All methodologies involving experimental animals received approval from the Department of Laboratory Animal Resources Committee of Yonsei University College of Medicine (no. 2020-0055).

A single Chub mackerel (Scomber japonicus) was acquired from the Saruga shopping center, Seoul, Korea; Anisakis third-stage larvae were manually collected from the abdominal cavity, yielding a total of 150 larvae, which were washed before DNA and protein extraction processes. DNA from 15 Anisakis larvae was extracted using the NucleoSpin DNA Insect Kit (Macherey-Nagel, Düren, Germany). The presence of A. pegreffii was confirmed through polymerase chain reaction (PCR) using primers targeting the ITS region: Primer A (forward: 5′-GTCGAATTCGTAGGTGAACCTGCGGAAGGATCA-3′) and Primer B (reverse: 5′-GCCGGATCCGAATCCTGGTTAGTTTCTTTTCCT-3′) [11]. PCR amplifications were sequenced and aligned with A. pegreffii reference sequences obtained from GenBank, confirming that the 15 tested parasites were A. pegreffii. Protein extraction from the remaining 135 larvae was performed using sonication, followed by centrifugation at 10,000×g for 30 min and filtration through a sterilized 0.22-μm filter (Merck Millipore, Darmstadt, Germany). The protein concentration was determined using the Bradford assay (Bio-Rad, Hercules, CA, USA). The extract was maintained on ice throughout the extraction process and stored at −80°C.

Female BALB/c mice (n=15; 6 weeks old) were purchased from Orient Bio (Seongnam, Korea) and tested in 3 groups (5 mice per group): phosphate-buffered saline (PBS) group, Anisakis-immunized without ampicillin treatment group, and Anisakis-immunized with ampicillin group; the number of mice per group was consistent with that used in our previous Anisakis allergy research [12]. To ensure optimum conditions, mice were housed in a specific pathogen-free environment with a 12-h light/dark cycle and given a week to acclimatize before experimentation. A daily health assessment was systematically conducted per mouse.

For immunization, 100 μg of A. pegreffii protein extract was suspended in PBS while ampicillin was prepared at a concentration of 1 g/L in distilled water [13]. This concentration was then provided to specific groups as part of their daily water consumption. The ampicillin regimen was commenced a week before the primary sensitization and maintained throughout the experiment. Mice were intraperitoneally injected with the A. pegreffii extract (100 μg) once weekly for a total of 4 weeks [14]. Mice were euthanized a fortnight after the final dosage. This study was repeated 3 times.

Mice were euthanized using CO₂ gas, and blood was collected from the inferior vena cava. For serum separation, blood samples were centrifuged for 30 min at 3,000×g, and the supernatant was collected. The serum levels of A. pegreffii-specific IgG1 and IgG2a were measured using an enzyme-linked immunosorbent assay (ELISA). Briefly, 1×8 Stripwell 96-well plates (Corning, Corning, NY, USA) were coated with 100 μl/well coating buffer containing 0.3 μg A. pegreffii protein extract and incubated overnight at 4°C [12]. The plate was then washed with PBS+0.05% of Tween-20 (PBST) and blocked with 200 μl/well diluent buffer (PBST containing 1% of bovine serum albumin (BSA)) for 1 h at 25°C. Samples were diluted 1:100 with diluent buffer. After blocking, 100 μl of samples were dispensed per well and incubated for 2 h. Biotinylated goat anti-mouse IgG1 or biotin goat anti-mouse IgG2a (both diluted 1:1,000; NBP1-69914B and NBP1-69915B, respectively; Novus Biologicals, Littleton, CO, USA) were used as secondary antibodies. Wells were then incubated with an avidin-horseradish peroxidase (HRP) conjugate (BioLegend, San Diego, CA, USA) for 30 min, followed by incubation with 3,3,5,5-tetramethylbenzidine (Sigma-Aldrich) substrate (50 μl) in the dark for 5 min. The reaction was inhibited by adding 0.5M H₂SO₄. Absorbance was measured at 450 nm using VersaMax (Molecular Devices, Seoul, Korea). Statistical analyses were performed using GraphPad Prism software (version 8.0, GraphPad Software, San Diego, CA, USA).

Western blotting was performed using A. pegreffii protein. Proteins were separated by gel electrophoresis. A. pegreffii protein (10 μg) was loaded into each well of a mini format 10% sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel. Following gel electrophoresis, proteins were transferred from the gel onto a polyvinylidene difluoride membrane. After transfer, the membrane was blocked with 5 ml of Tris-buffered saline Tween-20 (TBST) containing 5% of BSA for 2 h at 25°C. Pooled samples were diluted 1:1,000 with TBST containing 1% of BSA and incubated overnight at 4°C. The goat anti-mouse IgG1 and goat anti-mouse IgG2a heavy chain antibodies (both diluted 1:1,000; ab97240 and ab97245, respectively; Abcam, Cambridge, UK) were used as secondary antibodies conjugated with HRP (1:10,000; ab2116, Abcam) for detection. Bands were developed using the ECL reaction (Bio-Rad), and images were captured using Chemidoc XRS (Bio-Rad).

Protein concentrations were determined using the Bradford method (Bio-Rad). For 2-D analysis, pH 3–10 IPG strips (GE Healthcare Life Sciences, Pittsburgh, PA, USA) were rehydrated in swelling buffer containing 7 M urea, 2 M thiourea, 2.5% (w/v) dithiothreitol, and 4% (w/v) CHAPS. Protein lysates (600 μg) were loaded into the rehydrated IPG strips using an IPGphor III (GE Healthcare Life Sciences), and the 2-D separation was performed on a 10% SDS-PAGE gel. Following fixation of the gels for 1 h in a solution of 40% (v/v) methanol containing 5% (v/v) phosphoric acid, gels were stained with colloidal Coomassie blue G-250 solution (ProteomeTech, Seoul, Korea). Gels were destained using deionized water, and images were acquired via an image scanner (Bio-Rad). Image analysis was performed using ImageMaster 2-D Platinum software (Amersham Biosciences, Hercules, CA, USA). To compare protein spots, >25 spots in all gels were marked and normalized. Data for the normalized spot intensities are shown in Table 1.

Spots from 2-D gels were excised and in-gel digested with trypsin [15]. Briefly, protein spots were excised from the stained gels and cut into pieces, which were washed for 1 h at 25°C in 25 mM ammonium bicarbonate buffer, pH 7.8, containing 50% (v/v) acetonitrile (ACN). Following dehydration of gel pieces in a centrifugal vacuum concentrator (Biotron, Inc., Incheon, Korea) for 10 min, they were rehydrated in 50 ng of sequencing-grade trypsin solution (Promega, Madison, WI, USA). After incubation in 25 mM ammonium bicarbonate buffer, pH 7.8, at 37°C overnight, tryptic peptides were extracted in 100 μl of 1% formic acid (FA) containing 50% (v/v) ACN for 20 min with mild sonication. The extracted solution was then concentrated using a centrifugal vacuum concentrator. Before mass spectrometry (MS) analysis, the peptide solution was desalted using a reversed-phase column [16]. Briefly, after an equilibration step with 10 μl of 5% (v/v) FA, the peptide solution was loaded onto the column and washed with 10 μl of 5% (v/v) FA. Bound peptides were eluted in 8 μl of 70% ACN in 5% (v/v) FA.

Liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis was performed using a nano ACQUITY UPLC and LTQ-orbitrap-mass spectrometer (Thermo Electron, San Jose, CA, USA) with a BEH C18 1.7 μm, 100 μm×100 mm column (Waters, Milford, MA, USA). The mobile phase A for the LC separation was 0.1% FA in deionized water, and the mobile phase B was 0.1% FA in ACN. The chromatography gradient provided a linear increase from 10% B to 40% B for 16 min, from 40% B to 95% B for 8 min, and from 90% B to 10% B for 11 min. The flow rate was 0.5 μl/min. For tandem MS, mass spectra were acquired using data-dependent acquisition with a full mass scan (300–2,000 m/z), followed by MS/MS scans. Each MS/MS scan acquired was the average of 1 microscan of the LTQ. The temperature of the ion transfer tube was controlled at 275°C, and the spray voltage was 2.3 kV. The normalized collision energy was set to 35% for MS/MS. Individual spectra from MS/MS were processed using the SEQUEST software (Thermo Quest, San Jose, CA, USA), and the generated peak lists were used to query the in-house database using the MASCOT program (Matrix Science, London, UK). The modifications of carbamidomethyl, deamidated, and oxidation were set for MS analysis, and the tolerance of the peptide mass was 10 ppm. The MS/MS ion mass tolerance was 0.8 Da, the allowance of missed cleavage was 2, and charge states +2 and +3 were considered for data analysis. We considered only significant hits as defined by the MASCOT probability analysis [17].

Mouse ceca were collected and stored at −80°C. Cecal DNA was extracted using the FastDNA SPIN Kit for Soil (MP Biomedicals, Carlsbad, CA, USA). DNA corresponding to the V3–V4 region of the 16S rRNA gene was PCR amplified using the following bacterial universal primer pair: forward primer, 5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3′, and reverse primer, 5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3′ [18]. Limited-cycle amplification was performed using the Nextera XT Index Kit (Illumina, San Diego, CA, USA) to add multiplexing indices and Illumina sequencing adapters to PCR products. Libraries were normalized, pooled, and sequenced using the MiSeq platform (600 cycles, Illumina MiSeq V3 cartridge; Illumina).

Bioinformatics analyses were performed as described by Kim et al. [18] Raw reads were processed through a quality check, and low-quality (Q<25) reads were filtered using Trimmomatic 0.32 [19]. Paired-end sequence data were merged using PandaSeq [20]. Primers were trimmed using the ChunLab in-house program (ChunLab, Inc., Seoul, Korea), applying a similarity cutoff threshold of 0.8. Sequences were denoised using the Mothur preclustering program, which merges and extracts unique sequences, allowing up to 2 differences between them [21]. The EzBioCloud database [22] was used for taxonomic assignment using BLAST 2.2.22 [23], and pairwise alignments were generated to calculate the similarity [24]. The UCHIME algorithm and nonchimeric 16S rRNA database from EzBioCloud were used to detect chimeric sequences for reads with the best hit similarity rate of <97% [25]. Sequence data were then clustered using CD-Hit and UCLUST [26,27]. All aforementioned analyses were performed using EzBioCloud, a commercially available ChunLab bioinformatics cloud platform for microbiome research (https://www.ezbiocloud.net/). Reads were normalized by random subsampling to 7,666 to perform the analyses. The Shannon index [28] and principal coordinate analysis (PCoA) [29] were computed based on the generalized UniFrac distance [30]. The Kruskal–Wallis test was used to compare differences between the number of operational taxonomic units (OTUs), and the Shannon index was used to compare microbiome diversity among the 3 groups. Linear discriminant analysis effect size (LEfSe) analysis was used to identify significantly different taxa [31].

In the cecal microbiome study, the number of OTUs and Shannon index significantly decreased in the ampicillin-treated and Anisakis-immunized mice (Fig. 1A, B), indicating that the cecal microbiome diversity decreased after ampicillin treatment. In PCoA, the ampicillin-treated samples were distinctly clustered from untreated counterparts along the PCoA1 axis (Fig. 1C). Composition analysis suggested that the ampicillin-treated group exhibited the pronounced presence of a bacterial species resembling Proteus while the relative abundance of Bacteroides was markedly different across the groups (Fig. 2). We then performed LEfSe analysis to identify significant differences in bacterial abundance among the control and ampicillin-treated and untreated Anisakis-immunized groups. The LDA score for Proteus vulgaritus was 5.54, and the average relative abundance was 63.12%, 0.04%, and 0.02%, respectively in the ampicillin-treated Anisakis-immunized, control, and ampicillin-untreated Anisakis-immunized groups. The LDA score for Bacteroides_uc was 5.16, and its average relative abundance was 21.92%, 28.28%, and 0.08% in the control, ampicillin-untreated Anisakis-immunized, and ampicillin-treated Anisakis-immunized groups, respectively (Table 2).

ELISA analysis showed that the Anisakis-specific IgG1 and IgG2a levels increased in the Anisakis-immunized mice with or without ampicillin treatment, although the difference in the IgG1 and IgG2a levels between Anisakis-immunized ampicillin-treated and untreated groups was not significantly different (Fig. 3A, B).

To determine which Anisakis antigens were immunized in mice, immunoblotting was performed using pooled sera. In a 1-dimensional (1-D) immunoblot using Anisakis-specific anti-mouse IgG1, a 56-kDa band was detected only in the ampicillin treatment group and identified as an “unnamed protein product (VDK17787.1 and VDK17834.1)” in MS analysis (Fig. 4A; Table 3). In an immunoblot with Anisakis-specific anti-mouse IgG2a, a band at 70 kDa disappeared in the ampicillin treatment group and was identified as an unnamed protein product (VDK17973.1) and heat shock protein 70 (HSP70) (AIU38247.1) (Fig. 4B; Table 3).

For improved resolution, we performed a 2-D immunoblot against Anisakis extract with immunized mouse sera. Using Anisakis-specific anti-mouse IgG1, spots 13 at 55 kDa and 19 at 46 kDa appeared only in the ampicillin-untreated Anisakis-immunized group while spots 4 at 70 kDa and 7 at 70 kDa and 14 kDa appeared only in the ampicillin-treated Anisakis-immunized group (Fig. 5A, B). Using Anisakis-specific anti-mouse IgG2a, spots 1, 16, 17, and 27 at 100, 48, 48, and 27 kDa, respectively, that were present in the ampicillin-untreated group disappeared in the ampicillin-treated group (Fig. 5C, D). Furthermore, spots 5 and 6 at 70 kDa were reduced in the ampicillin-treated group. MS was performed to identify these proteins of selected spots in the 2-D immunoblot assay (Table 1); spot 13 was identified at HSP70 (AIU38247.1) while spots 11 and 12 were identified as an unnamed protein product (VDK17787.1); these spots were detected in the 1-D western blotting MS analysis (Table 3). These observations suggest that ampicillin treatment altered the immunized patterns of Th1- and Th2-inducing Anisakis antigens.