Nox1 NADPH oxidase is necessary for late but not early myocardial ischaemic preconditioning

Abstract

Aims: Ischaemic preconditioning (IPC) is an adaptive mechanism that renders the myocardium resistant to injury from subsequent hypoxia. Although reactive oxygen species (ROS) contribute to both the early and late phases of IPC, their enzymatic source and associated signalling events have not yet been understood completely. Our objective was to investigate the role of the Nox1 NADPH oxidase in cardioprotection provided by IPC.

Methods and results: Wild-type (WT) and Nox1-deficient mice were treated with three cycles of brief coronary occlusion and reperfusion, followed by prolonged occlusion either immediately (early IPC) or after 24 h (late IPC). Nox1 deficiency had no impact on the cardioprotection afforded by early IPC. In contrast, deficiency of Nox1 during late IPC resulted in a larger infarct size, cardiac remodelling, and increased myocardial apoptosis compared with WT hearts. Furthermore, expression of Nox1 in WT hearts increased in response to late IPC. Deficiency of Nox1 abrogated late IPC-mediated activation of cardiac nuclear factor-κB (NF-κB) and induction of tumour necrosis factor-α (TNF-α) in the heart and circulation. Finally, knockdown of Nox1 in cultured cardiomyocytes prevented TNF-α induction of NF-κB and the protective effect of IPC on hypoxia-induced apoptosis.

Conclusions: Our data identify a critical role for Nox1 in late IPC and define a previously unrecognized link between TNF-α and NF-κB in mediating tolerance to myocardial injury. These findings have clinical significance considering the emergence of Nox1 inhibitors for the treatment of cardiovascular disease.

Keywords: Ischaemic preconditioning; Myocardial infarction; NF-κB; Nox1; TNF-α.

Figure 1

Nox1 is necessary for the effect of late but not early IPC on infarct size reduction. Representative images and summary data of TTC-stained myocardium after early (A) or late (B) IPC. Distance between hatch marks within images = 1 mm. Infarct size is presented at the percentage of total area. Occl, occlusion; Rep, reperfusion. n = 6–14; *P < 0.05 vs. sham, ^ΔP < 0.05 vs. WT late IPC.

Figure 2

Late IPC protection from cardiac remodelling following prolonged ischaemia requires Nox1. (A) Representative echocardiographic images at end-diastole and end-systole. (B–E) Summary data of LV end-diastolic volume (EDV, B), end-systolic volume (ESV, C), ejection fraction (EF, D), and ischaemic zone (IZ, E). Control mice had no coronary occlusion, and sham mice were subjected to 30 min coronary occlusion without IPC; and IPC mice had the late IPC protocol described in Figure 1B. n = 6–12; *P < 0.05 vs. WT control; ⁺P < 0.05 vs. WT sham, ^ΔP < 0.05 vs. WT IPC.

Figure 3

Nox1 mediates the late IPC reduction in cell death following hypoxic injury. (A) Representative images of TUNEL-stained myocardium. Nuclei were counterstained with DAPI. Scale bar = 50 µm. (B) PARP cleavage as assessed by western blotting in whole heart lysates. GAPDH, loading control. Relative densitometric values corrected for GAPDH and normalized to sham are summarized below the blot (n = 3, *P < 0.05 vs.WT sham). (C and D) Following knockdown of Nox1, apoptosis (C) or cell viability (D) was examined in cultured cardiomyocytes after exposure to hypoxia in the absence or presence of preconditioning (PC). shGFP, non-silencing control. n = 3; *P < 0.05 vs. shGFP no treatment, ^‡P < 0.05 vs. shGFP hypoxia, ^ΔP < 0.05 vs. shGFP hypoxia/PC.

Figure 4

Cardiac Nox1 expression is induced in response to late IPC. (A and B) Baseline mRNA (A) and protein (B) levels in whole heart lysates. Data in (A) are expressed relative to Nox1 levels in WT hearts. (C and D) Effect of late IPC on mRNA (C) and protein (D) levels in WT hearts. Lysates were collected after late IPC and prior to prolonged occlusion. (E) Immunostaining for Nox1 in the ischaemic risk area in WT or Nox1^−/y hearts after sham or late IPC (prior to prolonged occlusion); scale bar = 40 µm. In (B) and (D), relative densitometric values corrected for GAPDH and normalized to WT (B) or sham only (D) are listed below the blots (n = 3 and 4–6, respectively). Data in (C) are expressed relative to sham levels (n = 7–9); *P < 0.05 vs. sham.

Figure 5

Late IPC activation of NF-κB is Nox1-dependent. (A) Representative images and summary data of NF-κB activation as determined by electrophoretic mobility shift assay. Lysates were collected 24 h after late IPC and prior to prolonged occlusion; n = 3 independent experiments; *P < 0.05 vs. WT sham. (B) Light micrographs of myocardium after immunostaining for the active form of NF-κB (phospho-p65) 24 h after late IPC. The no primary antibody control showed no staining; scale bar = 100 µm. (C) NF-κB p65 subunit phosphorylation (p-p65) in whole heart lysates collected as in (A). Data were normalized to GAPDH and expressed relative to sham levels in WT hearts; n = 4–5.

Figure 6

Nox1 is necessary for induction of TNF-α in late IPC. (A and B) TNF-α mRNA (A) and protein (B) levels in myocardium 24 h after late IPC and prior to prolonged occlusion. Data are normalized to WT sham; n = 5–7; *P < 0.05 vs. WT sham, ^ΔP < 0.05 vs. WT IPC only. (C) Change in serum TNF-α levels at early (30 m) and late (24 h) time points after IPC as measured by ELISA. There was no prolonged occlusion after IPC. Data are presented as the change from sham; n = 4–7; *P < 0.05 vs. WT 30 min, ⁺P < 0.05 vs. WT 24 h. (D) NF-κB activity was measured by luciferase reporter assay in cultured cardiomyocytes after Nox1 knockdown (Ad-shNox1) and treatment with TNF-α (10 µmol/L). Data are normalized to cells infected with Ad-shGFP and treated with vehicle; n = 3; *P < 0.05 vs. shGFP vehicle; ⁺P < 0.05 vs. shGFP TNF-α, ^ΔP < 0.05 vs. shGFP vehicle, ^‡P < 0.05 vs. shNox1 vehicle. (E) Nox1 mRNA levels in cultured cardiomyocytes after treatment with TNF-α (10 µmol/L); n = 3; *P < 0.05 vs. vehicle.