Altered Expression of Zonula occludens-1 Affects Cardiac Na+ Channels and Increases Susceptibility to Ventricular Arrhythmias

Figure 1. Reduction of ZO-1 protein expression in human tissue samples with diverse etiologies of heart failure. (A) Transcript protein maps of the TJP1 gene. PDZ: Psd95/discs large/zonula occludens-1 domain, SH3: Src homology3 domain, GUK: Guanylate kinase-like domain. (B) Expression of TJP1 isoforms in human non-failing and failing ventricular tissues (n = 3 for failing and non-failing tissues * p < 0.05) (C–E) Immunoblotting and quantitative analysis of normalized ZO-1 protein expression (relative to total protein) illustrating reduced ZO-1 protein expression in ischemic and non-ischemic heart failure samples (n = 7 for patients with non-failing hearts, n = 7 for patients with ischemic failing hearts, n = 11 for non-ischemic failing hearts and n = 18 for all failing heart samples).

Figure 2. Characterization of ZO-1+/− haploinsufficient mouse model. (A) Schematic design illustrating the location of the LacZ-neo-cassette and loxP sites on a single allele of the Tjp1 gene. (B) qPCR data showing a significant reduction of Tjp1 transcript in ZO-1 deficient mice (n=5/genotype and mice were 19 weeks old), WT (two male and three female mice) and ZO-1+/− (one male and four female mice). (C,D) Immunoblotting and quantitative analysis of heart lysates from control and ZO-1+/− mice confirming a reduction of ZO-1 protein levels in ZO-1 deficient mice (n = 4 for each genotype, male mice).

Figure 3. Effect of ZO-1 haploinsufficiency on connexin-43 expression (Cx43) and localization. (A,B). Immunoblotting and quantitative analysis of Cx43 illustrating a significant upregulation of normalized protein levels in the hearts of ZO-1 haploinsufficient mice (n = 4 for each genotype), WT (two male and two female mice) and ZO-1+/− (four male mice). (C,D) Immunostaining of ZO-1 confirming a reduction of ZO-1 protein expression in the hearts of the ZO-1+/− mice with no changes in subcellular localization within the single isolated ventricular myocytes. (E,F) Immunostaining denoting an increased expression and an aggregation of Cx43 at the ID of ZO-1+/− myocytes. Scale bars equal 20 μm (C,E) and 10 μm (D,F) (n = 3/genotype, heart sections: WT (three male mice) and ZO-1+/− (three male mice) and isolated myocytes: WT (one male and two female mice) and ZO-1+/− (three male mice)).

Figure 4. Electrocardiographic analysis of ZO-1+/− mice. (A–C) Adult ZO-1+/− mice showing a reduction in peak heart rate, prolonged PR interval and wider QRS duration compared to the WT control mice post ISO-stimulation. (* p < 0.05 for A–C) (D,E) Radio-implanted telemeter recordings of WT and ZO-1+/− mice after intraperitoneal (IP) administration of carbachol (0.15 mg/kg) followed by 2 mg/kg of epinephrine and an IP caffeine dose of 120 mg/kg, ZO-1+/− mice alone, display multiple PVCs and short runs of bigeminy (denoted by # in F) (WT, n = 3, two male and one female mice and ZO-1+/−, n = 3, male mice)).

Figure 5. ZO-1+/− ventricular myocytes (VMs) display prolonged action potential duration and spontaneous activity. (A) Representative AP recordings from WT and ZO-1+/− VMs. (B,C) Action potential duration at 50, 75 and 90% repolarization are significantly prolonged in ZO-1+/− myocytes while action potential amplitude (APA) and the maximum rate of depolarization (dV/dt) are unaffected. (D) Paced APs in ZO-1+/− VMs display early after depolarizations (EADs) often triggering ectopic activity. (E) ZO-1+/− VMs at rest display spontaneous diastolic depolarization (DD) leading to triggered activity (for (B,C), WT, n = 4; n = 16 and ZO-1+/−, n = 4; n = 11).

Figure 6. Reduced ZO-1 expression reduces NaV1.5 mediated current density (INa). (A) Representative whole cell recordings of INa in WT and ZO-1+/− ventricular myocytes. (B) ZO-1+/− VMs display reduced current density, * p < 0.05 while (C) voltage dependent inactivation and (D) time dependent recovery are unaltered (for (B–D), WT, n = 3; n = 10 and ZO-1+/−, n = 3; n = 12).

Figure 7. NaV1.5 expression and localization in ZO-1 deficiency. (A) Localization of NaV1.5 (green) and N-cadherin (red) in WT (n = 5, three male and two female mice) and ZO-1+/− (n = 5, three male and two female mice) ventricular myocytes (scale bars = 20 μm; arrows indicate ID). DAPI (blue) denotes nuclei. Secondary antibodies were used as negative controls. (B,C) Immunoblotting and quantitative analysis of normalized NaV1.5 protein expression reveal no significant changes between WT and ZO-1+/− heart lysates (n = 4 for each genotype, WT, two male and two female mice and ZO-1+/−, four male mice).

Figure 8. ZO-1 deficiency displays an enhanced sensitivity to adrenergic stimulation. (A,B) Representative whole recordings of INa before and after application of 1 µM isoproterenol (ISO). (C) Schematic presentation of the INa, L analysis. (D–E) Integration of INa, L adjusted to peak in WT and ZO-1+/− myocytes displays a sensitivity to 1 µM ISO. (F) The percent increase in INa, L integration following exposure to 1 µM ISO is significantly larger in ZO-1+/− VMs. Horizontal Bars represent mean and vertical bars represent SD (for (D–F), WT, n = 3 (two female and one male mice); n = 10 and ZO-1+/−, n = 3 (two female and one male mice; n = 11).

Figure 9. ZO-1+/− myocytes display an increase in CaV1.2 mediated current (ICa). (A) Representative whole cell recordings of ICa in WT and ZO-1+/− ventricular myocytes. (B) ZO-1+/− myocytes display an upregulated current density at several membrane voltages, # p < 0.05 while (C) voltage dependent inactivation and (D) time dependent recovery remain unaffected (for (B), WT, n = 4 (four male mice); n = 12 and ZO-1+/−, n = 3 (two male and one female mice); n = 11. For (C,D), WT, n = 4 (four male mice); n = 11 and ZO-1+/−, n = 3 (two male and one female mice); n = 9). (E) Immunoblotting and quantitative analysis of normalized CaV1.2 protein expression reveal no changes between WT and ZO-1+/− heart lysates (n = 4 for each genotype), WT (two male and two female mice) and ZO-1+/− (four male mice).