Acid Sphingomyelinase Promotes Cellular Internalization of Clostridium perfringens Iota-Toxin

Clostridium perfringens iota-toxin is a binary actin-ADP-ribosylating toxin composed of the enzymatic component Ia and receptor binding component Ib. Ib binds to a cell surface receptor, forms Ib oligomer in lipid rafts, and associates with Ia. The Ia-Ib complex then internalizes by endocytosis. Here, we showed that acid sphingomyelinase (ASMase) facilitates the cellular uptake of iota-toxin. Inhibitions of ASMase and lysosomal exocytosis by respective blockers depressed cell rounding induced by iota-toxin. The cytotoxicity of the toxin increased in the presence of Ca2+ in extracellular fluids. Ib entered target cells in the presence but not the absence of Ca2+. Ib induced the extracellular release of ASMase in the presence of Ca2+. ASMase siRNA prevented the cell rounding induced by iota-toxin. Furthermore, treatment of the cells with Ib resulted in the production of ceramide in cytoplasmic vesicles. These observations showed that ASMase promotes the internalization of iota-toxin into target cells.

We recently reported that C. botulinum C2 toxin needs ASMase for cell entry [22]. Ca 2+ influx induced by C2 toxin facilitates the exocytosis of lysosomal ASMase. The extracellularly released ASMase cleaves sphingomyelin in the outer membrane, which results in the production of ceramide-rich microdomains. The microdomains bud into cytoplasmic membranes, initiating endocytosis [6]. It is unclear whether ASMase is involved in the internalization of iota-toxin in host cells. Madin-Darby canine kidney (MDCK) cells offer an excellent model for studying the uptake of iota-toxin [12]. In the present study, we examined whether ASMase participates in the endocytosis of iota-toxin.

Calcium Ion Accelerates Endocytosis of Iota-Toxin
We investigated the role of Ca 2+ in target cell entry of iota-toxin. Having previously demonstrated that the cell rounding activity of iota-toxin was enhanced by increased extracellular Ca 2+ concentrations (0.2-1.8 mM) [23], we evaluated the cytotoxicity of iota-toxin in the absence or presence of Ca 2+ . MDCK cells treated with iota-toxin in extracellular Ca 2+ medium caused quick cell rounding ( Figure 1A). In contrast, in calcium-free extracellular medium, the toxin evoked weak cell rounding. An in vitro ADP-ribosylation assay demonstrated that iota-toxin was able to ADP-ribosylate the G-actin both in the absence and presence of Ca 2+ ( Figure 1B). To examine the role of Ca 2+ on the entry of Ib, MDCK cells were treated with Ib in the absence or presence of Ca 2+ at 37 • C ( Figure 1C). Ib effectively entered the cytoplasm in the presence of Ca 2+ . In contrast, in medium that lacked Ca 2+ , Ib was present on the cell membrane. We previously showed that the binding and oligomerization of Ib to MDCK cells was detected in Ca 2+ -free buffer, as well as in Ca 2+ -containing buffer [23], indicating that the binding and oligomer formation of Ib to target cells do not need extracellular Ca 2+ . Pore-forming toxin-induced Ca 2+ influx initiates endocytosis during membrane repair [20,21]. We already reported that Ib causes Ca 2+ influx into MDCK cells [12]. These results indicate that Ib facilitates Ca 2+ uptake from the extracellular fluid during endocytic internalization.
lacked Ca 2+ , Ib was present on the cell membrane. We previously showed that the binding and oligomerization of Ib to MDCK cells was detected in Ca 2+ -free buffer, as well as in Ca 2+ -containing buffer [23], indicating that the binding and oligomer formation of Ib to target cells do not need extracellular Ca 2+ . Pore-forming toxin-induced Ca 2+ influx initiates endocytosis during membrane repair [20,21]. We already reported that Ib causes Ca 2+ influx into MDCK cells [12]. These results indicate that Ib facilitates Ca 2+ uptake from the extracellular fluid during endocytic internalization. Cells were fixed, permeabilized, and stained with an anti-Ib antibody and 4 ,6 -diamino-2-phenylindole (DAPI). Ib (red) and the nucleus (blue) were viewed using a confocal microscope. The same studies were performed four times, and a typical result is shown. Bar, 7.5 µm.

Ib Promotes Exocytosis of Lysosomal Acid Sphingomyelinase
Ca 2+ -triggered exocytosis of the lysosomal enzyme ASMase promotes endocytosis of lesions [19][20][21]. As depicted in Figure 2A,B, Ib elevates ASMase activity in the extracellular medium in a concentrationand time-dependent fashion in Ca 2+ medium, but not in Ca 2+ -free medium. The origin of ASMase in the extracellular medium is proposed to be lysosomal exocytosis. To examine whether Ib causes the exocytosis of lysosomes, the extracellular medium from the cells treated with Ib was evaluated for the presence of β-hexosaminidase (βHex) as a lysosomal marker protein [22] ( Figure 2C). βHex activity was time-dependently elevated in the extracellular medium from the Ib-incubated cells compared to the control cells. This result indicated that Ib facilitates lysosomal exocytosis. We next examined whether Ib induces the activation of neutral sphingomyelinase (NSMase) ( Figure 2D). The extracellular medium from Ib-treated cells exhibited increased ASMase activity. In contrast, Ib did not induce an increase in NSMase activity. evaluated for the presence of β-hexosaminidase (βHex) as a lysosomal marker protein [22] ( Figure  2C). βHex activity was time-dependently elevated in the extracellular medium from the Ib-incubated cells compared to the control cells. This result indicated that Ib facilitates lysosomal exocytosis. We next examined whether Ib induces the activation of neutral sphingomyelinase (NSMase) ( Figure 2D). The extracellular medium from Ib-treated cells exhibited increased ASMase activity. In contrast, Ib did not induce an increase in NSMase activity. (1.8 mM) medium or Ca 2+ -free medium with Ib (400 ng/mL) at 37 • C for the indicated incubation periods; (B) MDCK cells were treated either in Ca 2+ (1.8 mM) medium or Ca 2+ -free medium with various amounts of Ib at 37 • C for 30 min. Acid sphingomyelinase (ASMase) activity in the culture medium was measured as described in the Materials and Methods. Non-treated cells employed as controls are set at a basal level of 1.0. Results are indicated as relative values of the values acquired from non-treated controls (means ± standard deviations (SD) (n = 4)); (C) MDCK cells were treated with Ib (500 ng/mL) at 37 • C for the indicated incubation periods. Supernatant fluids from the culture were evaluated for β-hexosaminidase (βHex) activity. βHex activities are represented as the percentage of the total enzymatic activity detected in supernatant fluid and cells (means ± standard deviations (SD) (n = 4)); (D) Cells were treated with Ib (400 ng/mL) for 60 min at 37 • C. ASMase activity in supernatant fluids from the culture and neutral sphingomyelinase (NSMase) activity in cell were evaluated as described in the Materials and Methods. Results are expressed as percentages of the values acquired from non-treated controls (means ± standard deviations (SD) (n = 4)).

Role of Acid Sphingomyelinase on Cellular Uptake of Iota-Toxin
We investigated whether ASMase is involved in the cellular internalization of iota-toxin. MDCK cells were treated with ASMase inhibitors amitriptyline and imipramine and evaluated for susceptibility to iota-toxin ( Figure 3A). The cell rounding activity caused by iota-toxin was suppressed by treatments with amitriptyline (Ami) and imipramine (Imi). Bromoenol lactone (BEL), an inhibitor of lysosomal exocytosis, also inhibited the toxin-induced cytotoxicity. In contrast, GW4869 (GW), which is a specific inhibitor of NSMase, did not influence the cell rounding caused by the toxin. Then, we examined the effect of amitriptyline on cellular uptake of Ib ( Figure 3B). Ib was localized in intracellular vesicles in the presence of a vehicle. In contrast, incubation of Ib with MDCK cells that were preincubated with amitriptyline decreased the Ib internalization, and Ib was instead observed in the cell membrane. These results reveal that ASMase is involved in Ib cell entry into target cells.
the toxin. Then, we examined the effect of amitriptyline on cellular uptake of Ib ( Figure 3B). Ib was localized in intracellular vesicles in the presence of a vehicle. In contrast, incubation of Ib with MDCK cells that were preincubated with amitriptyline decreased the Ib internalization, and Ib was instead observed in the cell membrane. These results reveal that ASMase is involved in Ib cell entry into target cells.  4)). Significant differences (Student's t-test) from control cells are shown. * p < 0.01, significantly different from DMSO plus iotatoxin. Abbreviations: DMSO, dimethyl sulfoxide; Ami, amitriptyline; Imi, imipramine; BEL, bromoenol lactone; (B) MDCK cells were preincubated with amitriptyline (10 M) at 37 °C for 1 h. Cells were incubated with Ib (1 g/mL) at 37 °C for 30 min. Cells were fixed, permeabilized, and stained with the anti-Ib antibody, Alexa Fluor488-phallodin and DAPI. Ib (red), actin (green), and the nucleus (blue) were analyzed using a confocal microscope. A representative data from one of three studies is shown. Bar, 7.5 m.

Effects of ASMase-siRNA on Iota-Toxin-Caused Cytotoxicity
To examine the influence of ASMase on iota-toxin-caused cell rounding activity, an RNAi analysis was used to silence ASMase. Treating cells with ASMase siRNA resulted in decreased expression of ASMase as compared to intact cells or negative control (NC)-siRNA-transfected cells ( Figure 4A). Compared with intact or NC-siRNA, ASMase siRNA transfection could markedly reduce iota-toxin-induced cell rounding ( Figure 4A). These results showed that ASMase plays a role in iota-toxin-induced cytotoxicity.

Effects of ASMase-siRNA on Iota-Toxin-Caused Cytotoxicity
To examine the influence of ASMase on iota-toxin-caused cell rounding activity, an RNAi analysis was used to silence ASMase. Treating cells with ASMase siRNA resulted in decreased expression of ASMase as compared to intact cells or negative control (NC)-siRNA-transfected cells ( Figure 4A). Compared with intact or NC-siRNA, ASMase siRNA transfection could markedly reduce iota-toxin-induced cell rounding ( Figure 4A). These results showed that ASMase plays a role in iota-toxin-induced cytotoxicity.

Iota-Toxin Causes An Increase in Ceramide
To confirm the role of ASMase in the toxin-caused cytotoxicity, we investigated whether Ib causes the production of ceramide in target cells. Subcellular localization of ceramide production caused by Ib was visualized by immunofluorescence staining using an anti-ceramide antibody. As shown in Figure 4B, Ib caused an elevation of ceramide production in a time-dependent fashion. Ib colocalized with ceramide in intracellular vesicles after 30 min, indicating that the generation of ceramide in the outer leaflet of the plasma membrane by Ib promotes Ib entry. Then, we examined whether Ib causes the production of ceramide. When MDCK cells were incubated with Ib at 37 • C for 30 min, Ib induced the production of ceramide in a dose-dependent fashion ( Figure 4C). Ib-induced ceramide production was inhibited by the addition of amitriptyline. Following heat denaturation, inactivated Ib did not cause the ceramide production.

Iota-Toxin Causes An Increase in Ceramide
To confirm the role of ASMase in the toxin-caused cytotoxicity, we investigated whether Ib causes the production of ceramide in target cells. Subcellular localization of ceramide production caused by Ib was visualized by immunofluorescence staining using an anti-ceramide antibody. As shown in Figure 4B, Ib caused an elevation of ceramide production in a time-dependent fashion. Ib colocalized with ceramide in intracellular vesicles after 30 min, indicating that the generation of ceramide in the outer leaflet of the plasma membrane by Ib promotes Ib entry. Then, we examined whether Ib causes the production of ceramide. When MDCK cells were incubated with Ib at 37 °C for 30 min, Ib induced the production of ceramide in a dose-dependent fashion ( Figure 4C). Ib-induced ceramide production was inhibited by the addition of amitriptyline. Following heat denaturation, inactivated Ib did not cause the ceramide production.

Discussion
In this study, Ib facilitated ASMase release and then ceramide production in the exterior cell surface. Formation of ceramide platforms in the plasma membrane induces a negative curvature in the membrane, leading to an inward vesiculation for its endocytic uptake. The present results indicated that ASMase plays a critical role in the entry of iota-toxin into target cells.
Pore formation of PFT in the cell membrane causes the host cell injury. Nucleated cells repair plasma membrane wounding through a molecular mechanism underlying Ca 2+ -dependent lysosomal

Discussion
In this study, Ib facilitated ASMase release and then ceramide production in the exterior cell surface. Formation of ceramide platforms in the plasma membrane induces a negative curvature in the membrane, leading to an inward vesiculation for its endocytic uptake. The present results indicated that ASMase plays a critical role in the entry of iota-toxin into target cells.
Pore formation of PFT in the cell membrane causes the host cell injury. Nucleated cells repair plasma membrane wounding through a molecular mechanism underlying Ca 2+ -dependent lysosomal exocytosis, ASMase release, and rapid endocytosis of lesion sites. Some bacteria gain access to target cells by producing PFTs and SMases. In a previous study, Ib formed cation-selective channels in lipid bilayer membranes [4]. We reported that Ib causes K + release from Vero cells [10] and shows cytotoxicity in A549 cells [13]. Moreover, Ib facilitates Ca 2+ influx during cell entry [12]. In this research, Ib was readily endocytosed under conditions of Ca 2+ buffer but not Ca 2+ -free buffer. We speculate that extracellular release of ASMase by Ca 2+ -mediated lysosomal exocytosis is involved in an early stage of Ib entry to the host cells. In this experiment, Ib induced the extracellular release of lysosomal ASMase. The cytotoxic activity of iota-toxin toward target cells was inhibited by ASMase inhibitors and ASMase knockdown. Moreover, Ib increased the production of ceramide. It has been reported that Ca 2+ influx in injured cells stimulates a repair pathway that leads to resealing of the cell membrane [24]. Ca 2+ -induced exocytosis of lysosomes facilitates the membrane repair of host cells wounded by PFTs [25,26]. Exocytosis of lysosomes in injured cells is associated with an increased endocytosis that eliminates pores formed by PFTs and lesion sites from the cell membrane [26]. Ca 2+ -induced exocytosis of the lysosomal ASMase accelerates the endocytosis that facilitates active membrane repair pathway in injured cells. ASMase hydrolyses the cell outer membrane sphingomyelin into ceramide [27], and newly generated ceramide-rich platforms can cause inward bending of the affected plasma membrane [28,29]. Therefore, Ib utilizes this plasma membrane repair mechanism as an entry pathway. These studies indicate that iota-toxin is incorporated into host cells through Ca 2+ -triggered lysosomal ASMase exocytosis.
Iota-toxin-caused cytotoxicity was not blocked by NSMase inhibitor GW4869. Additionally, NSMase was not activated by Ib treatment. According to these findings, NSMase does not play a role in the cytotoxicity of iota-toxin. In the case of iota-toxin, hydrolysis of sphingomyelin by ASMase delivered by Ib produces a negative curvature in the outer layer. This negative curvature leads to an inward vesiculation for endocytosis. ASMase is responsible for cellular uptake of iota-toxin, while NSMase is involved in the generation of ceramide-riched microdomain in the inner layer of the plasma membrane, thereby leading to an outward curvature that is implicated in the exosomal shedding of PFTs [30]. Ib treatment of MDCK cells cannot induce cell blebbing relating to the exosomal shedding. Again, NSMase has no involvement in the cytotoxicity of iota-toxin.
Sphingomyelin is abundant in the cytoplasmic membrane and is observed in the cholesteroland sphingolipid-rich plasma membrane domains, termed lipid rafts [21]. Enzymatic hydrolysis of sphingomyelin by exocytosed ASMase produces ceramide. Ceramide supports lipid raft enlargement into large membrane domains in which cell-surface receptors cluster. LSR is a single-pass transmembrane receptor expressed in various tissues including the liver and intestine [31]. LSR serves as the cellular receptor for clostridial binary toxins, namely iota-toxin, C. difficile CDT, and C. spiroforme toxin [9,32]. It has been reported that the binding component of CDT causes accumulation of LSR into lipid rafts [31]. Previously, we reported that Ib binds to a host cell receptor, accumulates in plasma membrane lipid rafts, and internalizes into the cells [11]. Our findings indicate that Ib induces the clustering of LSR into lipid rafts and is endocytosed from ceramid-rich lipid rafts.

Conclusions
We demonstrated an important function of ASMase in cellular uptake of iota-toxin. Ib-induced elevation of intracellular Ca 2+ concentration promotes lysosomal exocytosis. An extracellularly secreted ASMase generates the invaginated ceramide microdomain into the cell, triggering endocytosis that leads to the cellular uptake of iota-toxin.

Cell Culture Assay
MDCK (Madin Darby Canine Kidney) cells were purchased from the Riken BRC (Tsukuba, Japan). Cells were cultivated in DMEM containing 10% fetal bovine serum (FBS), 100 µg/mL of streptomycin, 100 units/mL of penicillin, and 2 mM L-glutamine (FBS-DMEM). Cells were incubated at 37 • C in a 5% CO 2 atmospheres under humidified conditions. For cytotoxicity experiments, cells were seeded in 48-well tissue culture plates and incubated in FCS-DMEM. Various concentrations of Ia and Ib were added and cells were further incubated at 37 • C. After 4 h, cells were observed for morphological changes as described earlier [13]. For evaluation of the role of Ca 2+ during iota-toxin-induced cytotoxicity, culture media were removed and replaced with HBSS with or without 1.8 mM Ca 2+ . For inhibitor treatment, cells were incubated for 30 min with 25 µM imipramine, 25 µM amitriptyline, 25 µM GW4869, or 25 µM BEL at 37 • C. Thereafter, iota-toxin was added and cells were additionally incubated at 37 • C in the presence of toxin plus inhibitor. In some experiments, an inactive form of Ib was produced by heating for 5 min at 95 • C. For detection of ASMase and beta-actin, Western blot analysis was performed using specific antibodies as described earlier [22].

Assay of β-Hexosaminidase
MDCK cells were seeded in 24-well plates and incubated with Ib or PBS at 37 • C. After the given incubation periods, the culture supernatants were removed, and centrifuged at 10,000 g for 20 min. Activity of β-hexosaminidase in the supernatants was determined by incubation with 1 mM p-nitrophenyl N-acetyl-β-D-glucosaminide in 0.1 M citrate buffer (PH 4.5) at 37 • C for 2 h. To stop the reaction, 0.1 M sodium carbonate buffer (pH 9.8) was added to the reaction mixture. Absorbance at 405 nm was measured. Activity of β-hexosaminidase was indicated as a percentage of the total enzyme activity detected in the culture medium and cell lysate generated using 1% Triton X-100 as described earlier [22].

Assay of ADP-Ribosylation
Five micrograms of beta-actin were incubated at 37 • C for 60 min in a reaction buffer solution containing 25 mM Tris-HCl buffer (pH 7.6), 2.5 mM MgCl 2 , 0.5 mM DTT, 0.5 mM EDTA, protease inhibitor mixture, 5 µM biotin-labelled NAD + (Trevigen Inc., Gaithersburg, MD, USA) and 200 ng of Ia protein. The reaction was terminated by addition of 2× SDS-sample buffer and heating at 95 • C for 5 min. The samples were subjected to SDS-PAGE and blotted to a PVDF membrane. The biotin-labelled ADP-ribosytated actin was determined using peroxidase-conjugated streptavidin and a subsequent ECL kit [8].

Si RNA Transfection
siRNA negative controls and siRNAs for SMPD1 were purchased from Qiagen (Tokyo, Japan). MDCK cells (5 × 10 6 cells) were incubated with 500 pmol siRNA and then transfected using a Neon TM transfection system (Thermo Fisher Sci, Tokyo, Japan) [22]. After electroporation, cells were seeded to 24-well plates and incubated at 37 • C. Experiments were performed at 48 h after siRNA transfection [22].

Immunocytochemistry
MDCK cells were seeded on a poly-D-lysine-coated glass bottom dish (MatTek, Ashland, MA, USA) and cultured in a 5% CO 2 at 37 • C atmosphere for 24 h in FCS-DMEM. To examine the cellular uptake of Ib, cells were treated with Ib at 37 • C for the given incubation periods in FCS-DMEM. Cells were washed three times with phosphate-buffered saline (PBS) and treated with 2% paraformaldehyde at room temperature. The bottom dishes were incubated for 15 min in 20 mM NH 4 Cl, and incubated in PBS supplemented with 0.1% Triton X-100 at room temperature for 15 min. Cells were blocked in PBS supplemented with 4% BSA at room temperature for 2 h. For Ib staining, cells were incubated with primary antibody (rabbit anti-Ib antibody) at 4 • C for 12 h. The cells were then rinsed with PBS containing 0.02% Triton X-100, and incubated with secondary antibody (Alexa Fluor 488-conjugated anti-rabbit IgG or Alexa Fluor 568-conjugated anti-rabbit IgG or anti-mouse IgG-FITC) at room temperature for 1 h. Nuclei were labeled with DNA dye DAPI (0.4 µg/mL). For ceramide staining, cells were treated with primary antibody (mouse anti-ceramide IgM antibody) for 2 h, and with secondary antibody (anti-mouse IgM Alexa Fluor594) for 1 h. Cells were analyzed using a Nikon A1 laser scanning confocal microscope (Tokyo, Japan) as described earlier [12,22].

Statistical Analysis
Statistical analyses were carried out by EZR software (Easy R, Saitama Medical Center, Jichi Medical Univeristy, Japan) [33]. Differences between the two groups were assessed by a two-tailed Student's t-test. A one-way analysis of variance (ANOVA) followed by the Tukey test were utilized to check for differences among three or more groups. P values < 0.01 were considered statistically significant.