ABSTRACT

Poly(ADP-ribose) polymerase-1 (PARP1) is a nuclear enzyme that can trigger caspase-independent necrosis. Two main mechanisms for this have been proposed: one involving RIP1 and JNK kinases and mitochondrial permeability transition (MPT), the other involving calpain-mediated activation of Bax and mitochondrial release of apoptosis-inducing factor (AIF). However, whether these two mechanisms represent distinct pathways for PARP1-induced necrosis, or whether they are simply different components of the same pathway has yet to be tested. Mouse embryonic fibroblasts (MEFs) were treated with either N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) or β-Lapachone, resulting in PARP1-dependent necrosis. This was associated with increases in calpain activity, JNK activation and AIF translocation. JNK inhibition significantly reduced MNNG- and β-Lapachone-induced JNK activation, AIF translocation, and necrosis, but not calpain activation. In contrast, inhibition of calpain either by Ca2+ chelation or knockdown attenuated necrosis, but did not affect JNK activation or AIF translocation. To our surprise, genetic and/or pharmacological inhibition of RIP1, AIF, Bax and the MPT pore failed to abrogate MNNG- and β-Lapachone-induced necrosis. In conclusion, although JNK and calpain both contribute to PARP1-induced necrosis, they do so via parallel mechanisms.

INTRODUCTION

Apoptosis is regarded as a ‘programmed’ form of cell death due to the fact that there are specific, genetically determined pathways that mediate this process. In contrast, until recently it was believed that necrosis was a random, uncontrolled process that leads to the ‘accidental’ death of the cell. However, recent evidence has indicated that distinct molecular pathways also mediate necrosis (Vanlangenakker et al., 2012). In particular, a great number of recent studies have focused on death-receptor-induced mechanisms of programmed necrosis, or necroptosis. Activation of tumor necrosis factor (TNF) receptors under conditions of caspase-8 inhibition leads to the formation of the necrosome, a complex consisting of the serine/threonine kinases receptor interacting proteins (RIP, or RIPK) 1 and 3 (Cho et al., 2009; He et al., 2009; Zhang et al., 2009; Li et al., 2012). This complex in turn has been proposed to facilitate necrotic killing through a variety of mechanisms, including activation of NADPH oxidases (Kim et al., 2007), induction of mitochondrial reactive oxygen species (Irrinki et al., 2011; Vanlangenakker et al., 2011) and activation of the pseudokinase mixed lineage kinase like (MLKL) with subsequent activation of the mitochondrial-associated phosphatase PGAM5 (Sun et al., 2012; Wang et al., 2012).

In addition to the necroptosis pathway, another necrotic program involving the DNA repair enzyme poly(ADP-ribose) polymerase-1 (PARP1) has also emerged. Genotoxic stresses, such as oxidants and alkylating agents, have long been known to cause necrotic cell death that is associated with an overstimulation of PARP1 (Hassa, 2009). Moreover, ischemia/reperfusion-induced myocardial and cerebral necrosis is markedly attenuated by genetic inhibition of PARP1 (Zingarelli et al., 1998; Li et al., 2010). It was originally thought that necrosis induced by PARP1 hyperactivation was simply due to metabolic catastrophe, where the overactive PARP1 used up the supply of NAD+ in the cell, and subsequently ATP (van Wijk and Hageman, 2005). However, it is now appreciated that PARP1-induced necrosis follows more specific molecular pathways (Xu et al., 2006; Moubarak et al., 2007; Artus et al., 2010; Dong et al., 2010; Chiu et al., 2011; Park et al., 2011).

Despite this, the signaling networks responsible for PARP1-mediated cell death have yet to be clearly defined. Moubarak and colleagues (Moubarak et al., 2007) have reported that PARP1-induced necrosis is dependent on activation of the Ca2+-activated protease calpain, which in turn induces Bax translocation to the mitochondrion where it elicits the release and nuclear translocation of apoptosis-inducing factor (AIF). In contrast, Xu et al. (Xu et al., 2006) have suggested that PARP1 activation activates the pro-necrotic kinase RIP1. RIP1 in turn activates JNK1, which induces necrosis through a mechanism that can be blocked by the mitochondrial permeability transition (MPT) pore inhibitor cyclosporine-A. However, whether these two mechanisms represent distinct pathways for PARP1-induced necrosis, or whether they are simply different components of the same pathway has yet to be tested. Moreover, the majority of studies have relied on only a single pharmacological PARP1 activator, and direct comparisons with another distinct activator are lacking.

Consequently the purpose of the present study was to systematically investigate the roles and functional interrelationships of RIP1, JNK, calpain, Bax and MPT in PARP1-mediated necrosis induced by two distinct PARP1 activators, N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) and β-Lapachone. Although we were able to confirm independent roles for JNK and calpain in PARP1 necrosis, RIP1, Bax, AIF, and MPT appear to be dispensable.

RESULTS

β-Lapachone and MNNG induce PARP1-dependent, caspase-independent necrosis

We first verified that β-Lapachone and MNNG were indeed acting through activation of PARP1. Treatment of wild-type mouse embryonic fibroblasts (MEFs) with both β-Lapachone and MNNG induced substantial cellular necrosis (Fig. 1A). Importantly, the cytotoxic effect of both compounds was ablated in PARP1-deficient MEFs (Fig. 1A). We observed no evidence of caspase-3 cleavage in the treated cells in contrast to treatment with staurosporine, a canonical apoptosis inducer (Fig. 1B), and β-Lapachone- and MNNG-induced cell death was insensitive to the pan-caspase inhibitor benzyloxycarbonyl-Val-Ala-Asp(OMe)-fluoromethylketone (zVAD-FMK) (Fig. 1C). These data confirmed that both β-Lapachone and MNNG induced PARP1-dependent necrosis, rather than apoptosis, in MEFs.

Fig. 1.

β-Lapachone and MNNG induce PARP1-dependent, but caspase-independent, necrosis. (A) Necrosis, as measured by Sytox Green staining, in wild-type (Parp1+/+) and PARP1-deficient (Parp1−/−) MEFs treated with 20 µM β-lapachone or 1 mM MNNG for 4 h. (B) MEFs were treated with 20 µM β-lapachone or 1 mM MNNG for 4 h, and then western blotted for cleaved caspase-3 and GAPDH. Treatment with 300 nM staurosporine for 18 h was used as a positive control. (C) Necrosis, as measured by Sytox Green staining, in MEFs treated with 20 µM β-lapachone or 1 mM MNNG for 4 h with or without the pan-caspase inhibitor zVAD-FMK (20 µM). Results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle; †P<0.05 versus Parp1+/+.

Fig. 1.

β-Lapachone and MNNG induce PARP1-dependent, but caspase-independent, necrosis. (A) Necrosis, as measured by Sytox Green staining, in wild-type (Parp1+/+) and PARP1-deficient (Parp1−/−) MEFs treated with 20 µM β-lapachone or 1 mM MNNG for 4 h. (B) MEFs were treated with 20 µM β-lapachone or 1 mM MNNG for 4 h, and then western blotted for cleaved caspase-3 and GAPDH. Treatment with 300 nM staurosporine for 18 h was used as a positive control. (C) Necrosis, as measured by Sytox Green staining, in MEFs treated with 20 µM β-lapachone or 1 mM MNNG for 4 h with or without the pan-caspase inhibitor zVAD-FMK (20 µM). Results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle; †P<0.05 versus Parp1+/+.

β-Lapachone- and MNNG-induced necrosis is dependent on JNK but not RIP1

As a PARP1–RIP1–JNK pathway has been proposed (Xu et al., 2006; Chiu et al., 2011), we next examined the role of the pro-necrotic signaling kinases RIP1 and JNK in PARP1-mediated necrosis. Transfection of MEFs with a RIP1-specific small interfering (si)RNA reduced RIP1 protein levels by ∼80% (Fig. 2A). In control siRNA-transfected cells β-Lapachone and MNNG induced necrotic death in a dose-dependent manner (Fig. 2B,C). Depletion of RIP1, however, failed to attenuate the cell death effect of either agent (Fig. 2B,C). Pharmacological inhibition of RIP1 with necrostatin-1 similarly failed to attenuate β-Lapachone- and MNNG-induced necrotic death (supplementary material Fig. S1A,B). We repeated these experiments using Ripk1−/− MEFs which are completely devoid of RIP1 (Fig. 2D). However, even the complete absence of RIP1 did not affect the ability of β-Lapachone and MNNG to evoke necrosis (Fig. 2E,F).

Fig. 2.

β-Lapachone- and MNNG-induced necrosis is independent of RIP1. (A) MEFs were transfected with 100 nM of either a control (CONsi) or RIP1-specific (RIP1si) siRNA for 48 h and then western blotted for RIP1. GAPDH was used as a loading control. (B) Necrosis, as measured by Propidium Iodide (PI) staining, in MEFs transfected with control and RIP1 siRNA and treated with increasing concentrations of β-Lapachone for 4 h. (C) Necrosis, as measured by Propidium Iodide staining, in MEFs transfected with control and RIP1 siRNA and treated with increasing concentrations of MNNG for 4 h. (D) Western blotting for RIP1 in wild-type (Ripk1+/+) and RIP1-deficient (Ripk1−/−) 3T3-transformed MEFs. GAPDH was used as a loading control. (E) Percentage of cells displaying necrosis, as measured by PI staining, in Ripk1+/+ and Ripk1−/− MEFs treated with increasing concentrations of β-lapachone for 4 h. (F) Necrosis, as measured by PI staining, in Ripk1+/+ and Ripk1−/− MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of four independent experiments performed in duplicate. Results are mean±s.e.m.

Fig. 2.

β-Lapachone- and MNNG-induced necrosis is independent of RIP1. (A) MEFs were transfected with 100 nM of either a control (CONsi) or RIP1-specific (RIP1si) siRNA for 48 h and then western blotted for RIP1. GAPDH was used as a loading control. (B) Necrosis, as measured by Propidium Iodide (PI) staining, in MEFs transfected with control and RIP1 siRNA and treated with increasing concentrations of β-Lapachone for 4 h. (C) Necrosis, as measured by Propidium Iodide staining, in MEFs transfected with control and RIP1 siRNA and treated with increasing concentrations of MNNG for 4 h. (D) Western blotting for RIP1 in wild-type (Ripk1+/+) and RIP1-deficient (Ripk1−/−) 3T3-transformed MEFs. GAPDH was used as a loading control. (E) Percentage of cells displaying necrosis, as measured by PI staining, in Ripk1+/+ and Ripk1−/− MEFs treated with increasing concentrations of β-lapachone for 4 h. (F) Necrosis, as measured by PI staining, in Ripk1+/+ and Ripk1−/− MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of four independent experiments performed in duplicate. Results are mean±s.e.m.

With regards to JNK signaling, we first examined whether PARP1 activation elicited activation of JNK. Treatment of MEFs with either β-Lapachone or MNNG caused a dose-dependent increase in JNK phosphorylation (Fig. 3A), indicative of activation. Moreover, this was significantly attenuated by SP600125, an inhibitor of JNK signaling (Fig. 3A). Co-incubation with SP600125 was also able to significantly inhibit β-Lapachone- and MNNG-induced cell death (Fig. 3B,C). Thus, unlike RIP1, JNK activation appears to be a crucial step in PARP1-mediated necrosis. We then tested which JNK isoform is involved by knocking down either JNK1 or JNK2 in the MEFs (Fig. 3D). Interestingly, silencing of JNK1 did not greatly affect either β-Lapachone- or MNNG-induced necrosis with only a small reduction observed at the highest concentration of the agents (Fig. 3E,F). In contrast JNK2 knockdown considerably attenuated cell death in response to both compounds suggesting that it is this isoform that plays a causative role in PARP1-mediated necrotic death.

Fig. 3.

β-Lapachone and MNNG-induced necrosis is dependent on JNK activation. (A) Western blotting for phosphorylated JNK (pJNK) and total JNK in MEFs treated with increasing concentrations of β-Lapachone (upper panels) or MNNG (lower panels) for 2 h with or without the JNK inhibitor SP600125 (20 µM). (B) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of β-lapachone for 4 h, with or without 20 µM SP600125. (C) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of MNNG for 4 h, with or without 20 µM SP600125. (D) MEFs were transfected with 100 nM of a control (CONsi), JNK1-specific (JNK1si) or JNK2-specific (JNK2si) siRNAs for 48 h and then western blotted for JNK1/2. GAPDH was used as a loading control. (E) Percentage of cells displaying necrosis, as measured by Sytox Green staining, in control, JNK1 and JNK2 siRNA-transfected MEFs treated with increasing concentrations of β-Lapachone for 4 h. (F) Percentage of cells displaying necrosis, as measured by Sytox Green staining, in control, JNK1 and JNK2 siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle or CONsi.

Fig. 3.

β-Lapachone and MNNG-induced necrosis is dependent on JNK activation. (A) Western blotting for phosphorylated JNK (pJNK) and total JNK in MEFs treated with increasing concentrations of β-Lapachone (upper panels) or MNNG (lower panels) for 2 h with or without the JNK inhibitor SP600125 (20 µM). (B) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of β-lapachone for 4 h, with or without 20 µM SP600125. (C) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of MNNG for 4 h, with or without 20 µM SP600125. (D) MEFs were transfected with 100 nM of a control (CONsi), JNK1-specific (JNK1si) or JNK2-specific (JNK2si) siRNAs for 48 h and then western blotted for JNK1/2. GAPDH was used as a loading control. (E) Percentage of cells displaying necrosis, as measured by Sytox Green staining, in control, JNK1 and JNK2 siRNA-transfected MEFs treated with increasing concentrations of β-Lapachone for 4 h. (F) Percentage of cells displaying necrosis, as measured by Sytox Green staining, in control, JNK1 and JNK2 siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle or CONsi.

β-Lapachone- and MNNG-induced necrosis is dependent on Ca2+ and calpain

In addition to JNK, previous studies have implicated the mobilization of Ca2+ and activation of the Ca2+-dependent protease calpain as key proximal signals in β-Lapachone- and MNNG-induced necrosis (Tagliarino et al., 2003; Moubarak et al., 2007; Dong et al., 2010). Consistent with this, co-treatment with the Ca2+ chelating agent BAPTA-AM significantly attenuated the degree of necrotic cell death in response to β-Lapachone and MNNG (Fig. 4A,B). Calpain activity was also dose-dependently increased by both β-Lapachone and MNNG (Fig. 4C,D). To genetically inhibit the μ- and m-calpains we transfected the MEFs with siRNA against Capn4, the small subunit required for the activity of both isoforms. This siRNA reduced Capn4 levels to ∼35% that of control transfected cells (Fig. 4E) and markedly reduced β-Lapachone- and MNNG-induced necrosis (Fig. 4F,G).

Fig. 4.

β-Lapachone and MNNG-induced necrosis is dependent on Ca2+ and calpain. (A) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of β-Lapachone for 4 h, with or without 1 µM BAPTA-AM. (B) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of MNNG for 4 h, with or without 1 µM BAPTA-AM. (C) Calpain activity in MEFs treated with increasing concentrations of β-Lapachone for 2 h. (D) Calpain activity in MEFs treated with increasing concentrations of MNNG for 2 h. (E) MEFs were transfected with 100 nM of either a control (CONsi) or Capn4-specific (CAPN4si) siRNA for 48 h and then western blotted for Capn4. GAPDH was used as a loading control. (F) Necrosis, as measured by Sytox Green staining, in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of β-lapachone for 4 h. (G) Necrosis, as measured by Sytox Green staining, in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle or CONsi.

Fig. 4.

β-Lapachone and MNNG-induced necrosis is dependent on Ca2+ and calpain. (A) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of β-Lapachone for 4 h, with or without 1 µM BAPTA-AM. (B) Necrosis, as measured by Sytox Green staining, in MEFs treated with increasing concentrations of MNNG for 4 h, with or without 1 µM BAPTA-AM. (C) Calpain activity in MEFs treated with increasing concentrations of β-Lapachone for 2 h. (D) Calpain activity in MEFs treated with increasing concentrations of MNNG for 2 h. (E) MEFs were transfected with 100 nM of either a control (CONsi) or Capn4-specific (CAPN4si) siRNA for 48 h and then western blotted for Capn4. GAPDH was used as a loading control. (F) Necrosis, as measured by Sytox Green staining, in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of β-lapachone for 4 h. (G) Necrosis, as measured by Sytox Green staining, in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle or CONsi.

β-Lapachone- and MNNG-induced necrosis is not dependent on Bax

Having established JNK and Ca2+/calpain as necessary for PARP1-mediated necrosis we next investigated the distal mitochondrial mechanisms. Activation of the Bcl-family protein Bax has been postulated to play a key role in MNNG-induced necrosis, downstream of calpain (Moubarak et al., 2007). We therefore treated MEFs with β-Lapachone and MNNG and determined the level of Bax activation by immunofluorescence staining for the open conformation of the protein (Fig. 5A). Both agents induced an increase in active Bax staining, although not as much as that seen with staurosporine. We then exposed wild-type MEFs and MEFs null for both Bax and Bak (also known as BAK1) to β-Lapachone and MNNG. However, to our surprise the Bax/Bak-deficient cells (Fig. 4B) still exhibited a robust dose-dependent death response to β-Lapachone and MNNG that was of similar magnitude to their wild-type counterparts (Fig. 5C,D). We were concerned that these particular cell lines were transformed and were also lacking Bak, either one of which could have altered the Bax-dependency of the system. We therefore took a parallel approach where we inhibited Bax activation by overexpressing Bcl2. Infection of wild-type MEFs with a Bcl2-encoding adenovirus resulted in a significant increase in Bcl2 protein levels (Fig. 5E) However, similar to the cells that were deficient for both Bax and Bak, β-Lapachone- and MNNG-induced necrosis was not attenuated by Bcl2 overexpression (Fig. 5F,G). Taken together, these data indicate that Bax is not required for PARP1-mediated necrosis in MEFs.

Fig. 5.

β-Lapachone and MNNG-induced necrosis is independent of Bax. (A) Immunostaining for active Bax in MEFs treated with either 20 µM β-lapachone or 1 mM MNNG for 2 h. Treatment with 300 nM staurosporine for 18 h was used as a positive control. Scale bar: 50 µm. (B) Western blotting for Bax in wild-type (Bax+/+/Bak+/+) and Bax−/−/Bak−/− (Bax and Bak deficient) SV40-transformed MEFs. GAPDH was used as a loading control. (C) Necrosis, as measured by Sytox Green staining, in Bax+/+/Bak+/+ and Bax−/−/Bak−/− MEFs treated with increasing concentrations of β-Lapachone for 4 h. (D) Necrosis, as measured by Sytox Green staining, in Bax+/+/Bak+/+ and Bax−/−/Bak−/− MEFs treated with increasing concentrations of MNNG for 4 h. (E) MEFs were infected with adenoviruses encoding for either β-galactosidase (βGal) or Bcl2 for 48 h and then western blotted for Bcl2 and GAPDH. (F) Necrosis, as measured by Sytox Green staining, in βGal- or Bcl2-infected MEFs treated with increasing concentrations of β-Lapachone for 4 h. (G) Necrosis, as measured by Sytox Green staining, in βGal- or Bcl2-infected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus Bax+/+/Bak+/+.

Fig. 5.

β-Lapachone and MNNG-induced necrosis is independent of Bax. (A) Immunostaining for active Bax in MEFs treated with either 20 µM β-lapachone or 1 mM MNNG for 2 h. Treatment with 300 nM staurosporine for 18 h was used as a positive control. Scale bar: 50 µm. (B) Western blotting for Bax in wild-type (Bax+/+/Bak+/+) and Bax−/−/Bak−/− (Bax and Bak deficient) SV40-transformed MEFs. GAPDH was used as a loading control. (C) Necrosis, as measured by Sytox Green staining, in Bax+/+/Bak+/+ and Bax−/−/Bak−/− MEFs treated with increasing concentrations of β-Lapachone for 4 h. (D) Necrosis, as measured by Sytox Green staining, in Bax+/+/Bak+/+ and Bax−/−/Bak−/− MEFs treated with increasing concentrations of MNNG for 4 h. (E) MEFs were infected with adenoviruses encoding for either β-galactosidase (βGal) or Bcl2 for 48 h and then western blotted for Bcl2 and GAPDH. (F) Necrosis, as measured by Sytox Green staining, in βGal- or Bcl2-infected MEFs treated with increasing concentrations of β-Lapachone for 4 h. (G) Necrosis, as measured by Sytox Green staining, in βGal- or Bcl2-infected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus Bax+/+/Bak+/+.

MPT is not required for β-Lapachone and MNNG-induced necrosis

Given that Bax did not appear to play a role in PARP1-driven necrosis, we next examined whether another mitochondrial mediator of necrosis, the MPT pore, was part of the signaling mechanism. We first assessed the induction of MPT using the calcein-Co2+ fluorescence method, whereby a reduction in mitochondrial calcein fluorescence is indicative of MPT. Neither β-Lapachone nor MNNG elicited any significant reduction in calcein fluorescence in wild-type MEFs (Fig. 6A,B), indicating that PARP1-induced necrosis is independent of the MPT pore. To assess this, we knocked down CypD, a key activator of the MPT pore in wild-type MEFs (Fig. 6C), resulting in a ∼85% reduction in protein levels. Depletion of CypD did not affect the ability of either β-Lapachone or MNNG to dose-dependently induce necrosis in wild-type MEFs (Fig. 6D,E). Similar results were obtained in MEFs completely deficient in CypD (Fig. 6F,G) or in MEFs treated with the ANT ligand bongkrekic acid (supplementary material Fig. S2A,B).

Fig. 6.

β-Lapachone and MNNG-induced necrosis is independent of mitochondrial permeability transition. (A) Mitochondrial permeability transition (MPT) determined by calcein/CoCl2 fluorescence in MEFs exposed to vehicle or 20 µM β-lapachone for 2 h. (B) MPT determined by calcein/CoCl2 fluorescence in MEFs exposed to vehicle or 1 mM MNNG for 2 h. Scale bars: 50 µm. (C) MEFs were transfected with 100 nM of either a control (CONsi) or CypD-specific (CypDsi) siRNA for 48 h and then western blotted for CypD. GAPDH was used as a loading control. (D) Necrosis, as measured by Propidium Iodide (PI) staining, in control and RIP1 siRNA-transfected MEFs treated with increasing concentrations of β-lapachone for 4 h. (E) Necrosis, as measured by PI staining, in control and CypD siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. (F) Western blotting for CypD in wildtype (Ppif+/+) and CypD-deficient (Ppif−/−) MEFs. GAPDH was used as a loading control. (G) Necrosis, as measured by PI staining, in Ppif+/+ and Ppif−/− MEFs treated with increasing concentrations of β-Lapachone for 4 h. (H) Necrosis, as measured by PI staining, in Ppif+/+ and Ppif−/− MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus Ppif+/+.

Fig. 6.

β-Lapachone and MNNG-induced necrosis is independent of mitochondrial permeability transition. (A) Mitochondrial permeability transition (MPT) determined by calcein/CoCl2 fluorescence in MEFs exposed to vehicle or 20 µM β-lapachone for 2 h. (B) MPT determined by calcein/CoCl2 fluorescence in MEFs exposed to vehicle or 1 mM MNNG for 2 h. Scale bars: 50 µm. (C) MEFs were transfected with 100 nM of either a control (CONsi) or CypD-specific (CypDsi) siRNA for 48 h and then western blotted for CypD. GAPDH was used as a loading control. (D) Necrosis, as measured by Propidium Iodide (PI) staining, in control and RIP1 siRNA-transfected MEFs treated with increasing concentrations of β-lapachone for 4 h. (E) Necrosis, as measured by PI staining, in control and CypD siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. (F) Western blotting for CypD in wildtype (Ppif+/+) and CypD-deficient (Ppif−/−) MEFs. GAPDH was used as a loading control. (G) Necrosis, as measured by PI staining, in Ppif+/+ and Ppif−/− MEFs treated with increasing concentrations of β-Lapachone for 4 h. (H) Necrosis, as measured by PI staining, in Ppif+/+ and Ppif−/− MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus Ppif+/+.

β-Lapachone and MNNG induce mitochondrial–nuclear AIF translocation through JNK but not Ca2+/calpain

The translocation of AIF from the mitochondrial inter-membrane space to the nucleus is thought to be the pivotal event in PARP1-induced necrotic death (Xu et al., 2006; Moubarak et al., 2007; Artus et al., 2010; Chiu et al., 2011; Park et al., 2011). Immunocytochemistry revealed that both β-Lapachone and MNNG induced the nuclear translocation of AIF (Fig. 7A,B), again in a concentration-dependent manner (Fig. 7C,D). Importantly, the JNK inhibitor SP600125 almost completely blocked the ability of both compounds to induce AIF translocation (Fig. 7A–D), consistent with its protective effects against β-Lapachone and MNNG cytotoxicity. In direct contrast, inhibition of Ca2+/calpain by BAPTA-AM or Capn4 knockdown did not prevent AIF translocation (Fig. 7E,F). We next wanted to confirm that AIF was indeed necessary for both β-Lapachone- and MNNG-induced necrosis. Transfection of MEFs with an AIF-specific siRNA reduced AIF levels by ∼90% (Fig. 8A). However, very much to our surprise this had no effect on the ability of either β-Lapachone or MNNG to induce cellular necrosis (Fig. 8B,C).

Fig. 7.

β-Lapachone- and MNNG-induced AIF translocation is dependent on JNK but not Ca2+/calpain. (A) Immunostaining for AIF in MEFs treated with 20 µM β-Lapachone for 2 h, with or without 20 µM SP600125. (B) Immunostaining for AIF in MEFs treated with 1 mM MNNG for 2 h, with or without 20 µM SP600125. Scale bars: 50 µm. (C) Quantification of AIF translocation in MEFs treated with increasing concentrations of β-lapachone for 2 h, with or without 20 µM SP600125. (D) Quantification of AIF translocation in MEFs treated with increasing concentrations of MNNG for 2 h, with or without 20 µM SP600125. (E) Quantification of AIF translocation in MEFs treated with increasing concentrations of β-lapachone for 2 h, with or without 1 µM BAPTA-AM. (F) Quantification of AIF translocation in MEFs treated with increasing concentrations of MNNG for 2 h, with or without 1 µM BAPTA-AM. (G) Quantification of AIF translocation in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of β-lapachone for 2 h. (H) Quantification of AIF translocation in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of MNNG for 2 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle.

Fig. 7.

β-Lapachone- and MNNG-induced AIF translocation is dependent on JNK but not Ca2+/calpain. (A) Immunostaining for AIF in MEFs treated with 20 µM β-Lapachone for 2 h, with or without 20 µM SP600125. (B) Immunostaining for AIF in MEFs treated with 1 mM MNNG for 2 h, with or without 20 µM SP600125. Scale bars: 50 µm. (C) Quantification of AIF translocation in MEFs treated with increasing concentrations of β-lapachone for 2 h, with or without 20 µM SP600125. (D) Quantification of AIF translocation in MEFs treated with increasing concentrations of MNNG for 2 h, with or without 20 µM SP600125. (E) Quantification of AIF translocation in MEFs treated with increasing concentrations of β-lapachone for 2 h, with or without 1 µM BAPTA-AM. (F) Quantification of AIF translocation in MEFs treated with increasing concentrations of MNNG for 2 h, with or without 1 µM BAPTA-AM. (G) Quantification of AIF translocation in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of β-lapachone for 2 h. (H) Quantification of AIF translocation in control and Capn4 siRNA-transfected MEFs treated with increasing concentrations of MNNG for 2 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. *P<0.05 versus vehicle.

Fig. 8.

AIF is dispensable for β-Lapachone- and MNNG-induced necrosis. (A) MEFs were transfected with 100 nM of either a control (CONsi) or AIF-specific (AIFsi) siRNA for 48 h and then western blotted for AIF. GAPDH was used as a loading control. (B) Necrosis, as measured by Sytox Green staining, in control and AIF siRNA-transfected MEFs treated with increasing concentrations of β-Lapachone for 4 h. (C) Necrosis, as measured by Sytox Green staining, in control and AIF siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of six independent experiments performed in duplicate. Results are mean±s.e.m.

Fig. 8.

AIF is dispensable for β-Lapachone- and MNNG-induced necrosis. (A) MEFs were transfected with 100 nM of either a control (CONsi) or AIF-specific (AIFsi) siRNA for 48 h and then western blotted for AIF. GAPDH was used as a loading control. (B) Necrosis, as measured by Sytox Green staining, in control and AIF siRNA-transfected MEFs treated with increasing concentrations of β-Lapachone for 4 h. (C) Necrosis, as measured by Sytox Green staining, in control and AIF siRNA-transfected MEFs treated with increasing concentrations of MNNG for 4 h. The results shown are representative of six independent experiments performed in duplicate. Results are mean±s.e.m.

β-Lapachone and MNNG induce JNK and Ca2+/calpain activation through independent pathways

The fact that JNK, but not calpain, inhibition blocked AIF translocation suggested that these mediators act in parallel rather than in series. We therefore examined whether JNK inhibition could block calpain activation. JNK inhibition, however, did not attenuate calpain activity induced by either β-Lapachone or MNNG (Fig. 9A,B). Conversely, inhibition of the Ca2+/calpain pathway by Capn4 knockdown failed to block β-Lapachone- and MNNG-induced JNK phosphorylation (Fig. 9C). Finally, we tested whether inhibition of JNK2 and calpain had additive protective effects against PARP1-mediated necrosis. Knockdown of JNK2 had the predicted reduction in cell death induced by β-Lapachone and MNNG (Fig. 9D,E), as shown above. However, this protection was significantly enhanced with the additional knockdown of Capn4 (Fig. 9D,E), again indicating that these pathways are operating in parallel.

Fig. 9.

JNK and calpain activation are independent of one another. (A) Calpain activity in MEFs treated with increasing concentrations of β-Lapachone for 2 h, with or without 20 µM SP600125. (B) Calpain activity in MEFs treated with increasing concentrations of MNNG for 2 h, with or without 20 µM SP600125. (C) Western blotting for phosphorylated JNK (pJNK) and total JNK in control (CONsi) and Capn4 (CAPN4si) siRNA-transfected MEFs treated with increasing concentrations of β-Lapachone (upper panels) or MNNG (lower panels) for 2 h. (D) Necrosis, as measured by Sytox Green staining, in MEFs transfected with control, JNK2 or JNK2+Capn4 siRNAs and then treated with increasing concentrations of β-Lapachone for 4 h. (E) Necrosis, as measured by Sytox Green staining, in MEFs transfected with control, JNK2, or JNK2+Capn4 siRNAs and then treated with increasing concentrations of MNNG for 4 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. (F) Schematic illustrating the proposed parallel JNK2 and Ca2+/calpain pathways mediating PARP1-induced necrosis.

Fig. 9.

JNK and calpain activation are independent of one another. (A) Calpain activity in MEFs treated with increasing concentrations of β-Lapachone for 2 h, with or without 20 µM SP600125. (B) Calpain activity in MEFs treated with increasing concentrations of MNNG for 2 h, with or without 20 µM SP600125. (C) Western blotting for phosphorylated JNK (pJNK) and total JNK in control (CONsi) and Capn4 (CAPN4si) siRNA-transfected MEFs treated with increasing concentrations of β-Lapachone (upper panels) or MNNG (lower panels) for 2 h. (D) Necrosis, as measured by Sytox Green staining, in MEFs transfected with control, JNK2 or JNK2+Capn4 siRNAs and then treated with increasing concentrations of β-Lapachone for 4 h. (E) Necrosis, as measured by Sytox Green staining, in MEFs transfected with control, JNK2, or JNK2+Capn4 siRNAs and then treated with increasing concentrations of MNNG for 4 h. The results shown are representative of three or four independent experiments performed in duplicate. Results are mean±s.e.m. (F) Schematic illustrating the proposed parallel JNK2 and Ca2+/calpain pathways mediating PARP1-induced necrosis.

DISCUSSION

In the present study, we thoroughly tested the roles of the various proposed mediators of PARP1-induced necrosis (RIP1, JNK, calpain, Bax, MPT and AIF) and how they might or might not be functionally coupled to one another. Our findings demonstrate that (1) activation of the stress kinase JNK is required for PARP1-induced cell necrosis, and (2) that activation of the protease calpain activation is also crucial for cell death but that it is acting in parallel to the JNK pathway. However, unlike previous studies, we could find no role for RIP1, Bax, MPT and AIF in necrosis induced by the either of the PARP1 activators. Importantly, we obtained the same results whether we used MNNG or β-Lapachone, two distinct PARP1 activators. This would indicate that their common ability to activate PARP1 is what is responsible for their cytotoxic effects. Indeed, ablation of Parp1, the PARP1-encoding gene, effectively abolished death induced by either compound.

Xu et al. (Xu et al., 2006) have proposed a RIP1–JNK–MPT–AIF sequence of events as being responsible for necrosis induced by MNNG, and hence PARP1. A more recent study has also implicated RIP1 and JNK as being part of this process (Chiu et al., 2011). Similarly, JNK activation has been reported as being crucial for β-Lapachone-induced cell death (Shiah et al., 1999; Park et al., 2011), although whether RIP1 is also involved in β-Lapachone's effects has never been tested until the present study. Despite this, we were unable to recapitulate these data in 3T3-transformed Ripk1−/− MEFs. Of course it is possible that the transformation procedure altered the pathways, such that RIP1 became dispensable, or the chronic loss of RIP1 was overcome by compensatory upregulation of another protein. However, acute inhibition of RIP1, either with siRNA or the RIP1 inhibitor necrostatin also failed to prevent MNNG-induced necrosis. Similarly negative results were obtained with β-Lapachone. One possibility is that PARP1 is actually downstream of RIP1 in the RIP1–RIP3-dependent necroptotic pathway as has been recently reported for TRAIL-induced necrosis (Jouan-Lanhouet et al., 2012). However, we have found that the protective effects of RIP1 and PARP1 inhibition against oxidative-stress-induced necrosis are additive (data not shown), suggesting that they do in fact represent separate necrotic pathways. Consistent with our results, Sosna et al. found that inhibition of PARP1 fails to block TNF-induced necrosis and that inhibition of RIP1 or RIP3 could not prevent MNNG-induced necrosis (Sosna et al., 2014).

In contrast to RIP1, we did confirm that JNK is required for both MNNG- and β-Lapachone-induced AIF translocation and cytotoxicity. Both compounds dose-dependently activated JNK and induced AIF translocation, and pharmacological JNK inhibition was efficient in blocking both AIF translocation and cell necrosis elicited by the PARP1 activators. Xu et al. (Xu et al., 2006) found that JNK1-deficient cells were resistant to MNNG. In contrast, another study indicated that it was JNK2 rather than JNK1 that was important (Park et al., 2011). Using siRNA we found that it was the JNK2 isoform that predominantly contributed to both MNNG- and β-Lapachone-induced necrosis, with only a minor contribution from JNK1. The exact mechanisms by which PARP1 activity leads to JNK activation is not entirely clear. MNNG-induced JNK activation is dependent on MEKK1 and MKK7 (Parra et al., 2000), and dominant-negative MEKK1 could also block cell death elicited by β-Lapachone (Shiah et al., 1999). This would indicate that the PARP1 activators are acting at some point much further upstream of JNK. Whether this involves direct ADP-ribosylation of MEKK1 or whether other more indirect mechanisms are involved is unknown.

As the last step in the proposed PARP1–RIP1–JNK1 pathway, Xu et al. indicated that MPT was the process ultimately responsible for the release of AIF (Xu et al., 2006). Other studies have also suggested such a role for MPT in PARP1-induced AIF translocation and necrosis (Alano et al., 2004), and indeed ablation of CypD, a major regulatory component of the MPT pore, renders cells resistant to necrotic stimuli (Baines et al., 2005; Nakagawa et al., 2005; Schinzel et al., 2005). Dodoni et al. have also indicated that MNNG could directly induce MPT in isolated mitochondria, independent of PARP1 (Dodoni et al., 2004). However, we found no evidence that MPT is a necessary step in either MNNG- or β-Lapachone-induced necrosis and neither chronic (knockout) nor acute (knockdown) deletion of CypD affected cell death in response to these agents. One concern was that the PARP1 activators were simply bypassing the CypD-dependent step of MPT pore opening. However, treatment with the ANT inhibitor bongkrekic acid, which also inhibits MPT, similarly failed to block PARP1-mediated necrosis. This is consistent with other previous studies demonstrating that the MPT pore inhibitor cyclosporine-A and CypD deletion are both ineffective in blocking PARP1-mediated necrosis (Liu et al., 2002; Loor et al., 2010; Abeti et al., 2011).

The role of calpain in mediating PARP1-induced necrosis is somewhat controversial. First proposed by Susin's group as being essential for MNNG-induced AIF translocation and necrosis (Moubarak et al., 2007), several papers have reported a crucial role for calpain in mediating PARP1-induced cytotoxicity (Tagliarino et al., 2003; Vosler et al., 2009; Chiu et al., 2011). This is consistent with findings that chelation of Ca2+ can abrogate the necrosis elicited by MNNG and β-Lapachone (Tagliarino et al., 2001; Bentle et al., 2006; Chiu et al., 2011). However, a report from Dawson's group has indicated that, although MNNG could indeed elevate calpain activity, its toxic effects were not abrogated in MEFs lacking the small regulatory subunit of m- and μ-calpains (Wang et al., 2009). In the present study, we found that both MNNG and β-Lapachone dose-dependently increased calpain activity in MEFs. Importantly, BAPTA could significantly reduce both MNNG- and β-Lapachone-induced necrosis, a finding that was phenocopied by knocking down the small calpain subunit. Thus, at least in our hands, activation of the Ca2+/calpain system does indeed appear to be necessary for PARP1-induced necrosis. However, Ca2+ chelation had no effect on mitochondrial-to-nuclear AIF translocation evoked by either of the PARP1 activators. This suggests that, unlike JNK, activation of the Ca2+/calpain system is dispensable for AIF release and translocation. These results are in contrast to previous reports and we are not sure why our results are different. We were also unable to block calpain activation with the JNK inhibitor; conversely inhibition of calpain did not affect JNK activation. Taken together, these data would indicate that although both JNK and calpain are important mediators of PARP1-induced necrosis, they are acting in parallel rather than in series. Consistent with this notion was the finding that protection offered by knockdown of JNK2 and Capn4 was additive.

Despite corroborating a role for JNK and calpain in PARP1-induced necrosis, we could not confirm a role for Bax. We found that both MNNG and β-Lapachone induced a mild activation of Bax in the MEFs. However, inhibition of Bax, either through gene targeting or overexpression of Bcl2, did not affect the ability of the PARP1 activators to evoke necrotic death. The gene-targeted MEFs also lacked Bak, thus ruling out any compensation by this related protein. Consistent with this finding we also found that JNK inhibition with SP600125, which blocked cell death, did not block activation of Bax by MNNG and β-Lapachone (data not shown), indicating that the two events were independent. This is interesting as PARP1 activators might be able to induce Bax independently of PARP1 activation (Byun et al., 2009). Alternatively, calpain activation might have induced Bax activation in our cells, but this was not required for the death effects of calpain. Either way these data, coupled with those ruling out a role for MPT, raise the question of how JNK is inducing the release of AIF from the mitochondria in response to either MNNG or β-Lapachone. It is plausible that other Bcl2 family members might be responsible. For example, Bmf and BNIP3 have recently been proposed as mediators of TNF-induced necrosis (Hitomi et al., 2008; Kim et al., 2011), and BNIP3 can induce mitochondrial permeabilization in an MPT-independent manner (Quinsay et al., 2010). Moreover, both of these proteins are known to be activated and/or upregulated by JNK (Lei and Davis, 2003; Chaanine et al., 2012). Alternatively, ceramide channels might be involved, as ceramide can induce AIF efflux from the mitochondria (Scharstuhl et al., 2009) and an inhibitor of ceramide synthesis can block MNNG-induced death in MEFs (Yang and Duerksen-Hughes, 2001).

All the above is predicated on the concept that AIF release and nuclear translocation is a prerequisite for PARP1-driven cell necrosis. However, very much to our surprise was the finding that depletion of AIF was not sufficient to block either MNNG- or β-Lapachone-induced cell death. We actually tried two different siRNAs and obtained similar results each time (not shown). The reason for this is not clear. One possibility is that it might be that AIF is required for the chronic cytotoxic effects of the PARP1 activators but is not necessary for the more acute effects observed in our study. Alternatively, a more technical reason for this discrepancy and the others noted in our study is our use of primary cultured cells versus immortalized ones. It might be that the signaling pathways are altered in the transformed cells such that, for example, RIP1 and AIF become involved in the toxic effects of PARP1 activation, whereas they are dispensable in non-transformed cells. This would in fact be very interesting as it would potentially indicate that PARP1-driven signaling is different in cancer cells, a fact that could be exploited therapeutically. Finally it is conceivable that the knockdowns of the various proteins in our study, although efficient, left sufficient protein available to mediate necrosis.

In summary, we found that RIP1, AIF, Bax and MPT are dispensable for acute PARP1-induced necrotic cell death. Instead it appears that both JNK and calpain act in parallel to mediate the necrotic effects of PARP1 activation. Thus, dual targeting of both of these enzymes is more likely to be of clinical benefit than targeting each one individually.

MATERIALS AND METHODS

Reagents

Propidium Iodide, Sytox Green and Lipofectamine RNAiMAX were purchased from Life Technologies; Dulbecco's modified Eagle's medium (DMEM), Hanks's buffered saline solution (HBSS), penicillin/streptomycin and fetal bovine serum (FBS) were purchased from Hyclone; β-Lapachone, SP600125, necrostatin-1, bongkrekic acid and BAPTA-AM (acetoxymethyl ester) were purchased from Axxora; zVAD-FMK and the calpain-GLO activity assay were purchased from Promega; MNNG was from Chem Service. All other chemicals and reagents were from Sigma-Aldrich.

Cell culture

All experiments involving the harvesting of mouse tissues and embryos were approved by the University of Missouri-Columbia Animal Care and Use Committee and conformed to the NIH guidelines for the use and care of animals. Primary cultured wild-type, Ppif−/−, and Parp1−/− MEFs were obtained from E13.5–15.5 embryos by trypsin digestion as previously described (Baines et al., 2005; McGee and Baines, 2011). The 3T3-transformed wild-type and Ripk1−/− MEFs were a generous gift from Michelle Kelliher (University of Massachusetts Medical School, Worcester, MA). The SV40-transformed wild-type and Bax−/−/Bak−/− MEFs were a generous gift from the late Stanley Korsmeyer. Cells were maintained in DMEM medium supplemented with 10% (v/v) FBS, penicillin (100 U/ml) and streptomycin (0.1 mg/ml).

Adenovirus and siRNAs

Replication-deficient adenovirus for Bcl2 was a generous gift from Lorrie Kirshenbaum (St-Boniface Hospital Research, Winnipeg, Manitoba, Canada). MEFs were infected with the adenovirus at a multiplicity of infection (MOI) of 30 plaque-forming units for 48 h. The siRNA duplexes against mouse RIP1 (5′-GGGCCAGUAGCAGAUGACCUCAUAA-3′), CypD (5′-UCACAGACUGUGGCCAGUUGAGUA-3′), Capn4 (5′-ACUGACCGAUCAGGGACUAUCGGUA-3′), and AIF (5′-GGGAUGCUGCAUGCUUCUAUGAUAU-3′), as well as a non-specific control siRNA, were obtained from Life Technologies. Pools of siRNA duplexes against JNK1 (5′-GAAACUACAACCAACAGUA-3′, 5′-GAUUGGAGAUUCUACAUUC-3′, 5′-GGAGUUAGAUCAUGAAAGA-3′, 5′-GGAAAGAACUGAUAUACAA-3′) and JNK2 (5′-GAAGUUAAGUCGUCCUUUU-3′, 5′-ACUAAUGGAUGCUAACUUA-3′, 5′-GCAUUCAGCUGGUAUCAUU-3′, 5′-GGAAAGAGCUAAUUUACAA-3′) were obtained from Dharmacon. MEFs were transfected with 100 nM siRNA using Lipofectamine RNAiMAX and then cultured for 48 h prior to the various experiments.

Fluorescence assays

Necrosis was determined by Propidium Iodide or Sytox Green exclusion, as previously described (Baines et al., 2005; McGee and Baines, 2011). Briefly, treated cells on 12-well plates were incubated with either Propidium Iodide or Sytox Green in HBSS for 30 min at 37°C. Cells were then fixed in 4% paraformaldehyde and counterstained with bis-benzamide to stain all nuclei. The resultant fluorescence images were collected using an inverted fluorescence microscope (Olympus IX51) connected to a digital camera. For the determination of MPT, MEFs were incubated with 1 µM calcein-AM plus 1 mM CoCl2 in HBSS for 30 min at 37°C. Dye-loaded cells were then washed twice with HBSS and fluorescence images collected using an inverted fluorescence microscope (Olympus IX51) connected to a digital camera.

Calpain activity assay

Calpain activity, both at baseline and in response to 2 mM CaCl2, was determined using the Calpain-GLO luminescence activity assay from Promega. Activity was calculated as the difference between the basal and Ca2+-induced activities.

Immunocytochemistry

β-Lapachone- or MNNG-treated MEFs were fixed with 4% paraformaldehyde, blocked, and stained either with an antibody against AIF (BD Biosciences, 1∶500) or an antibody that recognizes the active form of Bax (BD Biosciences clone 6A7, 1∶100). Cells were then incubated with the respective FITC- or TRITC-conjugated secondary antibody (Alexa Invitrogen, 1∶500). Images of the cells were collected with an inverted fluorescence microscope (Olympus IX51) connected to a digital camera.

Western blotting

Cells were lysed in buffer containing 150 mM NaCl, 10 mM Tris-HCl pH 7.4, 1 mM EDTA, and 1% Triton X-100. Proteins were resolved by SDS-PAGE using 10–15% acrylamide, transferred onto PVDF membranes, and blotted using the following commercially available antibodies against the following proteins: RIP and AIF from BD Biosciences; phospho-JNK from Promega; JNK and cleaved caspase-3 from Cell Signaling; Bax and Bcl2 from Santa Cruz Biotechnology; Capn4 and GAPDH from Millipore; and CypD from Abcam. Membranes were then incubated with the appropriate alkaline-phosphatase-linked secondary antibody (Santa Cruz Biotechnology) and visualized by enhanced chemifluorescence (Amersham).

Statistical analyses

Statistical significance was calculated by the Student's t-test. P<0.05 was considered statistically significant.

Acknowledgements

We thank Kathryn Crombie (University of Missouri-Columbia, Columbia, MO) for excellent technical assistance. We also thank Michelle Kelliher, Stanley Korsmeyer and Lorrie Kirshenbaum for their generous gifts of the Ripk1−/− MEFs, Bax−/−/Bak−/− MEFs, and Bcl2 adenovirus, respectively.

Author contributions

C.P.B. conceived and designed the experiments. D.L.D. and C.P.B. conducted the experiments and analyzed the data. C.P.B. wrote the manuscript.

Funding

This work was supported by National Institutes of Health [grant numbers HL092327, HL094404, HL094404 to C.P.B.]. Deposited in PMC for release after 12 months.

References

Abeti
R.
,
Abramov
A. Y.
,
Duchen
M. R.
(
2011
).
β-amyloid activates PARP causing astrocytic metabolic failure and neuronal death.
Brain
134
,
1658
1672
.
Alano
C. C.
,
Ying
W.
,
Swanson
R. A.
(
2004
).
Poly(ADP-ribose) polymerase-1-mediated cell death in astrocytes requires NAD+ depletion and mitochondrial permeability transition.
J. Biol. Chem.
279
,
18895
18902
.
Artus
C.
,
Boujrad
H.
,
Bouharrour
A.
,
Brunelle
M. N.
,
Hoos
S.
,
Yuste
V. J.
,
Lenormand
P.
,
Rousselle
J. C.
,
Namane
A.
,
England
P.
et al. (
2010
).
AIF promotes chromatinolysis and caspase-independent programmed necrosis by interacting with histone H2AX.
EMBO J.
29
,
1585
1599
.
Baines
C. P.
,
Kaiser
R. A.
,
Purcell
N. H.
,
Blair
N. S.
,
Osinska
H.
,
Hambleton
M. A.
,
Brunskill
E. W.
,
Sayen
M. R.
,
Gottlieb
R. A.
,
Dorn
G. W.
et al. (
2005
).
Loss of cyclophilin D reveals a critical role for mitochondrial permeability transition in cell death.
Nature
434
,
658
662
.
Bentle
M. S.
,
Reinicke
K. E.
,
Bey
E. A.
,
Spitz
D. R.
,
Boothman
D. A.
(
2006
).
Calcium-dependent modulation of poly(ADP-ribose) polymerase-1 alters cellular metabolism and DNA repair.
J. Biol. Chem.
281
,
33684
33696
.
Byun
J. Y.
,
Kim
M. J.
,
Eum
D. Y.
,
Yoon
C. H.
,
Seo
W. D.
,
Park
K. H.
,
Hyun
J. W.
,
Lee
Y. S.
,
Lee
J. S.
,
Yoon
M. Y.
et al. (
2009
).
Reactive oxygen species-dependent activation of Bax and poly(ADP-ribose) polymerase-1 is required for mitochondrial cell death induced by triterpenoid pristimerin in human cervical cancer cells.
Mol. Pharmacol.
76
,
734
744
.
Chaanine
A. H.
,
Jeong
D.
,
Liang
L.
,
Chemaly
E. R.
,
Fish
K.
,
Gordon
R. E.
,
Hajjar
R. J.
(
2012
).
JNK modulates FOXO3a for the expression of the mitochondrial death and mitophagy marker BNIP3 in pathological hypertrophy and in heart failure.
Cell Death Dis.
3
,
265
.
Chiu
L. Y.
,
Ho
F. M.
,
Shiah
S. G.
,
Chang
Y.
,
Lin
W. W.
(
2011
).
Oxidative stress initiates DNA damager MNNG-induced poly(ADP-ribose)polymerase-1-dependent parthanatos cell death.
Biochem. Pharmacol.
81
,
459
470
.
Cho
Y. S.
,
Challa
S.
,
Moquin
D.
,
Genga
R.
,
Ray
T. D.
,
Guildford
M.
,
Chan
F. K.
(
2009
).
Phosphorylation-driven assembly of the RIP1-RIP3 complex regulates programmed necrosis and virus-induced inflammation.
Cell
137
,
1112
1123
.
Dodoni
G.
,
Canton
M.
,
Petronilli
V.
,
Bernardi
P.
,
Di Lisa
F.
(
2004
).
Induction of the mitochondrial permeability transition by the DNA alkylating agent N-methyl-N′-nitro-N-nitrosoguanidine. Sorting cause and consequence of mitochondrial dysfunction.
Biochim. Biophys. Acta
1658
,
58
63
.
Dong
Y.
,
Bey
E. A.
,
Li
L. S.
,
Kabbani
W.
,
Yan
J.
,
Xie
X. J.
,
Hsieh
J. T.
,
Gao
J.
,
Boothman
D. A.
(
2010
).
Prostate cancer radiosensitization through poly(ADP-Ribose) polymerase-1 hyperactivation.
Cancer Res.
70
,
8088
8096
.
Hassa
P. O.
(
2009
).
The molecular “Jekyll and Hyde” duality of PARP1 in cell death and cell survival.
Front. Biosci.
14
,
72
111
.
He
S.
,
Wang
L.
,
Miao
L.
,
Wang
T.
,
Du
F.
,
Zhao
L.
,
Wang
X.
(
2009
).
Receptor interacting protein kinase-3 determines cellular necrotic response to TNF-alpha.
Cell
137
,
1100
1111
.
Hitomi
J.
,
Christofferson
D. E.
,
Ng
A.
,
Yao
J.
,
Degterev
A.
,
Xavier
R. J.
,
Yuan
J.
(
2008
).
Identification of a molecular signaling network that regulates a cellular necrotic cell death pathway.
Cell
135
,
1311
1323
.
Irrinki
K. M.
,
Mallilankaraman
K.
,
Thapa
R. J.
,
Chandramoorthy
H. C.
,
Smith
F. J.
,
Jog
N. R.
,
Gandhirajan
R. K.
,
Kelsen
S. G.
,
Houser
S. R.
,
May
M. J.
et al. (
2011
).
Requirement of FADD, NEMO, and BAX/BAK for aberrant mitochondrial function in tumor necrosis factor alpha-induced necrosis.
Mol. Cell. Biol.
31
,
3745
3758
.
Jouan-Lanhouet
S.
,
Arshad
M. I.
,
Piquet-Pellorce
C.
,
Martin-Chouly
C.
,
Le Moigne-Muller
G.
,
Van Herreweghe
F.
,
Takahashi
N.
,
Sergent
O.
,
Lagadic-Gossmann
D.
,
Vandenabeele
P.
et al. (
2012
).
TRAIL induces necroptosis involving RIPK1/RIPK3-dependent PARP-1 activation.
Cell Death Differ.
19
,
2003
2014
.
Kim
Y. S.
,
Morgan
M. J.
,
Choksi
S.
,
Liu
Z. G.
(
2007
).
TNF-induced activation of the Nox1 NADPH oxidase and its role in the induction of necrotic cell death.
Mol. Cell
26
,
675
687
.
Kim
J. Y.
,
Kim
Y. J.
,
Lee
S.
,
Park
J. H.
(
2011
).
BNip3 is a mediator of TNF-induced necrotic cell death.
Apoptosis
16
,
114
126
.
Lei
K.
,
Davis
R. J.
(
2003
).
JNK phosphorylation of Bim-related members of the Bcl2 family induces Bax-dependent apoptosis.
Proc. Natl. Acad. Sci. USA
100
,
2432
2437
.
Li
X.
,
Klaus
J. A.
,
Zhang
J.
,
Xu
Z.
,
Kibler
K. K.
,
Andrabi
S. A.
,
Rao
K.
,
Yang
Z. J.
,
Dawson
T. M.
,
Dawson
V. L.
et al. (
2010
).
Contributions of poly(ADP-ribose) polymerase-1 and -2 to nuclear translocation of apoptosis-inducing factor and injury from focal cerebral ischemia.
J. Neurochem.
113
,
1012
1022
.
Li
J.
,
McQuade
T.
,
Siemer
A. B.
,
Napetschnig
J.
,
Moriwaki
K.
,
Hsiao
Y. S.
,
Damko
E.
,
Moquin
D.
,
Walz
T.
,
McDermott
A.
et al. (
2012
).
The RIP1/RIP3 necrosome forms a functional amyloid signaling complex required for programmed necrosis.
Cell
150
,
339
350
.
Liu
T. J.
,
Lin
S. Y.
,
Chau
Y. P.
(
2002
).
Inhibition of poly(ADP-ribose) polymerase activation attenuates β-lapachone-induced necrotic cell death in human osteosarcoma cells.
Toxicol. Appl. Pharmacol.
182
,
116
125
.
Loor
G.
,
Kondapalli
J.
,
Schriewer
J. M.
,
Chandel
N. S.
,
Vanden Hoek
T. L.
,
Schumacker
P. T.
(
2010
).
Menadione triggers cell death through ROS-dependent mechanisms involving PARP activation without requiring apoptosis.
Free Radic. Biol. Med.
49
,
1925
1936
.
McGee
A. M.
,
Baines
C. P.
(
2011
).
Complement 1q-binding protein inhibits the mitochondrial permeability transition pore and protects against oxidative stress-induced death.
Biochem. J.
433
,
119
125
.
Moubarak
R. S.
,
Yuste
V. J.
,
Artus
C.
,
Bouharrour
A.
,
Greer
P. A.
,
Menissier-de Murcia
J.
,
Susin
S. A.
(
2007
).
Sequential activation of poly(ADP-ribose) polymerase 1, calpains, and Bax is essential in apoptosis-inducing factor-mediated programmed necrosis.
Mol. Cell. Biol.
27
,
4844
4862
.
Nakagawa
T.
,
Shimizu
S.
,
Watanabe
T.
,
Yamaguchi
O.
,
Otsu
K.
,
Yamagata
H.
,
Inohara
H.
,
Kubo
T.
,
Tsujimoto
Y.
(
2005
).
Cyclophilin D-dependent mitochondrial permeability transition regulates some necrotic but not apoptotic cell death.
Nature
434
,
652
658
.
Park
M. T.
,
Song
M. J.
,
Lee
H.
,
Oh
E. T.
,
Choi
B. H.
,
Jeong
S. Y.
,
Choi
E. K.
,
Park
H. J.
(
2011
).
β-lapachone significantly increases the effect of ionizing radiation to cause mitochondrial apoptosis via JNK activation in cancer cells.
PLoS ONE
6
,
e25976
.
Parra
M.
,
Lluís
F.
,
Miralles
F.
,
Caelles
C.
,
Muñoz-Cánoves
P.
(
2000
).
The cJun N-terminal kinase (JNK) signaling pathway mediates induction of urokinase-type plasminogen activator (uPA) by the alkylating agent MNNG.
Blood
96
,
1415
1424
.
Quinsay
M. N.
,
Lee
Y.
,
Rikka
S.
,
Sayen
M. R.
,
Molkentin
J. D.
,
Gottlieb
R. A.
,
Gustafsson
A. B.
(
2010
).
Bnip3 mediates permeabilization of mitochondria and release of cytochrome c via a novel mechanism.
J. Mol. Cell. Cardiol.
48
,
1146
1156
.
Scharstuhl
A.
,
Mutsaers
H. A.
,
Pennings
S. W.
,
Russel
F. G.
,
Wagener
F. A.
(
2009
).
Involvement of VDAC, Bax and ceramides in the efflux of AIF from mitochondria during curcumin-induced apoptosis.
PLoS ONE
4
,
e6688
.
Schinzel
A. C.
,
Takeuchi
O.
,
Huang
Z.
,
Fisher
J. K.
,
Zhou
Z.
,
Rubens
J.
,
Hetz
C.
,
Danial
N. N.
,
Moskowitz
M. A.
,
Korsmeyer
S. J.
(
2005
).
Cyclophilin D is a component of mitochondrial permeability transition and mediates neuronal cell death after focal cerebral ischemia.
Proc. Natl. Acad. Sci. USA
102
,
12005
12010
.
Shiah
S. G.
,
Chuang
S. E.
,
Chau
Y. P.
,
Shen
S. C.
,
Kuo
M. L.
(
1999
).
Activation of c-Jun NH2-terminal kinase and subsequent CPP32/Yama during topoisomerase inhibitor β-lapachone-induced apoptosis through an oxidation-dependent pathway.
Cancer Res.
59
,
391
398
.
Sosna
J.
,
Voigt
S.
,
Mathieu
S.
,
Lange
A.
,
Thon
L.
,
Davarnia
P.
,
Herdegen
T.
,
Linkermann
A.
,
Rittger
A.
,
Chan
F. K.
et al. (
2014
).
TNF-induced necroptosis and PARP-1-mediated necrosis represent distinct routes to programmed necrotic cell death.
Cell. Mol. Life Sci.
71
,
331
348
.
Sun
L.
,
Wang
H.
,
Wang
Z.
,
He
S.
,
Chen
S.
,
Liao
D.
,
Wang
L.
,
Yan
J.
,
Liu
W.
,
Lei
X.
et al. (
2012
).
Mixed lineage kinase domain-like protein mediates necrosis signaling downstream of RIP3 kinase.
Cell
148
,
213
227
.
Tagliarino
C.
,
Pink
J. J.
,
Dubyak
G. R.
,
Nieminen
A. L.
,
Boothman
D. A.
(
2001
).
Calcium is a key signaling molecule in β-lapachone-mediated cell death.
J. Biol. Chem.
276
,
19150
19159
.
Tagliarino
C.
,
Pink
J. J.
,
Reinicke
K. E.
,
Simmers
S. M.
,
Wuerzberger-Davis
S. M.
,
Boothman
D. A.
(
2003
).
μ-calpain activation in β-lapachone-mediated apoptosis.
Cancer Biol. Ther.
2
,
141
152
.
van Wijk
S. J.
,
Hageman
G. J.
(
2005
).
Poly(ADP-ribose) polymerase-1 mediated caspase-independent cell death after ischemia/reperfusion.
Free Radic. Biol. Med.
39
,
81
90
.
Vanlangenakker
N.
,
Vanden Berghe
T.
,
Bogaert
P.
,
Laukens
B.
,
Zobel
K.
,
Deshayes
K.
,
Vucic
D.
,
Fulda
S.
,
Vandenabeele
P.
,
Bertrand
M. J.
(
2011
).
cIAP1 and TAK1 protect cells from TNF-induced necrosis by preventing RIP1/RIP3-dependent reactive oxygen species production.
Cell Death Differ.
18
,
656
665
.
Vanlangenakker
N.
,
Vanden Berghe
T.
,
Vandenabeele
P.
(
2012
).
Many stimuli pull the necrotic trigger, an overview.
Cell Death Differ.
19
,
75
86
.
Vosler
P. S.
,
Sun
D.
,
Wang
S.
,
Gao
Y.
,
Kintner
D. B.
,
Signore
A. P.
,
Cao
G.
,
Chen
J.
(
2009
).
Calcium dysregulation induces apoptosis-inducing factor release: cross-talk between PARP-1- and calpain-signaling pathways.
Exp. Neurol.
218
,
213
220
.
Wang
Y.
,
Kim
N. S.
,
Li
X.
,
Greer
P. A.
,
Koehler
R. C.
,
Dawson
V. L.
,
Dawson
T. M.
(
2009
).
Calpain activation is not required for AIF translocation in PARP-1-dependent cell death (parthanatos).
J. Neurochem.
110
,
687
696
.
Wang
Z.
,
Jiang
H.
,
Chen
S.
,
Du
F.
,
Wang
X.
(
2012
).
The mitochondrial phosphatase PGAM5 functions at the convergence point of multiple necrotic death pathways.
Cell
148
,
228
243
.
Xu
Y.
,
Huang
S.
,
Liu
Z. G.
,
Han
J.
(
2006
).
Poly(ADP-ribose) polymerase-1 signaling to mitochondria in necrotic cell death requires RIP1/TRAF2-mediated JNK1 activation.
J. Biol. Chem.
281
,
8788
8795
.
Yang
J.
,
Duerksen-Hughes
P. J.
(
2001
).
Activation of a p53-independent, sphingolipid-mediated cytolytic pathway in p53-negative mouse fibroblast cells treated with N-methyl-N-nitro-N-nitrosoguanidine.
J. Biol. Chem.
276
,
27129
27135
.
Zhang
D. W.
,
Shao
J.
,
Lin
J.
,
Zhang
N.
,
Lu
B. J.
,
Lin
S. C.
,
Dong
M. Q.
,
Han
J.
(
2009
).
RIP3, an energy metabolism regulator that switches TNF-induced cell death from apoptosis to necrosis.
Science
325
,
332
336
.
Zingarelli
B.
,
Salzman
A. L.
,
Szabó
C.
(
1998
).
Genetic disruption of poly (ADP-ribose) synthetase inhibits the expression of P-selectin and intercellular adhesion molecule-1 in myocardial ischemia/reperfusion injury.
Circ. Res.
83
,
85
94
.

Competing interests

The authors declare no competing interests.

Supplementary information