PAGln, an Atrial Fibrillation-Linked Gut Microbial Metabolite, Acts as a Promoter of Atrial Myocyte Injury

Figures

• 

  Figure 1. PAGln was significantly associated with AF. (A) Increased relative abundance of gut microbial porA enzyme in AF fecal samples. p = 0.493, Student’s t-test. (B) Higher plasma PAGln levels in AF patients than controls without AF. (C) Levels of plasma PAGln in PAF, psAF and control groups. PAF, paroxysmal atrial fibrillation; psAF, persistent atrial fibrillation. *** p < 0.001; ns, no significance.

• 

  Figure 2. Predictive potential of PAGln for AF. (A) ROC analysis on plasma PAGln in predicting presence of AF (AUC = 0.774, 95% CI: 0.693–0.841, p < 0.001). (B) ROC curve of the Taiwan AF score (blue) (AUC = 0.602, 95% CI: 0.513–0.685; p = 0.053) and Taiwan AF-PAGln score (orange) (AUC = 0.795, 95% CI: 0.716–0.860, p < 0.001).

• 

  Figure 3. Correlations between left atrial dimension, LVEF and plasma PAGln. (A) Higher left atrial diameters (LAAPD, LALRD and LAUDD) in the high-PAGln group (>66.86 ng/L) than in the low-PAGln group (≤66.86 ng/L). LAAPD, left atrial anteroposterior diameter, p = 0.002; LAUDD, left atrial up and down diameter, p = 0.002; LALRD, left atrial left and right diameter, p = 0.056. (B) Plasma levels of PAGln were elevated in AF patients with LAE compared to the levels in AF patients without LAE. p = 0.337. (C–E) Plasma PAGln was positively associated with LAAPD (R = 0.317, p < 0.001), LALRD (R = 0.231, p = 0.011) and LAUDD (R = 0.305, p < 0.001) based on Pearson correlation analysis. (F) Plasma PAGln was negatively related to LVEF (R = −0.175, p = 0.055). Pearson correlation analysis. **, p < 0.01; ns, no significance.

• 

  Figure 4. The levels of apoptosis, ROS production, CaMKII and RyR2 activation and cell viability were detected in mouse atrial myocytes treated with or without PAGln. (A) Representative flow cytometry results of apoptosis and ROS levels in control and PAGln groups. (B) PAGln treatment enhanced ROS production and apoptosis. n = 3. (C) Increased NOX activity in HL-1 cells treated with PAGln. n = 4. (D,E) PAGln treatment decreased SOD activity and cell viability. n = 4. (F,G) Representative Western blot images and the expression levels of p-CaMKII, CaMKII, p-RyR2 and RyR2 in HL-1 cells treated with or without PAGln. GAPDH was used as an endogenous control. n = 4–8; *, p < 0.05; **, p < 0.01; ***, p < 0.001; ns, no significance.