Oxidant-induced apoptosis is mediated by oxidation of the actin-regulatory protein cofilin

Abstract

Physiological oxidants that are generated by activated phagocytes comprise the main source of oxidative stress during inflammation. Oxidants such as taurine chloramine (TnCl) and hydrogen peroxide (H(2)O(2)) can damage proteins and induce apoptosis, but the role of specific protein oxidation in this process has not been defined. We found that the actin-binding protein cofilin is a key target of oxidation. When oxidation of this single regulatory protein is prevented, oxidant-induced apoptosis is inhibited. Oxidation of cofilin causes it to lose its affinity for actin and to translocate to the mitochondria, where it induces swelling and cytochrome c release by mediating opening of the permeability transition pore (PTP). This occurs independently of Bax activation and requires both oxidation of cofilin Cys residues and dephosphorylation at Ser 3. Knockdown of endogenous cofilin using targeted siRNA inhibits oxidant-induced apoptosis, which is restored by re-expression of wild-type cofilin but not by cofilin containing Cys to Ala mutations. Exposure of cofilin to TnCl results in intramolecular disulphide bonding and oxidation of Met residues to Met sulphoxide, but only Cys oxidation causes cofilin to induce mitochondrial damage.

Figure 1

TnCl causes dissociation of actin/cofilin complexes and translocation of cofilin to the mitochondria independently of Bax activation. (a) Subcellular fractionation was carried out on control (–) or TnCl-treated (+) JLP-119 lymphoma cells (0.5 mM TnCl in PBS for 1 h) before analysis by a western blot immunoassay using anti-cofilin, anti-Bax, anti-MnSOD (mitochondrial isoform of superoxide dismutase) or anti-GAPDH (as a cytosolic marker) as described in the Methods. (b) Densitometric analysis of cofilin (left panel) or Bax (right panel) bands in the different subcellular compartments shown in a. Bands from treated cells were normalized to each respective control band and to the internal controls from untreated cells. Data are mean ± s.d. from three independent experiments; *P < 0.005 compared with control, **P < 0.001 compared with control (Student's t-test). (c) Wild-type or Bax^–/–/Bak^–/– double knockout mouse embryonic fibroblasts were treated with TnCl (1.5 mM) or VP-16 (100 μM) in complete media for 1 h, after which the medium was replaced. Cell death was evaluated 24 h later by flow cytometry after staining with annexin-V/FITC and propidium iodide. The data are the means from two independent experiments. (d) JLP-119 lymphoma cells were treated with TnCl (0.5 mM) in PBS for 1 h and then cell extracts were subjected to immunoprecipitation (IP) using an anti-cofilin antibody as described in the Methods. Western blot immunoassay (IB) used either anti-actin or anti- cofilin, as indicated in the figure. Mito, mitochondria. See Supplementary Information, Fig. S8 for full scans of a and d.

Figure 2

Oxidized cofilin binds to mitochondria, causing their swelling and cytochrome c release through opening of the PTP. (a) Recombinant human cofilin was oxidized by TnCl in vitro as described in the Methods. The figure shows detection of cofilin disulphides by 5-IAF-labelling and SDS–PAGE (fluorescein fluorescent band). (b) Rat liver mitochondria were exposed to purified, recombinant cofilin (5 μg for 10 min at 37 °C) that was either untreated or previously oxidized (ox-cofilin) with TnCl (100 μM). Mitochondria were treated in the presence or absence of the mitochondrial PTP inhibitors cyclosporine A (CsA; 2 μM) or bongkrekic acid (BA; 50 μM). A sample of ox-cofilin was treated with the sulphydryl reducing agent DTT and then dialysed before adding to the mitochondria to assess the role of cofilin disulphide bonds on its biological activity. The vehicle control contained TnCl (2 μM), which has no direct effect on the mitochondria. Cofilin protein in the mitochondrial (Mito) pellet and cytochrome c in the supernatant were assessed by a western blot immunoassay (30 μg protein per lane). Adenine nucleotide translocase (ANT) was used as a loading control for the mitochondrial pellet fraction. The remaining unbound cofilin and the cytochrome c (Cyt c) released by Ca²⁺ treatment (150 μM CaCl₂) were assessed in the mitochondrial supernatants. (c) Swelling of mitochondria exposed to native or oxidized cofilin, or oxidized bovine serum albumin (ox-BSA; 5 μg) was measured as described in the Methods. Ca²⁺ (150 μM CaCl₂) was used as a positive control for induction of the mitochondrial permeability transition. DTT and CsA or BA treatments were as described in b. Control (vehicle) mitochondria were stable and showed no swelling under the same incubation conditions. The data shown are representative of at least three independent experiments.

Figure 3

TnCl-induced apoptosis is dependent on oxidation of Cys residues in cofilin, as determined by site-directed mutagenesis and siRNA silencing. (a, b) Induction of apoptosis in COS-7 cells 24 h after treatment with TnCl as assessed by (a) flow cytometry and (b) nuclear morphology (see Methods). PI, propidium iodide.(c) Western blot immunoassay of cofilin levels in total cell lysates from COS-7 cells transfected with plasmids expressing wild-type cofilin or cofilin in which the indicated Cys residues were mutated to Ala residues. α-tubulin protein levels served as an internal loading control. The blot shown is representative of three independent experiments. (d) At 24 h after transfection, COS-7 cells were treated with TnCl for 1 h and then the medium was replaced. Apoptosis was assessed by flow cytometry 24 h later and is calculated relative to the levels in mock-transfected cells (lipofectamine plus empty plasmid vector). Data are mean ± s.d. from at least three independent experiments, n = 4, *P ≤ 0.01, **P ≤ 0.001 (compared to wild-type TnCl-treated cells; Student's t-test). Knockdown of endogenous cofilin with siRNA. COS-7 cells were treated with either control siRNA or cofilin siRNA as described in the Methods. After a 48 h incubation, the cells were transfected with an expression plasmid containing either no insert (Mock), the wild-type cofilin gene, or cofilin in which Cys 139 and 147 were mutated to Ala residues (Cys 139, 147). (e) Representative western blot for cofilin protein levels following knockdown and re-expression. (f) A day after transfection, the cells were treated for 1 h with TnCl (1.5 mM) and then washed. Apoptosis was assessed by flow cytometry 24 h after TnCl treatment and is calculated relative to the levels in control siRNA, mock transfected cells. Data are mean ± s.d. from three independent experiments, *P ≤ 0.01, **P ≤ 0.001 (compared to control siRNA, mock transfected TnCl-treated cells; Student's t-test). There was no significant difference between the levels of TnCl- induced apoptosis in cofilin-knockdown cells transfected with mock plasmid versus plasmid expressing Cys-Ala mutated cofilin (#). Conditions for cofilin silencing are shown in Supplementary Information, Fig. S3. See Supplementary Information, Fig. S8 for full scans of c and e.

Figure 4

Oxidation of cofilin is required for its translocation to the mitochondria. (a) COS-7 cells were transfected with expression plasmids containing wild-type cofilin (left) or cofilin with Cys to Ala mutations at residues 139 and 147 (C139, 147A; right). At 48 h after transfection, cells were treated with TnCl (1.5 mM) for 3 h in complete medium, washed, collected and lysed as described in the Methods. Subcellular fractionation was carried out on control (–) or TnCl-treated (+) cells before analysis by a western blot immunoassay using anti-cofilin, anti-COX IV (as a mitochondrial marker) or anti-GAPDH (as a cytosolic marker). (b) Densitometric analysis of cofilin bands in the different subcellular compartments shown in a. Data are normalized to the relative densities of the appropriate marker bands for each subcellular fraction. Mito, mitochondria. See Supplementary Information, Fig. S8 for full scan of a.

Figure 5

Cofilin must be dephosphorylated and oxidized for TnCl to cause apoptosis and loss of cellular mitochondrial membrane potential. (a) COS-7 cells were transfected for 6 h with expression plasmids containing different constructs of wild-type or mutated cofilin. Total cell lysates were extracted 48 h after transfection. Expression of wild-type and mutant cofilin proteins was followed by a western-blot immunoassay using an anti-cofilin antibody. α-tubulin protein was used as an internal control for protein loads. The blot shown is representative of two independent experiments. (b) The transfected cells described in a were treated with TnCl (1.5 mM) for 1 h. Apoptosis was assessed by flow cytometry 24 h after TnCl treatment and is calculated relative to the levels in mock transfected cells (lipofectamine plus empty vector). Data are the means from two independent experiments. (c) COS-7 cells transfected as in a were treated with TnCl (1.5 mM), washed and incubated with JC-1 dye (10 μg ml^–1) at 37 °C for 20 min. The loss of mitochondrial membrane potential (ΔΨm) as a function of incubation time was assayed using the ratio of JC-1 red (590 nm) to green (540 nm) fluorescence and expressed as a percentage of control (untreated) cells. Data are the means from two independent experiments carried out in duplicate. S3A, Ser 3 mutated to Ala; S3D, Ser 3 mutated to Asp; C139,147A, Cys 139 and 147 mutated to Ala; Cys S3A, combined mutation of Cys 139 and 147, and Ser 3 to Ala; Cys S3D, combined mutation of Cys 139 and 147 to Ala, and Ser 3 to Asp. See Supplementary Information, Fig. S8 for full scan of a.