The antibodies used in the experiments were purchased from Cell Signaling Technology (Maryland, USA), the triclosan (TCS) came from Sigma (Missouri, USA), and the other reagents, unless stated otherwise, were all purchased from Sigma (Missouri, USA). The pmRFP-LC3 is mammalian expression of rat LC3 fused to mRFP, the ptfLC3 is mammalian expression of rat LC3 fused to mRFP and EGFP, and the pEGFP-LC3 is mammalian expression of rat LC3 fused to EGFP (Kimura et al., 2007). Lactobacillus, Escherichia coli (E. coli), and Enterococcus faecalis (E. faecalis) were isolated from normal mouse feces. S. typhimurium ATCC 14028 and C. albicans SC5314 were kept by our laboratory.

The MICs were determined using the microdilution method as described previously (Moore et al., 2008). The MIC was defined as the lowest concentration for which bacterial growth did not occur. The MBC was defined as the lowest concentration of microbicide at which no growth occurred after 4 days of incubation.

The MICs of TCS against the C. albicans strains were determined by broth microdilution using two-fold serial dilutions in RPMI 1640 medium, as described by the CLSI method M27-A. The test was performed in 96-well, flat-bottomed microtitration plates according to previous studies (Sun et al., 2008).

Intestinal epithelial cells and intestinal macrophages were extracted according to previous studies (Macartney et al., 2000; Yu et al., 2011). Raw264.7 cells and HeLa cells used in our study came from ATCC (Maryland, USA) and were maintained in DMEM supplemented with 10% fetal bovine serum (FBS) under the conditions of 37°C with 5% CO₂. All cell culture reagents were purchased from Gibco Laboratories (NY, USA).

TCS was diluted with cell culture medium to concentrations from 0.25 to 32 μM. The cells were seeded into six-well flat-bottom plates (1 × 10⁶ cells/well) or 24-well glass-bottom plates (2 × 10⁵ cells/well) in serum-free and antibiotic-free DMEM medium.

For the transmission electron microscopy (TEM) observation, cells were incubated in six-well flat-bottom plates and were treated with TCS at a dose of 8 μM for 180 min. The cells were stained after immobilization. Autophagosome-like vesicles were observed under the TEM (Hitachi H-7650, Tokyo, Japan).

For the fluorescence microscope image collection, the cells needed to be transiently transfected with pEGFP-LC3 or ptfLC3 plasmids. The Raw264.7 cells were incubated in 24-well flat-bottom plates with glass slides. The instructions to complete cell transfection were followed. Then, we replaced the medium with new serum-free DMEM. The cells transfected with plasmids were used in other experiments.

Cells or transfected cells were seeded in 24-well glass-bottom plates. The cells were treated with TCS with or without S. typhimurium or C. albicans. The images were obtained by an Olympus FV1000 confocal laser-scanning microscope (Olympus, Tokyo, Japan) with a 60x objective lens. Image analyses and exports were performed using a Fluoview ver. 1.7.3.0 (Olympus, Tokyo, Japan).

The cells were treated with TCS as described above, and the total protein was collected by centrifugation and was quantified using the BCA reagent (Beyotim, P0012). The images were obtained by a CanoScan LiDE 100 scanner (Canon). Protein blots were measured using Image-J software.

For the assay of the phagocytosis ratio in RAW264.7 cells, the cells were incubated in 24-well flat-bottom plates and were treated with Rapa (200 nM, 12 h), TCS, or TCS and 3-MA (5 mM, 180 min). The cells were then exposed to S. typhimurium or C. albicans at the respective MOI of 25:1 and 1:1 for 120 and 30 min. The assay of S. typhimurium growth in RAW264.7 cells was performed as previously described (Chiu et al., 2009). The collected C. albicans were spread onto YPD agar plates for CFU counting.

Female BALB/c mice (20 ± 2 g) were kept in conditions of a constant photoperiod (12 h light, 12 h dark) and had free access to food and water. For autophagy observation, mice were injected with TCS from 3.125 to 40 mg/kg for 90 min through the enterocoelia. Then, the mice were sacrificed for liver and spleen harvesting. Western blotting was used for the relevant protein assays.

For the assays assessing the effect of the bacteria-killing abilities, mice were infected by an intraperitoneal injection of an overnight culture of S. typhimurium (1 × 10⁵ organisms in 0.1 ml PBS). After 24 h, the mice were treated with vehicle (0.5% methylcellulose-0.1% Tween 80 in sterile water), TCS, or TCS and 3-MA (24 mg/kg) by intraperitoneal injection once a day for 4 days. Then, the mice were euthanized, and the livers and spleens were collected for homogenization in 5 ml cold PBS. The homogenates were then serially diluted and plated onto LB agar plates. The CFU counting was conducted as described above.

To establish a mouse model that was infected with C. albicans, mice were infected with C. albicans (1 × 10⁷ organisms in 0.1 ml PBS) by oral administration. After 3 days, the mice were randomly assigned to control and treatment groups. C. albicans were treated the same way as those treated with S. typhimurium. The homogenates were then serially diluted and plated onto YPD agar plates. The CFU counting was conducted as previously described.

For the long-term sub-inhibitory concentration animal administration experiment, mice were randomly assigned to control, experiment, and treatment groups. The mice were treated with vehicle, TCS (1, 2, or 3 mg/day), or TCS (3 mg/day) by oral ingestion of 15 ml and 20 μl of 3-MA (24 mg/kg) by intraperitoneal injection for 30 days.

Representative bacteria in the feces of the mice treated with vehicle, TCS (1, 2, or 3 mg/day), or TCS (3 mg/day) and 20 μl of 3-MA (24 mg/kg), were counted with differential media on the 30th day. Intestinal bacteria of mice were cultivated with differential media, including MRS medium, MacConkey agar, and Enterococcus faecalis agar, which were purchased from Hopebio Co., Ltd. (Qingdao, China). The MRS plate was incubated anaerobically for 2–3 days at 37°C, while the MacConkey and E. faecalis plates were incubated aerobically overnight at 37°C.

16S rRNA gene high-throughput sequencing was used to determine the structure comparisons of the bacterial species in each of these samples using the Illumina HiSeq platform. PCR primers 515F and 806R targeting the bacterial 16S rRNA V4 region were selected for bacterial community analysis using the Illumina high-throughput sequencing method Sequences were analyzed using Quantitative Insights Into Microbial Ecology (QIIME) software and the UPARSE pipeline. The default settings for Illumina processing in QIIME were used and the UPARSE pipeline was then used to assign taxonomy at 97% similarity via the RDP classifier.

Mouse studies were performed in strict accordance with the Institutional Guiding Principles for Biomedical Research Involving Animals. The experiments were approved by the Jilin University Animal Care and Use Committee (no: IZ-2009-008). The gene copies indicated the absolute copy numbers present per unit of dry weight, whereas the normalized copy number based on 16S rRNA was regarded as the abundance. Sequences analysis were performed by Uparse software (Uparse v7.0.1001, http://drive5.com/uparse/). Sequences with ≥97% similarity were assigned to the same OTUs. The Silva Database (https://www.arb-silva.de/) was used based on RDP classifier (Version 2.2, http://sourceforge.net/projects/rdp-classifier/) algorithm to annotate taxonomic information.

All of the results are expressed as the means ± SD. The group means were compared using a one-way ANOVA, and Student's t-test was used to determine the significance of differences. For p values,^*p < 0.05, ^**p < 0.01, or ^***p < 0.001 compared with the control were considered statistically significant. The data are representative of triplicate experiments and are presented as the mean value ± the SD.

The MIC and MBC of TCS against the strain S. typhimurium ATCC14028 in this study were 0.5 μg/ml and 32 μM, respectively. The MIC and MBC of TCS against the strain C. albicans SC5314 tested in this study were 2 and 16 μg/ml, respectively, and this was consistent with a previous report that TCS against C. albicans is at 16 μg/ml (Higgins et al., 2012).

To explore whether TCS induces autophagy in cells, TCS with different concentrations from 0.5 to 32 μM was added to HeLa cells that were transfected with GFP-LC3 plasmid. The results demonstrated that the number of LC3 puncta was significantly increased in the TCS- or Rapa-treated cells, whereas in the control and 3-MA-added cells, few LC3 puncta were found (Figure S1A). RAW264.7 cells were administered TCS for different times or treated with different doses for 90 min for western blot assays. The results demonstrated that the ratio of LC3-II/LC3-I increased, which was followed by a decrease and reached a peak after 90 min of treatment (Figure 1A). The ratio of LC3-II/LC3-I increased with the TCS dose (Figure 1B). To seek in vivo evidence that TCS induces autophagy, mice were treated with different doses of TCS, and the livers and spleens were collected for western blot assays. The results showed that TCS at all of the concentrations tested (3.75–25 mg/kg) improved the ratio of LC3-II/LC3-I both in the livers (Figure S1B) and spleens (Figure S1C) in a dose-dependent manner.

To further verify that autophagy was induced by TCS, we observed the autophagosome formation in TCS-treated RAW264.7 cells by TEM. In the control groups (Figure S2A) or 3-MA-treated groups (Figure S2B), no autophagosome was found. In the TCS-treated cells, autophagosomes with a double membrane structure could be easily found, and the TEM captured the moment that the autophagosome was fusing with a lysosome (Figure S2C). The same structures were seen in the rapamycin-treated groups (Figure S2D).

To further reveal whether TCS induced an autophagic response, firstly, non-phagocytic cells (HeLa) were transfected with ptf-LC3 plasmids and were treated with TCS. The GFP-LC3 component is sensitive to lysosomal proteolysis, and therefore, the corresponding GFP fluorescence is quenched once it undergoes proteolysis, whereas the RFP-LC3 component will not be quenched with proteolysis. The results showed that compared with the control groups, the numbers of both the GFP-LC3 and RFP-LC3 puncta in TCS-treated cells increased, and in the merged images, red and yellow puncta were found (Figure S2E). This demonstrated that during this autophagic process, the fluorescence of some GFP-LC3 quenched, and thus, the autophagosome and lysosome fused in TCS-induced autophagy. Secondly, we use chloroquine (CQ) to verify that TCS induced autophagy. The results of western blot assays showed that in TCS- or CQ-treated RAW264.7 cells, the ratio of LC3-II/LC3-I increased compared with the TCS-alone-treated cells, and the ratio of LC3-II/LC3-I in the cells treated with TCS and CQ was significantly increased (Figure 1C). This demonstrated that during the process of TCS-induced autophagy, the addition of CQ prevented the fusion of the autophagosome and lysosome and resulted in the accumulation of LC3-II. Thus, the procedure of the autophagosome fused with a lysosome was included in TCS-induced autophagy. Overall, these results further suggest that TCS treatment induces an autophagic response.

To explore whether TCS-induced autophagy via the classical mTOR-dependent pathway, RAW264.7 cells were treated with TCS at different concentrations from 0.25 to 32 μM for 90 min. Western blot assays showed that with the increased concentration, the phosphorylation of mTOR, p70S6K, and 4E-BP1 increased compared to the untreated cells (Figure 1D). After treatment with TCS (8 μM), the western blot results showed that during the treatment period (0–720 min), the phosphorylation of mTOR, p70S6K, or 4E-BP1 was not downregulated (Figure 1E). Furthermore, the cells were treated with TCS and Torin1 (an mTOR inhibitor), and the results showed that Torin1 had no effect on the expression of LC3-II (Figure S3A). Therefore, TCS triggered mTOR-independent autophagy in MΦ.

Furthermore, the western blot assay showed that TCS up-regulated the phosphorylation of AMPK and ULK1 in RAW264.7 cells treated with TCS at concentrations from 0.25 to 32 μM for 90 min (Figure 1D). After the treatment with TCS (8 μM) for the different times, ranging from 0 to 720 min, the western blot results showed that with increased treatment time, the phosphorylation of AMPK and ULK1 trended to rise at first and then fall (Figure 1E). In addition, the cells were treated with TCS and compound C (an AMPK inhibitor) to better understand this pathway. The results showed that compound C decreased AMPK and ULK1 phosphorylation and decreased the level of LC3-II expression induced by TCS (Figure S3B). Therefore, TCS stimulated autophagy through AMPK/ULK1 pathway activation.

Next, the cells were treated with TCS at concentrations from 0.25 to 32 μM for 90 min. The western blot assay demonstrated that the levels of phosphorylation of JNK, p38, and ERK were up-regulated in a dose-dependent manner in general (Figure 1F). After treatment with TCS (8 μM) for different times, ranging from 0 to 720 min, the western blot results indicated that phosphorylation of JNK, p38, and ERK presented a trend that increased at first and then fell with time (Figure 1G).

To explore the regulatory relationship among JNK, p38, and ERK in TCS-induced autophagy, the corresponding inhibitors SP600126, SB203580, and PD98059 for JNK, p38, and ERK were used. The results showed that when PD98059 was added, the expression of p-ERK was inhibited, whereas the expression of p-JNK and p-p38 was not affected (Figure S3C). When SB203580 was added, the phosphorylation of p38 and ERK was inhibited, but it had no influence on the phosphorylation of JNK (Figure S3C). SP600126 was also used to pretreat the cells, and the phosphorylation levels of JNK, p38 and ERK were all inhibited (Figure S3D). Meanwhile, the ratio of LC3-II/LC3-I was reduced when the three types of inhibitors were added (Figures S3C,D). These results demonstrated that TCS stimulated autophagy through JNK/p38/ERK pathway activation. Moreover, JNK was upstream of p38 and ERK and regulated the levels of their phosphorylation. As the upstream regulator of ERK, p38 also played a regulatory role in the phosphorylation of ERK.

RAW264.7 cells were transfected with RFP-LC3 plasmids and were incubated with or without pathogens. The cells infected with pathogens were then treated with or without TCS. Then, the images were collected by a confocal laser-scanning microscope. The results showed that in the controls, few p62 or NDP52 and LC3 puncta could be found. There was a co-localization among intracellular pathogens, LC3 and p62, or NDP52 in the S. typhimurium or C. albicans infected cells, and these co-localizations were strengthened by TCS treatment (Figure 2A, Figure S4). After treatment with TCS (8 μM) at different times, ranging from 0 to 90 min, the western blot results showed that with the increase in treatment time, the expression of p62 increased (Figure 2B). Together, the results confirmed that TCS improved the p62 and NDP52 expression and strengthened the co-localization among intracellular pathogens, NDP52 or p62 and LC3.

For the phagocytosis rate assays, RAW264.7 cells were treated with TCS first and were then incubated with S. typhimurium or C. albicans. The number of extracellular pathogens was obtained by colony counting. The results demonstrated that the number of extracellular pathogens decreased when the cells were treated with TCS or Rapa, and the decrease was prevented by 3-MA treatment (Figure 3A, Figure S5A). In addition, for the higher doses of TCS, there was a lower number of extracellular pathogens (Figure 3B, Figure S5B). For the pathogen killing assays, RAW264.7 cells were incubated with S. typhimurium or C. albicans and were then treated with TCS. The results showed that the number of intracellular pathogens decreased with the increasing dose of TCS treatment (Figure 3C, Figure S5C), and the decrease in colony number was prevented by 3-MA treatment (Figure 3D, Figure S5D). Taken together, we concluded that the phagocytic and killing capacity of pathogens by RAW264.7 cells was enhanced by TCS treatment in a dose-dependent manner.

To seek in vivo evidence that TCS strengthens the killing of pathogens by inducing autophagy, we established a mouse model infected with S. typhimurium. The livers and spleens were harvested for bacterial colony counting after 4 days of TCS therapy. With higher doses of TCS, there were fewer surviving S. typhimurium in the livers (Figure 3E) and spleens (Figure 3F), whereas the 3-MA treatment prevented the S. typhimurium from being killed by TCS-induced autophagy both in the livers and spleens (Figures 3G,H). In addition, the mice were infected with C. albicans by oral administration, and the therapy by different concentrations of TCS was conducted three times a day. The tongues were collected for colony counting. The results demonstrated that TCS at concentrations from 6.25 to 20 mg/kg decreased the number of surviving C. albicans in the tongues in a dose-dependent manner (Figure S5E). 3-MA treatment inhibited the TCS-induced autophagy and protected the C. albicans from being killed (Figure S5F). Together, our results demonstrated that TCS strengthens the killing of pathogens by inducing autophagy in vivo.

To explore the effect of TCS on autophagy in the intestine and intestinal flora in vivo, we established a mouse model of a long-term sub-inhibitory concentration of TCS administration. The major intestinal flora in the feces of the mouse model were counted, and the total number of intestinal flora was significantly decreased with the increased TCS concentration. Moreover, the number of Lactobacillus and E. coli significantly decreased, but the E. faecalis significantly increased (Figure 4A). The shifts of the bacterial community compositions were further corroborated by clear clustering of the dominant bacterial genus corresponding to different treatments in the heat map as shown in Figure S6. Firmicutes and Bacteroidetes had higher relative abundances compared with other phyla. Lactobacillus and Escherichia-Shigella had the similar results with Figure 4A. In addition, the western blot results of LC3 confirmed that autophagy occurred in the intestine of the mouse model and that autophagy was more significant with the increased TCS concentration (Figure 4B). To verify the above results, mouse intestinal macrophages and non-phagocytic cells (epithelial cells) were extracted and treated with TCS to induce autophagy. The results showed that the ratio of LC3-II/LC3-I under treatment of TCS or Rapa increased compared to the treatment of the combination of TCS/3-MA or control in both the mouse intestinal macrophages and epithelial cells (Figure 4C). Furthermore, the intracellular pathogen clearance experiments showed that TCS, in a dose-dependent manner, enhanced the killing of the strains Lactobacillus, E. coli, and E. faecalis isolated from mouse intestine by extracting the intestinal macrophages and epithelial cells, whereas 3-MA treatment reversed the results, suggesting that TCS-induced autophagy played a role in enhancing intracellular pathogen clearance (Figure 4D). Taken together, the results showed that the intestinal flora balance of the mouse was broken by long-term sub-inhibitory concentrations of TCS treatment.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.