Immunochromatographic test strip for the rapid detection of tricaine in fish samples

ABSTRACT

Tricaine (TMS) is usually used as an anesthetic in the transportation of aquatic animals. However, when consumed, the numerous residual drug residues found in fish can cause a high risk to human health. Therefore, it is important to develop a rapid and sensitive method for the detection of this drug in food. Here we have developed a novel lateral-flow immunochromatographic strip (LF-ICS) method for the detection of TMS in fish samples, using a very effective anti-TMS monoclonal antibody with a half-maximal inhibitory concentration value of 31.79 ng/mL. After the K₂CO₃ and coating antigen concentrations were optimized in phosphate buffered saline (pH 7.4), the LF-ICS was used to determine TMS levels in fish samples spiked with varying concentrations of the drug. Following this, the values for the visual limit of detection and the cut-off for TMS were found to be less than 0.1 and 1 μg/mL, respectively. These results indicated that our developed LF-ICS is applicable for on-site screening of TMS in fish samples.

KEYWORDS:

• Anesthetic
• fish samples
• immunochromatographic
• strips
• tricaine

Introduction

Tricaine methanesulfonate (TMS), 3-Aminobenzoic acid ethyl ester methanesulfonate, also known as MS-222, finquel or metacaine, is a synthetic ester-type anesthetic with excellent water solubility (Figure 1). It is commonly used in the fisheries industry and can keep the organism in a state of low energy consumption, low sensory perception and slowed metabolism, by acting on central and peripheral nervous systems (Carter et al., 2011; Su et al., 2015). TMS is an ideal anesthetic, as it has a rapid effect and allows the fish to recover quickly (Cakir & Strauch, 2005; Cho & Heath, 2000). The drug is widely used in the form of an aqueous solution to sedate fish for handling or transportation, which can effectively reduce casualties and ensure freshness of the fish. Many researchers also use TMS to immobilize fish during invasive surgical procedures (Küçük, 2010; Palmer & Mensinger, 2004; Rombough, 2007).

Figure 1. The molecular structure of TMS and TMS-1.

Despite the outstanding advantages of TMS as an aquatic anesthetic, the drug residues it produces in fish destined for human consumption can produce detrimental effects to human health, including endocrine disorders, allergic responses, and altered cardiovascular function (Carter et al., 2011; Vera et al., 2010). Guidelines and regulations for TMS use vary among countries, for example, TMS is not approved by France, Greece, Spain, or China for public health reason (Su et al., 2015; Topic Popovic et al., 2012), but is allowed as an anesthetic for use in certain types of fish in the United States and in several other countries (Carter et al., 2011). In addition, the fish anesthetized by TMS are usually required to undergo a withdrawal period before slaughtering, which is 5, 10, and 21 days in Canada, New Zealand, and the United States, respectively (Topic Popovic et al., 2012). In order to reduce the harm of TMS residues to humans, it is vital to develop an effective method for monitoring the drug in fish. Previous research looking at the detection of TMS has used various methodologies including liquid chromatography (Nochetto et al., 2009), high performance liquid chromatography-tandem mass spectrometry (Li et al., 2014; Scherpenisse & Bergwerff, 2007; Xie et al., 2019), and liquid chromatography-tandem mass spectrometry (Zhao et al., 2017), with their advantage of accuracy and high sensitivity. However, these methods are not suitable for the detection of TMS in multiple samples because they are time consuming are of low-throughput, require complicated sample pre-treatment processes and are expensive. Therefore, the development of a rapid and sensitive method for the analysis of TMS is of great importance.

In recent years, immunoassays based on specific binding of antigen and antibody have been widely used for the analysis of harmful substances in food due to their obvious advantages as they are rapid, inexpensive, and simple to use (Ahmed et al., 2020; Wang et al., 2017b). The lateral-flow immunochromatographic strip (LF-ICS), is a novel detection method combining nanoparticles and enzyme-linked immunosorbent assay (ELISA) and is simpler and faster than ELISAs. To date, it has been successfully applied to the detection of several anesthetics, such as procaine (Hao et al., 2018), morphine (Chen et al., 2019), and eugenol (Shen et al., 2019). To the best of our knowledge, however, there are no studies using immunoassay methodologies to detect TMS residues.

In the present study, an anti-TMS monoclonal antibody (mAb) with high sensitivity and specificity was prepared, and then a LF-ICS method was developed based on the mAb for the detection of TMS in fish.

Materials and methods

Chemicals

TMS and 3-Aminobenzoic acid (TMS-1) were obtained from J&K Scientific Ltd (Beijing, China). Enzyme immunoassay-grade horseradish peroxidase-labelled goat anti-mouse immunoglobulin was purchase from Kangcheng Bioengineering Co. (Shanghai, China). Incomplete Freund’s adjuvant, complete Freund’s adjuvant, Keyhole limpet hemocyanin (KLH; MW 4,500,000 Da), ovalbumin (OVA; MW 45,000 Da), bovine serum albumin (BSA; MW 67,000 Da), 3, 3′, 5, 5′-tetramethylbenzidine (TMB), Tween-20, gelatin, Tributylamine, and Isobutyl chloroformate were purchased from Sigm-Aldrich (St. Louis, MO, USA). Life Technologies Corporation (Shanghai, China) supplied all cell culture reagents (e.g. polyethylene glycol solution, 1640 cell culture medium, fetal bovine serum, hypoxanthine aminopterin thymidine supplement, and hypoxanthine thymidine supplement).

The LF-ICS materials, including the glass fibre membrane acting as the sample pad, the polyvinylchloride plate serving as backing support material, and H5076 filter paper used for the absorbent pad were all supplied by Goldbio Tech Co., Ltd. (Shanghai, China). Nitrocellulose (NC) membranes used to hold the antigen and anti-IgG antibody were provided by Whatman-Xinhua Filter Paper Co., Ltd. (Hangzhou, China). In addition, the coating antigens linked with ovalbumin (OVA; TMS-1-OVA) and specific nanogold-labelled antibody (anti-TMS) were prepared in our laboratory.

Solutions

For the preparation of phosphate buffered saline (PBS), 8 g of NaCl, 0.2 g of KCl, 1.44 g of Na₂HPO₄, and 0.24 g of NaH₂PO₄ were dissolved in purified water to make 1L (PBS, 0.01 M, pH 7.4). The washing buffer solution was made by adding 0.05% Tween-20 (v/v) to PBS (0.01 M, pH 7.4). Next 1.59 g of Na₂CO₃ and 2.93 g of NaHCO₃ were dissolved into 1 L of purified water to make the coating buffer solution (CBS, 0.05 M, pH 9.6). The substrate solution consisted of 80% solution A (citric acid, Na₂HPO₄, and H₂O₂) and 20% solution B (0.06% v/v TMB in glycol). The stop solution was 2 M sulfuric acid and the blocking buffer was CBS (0.05 M, pH 9.6) containing 2% gelatin. Ultrapure water was generated from a Milli-Q Purification System (Millipore Co., Bedford, MA, USA).

Preparation of immunogen and coating antigen

Immunogen TMS-1-KLH

To prepare TMS-1-KLH, hapten of 3-aminobenzoic acid (TMS-1, Figure 1) was selected to bind to KLH using the dicyclohexylcarbodiimide (DCC) method (Kong et al., 2017; Wang et al., 2017a). Specifically, 6 g of TMS-1 was dissolved in 0.6 mL of N,N-dimethylformamide (DMF), and then 20 mg of N-Hydroxysuccinimide and 36 mg of DCC were added. After mixing at room temperature for 4 h, the solution was added dropwise into 10 mg of KLH solution diluted with an equal volume of CBS (0.05 M, pH 9.6), and the solution maintained at pH 9.0 with 0.5 M NaOH. The solution was continuously stirred at room temperature for 24 h, and then dialyzed against 0.01 M PBS for 3 days (The PBS was changed every 8 h). Finally, the conjugate was confirmed using a UV scanner.

Coating antigen TMS-1-OVA

The synthesis of the coating antigen TMS-1-OVA was performed using the mixed-anhydride method as described in our previous study (Lei et al., 2018b). Briefly, 4 mg of TMS-1 was dissolved in 0.4 mL of DMF, then 8.3 μL of tributylamine and 4.5 μL of isobutyl-chloroformate were added at intervals of 15 min. After stirring for 1 h, the solution was added dropwise to 7 mg OVA dissolved in 1.4 mL of 0.05 M CBS and the reaction was left for 6 h. All of the above reactions were performed at 4°C and the solution was dialyzed and confirmed as described earlier.

Preparation of mAb against TMS

BALB/c mice (females, aged 6–8 weeks) were immunized with the immunogen TMS-1-KLH by subcutaneous multisite injections, and specific immunization methods are as described previously (Guo et al., 2015; Lei et al., 2018a). After the sixth immunization, the mice were evaluated using an indirect competitive ELISA (ic-ELISA). Relevant mice, as judged by the titre and half-maximal inhibitory concentration (IC₅₀), were injected with 20 μg of the immunogen to the enterocoelia 3 days before sacrifice, and the spleen cells were extracted and fused with Sp2/0 myeloma cells isolated from respective mice. The hybridoma cells were injected into the mouse abdominal cavity and were inactivated with paroline; the ascites including anti-TMS mAb were collected a week later. Finally, the mAb against TMS was purified from ascites using the saturated ammonium sulfate method and tested by ic-ELISA as described by Ye et al. (2018).

Development of the LF-ICS test device

Colloidal gold nanoparticle (GNP) synthesis

The synthesis of the GNPs was performed as described by Chen et al. (2017a) Briefly, 50 mL of 0.1 g/L chloroauric acid solution in a clean conical flask was heated to boiling and continuously stirred. Then, 2 mL of fresh trisodium citrate solution (1% w/v) was rapidly added and stirring was continued. The solution was kept boiling until it appeared a wine-red colour, then the heat source was removed and allowed to cool to room temperature and the GNPs were stored at 4°C.

Preparation of colloidal gold-labelled mAb

The method of labelling anti-TMS mAb with GNPs was similar to previous studies (Chen et al., 2016; Wang et al., 2020). Firstly, anti-TMS mAb was diluted to 0.2 mg/mL with 0.01 M CBS, and the GNPs were adjusted to pH 7.0 with 0.1 M K₂CO₃. Then, 0.2 mL of the mAb was added slowly to the GNP solution (6 mL) and allowed to react for 50 min at room temperature. Next, 0.5 mL of 10% (m/v) BSA was added to block those GNP surfaces not bound by antibody, and 2 h after this the solution was centrifuged at 8000 rcf for 40 min. The sediment at the bottom was then washed two times using 0.02 M PBS solution (pH 7.4, containing 1% BSA, 5% sucrose, and 0.5% polyethylene glycol 6000) to remove excess reagent. Finally, the gold labelled anti-TMS mAb product was redissolved in 3 mL of basic resuspension buffer (0.02 M PBS, 0.1% polyethylene glycol, 2% sorbitol, 5% sucrose, 0.04% NaN₃, 0.1% Tween, and 1% mannitol) and stored at 4°C.

Preparation and principle of the LF-ICS

The LF-ICS was prepared based on established methodology derived from our laboratory (Chen et al., 2017b; Guo et al., 2018). As Figure 2(a) shows, the LF-ICS is comprised of a support plate, absorbent pad, NC membrane, and sample pad, which are assembled layer by layer in order. There are two capture lines immobilized on the NC membrane, a test line (T-line) sprayed with antigen TMS-1-OVA to capture the free gold-labelled mAb and a control line (C-line) sprayed with goat anti-mouse IgG antibody to capture both gold-labelled mAb and the conjugate of TMS and gold-labelled mAb. After this, the capture lines were dried for 4 h at 37°C, and then cut into strips of 2.8 mm diameter. Finally, the LF-ICS was stored at room temperature for future use.

Figure 2. Schematic illustrations of the LF-ICS. (a) The structure of the LF-ICS, (b) The principle of the LF-ICS test for TMS.

The principle of the LF-ICS is based on competitive binding between the target analyte and the coating antigen to gold-labelled mAb. As shown in Figure 2(b), the sample solution is mixed with the gold-labelled mAb in microwell plates, and incubated for 5 min at 37°C. When the strip is inserted into the well, the solution flows upwards towards the absorption pad by capillary action, and the result can be observed 7 min later. If the analyte in the sample solution is negative, a red band will appear on the T-line as a result of the binding of the gold-labeled mAb and coating antigen. If the analyte in the sample solution is positive, the red bands on the T-line will be weak because the free colloidal gold-labeled mAb is reduced after reacting with the analyte, and therefore the colour becomes lighter as the analyte increases. The C-line is always red even if analyte in the sample solution is absent as the goat anti-mouse IgG antibody captures not only excess colloidal gold-labeled mAb, but also the compounds within the analyte and the antibody. This implies that the LF-ICS is invalid when the C-line is colourless.

Determination of LF-ICS performance

Sensitivity of the test strip

In this study, the sensitivity of the LF-ICS was measured by using several TMS standard dilutions (0.1, 0.25, 0.5, 1, 2.5, and 5 μg/mL). For each standard 150 μL was added to 50 μL of the gold-labelled mAb solution in the microwell plates and mixed for 5 min at 37°C. Then the strips were put into the solutions, and the results were observed 7 min later. For the LF-ICS, the visual limit of detection (vLOD) was defined as the minimum concentration of TMS producing a visible T-line visible when compared to the negative control group. The cut-off value was defined as the minimum concentration of TMS in which the T-line was colourless.

Sample preparation and analysis using the LF-ICS

TMS-negative fish samples, confirmed by HPLC-MS/MS, were used to test the LF-ICS in this study by adding TMS standards. Specifically, 0.5 mg/mL of TMS stock solution was prepared by dissolving TMS into DMF, and then gradually diluting to 0.2, 0.1, 0.05, and 0.02 mg/mL with DMF. Ten microliters of the above five TMS dilutions and a blank DMF were added to 2 g of negative fish samples respectively. After this, the spiked fish samples were extracted with 10 mL of ethyl acetate using a rotating shaker for 10 min in centrifuge tubes and then centrifuged for 10 min at 8000×g. The supernatants were collected and dried over flowing nitrogen at 50°C. Finally, the extracts were dissolved in 10 ml of PBS (0.5% Tween-20, v/v) and used for the LF-ICS analysis.

Results and discussion

Characterization of antigens and antibody

In view of the weak immunogenicity displayed by small molecules, TMS was coupled with KLH and OVA as the immunogen and coating antigen respectively. As shown in Figure 3(a), the UV peak shape for TMS-1-KLH at 280 and 355 nm had changed significantly when compared to KLH, which can be influenced by TMS-1 attached to KLH. Similarly, TMS-1-OVA was also significantly different from OVA at 280 nm (Figure 3b). Therefore, we had confirmed that TMS-1-KLH and TMS-1-OVA had been prepared successfully.

Figure 3. Ultraviolet−visible absorption spectra of immunogen TMS-1-KLH (a) and coating antigen TMS-1-OVA (b).

It is well known that a good monoclonal antibody is essential for the preparation of test strips, so our excellent anti-TMS mAb was screened and characterized by ic-ELISA. As shown in Figure 4, a standard curve was created based on 0.1 μg/mL of TMS-1-OVA and 0.05 μg/mL of the mAb, and the corresponding regression equation used was y = 0.28 + 1.47/(1 + 100x/3179)^0.95. The IC₅₀ of TMS was 31.79 ng/mL and the linear range was 7.38−136.95 ng/mL (IC₂₀ to IC₈₀).

Figure 4. Standard curve for TMS in ic-ELISA test.

Optimization of the test strip

In this study, the performance of the LF-ICS was optimized by testing different volume of 0.1 M K₂CO₃ added to the GNP solution and the concentration of TMS-1-OVA sprayed onto the T-line. The coating antigen concentration is closely related to the T-line colour and LF-ICS sensitivity, and we found that the red band was invisible when too few coating antigens were immobilized on T-line, whereas too much coating antigen lead to low sensitivity of the LF-ICS seeing that some antibodies were bond to the excess coating antigens fixed weakly on T-line and remove easily by capillarity. Therefore, different TMS-1-OVA concentrations (0.3 and 0.8 mg/mL) were used to test the strips using the TMS standard solutions (0 and 2.5 μg/mL). As shown in Figure 5, 0.8 mg/mL of TMS-1-OVA resulted in obvious red bands in the TMS-negative samples and a striking contrast was seen between 0 and 2.5 μg/mL of TMS standard solution on the T-line.

Figure 5. Optimization of the LF-ICS in 0.01 M PBS solution. The concentration of TMS-1-OVA (A) 0.3 mg/mL, (B) 0.8 mg/mL. The volume of K₂CO₃ (a) 8 μL, (b) 10 μL. The concentration of TMS standards: (1) 0 μg/mL, (2) 2.5 μg/mL.

Regardless of the amount of antibody, the main factor affecting assay performance was the amount of 0.1 M K₂CO₃ which influenced the characteristics of the GNP mAb by adjusting the pH of the solution. Specifically, the positively charged mAb (IgG) can tightly bind to the negatively charged GNPs by electrostatic forces in adaptive alkaline conditions. The result (Figure 5) showed that the colour intensity and sensitivity of the strips were well defined at 0.8 mg/ml of TMS-1-OVA, but the amount of 0.1 M K₂CO₃ (8 and 10 μL) had no significant effect on the performance of the LF-ICS. This indicated that the GNP mAb was well synthesized and stable in the solutions used both for 8 and 10 μL of 0.1 M K₂CO₃.

The sensitivity of the LF-ICS assay was optimal when 0.8 mg/ml of TMS-1-OVA was sprayed onto the T-line and 8 μL of 0.1 M K₂CO₃ was added to the GNP solution. The results are shown in Figure 6, the colour intensity of the red bands on the T-line gradually faded away with increasing TMS from 0 to 1 μg/mL and vanished completely at 2.5 μg/mL. Therefore, the values for the vLOD and the cut-off value of the LF-ICS for TMS were less than 0.1 and 2.5 μg/mL respectively in 0.01 M PBS (pH 7.4).

Figure 6. The sensitivity analysis of the LF-ICS in 0.01 M PBS solution. The concentration of TMS standards: (1) 0 μg/mL, (2) 0.1 μg/mL, (3) 0.25 μg/mL, (4) 0.5 μg/mL, (5) 1 μg/mL, (6) 2.5 μg/mL, (7) 5 μg/mL.

Detection of TMS in fish samples

Unlike PBS solutions, residues in fish samples and organic solvents left in the extraction solution affect capillarity action and the binding of antigens and antibodies in the LF-ICS. To reduce this matrix interference, the LF-ICS was optimized by adding six different surfactants (1% of polyethylene glycol, polyvinyl pyrrolidone, BSA, Tween-20, Brij-30, and On-870) to the basic resuspension buffers, and tested using TMS-negative fish extract (0 μg/mL) and TMS-positive fish extract (2.5 μg/mL). As shown in Figure 7, surfactants enhanced all colour intensities on the C-line of the LF-ICS tested with fish samples. The LF-ICS used with polyvinyl pyrrolidone performed best when compared to the other surfactants, based on colour intensity on the T-line and sensitivity.

Figure 7. Optimization of 1% surfactants in fish sample. Surfactants: (a) polyethylene glycol, (b) polyvinyl pyrrolidone, (c) BSA, (d) Tween-20, (e) Brij-30, (f) On-870. The concentration of TMS standards: (1) 0 μg/mL, (2) 2.5 μg/mL.

Sensitivity is an important feature for the evaluation of samples using LF-ICS. In this study, we tested our LF-ICS using spiked fish samples. As shown in Figure 8, TMS in the spiked samples weakened the colour intensity on T-line by competition when compared to the negative sample (0 ng/mL). The value of the vLOD of the LF-ICS for TMS were less than 0.1 μg/mL in fish samples, and the cut-off was 1 μg/mL.

Figure 8. The sensitivity analysis of the LF-ICS in fish samples. The concentration of TMS standards: (1) 0 μg/mL, (2) 0.1 μg/mL, (3) 0.25 μg/mL, (4) 0.5 μg/mL, (5) 1 μg/mL, (6) 2.5 μg/mL.

Conclusion

In general, an excellent mAb against TMS was produced based on the hapten TMS-1, and a novel LF-ICS method was developed to detect TMS in fish samples. The amount of K₂CO₃ added to the GNP solution and the concentration of TMS-1-OVA sprayed onto the T-line were optimized with the aim of strengthening the performance of the LF-ICS in 0.01 M PBS. Surfactants were also optimized to minimize matrix interference in assay samples. The fish samples only need a simple preparation procedure before the LF-ICS test, and results can be obtained within 15 min using the naked eye. As a simple, fast, high-throughput and reliable detection method for TMS, the LF-ICS is a promising assay and could be widely applied to the on-site screening of large numbers of fish samples in the future.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work is financially supported by National Key R&D Program [grant number 2018YFC1602604].