- Email: [email protected]
Mast cell apoptosis induced by siramesine, a sigma-2 receptor agonist
Biochemical Pharmacology 84 (2012) 1671–1680
Contents lists available at SciVerse ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Mast cell apoptosis induced by siramesine, a sigma-2 receptor agonist Jane Spirkoski 1, Fabio R. Melo 1, Mirjana Grujic, Gabriela Calounova, Anders Lundequist, Sara Wernersson, Gunnar Pejler * Swedish University of Agricultural Sciences, Dept. of Anatomy, Physiology and Biochemistry, BMC Box 575, 75123, Uppsala, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 20 August 2012 Accepted 26 September 2012 Available online 9 October 2012
Mast cells (MCs) are well known for their detrimental effects in the context of allergic disorders. Strategies that limit MC function can therefore have a therapeutic value. Previous studies have shown that siramesine, a sigma-2 receptor agonist originally developed as an anti-depressant, can induce cell death in transformed cells through a mechanism involving lysosomal destabilization. Since MCs are remarkably rich in lysosome-like secretory granules we reasoned that MCs might be sensitive to siramesine. Here we show that murine and human MCs are highly sensitive to siramesine. Cell death was accompanied by secretory granule permeabilization, as shown by reduced acridine orange staining and leakage of granule proteases into the cytosol. Wild type siramesine-treated MCs underwent cell death with typical signs of apoptosis but MCs lacking serglycin, a proteoglycan crucial for promoting the storage of proteases within MC secretory granules, died predominantly by necrosis. A dissection of the underlying mechanism suggested that the necrotic phenotype of serglycin / cells was linked to defective Poly(ADP-ribose) polymerase-1 degradation. In vivo, siramesine treatment of mice caused a depletion of the MC populations of the peritoneum and skin. The present study shows for the first time that MCs are highly sensitive to apoptosis induced by siramesine and introduces the possibility of using siramesine as a therapeutic agent for treatment of MC-dependent disease. ß 2012 Elsevier Inc. All rights reserved.
Keywords: Mast cells Apoptosis Siramesine Serglycin Secretory granules
1. Introduction Mast cells (MCs) are major effector cells in allergic conditions such as anaphylaxis and allergic asthma [1,2], but recent studies have strongly implicated MCs also in other types of disease [3–6]. A hallmark feature of MCs is their high content of densely packed secretory granules. These are filled with large quantities of various preformed compounds, including bioactive monoamines, cytokines, serglycin proteoglycan, lysosomal enzymes, as well as MCspecific proteases of chymase, tryptase and carboxypeptidase A3 (CPA3) type [7–10]. MC degranulation can be triggered by different means, including antigen-mediated crosslinking of immunoglobulin (Ig) E molecules bound to the high affinity IgE receptor [11]. Upon
Abbreviations: MC, mast cell; mMCP, mouse mast cell protease; BMMC, bone marrow-derived mast cell; PCMC, peritoneal cell-derived mast cell; PARP-1, Poly(ADP-ribose) polymerase-1; AO, acridine orange; WT, wild type; AnnV, Annexin V; CPA3, carboxypeptidase A3; Ig, immunoglobulin; RT-PCR, reverse transcription polymerase chain reaction; HMC-1, Human Mast Cell Line-1; LAD2, Laboratory of Allergic Diseases-2. * Corresponding author. Tel.: +46 18 4714090. E-mail address: [email protected] (G. Pejler). 1 Equal contribution. 0006-2952/$ – see front matter ß 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bcp.2012.09.028
degranulation, the preformed MC mediators are released into the surrounding tissue. Furthermore, MC activation will trigger de novo synthesis of various additional pro-inflammatory compounds, encompassing eicosanoids and a panel of cytokines and chemokines [12,13]. Given the wide array of detrimental MC actions, strategies that limit MC function may be utilized for therapeutic purposes. Such strategies include various stabilizers that prevent MC degranulation, as well as selective targeting of different products secreted by MCs [14–16]. However, the impact of MCs on a given disorder may be multifaceted, i.e. being mediated by multiple products secreted by MCs. Hence, an efficient blockade of detrimental MC effects may require global inhibition of activities mediated by MCs. To accomplish this, an emerging strategy is to eradicate MC populations altogether by selectively inducing MC apoptosis (reviewed in [17]). Siramesine is a sigma-2 receptor agonist that was originally intended as an anti-depressant [18]. The drug has been proven safe for use in humans but was abandoned as anti-depressant because of low efficacy. In addition to its proposed anti-depressant properties, it has been shown that siramesine has pro-apoptotic activity on various transformed cell types [19,20], and that the cytotoxic activity of siramesine on transformed cells involves lysosomal destabilization [19]. Lysosomes and secretory granules
1672
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
share many features, including acidic pH, membrane constituents and content of lysosomal hydrolases [9], and we therefore reasoned that agents (such as siramesine) causing permeabilization of lysosomes might have similar actions on secretory granules. MC granules have an extraordinary high content of fully active proteases [10], and damage to the MC granules will thus lead to the entry of vast amounts of proteolytic activity into the cytosol. Potentially, this can lead to apoptosis through proteolytic activation of cytosolic pro-apoptotic compounds and/or degradation of anti-apoptotic compounds [21]. Based on these notions, we decided to investigate whether siramesine had granule-permeabilizing activity on MCs, and whether such activity could cause cell death. Indeed, we here show that siramesine efficiently and selectively induces MC apoptosis. 2. Materials and methods
skin was dissected, paraffin-embedded and sectioned. The cytospin slides were stained with May Gru¨nwald/Giemsa, whereas the tissue sections were stained with toluidine blue and hematoxylin/eosin. Numbers of individual cell types were calculated from the total number of peritoneal lavage cells and from the percentage of individual cell types on cytospin slides. In skin specimens, the numbers of MCs in uniformly sized sections were enumerated based on staining with toluidine blue. 2.5. Cell viability measurement 100 ml from cell suspensions containing the various populations of MCs were collected and transferred to individual wells of a 96-well plate. 20 ml of CellTiter-Blue1 Reagent (Promega-Invitrogen, Carlsbad, CA) was added to the wells, followed by incubation for 4–5 h at 37 8C. Fluorescence was read using a microplate reader (Infinite M200 – TECAN) at 560 nm for excitation and 590 nm for emission.
2.1. Reagents 2.6. Staining of acidic compartments Siramesine was generously given by Lundbeck A/S (Copenhagen, Denmark). Z-DEVD-FMK, Z-VAD-FMK, E-64d and Pepstatin A were purchased from Sigma–Aldrich (Steinheim, Germany). Antib actin goat polyclonal antibodies were from Santa Cruz Biotechnology (sc1616, Santa Cruz, CA); anti-PARP rabbit polyclonal was from Abcam (ab6079, Cambridge, UK). Fluorescently labeled donkey anti-rabbit and -goat Ig were purchased from Li-cor (Bad Homburg, Germany). Acridine orange (AO) was from Sigma– Aldrich (Steinheim, Germany).
Cells (5 105 cells/ml) were incubated for 30 min with 5 mg/ml acridine orange (AO; Sigma–Aldrich). The cells were pelleted by centrifugation (300 g; 8 min), and washed three times with PBS and then resuspended in 500 ml PBS containing BSA (100 mg/ml). 100 ml from the cell suspensions was transferred to a 96-well plate. The fluorescence intensity was measured using a microplate reader (Infinite M200 – TECAN); 488 nm for excitation and 650 nm for emission.
2.2. Mice
2.7. Flow cytometry and Western blot analysis
Wild-type (WT) and serglycin / [22] mice were on C57BL/6J genetic background. All animal experiments were approved by the local ethical committee (Uppsala, Sweden).
Flow cytometry [26] and Western blot analysis [23] was performed as described. 2.8. DNA degradation
2.3. Cell culture Bone Marrow Derived Mast Cells (BMMCs) [23] and peritoneal cell-derived MCs (PCMCs) [24] were isolated and cultured as previously described. Human Mast Cell Line-1 (HMC-1, specifically HMC-1.2; kind gift from the Mayo Foundation for Medical Education and Research) and the cell line denoted Laboratory of Allergic Disease 2 (LAD2; gift from A. Kirshenbaum and Dean D. Metcalfe, National Institutes of Health) were cultured as described [25].
DNA was isolated as described [8]. Samples were separated using 1% agarose gel electrophoresis containing 0.1 ml/ml Gel Red Nucleic Acid Stain (Biotium, Hayward, CA) and visualized under UV light transillumination. 2.9. Real time reverse transcription polymerase chain reaction (RTPCR) Real time RT-PCR primers and conditions were as described [23].
2.4. Effects of siramesine 2.10. Statistical calculations For in vitro experiments, a stock solution of 5 mM siramesine in Polyethylene glycol (PEG) 400 was prepared. A 250 mM working solution was prepared by diluting the stock solution with 50% dimethyl sulfoxide (DMSO). Control samples were treated with vehicle containing the same concentration of PEG and DMSO as in samples treated with siramesine. Treatment of MCs with vehicle alone did not induce cell death (data not shown). Inhibitors were used at the following final concentrations: Z-DEVD-FMK (20 mM); Z-VAD-FMK (20 mM); E-64d (20 mM; inhibitor of cysteine proteases); Pepstatin A (50 mM; inhibitor of aspartic acid proteases). For in vivo experiments, siramesine was diluted in 10% HPbetacyclodextrin. WT mice were either non-treated or injected i.p. with 100 ml vehicle only (10% HPbeta-cyclodextrin) or with siramesine (3 and 20 mg/kg/day) for 3 consecutive days. Mice were then sacrificed followed by peritoneal lavage, counting of total cells in the lavage fluid and preparation of cytospin slides. Tissue from ear
The raw data were exported to the Microsoft Office Excel 2007 software program, where most of the calculations were done. Preparation of graphs, calculation of dose dependency and calculation the significance [One-Way ANOVA (Figs. 4–8) or unpaired t-test (Fig. 3)] was performed using GraphPad Prism 4 software (GraphPad Software, Inc, San Diego, CA). Results shown are from single experiments, representative of at least three individual experiments. 3. Results 3.1. Siramesine causes apoptosis of murine and human MCs Previous studies have shown that siramesine has antitumorigenic properties and, moreover, that the cytotoxic effect of siramesine was higher on cells that had undergone neoplastic
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1673
transformation than on untransformed cells [19]. Based on the notion that siramesine acts through lysosomal destabilization [19] and that MCs are exceptionally rich in lysosome-like secretory granules we hypothesized that siramesine also could be cytotoxic for MCs. Indeed, as shown in Fig. 1A and B, siramesine showed dose-dependent, potent cytotoxic activity on murine bone marrow-derived MCs (BMMCs) as determined by cell viability measurements. A time course experiment showed that cell death was achieved within 24 h (Fig. 1C). Notably, BMMCs are nontransformed primary cells, and these data thus suggest that the cytotoxic activity of siramesine is not limited to transformed cell types.
Fig. 2. Siramesine induces secretory granule permeabilization. (A) BMMCs (WT or serglycin / ) and LAD2 cells were incubated with 11 mM siramesine. At the time points indicated, cells were stained with AO. Results are expressed as % fluorescence in comparison with non-treated control cells. Data in (A) are expressed as mean values S.E.M. (n = 3). (B) BMMCs (WT or serglycin / ; 106 cells) were incubated with siramesine (11 mM) for the time periods indicated. Cytosolic extracts were then prepared and subjected to Western blot analysis for presence of granule proteases (mMCP-6 and CPA3). Note the low levels of mMCP-6 and CPA3 in cytosolic extracts of control samples (0 h). Note also that mMCP-6 and CPA3 as expected were undetectable in serglycin / cells, the latter in agreement with the strong dependence of mMCP-6 and CPA3 on serglycin for storage.
Next, we determined whether siramesine is cytotoxic for MC types in addition to BMMCs. In order to assess if siramesine is apoptotic also for terminally differentiated MCs we developed peritoneal cell-derived MCs (PCMCs) [24] and subjected these to siramesine. Indeed, siramesine was highly cytotoxic for PCMCs. In
Fig. 1. Siramesine is cytotoxic for MCs. (A) BMMCs (WT or serglycin / ) and LAD2 cells were incubated with different concentrations of siramesine as indicated. After 24 h, cell viability was measured. Results are expressed as % viable cells in comparison with non-treated control cells. (B) IC50 values for cytotoxicity of siramesine on various MC types. Data are expressed as mean values S.E.M. (C) Time course of siramesine-induced cell death. BMMCs (WT or serglycin / ) and LAD2 cells were incubated with 11 mM siramesine for the time periods indicated, followed by measurements of residual cell viability as compared with non-treated control cells (n = 3).
Fig. 3. Siramesine-induced cell death in BMMCs is caspase-independent and cysteine cathepsin-dependent. BMMCs (106 cells/ml) were left untreated or incubated with 11 mM siramesine (S). The indicated inhibitors were added 30 min prior to siramesine. After 24 h, cell death was measured by staining with AnnV and PI, followed by flow cytometry analysis. Data are expressed as mean values S.E.M. (n = 3). **P 0.01; ***P 0.001. UNT, untreated; PS-A, pepstatin A.
1674
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
Fig. 4. WT MCs undergo apoptotic cell death in response to siramesine but cells lacking serglycin die by necrosis. WT or serglycin / BMMCs were left either untreated or were incubated with 11 mM siramesine as indicated. (A) After 24 h, cells were stained with AnnV and PI followed by flow cytometry analysis. Note that the AnnV /PI population represents viable cells, that AnnV+/PI cells are apoptotic and that the AnnV+/PI+ population corresponds to necrotic cells. Data are expressed as mean values S.E.M. (n = 3). (B) Representative dot plots showing AnnV/PI staining of WT and serglycin / cells. (C) DNA was extracted from WT and serglycin / non-treated BMMCs (0 h) or BMMCs treated with siramesine (11 mM) for the time periods indicated, followed by separation on 1% agarose gels. *P 0.05; **P 0.01; ***P 0.001; ****P 0.0001.
fact, a comparison of the IC50 values showed that siramesine was even more toxic for PCMCs than for BMMCs (Fig. 1B). We also assessed whether siramesine induces cell death in human MCs by evaluating its effect on LAD2 and HMC-1 cells. LAD2 cells, derived from a MC sarcoma/leukemia patient, constitute relatively mature human MCs with several features shared by mature tissue MC populations, including cell surface expression of c-kit, dependence on stem cell factor for proliferation, and expression of the high affinity IgE receptor [27]. HMC-1, on the other hand, is a less mature MC-like cell line derived from a MC leukemia, lacking c-kit dependence and cell surface IgE receptor expression [28]. As shown in Fig. 1A–C, siramesine induced cell death of both of these human MC types, suggesting that siramesine is cytotoxic for MCs regardless of species of origin.
3.2. Siramesine causes secretory granule permeabilization in MCs To address the mechanism of siramesine action on MCs, we first investigated whether siramesine induces granule/lysosome damage. An established way to monitor lysosomal damage is by using acridine orange (AO), a fluorescent dye that stains acidic compartments (e.g. lysosomes and secretory granules) intensely [29]. When lysosomal damage is induced, AO will loose fluorescence intensity and loss of AO staining can thus be used to monitor lysosome/granule permeabilization. As shown in Fig. 2A, addition of siramesine to BMMCs and LAD2 cells caused a robust decrease in AO staining intensity, suggesting loss of granule/lysosome integrity. In contrast, siramesine caused only limited decreases in staining of BMMCs with nonyl-AO (NAO), a dye that marks
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1675
Fig. 5. PARP-1 degradation in response to siramesine-induced cell death is serglycin-dependent. (A) BMMCs (WT or serglycin / ; 106 cells) were left untreated or were incubated with siramesine for the time periods indicated. Cell extracts were then prepared and subjected to Western blot analysis for PARP-1 and b-Actin. (B) Quantification of the PARP-1 band intensities from independent experiments. Data are expressed as mean values S.E.M. (n = 3). ****P 0.0001.
mitochondrial integrity, indicating that siramesine induces cell death by mechanisms largely independent of mitochondrial pathways (data not shown). To obtain further proof for granule permeabilization, we analyzed cytosolic extracts from siramesinetreated cells for presence of mouse MC protease 6 (mMCP-6) and CPA3, both of these being MC granule-specific proteases that depend on serglycin proteoglycan for storage [22]. Whereas neither of these proteases was present in significant amounts in the cytosol of non-treated cells, they were both detected in the cytosol of siramesine-treated cells, supporting that siramesine causes granule permeabilization in MCs (Fig. 2B). 3.3. Siramesine-induced cell death involves cysteine cathepsins but is caspase-independent As a next step in elucidating the mechanism of siramesine action on MCs, we examined the effect of various protease inhibitors on siramesine-induced MC death. First, we investigated the impact of caspase inhibition by using either a pan-caspase inhibitor (Z-VAD) or a caspase-3/7 inhibitor (Z-DEVD). However, neither of these caspase inhibitors blocked siramesine-induced cell death, arguing that siramesine induces caspase-independent cell death in MCs (Fig. 3). As a control showing that the caspase inhibitors used were effective, caspase inhibition provided partial blockade of the cytotoxic effect of Leu-Leu-O-Me (data not shown), an agent that was previously shown to cause partially caspasedependent cell death of BMMCs [26]. MC secretory granules contain large amounts of lysosomal proteases, including cysteine cathepsins and cathepsins of aspartic acid protease class [9]. Granule permeabilization will thus lead to leakage of lysosomal enzymes into the cytosol and, since lysosomal cathepsins have previously been implicated in apoptosis [30], we assessed whether their inhibition may block siramesine-induced
MC death. As shown in Fig. 3, inhibition of cysteine cathepsins (E64d) caused a reduction in cell death whereas inhibition of aspartic acid proteases (pepstatin A) had only a minimal effect. These data suggest that siramesine-induced MC death is cysteine cathepsindependent. 3.4. MCs undergo apoptotic cell death in response to siramesine Major pathways of cell death include apoptosis and necrosis, which can be distinguished by Annexin V/propidium iodide (PI) staining, with apoptotic cells being Annexin V+/PI while necrotic cells typically are double Annexin V+/PI+. As indicated by Annexin V/PI staining, siramesine induced apoptotic cell death in BMMCs (Fig. 4A and B). In line with apoptotic cell death, DNA degradation was evident in siramesine-treated BMMCs (Fig. 4C). 3.5. Serglycin promotes apoptotic over necrotic cell death in response to siramesine A major component of MC secretory granules is serglycin, a proteoglycan of high anionic charge density that has a key role in promoting the storage of various MC-specific proteases [8]. Since serglycin-dependent proteases (mMCP-6, CPA3) were shown to enter the cytosol following siramesine treatment (see Fig. 2B) we reasoned that serglycin (and its associated proteases) may affect the cell death process, e.g. by promoting proteolysis of compounds of the apoptotic machinery. To evaluate this possibility, we compared the effect of siramesine on WT and serglycin / BMMCs. While siramesine had approximately equal cytotoxicity on WT and serglycin / BMMCs over a 24 h time period (Fig. 1A), cell death was somewhat delayed in serglycin / cells (Fig. 1C). More strikingly, whereas siramesine induced apoptotic cell death in WT BMMCs, non-viable serglycin / cells were predominantly
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1676
Fig. 6. Siramesine causes MC apoptosis in vivo. Mice (n = 5) were treated with vehicle only (VEH), vehicle containing siramesine (3 or 20 mg/kg/day; i.p. injection) or were untreated (UNT). (A) After 3 days of treatment, mice were sacrificed followed by peritoneal lavage and quantification of peritoneal MCs after staining of cytospin slides with May Gru¨nwald/Giemsa. (B) Total RNA was extracted from peritoneal cells recovered from control- or siramesine (20 mg/kg/day)-treated mice and was analyzed by real time RT-PCR for levels of mRNA coding for mMCP-6, using Hprt as house keeping control. Note the increase in Ct value for mMCP-6 expression in response to siramesine, indicating reduced mMCP-6 expression. **P 0.01; ***P 0.001; ****P 0.0001. (C) Representative cytospin slides showing May Gru¨nwald/Giemsa staining of peritoneal cells, either untreated or treated with vehicle or siramesine as indicated. Arrows indicate MCs.
necrotic (Fig. 4A). Moreover, DNA degradation was markedly delayed in serglycin / cells as compared to WT counterparts (Fig. 4C). 3.6. PARP-1 is preserved in serglycin
/
MCs
A plausible explanation for the pro-apoptotic effect of serglycin may be that serglycin-dependent mechanisms have an impact on the proteolysis of pro- or anti-apoptotic compounds. Among candidate compounds whose differential proteolytic processing could explain the apoptotic and necrotic phenotype of WT and
serglycin / cells, respectively, is Poly(ADP-ribose) polymerase 1 (PARP-1), a protein involved in DNA repair. During apoptosis PARP1 undergoes degradation, resulting in diminished DNA repair and, consequently, DNA fragmentation. On the other hand, if PARP-1 degradation is defective, sustained DNA repair can lead to energy depletion resulting in necrotic cell death [31]. As shown in Fig. 5A and B, baseline PARP-1 levels were significantly higher in serglycin / than in WT cells. Moreover, while PARP-1 was effectively degraded in WT cells treated with siramesine, PARP1 was largely preserved in cells lacking serglycin. Hence, the necrotic phenotype of serglycin / cells is clearly compatible with
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1677
Fig. 7. Siramesine does not cause depletion of peritoneal macrophage or lymphocyte populations in vivo. Mice (n = 5) were treated with vehicle only, vehicle containing siramesine (3 or 20 mg/kg/day; i.p. injection) or were non-treated. After 3 days of treatment, mice were sacrificed followed by peritoneal lavage and quantification of peritoneal cell populations after staining of cytospin slides with May Gru¨nwald/Giemsa. *P 0.05; ***P 0.001; ****P 0.0001.
effects of preserved PARP-1 activity. In agreement with these findings, preservation of PARP-1 in serglycin / mast cells was recently shown to occur in response to various other cytotoxic regimens [32]. 3.7. Siramesine depletes MC populations in vivo In order to evaluate the in vivo efficacy of siramesine in inducing MC death, we injected mice i.p. with siramesine. As depicted in Fig. 6A, siramesine at 3 mg/kg/day (3 days treatment) caused a profound depletion of the peritoneal MC population, representing MCs of the connective tissue MC (CTMC) subclass, and MCs were completely absent after treatment with 20 mg/kg/day. To confirm
the reduction in MC numbers we assessed the expression of the constitutively expressed MC-specific marker mMCP-6 by real time RT-PCR. As depicted in Fig. 6B, the expression of the mMCP-6 gene was markedly decreased after treatment of mice with siramesine (as evidenced by a robust increase in Ct value), in agreement with siramesine having a cytotoxic effect on MCs. Although MC numbers where substantially reduced by siramesine treatment, other peritoneum-resident cell types, i.e. macrophages and lymphocytes, were not reduced (Fig. 7). Thus, siramesine causes depletion of MC populations in vivo, and shows selectivity for MCs over other cell types. To assess whether siramesine also has systemic effects on MC populations distant from the site of injection, MC populations in
1678
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
Fig. 8. Siramesine has systemic effects on MC populations. Mice (n = 5) were treated with vehicle only, vehicle containing siramesine (3 or 20 mg/kg/day; i.p. injection) or were non-treated. After 3 days of treatment, mice were sacrificed. Ear skin tissue was paraffin-embedded, sectioned and stained with toluidine blue, followed by quantification of MC numbers. *P 0.05; **P 0.01; ***P 0.001. (B) Representative tissue sections stained with toluidine blue. Arrows indicate MCs.
ear skin were examined. As shown in Fig. 8, the numbers of skin MCs were reduced after treatment of mice with siramesine. Hence, siramesine has systemic effects that reduce MC populations in vivo. There was no sign of tissue damage apart from the depletion of MCs (not shown). 4. Discussion Based on the detrimental actions of MCs in a wide array of devastating diseases (see Section 1), there is an urgent need to improve therapeutic regimens that counteract their harmful effects. One way to accomplish this would be to eliminate MC populations by inducing their apoptosis, either globally or
specifically at the site of lesion [33–36]. Here we explored a novel approach for inducing MC apoptosis by taking advantage of the high abundance or protease-containing secretory granules in MCs. Our hypothesis was that targeting of the MC granules would cause release of granular content, including various proteases, into the cytosol of the MCs and that this potentially could result in apoptosis. Moreover, since MCs are exceptionally rich in secretory granules, we hypothesized that MCs would be more sensitive to apoptosis induced by this type of regimen in comparison with most other cell types. In support of this concept, we previously showed that MCs are sensitive to the lysosomotropic agent Leu-Leu-O-Me, by a mechanism involving granule permeabilization [26,37]. However, since Leu-Leu-O-Me has not been approved for use in
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
vivo, and because relatively high concentrations of Leu-Leu-O-Me are required for cytotoxicity [38–41], an important goal of this investigation was to identify granule-targeting compounds that are of high efficacy and that can be administered in vivo. Here we identify siramesine as a compound fulfilling these criteria. For therapeutic purposes, a MC-targeting compound should ideally act by inducing apoptosis rather than necrosis, since necrotic cell death is associated with loss of cell membrane integrity and release of pro-inflammatory danger signals from dying cells. Indeed, out data show that siramesine induces cell death with typical signs of apoptosis rather than necrosis and it is thus possible that siramesine may be used without causing extensive inflammatory side effects. Intriguingly, although MC death was characterized as apoptosis, no effect of caspase inhibitors was seen, suggesting caspaseindependent cell death. Notably, this is in agreement with previous studies showing that siramesine causes caspase-independent cell death of various transformed cell lines [19]. While no effect of caspase inhibition was observed, inhibition of cysteine cathepsins significantly prevented cell death. This is clearly in line with numerous reports implicating cysteine cathepsins in apoptotic cell death (reviewed in [30]). Cysteine cathepsins are localized to the lysosomal compartment and are also found in MC secretory granules. A role for these enzymes in siramesine-induced cell death thus implies that lysosomal/granule compartments are damaged in response to siramesine. In support of this notion, siramesine treatment caused reduced AO fluorescence, a sign of lost lysosomal/granular integrity. Moreover, mMCP-6 and CPA3 were recovered in the cytosol of siramesine-treated MCs but were absent in the cytosol of untreated cells. Since both of these MCspecific proteases are localized to secretory granules, these findings argue strongly that siramesine indeed causes permeabilization of the MC secretory granules, followed by release of granule content into the cytosol. A striking finding was that, whereas WT MCs underwent apoptotic cell death in response to siramesine, cells lacking serglycin died predominantly by necrosis. Serglycin is a main component of MC secretory granules, and previous work has shown that serglycin holds a key role in maintaining and regulating secretory granule homeostasis by promoting the storage of a variety of granule compounds, including bioactive monoamines and different MCspecific proteases [8]. Clearly, the strong apoptosis-promoting effect of serglycin further reinforces the crucial role of secretory granule permeabilization in cell death caused by siramesine. Notably, serglycin has also been shown to promote apoptosis vs. necrosis in response various other cytotoxic agents [26,32]. An important aspect of this investigation was to evaluate whether siramesine has the potential to reduce MC populations in vivo. Indeed, i.p. injection of mice with siramesine depleted the peritoneal MC population, indicating that siramesine has cytotoxic activity toward MCs at the site of injection. It was observed that the i.p. administration of siramesine caused moderate signs of inflammation, as evidenced by increased numbers of macrophages and other inflammatory cells. Possibly, this finding could be explained by pro-inflammatory activity of siramesine, which is not related to its effects on MCs. Alternatively, the moderate inflammation may be caused by factors that are released in processes accompanying the apoptosis of MCs. For example, it has been demonstrated that the clearance of apoptotic bodies can be associated with release of monocyte and neutrophil attractants [42]. Potentially, this may aid to recruit additional phagocytic cells, thereby enhancing the efficiency of clearing the site from apoptotic bodies. As shown by the significant reductions of the MC populations in skin, siramesine has also systemic effects on MC populations distant from the site of administration. Importantly, whereas the
1679
peritoneal MC population was completely depleted, other resident cell populations of the peritoneum, i.e. macrophages and lymphocytes, were not reduced by siramesine in vivo. Hence, these findings suggest that siramesine shows selectivity for MCs over other leukocyte populations in vivo, introducing the possibility of using siramesine (or compounds with similar properties) in therapy aiming at counteracting harmful actions of MCs. However, the ability of siramesine to ameliorate MCdependent diseases, in both murine models and in humans, remains to be investigated. Acknowledgments We are grateful to Lundbeck A/S for providing siramesine and to Rene´ Holm (Lundbeck A/S) for helpful advice. This work was supported by grants from: The Swedish Research Council, The Swedish Cancer Foundation and King Gustaf V’s 80-year Anniversary Fund.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bcp.2012.09.028. References [1] Bradding P, Walls AF, Holgate ST. The role of the mast cell in the pathophysiology of asthma. J Allergy Clin Immunol 2006;117:1277–84. [2] Galli SJ, Tsai M. IgE and mast cells in allergic disease. Nat Med 2012;18: 693–704. [3] Dudeck A, Dudeck J, Scholten J, Petzold A, Surianarayanan S, Kohler A, et al. Mast cells are key promoters of contact allergy that mediate the adjuvant effects of haptens. Immunity 2011;34:973–84. [4] Soucek L, Lawlor ER, Soto D, Shchors K, Swigart LB, Evan GI. Mast cells are required for angiogenesis and macroscopic expansion of Myc-induced pancreatic islet tumors. Nat Med 2007;13:1211–8. [5] Sun J, Sukhova GK, Wolters PJ, Yang M, Kitamoto S, Libby P, et al. Mast cells promote atherosclerosis by releasing proinflammatory cytokines. Nat Med 2007;13:719–24. [6] Liu J, Divoux A, Sun J, Zhang J, Clement K, Glickman JN, et al. Genetic deficiency and pharmacological stabilization of mast cells reduce diet-induced obesity and diabetes in mice. Nat Med 2009;15:940–5. [7] Metcalfe DD, Baram D, Mekori YA. Mast cells. Physiol Rev 1997;77:1033–79. [8] Kolset SO, Pejler G. Serglycin-A structural and functional chameleon with wide impact on immune cells. J Immunol 2011;187:4927–33. [9] Lundequist A, Pejler G. Biological implications of preformed mast cell mediators. Cell Mol Life Sci 2011;68:965–75. [10] Pejler G, Ro¨nnberg E, Waern I, Wernersson S. Mast cell proteases: multifaceted regulators of inflammatory disease. Blood 2010;115:4981–90. [11] Rivera J, Fierro NA, Olivera A, Suzuki R. New insights on mast cell activation via the high affinity receptor for IgE. Adv Immunol 2008;98:85–120. [12] Boyce JA. Mast cells and eicosanoid mediators: a system of reciprocal paracrine and autocrine regulation. Immunol Rev 2007;217:168–85. [13] Galli SJ, Grimbaldeston M, Tsai M. Immunomodulatory mast cells: negative, as well as positive, regulators of immunity. Nat Rev Immunol 2008;8:478–86. [14] Peachell P. Regulation of mast cells by beta-agonists. Clin Rev Allergy Immunol 2006;31:131–42. [15] Takai S, Jin D, Miyazaki M. Targets of chymase inhibitors. Expert Opin Ther Targets 2011;15:519–27. [16] Sommerhoff CP, Schaschke N. Mast cell tryptase beta as a target in allergic inflammation: an evolving story. Curr Pharm Des 2007;13:313–32. [17] Karra L, Berent-Maoz B, Ben-Zimra M, Levi-Schaffer F. Are we ready to downregulate mast cells. Curr Opin Immunol 2009;21:708–14. [18] Heading C. Siramesine H Lundbeck. Curr Opin Investig Drugs 2001;2:266–70. [19] Ostenfeld MS, Fehrenbacher N, Hoyer-Hansen M, Thomsen C, Farkas T, Jaattela M. Effective tumor cell death by sigma-2 receptor ligand siramesine involves lysosomal leakage and oxidative stress. Cancer Res 2005;65:8975–83. [20] Zeng C, Rothfuss J, Zhang J, Chu W, Vangveravong S, Tu Z, et al. Sigma-2 ligands induce tumour cell death by multiple signalling pathways. Br J Cancer 2012;106:693–701. [21] Bird PI, Trapani JA, Villadangos JA. Endolysosomal proteases and their inhibitors in immunity. Nat Rev Immunol 2009;9:871–82. [22] A˚brink M, Grujic M, Pejler G. Serglycin is essential for maturation of mast cell secretory granule. J Biol Chem 2004;279:40897–905. [23] Ro¨nnberg E, Pejler G. Serglycin: the master of the mast cell. Methods Mol Biol 2012;836:201–17.
1680
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
[24] Malbec O, Roget K, Schiffer C, Iannascoli B, Dumas AR, Arock M, et al. Peritoneal cell-derived mast cells: an in vitro model of mature serosal-type mouse mast cells. J Immunol 2007;178:6465–75. [25] Lundequist A, Calounova G, Wensman H, Ronnberg E, Pejler G. Differential regulation of Nr4a subfamily nuclear receptors following mast cell activation. Mol Immunol 2011;48:1753–61. [26] Melo FR, Waern I, Ro¨nnberg E, A˚brink M, Lee DM, Schlenner SM, et al. A role for serglycin proteoglycan in mast cell apoptosis induced by a secretory granulemediated pathway. J Biol Chem 2011;286:5423–33. [27] Kirshenbaum AS, Akin C, Wu Y, Rottem M, Goff JP, Beaven MA, et al. Characterization of novel stem cell factor responsive human mast cell lines LAD 1 and 2 established from a patient with mast cell sarcoma/leukemia; activation following aggregation of FcepsilonRI or FcgammaRI. Leuk Res 2003;27:677–82. [28] Furitsu T, Tsujimura T, Tono T, Ikeda H, Kitayama H, Koshimizu U, et al. Identification of mutations in the coding sequence of the proto-oncogene c-kit in a human mast cell leukemia cell line causing ligand-independent activation of c-kit product. J Clin Invest 1993;92:1736–44. [29] Ivanova S, Repnik U, Bojic L, Petelin A, Turk V, Turk B. Lysosomes in apoptosis. Methods Enzymol 2008;442:183–99. [30] Turk B, Turk V. Lysosomes as suicide bags in cell death: myth or reality. J Biol Chem 2009;284:21783–87. [31] Ha HC, Snyder SH. Poly(ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc Natl Acad Sci USA 1999;96:13978–82. [32] Melo FR, Grujic M, Spirkoski J, Calounova G, Pejler G. Serglycin proteoglycan promotes apoptotic versus necrotic cell death in mast cells. J Biol Chem 2012;287:18142–52. [33] Karlberg M, Ekoff M, Huang DC, Mustonen P, Harvima IT, Nilsson G. The BH3mimetic ABT-737 induces mast cell apoptosis in vitro and in vivo: potential for therapeutics. J Immunol 2010;185:2555–62.
[34] Berent-Maoz B, Piliponsky AM, Daigle I, Simon HU, Levi-Schaffer F. Human mast cells undergo TRAIL-induced apoptosis. J Immunol 2006;176:2272–8. [35] Juurikivi A, Sandler C, Lindstedt KA, Kovanen PT, Juutilainen T, Leskinen MJ, et al. Inhibition of c-kit tyrosine kinase by imatinib mesylate induces apoptosis in mast cells in rheumatoid synovia: a potential approach to the treatment of arthritis. Ann Rheum Dis 2005;64:1126–31. [36] Hartmann K, Wagelie-Steffen AL, von Stebut E, Metcalfe DD. Fas (CD95, APO-1) antigen expression and function in murine mast cells. J Immunol 1997;159: 4006–14. [37] Melo FR, Lundequist A, Calounova G, Wernersson S, Pejler G. Lysosomal membrane permeabilization induces cell death in human mast cells. Scand J Immunol 2011;74:354–62. [38] Thiele DL, Lipsky PE. The immunosuppressive activity of L-leucyl-L-leucine methyl ester: selective ablation of cytotoxic lymphocytes and monocytes. J Immunol 1986;136:1038–48. [39] Thiele DL, Lipsky PE. Mechanism of L-leucyl-L-leucine methyl ester-mediated killing of cytotoxic lymphocytes: dependence on a lysosomal thiol protease, dipeptidyl peptidase I, that is enriched in these cells. Proc Natl Acad Sci USA 1990;87:83–7. [40] Cirman T, Oresic K, Mazovec GD, Turk V, Reed JC, Myers RM, et al. Selective disruption of lysosomes in HeLa cells triggers apoptosis mediated by cleavage of Bid by multiple papain-like lysosomal cathepsins. J Biol Chem 2004;279: 3578–87. [41] Droga-Mazovec G, Bojic L, Petelin A, Ivanova S, Romih R, Repnik U, et al. Cysteine cathepsins trigger caspase-dependent cell death through cleavage of bid and antiapoptotic Bcl-2 homologues. J Biol Chem 2008;283:19140–50. [42] Fries DM, Lightfoot R, Koval M, Ischiropoulos H. Autologous apoptotic cell engulfment stimulates chemokine secretion by vascular smooth muscle cells. Am J Pathol 2005;167:345–53.
Contents lists available at SciVerse ScienceDirect
Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm
Mast cell apoptosis induced by siramesine, a sigma-2 receptor agonist Jane Spirkoski 1, Fabio R. Melo 1, Mirjana Grujic, Gabriela Calounova, Anders Lundequist, Sara Wernersson, Gunnar Pejler * Swedish University of Agricultural Sciences, Dept. of Anatomy, Physiology and Biochemistry, BMC Box 575, 75123, Uppsala, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 20 August 2012 Accepted 26 September 2012 Available online 9 October 2012
Mast cells (MCs) are well known for their detrimental effects in the context of allergic disorders. Strategies that limit MC function can therefore have a therapeutic value. Previous studies have shown that siramesine, a sigma-2 receptor agonist originally developed as an anti-depressant, can induce cell death in transformed cells through a mechanism involving lysosomal destabilization. Since MCs are remarkably rich in lysosome-like secretory granules we reasoned that MCs might be sensitive to siramesine. Here we show that murine and human MCs are highly sensitive to siramesine. Cell death was accompanied by secretory granule permeabilization, as shown by reduced acridine orange staining and leakage of granule proteases into the cytosol. Wild type siramesine-treated MCs underwent cell death with typical signs of apoptosis but MCs lacking serglycin, a proteoglycan crucial for promoting the storage of proteases within MC secretory granules, died predominantly by necrosis. A dissection of the underlying mechanism suggested that the necrotic phenotype of serglycin / cells was linked to defective Poly(ADP-ribose) polymerase-1 degradation. In vivo, siramesine treatment of mice caused a depletion of the MC populations of the peritoneum and skin. The present study shows for the first time that MCs are highly sensitive to apoptosis induced by siramesine and introduces the possibility of using siramesine as a therapeutic agent for treatment of MC-dependent disease. ß 2012 Elsevier Inc. All rights reserved.
Keywords: Mast cells Apoptosis Siramesine Serglycin Secretory granules
1. Introduction Mast cells (MCs) are major effector cells in allergic conditions such as anaphylaxis and allergic asthma [1,2], but recent studies have strongly implicated MCs also in other types of disease [3–6]. A hallmark feature of MCs is their high content of densely packed secretory granules. These are filled with large quantities of various preformed compounds, including bioactive monoamines, cytokines, serglycin proteoglycan, lysosomal enzymes, as well as MCspecific proteases of chymase, tryptase and carboxypeptidase A3 (CPA3) type [7–10]. MC degranulation can be triggered by different means, including antigen-mediated crosslinking of immunoglobulin (Ig) E molecules bound to the high affinity IgE receptor [11]. Upon
Abbreviations: MC, mast cell; mMCP, mouse mast cell protease; BMMC, bone marrow-derived mast cell; PCMC, peritoneal cell-derived mast cell; PARP-1, Poly(ADP-ribose) polymerase-1; AO, acridine orange; WT, wild type; AnnV, Annexin V; CPA3, carboxypeptidase A3; Ig, immunoglobulin; RT-PCR, reverse transcription polymerase chain reaction; HMC-1, Human Mast Cell Line-1; LAD2, Laboratory of Allergic Diseases-2. * Corresponding author. Tel.: +46 18 4714090. E-mail address: [email protected] (G. Pejler). 1 Equal contribution. 0006-2952/$ – see front matter ß 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bcp.2012.09.028
degranulation, the preformed MC mediators are released into the surrounding tissue. Furthermore, MC activation will trigger de novo synthesis of various additional pro-inflammatory compounds, encompassing eicosanoids and a panel of cytokines and chemokines [12,13]. Given the wide array of detrimental MC actions, strategies that limit MC function may be utilized for therapeutic purposes. Such strategies include various stabilizers that prevent MC degranulation, as well as selective targeting of different products secreted by MCs [14–16]. However, the impact of MCs on a given disorder may be multifaceted, i.e. being mediated by multiple products secreted by MCs. Hence, an efficient blockade of detrimental MC effects may require global inhibition of activities mediated by MCs. To accomplish this, an emerging strategy is to eradicate MC populations altogether by selectively inducing MC apoptosis (reviewed in [17]). Siramesine is a sigma-2 receptor agonist that was originally intended as an anti-depressant [18]. The drug has been proven safe for use in humans but was abandoned as anti-depressant because of low efficacy. In addition to its proposed anti-depressant properties, it has been shown that siramesine has pro-apoptotic activity on various transformed cell types [19,20], and that the cytotoxic activity of siramesine on transformed cells involves lysosomal destabilization [19]. Lysosomes and secretory granules
1672
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
share many features, including acidic pH, membrane constituents and content of lysosomal hydrolases [9], and we therefore reasoned that agents (such as siramesine) causing permeabilization of lysosomes might have similar actions on secretory granules. MC granules have an extraordinary high content of fully active proteases [10], and damage to the MC granules will thus lead to the entry of vast amounts of proteolytic activity into the cytosol. Potentially, this can lead to apoptosis through proteolytic activation of cytosolic pro-apoptotic compounds and/or degradation of anti-apoptotic compounds [21]. Based on these notions, we decided to investigate whether siramesine had granule-permeabilizing activity on MCs, and whether such activity could cause cell death. Indeed, we here show that siramesine efficiently and selectively induces MC apoptosis. 2. Materials and methods
skin was dissected, paraffin-embedded and sectioned. The cytospin slides were stained with May Gru¨nwald/Giemsa, whereas the tissue sections were stained with toluidine blue and hematoxylin/eosin. Numbers of individual cell types were calculated from the total number of peritoneal lavage cells and from the percentage of individual cell types on cytospin slides. In skin specimens, the numbers of MCs in uniformly sized sections were enumerated based on staining with toluidine blue. 2.5. Cell viability measurement 100 ml from cell suspensions containing the various populations of MCs were collected and transferred to individual wells of a 96-well plate. 20 ml of CellTiter-Blue1 Reagent (Promega-Invitrogen, Carlsbad, CA) was added to the wells, followed by incubation for 4–5 h at 37 8C. Fluorescence was read using a microplate reader (Infinite M200 – TECAN) at 560 nm for excitation and 590 nm for emission.
2.1. Reagents 2.6. Staining of acidic compartments Siramesine was generously given by Lundbeck A/S (Copenhagen, Denmark). Z-DEVD-FMK, Z-VAD-FMK, E-64d and Pepstatin A were purchased from Sigma–Aldrich (Steinheim, Germany). Antib actin goat polyclonal antibodies were from Santa Cruz Biotechnology (sc1616, Santa Cruz, CA); anti-PARP rabbit polyclonal was from Abcam (ab6079, Cambridge, UK). Fluorescently labeled donkey anti-rabbit and -goat Ig were purchased from Li-cor (Bad Homburg, Germany). Acridine orange (AO) was from Sigma– Aldrich (Steinheim, Germany).
Cells (5 105 cells/ml) were incubated for 30 min with 5 mg/ml acridine orange (AO; Sigma–Aldrich). The cells were pelleted by centrifugation (300 g; 8 min), and washed three times with PBS and then resuspended in 500 ml PBS containing BSA (100 mg/ml). 100 ml from the cell suspensions was transferred to a 96-well plate. The fluorescence intensity was measured using a microplate reader (Infinite M200 – TECAN); 488 nm for excitation and 650 nm for emission.
2.2. Mice
2.7. Flow cytometry and Western blot analysis
Wild-type (WT) and serglycin / [22] mice were on C57BL/6J genetic background. All animal experiments were approved by the local ethical committee (Uppsala, Sweden).
Flow cytometry [26] and Western blot analysis [23] was performed as described. 2.8. DNA degradation
2.3. Cell culture Bone Marrow Derived Mast Cells (BMMCs) [23] and peritoneal cell-derived MCs (PCMCs) [24] were isolated and cultured as previously described. Human Mast Cell Line-1 (HMC-1, specifically HMC-1.2; kind gift from the Mayo Foundation for Medical Education and Research) and the cell line denoted Laboratory of Allergic Disease 2 (LAD2; gift from A. Kirshenbaum and Dean D. Metcalfe, National Institutes of Health) were cultured as described [25].
DNA was isolated as described [8]. Samples were separated using 1% agarose gel electrophoresis containing 0.1 ml/ml Gel Red Nucleic Acid Stain (Biotium, Hayward, CA) and visualized under UV light transillumination. 2.9. Real time reverse transcription polymerase chain reaction (RTPCR) Real time RT-PCR primers and conditions were as described [23].
2.4. Effects of siramesine 2.10. Statistical calculations For in vitro experiments, a stock solution of 5 mM siramesine in Polyethylene glycol (PEG) 400 was prepared. A 250 mM working solution was prepared by diluting the stock solution with 50% dimethyl sulfoxide (DMSO). Control samples were treated with vehicle containing the same concentration of PEG and DMSO as in samples treated with siramesine. Treatment of MCs with vehicle alone did not induce cell death (data not shown). Inhibitors were used at the following final concentrations: Z-DEVD-FMK (20 mM); Z-VAD-FMK (20 mM); E-64d (20 mM; inhibitor of cysteine proteases); Pepstatin A (50 mM; inhibitor of aspartic acid proteases). For in vivo experiments, siramesine was diluted in 10% HPbetacyclodextrin. WT mice were either non-treated or injected i.p. with 100 ml vehicle only (10% HPbeta-cyclodextrin) or with siramesine (3 and 20 mg/kg/day) for 3 consecutive days. Mice were then sacrificed followed by peritoneal lavage, counting of total cells in the lavage fluid and preparation of cytospin slides. Tissue from ear
The raw data were exported to the Microsoft Office Excel 2007 software program, where most of the calculations were done. Preparation of graphs, calculation of dose dependency and calculation the significance [One-Way ANOVA (Figs. 4–8) or unpaired t-test (Fig. 3)] was performed using GraphPad Prism 4 software (GraphPad Software, Inc, San Diego, CA). Results shown are from single experiments, representative of at least three individual experiments. 3. Results 3.1. Siramesine causes apoptosis of murine and human MCs Previous studies have shown that siramesine has antitumorigenic properties and, moreover, that the cytotoxic effect of siramesine was higher on cells that had undergone neoplastic
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1673
transformation than on untransformed cells [19]. Based on the notion that siramesine acts through lysosomal destabilization [19] and that MCs are exceptionally rich in lysosome-like secretory granules we hypothesized that siramesine also could be cytotoxic for MCs. Indeed, as shown in Fig. 1A and B, siramesine showed dose-dependent, potent cytotoxic activity on murine bone marrow-derived MCs (BMMCs) as determined by cell viability measurements. A time course experiment showed that cell death was achieved within 24 h (Fig. 1C). Notably, BMMCs are nontransformed primary cells, and these data thus suggest that the cytotoxic activity of siramesine is not limited to transformed cell types.
Fig. 2. Siramesine induces secretory granule permeabilization. (A) BMMCs (WT or serglycin / ) and LAD2 cells were incubated with 11 mM siramesine. At the time points indicated, cells were stained with AO. Results are expressed as % fluorescence in comparison with non-treated control cells. Data in (A) are expressed as mean values S.E.M. (n = 3). (B) BMMCs (WT or serglycin / ; 106 cells) were incubated with siramesine (11 mM) for the time periods indicated. Cytosolic extracts were then prepared and subjected to Western blot analysis for presence of granule proteases (mMCP-6 and CPA3). Note the low levels of mMCP-6 and CPA3 in cytosolic extracts of control samples (0 h). Note also that mMCP-6 and CPA3 as expected were undetectable in serglycin / cells, the latter in agreement with the strong dependence of mMCP-6 and CPA3 on serglycin for storage.
Next, we determined whether siramesine is cytotoxic for MC types in addition to BMMCs. In order to assess if siramesine is apoptotic also for terminally differentiated MCs we developed peritoneal cell-derived MCs (PCMCs) [24] and subjected these to siramesine. Indeed, siramesine was highly cytotoxic for PCMCs. In
Fig. 1. Siramesine is cytotoxic for MCs. (A) BMMCs (WT or serglycin / ) and LAD2 cells were incubated with different concentrations of siramesine as indicated. After 24 h, cell viability was measured. Results are expressed as % viable cells in comparison with non-treated control cells. (B) IC50 values for cytotoxicity of siramesine on various MC types. Data are expressed as mean values S.E.M. (C) Time course of siramesine-induced cell death. BMMCs (WT or serglycin / ) and LAD2 cells were incubated with 11 mM siramesine for the time periods indicated, followed by measurements of residual cell viability as compared with non-treated control cells (n = 3).
Fig. 3. Siramesine-induced cell death in BMMCs is caspase-independent and cysteine cathepsin-dependent. BMMCs (106 cells/ml) were left untreated or incubated with 11 mM siramesine (S). The indicated inhibitors were added 30 min prior to siramesine. After 24 h, cell death was measured by staining with AnnV and PI, followed by flow cytometry analysis. Data are expressed as mean values S.E.M. (n = 3). **P 0.01; ***P 0.001. UNT, untreated; PS-A, pepstatin A.
1674
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
Fig. 4. WT MCs undergo apoptotic cell death in response to siramesine but cells lacking serglycin die by necrosis. WT or serglycin / BMMCs were left either untreated or were incubated with 11 mM siramesine as indicated. (A) After 24 h, cells were stained with AnnV and PI followed by flow cytometry analysis. Note that the AnnV /PI population represents viable cells, that AnnV+/PI cells are apoptotic and that the AnnV+/PI+ population corresponds to necrotic cells. Data are expressed as mean values S.E.M. (n = 3). (B) Representative dot plots showing AnnV/PI staining of WT and serglycin / cells. (C) DNA was extracted from WT and serglycin / non-treated BMMCs (0 h) or BMMCs treated with siramesine (11 mM) for the time periods indicated, followed by separation on 1% agarose gels. *P 0.05; **P 0.01; ***P 0.001; ****P 0.0001.
fact, a comparison of the IC50 values showed that siramesine was even more toxic for PCMCs than for BMMCs (Fig. 1B). We also assessed whether siramesine induces cell death in human MCs by evaluating its effect on LAD2 and HMC-1 cells. LAD2 cells, derived from a MC sarcoma/leukemia patient, constitute relatively mature human MCs with several features shared by mature tissue MC populations, including cell surface expression of c-kit, dependence on stem cell factor for proliferation, and expression of the high affinity IgE receptor [27]. HMC-1, on the other hand, is a less mature MC-like cell line derived from a MC leukemia, lacking c-kit dependence and cell surface IgE receptor expression [28]. As shown in Fig. 1A–C, siramesine induced cell death of both of these human MC types, suggesting that siramesine is cytotoxic for MCs regardless of species of origin.
3.2. Siramesine causes secretory granule permeabilization in MCs To address the mechanism of siramesine action on MCs, we first investigated whether siramesine induces granule/lysosome damage. An established way to monitor lysosomal damage is by using acridine orange (AO), a fluorescent dye that stains acidic compartments (e.g. lysosomes and secretory granules) intensely [29]. When lysosomal damage is induced, AO will loose fluorescence intensity and loss of AO staining can thus be used to monitor lysosome/granule permeabilization. As shown in Fig. 2A, addition of siramesine to BMMCs and LAD2 cells caused a robust decrease in AO staining intensity, suggesting loss of granule/lysosome integrity. In contrast, siramesine caused only limited decreases in staining of BMMCs with nonyl-AO (NAO), a dye that marks
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1675
Fig. 5. PARP-1 degradation in response to siramesine-induced cell death is serglycin-dependent. (A) BMMCs (WT or serglycin / ; 106 cells) were left untreated or were incubated with siramesine for the time periods indicated. Cell extracts were then prepared and subjected to Western blot analysis for PARP-1 and b-Actin. (B) Quantification of the PARP-1 band intensities from independent experiments. Data are expressed as mean values S.E.M. (n = 3). ****P 0.0001.
mitochondrial integrity, indicating that siramesine induces cell death by mechanisms largely independent of mitochondrial pathways (data not shown). To obtain further proof for granule permeabilization, we analyzed cytosolic extracts from siramesinetreated cells for presence of mouse MC protease 6 (mMCP-6) and CPA3, both of these being MC granule-specific proteases that depend on serglycin proteoglycan for storage [22]. Whereas neither of these proteases was present in significant amounts in the cytosol of non-treated cells, they were both detected in the cytosol of siramesine-treated cells, supporting that siramesine causes granule permeabilization in MCs (Fig. 2B). 3.3. Siramesine-induced cell death involves cysteine cathepsins but is caspase-independent As a next step in elucidating the mechanism of siramesine action on MCs, we examined the effect of various protease inhibitors on siramesine-induced MC death. First, we investigated the impact of caspase inhibition by using either a pan-caspase inhibitor (Z-VAD) or a caspase-3/7 inhibitor (Z-DEVD). However, neither of these caspase inhibitors blocked siramesine-induced cell death, arguing that siramesine induces caspase-independent cell death in MCs (Fig. 3). As a control showing that the caspase inhibitors used were effective, caspase inhibition provided partial blockade of the cytotoxic effect of Leu-Leu-O-Me (data not shown), an agent that was previously shown to cause partially caspasedependent cell death of BMMCs [26]. MC secretory granules contain large amounts of lysosomal proteases, including cysteine cathepsins and cathepsins of aspartic acid protease class [9]. Granule permeabilization will thus lead to leakage of lysosomal enzymes into the cytosol and, since lysosomal cathepsins have previously been implicated in apoptosis [30], we assessed whether their inhibition may block siramesine-induced
MC death. As shown in Fig. 3, inhibition of cysteine cathepsins (E64d) caused a reduction in cell death whereas inhibition of aspartic acid proteases (pepstatin A) had only a minimal effect. These data suggest that siramesine-induced MC death is cysteine cathepsindependent. 3.4. MCs undergo apoptotic cell death in response to siramesine Major pathways of cell death include apoptosis and necrosis, which can be distinguished by Annexin V/propidium iodide (PI) staining, with apoptotic cells being Annexin V+/PI while necrotic cells typically are double Annexin V+/PI+. As indicated by Annexin V/PI staining, siramesine induced apoptotic cell death in BMMCs (Fig. 4A and B). In line with apoptotic cell death, DNA degradation was evident in siramesine-treated BMMCs (Fig. 4C). 3.5. Serglycin promotes apoptotic over necrotic cell death in response to siramesine A major component of MC secretory granules is serglycin, a proteoglycan of high anionic charge density that has a key role in promoting the storage of various MC-specific proteases [8]. Since serglycin-dependent proteases (mMCP-6, CPA3) were shown to enter the cytosol following siramesine treatment (see Fig. 2B) we reasoned that serglycin (and its associated proteases) may affect the cell death process, e.g. by promoting proteolysis of compounds of the apoptotic machinery. To evaluate this possibility, we compared the effect of siramesine on WT and serglycin / BMMCs. While siramesine had approximately equal cytotoxicity on WT and serglycin / BMMCs over a 24 h time period (Fig. 1A), cell death was somewhat delayed in serglycin / cells (Fig. 1C). More strikingly, whereas siramesine induced apoptotic cell death in WT BMMCs, non-viable serglycin / cells were predominantly
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1676
Fig. 6. Siramesine causes MC apoptosis in vivo. Mice (n = 5) were treated with vehicle only (VEH), vehicle containing siramesine (3 or 20 mg/kg/day; i.p. injection) or were untreated (UNT). (A) After 3 days of treatment, mice were sacrificed followed by peritoneal lavage and quantification of peritoneal MCs after staining of cytospin slides with May Gru¨nwald/Giemsa. (B) Total RNA was extracted from peritoneal cells recovered from control- or siramesine (20 mg/kg/day)-treated mice and was analyzed by real time RT-PCR for levels of mRNA coding for mMCP-6, using Hprt as house keeping control. Note the increase in Ct value for mMCP-6 expression in response to siramesine, indicating reduced mMCP-6 expression. **P 0.01; ***P 0.001; ****P 0.0001. (C) Representative cytospin slides showing May Gru¨nwald/Giemsa staining of peritoneal cells, either untreated or treated with vehicle or siramesine as indicated. Arrows indicate MCs.
necrotic (Fig. 4A). Moreover, DNA degradation was markedly delayed in serglycin / cells as compared to WT counterparts (Fig. 4C). 3.6. PARP-1 is preserved in serglycin
/
MCs
A plausible explanation for the pro-apoptotic effect of serglycin may be that serglycin-dependent mechanisms have an impact on the proteolysis of pro- or anti-apoptotic compounds. Among candidate compounds whose differential proteolytic processing could explain the apoptotic and necrotic phenotype of WT and
serglycin / cells, respectively, is Poly(ADP-ribose) polymerase 1 (PARP-1), a protein involved in DNA repair. During apoptosis PARP1 undergoes degradation, resulting in diminished DNA repair and, consequently, DNA fragmentation. On the other hand, if PARP-1 degradation is defective, sustained DNA repair can lead to energy depletion resulting in necrotic cell death [31]. As shown in Fig. 5A and B, baseline PARP-1 levels were significantly higher in serglycin / than in WT cells. Moreover, while PARP-1 was effectively degraded in WT cells treated with siramesine, PARP1 was largely preserved in cells lacking serglycin. Hence, the necrotic phenotype of serglycin / cells is clearly compatible with
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
1677
Fig. 7. Siramesine does not cause depletion of peritoneal macrophage or lymphocyte populations in vivo. Mice (n = 5) were treated with vehicle only, vehicle containing siramesine (3 or 20 mg/kg/day; i.p. injection) or were non-treated. After 3 days of treatment, mice were sacrificed followed by peritoneal lavage and quantification of peritoneal cell populations after staining of cytospin slides with May Gru¨nwald/Giemsa. *P 0.05; ***P 0.001; ****P 0.0001.
effects of preserved PARP-1 activity. In agreement with these findings, preservation of PARP-1 in serglycin / mast cells was recently shown to occur in response to various other cytotoxic regimens [32]. 3.7. Siramesine depletes MC populations in vivo In order to evaluate the in vivo efficacy of siramesine in inducing MC death, we injected mice i.p. with siramesine. As depicted in Fig. 6A, siramesine at 3 mg/kg/day (3 days treatment) caused a profound depletion of the peritoneal MC population, representing MCs of the connective tissue MC (CTMC) subclass, and MCs were completely absent after treatment with 20 mg/kg/day. To confirm
the reduction in MC numbers we assessed the expression of the constitutively expressed MC-specific marker mMCP-6 by real time RT-PCR. As depicted in Fig. 6B, the expression of the mMCP-6 gene was markedly decreased after treatment of mice with siramesine (as evidenced by a robust increase in Ct value), in agreement with siramesine having a cytotoxic effect on MCs. Although MC numbers where substantially reduced by siramesine treatment, other peritoneum-resident cell types, i.e. macrophages and lymphocytes, were not reduced (Fig. 7). Thus, siramesine causes depletion of MC populations in vivo, and shows selectivity for MCs over other cell types. To assess whether siramesine also has systemic effects on MC populations distant from the site of injection, MC populations in
1678
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
Fig. 8. Siramesine has systemic effects on MC populations. Mice (n = 5) were treated with vehicle only, vehicle containing siramesine (3 or 20 mg/kg/day; i.p. injection) or were non-treated. After 3 days of treatment, mice were sacrificed. Ear skin tissue was paraffin-embedded, sectioned and stained with toluidine blue, followed by quantification of MC numbers. *P 0.05; **P 0.01; ***P 0.001. (B) Representative tissue sections stained with toluidine blue. Arrows indicate MCs.
ear skin were examined. As shown in Fig. 8, the numbers of skin MCs were reduced after treatment of mice with siramesine. Hence, siramesine has systemic effects that reduce MC populations in vivo. There was no sign of tissue damage apart from the depletion of MCs (not shown). 4. Discussion Based on the detrimental actions of MCs in a wide array of devastating diseases (see Section 1), there is an urgent need to improve therapeutic regimens that counteract their harmful effects. One way to accomplish this would be to eliminate MC populations by inducing their apoptosis, either globally or
specifically at the site of lesion [33–36]. Here we explored a novel approach for inducing MC apoptosis by taking advantage of the high abundance or protease-containing secretory granules in MCs. Our hypothesis was that targeting of the MC granules would cause release of granular content, including various proteases, into the cytosol of the MCs and that this potentially could result in apoptosis. Moreover, since MCs are exceptionally rich in secretory granules, we hypothesized that MCs would be more sensitive to apoptosis induced by this type of regimen in comparison with most other cell types. In support of this concept, we previously showed that MCs are sensitive to the lysosomotropic agent Leu-Leu-O-Me, by a mechanism involving granule permeabilization [26,37]. However, since Leu-Leu-O-Me has not been approved for use in
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
vivo, and because relatively high concentrations of Leu-Leu-O-Me are required for cytotoxicity [38–41], an important goal of this investigation was to identify granule-targeting compounds that are of high efficacy and that can be administered in vivo. Here we identify siramesine as a compound fulfilling these criteria. For therapeutic purposes, a MC-targeting compound should ideally act by inducing apoptosis rather than necrosis, since necrotic cell death is associated with loss of cell membrane integrity and release of pro-inflammatory danger signals from dying cells. Indeed, out data show that siramesine induces cell death with typical signs of apoptosis rather than necrosis and it is thus possible that siramesine may be used without causing extensive inflammatory side effects. Intriguingly, although MC death was characterized as apoptosis, no effect of caspase inhibitors was seen, suggesting caspaseindependent cell death. Notably, this is in agreement with previous studies showing that siramesine causes caspase-independent cell death of various transformed cell lines [19]. While no effect of caspase inhibition was observed, inhibition of cysteine cathepsins significantly prevented cell death. This is clearly in line with numerous reports implicating cysteine cathepsins in apoptotic cell death (reviewed in [30]). Cysteine cathepsins are localized to the lysosomal compartment and are also found in MC secretory granules. A role for these enzymes in siramesine-induced cell death thus implies that lysosomal/granule compartments are damaged in response to siramesine. In support of this notion, siramesine treatment caused reduced AO fluorescence, a sign of lost lysosomal/granular integrity. Moreover, mMCP-6 and CPA3 were recovered in the cytosol of siramesine-treated MCs but were absent in the cytosol of untreated cells. Since both of these MCspecific proteases are localized to secretory granules, these findings argue strongly that siramesine indeed causes permeabilization of the MC secretory granules, followed by release of granule content into the cytosol. A striking finding was that, whereas WT MCs underwent apoptotic cell death in response to siramesine, cells lacking serglycin died predominantly by necrosis. Serglycin is a main component of MC secretory granules, and previous work has shown that serglycin holds a key role in maintaining and regulating secretory granule homeostasis by promoting the storage of a variety of granule compounds, including bioactive monoamines and different MCspecific proteases [8]. Clearly, the strong apoptosis-promoting effect of serglycin further reinforces the crucial role of secretory granule permeabilization in cell death caused by siramesine. Notably, serglycin has also been shown to promote apoptosis vs. necrosis in response various other cytotoxic agents [26,32]. An important aspect of this investigation was to evaluate whether siramesine has the potential to reduce MC populations in vivo. Indeed, i.p. injection of mice with siramesine depleted the peritoneal MC population, indicating that siramesine has cytotoxic activity toward MCs at the site of injection. It was observed that the i.p. administration of siramesine caused moderate signs of inflammation, as evidenced by increased numbers of macrophages and other inflammatory cells. Possibly, this finding could be explained by pro-inflammatory activity of siramesine, which is not related to its effects on MCs. Alternatively, the moderate inflammation may be caused by factors that are released in processes accompanying the apoptosis of MCs. For example, it has been demonstrated that the clearance of apoptotic bodies can be associated with release of monocyte and neutrophil attractants [42]. Potentially, this may aid to recruit additional phagocytic cells, thereby enhancing the efficiency of clearing the site from apoptotic bodies. As shown by the significant reductions of the MC populations in skin, siramesine has also systemic effects on MC populations distant from the site of administration. Importantly, whereas the
1679
peritoneal MC population was completely depleted, other resident cell populations of the peritoneum, i.e. macrophages and lymphocytes, were not reduced by siramesine in vivo. Hence, these findings suggest that siramesine shows selectivity for MCs over other leukocyte populations in vivo, introducing the possibility of using siramesine (or compounds with similar properties) in therapy aiming at counteracting harmful actions of MCs. However, the ability of siramesine to ameliorate MCdependent diseases, in both murine models and in humans, remains to be investigated. Acknowledgments We are grateful to Lundbeck A/S for providing siramesine and to Rene´ Holm (Lundbeck A/S) for helpful advice. This work was supported by grants from: The Swedish Research Council, The Swedish Cancer Foundation and King Gustaf V’s 80-year Anniversary Fund.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bcp.2012.09.028. References [1] Bradding P, Walls AF, Holgate ST. The role of the mast cell in the pathophysiology of asthma. J Allergy Clin Immunol 2006;117:1277–84. [2] Galli SJ, Tsai M. IgE and mast cells in allergic disease. Nat Med 2012;18: 693–704. [3] Dudeck A, Dudeck J, Scholten J, Petzold A, Surianarayanan S, Kohler A, et al. Mast cells are key promoters of contact allergy that mediate the adjuvant effects of haptens. Immunity 2011;34:973–84. [4] Soucek L, Lawlor ER, Soto D, Shchors K, Swigart LB, Evan GI. Mast cells are required for angiogenesis and macroscopic expansion of Myc-induced pancreatic islet tumors. Nat Med 2007;13:1211–8. [5] Sun J, Sukhova GK, Wolters PJ, Yang M, Kitamoto S, Libby P, et al. Mast cells promote atherosclerosis by releasing proinflammatory cytokines. Nat Med 2007;13:719–24. [6] Liu J, Divoux A, Sun J, Zhang J, Clement K, Glickman JN, et al. Genetic deficiency and pharmacological stabilization of mast cells reduce diet-induced obesity and diabetes in mice. Nat Med 2009;15:940–5. [7] Metcalfe DD, Baram D, Mekori YA. Mast cells. Physiol Rev 1997;77:1033–79. [8] Kolset SO, Pejler G. Serglycin-A structural and functional chameleon with wide impact on immune cells. J Immunol 2011;187:4927–33. [9] Lundequist A, Pejler G. Biological implications of preformed mast cell mediators. Cell Mol Life Sci 2011;68:965–75. [10] Pejler G, Ro¨nnberg E, Waern I, Wernersson S. Mast cell proteases: multifaceted regulators of inflammatory disease. Blood 2010;115:4981–90. [11] Rivera J, Fierro NA, Olivera A, Suzuki R. New insights on mast cell activation via the high affinity receptor for IgE. Adv Immunol 2008;98:85–120. [12] Boyce JA. Mast cells and eicosanoid mediators: a system of reciprocal paracrine and autocrine regulation. Immunol Rev 2007;217:168–85. [13] Galli SJ, Grimbaldeston M, Tsai M. Immunomodulatory mast cells: negative, as well as positive, regulators of immunity. Nat Rev Immunol 2008;8:478–86. [14] Peachell P. Regulation of mast cells by beta-agonists. Clin Rev Allergy Immunol 2006;31:131–42. [15] Takai S, Jin D, Miyazaki M. Targets of chymase inhibitors. Expert Opin Ther Targets 2011;15:519–27. [16] Sommerhoff CP, Schaschke N. Mast cell tryptase beta as a target in allergic inflammation: an evolving story. Curr Pharm Des 2007;13:313–32. [17] Karra L, Berent-Maoz B, Ben-Zimra M, Levi-Schaffer F. Are we ready to downregulate mast cells. Curr Opin Immunol 2009;21:708–14. [18] Heading C. Siramesine H Lundbeck. Curr Opin Investig Drugs 2001;2:266–70. [19] Ostenfeld MS, Fehrenbacher N, Hoyer-Hansen M, Thomsen C, Farkas T, Jaattela M. Effective tumor cell death by sigma-2 receptor ligand siramesine involves lysosomal leakage and oxidative stress. Cancer Res 2005;65:8975–83. [20] Zeng C, Rothfuss J, Zhang J, Chu W, Vangveravong S, Tu Z, et al. Sigma-2 ligands induce tumour cell death by multiple signalling pathways. Br J Cancer 2012;106:693–701. [21] Bird PI, Trapani JA, Villadangos JA. Endolysosomal proteases and their inhibitors in immunity. Nat Rev Immunol 2009;9:871–82. [22] A˚brink M, Grujic M, Pejler G. Serglycin is essential for maturation of mast cell secretory granule. J Biol Chem 2004;279:40897–905. [23] Ro¨nnberg E, Pejler G. Serglycin: the master of the mast cell. Methods Mol Biol 2012;836:201–17.
1680
J. Spirkoski et al. / Biochemical Pharmacology 84 (2012) 1671–1680
[24] Malbec O, Roget K, Schiffer C, Iannascoli B, Dumas AR, Arock M, et al. Peritoneal cell-derived mast cells: an in vitro model of mature serosal-type mouse mast cells. J Immunol 2007;178:6465–75. [25] Lundequist A, Calounova G, Wensman H, Ronnberg E, Pejler G. Differential regulation of Nr4a subfamily nuclear receptors following mast cell activation. Mol Immunol 2011;48:1753–61. [26] Melo FR, Waern I, Ro¨nnberg E, A˚brink M, Lee DM, Schlenner SM, et al. A role for serglycin proteoglycan in mast cell apoptosis induced by a secretory granulemediated pathway. J Biol Chem 2011;286:5423–33. [27] Kirshenbaum AS, Akin C, Wu Y, Rottem M, Goff JP, Beaven MA, et al. Characterization of novel stem cell factor responsive human mast cell lines LAD 1 and 2 established from a patient with mast cell sarcoma/leukemia; activation following aggregation of FcepsilonRI or FcgammaRI. Leuk Res 2003;27:677–82. [28] Furitsu T, Tsujimura T, Tono T, Ikeda H, Kitayama H, Koshimizu U, et al. Identification of mutations in the coding sequence of the proto-oncogene c-kit in a human mast cell leukemia cell line causing ligand-independent activation of c-kit product. J Clin Invest 1993;92:1736–44. [29] Ivanova S, Repnik U, Bojic L, Petelin A, Turk V, Turk B. Lysosomes in apoptosis. Methods Enzymol 2008;442:183–99. [30] Turk B, Turk V. Lysosomes as suicide bags in cell death: myth or reality. J Biol Chem 2009;284:21783–87. [31] Ha HC, Snyder SH. Poly(ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc Natl Acad Sci USA 1999;96:13978–82. [32] Melo FR, Grujic M, Spirkoski J, Calounova G, Pejler G. Serglycin proteoglycan promotes apoptotic versus necrotic cell death in mast cells. J Biol Chem 2012;287:18142–52. [33] Karlberg M, Ekoff M, Huang DC, Mustonen P, Harvima IT, Nilsson G. The BH3mimetic ABT-737 induces mast cell apoptosis in vitro and in vivo: potential for therapeutics. J Immunol 2010;185:2555–62.
[34] Berent-Maoz B, Piliponsky AM, Daigle I, Simon HU, Levi-Schaffer F. Human mast cells undergo TRAIL-induced apoptosis. J Immunol 2006;176:2272–8. [35] Juurikivi A, Sandler C, Lindstedt KA, Kovanen PT, Juutilainen T, Leskinen MJ, et al. Inhibition of c-kit tyrosine kinase by imatinib mesylate induces apoptosis in mast cells in rheumatoid synovia: a potential approach to the treatment of arthritis. Ann Rheum Dis 2005;64:1126–31. [36] Hartmann K, Wagelie-Steffen AL, von Stebut E, Metcalfe DD. Fas (CD95, APO-1) antigen expression and function in murine mast cells. J Immunol 1997;159: 4006–14. [37] Melo FR, Lundequist A, Calounova G, Wernersson S, Pejler G. Lysosomal membrane permeabilization induces cell death in human mast cells. Scand J Immunol 2011;74:354–62. [38] Thiele DL, Lipsky PE. The immunosuppressive activity of L-leucyl-L-leucine methyl ester: selective ablation of cytotoxic lymphocytes and monocytes. J Immunol 1986;136:1038–48. [39] Thiele DL, Lipsky PE. Mechanism of L-leucyl-L-leucine methyl ester-mediated killing of cytotoxic lymphocytes: dependence on a lysosomal thiol protease, dipeptidyl peptidase I, that is enriched in these cells. Proc Natl Acad Sci USA 1990;87:83–7. [40] Cirman T, Oresic K, Mazovec GD, Turk V, Reed JC, Myers RM, et al. Selective disruption of lysosomes in HeLa cells triggers apoptosis mediated by cleavage of Bid by multiple papain-like lysosomal cathepsins. J Biol Chem 2004;279: 3578–87. [41] Droga-Mazovec G, Bojic L, Petelin A, Ivanova S, Romih R, Repnik U, et al. Cysteine cathepsins trigger caspase-dependent cell death through cleavage of bid and antiapoptotic Bcl-2 homologues. J Biol Chem 2008;283:19140–50. [42] Fries DM, Lightfoot R, Koval M, Ischiropoulos H. Autologous apoptotic cell engulfment stimulates chemokine secretion by vascular smooth muscle cells. Am J Pathol 2005;167:345–53.