethylamino groups, thereby strengthening its interaction capabilities with antibodies. This design strategy not only improves the immune recognition efficiency of antigens but also lays a solid foundation for the subsequent development of highly sensitive and selective immunoassay methods.

H1-LF and H2-LF were used as immunogens, and H1-BSA and H2-BSA were used as coating antigens for mouse immunization, respectively. After the fourth immunization, mouse serum was collected to determine antibody titer and affinity for MG (Table 1). Absorbance and corresponding inhibition rates of MG were used as evaluation criteria. For H2-LF, serum from mouse 3 exhibited the strongest recognition of MG at 200 ng/mL, with inhibition rates of 27 and 30% for H1-BSA and H2-BSA, respectively. For H1-LF, serum from mouse 2 showed inhibition rates of 63 and 45% for H2-BSA and H1-BSA, respectively. Both H1 and H2 haptens induced strong immune responses against MG, but H1-LF, enhanced by H2-BSA, demonstrated significant improvements in titer and inhibition rate. Consequently, mouse 2 (H1-LF) was selected for cell fusion experiments.

Following cell fusion, monoclonal antibody M1 was obtained from H2-LF. Protein G affinity chromatography was used for purification due to its strong binding affinity with mammalian IgG. By adjusting the pH of the mobile phase, high-purity mAbs were achieved. The antibody was concentrated using dialysis bags covered with polyethylene glycol 20,000 at the bottom and surface. Observations were made every 30 min until the desired volume was reached. After a 4-h PBS dialysis exchange, the final concentration was determined to be 4.2 mg/mL.

As shown in Table 2, a comparison of the prepared monoclonal antibodies with previous studies revealed that the introduction of an arm onto the benzene ring that never contained a dimethylamino group resulted in a cross-reactivity with CV as high as 100%. Subsequently, the cross-reactivity decreased to 27% when an arm was introduced onto the dimethylamino group. This pattern also confirms that the dimethylamino group is the most crucial antigenic epitope for MG recognition. Based on this finding, this study further introduced a double bond to make the structure closer to MG. The successfully prepared monoclonal antibody not only outperformed previous studies in terms of sensitivity but also in specificity.

Theoretically, a decrease in the concentration of the coated antigen leads to a decrease in its own antibody’s competitive binding ability, while enhancing the competitive binding ability of the drug’s antibody. To maintain stable working conditions, the antibody’s working concentration needs to be adjusted upwards. In this experiment, the checkerboard method was used to optimize the concentrations of coated antigen and antibody. Under other consistent conditions, IC-ELISA was used to compare the IC₅₀ values under different combinations of coated antigen and antibody concentrations. The resulting standard curve had the lowest IC₅₀ value when the coated antigen concentration was 0.25 μg/mL and the antibody concentration was 125 ng/mL. Therefore, a coated antigen concentration of 0.25 μg/mL and an antibody concentration of 125 ng/mL were selected as the optimal working concentrations.

As shown in Figure 3, the buffer had little effect on MG. When the standards were diluted with PBST, the IC₅₀ values were relatively low, indicating that PBST improved the detection sensitivity compared with distilled water. This enhancement may be attributed to the compositional similarity between PBST and the antibody diluent, which maintains the pH and ionic concentration of the buffer system, thus avoiding adverse effects on the antigen–antibody reaction. In contrast, buffers such as distilled water and phosphate buffer (PB) may disrupt this equilibrium and affect detection sensitivity. Therefore, PBST was chosen as the optimal diluent for the standard.

Due to the strong lipophilicity of MG, its standard solution mainly needs to be prepared with acetonitrile as the solvent. In the actual sample detection process, the concentration of organic solvents in the reaction system has an important influence on the biological activity of the antibody, the binding efficiency of antigen and antibody, and the solubility of the drug. Therefore, the concentration of acetonitrile in the PBST buffer solution was optimized to study the organic solvent tolerance of the antibody. The results showed that when the buffer contained a low concentration of acetonitrile, the PBST buffer could not dissolve the MG well, resulting in a higher IC₅₀ value. When the acetonitrile content in the buffer was increased from 1 to 8%, the lowest IC₅₀ ratio was observed at 4% acetonitrile content. However, when the acetonitrile content in the buffer reached 8%, some of the antibodies were inactivated due to the high concentration of organic solvent. Therefore, the PBST buffer containing 4% acetonitrile was selected as the optimal buffer system condition. In summary, after a series of optimization, the IC-ELISA standard curve of MG was established under the optimal experimental conditions: as shown in Figure 4, the IC₅₀ value of MG detected by IC-ELISA was 0.83 ng/mL, and the linear ranges was 0.19–3.52 ng/mL, and IC₁₀ was 0.08 ng/mL.

The specificity of the antibodies was evaluated using Malachite Green (MG), Leucomalachite Green (LMG), Crystal violet (CV), Leuco Crystal Violet (LCV), Einecs 284-783-6 (BG) and Leucobrilliant green (LBG) as structural analogs. As shown in Table 3, the cross-reactivity of anti-MG mAb with other structural analogs was below 0.1. This indicates that the design of dimethylamino group effectively improves the selectivity of the antibody, and also proves that dimethylamino group is the key antigenic recognition epitope of malachite green, confirming the feasibility of our strategy. The antibody prepared in this study can be used to further develop highly sensitive and selective immunoassay methods (Figure 4).