Sparsely methylated mitochondrial cell free DNA released from cardiomyocytes contributes to systemic inflammatory response accompanied by atrial fibrillation

Abstract

Systemic inflammation is assumed to be the consequence and the cause of atrial fibrillation (AF); however, the underlying mechanism remains unclear. We aimed to evaluate the level of cell-free DNA (cfDNA) in patients with AF and AF mimicking models, and to illuminate its impact on inflammation. Peripheral blood was obtained from 54 patients with AF and 104 non-AF controls, and cfDNA was extracted. We extracted total cfDNA from conditioned medium after rapid pacing to HL-1 cells. Nuclear and mitochondrial DNA were separately extracted and fragmented to simulate nuclear-cfDNA (n-cfDNA) and mitochondrial-cfDNA (mt-cfDNA). The AF group showed higher cfDNA concentration than the non-AF group (12.6 [9.0-17.1] vs. 8.1 [5.3-10.8] [ng/mL], p < 0.001). The copy numbers of n-cfDNA and mt-cfDNA were higher in AF groups than in non-AF groups; the difference of mt-cfDNA was particularly apparent (p = 0.011 and p < 0.001, respectively). Administration of total cfDNA and mt-cfDNA to macrophages significantly promoted IL-1β and IL-6 expression through TLR9, whereas n-cfDNA did not. Induction of cytokine expression by methylated mt-cfDNA was lower than that by unmethylated mt-cfDNA. Collectively, AF was associated with an increased cfDNA level, especially mt-cfDNA. Sparsely methylated mt-cfDNA released from cardiomyocytes may be involved in sterile systemic inflammation accompanied by AF.

Figure 1

Elevation of total cfDNA in AF patients. (a) Comparison between non-AF and AF. (b) Comparison among the 4 subgroups. Mann–Whitney test was performed for a, Kruskal–Wallis test followed by post-hoc Steel–Dwass for b. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2

mt-cfDNA, rather than n-cfDNA, was increased in AF patients. Comparison between non-AF and AF for n-cfDNA (a) and mt-cfDNA (b). Comparison among the 4 subgroups for n-cfDNA (c) and m-cfDNA (d). Mann–Whitney test was performed for (a) and (b), Kruskal–Wallis test followed by post-hoc Steel–Dwass for (c) and (d). *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3

Total cfDNA, n-cfDNA, and mt-cfDNA in in vitro AF mimicking model. (a) Total cfDNA, (b) n-cfDNA copy number, (c) mt-cfDNA copy number in an in vitro tachycardia pacing model (n = 3 in each pacing duration). Absolute n-cfDNA and mt-cfDNA copy numbers were calculated using qPCR, using the created standard curves. Unpaired t test was performed. ANOVA test were followed by post-hoc Tukey for a in comparison among pacing groups (†). *p < 0.05, **p < 0.01, ***p < 0.001, † in comparison among pacing groups.

Figure 4

Total cfDNA, n-cfDNA, and mt-cfDNA in an in vivo AF mimicking model. (a) Total cfDNA, (b) n-cfDNA copy number, (c) mt-cfDNA copy number in an in vivo tachycardia pacing model (n = 7). Absolute n-cfDNA and mt-cfDNA copy number were calculated using qPCR using the created standard curves. * p < 0.05.

Figure 5

mt-cfDNA, but not n-cfDNA, promoted IL-1β and IL-6 expression in macrophages. (a) Total cfDNA (25 ng/mL, 250 ng/mL or 500 ng/mL) (n = 4–8), (b) n-cfDNA (500 ng/mL), and mt-cfDNA (500 ng/mL) (n = 5–8) were applied to J774.1, then pro-inflammatory cytokines expression was analyzed using qPCR. Nuclear and mitochondrial DNA were independently extracted and fragmented to simulate n-cfDNA and mt-cfDNA. ANOVA followed by post-hoc Tukey was performed. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 6

Sparsely methylated mt-cfDNA induced IL-1β and IL-6 expression in macrophages via TLR9. (a) Comparison of IL-1β and IL-6 expression levels between TLR9 inhibitor vs. ODN control under mitochondrial DNA exposure (500 ng/mL) (n = 4–6). (b) Comparison of IL-1β and IL-6 expression levels between fully methylated mitochondrial DNA (500 ng/mL) versus totally unmethylated mitochondrial DNA (500 ng/mL) (n = 6). Unpaired t test was performed for (a) and (b) *** p < 0.001.

Figure 7

Receiver operating curve (ROC) curve analysis for AF incidence and progression. ROC analysis for AF incidence (a) (non-AF vs. AF), and for AF progression (b) (non-AF and paroxysmal AF vs. persistent AF) using total cfDNA, n-cfDNA, and mt-cfDNA. AUC, area under the curve.