Salmonella and S. aureus Escape From the Clearance of Macrophages via Controlling TFEB

Phagosome- and xenophagosome-lysosome systems play a critical role in the defense of pathogenic bacteria, such as Salmonella and S. aureus, in macrophages. A great part of the bacteria escapes from the digestion and can survive through some mechanisms that are still poorly understood and which require further exploration. Here we identified that Salmonella inhibited the expression and activation of TFEB to blunt the functions of lysosomes and defense of clearance by activating caspase-1. The expression and activation of TFEB were enhanced early under the infection of S. aureus, which was followed by shrinkage to weaken lysosomal functions due to the delayed activation of ERK, mTOR, and STAT3. Thus, we have identified novel escape mechanisms for Salmonella and S. aureus to deepen and strengthen our strategies fighting with pathogens.


INTRODUCTION
Microbes have been alive in the world for thousands of years along with human beings, but the interplay between microbes and humans is still ongoing. When malignant bacteria infect the human body, the human immune system will work as a fighter against the invasion. The macrophage, as one of the innate immune cells, lends itself to protect us from bacterial infection first and foremost. In macrophages, phagosomes could act as a defense against a huge group of intracellular bacteria by fusing with lysosomes, but also a large group of invaders still survive through their processes of escape (Huang and Brumell, 2014). Meanwhile, the selective macro-autophagosome, known as xenophagosome, will capture the invaders or survivors to transport into lysosomes for clearance (Bauckman et al., 2015;Upadhyay and Philips, 2019;Xu et al., 2019). Obviously, lysosomes act as the terminator to suppress the survival of bacteria in macrophages.
The lysosome, known as "recycling center" of a cell, contains about 60 kinds of hydrolases within its single-membrane vesicles. Those hydrolases, such as cathepsins, collagenase, DNase, esterases, and so on work in an acidic pH (4.5-5.5) environment to degrade macromolecules, including lipid acids, DNA, and proteins (Perera and Zoncu, 2016;Ballabio and Bonifacino, 2020). Because of its powerful ability of digestion, disorders of lysosomal degradation result in a series of diseases, such as cancer, neurodegenerative diseases, metabolic disorders and so on, which are viewed as lysosomal storage disease (LSD) (Platt et al., 2018;Ballabio and Bonifacino, 2020). Specifically, inflammation and microbes infection are closely associated with lysosomes (Bauckman et al., 2015;Upadhyay and Philips, 2019;Xia et al., 2019;Xu et al., 2019).
TFEB, working as a critical regulator of autophagic and lysosomal functions, regulates the transcription of lysosomal membrane proteins and hydrolases, such as LAMP1/2, V-ATPase, cathepsin A/B/F, and so on (Settembre et al., 2011;Napolitano and Ballabio, 2016). The nuclear TFEB could bind to the 10-base E-box-like palindromic sequence (CLEAR) which is shared by the promoter sequence of lysosomal genes (Palmieri et al., 2011). While a group of protein kinases such as mTORC1, ERK2, and AKT could phosphorylate TFEB at ser142 or ser211, and the phosphorylated TFEB can be assembled to a 14-3-3 complex in cytoplasm (Settembre et al., 2011;Martina et al., 2012;Palmieri et al., 2017). When TFEB is dephosphorylated by calcineurin and released from the 14-3-3 complex, it will enter into nucleus for its transcriptional functions (Medina et al., 2015). Meanwhile, the expression level of TFEB also greatly contributes to its activities and functions.
Although it is well-studied that phagocytosis and xenophagy play a great role in the defense of bacteria, it is true that a lot of intracellular invaders survive and replicate (Huang and Brumell, 2014). Do Salmonella and Staphylococcus aureus (S. aureus), as two of common pathogenic bacteria, cross-talk with lysosomes? As reported, the process of phagocytosis could promote the activities of lysosomes, because pathogens bind to the Fcγ receptors and fuel the TFEB (Gray et al., 2016). S. aureus could activate TFEB to up-regulate the expression of inflammatory cytokines in RAW264.7 cells (Visvikis et al., 2014). Enhancing the activities of TFEB to restore the process of xenophagy to restrict the survival of Salmonella (Ammanathan et al., 2019). Yet, the detailed and critical mechanisms of how Salmonella and S. aureus regulate the functions of lysosomes and TFEB expression remain to be explored.
Here, we focused on how Salmonella and S. aureus regulated the activities of lysosomes and expression of TFEB through directly infecting bone marrow derived macrophages (BMDMs). We identified different regulating phenotypes that Salmonella down-regulated the critical genes of lysosomes and Tfeb striking as the activation of caspase1, while S. aureus enhanced the expression of TFEB early, and was reined in later because of the activation of ERK, mTOR, and STAT3 signals. If we inhibited the activation of caspase1 or ERK, mTOR, and STAT3 signals, BMDMs restored the full expression of tfeb and restrained the replication of Salmonella and S. aureus. Thus, we found two novel survival mechanisms of Salmonella and S. aureus.

S. aureus Boosts TFEB Early and Is Reined in Later
To explore how bacteria regulate the processes of lysosomal degradation, we infected BMDMs with Salmonella and S. aureus directly at a time gradient, and tested the lysosomal hydrolase genes, membrane genes and autophagic genes. Showing with histograms, we found that Salmonella remarkably restrained Tfeb, Lamp1, V-Atpase, hydrolase genes and autophagic genes, while S. aureus enhanced the transcription of those genes lightly (Figures 1A,B). Furthermore, under the infection of S. aureus, the level of Tfeb was up-regulated at 1 h, but gradually shrunk at 3 and 5 h ( Figure 1B). As mentioned, Tfeb is an important transcription factor of lysosomal and autophagic genes, and thus we supposed that Salmonella and S. aureus controlled the transcription of Tfeb, which resulted in regulation of lysosomal and autophagic genes.
To confirm this, we tested the protein level of lysosomes and LC3 upon the infection of the two bacteria. The TFEB, LAMP1, ATP6V1A, ATP6V0D2, cathepsin B and LC3 were strongly inhibited by Salmonella. It was same with this gene that the protein level of TFEB was first enhanced, when BMDMs were cocultured with S. aureus, and then reined back (Figure 2A and Supplementary Figures S1A-F). We also verified with Immunofluorescence, and found similar phenotypes (Figures 2B,C). Meanwhile, nuclear localization rates of TFEB were notably increased at 1, 3 h and declined at 5 h with the infection of S. aureus, but not obviously in Salmonella group (Figures 2B,D). This means a discrepant regulation of TFEB activity by the two bacteria. Furthermore, we used LysoTracker red to stain acid vacuoles, and the mean fluorescence intensities (MFI) might relatively represent the lysosomal acidic strength. After infection for 5 h, we found that Salmonella could weaken the lysosomal acidity, while S. aureus could not (Figures 2E-H). Within vivo, C57/B6 mice were infected with S. aureus or Salmonella via intraperitoneal injection, and then we collected the peritoneal macrophages for testing with western-blot. We found TFEB, ATP6V1A, ATP6V0D2, and LC3 were increased under the treatment of S. aureus and inhibited by Salmonella ( Figure 2I and Supplementary Figure S1G).
S. aureus Strikingly Activates ERK, mTOR, NFκB, and STAT3 Signaling Pathways, While Salmonella Activates Caspase-1 To figure out the detailed mechanisms, we infected BMDMs with S. aureus or Salmonella for 0, 1, 3, 5 h, and checked several signaling pathways that might regulate the expression and activity of TFEB. We found that ERK, mTOR, NFκB, and STAT3 could be activated obviously with treatment of S. aureus, but only slightly activated by Salmonella (Figures 3A,C-G). Meanwhile, as reported, The bacterium such as Salmonella or Listeria could activate inflammasomes (Wu et al., 2010;Zhao et al., 2011;Qu et al., 2016), and we tested the mature-caspase 1 level that was released to the supernatant. The results were that S. aureus could not activate caspase-1 but Salmonella could (Figures 3B,H).

VX-765 Restores the Expression of TFEB and Lysosomal Proteins With
Administration of Salmonella, While ERKi, Rapamycin and S31-201 Accelerate the Expression of TFEB and Lysosomal Proteins With Administration of S. aureus To verify which signal could regulate the expression of TFEB and lysosomal proteins, BMDMs were pre-treated with caspase-1 inhibitor, ERKi, rapamycin, S31-201 or IKK16 upon the infection of Salmonella. Because LAMP1 is a key marker of lysosome and ATP6V0D2 has a great role in the autophagylysosome degradation process (Xia et al., 2019), we tested the protein and gene level of the two and TFEB. We found that TFEB, LAMP1, and ATP6V0D2 did not decrease anymore with treatment of VX-765 ( Figures 4A,B and Supplementary Figure S2A), while there were no significant changes under the administration of ERK, mTOR, STAT3, or NFκB inhibitor (Supplementary Figures S2B,C). When BMDMs were infected with S. aureus, the LAMP1 and ATP6V0D2 were much more obviously increased and TFEB was continuously up-regulated in a time-dependent manner after treatments with ERK, mTOR and STAT3 inhibitor (Figures 4C-E and Supplementary Figure S3A). Meanwhile, BMDMs were treated with those inhibitors alone that could induce non-significant responses (Supplementary Figures S4A-D). Together, we suggested that Salmonella could inhibit the expression of TFEB through the activation of caspase-1, and the detailed mechanisms need further follow-up investigations. While S. aureus could block the sufficient expression of TFEB and lysosomal genes by activating ERK, mTOR, STAT3.
VX-765 Restores the Activity of TFEB With Administration of Salmonella, While ERKi, Rapamycin and S31-201 Promote the Activity of TFEB Under Infection of S. aureus The expression level of TFEB was positive-regulated by the inhibitors upon the infection of Salmonella and S. aureus. To explore the activity of TFEB under bacterial infection with the treatment of inhibitors, we isolated the cytoplasm and nucleus of BMDMs to check the level of TFEB. The results showed that VX-765 could promote the TFEB to transfer into the nucleus under infection of Salmonella (Figures 5A,C), while S31-201, rapamycin and ERKi could dramatically increase the nuclear TFEB upon infection of S. aureus (Figures 5B,D).  Meanwhile, to further verify the role of TFEB in the regulation of lysosomal proteins under bacterial infection, we constructed lentivirus to knock down Tfeb in BMDMs, and then infected with Salmonella and S. aureus. We found that if TFEB was knocked down, the ATP6V0D2 was obviously down-regulated on infection of bacteria alone or combined with inhibitors (Figures 5E-J).
VX-765 Reconstitutes the Defense of Salmonella, While ERKi, Rapamycin and S31-201 Enhance the Defense of S. aureus Next, to confirm if the defense of bacteria could be reconstituted or enhanced after administration with inhibitors, inhibitorsprimed BMDMs were cocultured with Salmonella or S. aureus, and measured the number of survivors in BMDMs with a gentamicin protection assay. The results announced that with treatment of caspase-1 inhibitor, less Salmonella survived than in the control group (Figure 6A), and less S. aureus were alive after administration of ERK, mTOR and STAT3 inhibitor ( Figure 6B). To exclude the different endocytosis treatment with inhibitors, inhibitor-primed BMDMs were cocultured with FITC-Dextrain, and the results showed that there were non-significant changes among those inhibitor treatments (Figures 6C,D). Thus, with reconstituting or promoting the expression of TFEB by caspase-1 inhibitor or ERK, mTOR and STAT3 inhibitors, BMDMs strengthened the defense against the invasion of Salmonella and S. aureus.

DISCUSSION
Phagosome-lysosome and autophagosome-lysosome, as two critical mechanisms of digestion defense against the invasion of pathogens, could clear most intruders. But there are a large group of bacteria that continue to survive under excessive infection. Meanwhile, these pathogens have evolved some escape approaches to resist the clearance of immune cells (Huang and Brumell, 2014). In this study, we examined novel escape mechanisms for Salmonella and S. aureus with controlling the expression and activity of TFEB, and then regulating the degradation function of lysosomes. We also verified the detailed signaling pathways that participated in the regulation of the expression and function of TFEB and lysosomes, and through blocking those pathways restored or enhanced the function of lysosomes and defense of bacteria.
Salmonella is a classical intracellular pathogenic germ which induces colitis and poses a huge health burden (Hartz et al., 1950;Herp et al., 2019). Several escape mechanisms have been found that are used by Salmonella in order to survive. For example, when they infect epithelial cells or macrophages, those intracellular bacteria are wrapped up in vacuoles, termed as Salmonella-containing vacuoles (SCVs), through its T3SSs (SPI-1 and SPI-2) for better replication and ability to survive (Birmingham et al., 2006). Meanwhile, at the early invasion, Using lentivirus to knock down TFEB in BMDMs, and those cells were treated with bacteria alone or combined with inhibitors for 3 h. TFEB and ATP6V0D2 were tested with western-blot (E,H) and the quantitative data were displayed by histograms (G-J). Histograms depict mean values (± SEM). *p < 0.05; **p < 0.01; ***p < 0.001. Representative bands were from three independent experiments (A,B,E,F). Salmonella could also trigger xenophagy as consuming energy and amino acid, but the process would quickly be suppressed via degrading Sirt1/LKB1/AMPK complexes (Ganesan et al., 2017). However, the detailed processes for escape should be further explored.
Under the infection of Salmonella, NLRC4 and NLRP3 inflammasomes could be activated within 1 h, cut the procaspase-1 into active caspase-1 in macrophages and induce pyroptosis (Monack et al., 2000;Lara-Tejero et al., 2006;Zhao et al., 2011;Qu et al., 2016). As reported, the totally knock-out Caspase1 in mice, Caspase1 −/− mice, even increased the susceptibility to Salmonella infection in vivo via other mechanisms, such as deficiency of cytokines or involvement of neutrophils (Lara-Tejero et al., 2006;Raupach et al., 2006;Miao et al., 2010). Although Denise M. Monack's group had published that intracellular living bacteria were equal between WT and Caspase1 −/− macrophages through gentamicin protection assay (Monack et al., 2000), the caspase1-depdent cell death was robustly decreased in the caspase1 −/− group (Lara-Tejero et al., 2006). Thus, whether or not caspase1 is involved in the defense of Salmonella in macrophages also needs to be investigated. In our present study, we had identified a critical role of caspase1 in regulating the expression of TFEB, and TFEB contributed to the resistance of pathogens involving Salmonella (Huang and Brumell, 2014;Ammanathan et al., 2019). In our gentamicin protection assay, we measured the living bacteria in each well by folding β-actin, and inhibition of caspase-1 activity obviously stunted the replication of Salmonella. Therefore, we introduced a new surviving pathway, "Salmonella-caspase1-TFEB-lysosome digestion-Salmonella survive, " for Salmonella.
S. aureus is a common pathogenic bacteria to induce systemic infections or abscesses for humans and animals, although it is commensal most of the time. Not only do neutrophils act as the first responders under the infection of S. aureus, but also macrophages have a critical role in clearance of the bacteria (Koziel et al., 2009;Flannagan et al., 2016;Moldovan and Fraunholz, 2019). The professional phagocytes, macrophages, uptake S. aureus within minutes to capture the bacteria with Rab-5 positive phagosomes, and those phagosomes need further maturation. Mature phagosomes can be fused with lysosomes to form phagolysosomes for killing bacteria (Moldovan and Fraunholz, 2019). But, a great part of those phagolysosomes could not completely digest S. aureus as they are lacking some necessary hydrolases (Jubrail et al., 2016). Those survivors could replicate in the acidic vacuoles and finally cause macrophage death (Jubrail et al., 2016;Tranchemontagne et al., 2016). Hence, we focused on the critical mechanisms for the escape of S. aureus in macrophages. Here, we found that S. aureus could activate some key protein kinases, such as ERK, mTOR, STAT3 and so on. ERK and mTOR had been well-studied to control the activity and expression of TFEB, while STAT3 was first verified to blunt the function of TFEB in macrophages under S. aureus infection (Li et al., 2018). Insufficient TFEB led to the disabling lysosomes and more bacteria survived.
In summary, in the battle of bacteria and macrophages, bacteria have generated some escape pathways for better survival. We have verified the great role of TFEB in the defense of bacteria, and explored critical mechanisms to regulate the expression and activity of TFEB during the infection of Salmonella and S. aureus. Those discoveries contribute to the strategies for the cure of infection dramatically.

Cell Culture and Stimulation
All experiments were performed in vitro with mouse primary macrophages that were derived from bone marrow cells (BMDMs). Those cells were cultured with DMEM (Thermo Fisher Scientific, 11965092) medium containing 10% FBS, penicillin, streptomycin and 50ng/ml M-CSF (PeproTech, 315-02) for 7 days. BMDMs were cocultured with Salmonella and S. aureus at a desired MOI (2:1) in the gentamicin protection assay, while MOI (5:1) in direct infection or inhibitor-primed infection experiments.
mRNA Isolation and Real-Time PCR Total mRNA was extracted from infected BMDMs with TRI Reagent (Sigma Aldrich; 93289) and reverse transcription using a kit from Thermo Fisher Scientific (4374966). Real-time PCR was performed with SYBR TM Green mix (Thermo Fisher Scientific, A25742). All real-time PCR primers were listed as follows, Tfeb

Western-Blot
Cellular proteins were extracted with RIPA buffer containing proteases and phosphatase inhibitor cocktail (Thermo Fisher Scientific, 78440) and quantified with a BCA kit (Thermo Fisher Scientific, A53226). Then, protein lysis was mixed with loading buffer and boiled at 100 • C for 5-10 min. The well-prepared lysis was loaded into 10% SDS-PAGE gel for electrophoresis and transferred on to a PVDF membrane. Before being incubated with antibodies against TFEB, LAMP1, ATP6V0D2, ATP6V1A, cathepsin B, LC3, p-STAT3, p-P65, p-PS6, p-ERK, p-AKT, caspase1 p10, and actin, the membranes should be blocked with 5% BSA, and followed by incubating with HRP-secondary antibodies (Cell Signaling Technology, 7074; Abcam, ab6885). Finally, the membranes were exposed under X ray and the bands were quantified with Image J.

Immunofluorescence Staining and Confocal Microscopy
The cells were seeded on glass slides and fixed with 4% formaldehyde for 30 min, permeabilized by 0.05% Triton X-100 about 20 min, blocked with 5% BSA for 1 h and incubated with anti-TFEB antibody overnight at 4 • C. The cells were incubated with FITC-anti-goat secondary antibody for 1 h and DAPI for 10 min in darkness. After fully washing, the slides were placed onto confocal microscopy (Zeiss, Germany) for taking fluorescent photos.

Infection Model in vivo and Bacteria Culture
The C57/B6 mice were purchased from shanghai model organisms. And our animal model was constructed according to the guidelines of the Institutional Animal Care. Salmonella (SL1344) and S. aureus (NCTC8325) were a gift from Xiangping Yang Lab. In the infection model, the single colonies of Salmonella and S. aureus were shaken in 2-4 ml LB medium for 5 h at 37 • C and the concentration of bacteria was quantified to reach an OD 600 of 0.5 with a spectrophotometer (Thermo Fisher Scientific TM ). Those sex-and age-matched mice were divided into three groups, including control group, Salmonella infection group, and S. aureus infection group. The experimental groups were administrated with 1 × 10 8 bacteria suspended in 0.5 ml DMEM medium through intraperitoneal injection for 8 h, and then the mice were sacrificed. Five milliliter germ-free PBS was injected into the peritoneum of the mice and kneaded gently. The macrophages were harvested from the ascites, and the cells lysed with RIPA buffer for western-blot.

Gentamicin Protection Assay
Salmonella and S. aureus were generated from single colonies as before. Mature BMDMs were cocultured with bacteria at an MOI of 2 for 1 h. After three rounds of washing with PBS, the infected BMDMs were incubated with 300 µg/ml gentamicin and 100 µg/ml gentamicin diluted in DMDM medium in turn for 1 h. After three rounds of washing with PBS, the BMDMs were lysed with 0.02% Triton x-100. Then one half of the lysis was diluted with LB into appropriate concentrations and seeded on LB plates for measuring the preliminary CFU. Another half of the protein lysis was quantified with a BCA kit and the final CFU per cell in each group was generated by the preliminary CFU folding protein concentration in each well.

Statistics
All of our data were quantified and drawn with GraphPad Prism 5 and presented as means (± SEM). The measurement of those data through a two tailed Student's t-test and p-values, considered as significant, should be less than 0.05.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

ETHICS STATEMENT
The animal study was reviewed and approved by the Wuhan Central Hospital, Huazhong University of Science and Technology, Wuhan. Written informed consent was obtained from the owners for the participation of their animals in this study.

AUTHOR CONTRIBUTIONS
SR and YX conceived and performed the experiments and analyzed the data. SR, TX, YX, and HZ wrote the manuscript. YX and HZ supervised the project. All authors contributed to the article and approved the submitted version.