Apoptotic Cells Release IL1 Receptor Antagonist in Response to Genotoxic Stress

Cells are subjected to various genotoxic insults throughout their life cycle, resulting in DNA lesions that affect genomic DNA replication and transcription (1). If not repaired properly, DNA lesions can result in mutagenesis and wide-scale genomic aberrations that endanger the survival of the individual cell and the organism. Cells express sensor proteins that recognize DNA lesions and stalled replication forks (2). These sensor proteins initiate a DNA damage response (DDR), which imposes cell-cycle arrest, allowing DNA repair, or induces apoptosis if the DNA damage is too extensive to be repaired (3).

The regulated degradation of cell contents during apoptosis prevents the release of inflammatory intracellular factors (4, 5). Inflammatory responses are also suppressed by the release of anti-inflammatory mediators, such as IL10, TGFβ, and prostaglandins by macrophage, upon phagocytosis of apoptotic cells. However, the anti-inflammatory effects mediated by apoptotic cells are still not well understood.

IL1 is a critical cytokine in inflammation, innate immunity, angiogenesis, and hematopoiesis (6). IL1 is encoded by two genes, IL1α and IL1β, which induce the expression of proinflammatory factors such as IL6 upon binding to IL1 receptor type 1 (IL1R1; ref. 7). IL1α and IL1β are synthesized as precursor proteins that are processed into mature forms. Both the pro-IL1α and the mature IL1α forms are biologically active, whereas IL1β requires processing by caspase 1 (8). IL1α precursors are constitutively expressed in fibroblasts, epithelial cells, and keratinocytes. Upon apoptosis, IL1α is sequestered intracellularly, which prevents the promotion of inflammation (9).

The IL1RA competes with IL1 for binding to the IL1R1 and disrupts the proinflammatory signaling pathways (10). IL1RA exists as a secreted soluble IL1RA (sIL1Ra) isoform and three intracellular (icIL1RA) isoforms. The ratio of IL1RA to IL1 was found to be critical for the control of inflammatory responses in animal and patients studies (11–13). IL1ra-deficient mice exhibit excessive inflammation when bred to various disease models, and patients with IL1RA deficiency show severe inflammatory conditions (14–19). The icIL1RA isoforms are considered a reservoir of biologic active IL1RA, but the factors that regulate the release of icIL1RA have not been studied in detail (6, 20–23).

Here, we report that apoptosis in response to genotoxic stress results in the release of intracellular IL1RA, but not IL1α or IL1β. The release of IL1RA depended on ATM, caspase-9, and caspase-3 activation, but did not require de novo protein synthesis, classical secretion pathways, or N-glycosylation. IL1RA released in response to bleomycin treatment impaired IL1-induced IL6 expression. In summary, we show that caspase-mediated apoptosis leads to the release of biologically active intracellular IL1RA.

Mice were housed and bred in pathogen-free conditions in strict compliance with the Institutional Animal Care and Use Committee guidelines at the National University of Singapore, in accordance with the National Advisory Committee for Laboratory Animal Research guidelines (Guidelines on the Care and Use of Animals for Scientific Purposes). The protocol was approved by the Institutional Animal Care and Use Committee of the National University of Singapore (Protocol number: 041/08), and steps were taken to minimize suffering.

C57BL/6 mice were purchased from the Centre for Animal Resources at the National University of Singapore (Singapore). IL1r-deficient mice were obtained from the Jackson Laboratory. Ears derived from IL1ra-deficient mice were kindly provided by Dr. C. Wakabayashi (National Institute of Neuroscience, Japan).

Primary fibroblasts from adult C57BL/6 mice were established as previously described (24). Isolated fibroblasts were cultured in RPMI 1640 medium (Invitrogen) supplemented with 5% heat-inactivated FCS (Invitrogen), 150 μmol/L β-mercaptoethanol (Sigma), 1.4 μmol/L of l-glutamine (Sigma), 20 U/mL of penicillin–streptomycin (Invitrogen), and 20 mmol/L of HEPES (Hyclone) for 12 to 24 days before being used for experiments. The mouse monocytic leukemic macrophage cell lines RAW 264.7 (<10 passages) and J774.2 (<10 passages) were obtained from ATCC. The TRAMP-C2 prostate cancer cell line (<10 passages) was kindly provided by Dr. D.H. Raulet (University of California, Berkeley, CA). BC2 cells were a generous gift by Dr. Corcoran (Wehi, Australia; <20 passages). All cells were maintained at 37°C in a humidified 5% CO₂ incubator (Thermo Scientific).

A total of 2.5 × 10⁵ primary fibroblasts or RAW 264.7 cells were treated with 20 μg/mL of cycloheximide (Merck Millipore), 5 μg/mL of tunicamycin (Sigma), or 20 mmol/L of 2-Deoxy-D-glucose (Sigma) for 1 hour, followed by treatment with 30 μg/mL of bleomycin (Merck Millipore) or 0.5 μg/mL of doxorubicin (Santa Cruz technology) for 20 hours.

To investigate the role of secretory pathways, 2.5 × 10⁵ primary fibroblasts were treated with 30 μg/mL of bleomycin (Merck Millipore) for 14 hours, followed by 1 μg/mL of BD GolgiStop (BD Biosciences) and 2 μmol/L of Protein Transport Inhibitor (BD Biosciences) for 6 hours.

To study the role of caspases, 2.5 × 10⁵ primary fibroblasts were treated with 20 μmol/L of the pan-caspase inhibitor, Z-VAD-FMK (Santa Cruz Technology), 10 μmol/L of caspase-9 inhibitor, Z-LEHD-FMK (BD Biosciences), or 10 μmol/L of caspase-3 inhibitor, Z-DEVD-FMK (Santa Cruz Technology) for 1 hour, followed by treatment with 30 μg/mL of bleomycin (Merck Millipore) for 20 hours. Supernatant was harvested from each well, centrifuged at 1,500 rpm for 5 minutes at 4°C to remove cell debris, and stored at –80°C until analysis.

To investigate the role of the DDR, 2.5 × 10⁵ primary fibroblasts were treated with 10 μmol/L of the ATM kinase-specific inhibitor, KU-60019 (R&D systems, USA) or 5 mmol/L of caffeine (Sigma, Singapore) for 1 hour, followed by treatment with 30 μg/ml of bleomycin (Merck Millipore) for 20 hours.

To assess bioactivity of IL1RA, primary fibroblasts were seeded at a density of 2.5 × 10⁵ cells per well in 6-well dishes and treated with 30 μg/mL of bleomycin (Merck Millipore) for 20 hours. The supernatant was transferred to a second set of fibroblasts. One hour later, recombinant IL1α (Peprotech) at the indicated concentrations was added for 6 hours. Alternatively, the supernatant was transferred to J774.2 cells transfected with 1 μg/mL of poly (dA:dT) LyoVec (Invivogen) overnight after pretreatment with 1 μmol/L of PMA (Sigma) for 3 hours and subsequent treatment with 2 μg/mL of LPS for 16 hours. The levels of IL6 transcripts were measured by real-time PCR.

The RNeasy Mini Kit (Qiagen) was used to isolate total RNA from treated cells. Reverse transcription was performed using 2 μg of total RNA, random hexamer primer, and moloney murine leukemia virus reverse transcriptase (M-MLV RT; Promega). For PCR, an amplification mixture consisting of 50 ng of reverse-transcribed RNA, 0.2 μmol/L forward primer, 0.2 μmol/L reverse primer, and 12.5 μL of iTaq SYBR Green Supermix with ROX (Biorad Laboratories) was prepared (25 μL). PCRs were performed in duplicates using the ABI PRISM 7700 Sequence Detection System from Applied Biosystems. The PCR thermocycling parameters were 50°C for 2 minutes, 95°C for 10 minutes, and 40 cycles of 95°C for 15 seconds and 60°C for 15 seconds and 72°C for 1 minute. All samples were normalized to their respective housekeeping gene hypoxanthine phosphoribosyltransferase (Hprt). The following primers were used: Hprt-5′, tgggaggccatcacattgt, Hprt-3′: gcttttccagtttcactaatgaca; IL1α-5′: gccagttgagtaggataaagg, IL1α-3′: cagtctgtctccttcttgagg; IL1β-5′: tcacagcagcacatcaacaa, IL1β-3′: tgtcctcatcctggaaggtc; IL1ra-5′: gggatactaaccagaagacc, IL1ra-3′: gacaggcacagcttgccccc; sIL1ra-5′: cttgctgtggcctcgggatgg, sIL1ra-3′: cttttcccagaagggcggcagg; icIL1ra-5′: tccaccctgggaaggtctgtgc, icIL1ra-3′: gctgcctctgaagccatgggtg; and IL6-5′: tgcaagagacttccatccagttg, IL6-3′: cctccgacttgtgaagtggt. Samples without RNA were included as negative controls.

Phycoerythrin-conjugated antibody to IL1R1 was purchased from BD Pharmingen. Dead cells were excluded from the analysis based on the side scatter versus forward scatter profile, and based on exclusion of propidium iodide (Sigma) fluorescence signal. Intracellular staining of caspase 3 was performed using the BD intracellular fixation and permeabilization Kit (BD Biosciences) according to the manufacturer's instructions. Briefly, cells were stained with the viability dye eFluor 780 (eBioscience) at 4°C for 30 minutes. Data acquisition was performed on a FACSCalibur (BD Biosciences) or a CyAn ADP analyzer (Beckman Coulter) and analyzed using Flowjo software 8.8.7 (Tree Star Inc.).

Apoptotic cells were detected using the Annexin V apoptosis detection Kit (eBioscience) according to the manufacturer's guidelines.

For the preparation of whole-cell extracts, cells were lysed in RIPA buffer consisting of 150 mmol/L NaCl, 1 mmol/L EDTA, 50 mmol/L Tris-HCl (pH 7.4), 1% NP-40, and 1% sodium deoxycholate (Sigma). The protease inhibitor cocktail set III and phosphatase inhibitor cocktail set V (Merck Millipore) were added to the lysis buffer according to the manufacturer's instructions prior to cell lysis. The whole-cell extracts were electrophoresed in 12% sodium dodecyl sulfate polyacrylamide gels and blotted onto nitrocellulose membranes (Amersham). The membranes were probed with the following antibodies: IL1RA/IL1F3 MAb (Clone 694204; RnD Systems) followed by Horseradish peroxidase (HRP)–conjugated goat anti-mouse or anti-rabbit IgG (Jackson Immunoresearch). GAPDH expression was detected with a GAPDH antibody conjugated to HRP purchased from Sigma.

The quantification of IL1α and IL1RA was determined using ELISA kits from R&D Systems. Quantification of IL1β and IL6 was determined using ELISA Ready Set Go kits (eBioscience) according to the manufacturer's instructions.

Caspase-Glo assays (Promega) were used to measure caspase-3/7, -8, and -9 activities as instructed by the manufacturer. Primary fibroblasts were seeded at a density of 2 × 10⁴ to 4 × 10⁴ cells/well of a 96-well plate and treated as indicated. Luminescent signals were analyzed using a Wallac 1420 VICTOR2 (Perkin Elmer).

Statistical analyses were performed using the Student t test in Prism 6.0f (Graphpad). Error bars denote ±SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; and ****, P < 0.0001.

The expression of IL1α and IL1β is increased in response to persistent DNA damage (25). To assess the effects of acute DNA damage on IL1 family members, we treated primary murine fibroblasts derived from adult tail and ear tissue with the radiomimetic bleomycin for 20 hours. Acute DNA damage in response to bleomycin failed to significantly increase IL1a and IL1b transcripts levels (Fig. 1A). In contrast, bleomycin treatment decreased IL1r (P < 0.002) and IL1ra (P < 0.001) transcripts. The presence of fewer IL1a and IL1b transcripts differentially affected protein amounts. The IL1R expression at the cell surface did not significantly change during bleomycin treatment of cells (Fig. 1B). In contrast, the amounts of IL1RA increased in the supernatant of bleomycin-treated cells in a dose-dependent manner (Fig. 1C and D; P < 0.05). Treatment of fibroblasts with doxorubicin, a genotoxic topoisomerase II inhibitor, also induced the release of IL1RA (P < 0.005), suggesting that IL1RA is released by fibroblasts in response to acute DNA damage (Fig. 1E). The release of IL1RA was not limited to fibroblasts as B-cell lymphoma, prostate cancer cells, and macrophage-like cancer cells also released IL1RA in response to bleomycin (Fig. 1F and data not shown).

A closer analysis of Il1ra and IL1RA transcript levels showed that secreted (sIL1ra) and intracellular (icIL1ra) Il1ra transcripts decreased within 4 hours (Fig. 2A). In contrast, the amounts of IL1RA gradually increased in the supernatant of bleomycin-treated fibroblasts after 16 hours, indicating that bleomycin-induced release of IL1RA is independent of transcription (Fig. 2B; P < 0.01). The increase of IL1RA in the supernatant was in contrast with reduced intracellular IL1RA 20 hours after bleomycin treatment (Fig. 2C; P < 0.01). Hence, our data indicate that the intracellular pool of IL1RA is released in response to DNA damage.

To test if bleomycin-induced IL1RA secretion requires de novo protein synthesis, we inhibited protein translation with cycloheximide. Pretreatment of fibroblasts with cycloheximide did not abrogate bleomycin-induced IL1RA levels (Fig. 3A; P < 0.004). Interestingly, cycloheximide also induced the release of IL1RA, and the effects of cycloheximide and bleomycin were additive (Fig. 3A; P < 0.002).

Next, we analyzed if the release of IL1RA depended on secretion via the endoplasmic reticulum (ER)–Golgi complex secretory pathway. The release of IL1RA was not significantly impaired by the addition, during the last 6 hours of bleomycin treatment, of brefeldin A and monensin, which disrupt the ER–Golgi complex secretory pathway (Fig. 3B).

N-glycosylation regulates the secretion and degradation of certain proteins (26). Inhibition of N-glycosylation by tunicamycin or 2-deoxy-d-glucose did not hamper the release of IL1RA, but rather enhanced it (Fig. 3C and D; P < 0.05).

To test if release of IL1RA in response to bleomycin is a consequence of genotoxic stress, we inhibited the ensuing activation of the DDR. Chemical inhibition of ATM (ataxia telangiectasia, mutated) impaired the release of IL1RA (Fig. 3E and F). Together, these observations suggest that IL1RA is released via DDR-dependent nonclassical secretion pathways in response to bleomycin.

To investigate if the release of IL1RA in response to bleomycin was a consequence of DDR-mediated apoptosis, we measured early- and late-stage events of apoptosis. Early apoptosis was measured by examining the binding of Annexin V to translocated phosphatidylserine on bleomycin-treated cells. Uptake of 7-amino-actinomycin D (7-AAD) by cells was used to analyze the loss of cell membrane integrity at the late stage of apoptosis. In response to bleomycin treatment, early- and late-stage events of apoptosis appeared after 8 hours of treatment with bleomycin, and the percentage of apoptotic cells increased further over the next 40 hours (Fig. 4A; P < 0.05). The loss of the mitochondrial membrane potential in response to DNA damage can promote the activation of caspase 9 and caspase 8 (27). Caspase 9 cleaves and activates caspase 3 and caspase 7 leading to apoptosis of cells. In accordance, the activation of caspase 9, caspase 7, and caspase 3 significantly increased after 4 hours (P < 0.01) of bleomycin treatment (Fig. 4B–D). A significant increase in caspase-8 activity was only observed after 16 hours of bleomycin treatment (Fig. 4C; P < 0.05). Bleomycin treatment also increased the percentage of apoptotic cancer cells that release IL1RA in response to genotoxic stress (Fig. 4E). In summary, our data show that the rate of apoptosis and the extracellular concentrations IL1RA correlate in response to treatment with bleomycin.

To test the role of caspases in the release of IL1RA in response to bleomycin, we blocked caspase activity prior to treatment of fibroblasts with bleomycin. Chemical inhibition of caspases using the general caspase inhibitor Z-VAD-FMK impaired the release of IL1RA in response to bleomycin (Fig. 5A; P < 0.05). Similarly, pretreatment with the irreversible caspase-9 inhibitor Z-LEHD-FMK or the caspase-3 inhibitor Z-DEVD-FMK reduced the release of IL1RA in response to bleomycin (Fig. 5B and C; P < 0.01). Collectively, our results demonstrate that bleomycin-induced release of IL1RA partially depends on caspase-9 and caspase 3–mediated apoptosis.

To investigate if the amounts of IL1RA released in response to bleomycin inhibit IL1-induced upregulation of IL6 transcripts (Fig. 6A), the supernatant of bleomycin and mock-treated fibroblasts was transferred to a new set of fibroblasts, which were subsequently treated with IL1α. Supernatant from bleomycin-treated fibroblast cultures impaired the IL1-induced upregulation of IL6 transcripts when compared with supernatant from mock-treated fibroblasts (Fig. 6A; P < 0.001). The inhibitory effect of the supernatant from bleomycin-treated fibroblasts depended on IL1RA, as transfer of supernatant from bleomycin-treated IL1RA^−/− fibroblasts failed to inhibit the upregulation of IL6 transcripts in response to IL1α (Fig. 6B). Similarly, IL1RA from bleomycin-treated fibroblast cultures impaired the IL1-induced upregulation of IL6 transcripts in J774.2 cells transfected with poly (dA:dT), an inflammasome inducer (Fig. 6C). Hence, our data show that IL1RA released in response to DNA damage is able to antagonize the binding of IL1 to the IL1R.

Intracellular IL1α is constitutively present in the cytosol of many cells (6). During apoptosis, cytosolic IL1α translocates to the nucleus and binds to chromatin (9). IL1α bound to chromatin fails to promote inflammation. In addition, phagocytosis of apoptotic cells by macrophages and immature dendritic cells is associated with reduced secretion of IL1 (4, 28). Here, we show that caspase-mediated apoptosis in response to DNA damage leads to the release of biologic significant amounts of IL1RA. The IL1RA is an important negative regulator of inflammatory responses (10). IL1RA exists as a secreted isoform (sIL1RA) and intracellular isoforms (icIL1RA). The secreted isoform is primarily produced by monocytes and macrophages. sIL1ra is the only gene in the IL1ra family, which encodes a signal peptide required for the secretion via the ER–Golgi complex secretory pathway (29–31). In contrast, icIL1RA is constitutively expressed in the cytoplasm of fibroblasts, keratinocytes, and other cells as a leaderless protein. icIL1RA consists of a number of isoforms generated by alternative splicing and translation initiation. Like its secreted counterpart, icIL1RA functions as an extracellular receptor antagonist of IL1 when released from cells (32). Our data suggest that icIL1RA is released nonclassical secretion pathways in response to bleomycin treatment as inhibitors of the classical endoplasmic secretion pathways or N-glycosylation, which can enhance secretion of proteins, did not abrogate the release of IL1RA (33). It is likely that presynthesized icIL1RA is released by fibroblasts and other cells as IL1ra transcripts and intracellular IL1RA levels decreased in response to bleomycin treatment. In addition, de novo translation was not required for the bleomycin-induced release of IL1RA. Many IL1 family members lack classical signal sequences necessary for secretion via the ER–Golgi complex pathway. These cytokines are released by autophagy, necrotic cells, or poorly defined nonclassical secretion pathways (34). No changes in the expression of the autophagy markers LC3-I, LC3-II, and p62 were observed after treatment with bleomycin (data not shown). Furthermore, pretreatment of fibroblasts with the autophagic sequestration blocker 3-methyladenine did not impair bleomycin-induced IL1RA secretion (data not shown), indicating that the release of IL1RA is not regulated by autophagy. IL1RA concentrations in the supernatant of bleomycin-treated fibroblasts correlated with the percentage of apoptotic but necrotic cells. Furthermore, inhibition of the apoptosis-inducing caspase 9 and caspase 3 impaired the release of IL1RA, supporting the conclusion that IL1RA is released by apoptotic cells.

IL1 contributes to the development of certain human cancers and has been suggested as a potential therapeutic target in human cancers (35, 36). Mutated p53 can sustain IL1β-driven tumor malignancy by suppressing the expression of IL1RA (37). p53 is frequently mutated in many different cancers, and mutations of p53 correlate with a decrease in apoptosis in some cancer models (38–40). Hence, our data indicate that the release of IL1RA by apoptotic cells expressing wild-type p53, but not mutated p53, may antagonize a proinflammatory tumor microenvironment promoting tumorigenesis. Consistent with this conclusion, IL1RA released by fibroblasts expressing wild-type p53 in response to bleomycin was able to impair IL1-induced IL6 expression. However, the tested murine cancer cells were able to release IL1RA in response to bleomycin, suggesting that only certain p53 mutations might be able to suppress the expression of IL1ra or that mouse and human IL1RA are regulated differentially by p53.

No potential conflicts of interest were disclosed.

Conception and design: J.Y. Chwee, M. Khatoo, N.Y. Jie Tan, S. Gasser

Development of methodology: J.Y. Chwee, M. Khatoo, N.Y. Jie Tan, S. Gasser

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.Y. Chwee

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.Y. Chwee, S. Gasser

Writing, review, and/or revision of the manuscript: M. Khatoo, N.Y. Jie Tan, S. Gasser

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S. Gasser

Study supervision: S. Gasser

Other (generated data for Figs. 1A–E, 2A and B, 3B and C, 5: J.Y. Chwee

The authors thank Dr. P. Hutchinson for technical assistance in flow cytometry.

This work was supported by Biomedical Research Council grant 07/1/21/19/513 and the National Research Foundation grant Hebrew University of Jerusalem-CREATE—Cellular and Molecular Mechanisms of Inflammation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.