A Lys49-PLA2 myotoxin of Bothrops asper triggers a rapid death of macrophages that involves autocrine purinergic receptor signaling

Abstract

Lys49-PLA(2) myotoxins, an important component of various viperid snake venoms, are a class of PLA(2)-homolog proteins deprived of catalytic activity. Similar to enzymatically active PLA(2) (Asp49) and to other classes of myotoxins, they cause severe myonecrosis. Moreover, these toxins are used as tools to study skeletal muscle repair and regeneration, a process that can be very limited after snakebites. In this work, the cytotoxic effect of different myotoxins, Bothrops asper Lys49 and Asp49-PLA(2), Notechis scutatus notexin and Naja mossambica cardiotoxin, was evaluated on macrophages, cells that have a key role in muscle regeneration. Only the Lys49-myotoxin was found to trigger a rapid asynchronous death of mouse peritoneal macrophages and macrophagic cell lines through a process that involves ATP release, ATP-induced ATP release and that is inhibited by various purinergic receptor antagonists. ATP leakage is induced also at sublytical doses of the Lys49-myotoxin, it involves Ca(2+) release from intracellular stores, and is reduced by inhibitors of VSOR and the maxi-anion channel. The toxin-induced cell death is different from that caused by high concentration of ATP and appears to be linked to localized purinergic signaling. Based on present findings, a mechanism of cell death is proposed that can be extended to other cytolytic proteins and peptides.

Figure 1

Cell death induced by Mt-I, Mt-II, Ctx and Ntx. Cytotoxicity was measured with the MTS assay on mouse peritoneal macrophages (a) and on macrophagic cell lines RAW264.7 and J774.A1 (b) as a function of the toxin concentration in the medium. Macrophages were incubated with the different toxins in the mKRB medium (see Materials and Methods section) for 1 h, and cell viability was determined. Values are mean±S.D.; n=4–5

Figure 2

ATP release induced by Mt-I, Mt-II, Ctx and Ntx. (a) Mt-II induces a dose-dependent release of ATP from RAW264.7 and J774.A1 cells. Mt-I, Ctx and Ntx (6 μM) do not induce a significant release of ATP. ATP was measured 5 min after intoxication with a luciferin/luciferase-based assay. (b) Kinetics of ATP (top panel) and LDH (bottom panel) release from J774.A1 and RAW264.7 cells induced by 0.35 and 1.5 μM Mt-II. Notice that ATP release temporally precedes that of LDH. Values are mean±S.D.; n =4

Figure 3

Characterization of the ATP release induced by Mt-II (1.5 μM) on macrophages. (A) Effect of bafilomycin and inhibitors of ATP channels. ATP present in the extracellular medium was determined 5 min after the intoxication. Values are mean±S.D. of four independent experiments on J774.A1. RAW264.7 cells give identical results (data not shown). Statistically significant differences were determined via an one-tailed Student's t-test (*P<0.01; **P<0.001). Inhibitors were used at the same concentrations indicated in Table 1. Bafilomycin A1 was used at 0.2 μM. (B) Calcium imaging of RAW264.7 cells loaded with fura-2 AM and treated with Mt-II (1.5 μM) in mKRB (top-left panel) or in Ca²⁺-free EGTA (0.2 mM)-containing medium (bottom-left panel). Top- right panel shows an addition of the same volume of mKRB without the toxin to fura-2 loaded cells. Bottom-right panel: fura-2 loaded cells were pretreated, before the intoxication with Mt-II in mKRB, with cyclopiazonic acid (20 μM, CPA) and EGTA (0.6 mM) to deplete intracellular calcium stores. The arrows indicate the time at which the toxin was added. The same results were obtained in at least three independent experiments for each condition. The inserts correspond to pictures of the evaluated cells before the addition of toxin or buffer, to correlate each colored trace to a specific cell. Representative ratios from each picture are shown. Similar results were obtained with J 774 cells (data not shown). (C) Release of ATP induced by Mt-II, 5 min after the intoxication, in cells preloaded with BAPTA-AM (2 μM) or in the presence of EGTA (1.5 mM) in the extracellular medium. Values are mean±S.D. of four independent experiments. Statistically significant differences were determined via an one-tailed Student's t-test (*P<0.01; **P<0.001)

Figure 4

Effect of some purinergic inhibitors on the ATP release induced by Mt-II in J774.A1 cells. Experiments were performed as reported in Figure 3A. All inhibitors were pre-incubated for 30 min. Statistically significant differences were determined via a one-tailed Student's t test (**P<0.001)

Figure 5

Observation by fluorescence microscopy of macrophage death induced by Mt-II (1.5 μM). Left: J744.A1 cells were loaded with the fluorescent dye calcein and intoxicated with Mt-II. Top panels: cell death appears as a rapid burst of cells in an asynchronous, ‘spotted-like' way (see also Supplementaty Video S1). Bottom panels: some cells swell and form characteristic membrane blebs (arrows). Images have been recorded with 10 '' time intervals (see also Supplementaty Video S2). Right panels: RAW264.7 cells intoxicated with Mt-II in the presence of annexin-V-fluorescein. Membrane blebs induced by Mt-II intoxication show surface exposure of phosphatidylserine (see also Supplementary Figures S6 A and B). Cell integrity was checked by staining with propidium iodide. All experiments were repeated at least three times

Figure 6

Effect of apyrase (a), of various purine derivatives (100 μM each) and of caffeine (10 μM) (b) on Mt-II-induced cytotoxicity. Apyrase activity was checked by determining ATP in the extracellular medium of J744.A1 cells incubated with Mt-II (1.5 μM, inset). Values are mean±S.D. of four independent experiments on J774.A1 and RAW264.7 cells

Figure 7

Model of action of Mt-II. The toxin interacts with a yet unidentified receptor (R), embedded in a membrane microdomain. R activates ATP channels by inducing the release of calcium from intracellular stores and/or by other mechanisms. The released ATP stimulates purinergic receptors within the same plasma membrane microdomain inducing a further release of ATP. The extracellular ATP activates a signaling cascade leading to changes in the composition of the plasma membrane, which in turn promotes the insertion of Mt-II into the lipid bilayer with consequent osmotic cell lysis. It is proposed that this hypothesis may be valid also for other membrane-active toxins and peptides