Inhibition of renin-angiotensin system (RAS) reduces ventricular tachycardia risk by altering connexin43

Abstract

Renin-angiotensin system (RAS) activation is associated with arrhythmias. We investigated the effects of RAS inhibition in cardiac-specific angiotensin-converting enzyme (ACE) overexpression (ACE 8/8) mice, which exhibit proclivity to ventricular tachycardia (VT) and sudden death because of reduced connexin43 (Cx43). ACE 8/8 mice were treated with an ACE inhibitor (captopril) or an angiotensin receptor type-1 blocker (losartan). Subsequently, electrophysiological studies were performed, and the hearts were extracted for Cx43 quantification using immunoblotting, immunohistochemistry, fluorescent dye spread method, and sodium current quantification using whole cell patch clamping. VT was induced in 12.5% of captopril-treated ACE 8/8 and in 28.6% of losartan-treated mice compared to 87.5% of untreated mice (P < 0.01). Losartan and captopril treatment increased total Cx43 2.4-fold (P = 0.01) and the Cx43 phosphorylation ratio 2.3-fold (P = 0.005). Treatment was associated with a recovery of gap junctional conductance. Survival in treated mice improved to 0.78 at 10 weeks (95% confidence interval 0.64 to 0.92), compared to the expected survival of less than 0.50. In a model of RAS activation, arrhythmic risk was correlated with reduced Cx43 amount and phosphorylation. RAS inhibition resulted in increased total and phosphorylated Cx43, decreased VT inducibility, and improved survival.

Fig. 1

a–d Surface and intracardiac electrograms from WT (a), untreated ACE 8/8 (b), captopril-treated ACE 8/8 (c), and losartan-treated ACE 8/8 (d) mice. Note the prolonged AH and HV intervals in ACE 8/8 mouse, reversed after treatment with either captopril or losartan. The surface ECGs are of the same scale; however, intracardiac signals are depicted with different vertical scales. e An example of induced VT during programmed ventricular stimulation in untreated ACE 8/8 mice

Fig. 2

a An immunoblot comparing Cx43 in WT and ACE 8/8 mice. The top row shows total Cx43. P0 runs at 41 kDa, whereas P1 and P2 are mostly phosphorylated and migrate at 44–46 kDa. Note the reduction in total and phosphorylated Cx43 in ACE 8/8. GAPDH staining was used for loading control. b Immunoblot showing the effect of captopril treatment on Cx43 in ACE 8/8 mice. c Immunoblot showing the effect of losartan treatment on Cx43. d Normalized total Cx43 in WT vs. ACE 8/8 mice ventricles. e (P1 + P2)/P0 ratio in WT vs. ACE 8/8 mice. f Normalized total Cx43 in ACE 8/8 mice after treatment with captopril and losartan (n=12 for untreated, n=10 for captopril, and n=8 for losartan-treated groups). g (P1 + P2)/P0 ratio in ACE 8/8 mice after treatment with captopril and losartan (n=13 for untreated, n=10 for captopril, and n=11 for losartan-treated groups). Student t test was used to analyze (c) and (d), one-way ANOVA was used for (e) and (f), with Dunnett’s test of significance to compare each treatment group with the control (*P<0.01, **P<0.05). Error bars represent mean ± SEM

Fig. 3

IHC-stained images showing representative fields from left ventricular myocardium of WT (a), ACE 8/8 (b), captopril-treated ACE 8/8 (d), and losartan-treated ACE 8/8 (d) mice. Cx43 is stained brown. (e) Cx43 area percentage (see text) is significantly reduced in ACE 8/8 ventricle compare to WT and recovered after treatment (combined captopril and losartan group)

Fig. 4

Results of merged fluorescent dye transfer images in WT (a), ACE 8/8 (b), and treated ACE 8/8 (C) mice. Red represents TXD staining that is confined to the injected cell, whereas green depicts LY staining that diffuses through gap junctions. Gap junctional conductance correlates with TXD-corrected LY spread. There is a significant reduction in dye spread in ACE 8/8 mice, which partially recovers after treatment (d). This effect is more pronounced in the longitudinal direction (e) (n=7 for WT, n=7 for ACE 8/8 no treatment group, and n=7 for the combined treatment group)

Fig. 5

a Current–voltage plot for sodium current in WT and ACE 8/8 mice. The experiments were performed with 10 ms pulses over a voltage range of −90 to +55 mV following a holding potential of −110 mV. The peak current density was −38.6±2.5 pA/pF (n=7) in WT compared to −42.7±12.6 pA/pF (n=3) in ACE 8/8 (P=NS). b Comparison of sodium current conductance (right) and steady-state inactivation (left) in WT and ACE 8/8 mice. The conductance was calculated based on the protocol presented in (a), whereas for the steady-state inactivation, holding potentials ranging from −110 to −20 mV were sustained for 1 s and followed by a 10 ms test pulse of −10 mV. The membrane voltage at half-maximal conductance was −34.2±1.5 mV (n=7) in WT and −42.6±2.9 mV (n=3) in ACE 8/8 mice (P=0.02). The voltage of half-maximal inactivation was −70.8±2.1 mV (n=4) and −76.0±4.0 mV (n=4), respectively (P=NS)

Fig. 6

a A representative immunoblot showing the association between VT inducibility and Cx43 expression in ACE 8/8 mice. GAPDH was used as a loading control. b Total Cx43 as a function of VT inducibility (n=6 for non-inducible and n=8 for inducible groups). c (P1 + P2)/P0 as a function of VT inducibility (n=7 for non-inducible and n=9 for inducible groups). Error bars represent mean ± SEM