RELATIVE CONTRIBUTIONS OF FIBROSIS AND CONNEXIN CHANGES TOTHE ATRIAL FIBRILLATION SUBSTRATE DURING THE DEVELOPMENT AND REVERSAL OF CONGESTIVEHEART FAILURE

Background: Connexin alterations occur in various atrial fibrillation (AF) paradigms, but their functional significance remains unclear. No data are available regarding the effects of CHF on atrial connexin expression and phosphorylation. We therefore analyzed connexin changes and their contribution to the AF substrate during the development and reversal ofCHF. Methods and Results: Dogs were allocated to three groups: CHF induced by 2-week ventricular tachypacing (CHF, n=15); CHF dogs allowed to recover for 4 weeks after 2-week tachypacing (REC, n=15) and non-paced shams (CTL, n=11). Left ventricular end-diastolic pressure increased with CHF (14.5±1.0*** vs.3.7±0.7, ***P < 0.001 vs. CTL) and normalized upon CHF recovery (5.1±1.0^†††, ^††† P < 0.001 vs. CHF). Real-time PCR and Western-blot analyses revealed connexin43 (Cx43) and connexin40 (Cx40) mRNA and protein expression to be unchanged by CHF and REC. However, CHF caused Cx43 dephosphorylation(by ~73%***) and increased Cx40/Cx43 protein ratio (by ~35%***), with both alterations completely reversing in REC. Immunofluorescent confocal microscopy confirmed connexin protein trends, with a reduction in phosphorylated Cx43 (by ~68%*** in CHF) that returned to control in REC. CHF caused conduction abnormalities (phasedelay-range and heterogeneity index, both P < 0.01) and burst pacing-induced AF prolongation (CTL 22±7s, CHF 1100±171s***, REC 884±220s***) which persisted in the recovery period, along with residual fibrosis (CTL 3.6±0.7%, CHF 14.7±1.5%***, REC13.3±2.3%***). Fibrosis physically interrupted muscle bundle continuity and anionically-based action potential model of canine atrium showed that fibrosiswas able to account for the observed conduction abnormalities. Conclusions: CHF causes connexin-dephosphorylation and Cx40/Cx43ratio increases. With CHF reversal, atrial connexin alterations recover completely, but tissue fibrosis, conduction abnormalities and a substrate forAF remain with fibrosis accounting for conduction abnormalities. Thus, althougha trial connexin changes occur with CHF, they are not essential for conduction disturbances and AF promotion, which appear rather to be related primarily tofibrotic interruption of muscle-bundle continuity.