Neutralization of interleukin-18 ameliorates ischemia/reperfusion-induced myocardial injury

Abstract

Ischemia/reperfusion (I/R) injury is characterized by the induction of oxidative stress and proinflammatory cytokine expression. Recently demonstrating that oxidative stress and TNF-alpha each stimulate interleukin (IL)-18 expression in cardiomyocytes, we hypothesized that I/R also induces IL-18 expression and thus exacerbates inflammation and tissue damage. Neutralization of IL-18 signaling should therefore diminish tissue injury following I/R. I/R studies were performed using a chronically instrumented closed chest mouse model. Male C57BL/6 mice underwent 30 min of ischemia by LAD coronary artery ligation followed by various periods of reperfusion. Sham-operated or ischemia-only mice served as controls. A subset of animals was treated with IL-18-neutralizing antibodies 1 h prior to LAD ligation. Ischemic LV tissue was used for analysis. Our results demonstrate that, compared with sham operation and ischemia alone, I/R significantly increased (i) oxidative stress (increased MDA/4-HNE levels), (ii) neutrophil infiltration (increased MPO activity), (iii) NF-kappaB DNA binding activity (p50, p65), and (iv) increased expression of IL-18Rbeta, but not IL-18Ralpha or IL-18BP transcripts. Administration of IL-18-neutralizing antibodies significantly reduced I/R injury measured by reduced infarct size (versus control IgG). In isolated adult mouse cardiomyocytes, simulated ischemia/reperfusion enhanced oxidative stress and biologically active IL-18 expression via IKK-dependent NF-kappaB activation. These results indicate that IL-18 plays a critical role in I/R injury and thus represents a promising therapeutic target.

FIGURE 1.

Reperfusion following 30 min of ischemia stimulates free radical generation and neutrophil infiltration. Male C57Bl/6 mice underwent 30 min of LAD coronary artery ligation, followed by reperfusion for various periods of time. A, details of the I/R protocol are shown. B, I/R induced free radical generation. Free radical generation was analyzed at 2 h of reperfusion by measuring the lipid peroxidation products MDA and 4-HNE in postischemic reperfused LV as described under “Experimental Procedures.” Ischemia alone and sham-operated animals served as controls. ^*, p < 0.001 (versus sham-operated; n = 4/group). C, ischemia/reperfusion induced neutrophil infiltration. MPO, a biochemical marker for neutrophil infiltration into tissues, was measured in postischemic reperfused myocardium using a colorimetric assay, as described under “Experimental Procedures.” ^*, p < 0.001 (versus sham and 30 min of ischemia alone).

FIGURE 2.

Ischemia/reperfusion activates NF-κB in postischemic myocardium. A, I/R induced NF-κB DNA-binding activity. Following 30 min of ischemia/2 h of reperfusion, nuclear proteins were extracted from postischemic reperfused LV tissue and analyzed for NF-κB DNA binding activity by EMSA as described under “Experimental Procedures.” Sham-operated and ischemia-alone groups served as controls (n = 4/group). Control studies determining the specificity of NF-κB oligonucleotide was shown in the completion experiments (lanes 1–4). Lane 1, protein extract from 30-min ischemia/2-h reperfusion myocardium was preincubated with a 75-fold molar excess of unlabeled double-stranded consensus NF-κB oligonucleotide, followed by the addition of labeled κB-specific probe. Lane 2, competition with cold mutant NF-κB oligonucleotide. Protein extract from 30-min ischemia/2-h reperfusion myocardium was preincubated with a 75-fold molar excess of unlabeled double-stranded mutant NF-κB oligonucleotide, followed by the addition of labeled κB-specific probe. Lane 3, protein extract from 30-min ischemia/2-h reperfusion myocardium was preincubated with a 75-fold molar excess of unlabeled double-stranded consensus Oct1 oligonucleotide, followed by the addition of labeled κB-specific probe. Lane 4, no nuclear protein, but the sample contains [γ-³²P]ATP-labeled κB-specific probe. Lane 5, protein extract from 30-min ischemia/2-h reperfusion myocardium was incubated with ³²P-labeled mutant NF-κB oligonucleotide. Lanes 6–9, protein extract from LV tissue from Sham-operated animals + κB-specific probe. Lanes 10–13, protein extract from ischemic LV tissue from 30 min of ischemia + κB-specific probe. Lanes 14–17, protein extract from LV tissue from postischemic reperfused LV tissue + κB-specific probe. Solid arrow, NF-κB-specific DNA-protein complexes. Solid circle, free probe. B, I/R-induced NF-κB activation was composed of p50 and p65. The nuclear extracts used in lane 14 in A were preincubated with p50 or p65 antibodies for 40 min prior to incubation with κB-specific probe followed by EMSA (n = 3). Normal IgG served as a control (Control IgG). Solid arrow, NF-κB-specific binding region. Open arrow, supershifted bands. Solid circle, free probe. C, I/R had no effect on basal Oct1 DNA binding activity. Nuclear protein extracts described in A were analyzed for Oct1 DNA binding by EMSA (n = 3). Solid arrow, Oct1-specific DNA-protein complexes. Solid circle, free probe. D, I/R-induced NF-κB activation was quantified by ELISA. Nuclear extracts described in A were analyzed for NF-κB activation and subunit composition by ELISA, as described under “Experimental Procedures” (n = 4/group, p < at least 0.001 versus corresponding Sham or ischemia (I) alone).

FIGURE 3.

Ischemia/reperfusion induces IL-18 expression in serum and postischemic myocardium. A, I/R induced IL-18 mRNA expression in a time-dependent manner. IL-18 mRNA in total RNA extracts was quantified by RT-qPCR. β-actin served as an internal control. ^*, p < 0.05; ^**, p < 0.001 versus sham and ischemia (I) alone (n = 4/group). B, I/R-induced IL-18 mRNA expression at 2 h of reperfusion was confirmed by Northern blot analysis. Each lane contains 20 μg of total RNA from an individual animal. 28 S rRNA served as an internal control and shows similar levels of RNA loading in each lane. C, I/R induced IL-18 protein expression in a time-dependent manner. IL-18 protein levels were quantified by immunoblotting. ^*, p < 0.01; ^**, p < 0.001 versus sham and ischemia (I) alone (n = 4/group). Representative immunoblotting at 2 h of reperfusion is shown in D. Each lane contains 30 μg of protein extract from an individual animal. β-Actin served as an internal control and shows similar levels of protein loading in each well. E, IL-18 levels in myocardial protein extracts were quantified by ELISA. ^*, p < 0.05; ^**, p < at least 0.01 versus Sham and ischemia (I) alone (n = 4/group). F, I/R enhanced systemic IL-18 levels in a delayed manner. Serum levels of IL-18 at the indicated time periods were quantified by ELISA. ^*, p < 0.05; ^**, p < 0.01 versus sham and ischemia (I) alone (n = 4/group).

FIGURE 4.

I/R induces robust expression of IL-18Rβ, but not IL-18Rα, and causes delayed expression of IL-18BP mRNA. A, I/R failed to modulate IL-18Rα mRNA expression. IL-18Rα mRNA expression was quantified by RT-qPCR (n = 4). β-Actin served as a control. B, I/R induced IL-18Rβ mRNA expression in a time-dependent manner. IL-18Rβ mRNA expression was quantified by RT-qPCR. β-Actin served as a control. ^*, p < 0.05; ^**, p < 0.01 versus sham at 3 h and 30 min of ischemia (I) alone (n = 4/group). C, I/R-induced IL-18Rα and IL-18Rβ mRNA expressions at 3 h of reperfusion were confirmed by Northern blot analysis. Each lane contains 20 μg of total RNA from an individual animal. 28 S rRNA served as an internal control and shows similar levels of RNA loading in each lane. D, I/R induced IL-18BP mRNA expression in a delayed manner. IL-18BP mRNA expression was quantified by RT-qPCR. β-Actin served as a control. ^*, p < 0.01 versus Sham at 6 h and 30 min of ischemia alone (n = 4/group). E, I/R-induced IL-18BP expression was confirmed at 6 h of reperfusion by Northern blot analysis. Each lane contains 20 μg of total RNA from an individual animal. 28 S rRNA served as an internal control and shows similar levels of RNA loading in each lane.

FIGURE 5.

Administration of IL-18-neutralizing antibodies attenuates I/R-induced tissue injury. A, male C57Bl/6 mice underwent 30 min of LAD coronary artery ligation followed by reperfusion for 24 h. Mice were administered with anti-IL-18-neutralizing antibodies 1 h prior to I/R. Details of the I/R protocol and that of intervention are shown. B, IL-18-neutralizing antibodies attenuated I/R-induced tissue injury. IL-18-neutralizing antibodies (500 μg/mouse, intravenously) were administered 1 h prior to LAD coronary artery ligation. Normal rat IgG served as a control. After 30 min of ischemia/24 h of reperfusion, animals were sacrificed, and infarct size was measured (n = 12/group). A representative TTC-stained tissue is shown in the inset.^*, p < 0.01 versus control (untreated) or normal rat IgG-treated.

FIGURE 6.

sI/R activates NF-κB in isolated adult mouse cardiomyocytes. A, adult mouse cardiomyocytes underwent 30 min of ischemia followed by reoxygenation for up to 4 h. Details of the sI/R and of intervention are shown. B, sI/R did not induce cell death during the 4-h study period. Cardiomyocytes were exposed to normoxia, ischemia for 30 min, or sI/R for 4 h. Cell death was analyzed by ELISA as described under “Experimental Procedures.” Doxorubicin hydrochloride (Dox.; 1 μm for 24 h) was used as a positive control. ^*, p < 0.001 versus control (normoxia) and ischemia alone. C, sI/R induced oxidative stress. Oxidative stress was quantified at 4 h after reoxygenation by quantitating the lipid peroxidation products MDA and 4-HNE, as described under “Experimental Procedures.” ^*, p < 0.001 versus normoxia; n = 6). D, sI/R-mediated oxidative stress was confirmed by DCF fluorescence. Cardiomyocytes loaded with the ROS-sensitive fluorophore dichlorofluorescein diacetate were exposed to ischemia alone or sI/R. Representative fluorescent micrographs at the indicated time periods are shown (n=3). E, sI/R activated NF-κB DNA binding activity. Cardiomyocytes were exposed to sI/R for upto 2 h, and nuclear extracts were assessed for NF-κB DNA binding activity by EMSA, as described under “Experimental Procedures” (n = 3/group). A representative of three independent experiments is shown. Arrow, NF-κB-specific DNA-protein complexes. F, sI/R-mediated NF-κB p65 translocation into the nucleus. Cardiomyocytes underwent sI/R for up to 2 h, and nuclear protein extracts were analyzed for p65 levels by ELISA (n = 3; ^*, p < 0.05; ^**, p < 0.01 versus normoxia). G, ROS were generated prior to NF-κB activation. Results from D and E are compared to demonstrate that ROS (oxidative stress) are generated prior to activation of NF-κB in cardiomyocytes following sI/R. ^*, p < 0.01 versus normoxia; †, p < 0.05; ††, p < 0.001 versus normoxia (n = 3). H, sI/R stimulated NF-κB-driven reporter gene activity. Cardiomyocytes were transduced with adenoviral NF-κB-Luc vector and 24 h later were exposed to sI/R (n = 12). Cotransfection with a β-galactosidase vector (Ad.β-galactosidase) was used to control for transfection efficiency. Ad.MCS-Luc served as a control. Firefly luciferase andβ-galactosidase levels were assayed after 12 h. ^*, p < 0.001 versus normoxia. I, sI/R induces p65 phosphorylation. Cardiomyocytes were exposed to sI/R for 2 h, and nuclear extracts were analyzed for phospho-p65 levels by immunoblotting. A representative of three independent experiments is shown. Lamin A/C levels demonstrate the purity and equal loading of the nuclear extracts in each lane.

FIGURE 7.

Simulated ischemia/reoxygenation induces NF-κB activation via IKK. A, adenoviral transduction of dnIKKβ blunted sI/R-mediated IKKβ activity. Cardiomyocytes were transfected with Ad.dnIKKβ (100 MOI). Ad.GFP served as a control. After 24 h, cells were exposed to sI/R for 2 h. IKK activity was analyzed by an in vitro kinase assay using glutathione S-transferase-IκBas the substrate (n = 3). Actin served as a loading control. B, sI/R induced IκB-α degradation. Cardiomyocytes were treated as in A but for 1 h and then analyzed for IκB-α degradation by immunoblotting (n = 3). Actin served as a loading control. C, sI/R induced NF-κB activation via IKKβ and IκB-α degradation. Cardiomyocytes were pretreated with the free radical scavenger PDTC (100 μm in PBS for 1 h) or transduced with adenoviral dnIKKβ or dnIκB-α for 24 h, followed by sI/R for 2 h. PBS and Ad.GFP served as controls. NF-κB DNA binding activity was analyzed by EMSA (n = 3). D, the proteasomal inhibitor MG-132 blunted sI/R-mediated NF-κB activation. Cardiomyocytes were pretreated with MG-132 (5 μm in DMSO for 1 h) prior to sI/R for 2 h. DMSO served as a control. NF-κB DNA binding activity was analyzed by EMSA (n = 3). The arrows in C and D indicate NF-κB-specific DNA-protein complexes.

FIGURE 8.

Simulated ischemia/reperfusion induces biologically active IL-18 via IKK-NF-κB signaling. A, PDTC inhibited sI/R-mediated IL-18 mRNA expression. Cardiomyocytes pretreated with PDTC (100 μm in PBS for 1 h) were exposed to sI/R for 4 h, and IL-18 mRNA expression was analyzed by Northern blotting. 28 S rRNA served as a control. A representative of three independent experiments is shown. B, sI/R induced IL-18 mRNA expression via IKK and IκB-α degradation. Cardiomyocytes transduced with adenoviral dnIKKβ, dnIκB-α, or GFP for 24 h were exposed to sI/R for 4 h. IL-18 mRNA expression was analyzed as in A (n = 3). TNF-α (10 ng/ml for 4 h) served as a positive control. C, sI/R-mediated IL-18 expression was independent of TNF-α. Cardiomyocytes were incubated with TNF-α-neutralizing antibodies or control IgG (10 μg/ml for 1 h) prior to sI/R. IL-18 protein levels were analyzed by immunoblotting (n = 3). Actin served as a loading control. D, TNF-α-neutralizing antibodies attenuated TNF-α-induced IL-18 expression. To demonstrate the efficacy of TNF-α-neutralizing antibodies, cardiomyocytes were incubated with αTNF-α antibodies or control IgG (10 μg/ml for 1 h) prior to TNF-α addition (10 ng/ml for 4 h). IL-18 protein levels were quantified by immunoblotting (n = 3). Actin served as a loading control. E, PDTC inhibited sI/R-mediated IL-18 protein expression. Cardiomyocytes treated as in A were analyzed for IL-18 protein expression by immunoblotting (n = 3). Actin served as a control. F, sI/R induced IL-18 protein expression via IKKβ and IκB-α degradation. Cardiomyocytes treated as in B were analyzed for IL-18 protein levels by immunoblotting. TNF-α served as a positive control. G, the proteasomal inhibitor MG-132 and the IKK inhibitor SC-514 inhibited sI/R-mediated IL-18 protein expression. Cardiomyocytes pretreated with MG-132 (5 μm in DMSO for 1 h) or SC-514 (10 μm in DMSO for 1 h) were exposed to sI/R for 4 h. IL-18 protein expression was analyzed by immunoblotting (n = 3). Actin served as a control. H, dnIKK and dnIκB-α blunted TNF-α-induced NF-κB activation. Cardiomyocytes transfected as in B were treated with TNF-α (10 ng/ml for 1 h) and were analyzed for nuclear NF-κB p65 levels by ELISA (n = 3). ^*, p < 0.001 versus untreated; †, p < 0.01 versus TNF-α. I, dnIKK and dnIκB-α failed to modulate sI/R-mediated JNK phosphorylation. Cardiomyocytes transfected as in B were exposed to sI/R for 2 h. Total and phospho-JNK levels in cleared cell lysates were analyzed by immunoblotting (n = 3). J, sI/R stimulated IL-18 secretion via IKK and IκB-α degradation. Cardiomyocytes were treated as in A and B, and IL-18 levels in culture supernatants were analyzed for IL-18 levels by ELISA (n = 6). TNF-α served as a positive control (10 ng/ml for 4 h). The levels in the treated samples were normalized to the normoxic samples (equal to 1), and the results are expressed as -fold induction from untreated (normoxic). ^*, p < 0.01 versus untreated; †, p < 0.05 versus sI/R. K, sI/R-stimulated IL-18 was biologically active. Mouse splenocytes (5 × 10⁶/ml) were incubated with and without cardiomyocyte-derived culture supernatants from G and LPS (1 μg/ml) for 24 h at 37 °C. After incubation, supernatants were analyzed for IFN-γ production by ELISA (n = 6). To demonstrate IL-18 specificity, culture supernatants were preincubated with anti-IL-18-neutralizing antibodies (10 μg/ml for 1 h at 37 °C). Normal IgG at similar concentrations served as a control. ^*, p < 0.001 versus untreated; †, p < 0.05 versus 200 μl of culture supernatant + LPS (n = 6).