Pleiotropic Effects of Myocardial MMP-9 Inhibition to Prevent Ventricular Arrhythmia

Figures

• 

  Figure 1. MMP-9 deficiency prevents ventricular arrhythmia. (A) MMP-9 enzymatic activity in the ventricular tissue after Ang II treatment in WT mice was assessed by zymography (n = 4, *P < 0.05). (B) Representative image of MMP-9 ventricular activity by zymography. Cropped blots are displayed, and full-length blots are included in Supplementary Fig. S10. (C) MMP-9 protein expression in the isolated ventricular cardiomyocytes as determined by Western blot analysis (n = 4, *P < 0.05). (D) Colocalization of MMP-9 and myofibrils. (E to H) The incidence (E) and duration (F) of ventricular arrhythmia episodes after the burst ventricular stimulation and ventricular arrhythmia induction threshold (G) were partially reduced in the MMP-9+/− mice and completely prevented in the MMP-9−/− mice (n = 10 for each group, #P < 0.05 vs. WT + Ang II). (H) Representative ECG and intracardiac tracing after the burst ventricular stimulation. Polymorphic ventricular tachycardia was induced in the WT mice after Ang II treatment. However, ventricular arrhythmia was not inducible in the MMP-9−/− mice after Ang II treatment. Data are shown as the mean ± SEM.

• 

  Figure 2. MMP-9 deficiency partially decreases cardiac fibrosis. (A) Quantitative analysis of cardiac fibrosis (n = 4–5, #P < 0.05 vs. WT + Ang II, ≠P < 0.05 vs. MMP-9−/− + Ang II). (B and C) The collagen III and collagen I mRNA levels did not differ between WT and MMP-9−/− mice after Ang II treatment (n = 4–6). (D) Left ventricular thickness was increased in WT mice but not in MMP-9−/− mice (n = 6–12, #P < 0.05 vs. WT + Ang II). (E) Representative image of Masson’s trichrome stain used to delineate cardiac fibrosis. (F) Representative image of cardiac hypertrophy illustrated by H&E staining. (G) Quantitative analysis of Cx43 protein expression and representative image of a Western blot. Cropped blots are displayed, and full-length blots are included in the Supplementary Fig. S11. (H) Cx43 lateralization was prevented in MMP-9−/− mice after Ang II treatment (n = 8–11, #P < 0.05 vs. WT + Ang II). (I) Representative image of Cx43 lateralization (denoted by white arrows; Green: Cx43, red: troponin T, blue: DAPI). Data are shown as the mean ± SEM.

• 

  Figure 3. MMP-9 deficiency prevents RyR2 phosphorylation in cardiomyocytes. (A) Expression of RyR2 receptor phosphorylated at serine 2808 (pS2808) and 2814 (pS2814) in mouse ventricular tissue. The increased level of pS2808 in the WT mice after Ang II treatment that correlated with a ventricular arrhythmia attack was prevented in the MMP-9−/− mice (n = 6, *P < 0.05 vs. WT + vehicle; #P < 0.05 vs. WT + Ang II). (B) Corresponding changes in PKA and CaMKII in mouse ventricular tissue (n = 6, *P < 0.05 vs. WT + vehicle; #P < 0.05 vs. WT + Ang II). (C) Representative Western blot showing phosphorylated RyR2 and PKA in mouse ventricular tissue. (D to G) MMP-9 expression was knocked down by siRNA in HL-1 cardiomyocytes (D), which prevented RyR2 hyperphosphorylation (E) and regulated PKA and CaMKII expression (F). (G) Representative Western blot in HL-1 cardiomyocytes. (D) to (G), (n = 4–6, *P < 0.05 vs. WT + vehicle, #P < 0.05 vs. WT + Ang II). Data are presented as the mean ± SEM.

• 

  Figure 4. MMP-9 deficiency reduces Ca2+ leakage and irregular calcium transients in isolated ventricular cardiomyocytes. (A and B) Amplitude of Ca2+ transients and decay over time (n = 17–38, from 4–5 experiments, P = 0.73). (C) Amplitude of Ca2+ storage induced by caffeine (20 mM) (n = 11–17, from 4–5 experiments, P = 0.14). (D to I) A summary of Ca2+ sparks, including the incidence (D), frequency (E), amplitude (F), Ca2+ spark duration (FDHM, G), and width (FWHM, H), as well as a representative image (I). A calcium spark is denoted by a white arrow (n = 21–87, from 4–6 experiments, #P < 0.05 vs. WT + Ang II). (J to L) A summary of irregular Ca2+ transients, including their incidence (J) and duration (K). Irregular, arrhythmia-like calcium transients (asterisk) early and delayed after depolarization were preceded by a calcium spark after ventricular pacing (arrow head), suggesting that calcium leakage might induce abnormal Ca2+ transients (L) in cardiomyocytes from WT mice after Ang II treatment. These irregular Ca2+ transients were absent in the MMP-9−/− mice after Ang II treatment (n = 18–39, from 4–5 experiments, #P < 0.05 vs. WT + Ang II, *P = 0.057 vs. WT + Ang II). Data are shown as…

• 

  Figure 5. MMP-9 inhibition prevented Ca2+ leakage and irregular calcium transients in hiPSC-CMs. (A and B) Amplitude and decay of Ca2+ transients over time (n = 28–40, from 4 experiments, P = 0.48). (C to H) Summary of Ca2+ spark, including the incidence (C), frequency (D), amplitude (E), duration (FDHM, F), and width (FWHM, G), as well as a representative image (H). Ca2+ sparks (denoted by white arrows) after treatment with Ang II are prevented by an MMP-9 inhibitor (n = 50–145, from 6 to 8 experiments, #P < 0.05 vs. WT + Ang II). (I to K) A summary of irregular Ca2+ transients, including their incidence (I) and duration (J), and a representative image (K). Irregular calcium transients were induced only in a high-Ca2+ (3.6 mM) Tyrode solution. Irregular, arrhythmia-like calcium transients, denoted by black arrows, were induced after treatment with Ang II and were prevented by MMP-9 or PKA inhibition (n = 33–73, from 3 to 5 experiments, #P < 0.05 vs. WT + Ang II). Data are shown as the mean ± SEM.

• 

  Figure 6. Mechanisms of MMP-9 inhibition to prevent ventricular arrhythmia. MMP-9 does not change collagen and Cx43 production but increases Cx43 lateralization. These minor changes might not significantly alter the conduction heterogeneity. In addition, MMP-9 can act as an intracellular signal regulator through the degradation of CD36 and increase calcium leakage, ventricular triggers, and arrhythmia through PKA and the hyperphosphorylation of RyR2.