A Novel Role of Claudin-5 in Prevention of Mitochondrial Fission Against Ischemic/Hypoxic Stress in Cardiomyocytes

Abstract

Background: Downregulation of claudin-5 in the heart is associated with the end-stage heart failure. However, the underlying mechanism ofclaudin-5 is unclear. Here we investigated the molecular actions of claudin-5 in perspective of mitochondria in cardiomyocytes to better understand the role of claudin-5 in cardioprotection during ischemia.

Methods: Myocardial ischemia/reperfusion (I/R; 30 min/24 h) and hypoxia/reoxygenation (H/R; 24 h/4 h) were used in this study. Confocal microscopy and transmission electron microscope (TEM) were used to observe mitochondrial morphology.

Results: Claudin-5 was detected in murine heart tissue and neonatal rat cardiomyocytes (NRCM). Its protein level was severely decreased after myocardial I/R or H/R. Confocal microscopy showedclaudin-5 presented in the mitochondria of NRCM. H/R-induced claudin-5 downregulation was accompanied by mitochondrial fragmentation. The mitofusin 2 (Mfn2) expressionwas dramatically decreased while the dynamin-related protein (Drp) 1 expression was significantly increased after H/R. The TEM indicatedH/R-induced mitochondrial swelling and fission. Adenoviral claudin-5 overexpression reversed these structural disintegration of mitochondria. The mitochondria-centered intrinsic pathway of apoptosis triggered by H/R and indicated by the cytochrome c and cleaved caspase 3 in the cytoplasm of NRCMs was also reduced by overexpressing claudin-5. Claudin-5 overexpression in mouse heart also significantly decreased cleaved caspase 3 and the infarct size in ischemic heart with improved systolic function.

Conclusion: We demonstrated for the first time the presence of claudin-5 in the mitochondria in cardiomyocytes and provided the firm evidence for the cardioprotective role of claudin-5 in the preservation of mitochondrial dynamics and cell fate against hypoxia- or ischemia-induced stress.

Fig. 1. The expression of claudin-5 in human cardiomyocyte.

(A) Subcellular localization of claudin-5 in AC16 human cardiomyocyte. Anti-claudin-5 antibody (Alexa Fluor 488, green) was used to detect endogenous claudin-5; Anti-TOMM20 antibody, a mitochondrial marker, was used to stain mitochondria (red). Scale bar, 5 μm. (B) Subcellular localization of claudin-5 in human heart tissue. Scale bar, 20 μm.

Fig. 2. The protein level of claudin-5 in the heart tissue.

(A) Western blot of claudin-5 in mouse heart tissue. mES (mouse embryonic stem cells) was used as positive control. (B) Western blot and quantitative analysis of the protein level of claudin-5 in the mouse heart after myocardial ischemia/reperfusion (I/R). * P < 0.05 vs. Sham; n = 5.

Fig. 3. The mRNA and protein level of claudin-5 in cardiomyocytes.

(A) mRNA level of claudin-5 in neonatal rat cardiomyocytes (NRCM). M indicates DNA marker. (B) Western blot of claudin-5 in NRCM. A549 and H293 cell lines are used as positive control. (C) Western blot and quantitative analysis of the protein level of claudin-5 in NRCM after hypoxia/reoxygeneration (H/R). * P < 0.05 vs. Normoxia (N); n = 5.

Fig. 4. Downregulation of claudin-5 and mitochondrial fragmentation in cardiomyocyte induced by hypoxia/reoxygenation (H/R).

(A) Subcellular localization of claudin-5 in neonatal rat cardiomyocytes (NRCM). A549 cell lines without anti-claudin-5 antibody incubation were used as negative control; A549 cell lines incubated with anti-claudin-5 antibody was used as positive control. Anti-claudin-5 antibody (green) was used to detect endogenous claudin-5 in cells; Mito-tracker was used to stain mitochondria (red) in cells. The images are merged to co-localize claudin-5 and mitochondria (orange). Scale bar, 50 μm. (B) Western blot and quantitative analysis of claudin-5 in mitochondria in NRCM. * P<0.05 vs. Normoxia (N); n=5. M indicates protein marker. (C) Downregulation of claudin-5 and mitochondrial fragmentation in NRCM after H/R. Anti-claudin-5 antibody (green) was used to detect endogenous claudin-5 in cells; Mito-tracker was used to stain mitochondria (red) in cells. The images are merged to co-localize claudin-5 and mitochondria (orange). Scale bar, 50 μm.

Fig. 5. The role of claudin-5 on the expression of Mfn2 and Drp-1 in cardiomyocytes.

(A). Effective knockdown of claudin-5 expression by siRNA. (B) The effect of claudin-5 knockdown on the protein level of Mfn2. (C) The effect of claudin-5 knockdown on the protein level of Drp-1. (D) Effective overexpression of claudin-5 by adenovirus (Adv). (E) The effect of claudin-5 overexpression on the protein level of Mfn2. (F) The effect of claudin-5 overexpression on the protein level of Drp-1. *P < 0.05 vs. Control (C), n=5. In control group, equal amount of PBS was added into the cell culture medium; Cont siRNA indicates nontargeted control siRNA. Cont Adv indicates LacZ adenovirus. Claudin-5 Adv indicates claudin-5 adenovirus.

Fig. 6. The role of claudin-5 on mitochondrial dynamics in cardiomyocytes.

(A) The western blot and quantitative analysis of the protein level of Mfn2 and Drp-1 in neonatal rat cardiomyocytes (NRCM). *P < 0.05 vs. normoxia (N); ^†P < 0.05 vs. hypoxia/reoxygenation (H/R); ^‡ P < 0.05 vs. H/R+Claudin-5 siRNA; n=5 in each group. (B) Representative pictures of mitochondrial morphology observed under transmission electron microscope. Red arrows indicate mitochondrial fission. Amplification, 11000 folds; Scale bar, 200 nm. (C) Quantitative analysis of the mean area of mitochondria. *P < 0.05 vs. N; ^†P < 0.05 vs. H/R, n=10; (D) Quantitative analysis of the mitochondrial fission. *P < 0.05 vs. N; ^†P < 0.05 vs. H/R.

Fig. 7. The role of claudin-5 on cardiomyocyte apoptosis induced by hypoxia/reoxygenation (H/R) in vitro.

(A) The western blot and quantitative analysis of the cytochrome c expression in cytoplasm and in mitochondria in neonatal rat cardiomyocytes (NRCM) (top, lower left and lower right). *P < 0.05 vs. normoxia (N); ^† P < 0.05 vs. H/R; n = 5 in each group. (B) The western blot and quantitative analysis of the caspase 3 cleavage in NRCM. *P<0.05 vs. N; ^† P < 0.05 vs. H/R; n = 5 in each group.

Fig. 8. The role of claudin-5 on infarct size and heart function after myocardial ischemia/reperfusion (I/R) in vivo.

(A) The western blot and quantitative analysis of claudin-5 after adenovirus (Adv) infection in left ventricle. *P < 0.05 vs. Control. (B) The western blot and quantitative analysis of claudin-5, cleaved caspase 3 and caspase 3 in the heart tissue in wild-type and claudin-5 adenovirus (Claudin-5 Adv) mice with sham operation and I/R. *P < 0.05 vs. Sham; ^† P < 0.05 vs. I/R in wild-type group; n = 5 in each group. (C) Representative 2,3,5-Triphenyl-2H-tetrazolium chloride (TTC) staining of the heart sections in wild-type and claudin-5 Adv mice with sham operation and I/R. *P < 0.05 vs. Sham; n = 5 in each group. (D) Representative echocardiographic imaging and quantitative analysis of fractional shortening (FS) in wild-type and claudin-5 Adv mice with sham operation and I/R. *P < 0.05 vs. Sham; ^† P < 0.05 vs. Wild-type I/R; n = 5 in each group.

Fig. 9. Diagram of claudin-5-mediated signaling.

Myocardial ischemia/ reperfusion (I/R) or hypoxia/ reoxygenation (H/R) decreases the expression of mitochondrial claudin-5, which in turn down-regulates the expression of Mfn2 (responsible for mitochondrial outer membrane fusion) and up-regulates the expression of Drp-1 (responsible for mitochondrial fission). The increased mitochondrial fission leads mitochondrial swelling and mitochondrial dependent apoptosis in cardiomyocyte, which finally impairs the systolic function of the heart.