CLOCK-BMAL1 regulates circadian oscillation of ventricular arrhythmias in failing hearts through β1 adrenergic receptor

Abstract

The incidence of ventricular arrhythmias (VAs) in chronic heart failure (CHF) exhibits a notable circadian rhythm, for which the underlying mechanism has not yet been well defined. Thus, we aimed to investigate the role of cardiac core circadian genes on circadian VAs in CHF. First, a guinea pig CHF model was created by transaortic constriction. Circadian oscillation of core clock genes was evaluated by RT-PCR and was found to be unaltered in CHF (P > 0.05). Using programmed electrical stimulation in Langendorff-perfused failing hearts, we discovered that the CHF group exhibited increased VAs with greater incidence at CT3 compared to CT15 upon isoproterenol (ISO) stimulation. Circadian VAs was blunted by a β1-AR-selective blocker rather than a β2-AR-selective blocker. Circadian oscillation of β1-AR was retained in CHF (P > 0.05) and a 4-h phase delay between β1-AR and CLOCK-BMAL1 was recorded. Therefore, when CLOCK-BMAL1 was overexpressed using adenovirus infection, an induced overexpression of β1-AR also ensued, which resulted in prolonged action potential duration (APD) and enhanced arrhythmic response to ISO stimulation in cardiomyocytes (P < 0.05). Finally, chromatin immunoprecipitation and luciferase assays confirmed that CLOCK-BMAL1 binds to the enhancer of β1-AR gene and upregulates β1-AR expression. Therefore, in this study, we discovered that CLOCK-BMAL1 regulates the expression of β1-AR on a transcriptional level and subsequently modulates circadian VAs in CHF.

Keywords: Arrhythmia; chronic heart failure; circadian clock; β1 adrenergic receptor.

Figure 1

Core circadian gene oscillations in an HF model. A-E: CLOCK, BMAL1, NR1D2, PER1, and DBP all exhibited significant circadian rhythms in control group (P < 0.05). Circadian oscillations of these core clock genes were not altered in CHF (N = 5 for each sampling time point; P > 0.05). Data are expressed as mean ± SEM.

Figure 2

Circadian variation in ventricular arrhythmias (VAs) is blunted by selective β1-adrenoreceptor antagonist. A: CHF group exhibited greater incidence of VAs. Circadian variations in response to ISO stimulation in failing hearts translated to greater incidence of VAs at CT3 than CT15, whereas circadian VAs were blunted by β1-AR-selective blocker (CGP). In addition, circadian VAs were not altered by β2-AR-selective blocker (ICI). B: Protocol for programmed electrical stimulation (PES); Premature extrastimulus (S2) was delivered after eight basic stimuli (S1). S1-S1 interval was 150 ms. C: Example of VF induced by PES.VF was induced by S1-S2 (8:1) programmed stimulation. N = 15 for each group. *P < 0.05, **P < 0.01.

Figure 3

Circadian oscillations for β-ARs mRNA level in CHF model. A: β1-AR exhibited significant circadian rhythms in control group (P < 0.05). Circadian oscillations of β1-AR were maintained in CHF model (P > 0.05). However, the mean expression of β1-AR was significantly attenuated in guinea pigs with CHF (N = 5 for each sampling time point, P < 0.01). B: β2-AR did not follow a circadian rhythm (N = 5 for each sampling time point, P > 0.05). Data are expressed as mean ± SEM.

Figure 4

Circadian oscillations for protein level of CLOCK-BMAL1 and β-ARs in CHF model. BMAL1, CLOCK, and β1-AR exhibited circadian oscillations in control group. Compared to control group, Circadian rhythms of clock genes (BMAL1 and CLOCK) and β1-AR were maintained in CHF group (P > 0.05). However, the mean expression of β1-AR was attenuated in CHF group (P < 0.01). Nevertheless, β2-AR did not follow a circadian rhythm (P > 0.05). (N = 5 for each sampling time point). Data are expressed as mean ± SEM.

Figure 5

CLOCK-BMAL1 overexpression induced arrhythmic response to ISO stimulation in cardiomyocytes. A: Percentage of asynchronous activity in the control (white bars), BMAL1 overexpression (black bars), CLOCK overexpression (blue bars), and CLOCK-BMAL1 overexpression (red bars) groups after superfusion with Tyrode’s solution containing ISO (10, 50, and 100 nmol/L) for 5 min. CLOCK-BMAL1 overexpression upregulated the percentage of arrhythmic myocytes. The number of cells = 80-120; hearts = 6; *P < 0.05. B: An example of action potential duration recorded from the control and CLOCK-BMAL1 overexpression groups showing the presence of early afterdepolarization (**), delayed afterdepolarization (*), and extrasystoles (#) during superfusion with ISO (100 nmol/L).

Figure 6

CLOCK-BMAL1 overexpression prolonged cardiomyocyte APD with ISO stimulation through β1-AR. A-C: Representative APDs (1-Hz) from myocytes with added ISO in the control, BMAL1 overexpression, CLOCK overexpression, CLOCK-BMAL1 overexpression, and sh-BMAL1 groups; APD at 50% (APD50) and 90% (APD90) repolarization. D-F: Representative APDs from myocytes with added ISO and highly selective β1 antagonists (CGP-20712A). G-I: Representative APDs from myocytes with added ISO and highly selective β2 antagonists (ICI-118,551). N = 10 cells from eight hearts per group. Data are expressed as mean ± SEM; *P < 0.05.

Figure 7

CLOCK-BMAL1 overexpression induced the expression of β1-AR rather than β2-AR. Altered protein expression of CLOCK, BMAL1, β1-AR, and β2-AR were determined with western blotting in the control, BMAL1 overexpression, CLOCK overexpression, CLOCK-BMAL1 overexpression (CB), and sh-BMAL1 groups. N = 5 per group. Data are expressed as mean ± SEM; *P < 0.05.

Figure 8

CLOCK-BMAL1 bound to the enhancer of β1-AR and upregulated β1-AR expression. A: Three putative E-box sequences in ADRB1 promoter located at -741 bp (A1), -806 bp, and -1005 bp (A2); PP: PER1 promoter E-box area. B: DNA expressions of A1 and A2 in anti-BMAL1 antibody-precipitated DNA samples; Rabbit IgG was used as a negative immunoprecipitation control. C: Three mutated E-box sequences in ADRB1 promoter. D: Luciferase activity of ADRB1 promoter and/or mutant ADRB1 promoters in overexpression of CLOCK-BMAL1. N = 4 for each group. Data are expressed as mean ± SEM; *P < 0.05, **P < 0.01.