TTNmv prevalence and association with hospitalization in a multiethnic atrial fibrillation (AF) cohort.

(A) Distribution of TTNmv in multiethnic AF cohort based on amino acid position in the TTN gene, stratified by REVEL in silico score for prediction of deleterious effect, defined by REVEL ≥ 0.70. (B) Mean cumulative incidence of AF and heart failure (HF)-related hospitalizations in subjects with AF stratified by presence of TTNmv. Hazard ratio (HR), 95% confidence interval (CI), and P-value were obtained from univariable Cox proportional hazard modeling. (C) Diagram denoting the location of TTNmv-T23756I. (D) Sequence alignment shows that the region of the T23756I is highly conserved across vertebrate species. (E) Allele frequencies of TTN-T3265I in various ethnic groups (gnomAD).

Clinical characteristics of ethnic minority subjects with AF stratified by presence of rare missense TTN variants.

*Data are missing for the following variables: eGFR (1), electrocardiogram within 3 months of AF diagnosis (11), LVEDD (19), left atrial size (6), left atrial diameter (21). Left ventricular dilatation is defined as left ventricular end diastolic diameter greater than 2 standard deviations above the normal sex-specific mean value. Variants with a REVEL score ≥ 0.7 were defined as predicted deleterious. Continuous data are represented as mean (standard deviation) and categorical data are represented as count (%).

Clinical characteristics of the early-onset AF patients with TTN-T32756I variation.

Human induced pluripotent stem cell-derived atrial cardiomyocytes (iPSC-aCMs) carrying TTN-T32756I variants have atypical contractility but no sarcomere disorganization.

(A) Workflow to generate the CRISPR/Cas9-mediated iPSC line carrying TTN-T32756I missense variation. (B-E) Contraction profile of wild type (black) and TTN-T32756I (Red) iPSC-aCMs showing increased beating frequency (C), Peak-to-Peak time (D), and Contraction duration (E) in the mutant. (F) Representative sarcomeric organization of wild-type (WT) and TTN-T32756I iPSC-aCM by Transmission electron microscopy (TEM). (G) There is no significant change in the sarcomere length (H). n.s.; P>0.05; *P<0.05; **P< 0.01.

Effect of T32756I on action potential (AP) and calcium-handling in iPSC-aCMs.

(A-C) Representative optical AP recordings of WT and TTN-T32756I showing reduction of AP duration (APD) at the 50% (APD50) (B) and 90% (APD90) repolarization (C). (D) Current-voltage (I-V) curves of the slow delayed rectifier potassium current (Iks) in WT and TTN_T32756I iPSC-aCMs. control (n=7) (E-F) Comparison of Iks current density at 50 mV (mean ± SEM). N.s.; P>0.05; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001. (G) Representative tracings of spontaneous calcium transients of WT and TTN-T32756I iPSC-aCMs. (H-I) Calcium kinetics show that the TTN-T32756I iPSC-aCMs have increased frequency (B) and decreased transient durations (I) compared with the WT iPSC-aCMs.

Transcriptomic profile and pathway enrichment analysis comparing TTN-T32756I iPSC-aCMs with the WT.

(A) Heatmaps of cardiac-related upregulated and downregulated differentially expressed genes (DEGs) (B) Top significantly enriched downregulated cardiac-related Gene-Ontology Biological process (GO-BP) pathways in the TTN-T32756I iPSC-aCMs. (C) Top significantly enriched downregulated cardiac-related Gene-Ontology Molecular Function (GO-MF) pathways in the TTN-T32756I iPSC-aCMs. (D) Top significantly enriched downregulated cardiac-related Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways in the TTN-T32756I iPSC-aCMs.(E) Significantly enriched upregulated and downregulated transcription factors (TFs) (F) Network diagram showing the upregulation of KCNQ1 by FHL2 predicted by the Ingenuity pathway enrichment analysis (IPA).

Inhibition of FHL2 rescues enhanced Iks in TTN-T32756I iPSC-aCMs.

(A) Co-immunoprecipitation revealed increased interaction between FHL2 and KCNQ1-KCNE1 (MinK) complex. (IP: KCNE1). Immunoblotting (IB) was performed with antibodies against FHL2 (32 kDa) and MinK (32 kDa) (n = 3) (B) qPCR data showing the inhibition of FHL2 gene by the siRNA in the WT and TTN-T32756I iPSC-aCMs (n=7). (C) I-V curves showing the rescue of the Iks TTN-T32756I iPSC-aCMs by the suppression of FHL2 (n=4-8). (D) Comparison of Iks current density at 50 mV (mean ± SEM). I Schematic showing the TTN-T32756I results in increased FHL2 binding with the KNCQ1-KCNE1 complex and enhanced Iks activity. *P<0.05, **P<0.01, ***P<0.001.

TTN-T32756I position and distribution.

(A) Location of the T23756I at Ig139 domain in the A-band of titin. (B) Allele frequencies of T3265I between sexes. (C) Age distribution of the variant carriers. (D) Genotype quality. (E) Allele balance for heterozygotes. Source: https://gnomad.broadinstitute.org/variant/2-179404525-G-A?dataset=gnomad_r2_1.

Generation of iPSC-aCMs with TTN-T32756I.

(A) TTN gene locus for generation of isogenic iPSCs with the T32756I variation. Guide sequence (gRNA) shown in the bottom gray box was cloned into vector to express gRNA guiding Cas9 exonuclease to the targeted protospacer adjacent motif sequence. (B) Next generation sequencing of the confirming T32756I mutation. (C) Representative immunostaining of pluripotency markers OCT4 and SOX2 in iPSCs. The 4′,6-diamidino-2-phenylindole (DAPI) indicates the nucleus. (D) Karyotype analysis of the T32756I iPSCs.

Contractility and sarcomere organization of TTN-T32756I iPSC-aCMs.

(A-C) Compared to WT, TTN-T32756I (Red) iPSC-aCMs show increased decreased Relaxation time (A), Contraction Amplitude (B), but no significant change to Time-to-Peak (C). (D-E) Immunostaining showing the sarcomeric organization of WT and TTN-T32756I iPSC-aCM by the pan-cardiomyocyte (CM) marker cardiac troponin T (cTnT; green) and α-actinin (orange). The DAPI staining indicates the nucleus. (F) Bar graph showing no change in the sarcomere length. n.s. P>0.05; **P<0.01.

TTN-T32756I iPSC-aCMs display anomalous action potentials, potassium currents, and calcium-handling.

(A) Compared to the WT, TTN- T32756I shows reduction of action potential duration at the 10% (APD10) repolarization. (B) Bar graph showing no change in the amplitude of the AP. (C) Total potassium current (IK) and voltage relationship (I-V curves) in WT and TTN-T32756I iPSC-aCMs. (D) Total IK current density at 50 mV. (E) Representative current traces at different voltages showing the isolation of the Iks current with the selective blocker HMR-1556 in both WT and TTN-T32756I iPSC-aCMs. (F-G) Bar graph showing that the TTN-T32756I iPSC-aCMs have decreased transient durations (F), but no change in the transient peak amplitudes (G) compared with the WT iPSC-aCMs. n.s. P>0.05; *P<0.05; ***P<0.001.

Upregulated pathways in TTN-T32756I iPSC-aCMs with the WT.

(A) Volcano plot showing spread of downregulated and upregulated differentially expressed genes (DEGs) (B) Top significantly enriched upregulated Gene-Ontology Biological process (GO-BP) pathways in the TTN-T32756I iPSC-aCMs. (C) Top significantly enriched upregulated Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways in the TTN-T32756I iPSC-aCMs. (D) Top significantly enriched TTRUST transcription factors (TFs).

List of TTN missense variants.

Age range represents patient’s approximate age at AF diagnosis in years. M=male, F=female, HL = Hispanic/Latinx, NHB = non-Hispanic Black. Variants with a blank value in the dbSNP or gnomAD columns represent variants not present in those respective databases.

Clinical characteristics of ethnic minority subjects with AF stratified by presence of predicted deleterious rare missense TTN variants.

*Data are missing for the following variables: eGFR (1), electrocardiogram within 3 months of AF diagnosis (11), LVEDD (19), left atrial size (6), left atrial diameter (21). Left ventricular dilatation is defined as left ventricular end diastolic diameter greater than 2 standard deviations above the normal sex-specific mean value. Variants with a REVEL score ≥ 0.7 were defined as predicted deleterious. Continuous data are represented as mean (standard deviation) and categorical data are represented as count (%).

TTN missense variants in subjects meeting criteria for nonischemic dilated cardiomyopathy.

Nonischemic dilated cardiomyopathy was defined by left ventricular ejection fraction <50% and left ventricular end diastolic diameter (LVEDD) greater than 2 standard deviations above the sex-specific mean, as well as coronary angiogram confirming the absence of obstructive coronary artery disease.

Parameter estimates for univariable and multivariable Cox proportional hazard models of atrial fibrillation and heart failure-related hospitalizations.

A partially adjusted multivariable model contained covariates of age and sex, and the fully adjusted model additionally accounted for race-ethnicity and ejection fraction <50% closest to AF diagnosis.

Cox proportional hazard models of hospitalizations related to TTN missense variant based on in silico prediction of impact.

REVEL score of ≥0.70 indicates potentially deleterious effect. A partially adjusted multivariable model contained covariates of age and sex, and the fully adjusted model additionally accounted for race-ethnicity and ejection fraction <50% closest to AF diagnosis.

Cox proportional hazard models of hospitalizations excluding cases with nonischemic dilated cardiomyopathy.

A total of 12 subjects were excluded. A partially adjusted multivariable model contained covariates of age and sex, and the fully adjusted model additionally accounted for race-ethnicity and ejection fraction <50% closest to AF diagnosis.

TTN-T32756I variant information.