All of the methods were previously approved by the Seoul National University Institutional Animal Care and Use Committee (Approval No. SNU-220621-3).

The AA evaluated in this study were blended to contain L-methionine (L-MET eco, 95% purity, CJ Cheiljedang Co., Ltd.), L-threonine (THR Pro, 80% purity, CJ Cheiljedang Co., Ltd.), and L-tryptophan (TRP Pro, 60% purity, CJ Cheiljedang Co., Ltd.). Tryptophan and threonine were granule-type feed-grade AA that were produced in Corynebacterium glutamicum. The AA blend was created at a weight ratio of 55 (Met):43 (Thr):2 (Trp).

Intestinal porcine epithelial cells (IPEC-J2, DSMZ No. ACC701, BWE, Germany) were seeded in T75 cell culture flasks (70,075, SPL Lifesciences, Gyeonggi-do, Korea) and cultured with Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F-12, Thermo Fisher Scientific, Waltham, United States) containing 10% fetal bovine serum (FBS, SV30207.02, HyClone, Cytiva, Australia), 1% insulin–transferrin–selenium–ethanolamine (ITS-X, 51500-056, Thermo Fisher Scientific, Waltham, United States), and 1% penicillin–streptomycin (15140-122, Thermo Fisher Scientific, Waltham, United States) at 37°C in a humidified atmosphere with 5% CO₂. The cells were sub-cultured once every 3 to 4 days (twice a week) until the confluence reached 90%. For sub-culture, the cells were detached from the flask by trypsin/EDTA (GIBCO, NY, United States), centrifuged at 300 × g for 3 min, re-suspended, and re-seeded at a concentration of 5.0 × 10⁵ cells in 25 mL of complete cell culture medium per T75 flask (17).

IPEC-J2s were seeded at 1.0 × 10⁵ cells/well (1 mL per well) in 24-well plates and cultured overnight. Cells were pretreated with or without AA blend for 24 h and challenged with 1 μg/mL of deoxynivalenol (DON, D0156, Sigma Aldrich, St. Louis, United States) and 1 μg/mL of lipopolysaccharide (LPS, L2630, Sigma Aldrich) for 3 h.

After these treatments, the cells were washed with cold PBS and harvested for total RNA extraction and the quantification of IL-8 mRNA. The percent reduction of IL-8 was calculated following the equation below:

where RGE is relative gene expression.

Total RNA was extracted from IPEC-J2 using the easy-spin RNA extraction kit (17,221, iNtRON, Korea). The RNA purity and concentration were measured using spectrophotometry (QIAxpert System, Qiagen, Hilden, Germany).

A quantitative reverse transcription PCR (RT-qPCR) was performed on the Rotor-Gene Q 2plex platform (Qiagen) in a final volume of 20 μL using the AccuPower GreenStar RT-qPCR PreMix (K-6403, Bioneer, Korea). The PCR reaction mixture contained 10 μL of 2X master mix, 2 μL of each primer (10 pmol/μL), 2 μL of template RNA, and 4 μL of DEPC-DW. The mixture was added into a PCR tube, and the PCR consisted of cDNA synthesis at 50° C for 15 min and pre-denaturation at 95° C for 5 min, followed by 40 cycles at 95\u00B0C for 15 s, 55° C for 30 s, and 72° C for 30 s. The primers were designed with Primer-Blast¹ based on the published cDNA sequence in the Gene Bank. The relative abundance of targeted genes was calculated according to the 2^−ΔΔCt method and normalized to the mean expression of GAPDH, which serves as the internal reference gene. Information on the detected genes and primers is shown in Table 1.

IPEC-J2s were seeded at 1.5 × 10⁴ cells/well (100 μL per well) in a 96-well plate. After overnight incubation at 37°C, cells were pretreated with or without the AA blend for 18 h and then washed with Hank’s Balanced Salt Solution (HBSS, 14175095, Thermo Fisher Scientific, Waltham, United States). To determine the amount of ROS, cells were incubated with 10 μM DCF-DA (2′, 7′-dichlorofluorescin-diacetate, D6883, Sigma Aldrich, St. Louis, United States) for 30 min and then treated with 0.5 mM H₂O₂ for 30 min. The fluorescence was read at 485 nm for excitation and 530 nm for emission with a fluorescence microplate reader (BioTek Synergy H1). The percent reduction of ROS was calculated following the equation below:

where RFU is the relative fluorescence unit.

The strain of S. enterica used in this study belongs to serovar Typhimurium, derived from the pig-virulent strain KVCC-BA1300432. This strain was kindly supplied by the Korean Veterinary Culture Collection (Gimcheon-si, Gyeongsangbuk-do, Republic of Korea).

Twenty-four 21-day-old pigs were purchased from a Salmonella-free herd. A commercial ELISA test (Swine Salmonella Ab Test, IDEXX Laboratories Inc., Westbrook, ME, United States) was used to test pig serology for Salmonella upon arrival at the Seoul National University facility. Twenty-four, 21-day-old pigs were randomly distributed into three groups (8 pigs per group). The diet was formulated according to the nutritional recommendations (18, 19).

At −14 days post-challenge (dpc, 21 days of age), pigs in the AA+ST group were started on a controlled standard diet feed supplemented with 0.3% of an AA blend. At 0 days post-challenge (dpc, 35 days of age), pigs in the AA+ST and ST groups were orally inoculated twice within 4 h using 1 mL of a growth medium containing 3.3 × 10⁹ CFU/mL of S. enterica serovar Typhimurium. Pigs in the control group were orally inoculated twice within 4 h with 1 mL of sterile saline solution. At 14 dpc (49 days of age), pigs were sedated by an intravenous injection of sodium pentobarbital and then euthanized by electrocution as previously described (20). Tissues were collected from each pig at necropsy.

Blood samples were collected from all pigs at −14, 0, 3, 7, and 14 dpc.

Rectal temperatures were recorded at 0, 1, 2, 3, 4, 5, 6, 7, and 14 dpc at the same time by the same personnel. A fecal scoring system was defined according to the following scale: 0 (normal), 1 (semisolid feces without blood), 2 (watery feces without blood), and 3 (blood-tinged feces) (4).

Pig weight was measured at −14 (21 days of age), 0 (35 days of age), and 14 (49 days of age) dpc throughout the study. An average daily weight gain (ADWG = grams/pig/day) was calculated over three time points: (i) between −14 and 0 dpc, (ii) between 0 and 14 dpc, and (iii) between −14 and 14 dpc at the study conclusion. The difference between the initial final weights was divided at each of these three time points by the number of days in the corresponding period to calculate ADWG. All data were obtained in a blinded manner.

Serum samples were collected for the quantification of tumor necrosis factor-α (TNF-α), IL-6, IL-8, and IL-10, as assayed using a commercial ELISA kit (Porcine TNF-alpha Quantikine ELISA Kit, Porcine IL-6 Quantikine ELISA Kit, Porcine IL-8/CXCL8 Quantikine ELISA Kit, and Porcine IL-10 Quantikine ELISA Kit, R&D Systems, Inc., Minneapolis, MN, United States). The results were expressed as pg./mL.

Serum superoxide dismutase (SOD) activity was measured using commercial kits (OxiSelect™ Superoxide Dismutase Activity Assay Kit, Cell Biolabs, Inc., San Diego, CA, United States).

A commercial protein carbonyl fluorometric assay was used for the measurement of protein carbonyl content in serum (Cell Biolabs, Inc., San Diego, CA, United States). A commercial thiobarbituric acid reactive substances assay kit was used for the direct quantitative measurement of malondialdehyde in serum samples.

A scoring system for microscopic cecal lesions was defined according to the following scale: 0 (normal), 1 (mild neutrophilic infiltrate without submucosal infiltrates), 2 (moderate neutrophilic infiltrate with or without submucosal infiltrate), and 3 (marked neutrophilic infiltrate with or without submucosal infiltrate) (21).

For the in vitro experiment, differences among the experimental data were assessed using a one-way ANOVA, followed by Tukey’s post-hoc test and F-protected test. For the in vivo experiment, a normal distribution was determined with the Shapiro–Wilk on these data. Whether or not the groups had statistically significant differences between them at various time points was then determined by performing a one-way ANOVA. For further evaluation, a post-hoc test for a pairwise comparison with Tukey’s adjustment was conducted with a statistical significance result from the one-way ANOVA test. A Kruskal–Wallis test was additionally performed only in cases where the normality assumption was not met. The results that showed statistical significance from the Kruskal–Wallis test were further evaluated with the Mann–Whitney test to compare the differences among the groups. The results were reported in p-values, and p-values of <0.05 were considered significant.

AA blend treatment significantly reduced (p < 0.05) LPS/DON-induced IL-8 expression to 49.7% ± 2.3% (2,000 ppm of AA blend), 19.1% ± 0.3% (500 ppm of AA blend), and 22.5% ± 12.1% (125 ppm of AA blend) compared to that of control (Figure 1A).

AA blend treatment significantly reduced (p < 0.05) H₂O₂-induced ROS to 39.7% ± 12.1% (2,000 ppm of AA blend), 29.6% ± 12.1% (1,000 ppm of AA blend), and 17.7% ± 6.3% (500 ppm of AA blend) compared to that of control (Figure 1B).

The mean rectal temperature was significantly higher (p < 0.05) in pigs (40.98 ± 0.66) from the ST group at 1 dpc compared to those of the control group pigs (40.20 ± 0.28) (Figure 2A). Pigs in the AA+ST and control groups had significantly (p < 0.05) lower fecal scores than those in the ST group at 1 to 7 dpc. Pigs in the control group had significantly (p < 0.05) lower fecal scores than those in the ST group at 14 dpc (Figure 2B).

The body weight and ADWG of the pigs were measured at −14, 0, and 14 dpc, where significant differences among the three groups were not found.

Serum TNF-α levels were significantly (p < 0.05) lower in the AA+ST and control groups at 7 and 14 dpc than those from the ST groups (Figure 3A). Serum IL-6 levels were significantly (p < 0.05) higher in the ST group at 3, 7, and 14 dpc than those from the control group. Serum IL-6 levels were significantly (p < 0.05) lower in the AA+ST and control groups at 14 dpc than those from the ST group (Figure 3B). Serum IL-8 levels were significantly (p < 0.05) lower in the AA+ST and control groups at 3, 7, and 14 dpc than those from the ST group (Figure 3C). Serum IL-10 levels were significantly (p < 0.05) lower in the AA+ST and control groups at 3 dpc than those from the ST group (Figure 3D).

Superoxide dismutase activity.

Serum SOD activities were significantly (p < 0.05) higher in the ST group at 3 and 7 dpc than those from the control groups and tend to increase in the AA+ST and control groups at 3 and 7 dpc than those from the ST group (Figure 4A).

Serum protein carbonyl (Figure 4B) and malondialdehyde (Figure 4C) levels were significantly (p < 0.05) higher in the ST group at 3 and 7 dpc than those from the control groups and tend to increase in the AA+ST and control groups at 3 and 7 dpc than those from the ST group.

Pigs in the AA+ST (lesion score 1.28 ± 0.44, Figure 5A) and control lesion score 0 ± 0, Figure 5B) groups had significantly (p < 0.05) lower Salmonella lesion scores than that in the ST ((lesion score 2.05 ± 0.67, Figure 5C) group at 14 dpc. Pigs in the AA+ST group had significantly higher Salmonella lesion scores than those in the control group at 14 dpc.