Fibroblast-Restricted Inflammasome Activation Promotes Atrial Fibrillation and Heart Failure With Diastolic Dysfunction

Abstract

Atrial fibrillation (AF) often coexists with heart failure, both involving inflammatory signaling and cardiac fibroblasts. To understand the role of fibroblast NLR family pyrin domain containing 3 (NLRP3) inflammasome in cardiac function, we found that NLRP3 was up-regulated in atrial fibroblasts from AF patients. Fibroblast-specific activation of NLRP3 in mice induced AF-promoting atrial myopathy and heart failure with diastolic dysfunction, accompanied by increased fibrosis, and reduced conduction velocity. Knockdown of NLRP3 prevented the AF-promoting atrial substrate and cardiomyopathy in the context of NLRP3 activation in fibroblasts. We identify the fibroblast NLRP3 inflammasome as a key pathway governing the promotion of proarrhythmic fibrosis in AF and cardiomyopathy.

Keywords: atrial cardiomyopathy; atrial fibrillation; cardiac fibroblast; fibrosis; inflammasome.

Graphical abstract

Figure 1

Enhanced NLRP3 Inflammasome Signaling in Fibroblasts of AF Patients and Canine Model of AF (A) Representative digital PCR photos from wells containing 8,500 partitions and quantification of NLR family pyrin domain containing 3 (NLRP3) (B), ASC (C), and IL1B (D) in human atrial fibroblasts isolated from right atrial appendage of normal sinus rhythm (NSR) and persistent atrial fibrillation (perAF) patients. (E, F) Representative Western blots and quantification of NLRP3 (E), ASC, and active interleukin (IL)-1β (F) in atrial fibroblast samples of control (Ctl) and AF dogs. Data are presented as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01. P values were determined with unpaired Student’s t-test (2-tailed) in B, C, and D, and Mann-Whitney U test in E and F.

Figure 2

FB-Specific Activation of NLRP3 in a Transgenic Murine Model (A) Schematic representation of the transgenic approach to establish the inducible Tcf21^iCre;Nlrp3^A350V/A350V (FB-KI) mouse model. Tamoxifen injected Tcf21^iCre mice were used as control (Ctl). (B-D) Representative Western blots and quantifications of precursor caspase-1 (Pro-Casp1) and cleaved caspase-1 (p20) in separated fibroblast (FB)- and cardiomyocyte (CM)-fractions of Ctl and FB-KI mice at 1-month post-tamoxifen injection. (E, F) Representative Western blots and quantifications of precursor and active IL-1β and -18 in atrial tissues of Ctl and FB-KI mice. Data are presented as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. P values were determined with unpaired Student’s t-test (2-tailed) in C, E, and F. Abbreviations as in Figure 1.

Figure 3

FB-Specific Activation of NLRP3 Promotes Atrial Cardiomyopathy and Enhances Susceptibility to Atrial Arrhythmia (A) Representative long-axis echocardiography images. (B-D) Quantifications of the superoinferior (SI) (B) and anteroposterior (AP) (C) dimensions and the area (D) of the left atrium (LA). (E) Representative M-mode echocardiography of the LA. (F-H) Quantifications of left atrial diameters (LADs) (F), (LADd) (G), and fractional shortening (FS%) (H). (I) Representative simultaneous recording of surface electrocardiogram (ECG) (lead 1) and intracardiac electrograms showed sinus rhythm in Ctl and AF in FB-KI mice following rapid atrial pacing. (J) Incidence of the pacing-induced AF. Data were obtained at 1-month post-tamoxifen injections. Data are presented as mean ± SEM in B, C, D, F, G, and H, or as percentage in J. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. P values were determined with unpaired Student’s t-test (2-tailed) in B, C, D, F, G, and H, and Mann Whitney U test in J. Abbreviations as in Figures 1 and 2.

Figure 4

FB-Specific Activation of NLRP3 Promotes Heart Failure With Diastolic Dysfunction (A) Representative M-mode echocardiography of left ventricles in Ctl and FB-KI mice. (B-D) Age-dependent reduction in ejection fraction (EF%) (B) and increases in end-systolic diameter (ESD) (C), and end-diastolic diameter (EDD) (D) of left ventricles of FB-KI mice (vs Ctl mice). (E-G) Representative tissue Doppler imaging (E) and quantifications of E/A ratio (F) and E/e’ ratio (G) in Ctl and FB-KI mice. (H) Atrial weight normalized to tibial length. (I) Ventricular weight normalized to tibial length. (J) Lung weight normalized to tibial length. (K) Levels of N-terminal-pro-brain natriuretic peptide (NT-proBNP) in Ctl and FB-KI mice. Data were obtained at 3 months post-tamoxifen injections. Data are presented as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. P values were determined with unpaired Student’s t-test (2-tailed) in B, C, D, F, G, H, I, J, and K. Abbreviations as in Figures 1 and 2.

Figure 5

FB-Specific Activation of NLRP3 Promotes Cardiac Fibrosis (A) Representative images of αSMA staining and (B) quantification of αSMA intensity in isolated cardiac fibroblasts of Ctl and FB-KI mice, after 48 hours’ culture. (C) Representative images of gel contraction assay in isolated FBs with and without TGFβ challenge. White dash lines outlined the edges of gels after 24 hours of culture. (D) Quantification of the contraction of gels (%) cocultured with FBs of Ctl and FB-KI mice. (E) Representative images of scratch assay with the FBs of Ctl and FB-KI (left) and quantification of migration area (right). (F) Representative images of Edu staining (pink) in culture FBs of Ctl and FB-KI (left) and quantification of the Edu+ cells normalized to total cells. (G) Representative images of Picrosirius Red staining in whole heart. (H) Representative images of Picrosirius Red staining in atria and quantification of atrial fibrosis. (I) Representative images of Picrosirius Red staining in ventricles and quantification of ventricular fibrosis. Red color indicated the fibrosis region. (J) Representative Western blots and (K) quantifications of collagen 1a (Col1a) and fibrogenic proteins in atria of Ctl and FB-KI mice. Data are presented as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. p-values were determined with unpaired Student’s t-test (two-tailed) in B, E, F, H, I, and K, and one-way analysis of variance with Sidak’s multiple comparison in D. Abbreviations as in Figures 1 and 2.

Figure 6

Altered Intercellular Communications in FB-KI Mice (A) Uniform Manifold Approximation and Projection representation of 12 cell clusters identified by the known signature genes. (B) Distribution of each cell type in Ctl and FB-KI samples. (C) Weighted interaction networks in Ctl and FB-KI samples. (D) Differential interaction strength map between FB-KI and Ctl. Blue color indicates reduced communication. Red color indicates enhanced communication. (E) Altered cell–cell communications events among aFB-, MyoFB-, and Mac-clusters, and the altered pathways in FB-KI compared with Ctl. (F) Representative plots for the COLLAGEN pathway-mediated communications among the aFB-, MyoFB-, and Mac-clusters in the Ctl and FB-KI mice. aCM = atrial cardiomyocyte; aFB = atrial fibroblast; Mac = macrophage; MyoFB = myofibroblast; vCM = ventricular cardiomyocyte; vFB = ventricular fibroblast; other abbreviations as in Figures 1 and 2.

Figure 7

FB-Specific Activation of NLRP3 Impairs Electrical Coupling (A) Representative Western blots and quantifications of Cx43 and phosphorylated Cx43-S368 in atria. (B) Representative images of immunohistochemistry of Cx43 in atria of Ctl and FB-KI mice. (C) Representative images of coimmunostaining of Cx43/WGA or Cx43/ZO-1 in atrial CMs of Ctl and FB-KI mice. (D) Quantification of the ratio of lateral Cx43 to intercalated disc (ID) Cx43 in atrial CMs of Ctl and FB-KI mice. (E) Representative optical di-4-ANEPPS signal and activation maps. (F) Conduction velocity (CV) in right atrium of Ctl and FB-KI mice. Data were obtained at 3 months post-tamoxifen injections. The data are presented as mean ± SEM in A and F, or as median with whiskers representing the minimum and maximum values in panel D. ∗P < 0.05, ∗∗P < 0.01. P values were determined with unpaired Student’s t-test (2-tailed) in A and F, and Mann-Whitney U test in D. Abbreviations as in Figures 1 and 2.

Figure 8

Knocking Down Nlrp3 In Fibroblasts Attenuates Cardiac Dysfunction and Proarrhythmic Substrate (A) Representative simultaneous recording of surface ECG (lead 1) and intracardiac electrograms following rapid atrial pacing in each group. (B) Incidence of reproducible pacing-induced AF in Ctl and FB-KI mice that received AAVdj/8-scramble or AAVdj/8-shNlrp3 virus, respectively. (C) Representative echocardiography of LA. (D) Quantifications of LA fractional shortening (FS%). (E) Atrial weight (AW) normalized to tibial length (TL). (F) Representative M-mode echocardiography of left ventricles in Ctl and FB-KI mice that received AAVdj/8-scramble or AAVdj/8-shNlrp3, respectively. (G) Quantification of ejection fraction (EF%). (H) Lung weight normalized to tibial length. (I) Representative tissue Doppler imaging in Ctl and FB-KI mice that received AAVdj/8-scramble or AAVdj/8-shNlrp3 virus, respectively. (J) Quantification of E/A ratio. (K) Quantification of E/e’ ratio. Data were obtained at 3 months post-tamoxifen injections. Data are presented as percentage in B, or as mean ± SEM in D, E, G, H, J, and K. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. P values were determined with Fisher exact test in B, and 1-way analysis of variance and Tukey’s multiple comparisons test in D, E, G, H, J, and K. Abbreviations as in Figures 1 and 2.