Atrial fibrillation (AF) is the most common sustained arrhythmia presenting at cardiology departments. A limited understanding of the molecular mechanisms responsible for the development of AF has hindered treatment strategies. The purpose of this study was to assess whether reduced activation of phosphoinositide 3-kinase (PI3K, p110␣) makes the compromised heart susceptible to AF. Risk factors for AF, including aging, obesity, and diabetes, have been associated with insulin resistance that leads to depressed/defective PI3K signaling. However, to date, there has been no link between PI3K(p110␣) and AF. To address this question, we crossed a cardiac-specific transgenic mouse model of dilated cardiomyopathy (DCM) with a cardiac-specific transgenic mouse expressing a dominant negative mutant of PI3K (dnPI3K; reduces PI3K activity). Adult (ϳ4.5 months) double-transgenic (dnPI3K-DCM), single-transgenic (DCM-Tg, dnPI3K-Tg), and nontransgenic mice were subjected to morphological, functional/ECG, microarray, and biochemical analyses. dnPI3K-DCM mice developed AF and had depressed cardiac function as well as greater atrial enlargement and fibrosis than DCM-Tg mice. AF was not detected in other groups. Aged DCM-Tg mice (ϳ15 months) with a similar phenotype to dnPI3K-DCM mice (4.5 months) did not develop AF , suggesting loss of PI3K activity directly contributed to the AF phenotype. Furthermore, increasing PI3K activity reduced atrial fibrosis and improved cardiac conduction in DCM-Tg mice. Finally, in atrial appendages from patients with AF, PI3K activation was lower compared with tissue from patients in sinus rhythm. These results suggest a link between PI3K(p110␣) and AF. (Am J Pathol