Human epicardial adipose tissue acutely increases atrial fibrillation

Epicardial adipose tissue (EAT) is a fat layer that directly contacts the atrial myocardium. Increased EAT mass has emerged as a new risk factor for developing the most common type of irregular heart rhythm, atrial fibrillation (AF). The occurrence of AF is determined by the interaction between atrial triggers (abnormal electrical impulse initiation) and atrial substrate (electro-structural remodelling). EAT has been shown to induce atrial substrate; however, whether EAT directly promotes atrial triggers, remains unknown. We aimed to determine if human EAT would acutely interact with the atrial myocardium for developing AF. The development of spontaneous contractions (SCs) was established as a proxy for AF in isolated human right atrial (RA) muscles (trabeculae) (n = 15) after different stimulation frequencies (1, 2 and 3 Hz) and using two arrhythmogenic triggers: increased external calcium ([Ca2+]o) and beta (β)-adrenergic stimulation with isoproterenol (ISO). The development of SCs was examined one hour after superfusing the trabeculae with the perfusate of fresh human EAT (with EAT) (n = 39). At baseline (1 mM [Ca2+]o), the proportion of trabeculae with SCs significantly increased with EAT compared to no EAT (3 Hz/with EAT = 30% vs. 0% at 1 Hz/no EAT, p < 0.001; % of trabeculae with SCs). However, there was no significant difference in the proportion of trabeculae with SCs during high [Ca2+]o (3 mM) or ISO (10-5 M) between no EAT and with EAT. Specific β3-adrenergic stimulation of the EAT-trabecula, with agonist BRL37344 (10-6 M), significantly increased the proportion of trabeculae with SCs (1 Hz/BRL37344 with EAT = 30%, p = 0.04, vs. 0% at 1 or 3 Hz/baseline of no EAT; % of trabeculae with SCs). In addition to the acute effect, EAT could release different factors over a longer period of time. To this end, the effect of the secretome of untreated and treated 24-hour cultured human EAT with 10-5 M ISO or 10-6 M BRL37344 was examined on developing SCs in trabeculae. There was a significant increase in the proportion of trabeculae with ISO-induced SCs during superfusion with secretome of BRL37344-treated cultured EAT (n = 8) compared to baseline (62% at each of 1 and 3 Hz/baseline and 1 Hz/ISO with BRL37344-treated secretome vs. 0% at 1 Hz/baseline with no secretome, p = 0.02; % of trabeculae with SCs). There was no significant increase in the proportion of trabeculae with ISO-induced SCs during superfusion with the secretome of untreated, or ISO-treated, cultured EAT compared to baseline. Together, the results of the fresh and 24-hour cultured EAT suggested an acute functional interaction between EAT and trabecula for developing SCs. The acute EAT-trabecula interaction could be due to the factors released by EAT, such as resistin, an EAT-released adipokine. Using ELISA, the levels of resistin were undetectable in the perfusate of fresh EAT (n = 30), whereas the detectable levels (3847 ± 847 pg/mL) of resistin in the secretome of cultured EAT (n = 21) did not correlate with the development of SCs. Moreover, the development of SCs in human trabeculae was tested after superfusion with human recombinant resistin protein (n = 10). The results demonstrated that SCs at baseline or during ISO-triggering did not increase with high concentrations (12,000 or 20,000 pg/mL) of resistin compared to low concentration (7,000 pg/mL) or no resistin. Thus, resistin was unlikely to acutely enhance the development of SCs. Obesity is associated with the increase in EAT mass, but with unknown impact on the size of adipocytes in this fat depot. Haematoxylin and eosin stained EAT from obese patients (BMI > 30 kg/m2, n = 7) demonstrated smaller size (p = 0.0002) adipocytes compared to adipocytes in EAT from overweight (BMI = 25-29.9 kg/m2, n = 12) or lean (18.5-24.9 kg/m2, n = 7) patients. However, BMI, the development of SCs, and resistin released by the cultured EAT did not correlate with the size of adipocytes in EAT. The patients who provided the RA trabeculae were burdened with different clinical conditions. Analysing the relationship between the clinical conditions with the function (contraction and relaxation) of trabeculae showed that postoperative AF, diabetes mellitus and female gender were associated with the decreased contraction and relaxation, but not with the development of SCs, of trabeculae. Hence, these clinical conditions and female gender may play an important role in determining the overall functional performance of RA myocardium in patients undergoing cardiac surgery. In brief, there is an acute interaction between EAT and atrial myocardium for developing SCs, suggesting EAT can be an acute arrhythmogenic trigger for developing AF. Resistin is unlikely to have an acute arrhythmogenic effect in developing AF. The results in this thesis provide the foundation for the future therapeutic targets to obliterate the acute arrhythmogenicity of EAT on developing AF