Suppression of Starvation-Induced Autophagy by Recombinant Toxic Shock Syndrome Toxin-1 in Epithelial Cells

Toxic shock syndrome toxin-1 (TSST-1), a superantigen produced from Staphylococcus aureus, has been reported to bind directly to unknown receptor(s) and penetrate into non-immune cells but its function is unclear. In this study, we demonstrated that recombinant TSST-1 suppresses autophagosomal accumulation in the autophagic-induced HeLa 229 cells. This suppression is shared by a superantigenic-deficient mutant of TSST-1 but not by staphylococcal enterotoxins, suggesting that autophagic suppression of TSST-1 is superantigenic-independent. Furthermore, we showed that TSST-1-producing S. aureus suppresses autophagy in the response of infected cells. Our data provides a novel function of TSST-1 in autophagic suppression which may contribute in staphylococcal persistence in host cells.


Introduction
Autophagy is a fundamental cellular homeostatic mechanism that provides for bulk degradation of organelles and cytosolic proteins [1]. During autophagy, parts of cytoplasm and organelles are encapsulated into double membrane vacuoles called autophagosomes, which encounter the lysosomes to digest the sequestered recycling components for self-support [2]. Autophagy-mediated proteolysis plays a crucial role in survival, growth, proliferation and differentiation of eukaryotic cells [3,4]. In addition, autophagy is involved in the defense against several pathogenic microorganisms [5]. It was previously postulated that some intracellular bacteria are targeted by autophagic degradation system [6]. They are sequestrated within autophagosomes, which ultimately deliver the microorganisms to lysosome to be eliminated. However, successful pathogens have evolved strategies to avoid autophagy, or to actively subvert its components, to promote their own replication [6,7].
Staphylococcus aureus is an important human pathogen which causes a variety of infection ranging from superficial infections to more life-threatening diseases [8]. S. aureus has been classically considered as an extracellular pathogen but numerous studies have shown that S. aureus can invade cells and replicate intracellularly [9]. This bacterium is able to infect various types of nonprofessional phagocytic host cells such as keratinocytes, fibroblasts, endothelial and epithelial cells [10,11]. One of the key features of S. aureus infection is the production of series of virulence factors, including secreted enzymes and toxins whose expression is regulated by a set of global virulence regulators [12,13]. Previous studies suggested a connection between autophagic response and S. aureus infection which occurs via the bacterial agr-virulence factor [14,15,16]. Pore-forming a-hemolysin, regulated by the agr is shown to participate in the activation of the autophagic pathway [15].
Toxic shock syndrome toxin-1 (TSST-1) is one of pyrogenic superantigens secreted by S. aureus. Potent effects of TSST-1 on host immune system have been largely elucidated [12]. This toxin directly crosslinks between the major histocompatibility complex class II molecule on antigen-presenting cell and T cell receptor bearing specific Vb element. This binding subsequently leads to a massive proliferation of T cells and the uncontrolled release of proinflammatory cytokines [17,18]. Previous studies have shown that TSST-1 binds to an uncharacterized receptor(s) on endothelial cells and epithelial cells [19,20] and penetrate into epithelial cells [21]. In addition, immunization with recombinant and/or mutant TSST-1 protects mice against systemic S. aureus infection [22,23]. Except for neutralization of superantigenic activity, these toxin-specific antibodies alter bacterial growth in the organs of mice. These data suggest another biological function of this toxin in the non-immune cells. Although the production of this toxin is under the control of several regulatory proteins, its expression is also partially regulated by the agr [24]. In this study, we investigated the effect of TSST-1 on autophagy in HeLa 229 cells. Our results suggest that TSST-1 suppresses autophagy. Furthermore, this suppression is superantigenic activity-independent.

Materials and Methods
Bacterial strains and growth conditions S. aureus 834 wild type (WT), a clinical septic isolate that produces TSST-1 [25], and its derivative TSST-1-deficient mutant (Dtst) were cultured at 37uC in tryptic soy broth (BD Bioscience, Sparks, MD) or tryptic soy agar for 16 h. The bacterial cells were collected, suspended in phosphate-buffered saline (PBS) and adjusted spectrophotometrically at 550 nm.

Electron microscopy
HeLa 229 cells were cultivated on sterilized glass slides and the autophagy was induced under nutrient-starvation condition with or without 10 mg/ml rTSST-1 in the presence of lysosomal protease inhibitors. At 4 h after induction, the cells were fixed with 4% paraformaldehyde, 1% glutaraldehyde (Wako) in PBS, and then post-fixed with 1% osmium tetroxide (Heraeus Chemicals, Port Elisabeth, South Africa) in 0.1 M phosphate buffer (pH 7.4). The samples were dehydrated through a graded series of ethanol (Wako) and propylene oxide (Wako) at room temperature, and embedded in Epon 812 resin (TAAB Laboratories Equipment Ltd., Berkshire, UK). They were then polymerized with the resin in gelatin capsules (No. 0; Eli Lilly Co., Indianapolis, IN) at 60uC for 48 h. After polymerization, the samples on glass slides were transferred to resin block. Ultra-thin sections (70-80 nm) were cut   with a diamond knife, stained with Sato's lead citrate [26] and uranyl acetate (Merck, Darmstadt, Germany), and observed under a transmission electron microscope JEM 1250 (JEOL Ltd., Tokyo, Japan) at 80 kV.

SDS-PAGE and Western blotting
After induction of autophagy in the presence of rTSST-1, mTSST-1, SEA, SEB or SEC with and without lysosomal protease inhibitors, crude proteins from HeLa 229 cells were collected in lysis buffer [2% triton X-100 in PBS containing complete protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany)] and applied to 12.5% polyacrylamide gel. The proteins were transferred to polyvinylidene fluoride membrane (Immobilon-P, Millipore, Bedford, MA). The membrane was then blocked for 2 h with 5% skim milk in Tris-buffered saline (20 mM Tris-pH 7.5, 150 mM NaCl, 0.05% Tween 20), washed twice with Trisbuffered saline, and incubated with a primary antibody anti-LC3 (Sigma) or b-tubulin (Santa Cruz Biotechnology, Inc., CA). The signal was detected by peroxidase-conjugated anti-rabbit IgG (MP Biomedicals) and SuperSignal West Dura Extended Duration Substrate (Pierce Biotechnology Inc., Rockfored, IL). The intensity of LC3-II band was quantified using Image Lab software normalized with intensity of b-tubulin band. The amount of LC3-II was calculated relatively to the amount of LC3-II from autophagic induction condition with lysosomal protease inhibitors, which set to 1.

S. aureus infection
S. aureus Dtst was constructed from the WT (File S1). GFP-LC3-expressing HeLa 229 cells were infected with S. aureus 834 or Dtst at multiplicity of infection of 100. After incubation for 45 min, the extracellular bacteria were eliminated with 100 mg/ml lysostaphin (Wako). At 6 h of infection, the cells were fixed, washed, lysed and blocked as described above. S. aureus cells were immunostained with anti-Staph. aureus antibody (ViroStat, Inc, Portland, ME) and rhodamine-conjugated anti-rabbit IgG (MP Biomedical). S. aureus cells and GFP-LC3 puncta were observed from under confocal microscope.

Statistical analysis
Data were expressed as means 6 standard deviations, and P, 0.05 from student's t test analysis was used to determine the significance of the differences.

rTSST-1 suppresses autophagosome accumulation in nutrient-starved HeLa 229 cells
To investigate the effect of TSST-1 on autophagy, LPS-free rTSST-1 was prepared and HeLa 229 cells were transfected with pEGFP-hLC3 plasmid. The effect of rTSST-1 on autophagy was then investigated in the GFP-LC3 expressing cells under both nutrient-rich (MEM) and nutrient-starvation (KRB) condition. As shown in Figure 1A-1B, the GFP-LC3 puncta in mock-transfected cells (transfection with pEGFP-C2 plasmid) was not observed in any conditions (MEM and KRB with and without rTSST-1). For the pEGFP-hLC3-transfected cells under nutrient-rich condition (MEM) in which autophagy was not induced, only small amount of GFP-LC3 puncta in these cells was observed. On the other hand, in the pEGFP-hLC3-transfected cells under nutrientstarvation condition (KRB) in which autophagy was induced, the average of GFP-LC3 puncta up to 11 per cell was found. Importantly, rTSST-1 did not significantly alter and/or induce the amount of GFP-LC3 puncta in the cells under nutrient-rich condition. In contrast, the amount of GFP-LC3 puncta in the autophagic-induced cells was significantly reduced by the addition of 10 mg/ml rTSST-1. The results indicate that rTSST-1 suppresses the autophagosomal accumulation in the autophagicinduced HeLa 229 cells.

rTSST-1 does not enhance lysosomal fusion and autophagosomal degradation
Autophagy is a dynamic process which comprises autophagosomal synthesis and autophagosomal degradation. In order to determine whether the autophagosome suppression by rTSST-1 is involved in lysosomal fusion process, lysosomes in the GFP-LC3expressing HeLa 229 cells were immunostained with LAMP1 and the overlapping between GFP-LC3 and lysosome was observed. As shown in Figure 2A-2B, total amount of GFP-LC3 puncta was reduced by the addition of rTSST-1, correlating to the results obtained in Figure 1. However, the percent of lysosomes fused with GFP-LC3 puncta was not significantly changed or enhanced by the addition of rTSST-1. It was around 41-47% of total GFP-LC3 puncta per cell. Similar results were also obtained by staining acidic pH of lysosomes using LysoTracker Red ( Figure S1). We further observed whether rTSST-1-dependent autophagosome suppression is involved in an enhancement of autophagosomal degradation. Autophagic flux was observed at various times of autophagic induction with or without the addition of the lysosomal protease inhibitors. As expected, the GFP-LC3 puncta which increased upon the time of autophagic induction was suppressed by rTSST-1 ( Figure 3A-3B). Furthermore, the addition of lysosomal protease inhibitors failed to restore autophagosomes in the rTSST-1-treated cells. These results indicate that rTSST-1 does not enhance autophagosome-lysosome fusion and autophagosomal degradation. To avoid false interpretation that may occur by overexpression of GFP-LC3, the autophagic-suppressing activity of rTSST-1 was confirmed by immunostaining of LC3 and electron micrographs ( Figure 4A-4C). The number of LC3 puncta from immunostaining and autophagosome-like vacuoles in electron micrographs was reduced by the addition of rTSST-1 to the nutrient-starved cells that supplemented with lysosomal protease inhibitors. rTSST-1 suppresses LC3-II accumulation in both KRB and rapamycin treatment LC3-II accumulation is an important marker for autophagosome. We also examined the accumulation of LC3-II in the HeLa 229 cells by Western blotting using anti-LC3 antibody. As expected, the results in Figure 5A indicated that the amount of LC3-II in the nutrient-starved cells (KRB) was significantly higher than that of nutrient-rich condition (MEM). In addition, rTSST-1 suppressed the accumulation of LC3-II in these cells in a dosedependent manner. A similar result was also found in the cells treated with rapamycin ( Figure 5B). The amount of LC3-II in the cells treated with rapamycin was significantly higher than that of DMSO control and the amount of LC3-II in the rapamycintreated cells was reduced by the addition of rTSST-1 in a dosedependent manner ( Figure 5C). A dose-dependent response of rTSST-1 analyzed by GFP-LC3 puncta formation in the nutrientstarved cells was also shown in Figure S2.

TSST-1-producing S. aureus suppresses autophagy
To confirm whether TSST-1 produced by S. aureus suppresses autophagy, Dtst was constructed and characterized (see File S1 and Figure S3-S7). After infecting the pEGFP-hLC3-transfected HeLa 229 cells with WT or Dtst for 6 h, S. aureus cells were immunostained with anti-S. aureus antibody. The amount of GFP-LC3 puncta in the cells with an equivalent number of S. aureus was analyzed. As shown in the Figure 6A-6D, the number of S. aureus between WT and Dtst in the selected cell was not significantly different but the GFP-LC3 puncta in the cells infected with Dtst were higher than those with the WT. In addition, colocalization of Dtst cells with GFP-puncta was also higher than that with the WT. The results suggest that the TSST-1-producing S. aureus suppresses autophagy in the response of infection.
The enhancement of GFP-LC3 puncta due to Dtst-infection was not only found in HeLa 229 but also found in the human epithelial kidney HEK293 cells and human intestinal epithelial 407 cells ( Figure S8). The effect of TSST-1 on autophagy suppression The results demonstrated that LC3-II accumulation in the Dtstinfected cells was reduced by addition of rTSST-1 ( Figure S9).

Suppression of autophagy by TSST-1 does not depend on superantigenic activity
In order to determine whether suppression of autophagosome depends on superantigenic activity of TSST-1, the accumulation of LC3-II in the nutrient-starved cells was observed by addition of mTSST-1 or SEs (SEA, SEB and SEC). mTSST-1 is a H135A mutant of rTSST-1 lacking of superantigenic activity, whereas the SEA, SEB and SEC are enterotoxins that exhibit superantigenic activity. The results in Figure 7A-7B demonstrated that the rTSST-1 and mTSST-1 had a similar autophagic-suppressing effect. In contrast, SEA, SEB and SEC did not suppress the accumulation of LC3-II in the autophagic induced cells. These results indicate that the autophagic suppression is specific to TSST-1 and does not depend on its superantigenic activity.

Discussion
Besides well-clarified superantigenic activity of TSST-1 in immune cells, our results suggest a novel function of TSST-1 in epithelial cells that is participating in autophagy. We demonstrated that TSST-1 suppresses autophagy in the autophagic-induced HeLa 229 cells. Non-increasing of autophagosome-lysosome fusion and non-restoring of autophagy by the addition of lysosomal protease inhibitors suggested that rTSST-1 may inhibit autophagosomal synthesis rather than enhance autophagosome degradation. In addition, this autophagic suppression was similarly found in the cells induced with nutrient-starvation and rapamycin treatment, suggesting that TSST-1 may suppress canonical autophagy pathway. Although TSST-1 shares superantigenic activity with SEs, its primary sequence is shorter and has a homologous limitation. Unlike rTSST-1 and mTSST-1, SEs did not suppress autophagy. These results suggest that autophagic suppressing activity of TSST-1 does not depend on superantigenic activity. In contrast, it requires a specific structure of TSST-1 that does not share with SEs.
The reason of autophagic suppression by TSST-1 is still elusive. In the current report, we also presented evidence that not only purified rTSST-1 is able to suppress autophagy in the autophagic-induced cells but also the TSST-1-secreting S. aureus suppresses autophagy in the response of infection. Thus, autophagic suppression by TSST-1 might contribute in staphylococcal infection. S. aureus is known as a major human pathogen which can be carried by healthy persons [27]. To escape host immune response, S. aureus has an effective strategy of persistence on mucosal surface and hiding within the host cells. Previous studies demonstrated that superantigenic activity of TSST-1 can modulate immune response, leading to an immunosuppressive state [12,18]. However, local production of TSST-1 may be insufficient to cause large-scale systemic immunosuppression. Thus, the local effect of TSST-1 at the colonization site might promote persistence of organism. Tuchscherr and coworkers demonstrated that S. aureus is able to invade and persist within non-phagocytic cells for several weeks after the infection [28]. To invade into the non-phagocytic cells, actin cytoskeleton reorganization regulated by integrin-linked kinase is required [29]. Our results indicated that entry of S. aureus into epithelia does not interfere with TSST-1 ( Figure S3-S5). After entry into the host cells, S. aureus requires appropriate characteristics to survive intracellularly including not killing the host cells and resisting or non-activating intracellular host defenses. Although S. aureus has been shown to induce autophagy via cAMP down regulation [30], the effect of autophagy to intracellular S. aureus is still unclear. In the model of Schnaith and coworkers, agr-positive S. aureus localizes in autophagosome-like vesicles, where S. aureus replicates and subsequently escapes into the cytoplasm, to promote host cell death [14]. On the other hand, Mauthe and coworkers found that S. aureus cells were entrapped in autophagosome-like vesicles which then are targeted for lysosomal degradation [16]. In fact, to survive in the host cells without induction of host cell death, S. aureus needs to down-regulate autophagy. It has been shown that percent persistence of TSST-1 producing strains is higher than non-producing strains [31]. Thus suppression of autophagy by TSST-1 might be an alternative strategy of S. aureus for persistence in the host cells. To demonstrate this idea, the intracellular bacterial number as well as the viability of S. aureusinfected cells should be investigated and compared between S. aureus WT and Dtst in further experiments. Although these intracellular persisting behaviors remained to be compared, the data in this study provides a novel function of TSST-1 participating in autophagic suppression. Figure 7. Suppression of LC3-II accumulation by rTSST-1 in the cells is superantigenic activity-independent. Autophagy in HeLa 229 cells was induced by nutrient-starvation (KRB) with or without lysosomal protease inhibitors and 10 mg/ml rTSST-1, mTSST-1, SEA, SEB or SEC. Cells in MEM were used control. At 4 h of induction, LC3-II was detected by Western blotting (A) and the intensity of LC3-II band was quantified (B) as described in Figure 5C. The data is provided as SD of at least 3-independent experiments. doi:10.1371/journal.pone.0113018.g007