Lysosomal ceramide generated by acid sphingomyelinase triggers cytosolic cathepsin B-mediated degradation of X-linked inhibitor of apoptosis protein in natural killer/T lymphoma cell apoptosis

We previously reported that IL-2 deprivation induced acid sphingomyelinase-mediated (ASM-mediated) ceramide elevation and apoptosis in an NK/T lymphoma cell line KHYG-1. However, the molecular mechanism of ASM–ceramide-mediated apoptosis during IL-2 deprivation is poorly understood. Here, we showed that IL-2 deprivation induces caspase-dependent apoptosis characterized by phosphatidylserine externalization, caspase-8, -9, and -3 cleavage, and degradation of X-linked inhibitor of apoptosis protein (XIAP). IL-2 re-supplementation rescued apoptosis via inhibition of XIAP degradation without affecting caspase cleavage. However, IL-2 deprivation induced ceramide elevation via ASM in lysosomes and activated lysosomal cathepsin B (CTSB) but not cathepsin D. A CTSB inhibitor CA-074 Me and knockdown of CTSB inhibited ceramide-mediated XIAP degradation and apoptosis. Inhibition of ceramide accumulation in lysosomes using an ASM inhibitor, desipramine, decreased cytosolic activation of CTSB by inhibiting its transfer into cytosol from the lysosome. Knockdown of ASM also inhibited XIAP degradation and apoptosis. Furthermore, cell permeable N-acetyl sphingosine (C2-ceramide), which increases mainly endogenous d18:1/16:0 and d18:1/24:1 ceramide-like IL-2 deprivation, induced caspase-dependent apoptosis with XIAP degradation through CTSB. These findings suggest that lysosomal ceramide produced by ASM mediates XIAP degradation by activation of cytosolic CTSB and caspase-dependent apoptosis. The ASM–ceramide–CTSB signaling axis is a novel pathway of ceramide-mediated apoptosis in IL-2-deprived NK/T lymphoma cells.

M Taniguchi 1 , H Ogiso 2 , T Takeuchi 2 , K Kitatani 3 , H Umehara 4 and T Okazaki* , 2 We previously reported that IL-2 deprivation induced acid sphingomyelinase-mediated (ASM-mediated) ceramide elevation and apoptosis in an NK/T lymphoma cell line KHYG-1. However, the molecular mechanism of ASM-ceramide-mediated apoptosis during IL-2 deprivation is poorly understood. Here, we showed that IL-2 deprivation induces caspase-dependent apoptosis characterized by phosphatidylserine externalization, caspase-8, -9, and -3 cleavage, and degradation of X-linked inhibitor of apoptosis protein (XIAP). IL-2 re-supplementation rescued apoptosis via inhibition of XIAP degradation without affecting caspase cleavage. However, IL-2 deprivation induced ceramide elevation via ASM in lysosomes and activated lysosomal cathepsin B (CTSB) but not cathepsin D. A CTSB inhibitor CA-074 Me and knockdown of CTSB inhibited ceramide-mediated XIAP degradation and apoptosis. Inhibition of ceramide accumulation in lysosomes using an ASM inhibitor, desipramine, decreased cytosolic activation of CTSB by inhibiting its transfer into cytosol from the lysosome. Knockdown of ASM also inhibited XIAP degradation and apoptosis. Furthermore, cell permeable N-acetyl sphingosine (C 2 -ceramide), which increases mainly endogenous d18:1/16:0 and d18:1/24:1 ceramide-like IL-2 deprivation, induced caspase-dependent apoptosis with XIAP degradation through CTSB. These findings suggest that lysosomal ceramide produced by ASM mediates XIAP degradation by activation of cytosolic CTSB and caspase-dependent apoptosis. The ASM-ceramide-CTSB signaling axis is a novel pathway of ceramide-mediated apoptosis in IL-2-deprived NK/T lymphoma cells. KHYG-1 natural killer T (NK/T) cells established from a patient with aggressive NK/T cell lymphoma displayed morphology of a large granular lymphocyte with a large nucleus, rough chromatin, and bulk basophilic cytoplasm. 1 Suck et al. 2 reported that KHYG-1 cells had in vitro cytotoxicity against lymphoma cell lines such as HL-60 and induced apoptosis of tumor cells via the granzyme M/perforin pathway. Thus KHYG-1 cells have a cytotoxic ability similar to NK cells against malignant cells. However, because interleukin-2 (IL-2) is essential for clonal expansion of KHYG-1 cells, this cell line is a useful model to investigate the mechanism by which NK/T lymphoma cells undergo programmed cell death. Previously, we reported that IL-2 deprivation (IL-2(− )) promoted ceramide generation due to the activation of acid sphingomyelinase (ASM), resulting in apoptosis of KHYG-1 cells. 3 The mechanisms of ASM-ceramide-mediated apoptotic signal in IL-2 deprived NK/T lymphoma cells has not been clarified.
Apoptosis uses two major signaling pathways to induce programmed cell death; extrinsic and intrinsic pathways. 4 The extrinsic pathway is mediated by extracellular death ligands (Fas ligand, tumor necrosis factor-α (TNF-α), TNF-related apoptosis-inducing ligand (TRAIL), or CD95) and activates death-inducing signaling complex-containing caspase-8 through those receptors. The intrinsic pathway mediates the disruption of mitochondria and induced the formation of an apoptosome complex composed of cytochrome C secreted from mitochondria, apoptotic protease-activating factor 1 (Apaf-1), and caspase-9. 4 Both extrinsic and intrinsic pathways lead to the activation of caspase-3/-7, which trigger various apoptotic phenomena such as phosphatidylserine (PS) externalization or DNA fragmentation. In addition to these mechanisms that induce apoptosis, it is important to clarify the role of lysosomal proteases in the regulation of antiapoptotic proteins such as Bcl-2 family members and pro-apoptotic proteins such as caspases.
Cathepsins (CTSs) in lysosomes consist of cysteine protease, aspartic protease cathepsin D (CTSD) and serine protease cathepsin B (CTSB). Upon extracellular and intracellular stressors, CTSs are released into the cytosol and are activated enzymatically by optimal pH conditions. 5 Apoptotic signals mainly inhibit antiapoptotic molecules such as Bcl-2 to activate pro-apoptotic Bax/Bak molecules by their degradation through CTSs. 6,7 Currently, the role of CTSD and CTSB in apoptosis is controversial. The activation of CTSB through transforming growth factor-β signaling was reported to increase the proliferation of melanoma cells and short hairpin RNA (shRNA) of CTSB had an apoptotic effect mediated through the degradation of X-linked inhibitor of apoptosis protein (XIAP) in invasive meningioma cells, suggesting the positive effect of CTSB in cell proliferation. 8,9 In contrast, it was reported that CTSB induced apoptosis by activating caspase-3 and -9 in dengue virus-infected HepG2 hepatocytes. 10 Caspase-3, -7, and -9 are inhibited by XIAP, an IAP family member 11,12 that directly binds to and inactivates caspase-3 or caspase-9 to inhibit their degradation, resulting in suppression of apoptosis. [13][14][15][16] Downregulation of XIAP increases the sensitivity of cancer cells to apoptotic stimuli, such as TRAIL or hypoxia. 17,18 In hematological malignancies, anti-CD33 antibodies induced apoptosis by decreasing XIAP in acute myeloid lymphoma (AML), 19 and AML patients with overexpression of XIAP showed unfavorable responses to induction chemotherapy. 20 Anticancer drug-resistant lymphoma cells also had overexpression of XIAP through the NF-κBdependent MEK/MAPK pathway. 21 In general, cytosolic proteins such as XIAP are regulated at the transcriptional level or by enzymatic degradation via proteases. However, the molecular mechanism by which the protein levels of XIAP are regulated in ceramide-induced NK/T cell apoptosis has not been investigated.
Ceramide is at the center of sphingolipid metabolism and acts as a substrate of other sphingolipids, such as sphingomyelin, sphingosine-1-phosphate (S1P), and glycosphingolipids. 22,23 Ceramide is also a lipid mediator that induces programmed cell death, differentiation, senescence, cell cycle arrest, and autophagy. [22][23][24] There are three pathways in intracellular ceramide generation: (i) de novo synthesis from L-serine and a palmitoyl-coenzyme A, (ii) the sphingomyelin cycle consisting of sphingomyelin synthase (SMS) and sphingomyelinase, and (iii) the salvage pathway where ceramide synthases utilize sphingosine degraded from SM, glycolipids, and S1P as a substrate of ceramide. 22 These pathways are mutually involved in the generation of ceramide induced by various stimuli. 23 Especially, ASM-generated ceramide has been well investigated in numerous types of cell death. Stimulation of TRAIL or CD95 ligands induces rapid ASM activation and formation of ceramide-enriched platforms in the plasma membrane. 25,26 ASM-generated ceramide provides a place to form clusters between ligands and their transmembrane receptors, which transduce an efficient death signal to the intracellular compartment. [25][26][27] However, stimulation of TNF-α or gemcitabine generated ceramide in lysosomes through ASM activation. Lysosomal ceramide was reported to trigger the CTSD-mediated apoptotic pathway. [28][29][30] Recently, arsenic trioxide induced the degradation of XIAP through the ubiquitin-proteasome pathway and treatment with valproic acid increased CTSB-induced apoptosis of chronic lymphoid lymphoma cells. 31,32 However, how ASM-generated lysosomal ceramide is related to the cathepsin family, including CTSB and CTSD and XIAP in NK/T lymphoma cell apoptosis, is poorly understood. In this study, we demonstrated that IL-2( − ) activated the ASM-ceramide pathway in lysosomes and that generation of d18:1/16:0 and d18:1/24:1 lysosomal ceramides caused the release of CTSB, but not CTSD, into the cytosol. CTSB-mediated degradation of XIAP subsequently induced activation of caspase-3 and its nuclear entry to execute apoptosis. These results suggest the ASM/ceramide/CTSB axis is a novel pathway for the degradation of XIAP in IL-2deprived NK/T lymphoma cell apoptosis.
To elucidate whether CTSB activation was downstream of increased lysosomal ceramide, we observed ceramide accumulation in IL-2( − ) conditions with or without CA-074 Me treatment. As shown in Figure 4c, CA-074 Me did not prevent lysosomal ceramide accumulation. However, CA-074 Me treatment inhibited apoptosis (Figure 4d). To confirm the effects of CA-074 Me, we performed knockdown of CTSB. KHYG-1 cells were quite difficult to transduce genes transiently. Thus we established CTSB knockdown cell lines by using CTSB-specific shRNA lentivirus (Figure 4e). CTSB knockdown (shCTSB) blocked XIAP degradation and PARP cleavage induced by IL-2( − ) (Figure 4f). In addition, cell survival was also increased in shCTSB cells compared with control cells (shSCR) (Figure 4f). These data suggested that accumulation of lysosomal ceramide during IL-2( − ) activated CTSB but not CTSD and the subsequent apoptosis. ASM inhibitor suppressed ceramide accumulation and CTSB-mediated apoptosis. The above data showed that IL-2( − ) activated ASM, which induced lysosomal ceramide accumulation. We tested the effects of an ASM inhibitor on IL-2( − )-mediated apoptosis. Preincubation with ASM inhibitor desipramine blocked ASM activation but not NSM during IL-2( − ) (Figures 5a and b). Next, we observed IL-2 ( − )-induced lysosomal ceramide accumulation in the presence or absence of desipramine by immunocytochemistry. As shown in Figure 5c, ceramide accumulation induced by IL-2( − ) was suppressed by desipramine treatment. Moreover, desipramine also blocked CTSB activation, which is downstream of lysosomal ceramide (Figure 5d).
CTSB or CTSD were released from lysosomes to the cytosol. [36][37][38] Lysosomal ceramide activates CTSD through its release from lysosome membranes to the cytosol. 28 To examine CTSB release by IL-2( − ) and the effect of desipramine on CTSB release, we separated cytosol and heavy membrane fractions, which contain lysosome membranes. Interestingly, IL-2( − ) induced CTSB release from lysosome to cytosol, and its release was inhibited by desipramine treatment (Figure 5e). According to inhibition of CTSB, XIAP degradation, which elicits cell survival, was also suppressed by desipramine treatment (Figure 5g).
To confirm the effect of desipramine treatment, we established ASM-knockdown cells by using ASM-specific shRNA lentivirus. As shown in Figures 6a and b, ASM protein and activity were significantly reduced by ASM-knockdown cells (shASM) compared with control cells (shSCR). According to knockdown of ASM, XIAP degradation and PARP cleavage by IL-2( − ) were blocked in shASM cells (Figure 6c). In addition, cell survival was also improved by ASM knockdown (Figure 6d). These results suggested that IL-2 ( − )-activated ASM increased lysosomal ceramide, which activates CTSB via its release from the lysosome to cytosol for XIAP degradation.
As shown in Figure 7d, C 2 -ceramide induced decrease of XIAP protein. Furthermore, CA-074 Me inhibited the degradation of XIAP induced by C 2 -ceramide ( Figure 7d). These results suggested that exogenous ceramide also accelerates release and activation of CTSB with endogenous ceramide accumulation and mediates CTSB-dependent XIAP degradation.
Exogenous ceramide induced XIAP degradation and caspase-dependent apoptosis. Finally, we examined the effect of exogenous ceramide on XIAP degradation and caspase-dependent apoptosis. C 2 -ceramide treatment reduced cytosolic XIAP protein by western blotting analysis and immunocytochemistry (Figures 8a and b). Active caspase-3 was detected by C 2 -ceramide treatment as demonstrated by XIAP degradation (Figure 8b). Moreover, C 2 -ceramide induced in vivo caspase-3 activation and apoptosis (Figures 8c and d). These data suggested that C 2 -ceramide mimicked IL-2( − )-induced apoptosis by XIAP degradation and caspase-3 activation.

Discussion
KHYG-1 NK/T lymphoma cells require IL-2 for their proliferation and survival and undergo apoptosis by IL-2( − ) even in the presence of serum. Thus it is important to understand the mechanism of IL-2( − )-induced apoptosis to develop novel treatments for drug-resistant NK/T cell lymphoma. We first showed that IL-2( − ) induced caspase-dependent apoptosis, characterized by PS externalization and cleavage   of caspase-8, -9, and -3, and the addition of IL-2 rescued cell death. Interestingly, although proteinase activity of caspase-3 was inhibited by IL-2 rescue, the 17-19-kD cleaved form of caspase-3 was observed. Furthermore, cleaved caspase-3 p17 did not traffic into the nucleus after IL-2 rescue while IL-2( − )-induced active caspase-3 accumulated in the nucleus to mediate the degradation of PARP. The precise mechanism of nuclear transport of caspase-3 has been clarified, but the translocation of caspase-3 p17 with more processing form p12 into the nucleus might be involved in Fas-induced apoptosis. Previous studies have demonstrated that p3-recognition site of caspase-3 is important for its nuclear entry. 42,43 Indeed, our data showed that more short form of cleaved caspase-3 (12 kD) was slightly but reliably detected in only IL-2( − ), suggesting that caspase-3 was not cleaved completely in IL-2 rescue. In addition, caspase-3 p17 remained in the cytosol and did not move into the nucleus after rescue of apoptosis by IL-2 supplementation. These results suggested that the mechanism that inhibits the entry of caspase-3 p17 into the nucleus exists in the cytosol.
It is well known that ASM localizes in lysosomes. During IL-2( − ) conditions, ASM-induced increase in ceramide was detected in lysosomes using anti-ceramide antibodies that colocalized with a lysosome marker Lamp1. After IL-2 rescue, the ceramide content in lysosomes decreased to the IL-2(+) control level. The molecular species of IL-2( − )-increased ceramide are of d18:1/16:0 and d18:1/24:1 types. We can detect that the increase of molecular species of ceramide by a diverse of stressors appears to be similar in pattern regardless of the mechanisms of PCD (data not shown). The localization of ceramide generated by the different pathways may be more critical for the regulation of cell death by ceramide signal, but at present evidence is ambiguous.
We detected in vitro activities of caspases in lysates of IL-2 ( − ) and IL-2 rescued cells (data not shown). However, the in vivo activity of caspase-3 was significantly suppressed by IL-2 rescue. Because XIAP can directly bind to caspase-3 to inhibit cleaved activation, 14 we focused on the regulation of XIAP protein, a member of the IAP family, by ASM-induced lysosomal ceramide. 14,16 A decrease of XIAP by IL-2( − ) was restored by IL-2 rescue. This reduction was due to protein degradation but not the suppression of mRNA expression in XIAP (data not shown). It was reported that XIAP expression was controlled by activated NF-κB 44 and that the PI3-K/Akt pathway, upstream of NF-κB, regulated XIAP. 45 We also detected NF-κB activation and enhanced XIAP mRNA expression by IL-2 rescue (data not shown). This suggests that the increase of XIAP in cell proliferation is regulated transcriptionally, whereas the decrease of XIAP induced by IL-2( − ) is due to degradation at the protein level.
Why could not XIAP inhibit caspase cleavage? Recently, some reports showed that cleaved caspase has other effects such as inflammatory activation without apoptosis. Kavanagh et al. 46 demonstrated that cytoplasmic cleaved caspase-3 induces pro-inflammatory activation through PKCδ but not apoptosis in microglia. Pro-inflammatory response activates NF-kB pathway. Moreover, Cheng et al. 47 showed that ionizing radiation activates PKCδ through caspase-3/7 and leads production of growth factors via Akt pathway in pancreatic cancer cells. Indeed, rescue activated NF-κB pathway and cell proliferation (data not shown). Thus cytoplasmic cleaved caspase might be related in cell proliferation but not apoptosis in rescue condition.
Recently, it was reported that the ceramide analogue LCL85 enhanced Fas-induced apoptosis by inhibiting XIAP. 48 We here demonstrated that regulation of ceramide in lysosomes has a role in the degradation of XIAP. What is the mechanism by which XIAP is degraded by lysosomal ceramide? Mechanisms that decrease XIAP protein include degradation by ubiquitination and proteases and/or transcriptional regulation. 49 Ceramide mediates apoptosis, and many molecules, including ceramide-activated protein phosphatases and protein kinase C ζ, are involved in its signaling. 50,51 Especially, lysosomal aspartic protease CTSD is reported to be activated by ASM-generated ceramide to induce apoptotic cell death in response to TNF signaling. 28,35 Lysosomal ceramide generated by ASM bound to CTSD mediated its translocation to the cytosol, following activation of the mitochondrial apoptotic pathway. 28 Interestingly, we found that an inhibitor of cysteine protease CTSB (CA-074 Me), but not CTSD (pepstatin A), inhibited XIAP degradation, and activation of CTSB was detected after IL-2( − ) treatment. Inhibitors of CTSB suppressed IL-2( − )-mediated apoptotic cell death, suggesting a role for CTSB, but not CTSD, in apoptosis. However, previous studies indicated a role for CTSB in tumor progression. 52 The increased expression of CTSB was related to the invasion and proliferation of melanoma cells, 9 and CTSB decreased dramatically after serum-deprivation-induced apoptosis, whereas CTSD increased, suggesting a balance between CTSB and CTSD. 53 In the current study, we showed that lysosomal ceramide-activated CTSB induced caspase activation through XIAP degradation because an inhibitor of CTSB CA-074 Me was effective in blocking the apoptotic pathway. However, we could not detect the cleavage site of CTSB in XIAP protein or direct digestion of XIAP using cell lysates and recombinant CTSB (data not shown). Bien et al. 54 showed that doxorubicin activated CTSB and CTSB-mediated caspase-dependent apoptosis in HeLa cells. In addition, the downregulation of XIAP was also induced by doxorubicin and was inhibited by CTSB siRNA or CA-074 Me treatment. These data support our finding that CTSB regulates XIAP degradation.
CA-074 Me did not inhibit the increase of ceramide in lysosomes during IL-2( − ), suggesting that ASM-induced ceramide generation is upstream of CTSB activation. Although CTSB has not been reported to bind to ceramide, previous studies showed that CTSB release is induced by alterations of lysosomal membrane permeabilization (LMP) in response to various stimuli, which activates caspasemediated apoptosis. 55 Thus lysosomal ceramide might induce LMP to release CTSB to the cytosol, but the precise mechanism of CTSB activation by lysosomal ceramide is unknown.
ASM activation and ceramide generation have been reported in various signaling pathways triggered by numerous stimuli, 25,56 chemotherapeutic drugs, 30,57 irradiation, 58,59 or pathogens. 27,60 In the present study, we showed IL-2( − ) activated ASM, but not NSM, and induced ceramide accumulation in lysosomes, suggesting that lysosomal ceramide is critical for IL-2( − )-induced apoptosis. Indeed, an inhibitor of ASM, desipramine, inhibited IL-2( − )-induced apoptosis by suppressing ceramide generation in lysosomes. Desipramine also inhibited the nuclear activation of caspase-3 through the degradation of XIAP by CTSB activation.
Finally, we investigated whether exogenous C 2 -ceramide mimics IL-2( − )-mediated CTSB activation, XIAP degradation, and caspase-dependent NK/T lymphoma cell apoptosis. C 2ceramide increased apoptosis by generating physiological d18:1/16:0 and d18:1/24:1 ceramides similar to IL-2 deprivation. The increase of ceramides induced activation of CTSB and subsequent XIAP degradation. Huang et al. 61 described that exogenous treatment with C 2 -ceramide triggered lysosomal pathways in T cell hybridomas and A549 lung adenocarcinoma cells and that C 2 -ceramide might have no direct effect on lysosomal function because the inhibition of ASM did not affect LMP or apoptosis. Recently, sphingosine, a metabolite of ceramide via ceramidase, was reported to mediate LMP and relocation of CTSB in TNF-α-treated hepatoma cells. 55 Thus sphingosine generated by acid ceramidase from ASM-generated ceramide might activate CTSB during IL-2( − ). In fact, IL-2( − ) also increased sphingosine (18 : 1) (from 3.35 ±0.85 pmol/1 × 10 6 cells in IL-2(+) to 4.97 ± 0.27 pmol/1 × 10 6 cells in IL-2(− )) (data not shown). However, amount of ceramide (459.3 ±21.4 pmol/ 1 × 10 6 cells in IL-2(− )) was more than 100-fold of sphingosine. In addition, we investigated the effects of inhibitors of enzymes related in ceramide metabolism on C 2 -ceramideinduced cell death (Supplementary Figure S1). At first, the treatment of desipramine with C 2 -ceramide had no effect on cell survival. This result suggested that ASM is not involved in cellular metabolism of C 2 -ceramide in our case. Next, we used inhibitors of ceramide synthase and acid ceramidase, which inhibit the metabolism of ceramide from sphingosine and of sphingosine from ceramide, respectively. Fumonisin B1, which is ceramide synthase inhibitor to block ceramide production from sphingosine, recovered cell survival compared with C 2 -ceramide only. Inversely, acid ceramidase inhibitor D-NMAPPD enhanced C 2 -ceramide-medaited cell death. Acid ceramidase produces sphingosine from ceramide in lysosome. These results demonstrated that ceramide but not sphingosine is related in C 2 -ceramide-induced apoptosis. In addition, some reports that acid ceramidase is implicated in cell survival of malignant cells supported our data that acid ceramidase has inhibitory effect on C 2 -ceramide-mediated apoptosis. [62][63][64] Thus we believe ASM-generated ceramide affected IL-2( − )-induced apoptosis.
In summary, our present work demonstrated a novel pathway related to lysosomal ceramide, CTSB, and XIAP in IL-2( − )-induced NK/T lymphoma cell apoptosis. Lysosomal ceramide generated by ASM induced the release of CTSB into the cytosol and the degradation of XIAP, resulting in the nuclear entry of active caspase-3 and subsequent apoptosis. The pathophysiological implication of this lysosomal ceramide/CTSB/XIAP axis in apoptotic cell death should be clarified in the future to develop a targeting therapy for NK/T lymphoma.
For establishments of ASM or CTSB knockdown cells, lentiviral particles of shRNA were obtained from Santa Cruz. Cells were infected with control shRNA (shSCR, sc-108080), ASM-specific shRNA (shASM, sc-41650-v), and CTSB-specific shRNA (shCTSB, sc-29238-v) lentiviral particles after the treatment with 1 μg/ml polybrene. Then cells were treated with 2 μg/ml puromycin for establishment of knockdown cell lines. Efficiency of knockdown was checked by western blotting analysis. Analysis of in vivo caspase and CTSB activities. In vivo caspase-3/7 activity was measured by FAM-fluorochrome inhibitor of caspases (FLICA) Caspase 3&7 Assay Kit (Immunochemistry Technologies, Bloomington, MN, USA) according to the manufacturer's information. Briefly, cells were incubated with the FAM-DEVDfluoromethylketone (FMK) reagent for 1 h at 37°C. After washing with PBS, cells were fixed with 0.1% (w/v) formaldehyde solution. In vivo CTSB activity was performed using Magic Red CTSB Activity Kit (Immunochemistry Technologies) according to the manufacturer's protocol. Cells were incubated with Magic Red CTSB Substrate (Immunochemistry Technologies) for 1 h at 37°C. The cells were pelleted after centrifugation, washed with PBS, and fixed with 0.1% formaldehyde. Fluorescence was measured by a fluorescent microplate reader Infinite 500 (Tecan Group Ltd., Männedorf, Switzerland), or at least 200 cells were counted and indicated as the percentage of positive cells per total cells. Images were obtained by fluorescent microscopy Leica DMRB or confocal microscopy Leica TSC SP2 (Leica Microsystems, Wetzlar, Germany).
Western blotting analysis. KHYG-1 cells were harvested, washed with PBS, and lysed in lysis buffer (10 mM Tris-HCl, pH 7.4, 10 mM KCl, 1.5 mM MgCl 2 , 1% (w/v) Triton X-100, 1 mM PMSF, 10 μg/ml leupeptin, and 10 μg/ml aprotinin). After incubation on ice for 20 min, debris was removed by centrifugation at 2000 × g for 10 min at 4°C. Supernatant was used as a loading sample. Proteins (30 μg) were subjected to SDS-poly-acrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA). Nonspecific binding was blocked by incubation of the membrane with PBS containing 0.1% (w/v) Tween-20 (PBS-T) and 5% (w/v) non-fat dried milk for 20 min at room temperature. Then membrane was incubated with primary antibodies overnight at 4°C and with secondary antibodies for 45 min at room temperature. Immunoreactive protein bands were visualized using an ECL-peroxidase detection system (Amersham Biosciences, Piscataway, NJ, USA) and LAS-4000 (Fujifilm, Tokyo, Japan).
Immunocytochemistry. For detection of active caspase-3 and XIAP, cells were treated by the indicated conditions, washed with ice-cold PBS, cytospun onto slides, and then fixed with 2% formaldehyde for 15 min at 4°C. Fixed cells were permeabilized with PBS containing 0.1% Triton X-100 for 5 min and incubated with PBS containing 2% (w/v) BSA for 30 min at room temperature. Cells were washed with PBS and then incubated with primary antibodies for 90 min at room temperature. After washing with PBS, AF488-or AF546-conjugated anti-IgG antibodies were incubated for 45 min. For determination of lysosomal ceramide, fixed cells were treated with anti-ceramide antibody and anti-Lamp1 antibody for 90 min at room temperature. Then AF488-conjugated anti-IgM and AF546conjugated anti-IgG antibodies were incubated for 45 min at room temperature. Nuclei were counterstained with DAPI.
Cell fractionation. Lysosome fraction was collected by using the Lysosome Enrichment Kit for Tissue and Cultured Cells according to the manufacturer's protocol (Thermo Scientific, 89839, Rockford, IL, USA). Cells (1 × 10 7 cells) were treated with IL-2(+) or IL-2(− ) for 24 h, and cell pellets were used for lysosome isolation. Whole cell lysate before lysosome collection and isolated lysosome fractions were then subjected to western blotting analysis to check the purity of the isolated lysosome and measurement of SM by LC-MS/MS. Anti-Lamp2 (lysosome), anti-pan-cadherin (plasma membrane), and anti-GAPDH (cytosol) antibodies were used as markers of each organelle.
CTSB and CTSD release. The release of CTSB and CTSD from the endolysosomal compartment was analyzed by heavy membrane fractionation. 67 Cells were lysed in homogenization buffer (10 mM Hepes-KOH, pH 7.4, 1 mM EDTA, 0.25 M sucrose, 1 mM PMSF, inhibitor cocktail (Roche)), incubated on ice for 40 min, and homogenized by 27-G syringes. Homogenates were centrifuged at 300 × g for 5 min, and unbroken cells were removed. Then lysate was centrifuged at 20 000 × g for 30 min, and supernatants contained the cytoplasmic fraction. Pellets were resolved in homogenization buffer containing 1% Triton X-100 and used for heavy membrane fraction, including lysosomes. Proteins (40 μg) were used for western blotting to detect the release of CTSB and CTSD from the heavy membrane fraction to the cytosol fraction. Anti-Lamp2 antibody was used as a lysosomal marker.
Statistical analysis. Comparisons between two groups were carried out using the unpaired Student's t-test.