Proteolytic inactivation of poly (ADP-ribose) polymerase 1 (PARP1) by the Nlrp3 and Nlrc4 inflammasomes

Abstract

Caspase-mediated cleavage of the DNA damage sensor poly(adenosine 5′-diphosphate-ribose) polymerase 1 (PARP1) is a hallmark of apoptosis. However, whether PARP1 is processed during pyroptosis, a specialized cell death program that occurs upon activation of caspase-1 in inflammasome complexes remains unclear. Here, we show that activation of the Nlrp3 and Nlrc4 inflammasomes induces processing of full-length PARP1 into a fragment of 89 kDa in a stimulus-dependent manner. Macrophages deficient for caspase-1 and those lacking the inflammasome adaptors Nlrp3, Nlrc4 and ASC were highly resistant to cleavage, whereas macrophages lacking the downstream inflammasome effector caspase-7 were partially protected. A modest, but statistically significant reduction in Nlrp3 inflammasome-induced pyroptosis was observed in PARP1 knockout macrophages. Thus, protease-mediated inactivation of PARP1 is a shared feature of apoptotic, necrotic and pyroptotic cells.

Introduction

The DNA damage repair enzyme poly(adenosine 5′-diphosphate-ribose) polymerase 1 (PARP1) recognizes single-stranded DNA breaks, double-stranded DNA breaks, cross-overs and supercoils (1). Binding to damaged DNA activates to catalyze the synthesis of poly(ADP) ribose (PAR), a branched polymer of repeated ADP-ribose subunits linked by glycosidic bonds. Autocatalytic poly(ADP) ribosylation of enhances the recruitment of DNA repair factors in order to salvage DNA damage in a process that consumes NAD⁺ and ATP energy stores of the cell (2, 3). To prevent energy depletion, PARP1 is proteolytically inactivated during apoptosis and necrosis (4–8), the two best characterized programmed cell death programs. In apoptotic cells, executioner caspases and granzymes are responsible for PARP1 processing (4–7). Similarly, PARP1 is proteolytically inactivated by lysosomal cathepsins during necrosis (8) Moreover, PARP1 cleavage fragments have been demonstrated to act as dominant negative molecules preventing DNA repair by full-length PARP1 and thus contributing to efficient cell death execution (7). However, whether PARP1 is processed during pyroptosis, a specialized form of pro-inflammatory programmed cell death in macrophages and dendritic cells, remains unclear. Pyroptosis is induced when the inflammatory caspase-1 is activated in large cytosolic protein complexes termed ‘inflammasomes’ (9, 10). The Nlrp3 inflammasome represents the best characterized caspase-1-activating complex (9). The Nod-like receptor Nlrp3 recruits caspase-1 into this complex in response to conserved microbial components, crystalline substances and endogenous danger signals such as ATP and uric acid (11). In contrast, the Nod-like receptor Nlrc4 is required for caspase-1 activation in macrophages infected with Salmonella typhimurium (9, 12, 13). The bipartite adaptor protein ASC is essential for bridging the interaction between Nod-like receptors and caspase-1 in inflammasomes because caspase-1 activation is abolished in ASC-deficient macrophages (13, 14).

Because pyroptosis is accompanied by DNA damage and oligonucleosomal DNA fragmentation (13–17), we investigated whether PARP1 is processed during this pro-inflammatory cell death mode. We showed that caspase-1 and the downstream inflammasome effector caspase-7 are responsible for PARP1 cleavage during pyroptosis. PARP1 deficient macrophages were less sensitive to pyroptosis induced by activation of the Nlrp3 inflammasome, suggesting that inflammasome-mediated inactivation of PARP1 contributes to pyroptotic cell death.

Materials and methods

Mice and macrophages

Nlrp3^−/−, Nlrc4^−/−, Pycard^−/−, Casp7^−/− and Casp1^−/− mice in a C57BL/6 background have been described before (14, 18–20). PARP1^−/− mice were obtained from Jackson laboratory. Mice were housed in a pathogen-free facility and the animal studies were conducted under protocols approved by St. Jude Children’s Research Hospital Committee on Use and Care of Animals. Bone marrow-derived macrophages (BMDMs) were prepared as described before (21). Briefly, bone marrow was isolated from femurs of 6–12 weeks old mice and were cultured in IMDM containing 10% heat-inactivated FBS, 20% L cell-conditioned medium, 100 U/ml penicillin, and 100 mg/ml streptomycin at 37°C in a humidified atmosphere containing 5% CO₂. After 5–7 days of incubation, cells were collected and plated in 6-well plates or in 24-well plates in IMDM containing 10% heat-inactivated FBS and 100 mg/ml thymidine and antibiotics. Macrophages were cultured for an additional 24 h before use.

Bacteria and microbial ligands

Salmonella enterica serovar typhimurium cultures were grown to stationary phase under aerobic conditions at 37°C in 5 ml Luria-Bertani broth (Difco Laboratories) and subcultured to O.D₆₀₀ 0.5 before being used for infecting macrophage cultures (MOI 5). Bacterial lipopolysaccharide (LPS) and the Toll-like receptor 2 (TLR2) agonist Pam3-CSK4 were purchased from Invivogen, USA. The fungal cell wall component mannan was purchased from Sigma-Aldrich, USA. The ligands were used at a concentration of 10 μg/ml. ATP was from Roche and used at 5 mM, whereas nigericin was obtained from Sigma and used at 20 μM. Stimulation of BMDMs with microbial ligands, ATP and nigericin was performed as previously described (10, 22).

Immunoblotting

Cells were washed twice with phosphate-buffered saline and scraped in lysis buffer (150 mM NaCl, 10 mM Tris pH 7.4, 5 mM EDTA, 1 mM EGTA, 0.1% Nonidet P-40) supplemented with a protease inhibitor cocktail tablet (Roche). Samples were clarified, denatured with SDS buffer and boiled for 5 min. Proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes. The membranes were immunoblotted with primary antibodies and detected with a secondary anti-rabbit antibody conjugated to horseradish peroxidase (Jackson ImmunoResearch) followed by enhanced chemiluminescence (Thermo Scientific). Rabbit anti-mouse caspase-1 was a generous gift of Dr. P. Vandenabeele (Ghent University, Belgium). PARP1 and caspase-7 antibodies were from Cell Signaling Technologies.

In vitro PARP1 cleavage assays

Recombinant PARP1 which was purified to near homogeneity (Trevigen) was subjected to in vitro protease assay in a total reaction volume of 50μl. The reaction contents were incubated at 37°C in presence of 30 nM caspase-1 or casapse-7 in the protease assay buffer (20 mm HEPES-KOH, pH 7.5, 10 mm KCl, 1.5 mm MgCl2, 1 mm EDTA, 1 mm DTT and 1/100th/1ml buffer of a Complete protease inhibitor mixture tablet (Roche Applied Science)). The reactions were stopped by adding equal volume of 2X SDS buffer and boiled for 5 min. The resulting cleavage products were analyzed by SDS-PAGE and immunoblotting with anti-PARP1 antibodies.

Cell death assay

Induction of pyroptosis was quantified according to the manufacturer instructions by monitoring early membrane permeabilization using the commercial Live/Dead assay (Invitrogen). Data were analyzed with Student’s t-test. p<0.05 was considered statistically significant.

Results and discussion

PARP1 is cleaved and inactivated in cells undergoing apoptosis or necrosis (4–8). To determine whether PARP1 is also processed during pyroptosis, the Nlrp3 inflammasome was activated in BMDMs isolated from C57BL/6 mice through stimulation with the TLR4 ligand LPS, the TLR2 ligand Pam3-CSK4 or the fungal cell wall component mannan for 3 hours followed by 5 mM ATP treatment for another 30 minutes. These stimuli are known to induce activation of the Nlrp3 inflammasome and pyroptotic cell death of activated macrophages (10, 21, 22). In agreement, caspase-1 activation was observed in stimulated BMDMs, but not in the untreated control setup (Fig. 1A). Unlike in untreated cells, significant pyroptotic cell death was observed in macrophages stimulated with LPS, Pam3-CSK4 or mannan combined with ATP (Fig. 1B). Notably, full-length PARP1 was processed into an 89 kDa fragment in the setups where caspase-1 activation and pyroptosis induction were observed, but not in untreated macrophages (Fig. 1A). These results demonstrate that PARP1 is processed during pyroptosis following ATP-induced activation of the Nlrp3 inflammasome. ATP can be substituted for the cation ionophore nigericin to engage the Nlrp3 inflammasome (10, 23). As in activated macrophages stimulated with ATP (Fig. 1A), PARP1 cleavage was observed in BMDMs that were stimulated with a combination of LPS, Pam3-CSK4 or mannan and nigericin (Fig. 1C). PARP1 processing was accompanied by significant induction of pyroptotic cell death in nigericin-treated cells (Fig. 1D). These results demonstrate that ATP- and nigericin-mediated activation of the Nlrp3 inflammasome is associated with PARP1 cleavage.

Figure 1. PARP1 is cleaved during ATP- and nigericin-induced pyroptotic cell death of activated macrophages.

A–D, BMDMs were primed with LPS, Pam3 or mannan for 3 h and then stimulated with ATP (A, B) or nigericin (C, D) for 30 min. (A, C) Cell extracts were immunoblotted with antibodies against PARP1 and caspase-1. The bands corresponding with full-length PARP1 (113 kDa), the 89 kDa PARP1 fragment, procaspase-1 (45 kDa) and the large catalytic subunit (20 kDa) are indicated. (B, D) Membrane permeabilization was measured as a cell death parameter using the Live/Dead assay (Invitrogen). Cell death data represent the mean ± standard deviation of triplicates. Results are representative of at least three independent experiments. LPS, Lipopolysaccharide; Pam3, Pam3-GSK4

Caspase-1 functions as the central effector of the Nlrp3 inflammasome (24) and is essential for ATP- and nigericin-induced pyroptosis in LPS-primed macrophages (10). To confirm that caspase-1 activation by the Nlrp3 inflammasome is responsible for PARP1 processing, BMDMs from mice lacking Nlrp3 (Nlrp3^−/−), the inflammasome adaptor ASC (Pycard^−/−) or caspase-1 (Casp1^−/−) were stimulated with LPS, Pam3-CSK4 or mannan in combination with ATP as described above, and cellular lysates were probed for PARP1 processing. Whereas PARP1 was readily processed under these conditions in wild-type BMDMs, PARP1 cleavage was abrogated in macrophages lacking these essential components of the Nlrp3 inflammasome (Fig. 2A). In contrast to Nlrp3 and ASC, Nlrc4 is not required for caspase-1 activation in TLR-activated macrophages exposed to ATP (13, 23). In agreement, PARP1 processing was not affected in Nlrc4 deficient macrophages stimulated with LPS, Pam3-CSK4 or mannan in combination with ATP (Fig. 2B). Nlrc4 is essential for caspase-1 activation and pyroptosis induction in macrophages infected with Salmonella typhimurium (12, 13). Consistently, Salmonella-induced PARP1 processing was abrogated in Nlrc4 knockout macrophages, but not in those lacking Nlrp3 (Fig. 2C). These results demonstrate that the Nlrp3 and Nlrc4 inflammasomes are essential for stimulus-dependent PARP1 processing during pyroptosis.

Figure 2. The Nlrp3 and Nlrc4 inflammasomes mediate PARP1 processing during pyroptosis.

(A, B) Macrophages from wild-type, Nlrp3^−/−, Nlrc4^−/−, Pycard^−/− and Casp1^−/− mice were primed with LPS, Pam3 or mannan for 3 h and then stimulated with ATP for 30 min or (C) infected for 4 h with Salmonella. Cell extracts were immunoblotted with antibodies against PARP1. The bands corresponding with full-length PARP1 (113 kDa) and the 89 kDa PARP1 fragment are indicated. Results are representative of three independent experiments. LPS, Lipopolysaccharide; Pam3, Pam3-GSK4

It was previously reported that the executioner protease caspase-7 is a downstream effector of the Nlrp3 and Nlrc4 inflammasomes (14, 22). In agreement, we observed caspase-7 processing indicative of its activation in wild-type macrophages that were stimulated with LPS, Pam3-CSK4 or mannan combined with ATP, but not in Casp1^−/− macrophages (Fig. 3A). As expected, the caspase-7 antibody failed to detect immunoreactive bands in lysates of caspase-7 deficient (Casp7^−/−) macrophages, thus confirming its specificity (Fig. 3A). Because both caspase-7 and caspase-1 are activated upon stimulation of the Nlrp3 inflammasome, both caspases may contribute to PARP1 processing during pyroptosis. To test this hypothesis, we determined the extent of PARP1 processing following Nlrp3 inflammasome activation in Casp1^−/− and Casp7^−/− macrophages, respectively. PARP1 processing was abrogated in Casp1^−/− macrophages (Fig. 3A) due to defective activation of both caspase-1 and caspase-7 in these cells (Fig. 3B). In contrast, PARP1 was processed in Casp7^−/− macrophages, albeit at significantly reduced levels when compared to wild-type BMDMs (Fig. 3B). These results suggest that both caspases-1 and -7 contribute to PARP1 cleavage during pyroptosis. Indeed, PARP1 was processed into an 89 kDa fragment when in vitro-translated PARP1 was incubated with either recombinant caspase-1 (Fig. 3C upper panel) or caspase-7 (Fig. 3C lower panel). The band corresponding to full-length PARP1 gradually decreased as early as 30 minutes after incubation with either recombinant caspase-1 or -7, while the 89 kDa cleavage product gained significance (Fig. 3C).

Figure 3. PARP1 is processed by caspases-1 and -7 and PARP1 inactivation protects against pyroptosis.

(A, B) Macrophages from wild type, Casp1^−/− and Casp7^−/− mice were primed with LPS, Pam3 or Mannan for 3 h and then stimulated with ATP for 30 min. Cell extracts were immunoblotted with antibodies against caspase-7 (A) and PARP1 (B). The bands corresponding with respectively procaspase-7 (33 kDa) and its large catalytic subunit (19 kDa), and full-length PARP1 (113 kDa) and the 89 kDa PARP1 fragment are indicated. (C) Recombinant PARP1 was incubated with 30 nM recombinant caspase-1 or casapse-7 in protease assay buffer (20 mm HEPES-KOH pH 7.5, 10 mm KCl, 1.5 mm MgCl₂, 1 mm EDTA, 1 mm DTT and Complete protease inhibitor (Roche)). PARP1 cleavage products were detected by Western blotting with PARP1 antibodies. (D) Macrophages from WT and PARP1^−/− mice were primed with LPS for 3 h and then stimulated with ATP for indicated times. Membrane permeabilization was measured as a cell death parameter using the Live/Dead assay (Invitrogen). Cell death data represent the mean ± standard deviation of triplicates. Results are representative of three independent experiments.

PARP1 cleavage by apoptotic caspases and granzymes is thought to contribute to efficient apoptosis execution (4–7). To determine whether PARP1 is required for efficient induction of pyroptotic cell death, BMDMs from PARP1 deficient (PARP1^−/−) mice were stimulated LPS in combination with ATP and the extent of pyroptosis induction was compared relative to wild-type macrophages. A modest, but statistically significant (p<0.004) reduction in pyroptosis was observed in LPS-primed PARP1 knockout macrophages that have been exposed to 5 mM ATP for 20 min or 30 min, respectively (Fig. 3D). Protection from LPS+ATP-induced pyroptosis was not due to defective secretion of pro-inflammatory cytokines, because culture supernatants of PARP1 deficient macrophages contained normal levels of the inflammasome-dependent cytokines IL-1β and IL-18 and the inflammasome-independent cytokines IL-6 and TNF-α (Supplementary Fig. 1). Thus, Nlrp3 inflammasome-mediated inactivation of PARP1 contributes to pyroptosis induction in activated macrophages.

Taken together, these results demonstrate that PARP1 is cleaved by caspase-1 and caspase-7 upon inflammasome activation, thus identifying one of the first molecular mechanisms by which inflammasomes induce pyroptosis. Taken together with previous reports on PARP1 processing in apoptosis (4–7) and necrosis (8), our results suggest that PARP1 processing is a general strategy used by cells undergoing programmed cell death to preserve the cellular energy stores in order to allow proper execution of the cell death program, and possibly to generate dominant negative cleavage fragments that may further enhance cell death execution by inhibiting PARP1-mediated DNA repair.

Abbreviations

NLR

NOD-like receptor

PARP1

poly (ADP-ribose) polymerase 1

TLR

Toll-like receptor

WT

wild-type