Establishment of a monoclonal antibody-based indirect enzyme-linked immunosorbent assay for the detection of trimethoprim residues in milk, honey, and fish samples

ABSTRACT

A specific monoclonal antibody (mAb) against trimethoprim (TMP) was produced and a mAb-based indirect competitive enzyme-linked immunosorbent assay (ic-ELISA) was established to detect TMP residues in milk, honey, and fish samples. The working range of ic-ELISA and IC₅₀ was 1.83–9.36 and 4.14 µg L⁻¹. The average recoveries of TMP from TMP-spiked milk, honey, and fish sample were 103.5–118.0%, 100.3–107.1%, and 91.5–108.1%, respectively. The results revealed that our developed mAb-based ic-ELISA can be effectively applied to detect TMP residues in these food products.

KEYWORDS:

• Monoclonal antibody
• trimethoprim
• milk
• honey
• fish

Introduction

Trimethoprim (TMP, 5-[(3′, 4′, 5′-trimethoxyphenyl) methyl]-2, 4-pyrimidinediamine) is a member of the 2, 4-diamino pyrimidine class of dihydrofolate reductase inhibitors and an antibiotic synergists (Luo et al., 2013). In animal husbandry, the single administration of TMP causes microbial resistance; therefore, TMP is widely administered in combination with sulfonamides or other antibiotics for the therapy of animal diseases, including fowl cholera, respiratory tract secondary bacterial infection, and pullorum disease (Liu, Wu, Sun, & Wan, 2012). The potency of TMP is enhanced 10× when administered with sulfonamides. However, the excessive use of TMP has contributed to the presence of drug residue in foods, which result in antibiotic-resistance, nausea, emesis, headache, pruritus, and rush in consumers (Croubels, Wassink, & De Backer, 2002). Additionally, the long term and excessive use of TMP reduces serum leukocyte and thrombocyte and induces teratogenesis.

Several countries have established the maximum residue limits (MRLs) of TMP in foods. In the European Union and China, the MRL of TMP residues is 50 µg kg⁻¹ in pork, chicken, and fish (Märtlbauer, Meier, Usleber, & Terplan, 1992) (Andrade, de Moraes, Rocha-Filho, Fatibello-Filho, & Cass, 2009). In light of its adverse effects on consumer health, it is important to develop an effective and rapid detection method of TMP residues that can be used to screen a large number of samples.

Currently, there are several methods developed to analyze TMP residues in food, which are mainly dependent on instruments including high-performance liquid chromatography (HPLC)-ultraviolet (HPLC-UV) (Batzias, Botsoglou, Kotsaki-Kovatsi, & Kounenis, 2002; Batzias, Delis, & Koutsoviti-Papadopoulou, 2005), HPLC-diode array detector (Bedor et al., 2008), and liquid chromatography-mass spectrometry (Economou, Petraki, Tsipi, & Botitsi, 2012; Fernandez-Torres, Lopez, Consentino, Mochon, & Payan, 2011), and hydrophilic interaction chromatography (Fontanals, Marcé, & Borrull, 2011; Uchiyama et al., 2011). These methods, which have adequate sensitivity and specificity, are expensive and require laborious and time-consuming sample pre-treatments steps and highly skilled personnel.

The enzyme-linked immunosorbent assay (ELISA) is a sensitive, simple, and cost-effective method, which has been widely used in the analyses of large compounds such as microbes (Feng et al., 2013), proteins (Peng et al., 2014), and allergens (Peng et al., 2013) and small compounds such as antibiotics (Chen et al., 2016; Suryoprabowo, Liu, Peng, Kuang, & Xu, 2014), pesticides (Chen, Liu, Kuang, Song, & Xu, 2013; Chen, Xu, et al., 2013), hormones (Kong et al., 2015; Liu, Peng, Jin, & Xu, 2007), and toxins (Zhang et al., 2013). There are also some papers which have reported the detection of TMP in food-based antibodies. The polyclonal antibody against TMP was produced with limit of detection (LOD) of 6 ng mL⁻¹ (Märtlbauer, Usleber, Schneider, & Dietrich, 1994). An ELISA based on antibodies produced from rabbit was developed with LOD of 12.5 µg kg⁻¹ in raw milk (Albrecht, Hammer, & Heeschen, 1996). However, there is still no report about the monoclonal antibody (mAb). The objectives of this study were to produce a sensitive and specific mAb and develop a mAb-based indirect competitive ELISA (ic-ELISA) for TMP residue detection in food samples.

Materials and methods

Chemicals and reagents

TMP, sulfadiazine, sulfamethazine, sulfamethoxazole, sulfamethizole, and 25% glutaraldehyde (GA) solution were obtained from J&K Scientific Ltd. (Beijing, China). Diaveridine, sulfadimethoxine, sulfathiazole, bovine serum albumin (BSA), ovalbumin (OVA), Freund’s complete and incomplete adjuvant, 3,3′,5,5′-tetramethylbenzidine (TMB), and polyethylene glycol 1500 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Enzyme immunoassay-grade horseradish peroxidase labelled goat anti-mouse immunoglobulin was supplied by Hua Mei Co. (Shanghai, China). RPMI-1640 cell culture medium, 50× HAT supplement (containing hypoxanthine aminopterin thymidine), 100× HT supplement (containing hypoxanthine thymidine), and fetal bovine serum were obtained from Thermo Fisher Scientific Inc. (Waltham, MA, USA). All other chemicals and solvents were of analytical grade.

Instruments

Absorbance measurements were performed with a spectrophotometric microtiter plate reader (Thermo), and UV spectra were determined with an UV–VIS spectrophotometer (Agilent, LA, USA). Pure water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA, USA).

Buffer solutions

The buffer solutions used in this study referred to our previous work with some modifications (Xu, Xu, Ma, Kuang, & Xu, 2015). (1) coating buffer (0.05 M carbonate buffer, CB, pH 9.6); (2) blocking buffer (0.05 M CB with 0.2% gelatin, w/v); (3) washing buffer (0.01 M phosphate buffered solution (PBS) with 0.05% Tween-20 v/v, PBS-T, pH 7.2); (4) assay buffer (0.01 M PBS, pH 7.2); (5) antibody diluent (0.01 M PBS-T with 0.1% gelatin, w/v); (6) citrate buffer (100 mL of 0.1 M citrate phosphate buffer containing 18 µL of 30% H₂O₂, pH 5.0); (7) TMB (60 mg TMB dissolved in 100 mL ethylene glycol); (8) substrate solution (5:1 v/v mixture of citrate buffer and TMB solution); and (9) Stop buffer (2 M H₂SO₄).

Synthesis of immunogens and coating antigens

Immunogens and coating antigens were synthesized via GA method, in which amino groups of TMP are covalently attached to carrier proteins (e.g. BSA or OVA) (Chen et al., 2015). Briefly, 13 mg of TMP previously dissolved in 3 mL of acetonitrile was mixed with 50 µL of 25% GA solution for reacting for 5 min. Then, the mixture solution was added dropwise into 100 mg of BSA or OVA dissolved in 10 mL of 0.01 M PBS, allowed to react for 2 h in room temperature, and dialyzed against PBS for 3 d. The formation of conjugates was confirmed by UV–VIS.

Immunization schedule

The immunization schedule was performed based on standard protocol (Yin, Liu, Song, Kuang, & Xu, 2015). Briefly, five BALB/c female mice were subcutaneously injected at multiple points with 100 µg of immunogen emulsified with Freund’s complete adjuvant. Four weeks after the initial injection, booster immunizations were performed with 50 µg of immunogen emulsified with Freund’s incomplete adjuvant. After the third immunization, titer and IC₅₀ of sera collected from mice were evaluated by ic-ELISA. The mouse with the highest titer and lowest IC₅₀ was screened for spleen donation for cell fusion. Twenty days prior to cell fusion, a final booster intra-peritoneal injection (25 µg of immunogen directly dissolved in 100 µL of physiological saline) was administered.

Cell fusion and hybridoma screening

Cell fusion and hybridoma screening were performed as previously reported (Kuang, Xu, Cui, Ma, & Xu, 2010). Firstly, spleen was removed from mouse with the highest titer and lowest IC₅₀ and ground to yield splenocytes. In the presence of PEG 1500, splenocytes were fused with mouse Sp 2/0 myeloma cells. The hybridoma cells distributed in 96-well plates were cultured in HAT medium supplemented with 20% bovine fetal serum for 7 d at 37°C and 5% CO₂. The optimum cell line with the highest titer and lowest IC₅₀ was screened by evaluating the supernatants via ic-ELISA. Sub-clones from the cell lines were intraperitoneally administered to mice primed with paraffin to obtain ascites. Monoclonal antibodies were obtained following the purification of ascites by the saturated ammonium sulfate method. The concentration of mAb was measured by UV–VIS spectroscopy at 278 nm. The mAb was labeled and stored at −20°C.

ic-ELISA

The most appropriate mAb concentration and suitable coating concentration for ic-ELISA were determined by bi-dimensional titration assays (Jiang et al., 2011). Briefly, 100 µL of TMP-OVA serially diluted in coating buffer was coated on microtiter plates for 2 h at 37°C. Subsequently, the plates were blocked washed and blocked for 2 h at 37°C. Blank assay buffer (50 µL) and diluted mAb (50 µL) were added into every well and incubated for the antigen-antibody reaction for 30 min at 37°C. The reaction was terminated by washing steps. Peroxidase-labeled goat anti-mouse IgG (100 µL, diluted 3000 times with antibody diluent) was added and incubated for 30 min at 37°C. Following the 30-min incubation, the plates were washed to remove excess peroxidase-labeled goat anti-mouse IgG. TMB substrate solution (100 µL) was added, and the plates were incubated for 15 min at 37°C. Finally, the reaction was stopped with 50 µL of 2 M H₂SO₄. Optical density (OD) was measured at 450 nm in a microplate reader. The procedure of bi-dimensional titration assay was similar to ic-ELISA, except that the blank assay buffer was replaced with diluted analyte. ODs were plotted against the logarithm of analyte concentration to obtain the standard inhibition curve.

Sensitivity and specificity

IC₅₀, which represents the concentration of competing compound that produced a 50% inhibition of antibody binding to the coating antigen, is an important indicator of mAb sensitivity (Jiang et al., 2011). The concentration that resulted in 20–80% inhibition and 15% inhibition were defined as detection range and LOD of ic-ELISA, respectively (Wang, Zhang, Gao, Duan, & Wang, 2010).

Specificity was defined as the ability of structurally related compounds to combine with antibody. Generally, cross-reactivity (CR) was measured to assess specificity. CR was calculated according to the following equation (Liu, Kuang, Peng, Wang, & Xu, 2014):

Seven analogues (diaveridine, sulfadiazine, sulfamethazine, sulfadimethoxine, sulfathiazole, sulfamethoxazole, and sulfamethizole) were used to evaluate the specificity of mAb. These compounds, previously dissolved in dimethyl formamide at 1 mg mL⁻¹, were diluted to 5, 10, 20, 50, 100, and 200 µg L⁻¹ with PBS.

Sample pretreatment

Three different food matrices were used in this study: milk, honey, and fish tissue (crucian carps). TMP-negative samples confirmed by HPLC were provided by Jiangsu Entry-Exit Inspection and Quarantine Bureau. Milk (5 mL) and honey (5 g) samples were spiked with TMP resulting in three different TMP concentrations (20, 40, and 80 µg L⁻¹ in milk and 20, 40, and 80 µg kg⁻¹ in honey). The TMP-spiked milk and honey samples were subsequently diluted 10× with 0.01 M PBS to reduce matrix effects, resulting in final TMP concentrations of 2, 4, and 8 µ L⁻¹ in milk and 2, 4, and 8 µg kg⁻¹ in honey.

Edible muscle from crucian carps was used. The fish samples were eviscerated, peeled, minced, and homogenized. Fish tissue (5 g) in a 50-mL centrifuge tube was spiked with TMP resulting in three different TMP concentrations (8, 16, and 32 µg kg⁻¹). The spiked fish tissues were placed in an ultrasonic bath for 5 min to thoroughly disperse TMP into the matrix (Cháfer-Pericás, Maquieira, Puchades, Miralles, & Moreno, 2010). Twenty-five milliliter of acetonitrile was added into the sample and mixed on a votex for 10 min. The mixture was centrifuged at 875 g for 5 min. Organic solvent in the supernatant was evaporated under a nitrogen stream. The remainder of samples were re-suspended in 4 mL of PBS-T containing 10% acetonitrile and filtered through 0.22 µm membrane (Cháfer-Pericás et al., 2010). This procedure diluted the fish tissue 4× to final TMP concentrations of 2, 4, and 8 µg kg⁻¹.

Results and discussions

Synthesis of immunogen

Small molecules such as TMP are incapable of triggering immune responses and eliciting antibody synthesis in animals. To be immunogenic, TMP has to be coupled to a carrier protein such as BSA or keyhole limpet hemocyanin; OVA is often used as a coating antigen. The three methoxy groups of TMP on benzene ring are unsuitable for conjugation. In contrast, the amino groups on pyrimidine ring are active and easily conjugate with the amino groups of carrier proteins. GA is a common homobifunctional crossing-linking agent that has been widely applied in conjugation reactions. The maximum absorbance peak of BSA/OVA is at 278 nm, and the characteristic absorbance peak of TMP is at 285 nm. As shown in Figure 1, the maximum absorbance of TMP-BSA and TMP-OVA obviously blueshifted to 266 nm, which confirms that the conjugates were successfully synthesized.

Figure 1. The UV–VIS absorption spectra of TMP-BSA (a) and TMP-OVA (b). The concentration of BSA, OVA, TMP-BSA, and TMP-OVA were 0.5 mg mL⁻¹.

Optimization of ic-Elisa

Certain parameters of the assay buffer, for example, organic solvent content, ionic strength, and pH, affect protein interactions and antigen-antibody reactions. Organic solvent contributes to the solubility of hydrophobic compounds. Excessive volumes of organic solvent negatively affect the antibody, leading to low OD values. As shown in Figure 2(a), the maximum OD varied with acetonitrile concentration. Under condition of 20% acetonitrile (v/v) in 0.01 M PBS, the sensitivity of antibody was poor. At 10% acetonitrile (v/v) in 0.01 M PBS, OD values and sensitivity were optimum. As shown in Figure 2(b), IC₅₀ was the lowest and OD was low at 3.2%. Therefore, 0.8% ionic strength was optimal. As shown in Figure 2(c), the maximum OD was <1 at pH 9.6. At pH 7.2, OD was the highest and IC₅₀ was the lowest. Figure 2(d) shows the standard sigmoidal inhibition curve of TMP under optimized conditions. The results revealed that the IC₅₀ value and quantitative detection range (IC₂₀–IC₈₀) were 4.14 and 1.83–9.36 µg L⁻¹, respectively.

Figure 2. Optimization of assay buffer for ic-ELISA: (a) Effect of acetonitrile content in PBS on ic-ELISA performance; (b) Effect of ionic strength of assay buffer on ic-ELISA performance; (c) Effect of pH value of assay buffer on ic-ELISA performance; (d) Standard curve of inhibition under the optimized conditions. Coating antigen: 0.2 µg mL⁻¹; mAb: 1: 32000; standards: 0.5, 1, 2, 5, 10, 20, and 50 µg L⁻¹. Each point presents the mean of ± SD of three replicates.

Cross-reactivity

There was no CR between the analogues and TMP, which illustrated that the produced mAb was specific to TMP (Table 1). The difference in chemical structure between TMP and diaveridine was an extra methoxy group on the benzene ring. However, diaveridine has no CR with TMP, which proved that the methoxy groups were the antigenic determinant.

Table 1. CR of mAb.

Compound	Structure	IC50 (µg L^−1)	CR (%)
TMP		4.14	100
Diaveridine		>50	<10
Sulfadiazine		>50	<10
Sulfamethazine		>50	<10
Sulfadimethoxine		>50	<10
Sulfathiazole		>50	<10
Sulfamethoxazole		>50	<10
Sulfamethizole		>50	<10

Recovery tests

Dilution is commonly used to reduce the matrix interferences; however, this procedure reduce the sensitivity. In this study, the sensitivity of ic-ELISA still satisfied the recovery tests after dilution. The fish tissues were treated with organic buffers to extract TMP residues. Acetonitrile, which is poorly tolerated in immunoassays, was completely removed. The four-fold dilution can obviously reduce the matrix effects. As shown in Table 2, the average recoveries of TMP spiked in milk, honey, and fish tissue were 103.5–118.0%, 100.3–107.1%, and 91.5–108.1%, respectively, which revealed that mAb-based ic-ELISA can be used for TMP residue analysis.

Table 2. Recoveries of TMP from spiked samples by ic-ELISA.^a

Matrix	Spiked TMP (µg L^−1 or µg kg^−1)	Mean ± SD^b (µg L^−1 or µg kg^-1)	Intraassay recovery (%)	CV^c (%)	Mean ± SD (µg L^−1 or µg kg^−1)	Interassay recovery (%)	CV (%)
Milk	20	21.72 ± 0.32	108.6 ± 1.5	1.47	19.57 ± 0.27	97.9 ± 1.4	1.38
	40	47.18 ± 1.12	118.0 ± 2.8	2.25	42.19 ± 1.31	105.5 ± 3.3	3.11
	80	82.83 ± 2.81	103.5 ± 3.5	3.39	90.15 ± 4.51	112.7 ± 5.6	5.00
Honey	20	20.05 ± 0.45	100.3 ± 2.3	2.24	19.81 ± 0.23	99.1 ± 1.2	1.16
	40	42.85 ± 2.16	107.1 ± 5.4	5.04	45.27 ± 1.64	113.2 ± 4.1	3.62
	80	84.07 ± 3.13	105.0 ± 3.9	3.72	87.90 ± 2.95	109.9 ± 3.7	3.36
Fish tissue	8	7.32 ± 0.35	91.5 ± 4.4	4.78	8.27 ± 0.31	103.4 ± 3.9	3.74
	16	17.31 ± 0.53	108.1 ± 3.3	3.06	15.45 ± 0.69	96.6 ± 4.3	4.47
	32	31.92 ± 1.93	99.8 ± 6.0	6.05	33.92 ± 2.12	106.0 ± 6.6	6.25

^aFor each concentration, three samples were spiked and analyzed by ic-ELISA.

^bSD, standard deviation.

^cCV, coefficient of variance. Interassay precision was determined by using same batch of microtiter plates (n = 3). For interassay, three batches of microtiter plates were used to estimate the samples. The recovery and CV are calculated from triplicate assays.

Conclusions

In this paper, a sensitive and specific mAb against TMP was produced for the first time, and a rapid and simple ic-ELISA was developed for the detection of TMP in milk, honey, and fish samples. The simple sample pretreatment steps would allow the rapid screening of a large number of samples. Based on the results, mAb had excellent working range (1.83–9.36 µg L⁻¹), which can be employed for commercial ELISA kits and gold immunochromatographic strips.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Yanni Chen got her bachelor’s degree from Liaoning Medical University, Jinzhou, China, in 2012 and then she began to study in Jiangnan University (Wuxi, China) as a Ph.D. student in Food Science. Her research interest includes immunoassay applications in food.

Liqiang Liu got his Ph.D. in Food Science in 2014 from Jiangnan University, Wuxi, China, and then became a faculty in College of Food Science and Technology of Jiangnan University. His research interests are immunochromatographic strip design and application.

Shanshan Song got her master’s degree in Food Science in 2012 from Jiangnan University, Wuxi, China, and then became a research assistant in College of Food Science and Technology of Jiangnan University. Her research interest includes monoclonal antibody development.

Hua Kuang got her Ph.D. from China Agricultural University in 2009 and then began to work as a faculty in College of Food Science and Technology of Jiangnan University. She is currently a full professor in food safety. Her research interest includes biosensor development.

Chuanlai Xu is a full professor of Food Science and Technology of Jiangnan University. He got his Ph.D. in Food Science in 2002. His research interests are fast detection technology and food safety evaluation.

Additional information

Funding

This work is financially supported by Natural Science Foundation of Jiangsu Province, MOF and MOE [grant number BK20140003, BE2014672, 201310135].