The matricellular protein CCN5 prevents adverse atrial structural and electrical remodelling

Figures

• 

  FIGURE 1. CCN5 prevents AngII‐induced atrial fibrosis in CCN5 Tg mice. A, Experimental scheme is shown for (B) and (C). Mice were sham‐operated or AngII‐infused osmotic pump transplantation for 2 weeks. B, Atria were sectioned and stained with Masson's trichrome. Representative images are shown (Left). The percentage of interstitial fibrotic areas is plotted (Right). Original magnification: 20 X, Scale bar: 200 μm. C, mRNA was isolated from left atrial and subjected to qRT‐PCR. Relative expression of pro‐fibrotic and pro‐inflammatory genes is shown. (n = 5‐6 per group). Error bar = SD, *P < 0.05, **P < 0.01

• 

  FIGURE 2. CCN5 prevents AngII‐induced CaMKII activation in atrial myocytes and atrial fibroblast to myofibroblast trans‐differentiation. A, Experimental scheme is shown for (b). HL‐1 cells, atrial myocyte cell line, were cultured in the presence of AngII and control conditioned media (CM‐Con) or CCN5 containing conditioned media (CM‐CCN5) for 48 hours. B, Total cell lysates from HL‐1 were immunoblotted with antibodies against phospho‐CaMKII (Thr286), total CaMKII, phospho‐RyR2 (Ser2808), phospho‐RyR2 (Ser2814), total RyR2, calsequestrin‐2, NCX1 and GAPDH. Quantified protein levels were plotted (Right). C, Experimental scheme is shown for D, E and F. Atrial fibroblasts were isolated from adult SD rats and cultured in the presence of AngII and CM‐Con or CM‐CCN5 for 48 hours. D, Immunofluorescence images of atrial fibroblasts were stained with anti‐α‐SMA antibody and Hoechst dye. Original magnification: 60 Χ. E, Total cell lysates from atrial fibroblasts were immunoblotted with antibodies against α‐SMA, collagen I, TGF‐β1 and α‐tubulin. Quantified protein levels are plotted (Bottom). F, mRNA was isolated from atrial fibroblasts and subjected to qRT‐PCR. Relative expression of pro‐fibrotic…

• 

  FIGURE 3. CCN5 reverses AngII‐induced atrial fibrosis in AAV9‐mediated CCN5 overexpressed mice. A, Experimental scheme is shown for B, C and D. Mice were underwent AngII‐infused osmotic pump implantation for 2 weeks. AAV9‐VLP or AAV9‐CCN5 (1 × 1011 viral genomes per mouse) was injected via tail vein. 4 weeks later, hearts were analysed. (B) Total cell lysates from heart atrium were immunoblotted with antibodies against CCN5 and GAPDH (Left). Quantified protein level (Middle) and mRNA level (Right) of CCN5 are plotted. Heart was sectioned and stained with Masson's trichrome. Representative images are shown from left atrium. The percentage of interstitial fibrotic areas is plotted. Original magnification: 20 X, Scale bar: 200 μm (D) mRNA was isolated from left atrial and subjected to qRT‐PCR. Relative expression of pro‐fibrotic and pro‐inflammatory genes is shown. (n = 5 per group). Error bar =SD *P < 0.05, **P < 0.01

• 

  FIGURE 4. CCN5 decreases predilection for atrial‐induced arrhythmia in fibrotic hearts. A, Experimental scheme is shown. B, Representative maps for each group showing delays in atrial activation time at baseline (10 Hz) in AngII hearts (7.6 ± 2.4 ms) compared to sham (4.5 ± 1.6 ms) indicative of conduction slowing, which was partially reversed in AAV9‐CCN5–treated hearts (5.8 ± 1.5 ms). Scale bar: 0.5 mm, activation time: 0‐10 ms, LV = left ventricle, LA = left atrium, PE = pacing electrode. C, Representative optical mapping traces showing that a higher pacing frequency is necessary to induce atrial arrhythmia in AAV9‐CCN5–treated hearts compared to AngII, indicating lower arrhythmia vulnerability and reverse remodelling. Scale bar: 50 ms AF = atrial fibrillation. D, Bar graph shows a higher susceptibility to atrial arrhythmia at cut‐off frequency of 22 Hz (PCL 44 ms). E, Bar graph shows a higher susceptibility to atrial arrhythmia at cut‐off frequency of 22 Hz (PCL 44 ms). E, Bar graph shows a significantly lower threshold to induced atrial arrhythmia in the AngII hearts (21.9 ± 0.8 Hz in AngII, 28.3 ± 1.7 Hz in sham) *P < 0.05 (one‐way ANOVA followed by post hoc Tukey test). F and G,…