Kunming mice (male, 22–25 g) were purchased from China Three Gorges University, and the ethical permit approval of the Medical Animal Care and Welfare Committee of China Three Gorges University was obtained before using the mice for study. Every five mice were kept in an individual cage with a constant dark-light cycle in a conventional SPF animal house and were given free access to fundamental food as normal diet and water.

Yersinia pseudotuberculosis YpIII strains with different reporter systems on the virulence plasmid were constructed as described previously (Wang et al., 2016) and listed in Table 1. All experiments were carried out in the BSL2 laboratory.

Luria Bertani (LB) medium (EMD Millipore), 2-phenyl-1, 2-benzisoselenazol-3(2H)-one (EbSe) (Daiichi), protease inhibitor cocktails (Roche), anti-Trx1 polyclonal antiserum (IMCO), rabbit anti-sheep IgG-HRP (Santa Cruz), goat anti-mouse H&L antibodies (Santa Cruz), IgG2a mouse monoclonal antibody (VIROGEN), CellROX™ Deep Red Reagent (Invitrogen), Anti-IL-1β antibody (Proteintech), anti-TNF-α antibody (Proteintech), anti-IL-10 antibody (Proteintech). Protein inhibitor cocktail (MedChemExpress). Silver nitrate and all the other reagents were from Sigma-Aldrich.

The inhibitory effect of EbSe-Ag⁺ on the growth of YpIII was measured by UV-Vis at A₆₀₀. YpIII cells were grown (26°C, 200 rpm) 8 h and diluted 1,000 times and treated with a serial concentration of EbSe-Ag⁺ for 24 h at 26°C in a 96-well plate, and the A₆₀₀ were measured.

The bactericidal activity of EbSe-Ag⁺ against YpIII was detected by laser scanning confocal microscopy (LSCM, A1R⁺, Nikon). YpIII cells were grown (26°C, 200 rpm) until an A₆₀₀ of 0.4, and diluted 1,000 times, which were treated with 4 μM EbSe and 0.5 μM Ag⁺ for 24 h at 26°C. The bactericidal effect was observed by LSCM.

YpIII was grown until an A₆₀₀ of 0.4 and treated for 30 min with 80 μM EbSe and 5 μM Ag⁺. Cells were obtained by centrifugation (4°C, 12,000 rpm, 15 min) and fixed with 2.5% glutaraldehyde. The morphology of YpIII cells was observed by transmission electron microscopy (TEM, Hitachi H-7500).

The post-antibiotic effect (PAE) of EbSe-Ag⁺ against YpIII was detected by UV-Vis. YpIII cells were grown (26°C, 200 rpm) until an A₆₀₀ of 0.4, and diluted 1,000 times. Further, cells were treated with 4 μM EbSe and 0.5 μM Ag⁺ for 32 h at 26°C, 200 rpm. Then, drugs were removed by centrifugation (4°C, 10,000 rpm, 2 min) for three times using PBS (pH 7.4). The A₆₀₀ of YpIII cells were further measured every 1 h for 6 h.

YpIII cells were grown (26°C, 200 rpm) until an A₆₀₀ of 0.4, and diluted 1,000 times, which were treated with 4 μM EbSe and 0.5 μM Ag⁺ after pre-incubation with 4 μM dithiothreitol (DTT) for 30 min at 26°C, 200 rpm. The A₆₀₀ of YpIII cells were measured.

YpIII cells were cultured until an A₆₀₀ of 0.4 and incubated with 80 μM EbSe and 5 μM Ag⁺ for 30 min. YpIII cells were obtained by centrifugation (4°C, 5,000 rpm, 5 min), and a protein inhibitor cocktail (in 50 mmol/L Tris-EDTA buffer, pH 7.4) was added to decrease the protease activity. The cells were disrupted with sonication (240 W, 10 min), and the supernatants were obtained by centrifugation (4°C, 12,000 rpm, 10 min). The TrxR activity and total glutathione amount were detected in a 96-well plate by 5, 5’-dithiobis-(2-nitrobenzoic acid) assay. The activity of the untreated group was 100%. The protocols were performed as previously described (Lu et al., 2013; Zou et al., 2017; Wang P. et al., 2020).

YpIII cells were cultured until an A₆₀₀ of 0.4, which were incubated with 80 μM EbSe and 5 μM Ag⁺ for 30 min. After lysis by sonication, the cell lysates were obtained by centrifugation at 12,000 rpm for 20 min. Western blotting assay was performed with anti-Trx1 polyclonal antibody.

YpIII cells were cultured until an A₆₀₀ of 0.4, which were incubated with 80 μM EbSe and 5 μM Ag⁺ for 30 min. After lysis by sonication, the cell lysates were obtained by centrifugation at 12,000 rpm for 20 min. Total protein S-glutathionylation (S-PSSG) of 80 μM EbSe and 5 μM Ag⁺-treated YpIII cells were detected by Western blotting. Cells were cultured, washed, and resuspended in buffer containing 30 mmol/L Iodoacetamide (IAM). Western blotting assay was performed with IgG2a mouse monoclonal antibody for glutathione-protein complexes.

YpIII were grown until an A₆₀₀ of 0.4 in LB medium and incubated with 80 μM EbSe and 5 μM Ag⁺ for 30 min. The YpIII cells were stained with 5 μM CellROX™ Deep Red Reagent for 30 min at 37°C. After incubation, the reactive oxygen species (ROS) production was quantified by flow cytometry (BECKMAN COULTER, AW15093).

All experiments were performed in accordance with the relevant guidelines and regulations approved by China Three Gorges University. 60 healthy male Kunming mice were randomly divided into 5 groups randomly (n = 12). The mice were infected via intragastric gavage with 100 μL 2 × 10⁹CFU of YpIII as previously described (Wang et al., 2016) and were further administered i.p. with 20 mg/kg EbSe, 5 mg/kg Ag⁺, 20 mg/kg EbSe and 5 mg/kg Ag⁺ and DMSO on the 1-, 3-, and 5-days post-infection, separately.

Mice were deprived of food and water for 16 h prior to intragastric administration with YpIII-bioluminescent and were monitored for bioluminescent emission using IVIS lumina II (Caliper LifeSciences) at 1-, 4-, and 7-days post-infection. Mice were anesthetized using the XGI-8 gas anesthesia system (Caliper LifeSciences) prior to imaging with 2.5% IsoFluVet in oxygen (Orion Pharma Abbott Laboratories Ltd., Great Britain), and during imaging in 0.5% IsoFluVet. Images were acquired and analyzed using Living Image 4.5 (Caliper LifeSciences). To analyze bacterial localization within gastrointestinal tracts, mice were euthanized, the gastrointestinal tracts were removed, and imaged by bioluminescent imaging.

Mice intestinal tracts were rolled, fixed in formalin (10%), and embedded in paraffin. Paraffin sections (4 μm thick) were used to detect cytokine levels by H&E stain and immunohistochemical (IHC) analysis. Anti-IL-1β, anti-TNF-α and anti-IL-10 antibodies were used for the analysis of cytokines in intestinal tracts.

Mice blood was collected 7 days post-infection, and the white blood cells (WBCs) and neutrophils counts were detected. For investigation of liver and kidney function, blood was collected and centrifuged at 3,000 rpm for 10 min. Serum alanine transaminase (ALT), aspartate aminotransferase (AST), urea and creatinine were determined by fully automatic biochemical analyzer (Siemens, viva proE).

Statistical analyses were performed by GraphPad Prism 6.0 (GraphPad Software). Means of data between two groups were contrasted using unpaired Student’s t test. Sample rates between two groups were tested with chi-square analysis. p-values of < 0.05 were significant.

The antibacterial effects of EbSe-Ag⁺ on the growth of YpIII-GFP and YpIII-mCherry were detected by UV-Vis. As shown in Figure 1, Ag⁺ alone inhibited YpIII growth with a 90% minimal inhibition concentration (MIC₉₀) of 8 μM, while the addition of 2 μM EbSe effectively reduced the MIC₉₀ of Ag⁺ to 0.5 μM (16 times). Meanwhile, 2 μM EbSe and 0.5 μM Ag⁺ showed no synergistic toxicity on human cells as described previously (Zou et al., 2017). In addition, the formula S = (F_X0/F₀₀) * (F_Y0/F₀₀) – (F_XY/F₀₀) was used to calculate the synergistic effect of EbSe-Ag⁺. The results showed that the synergy indexes were 0.97 and 0.94, respectively, indicating EbSe-Ag⁺ has a strong synergistic antibacterial effect against YpIII (Supplementary Figure 1).

Whether the antibacterial effect of EbSe-Ag⁺ is bacteriostatic or bactericidal was further investigated by LSCM and bacterial plating. 4 μM EbSe and 1 μM Ag⁺ were inoculated with YpIII-GFP and observed by LSCM (Figure 2). Meanwhile, the minimum bactericidal concentration (MBC) was calculated by plating (Supplementary Figure 2). Both results showed that the combination of 4 μM EbSe and 1 μM Ag⁺ has significant bactericidal effects against YpIII-GFP.

Further, the effect of EbSe-Ag⁺ on the morphology of YpIII was assessed by TEM. Normal YpIII-GFP has a smooth surface and a complete cell membrane and cell wall. After 30 min treatment with 80 μM EbSe and 5 μM Ag⁺, YpIII-GFP cells were deformed, cell membranes and cell walls were ruptured, and overflow of cell contents; meanwhile, 80 μM EbSe or 5 μM Ag⁺ treated cells showed no obvious morphological changes (Figure 3). The results showed that the morphology of YpIII-GFP changed significantly in EbSe-Ag⁺ treated group compared with the control group.

YpIII was cultured and co-cultivated with 8 μM EbSe and 0.5 μM Ag⁺ for 32 h. During this period, the A₆₀₀ was measured every 4 h. 32 h post-treatment, the bacterial solution was centrifuged and washed in PBS by 3 times, and resuspended in LB. The A₆₀₀ was further detected every 1 h for 6 h. The results show that the post-antibiotic effect (PAE) of EbSe-Ag⁺ was 4 h, which is longer than many clinical antibiotics, including quinolone (Figure 4).

The protective effect of DTT against EbSe-Ag⁺-treated YpIII was evaluated by UV-Vis and TEM assays. DTT was an organic reducing agent that can remove ROS, and 4 mmol/L DTT can effectively rescue YpIII from the bactericidal effect of 4 μM EbSe and 1 μM Ag⁺ (Figures 5A–F). Meanwhile, the TEM results showed that DTT could protect YpIII from EbSe-Ag⁺ caused morphological damage (Figure 5G). Both results indicated that the death of YpIII caused by EbSe-Ag⁺ was related to ROS production.

The mean fluorescent intensity (MFI) of ROS in YpIII-mCherry cells was detected by flow cytometry (Ezraty et al., 2017; Zou et al., 2017). The result showed that the ROS production level in YpIII cells treated with 80 μM EbSe and 5 μM Ag⁺ was significantly regulated when compared with the control (Figure 6).

The effects of EbSe-Ag⁺ on bacterial TrxR activity and GSH amount in YpIII were measured by DTNB assay. The results showed that the combination of 80 μM EbSe and 5 μM Ag⁺ could efficiently inhibit the TrxR activity and deplete GSH amount when compared with the control (Figures 7A–D).

Whether EbSe-Ag⁺ could affect protein Trx1 and S-PSSG expression levels were also analyzed by Western blot (Figures 8A–H). The result showed that Trx1 protein expression level had no difference, indicating the treatment of EbSe-Ag⁺ only influences enzyme activity. Meanwhile, protein S-PSSG was reduced by EbSe-Ag⁺ treatment compared with Ag⁺ or EbSe, which further reflected the loss of GSH.

Taken together, the above results showed that EbSe-Ag⁺ inhibited TrxR activity and depleted GSH in YpIII. All the results suggested that EbSe-Ag⁺ could disrupt YpIII TDRS, and ROS production is one of the key mechanisms underlying its bactericidal activity.

Gastroenteritis is among the most common Y. pseudotuberculosis-caused infections (Amphlett, 2016). To evaluate whether EbSe-Ag⁺ could protect mice from YpIII-caused gastroenteritis, 60 mice were randomly divided into five groups, and 2 × 10⁹ CFU/100 μL bioluminescent-YpIII strain Xen4 (Wang et al., 2016) were orally administrated to construct gastroenteritis models. Mice were further injected i.p. with 20 mg/kg EbSe, 5 mg/kg Ag⁺, 20 mg/kg EbSe and 5 mg/kg Ag⁺ and DMSO on days 1-, 3-, 5-, post-infection. Mice were monitored for bioluminescent emission at 1-, 4-, and 7-days post-infection using IVIS. To further analyze bacterial localization within gastrointestinal tracts, mice were euthanized, the gastrointestinal tracts were removed, and imaged by bioluminescent imaging. The results showed that EbSe-Ag⁺ led to a significant reduction in bacterial load on-site compared with the control group (Figure 9). These findings demonstrated that EbSe-Ag⁺ could effectively protect mice from YpIII-caused gastroenteritis.

Hematoxylin and eosin (H&E) staining was performed using the new formative tissues. Staining revealed that control mice had increased number of inflammatory cells. Meanwhile, EbSe-Ag⁺-treated mice had less scattered lymphocytes (Figure 10).

An IHC assay was used to measure the presence of pro-inflammatory cytokines IL-1β and TNF-α and anti-inflammatory cytokine IL-10. As shown in Figure 10, EbSe-Ag⁺ could significantly down-regulate IL-1β and TNF-α and up-regulate IL-10 when compared to control mice.

The peripheral blood from different groups of mice was collected, and the results showed that EbSe-Ag⁺ statistically reduced the counts of WBCs (Figure 11A) and neutrophils (Figure 11B) compared with the control. Furthermore, ALT, AST, urea, and creatinine in mice blood serum were also detected, and the results showed that EbSe-Ag⁺ treatment reduced the ALT (Figure 11C), AST (Figure 11D) levels compared with Ag⁺ treatment.

Overall speaking, EbSe-Ag⁺ treatment exhibited significant effects in reducing the YpIII-caused inflammation with low toxicity.