Mechanisms of enhanced beta-adrenergic reserve from cardiac resynchronization therapy

Abstract

Background: Cardiac resynchronization therapy (CRT) is the first clinical heart failure treatment that improves chamber systolic function in both the short-term and long-term yet also reduces mortality. The mechanical impact of CRT is immediate and well documented, yet its long-term influences on myocyte function and adrenergic modulation that may contribute to its sustained benefits are largely unknown.

Methods and results: We used a canine model of dyssynchronous heart failure (DHF; left bundle ablation, atrial tachypacing for 6 weeks) and CRT (DHF for 3 weeks, biventricular tachypacing for subsequent 3 weeks), contrasting both to nonfailing controls. CRT restored contractile synchrony and improved systolic function compared with DHF. Myocyte sarcomere shortening and calcium transients were markedly depressed at rest and after isoproterenol stimulation in DHF (both anterior and lateral walls), and CRT substantially improved both. In addition, beta(1) and beta(2) stimulation was enhanced, coupled to increased beta(1) receptor abundance but no change in binding affinity. CRT also augmented adenylate cyclase activity over DHF. Inhibitory G-protein (Galpha(i)) suppression of beta-adrenergic stimulation was greater in DHF and reversed by CRT. Galpha(i) expression itself was unaltered; however, expression of negative regulators of Galpha(i) signaling (particularly RGS3) rose uniquely with CRT over DHF and controls. CRT blunted elevated myocardial catecholamines in DHF, restoring levels toward control.

Conclusions: CRT improves rest and beta-adrenergic-stimulated myocyte function and calcium handling, upregulating beta(1) receptors and adenylate cyclase activity and suppressing G(i)-coupled signaling associated with novel RGS upregulation. The result is greater rest and sympathetic reserve despite reduced myocardial neurostimulation as components underlying its net benefit.

Fig. 1

A) Example radial strain-vs-time tracings for septal and lateral walls in a healthy control, dyssynchronous failing heart (DHF), and heart treated with bi-ventricular pacing (CRT). Marked disparities in regional strain in DHF were ameliorated by CRT. B) Corresponding pressure-strain loops show disparities in regional work (loop area) in DHF that is rendered more homogeneous by CRT. C) Group dyssynchrony analysis (standard deviation of time at peak systolic radial strain from multiple segments) shows marked discoordination in both groups at 3-wks that is corrected by CRT but persists in DHF hearts. *p<0.0001 versus 6wk CRT and control (CON). D) Echo-derived ejection fraction (EF) and stroke volume (SV) at 3 and 6 wk time points in each group. Both rose in CRT compared to DHF at 6-wks (p-values shown for unpaired analysis; RMANOVA; p<0.01 for group×time interaction). E) Invasive pressures in both models at 6wk terminal study. EDP-end-diastolic pressure; SBP-systolic pressure. * p<0.01 versus CON; † - p<0.01 vs NL, and p<0.05 vs DHF.

Fig. 2

A) Myocyte sarcomere length-time tracings and Ca²⁺ transients obtained from control, DHF, or CRT hearts (37°, 1 Hz stimulation). Data are shown at rest (bold) and after ISO stimulation (thin). Both rest and ISO stimulated function and Ca²⁺ transients were depressed in DHF and improved by CRT. B) Summary results support these examples; showing depressed function and Ca²⁺ handling was seen in both anterior (A) and lateral (L) walls in DHF, and both were improved by CRT (Mean±SEM; n=10–35 cells from 3 to 6 hearts for each data point). *, P≤0.05 vs. control and CRT; † p<0.05 vs anterior. C) Radiolabeled affinity binding assays for β-AR. Upper panels show raw data and Scatchard plots from which total binding (B_max, receptor density) and binding affinity (K_m) were determined (lower panels). B_max declined with DHF (* p<0.05 versus con) and was restored towards normal by CRT. Binding affinity was unaltered.

Fig. 3

A) Change in sarcomere shortening (SS), mean shortening velocity (MSV) and mean re-lengthening velocity (MRV) with selective stimulation of β₁ or β₂-AR (data obtained at 27°, 0.5 Hz stimulation, n=15–30 cells/condition from 3–6 different hearts ). Both were depressed in DHF hearts, but became more similar to controls with CRT (* p<0.01 vs CON and CRT; † p≤0.05 vs. CON). B) Both β₁ and β₂ mRNA expression declined in DHF. β₁ increased with CRT whereas β₂ remained reduced (* p<0.05 vs DHF and CON; †-p<0.05 versus CON), increasing the net β₁/β₂ ratio (* p<0.01 vs DHF and CON). C) Receptor number based on competition binding assays with selective inhibitors confirmed differential upregulation of β₁ versus β₂ by CRT (* p<0.05 versus CON and CRT). D) Immune-blot of GRK-2 from membrane fraction. Expression rose in both DHF and CRT and was similar between groups (* p<0.05 versus CON). Equal protein loading was confirmed by Ponceau stain.

Fig. 4

A) Examples of influence of adenylate cyclase activation by forskolin (FSK) on sarcomere shortening and Ca²⁺ transients in myocytes from CON, DHF, and CRT hearts (n=15–30 cells in each group and region from n=3–6 different hearts). B) Summary data. In DHF, FSK stimulated shortening was very depressed, even with peak Ca²⁺ transients enhanced to control levels in the anterior wall (* p<0.05 versus CON, †-p<0.05 versus lateral). FSK-stimulated shortening was greatly improved in CRT though not quite to control levels († p<0.05 vs CON), while peak Ca²⁺ was restored to normal response levels. C) Adenylate cyclase activity assessed by cAMP generation assay in response to ISO or FSK. Data are shown normalized to control (15±1 – ISO, 201±14 – FSK, fMol cAMP/mg protein/min). In DHF hearts, AC activity was depressed in both regions and with both stimuli, and this was improved by CRT (* p<0.05 vs CON, †p<0.03 vs DHF; ‡p<0.001 vs CON, §p<0.02 vs DHF).

Fig. 5

A) Sarcomere shortening in DHF and CRT myoctes at rest (Baseline, top panels) and with ISO stimulation (lower panels), with or without pertussis toxin (PTX) pre-treatment. PTX did not alter rest shortening in either group; however, it markedly enhanced the ISO response in DHF myocytes, achieving levels observed in CRT myocytes without PTX. CRT myocytes, by contrast, showed no change in shortening magnitude despite PTX administration. B) Summary data (n=10–30 cells from 3–5 hearts in each condition; *, p<0.05 vs. DHF ± PTX; † p<0.05 vs. all other conditions).

Fig. 6

Protein regulation of Gαi and RGS proteins. A) Membrane GαI (1,2,3) increased in DHF (* p<0.01 versus controls), and slightly more in CRT hearts (†, <0.05 versus DHF). Equal loading was conformed by Ponceau stain. B) Differential expression of RGS proteins by DHF versus CRT. Protein is from lateral endocardium with four different animals shown for each group. Both RGS 2 and RGS3 were markedly upregulated in CRT hearts, but not DHF, whereas RGS4 increased in both groups similarly over control. Summary data to the right are normalized to GAPDH. †-p<0.05 versus other two groups. * p<0.05 versus control. C) Differential up-regulation of RGS2 and RGS3 in CRT hearts was similar in both anterior and lateral myocardium, and in endocardial (en) and epicardial (ep) layers. * p<0.05, † p<0.001 versus respective DHF data. D) Gene expression (rtPCR) of RGS proteins shown normalized to GAPDH. All increased in CRT compared to control, with changes in RGS3 and RGS4 only seen in CRT (*p<0.05 vs other groups; †p<0.01 versus other groups).

Figure. 7

Myocardial catecholamines increase more in DHF than CRT hearts. Data are shown for four different catecholamines measured by HPLC from frozen myocardial tissue. Results are provided for anterior and lateral regions separately, and both combined. There was a tendency for higher levels in the lateral wall, though this did not reach significance for norepinerphrine or dopamine. CRT generally reduced levels in both regions (* p<0.05 vs. control and CRT; †p≤0.05 vs. control).