Parasites and animals
The T. spiralis muscle larvae (ML) used in the present study were obtained by acid-pepsin digestion  of infected ICR mice at the Department of Helminthology, Faculty of Tropical Medicine, Mahidol University. In the experiment, 200 ML per mouse were maintained for the new ICR mice, which were purchased from the National Laboratory Animal Center, Mahidol University, and then cared for at the Animal Care Center, Faculty of Tropical Medicine. T. spiralis ML were obtained from those infected mice after 2 months of infection by acid-pepsin digestion and were collected using Baermann’s apparatus .
CSAg was prepared from frozen T. spiralis ML. Briefly, larvae were mixed with lysis buffer  and then subjected to an ultrasonic apparatus (Ultrasonic Processor XL 2020, USA). In collection of the supernatant, the homogenate of the larvae was centrifuged at 13,000 g for 60 min at 4°C. The protein concentration was determined by Coomassie® Plus Protein Assay Reagent Kit (Pierce, USA) and stored at −80°C until use.
The eluted antigens (molecular weights [MWs] 101, 79, and 43 kDa, named TsCSAg-101, TsCSAg-79, and TsCSAg-43, respectively), following Tattiyapong’s study , were prepared by separating CSAg via SDS-PAGE and localized with E-Zinc™ Reversible Stain Kit (Pierce, USA) following the manufacturer’s protocol. Electroelution was performed as described previously, but with some modification . Briefly, each small piece of gel that contained an antigen was cut and transferred into each well of the electroelution apparatus (ATTO, Japan). Each well was filled with 3 mL enzyme-linked immunosorbent assay (ELISA) coating buffer (0.05 M carbonate-bicarbonate buffer, pH 9.6). The eluted proteins were removed from the bottom of the wells at a constant power of 30 mA, for 3 h. After that, the eluted antigens (TsCSAg-101, TsCSAg-79, and TsCSAg-43) were collected, and the sensitivity and specificity were determined by IgG-ELISA. The eluted antigens of the supernatant were prepared and analyzed by 2-DE and immunoblotting. Proteins spots were submitted to LC MS-MS for peptide identification.
Swine serum samples were obtained from a previous study  but with additional sera. These swine sera were divided into three categories: positive trichinellosis sera from experimentally infected swine (n = 5), other parasitic infections with additional diseases (n = 192), and normal control (n = 30). Swine trichinellosis sera were selected at an increasing antibody level on day 20, which was tested using CSAg of T. spiralis larvae by IgG-ELISA. Detailed information of the serum samples is presented in Supplementary Table S1 (see Additional file 1).
SDS-PAGE and immunoblot
These antigens were subjected to 14% SDS-PAGE. After electrophoresis, one gel was stained with ProteoSilver™ Silver staining (Sigma-Aldrich, USA), whereas the other gel was electroblotted onto a nitrocellulose membrane (0.45 µm, PALL, Sigma-Aldrich, USA) at 400 mA for 5 h via Semi-Dry Transfer Cell (ATTO, Japan). The immunoblot was performed as described previously, with some modifications . Subsequently, the membrane was blocked with 3% skim milk for 1 h on a rocking platform, which was cut into long strips and incubated with pooled swine serum at a dilution of 1:100 at room temperature, overnight. After washing with phosphate buffered saline 0.5% Tween 20 (PBST), strips were incubated with peroxidase-labeled affinity purified antibody to swine IgG (1:1,000 dilution) for 4 h. The immunoreaction was detected with a substrate solution (2,6-dichlorophenol indophenol). Finally, the membrane was washed with distilled water to stop the reaction. The separated antigens were analyzed by SDS-PAGE and immunoblot (Figs. S1a and b, see Additional file 2).
The protocol of the indirect ELISA followed a previously described method , but with minor modifications. The optimal antigen, dilutions of serum, and conjugate obtained from checkerboard titration were 1 µg/mL, 1:400, and 1:4,000, respectively. Each 50 µL of diluted antigen with coating buffer in the duplicate well was filled, incubated at 37°C for 1 h, and left overnight at room temperature. The unbinding sites with antigens in wells were blocked by 75 µL 0.5% skim milk in PBS at 37°C for 1 h. Antigens reacted with antibodies in 50 µL of diluted serum at 37°C for 1.30 h. Each 50 µL of diluted conjugate (peroxidase-labeled affinity purified antibody to swine IgG) was added and incubated as described above. Then, the color reaction was established by adding 50 µL of 2,2-azino-di-(3-ethyl-benzothiazoline sulfonate) containing H2O2 (ABTS) substrate and left for 30 min. The reaction was stopped by adding 75 µL 1% SDS, and the absorbance was determined by an ELISA microplate reader (Tecan, Austria) at 405 nm.
The cut-off value of IgG-ELISA was evaluated for the three eluted antigen preparations on the basis of a receiver operating characteristic analysis performed using GraphPad Prism version 5.01 for Windows (GraphPad Software, La Jolla, California, USA, www.graphpad.com).
2-DE and immunoblot
CSAg (100 µg) was adsorbed onto 7 cm immobilized pH gradient 3–10 NL (IPG) strips (GE Healthcare, USA) and then focused on an Ettan IPGphor 3 (GE Healthcare, USA). After the equilibration steps, IPG strips were placed onto 12% SDS-PAGE gels and run at 400 mAm, after which the gels were silver-stained. Alternatively, the separated spots were transferred to nitrocellulose membranes following standard protocols. Swine sera were used in 1:100 dilution. Immunoreactive spots were detected using goat anti-swine IgG antibody conjugated with horseradish peroxidase at 1:1,000 dilution and an immunoreaction detection system with a substrate solution.
Mass spectrometric analysis
Protein bands recognized by swine trichinellosis sera were excised from the parallel gel and subjected to in-gel tryptic digestion following the protocols of Reamtong et al. . Briefly, the gels were destained with 50 mM NH4HCO3 and 200 µL acetonitrile (ACN). Then, the colorless gels were incubated with 100 µL of 4 mM dithiothreitol (DTT) (GE Healthcare, UK) in 50 mM NH4HCO3 at 60°C for 15 min and alkylated with 250 mM iodoacetamide (IAM) (GE Healthcare, UK) in Milli-Q water at RT for 30 min in the dark. Next, reactions were quenched by 4 mM DTT in Milli-Q water and left for 5 min. Consequently, the gels were dehydrated with 100 µL ACN and shacked for 15 min. Afterward, the supernatant was removed, and the gels were completely gel dried in a fume hood. Subsequently, the gel pieces were digested with 10 ng/200 µL trypsin in 50 mM NH4HCO3 (Sigma-Aldrich, USA). The reactions were incubated overnight at 37°C. The amount of peptides in each small gel piece was achieved by extraction with 50% (V/V) ACN and dried using a vacuum evaporator (Labconco, USA). Then, the proteins were separated by high-performance liquid chromatography and analyzed by tandem MS. The LC MS-MS data were searched against the T. spiralis protein database from NCBI using Mascot version 2.4.1 (Matrix Science, UK).
The signal peptide, transmembrane domains, and subcellular localization of the identified proteins were predicted by using online tools SignalP , TMHMM , and CELLO v.2.5 , respectively.