Intermittent hypoxia-induced cardiomyopathy and its prevention by Nrf2 and metallothionein

Abstract

The mechanism for intermittent hypoxia (IH)-induced cardiomyopathy remains obscure. We reported the prevention of acute and chronic IH-induced cardiac damage by selective cardiac overexpression of metallothionein (MT). Herein we defined that MT-mediated protection from IH-cardiomyopathy is via activation of nuclear factor erythroid 2-related factor 2 (Nrf2), a critical redox-balance controller in the body. For this, mice were exposed to IH for 3 days (acute) or 4 or 8 weeks (chronic). Cardiac Nrf2 and MT expression in response to IH were significantly increased acutely yet decreased chronically. Interestingly, cardiac overexpression (Nrf2-TG) or global deletion of the Nrf2 gene (Nrf2-KO) made mice highly resistant or highly susceptible, respectively, to IH-induced cardiomyopathy and MT expression. Mechanistically, 4-week IH exposure significantly decreased cardiac Nrf2 binding to the MT gene promoter, and thus, depressed both MT transcription and translation in WT mice but not Nrf2-TG mice. Likewise, cardiac MT overexpression prevented chronic IH-induced cardiomyopathy and down-regulation of Nrf2 likely via activation of a PI3K/Akt/GSK-3β/Fyn-dependent signaling pathway. These results reveal an integrated relationship between cardiac Nrf2 and MT expression in response to IH -- acute compensatory up-regulation followed by chronic down-regulation and cardiomyopathy. Cardiac overexpression of either Nrf2 or MT offered cardioprotection from IH via complicated PI3K/Akt/GSK3B/Fyn signaling. Potential therapeutics may target either Nrf2 or MT to prevent chronic IH-induced cardiomyopathy.

Keywords: Intermittent hypoxia; Metallothionein; Nuclear factor erythroid 2-related factor 2; Obstructive sleep apnea; Redox regulation.

Figure 1.. Expression of Nrf2 in response to IH exposures.

FVB WT mice were exposed to IH for indicated times. Cardiac Nrf2 (A) expression was measured by Western blots with its antibody from Abcam (ab137550) with molecular weight of ~ 95 – 110 as suggested by Lau et al. [60]. The NQO1 (B) and SOD2 (C) mRNA was measured by RT-PCR and Western blots. Data are presented as mean ± SD (n=5). *, p<0.05 vs control.

Figure 2.. Nrf2-KO mice exacerbate IH-induced cardiac dysfunction.

Nrf2-KO and C57BL/6J WT mice were exposed to IH for indicated times. LVID;s and LVID;d (A), and LVEF and LVFS (B), and LVPW;s and LVPW;d (C), and IVS;s and IVS;d (D) were measured by echocardiography. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 3.. Nrf2-KO mice exacerbate IH-induced oxidative stress, inflammation and fibrosis.

Nrf2-KO and C57BL/6J WT mice were exposed to IH for indicated times. Cardiac oxidative damage was measured by Western blots for 3-nitrotyrosine (3-NT, A) and 4-Hydroxynonenal (4-HNE, B). Cardiac inflammation and fibrosis were measured by Western blots for vascular cell adhesion molecule 1 (VCAM-1, C), plasminogen activator inhibitor-1 (PAI-1, D), and connective tissue growth factor (CTGF, F). Cardiac collagen was measured by Sirius-red staining (E). MT (G) expression was measured by Western blots. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 4.. Nrf2 and its down-stream gene expression in Nrf2-TG and WT mice exposed to 4-week IH.

Expression of Nrf2 (A), NQO1 (B), and SOD2 (C) was examined by Western blots. When WB was carried out whole full membrane was separated into five horizontal strips based on Nrf2, NQO1, SOD and GAPDH molecule weights. These strips were separately incubate with each specific antibody and associated following procedures. Under this condition, therefore, all Nrf2, NQO1, and SOD2 expression were compared with the same GAPDH expression. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 5.. Nrf2-TG mice are resistant to 4-week IH-induced cardiac dysfunction.

Nrf2-TG and FVB WT mice were exposed to IH for indicated times. LVID;s and LVID;d (A), and LVEF and LVFS (B), and LVPW;s and LVPW;d (C), and IVS;s and IVS;d (D) were measured by echocardiography. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 6.. Nrf2-TG mice are resistant to 4-week IH-induced cardiac oxidative stress, inflammation and fibrosis.

FVB and Nrf2-TG mice were exposed to IH for 4 weeks. Cardiac oxidative damage was measured by Western blots for 3-NT (A) and 4-HNE (B). Cardiac inflammation was measured by Western blots for VCAM (C) and PAI-1 (D). Cardiac fibrosis was measured by Sirius-red staining (E) and Western blots for CTGF (F). MT (G) expression was measured by Western blots. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 7.. Reciprocal regulation between Nrf2 and MT.

MT-KO, MT-TG and their WT mice were exposed to IH for indicated times (A-C) and 4 weeks (D-F). Nrf2 and its target genes NQO1 and SOD2 expression was measured by Western blots. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 8.. Banding of Nrf2 to the promoter of MT stimulates its expression.

FVB and Nrf2-TG mice were exposed to IH for 4 weeks. CHIP analysis of Nrf2 level at MT1 promoter with antibody against Nrf2 (A). MT expression in each group was determined at mRNA level with real-time-PCR (B). Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; &, p<0.05 vs WT/H.

Figure 9.. The effect of MT-KO mice and MT-TG mice on IH-induced PI3K/GSK-3β/Akt/ Fyn expression.

MT-KO and 129S1 mice were exposed to IH for indicated times. MT-TG mice and FVB mice were exposed to IH for 4 weeks. The expression of p-PI3K and t-PI3K (A and E), p-Akt and t-Akt (B and F), p-GSK-3β and t-GSK-3β (C and G), Fyn (D and H) in heart was detected by Western blot assay and the ratio of p-PI3K and t-PI3K, p-Akt and t-Akt, p-GSK-3β and t-GSK-3β was presented. Data are presented as mean ± SD (n=5). *, p<0.05 vs WT/C; & p<0.05 vs WT/H.

Figure 10.. Effect of PI3K inhibition on cardiac function .

FVB mice were exposed to IH for 3 days with either PI3K inhibitor (LY294002) or vehicle simultaneously. LVID;s and LVID;d (A), and LVEF and LVFS (B), and LVPW;s and LVPW;d (C), and IVS;s and IVS;d (D) were measured by echocardiography. Data are presented as mean ± SD (n=5).

Figure 11.. Effect of PI3K inhibition on IH-induced Nrf2 expression.

FVB mice were exposed to IH for 3 days with either PI3K inhibitor (LY294002) or vehicle simultaneously. The expression of p-PI3K and t-PI3K (A), p-Akt and t-Akt (B), p-GSK-3β and t-GSK-3β (C), Fyn (D), Nrf2 (E) and its target genes NQO1 (F), SOD2 (G) in heart was detected by Western blot assay and the ratio of p-PI3K and t-PI3K, p-Akt and t-Akt, p-GSK-3β and t-GSK-3β was presented. Data are presented as mean ± SD (n=5). *, p<0.05 vs control; &, p<0.05 vs H.

Figure 12.. Diagram of the mechanism of Nrf2 and MT in preventing IH–induced cardiac injury.

In response to acute IH exposure, cardiac expression of Nrf2 and MT was increased in parallel as a compensated mechanism to protect the heart from IH-induced injury; however, in response to chronic IH exposure, cardiac expression of both MT and Nrf2 at the late stage was gradually decreased from a compensated stage to a decompensated stage, which making the heart more susceptible to IH-induced oxidative damage, resulting in the late cardiac remodeling and dysfunction (IH cardiomyopathy). During compensative process, MT expression was significantly affected positively by Nrf2 expression via binding to the promoter of MT, suggesting that MT as one of Nrf2 down-stream targets. Interesting, MT has expression in response to IH has certain feed-back impact on the function of Nrf2 via PI3K/Akt/GSK-3β-mediated inhibition of Fyn, Nrf2-functional negative regulator by moving it from nuclei into cytosol.