Lee, Hon-Chi

Background: In heart failure with preserved ejection fraction (HFpEF), left ventricular (LV) systolic elastance (Ees) increases in tandem with increases in arterial elastance (Ea), which maintains ejection fraction but limits systolic reserve and precipitates volume sensitivity. The left atrium (LA) lies in series with the LV and contributes to LV stroke volume. We hypothesized that HFpEF is associated with increased LA systolic elastance (E LA ) and deranged atrioventricular (AV) coupling which may additionally compromise cardiac performance. Methods: HFpEF was modeled by experimental hypertension (bilateral renal wrap + DOCA) in elderly dogs (n=9). LA and LV structure, function (echo, MRI) and invasively assessed pressure-volume relationships (PVR, open-chest; pericardiotomy; admittance catheter) were compared to sham-operated young dogs (control; n=13). Results: As compared to controls, HFpEF dogs displayed concentric remodeling, with increased diastolic stiffness, Ea and Ees (p<0.05 for all) and reduced Ea/Ees (p=0.03). LA maximum volume (28±3 vs 23±3mL, p=0.007), LA stroke volume (4±1 vs2±1mL, p=0.006), LA active emptying fraction (18±6 vs 12±2%, p=0.03) and LA ejection rate (51±14 vs 30±16mL/s) were higher in HFpEF versus controls, suggesting increased LA stroke work and inotropy. At matched heart rate and LA pressure, E LA was steeper in HFpEF than control (2.7±0.9 vs 1.5±0.6mmHg/mL, p=0.01) and positively correlated with Ees (r=0.6, p=0.005) and Ea (r=0.66, p=0.002). However AV mismatch occurred as increases in E LA were not proportionate to increases in Ees (Ees/E LA 1.3±0.4 HFpEF vs 0.8±0.4 control, p<0.05). On linear regression, worse AV coupling (i.e. higher Ees/E LA ) was associated with lower LV stroke volume ( Figure ). Conclusion: In experimental HFpEF, despite enhanced LA contractility, AV coupling was deranged. Worse AV coupling may further contribute to reserve dysfunction, LA pressure rise, and susceptibility to pulmonary edema in HFpEF.

Background: Reduced left atrial (LA) compliance (LAC) contributes to pulmonary venous hypertension in heart failure with preserved ejection fraction (HFpEF); however the underlying mechanisms are poorly defined. We hypothesized that microvascular inflammation with rarefaction and impaired nitric oxide (NO) signaling, LA titin hypophosphorylation and fibrosis may each contribute to reduced LAC in HFpEF. Methods: Invasive LA pressure-volume (PV) analyses (open-chest; pericardiotomy; admittance catheter) were performed in 9 aged dogs with experimental hypertension (bilateral renal wrap + DOCA; HFpEF) compared to 13 sham-operated young adult dogs (control). LAC was measured at end-LA reservoir during acute preload reduction (cava occlusion). LA tissue and LA microvascular function were analyzed. Results: Compared to controls, the curvilinear end-LA reservoir PV relationship ( Fig A ) was shifted upward/leftward in HFpEF (LA diastolic stiffness constant 0.08 vs 0.16 mmHg/mL in controls, p=0.005). LA microvascular density (MVD; CD31 stain) tended to be lower in HFpEF than controls (2154±578 vs 2650±523 vessels/mm 2 , p=0.07) and LA MVD negatively correlated with the LA diastolic stiffness constant (r=-0.66, p=0.006). LA microvessels from HFpEF dogs showed impaired endothelium-dependent shear stress-mediated dilation compared to controls, consistent with impaired NO signaling ( Fig B ). Titin isoform expression was similar between groups, but isoform N2B (stiffer) was (p=0.03) and N2BA (compliant) tended to be (p=0.09) hypophosphorylated in HFpEF LA appendage (LAA) tissue. Fibrosis (Picrosirius Red) was similar between groups in the LA free wall and LAA, but increased in areas adjacent to the pulmonary veins in HFpEF (27±12% vs controls 15±8%, p=0.01). Conclusion: In experimental HFpEF, LAC was reduced in association with evidence of LA microvascular rarefaction and impaired NO signaling, titin hypophosphorylation and regional increases in LA fibrosis.

Angiotensin II (Ang II) type I receptor (ATR 1 ) trafficking into caveolae is essential for Ang II signaling, which is known to be abnormal in diabetic vessels. We have shown that the large conductance Ca 2+ actviated K + (BK) channels are also targeted to caveolae in vascular cells. The potential interaction between Ang II signaling and BK channel function in normal and diabetic vessels is unknown. Using whole-cell patch clamp recordings and molecular biology techniques, we examined the mechanisms through which caveolae targeting facilitates the regulation of BK channels by Ang II signaling. We found that in cultured human coronary arterial smooth muscle cells (CASMC) and in freshly isolated rat CASMC, BK channels, ATR 1 , and Src-family protein tyrosine kinases (Src-PTK) were colocalized and enriched in the low buoyant density, caveolae-rich fractions. 2 μM Ang II inhibited BK channel activity by ∼50% in rat and human CASMC and these effects were completely abolished by 2 μM Losartan (a selective ATR 1 inhibitor), 10 μM PP2 (a selective Src-PTK inhibitor), and by caveolin-1 (cav-1) knockdown using 60 nM siRNA. Similar results were obtained in HEK293 cells coexpressing hSlo, BK-β 1 subunit, ATR 1 , cav-1, and Src-PTK, indicating that inhibition of BK channels by Ang II was mediated through ATR 1 activation of Src-PTK and the integrity of caveolae is critical for Ang II signaling. Culturing human CASMC with high glucose (HG, 22 mM) enhanced Ang II-mediated BK channel inhibition (78.8±16.8% vs. 54.5±15.7% in 5 mM glucose, n=3, p<0.05). Analysis of ATR 1 , Src-PTK, and BK channel distribution by sucrose gradient fractionation and by co-immunoprecipitation with anti-cav-1 antibodies showed that expression of ATR 1 and Src-PTK were up-regulated in human CASMC cultured in HG and in CASMC from streptozotocin-induced diabetic rats. Total BK channel protein in these cells was diminished, but the amount of BK channels co-immunoprecipitated with anti-cav-1 antibody was increased, suggesting increased caveolae targeting of BK channels in diabetes, which leads to enhanced Ang II-mediated BK channel inhibition. These results indicate that Ang II-BK channel interaction is critically dependent upon caveolae targeting under normal conditions and in disease states such as diabetes.

Introduction: Left atrial appendage (LAA) flow depends largely on left ventricular compliance and may play a role in mediating the regulation of left atrial volume-pressure relationships. Hypothesis: We hypothesize that LAA emptying flow velocity (LAAEV) is a predictive factor of long-term outcomes (e.g. recurrent AF, stroke, and survival) after cardioversion for non-valvular AF. Method: We identified 3,251 consecutive patients with non-valvular AF who underwent successful TEE-guided electrical cardioversion (ECV) at our institution between May 2000 and March 2012. Successful ECV was defined as sinus rhythm at time of discharge from the cardioversion unit. Patients were monitored following their ECV procedure for first documentation of recurrent AF, stroke or death. Multivariate Cox proportional hazards models were used to identify independent predictors of long-term outcomes. Patients with >= moderate valvular regurgitation or stenosis were excluded. Results: Among the 3,251 patients who were successfully cardioverted to sinus rhythm, the mean (±SD) LAAEV was 38.43±23 cm/s and the median was 33 cm/s, (interquartile range [IQR], 20-50). Patients with LAAEV <=33 cm/s had higher CHA 2 DS 2 -VASc score (2.6±1.2 vs. 1.9±1.3, P =.009), larger LAVI (52.0±20.9 cc/m 2 vs. 43.3±13.6 cc/m 2 , P <.001) than those with LAAEV >33 cm/s. Pre or post-procedure antiarrhythmic drug use was similar between the two groups. During 1-year follow-up, patients with LAAEV <=33 cm/s had significantly higher rate of AF recurrence than those with LAAEV >33 cm/s (55% vs 45%, P <.001). Likewise, during a mean follow-up of 4.9±3.6 years, similar patterns in 5-year rates were observed for first recurrence of AF (81% vs 73%, P <.001), stroke (7% vs 4%, P =.003) and mortality (31% vs 23%, P <.001) for LAAEV <=33 vs > 33cm/s, respectively. Stepwise multivariate Cox regression analysis revealed that LAAEV <=33 cm/s, age, CHA 2 DS 2 -VASc score were independent predictors of AF recurrence, stroke and mortality. Conclusions: LAA emptying flow velocity is an effective and convenient method for risk stratification of patients undergoing cardioversion for AF. Our data showed that patients with reduced LAAEV have an increased risk for AF recurrence, stroke and death following electrical cardioversion.

Atrial fibrillation (AF) is the most common cardiac arrhythmia and is a major cause of thromboembolic stroke in the United States. The risk of stroke is a 5-fold increase in patients with AF. The Sorbin and SH3 domain-containing protein 2 (Sorbs2) is a scaffolding protein known to interact with actin and several other cytoskeletal proteins in the intercalated discs and Z band of cardiomyocytes. However, the role of Sorbs2 gene in cardiovascular diseases is largely unknown. Using the loss-of-function of Sorbs2 ( Sorbs2 e8/e8 ) mice, we found that Sorbs2 e8/e8 mice had an abnormal lifespan and die at 4 to 7 months of age. Echocardiogram examine revealed that there was a significant decrease in right ventricle (RV) and left ventricular (LV) ejection fraction (EF), increase in RV end-diastolic volume (EDV) and end-systolic volume (ESV), and progressive chamber enlargement in left atria (LA) and RV, particularly in LA with severe fibrosis and intramural thrombus formation, in Sorbs2 e8/e8 mice after 3 months of age. Surface electrocardiograph also showed a significant conduction delay and spontaneous arrhythmias including those of bifid P wave configuration, right bundle branch block (RBBB), spontaneous AF and non-sustained ventricular tachycardia (VT) in Sorbs2 e8/e8 mice. Intracellular recordings in the freshly isolated LA demonstrated that Sorbs2 e8/e8 myocardium had a delayed action potential activation in response to electrical impulse excitation, reduced amplitude and upstroke velocity (dV/dt) at phase 0, slowed repolarization at phase 1 and phase 2, prolonged atrial effective refractory period (aERP), action potential durations at 50% (APD 50 ) and 90% (APD 90 ) repolarization, and decreased resting potentials. Moreover, the mRNA and protein expressions of voltage-gated Na + channels, L-type Ca 2+ channels, transient outward K + channels and inward rectifier K + channels were significantly downregulated in the LA of Sorbs2 e8/e8 mice, compared wo those of wild type mice. Our results indicate that Sorbs2 gene plays an important role in atrial myocardium architecture and electrophysiology. Sorbs2 dysfunction leads to atrial cardiomyopathy, spontaneous AF and cardiac embolism in mice. Hence, Sorbs2 is a new genetic association with cardioembolic stroke.