Abstract 550: FOG2 Modulates a TBX5-Dependent Gene Regulatory Network for Cardiac Rhythm Homeostasis

Abstract

Atrial fibrillation (AF) is the most common cardiac arrhythmia. Recent work has implicated numerous transcription factors in AF risk, including ZFPM2 (FOG2), GATA4, and TBX5. FOG2 is overexpressed in human heart failure, a major AF risk factor. We found that genetic inducible physiologic overexpression of FOG2 in the adult mouse heart caused spontaneous atrial fibrillation. Single cell electrophysiology revealed action potential prolongation and inappropriate depolarizations in FOG2-OE atrial cardiomyocytes. RNA-seq of the FOG2-OE left atrium prior to AF onset demonstrated that FOG2 suppresses a network of calcium cycling genes, providing a possible mechanism for disrupted cardiomyocyte electrophysiology. FOG2 ChIP-seq demonstrated genomic localization only at locations co-occupied by GATA4. Unexpectedly, we observed little overlap between genes activated by GATA4 and repressed by FOG2. Instead, we observed highly significant overlap between genes repressed by FOG2 and activated by TBX5. To identify the FOG2-dependent atrial gene regulatory network, we performed differential deep sequencing of ncRNAs. Comparison of atrial FOG2-dependent non-coding transcripts with previously performed TBX5-dependent ncRNA profiling indicated that TBX5 activated and FOG2 repressed a shared atrial rhythm gene regulatory network. Integration of FOG2, GATA4, and TBX5 ChIP-seq revealed that FOG2 only affected ncRNA transcription, indicative of enhancer activity changes, at locations co-occupied by TBX5. TBX5-dependent activation of specific regulatory elements for calcium handling genes was abolished by FOG2 in vitro . The genomic TBX5/FOG2 interaction predicted a genetic interaction in-vivo , in which cardiac conduction and arrhythmia abnormalities caused by Tbx5 haploinsufficiency were rescued by FOG2 haploinsufficiency. Non-coding RNA profiling thereby predicted a novel functional TF interaction between FOG2 and TBX5. This work reveals a specific genomic model of atrial rhythm control in which FOG2 is recruited to GATA4 and TBX5 bound locations to modulate a TBX5-dependent atrial gene regulatory network for calcium handling and cardiac rhythm homeostasis.