Ammonium bicarbonate (NH₄HCO₃), dithiothreitol (DTT), and iodoacetamide (IAA) were supplied by Sigma-Aldrich (St. Louis, MO, USA). TM was prepared and purified in the laboratory. HPLC grade acetonitrile and formic acid were purchased from Merck (Darmstadt, Germany). Tryptase was got from AB Sciex. Distilled water (Watsons, Hong Kong, China) was used in the experiment. Two signature peptides and the internal standard of ANIQLVEK, IVELEEELR, and ANIQL (¹³C₆,¹⁵N)VEK were synthesized by Qiangyao Biotechnology Company (Shanghai, China). The purity of all the synthetic peptides was higher than 98%.

HAT, HT, Freund's complete adjuvant, Freund's incomplete adjuvant, PEG4000, and immunoglobulin subtype kit were purchased from Sigma; cell culture plate and DMEM medium were purchased from GIBCO company; HRP-labeled sheep anti-mouse IgG was purchased from Beijing ZhongShan Gold Bridge Biotechnology Limited Company (Beijing, China). Bovine serum albumen (BSA), ovalbumin (OVA), and fetal bovine serum were supplied by Shanghai Shenggong; caprylic acid and ammonium sulfate with analytical grade were obtained from Komil Chemical Reagent Limited Company (Tianjin, China). BALB/c mice aged 6 to 9 weeks were purchased from the Laboratory Animal Center of Hebei Medical University.

Low protein adsorption sample bottles were supplied by Waters. Triple Quad 6500⁺ liquid chromatography–tandem mass spectrometry equipped with MultiQuant 3.0.1 data processing system (AB Sciex, USA) was used for allergen quantification analysis. Orbitrap Fusion equipped with a nanoliquid chromatography system (Easy-nLC 1000) (Thermo Scientific, USA) was adopted for allergen identification.

Carbon dioxide incubator (Sanyo Company, Japan), purification table (Suzhou Purification Equipment limited company, China), inverted microscope (Olympus, Japan), electronic analytical balance (Mettler, Switzerland), liquid nitrogen tank (Chengdu Jinfeng Liquid Nitrogen Container limited company, China), high-speed refrigerated centrifuge (Beckman, Germany), electric thermostatic water bath (Shanghai Jinghong Equipment limited company, China), microplate reader (Bio-Tek, USA), vortex mixer (Shanghai Medical University Instrument Factory, China), and ultramicro nucleic acid protein tester (Thermo, USA) were used in our experiment.

Proper amount of ANIQLVEK, IVELEEELR, and ANIQL (¹³C₆,¹⁵N)VEK were accurately weighed and dissolved in water to 10 ml, and 100 μmol/L mixed standard storage solution was prepared. An appropriate amount of TM was weighed and dissolved in a 10 ml volumetric flask, dissolving it in water to a constant volume, and then 100 μmol/L TM standard storage solution was made.

Three female BALB/c mice aged 6–8 weeks were selected for the antibody preparation. TM was diluted with normal saline and injected into the neck and back with an equal volume of Freund's complete adjuvant. The immunization was enhanced every 2 weeks. Freund's incomplete adjuvant of equal volume was added for the second and third immunization. Blood samples were collected after 7–10 days for the third immunization sessions, and indirect ELISA was used to detect the serum antibody titer.

SP2/0 myeloma was fused with spleen cells as described by Yang et al. (26). High-specificity and stable monoclonal cell strains against TM were repeatedly screened out by confirmation detections. The cell strains were cultured to achieve a certain quantity and then injected into atoleine-pretreated BALB/c mice to produce ascites fluids. The ascites fluids were purified according to the octanoic acid–ammonium sulfate method. The TM antibody was freeze-dried and stored at −20°C, and the antibody was thawed and diluted with PBS prior to use. All animal experiments were carried out according to the guidelines of the National Institutes of Health for Care and Use of Laboratory Animals and was approved by the Animal Ethical and Welfare Review Committee of Hebei Food Inspection and Research Institute.

The antibody subtypes were determined using Sigma's immunoglobulin subtype kit. Extract of shrimps, Argentine red shrimps, Penaeus vannameri, and bread crab were respectively coated and used to determine the crossreaction of monoclonal antibody by indirect ELISA. The ELISA procedure was as follows: appropriately diluted antigen with a concentration of 5 μg/ml was added to a 96-well ELISA plate (100μl/well) and coated overnight at 4°C, where carbonate buffer was used as the coating solution. After three washings, the wells were blocked with phosphate-buffered saline (PBS) containing 10% gelatin (300 μl/well) at 37°C for 2 h. After washing, 100 μl of the sample extract was added, and the plates were incubated at 37°C for 45 min. Blank control well (PBS) and negative well (negative serum) were set. After washing, 300 μl of 1:10,000 diluted HRP enzyme-labeled goat anti-mouse IgG was added to each well and placed at 37°C for 30 min. After washing, 100 μl of substrate-developing solution was added to each well and reacted at 37°C for 15 min away from light. The A450nm value was measured after the reaction was terminated by the stop buffer.

Refer to the instructions of CNBr-activated Sepharose 4B gel, and some modifications were made.

The first step is washing. A total of 2 g CNBr activated Sepharose 4B dry gel was suspended in 20 ml 1 mmol/L HCl to swell, and then the swollen gel was washed with 200 ml 1 mmol/L HCl and fully washed with 200 ml reaction buffer (0.1 mol/L NaHCO₃ pH 8.1).

The second step is coupling. The pretreated gel was quickly transferred to 10 ml of 2.7 mg/ml TM-antibody coupling buffer (0.1 mol/LNaHCO₃, 0.5 mol/L NaCl, pH 8.3), stirring overnight at 4°C with the purpose of conjugating TM monoclonal antibody to CNBr-activated Sepharose 4B adequately. The uncoupled antibodies were washed off with a coupling buffer of more than five times the volume of gel and all the eluents were collected. The content of uncoupled proteins in the eluent was determined by ultraviolet spectrophotometer, and the coupling rate was calculated.

The conjugation rate = 1-leached monoclonal antibodies/the added monoclonal antibodies. In the experiment, the conjugation rate between TM antibody and gel was 97.6%.

The third step is washing, where five times of the volume of 0.1 mol/L acetic acid buffer (pH 4.0, containing 0.5 mol/LNaCl) and 0.1 mol/L Tris-HCl (pH 8.0, containing 0.5 mol/LNaCl) were used to alternately wash the gel 4–6 times successively.

The last step is to install the column, where 1 ml column was used to prepare the immunoaffinity column and a sieve was placed into the column. PBS buffer solution (0.01 mol/L pH 7.4) was suspended and the coupling glue was loaded into the column until the glue height was 0.5 ml. PBS was balanced and the bottom of the mouth was sealed and stored in the refrigerator at 4°C.

Raw data obtained from Orbitrap Fusion was analyzed by the software Thermo Proteome Discoverer 1.4, and the relative parameters were set as follows: MS1 precursor was adopted in precursor selection; minimum precursor mass was set as 350 Da; maximum precursor mass was set as 5,000 Da; the minimum peak count was 1; protein database, shrimp database, and crab database were downloaded from Uniprot (http://www.uniprot.org); trypsin was used in enzymatic hydrolysis; maximum missed cleavage site was set as 2; minimum peptide length was 6; maximum peptide length was set as 144; the precursor mass tolerance was 10 ppm; fragment mass tolerance was 0.02 Da; oxidation (+15.995 Da) was selected in dynamic modification; and carbamidomethyl (+57.021 Da) was chosen in static modification.

Solid samples were finely ground before extraction as reported before (27). Then 0.1 g of the homogenized sample was weighed, 10 ml distilled water was added, the sample was shaken for 1 h, and centrifuged for 10 min with 9,500 g. Then 3 ml extract was taken for subsequent purification treatment.

The immunoaffinity column was taken out and the plug was removed to allow the liquid to flow out by gravity. A total of 10 ml phosphate buffer was used to balance the immunoaffinity column when the fluid in the column was no longer dripping. A total of 3 ml extract was sampled after the phosphate buffer was close to the sieve plate. Then 10 ml phosphate buffer (PH 7.4) was added to wash the impurities in the affinity column after the level of the sample liquid entered the affinity column. After the level of the sample liquid entered the affinity column, 10 ml phosphate buffer (PH 7.4) was added to wash the impurities in the affinity column.

After leaching, phosphate buffer was removed clearly with washing ears ball or cylinder. A total of 2.7 ml elution buffer (0.1 M of glycine buffer, pH 2.5) was added as the eluent solution by gravity flow rate drop out and the flow rate should be less than one drop each second. After all the liquid falls into the sieve plate, the eluent in the column was brought out by the ear ball or needle. A total 300 μl neutralization buffer (1 M Tris, pH 9.0) was added to the eluent, quickly mixed with the liquid collected in the test tube for subsequent enzymatic hydrolysis treatment.

Enzymatic digestion was conducted according to literature report (21). An aliquot of 500 μl sample solution was spiked with 25 μl of 25 nM stable isotope-labeled internal standard solution and then mixed with 10 μl of 500 mM DTT solution in a 60°C water bath for 30 min. An alkylation was performed with the addition of 30 μl 500 mM IAA solution at room temperature for 30 min in the dark. Thereafter, 100 μl 500 mM NH₄HCO₃ solution and 40 μl of 250 μg/ml trypsin were added and incubated overnight at 37°C in a water bath vibrator. Then, 20 μl 0.1% formic acid was added to terminate the digestion and 275 μl pure water was added to fill the volume to 1 ml. The sample was centrifuged at 14,000 g at room temperature for 20 min and the supernatant solution was collected for determination.

The statistical significance was analyzed through one-way analysis of variance by IBM SPSS Statistics 26. All experiments were performed at least three times.

The characteristics of the TM antibody were evaluated by indirect competition ELISA (28, 29). The serum titers of immunized mice were more than 1:4,00,000. Seven cell lines of anti-shrimp TM monoclonal antibody were screened, which were named as 1H1, 2C1, 2H11, 2F12, 3B2, 3H11, and 4B9, respectively. The protein concentration was measured by ultrafine nucleic acid protein analyzer and the titer was determined by indirect ELISA. The results showed that 2H11 had the highest titer, which was 1:1 million, the titer of 1H11, 3H11, and 4B9 was more than 1:250 000, and the titer of 2C1, 2F12, and 3B2 was more than 1:30 000. 2H11 was selected in the preparation of the subsequent immunoaffinity column.

The extracts of base shrimp, Argentine red shrimp, P. vannamei, and bread crab (labeled TMJ, TMA, TMN, TMM, respectively) and TM, 2H11 was diluted at 1:200, 1:50000, and 1:200000 times, respectively, and indirect ELISA was used to detect the crossreaction rate. The crossreaction rate was calculated by A450nm (sample)/A450nm (TM) × dilution ratio. The results were shown in Table 1. When the dilution factor was 200, the crossreaction rates were higher than 86.5%; when the dilution factor was 50,000, the crossreaction rate of TMJ and TMA was 77.5%, 39.2%, respectively, and the crossreaction rate of TMN and TMM was lower than 20%. When the dilution factor was 2,00,000, the crossreaction rate of TMJ was 53.2%, the crossreaction rate of TMA was 25.3%, and the crossreaction rate of TMN and TMM was lower than 10%. The results showed that when the dilution factor was low, the crossreaction rate was serious, which indicated the high homology of TM in different shrimp and crab. Significant differences existed in the crossreaction rates among different species at different dilution ratios, and the p-values were lower than 0.001.

To test the column capacity of self-made immunoaffinity column, we tested the recovery rate of different sample loading. As shown in Table 2, when the loading quantity of TM was in the range of 40–80 μg, the recovery rate was above 95%, and when the loading quantity reached 100 μg, the recovery rate was 79.1%. This is because the immunoaffinity column purification is based on the specific binding of antigen and antibody. Therefore, when the amount of antibody filled in the immunoaffinity column is fixed, the amount of antigen that can be adsorbed has an extreme limit. When all the specific binding sites of antibody are occupied, the specific binding of more antigens cannot be carried out. To ensure the purification effect, it is recommended that the sample quantity is not higher than 80 μg.

The selection of signature peptides is very important in the LC-MS/MS method development, which could influence the specificity and sensitivity of the method for the different ionization response of different peptides (24). Trypsin was used in enzymatic hydrolysis, where the cleavage site of trypsin was specifically at the C-terminal of lysine and arginine (30). When the length of peptide is lower than five amino acids, the analytical specificity is poor; when the peptides are too long, it is difficult and expensive to synthesize and the response of mass spectrometry is also unfavorable, so the peptide length is usually seven and 16 amino acids. To prevent possible chemical modifications, some susceptible amino acids, such as cysteine and methionine, should be avoided in the signature peptides selection. The selected peptide should be reproducibly observed and detectable in different states of samples, including the digested sample (31, 32).

Nanoliquid chromatography system tandem Orbitrap Fusion was used to analyze the peptide fragments after enzymolysis of TM. Raw data was analyzed by Thermo Proteome Discoverer 1.4. Twenty-eight peptides of TM were identified, and ANIQLVEK (AK-8), IVELEEELR (IR-9) were selected as signature peptides according to the principles selection.

Internal peptide (IP) was designed in our method. Based on the enzymatic digestion technique, homologous peptide was employed as the IP for measuring shrimp and crab allergen proteins. The isotopically labeled IP ANIQL (¹³C₆, ¹⁵N)VEK was designed and synthesized, in which all the carbon and nitrogen atoms in leucine (L) residues were labeled with ¹³C and ¹⁵N.

To optimize the parameters of mass spectrometry, standard solutions of synthetic peptides were directly injected into mass spectrometry by the syringe pump. Full scan mode was used to find the precursor ion of peptide fragment, the declustering potential was also optimized, the product ions were confirmed in product scan mode, and the collision energy was optimized. Three product ions were selected for each precursor ion and precursor-to-product ion transitions were detected by multiple reaction monitoring (MRM) mode. AK-8 was selected as the quantitative peptide for TM. All parameters of MS are shown in Table 3. The chromatographic-mass spectrograms of signature peptides and internal standard of TM are shown in Figure 1.

A total of 13 samples purchased from local supermarkets were used to determine the applicability of the method, including prawn, shrimp sticks, shrimp balls, crab stick, fish balls, etc. As shown in Table 6, TM was detected in Penaeus vannamei and China shrimp samples, with a concentration of 3,291 and 3,730 μg/g. TM was not detected in other samples. The results showed that the developed method could be used for the determination of TM in different kinds of food samples.

To compare relevant papers in the field, immunoaffinity purification was used in the procedure of sample pretreatment for first time. TM extraction, purification, antibody preparation, and the preparation of immunoaffinity column were finished by members of the research group. Signature peptides and isotope-labeled internal standard were used in the quantitative analysis. The method has the advantages of simple pretreatment, less interference, high specificity, and perfect accuracy and precision.