Reversal of right ventricular failure by chronic α_1A-subtype adrenergic agonist therapy

Abstract

Right ventricular (RV) failure (RVF) is a serious disease with no effective treatment available. We recently reported a disease prevention study showing that chronic stimulation of α_1A-adrenergic receptors (α_1A-ARs), started at the time of RV injury, prevented the development of RVF. The present study used a clinically relevant disease reversal design to test if chronic α_1A-AR stimulation, started after RVF was established, could reverse RVF. RVF was induced surgically by pulmonary artery constriction in mice. Two weeks after pulmonary artery constriction, in vivo RV fractional shortening as assessed by MRI was reduced by half relative to sham-operated controls (25 ± 2%, n = 27, vs. 52 ± 2%, n = 13, P < 10⁻¹¹). Subsequent chronic treatment with the α_1A-AR agonist A61603 for a further 2 wk resulted in a substantial recovery of RV fractional shortening (to 41 ± 2%, n = 17, P < 10⁻⁷ by a paired t-test) along with recovery of voluntary exercise capacity. Mechanistically, chronic A61603 treatment resulted in increased activation of the prosurvival kinase ERK, increased abundance of the antiapoptosis factor Bcl-2, and decreased myocyte necrosis evidenced by a decreased serum level of cardiac troponin. Moreover, A61603 treatment caused increased abundance of the antioxidant glutathione peroxidase-1, decreased level of reactive oxygen species, and decreased oxidative modification (carbonylation) of myofilament proteins. Consistent with these effects, A61603 treatment resulted in increased force development by cardiac myofilaments, which might have contributed to increased RV function. These findings suggest that the α_1A-AR is a therapeutic target to reverse established RVF.

NEW & NOTEWORTHY Currently, there are no effective therapies for right ventricular (RV) failure (RVF). This project evaluated a novel therapy for RVF. In a mouse model of RVF, chronic stimulation of α_1A-adrenergic receptors with the agonist A61603 resulted in recovery of in vivo RV function, improved exercise capacity, reduced oxidative stress-related carbonylation of contractile proteins, and increased myofilament force generation. These results suggest that the α_1A-adrenergic receptor is a therapeutic target to treat RVF.

INTRODUCTION

Right ventricular (RV) failure (RVF) occurs when the RV is unable to provide adequate blood flow through the pulmonary circulation at a normal preload (8). RVF is a serious clinical problem with a poor prognosis (8, 22). RVF has diverse causes including primary cardiomyopathies, RV ischemia/infarction, pressure loading due to pulmonic stenosis, or pulmonary hypertension from multiple causes and volume loading due to congenital heart disease and valvular pathologies (13).

Management of RVF remains suboptimal, and new therapies for RVF are needed (13). Therapies developed to treat patients with left heart failure are ineffective for improving function or survival of patients with RVF (22).

Numerous lines of evidence suggest that cardiac α₁-adrenergic receptors (α₁-ARs), in particular the α_1A-AR subtype, represent an endogenous cardioprotective system that exerts powerful beneficial effects in multiple clinical and experimental settings (1, 11, 18). Consistent with this, we recently reported that in a model of RVF induced by pulmonary fibrosis, chronic treatment with the α_1A-AR agonist A61603 minimized the development of RV dysfunction and RV myocyte injury (6). However, this previous study used a disease prevention design. Thus, it remained unclear whether A61603 treatment could reverse already established RVF. This clinically relevant question was addressed in the present study.

We induced RVF by pulmonary artery constriction (PAC) in mice. Two weeks later, when RVF was already evident, chronic treatment with A61603 was given for a further 2 wk at a very low dose that does not change blood pressure (17). We found that chronic A61603 treatment reversed multiple indexes of RVF and was associated with reduced oxidative stress, reduced oxidative stress-mediated modification of myofilament proteins, and improved contraction of cardiac myofilaments.

We conclude that the α_1A-AR is a potential therapeutic target to reverse established RVF.

METHODS

The San Francisco Veterans Affairs Medical Center is accredited by the American Association for the Accreditation of Laboratory Animal Care (Institutional Public Health Service Assurance No. A3476-01). The study was approved by the Animal Care and Use Subcommittee of the San Francisco Veterans Affairs Medical Center and conformed with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Revised 2011).

RV failure model and A61603 therapy.

Adult male C57BL/6J mice (Jackson Laboratory) were used (n = 73, age: 10–14 wk, and body weight: ≈25 g) at the beginning of the experiment. We used a pulmonary artery constriction (PAC) model of RVF as previously described (21). Animals were anesthetized with isoflurane (3% induction, 1.5% maintenance) and intubated, and a left lateral thoracotomy was performed. A 7-0 suture was placed around the pulmonary artery and used to tie the artery to a 26-gauge needle. The needle was quickly removed, resulting in a ligature constricting the pulmonary artery. The chest and skin were sutured closed. Sham control surgeries were identical, including placement of the suture, except that the suture around the artery was not tied. All sham surgery controls (n = 20) survived to the end of the experiment. Of the 53 mice subjected to PAC, 41 mice survived after 2 wk (77% survival), and these were allocated to treatment (n = 17) or no treatment (n = 24) groups for a further 2 wk.

For the treatment group, 2 wk after PAC, 17 mice were chronically treated for a further 2 wk with A61603 (Tocris Bio-Techne, Minneapolis, MN), a potent and highly specific agonist for the α_1A-AR that does not stimulate the other two α₁-AR-subtypes (α_1B or α_1D). A low dose of A61603 (10 ng·kg⁻¹·day⁻¹) (17) was given by continuous subcutaneous infusion with an osmotic minipump (Alzet model 1002, Durect) implanted between the scapulae under isoflurane anesthesia. This dose has been reported not to increase systolic or diastolic blood pressure and to be close to the EC₅₀ for activation of cardiac ERK signaling (17). Previously, A61603 has been reported to increase blood pressure, as expected for an α₁-AR agonist, with an EC₅₀ equal to 300 ng/kg during acute intravenous infusion in mice (23). Thus, the A61603 dose used in the present study was far below that required to increase blood pressure (17). All mice survived during the 2-wk A61603 treatment.

For the no treatment group, 2 wk after PAC, 24 mice were implanted with minipumps that delivered only saline for a further 2 wk. Of these, 22 mice survived after the 2 wk no treatment period.

MRI assessment of in vivo hemodynamics.

Cine cardiac magnetic resonance data were acquired with a 1-T spectrometer (Buker, ICON) from anesthetized mice (isoflurane at 3% induction and 1.25–1.5% maintenance to keep respiration at 90–110 breaths/min). Images were obtained using a retrospective gated cine gradient echo sequence with the following parameters: flip angle, 25°; echo time, 7 ms; repetition time, 10 ms; image size, 128 × 128 pixels; field of view, 22.5 × 22.5 mm; slice thickness, 1.25 mm; and scan time, 12 min. A single axial slice image was obtained midway between the apex and base. Postacquisition, the area of the RV chamber at end systole and end diastole was determined. RV fractional shortening (FS) was calculated from the diastolic minus systolic chamber area and expressed as a percentage of the diastolic chamber area. An estimate of RV cardiac output was computed from the difference between the diastolic versus systolic chamber area multiplied by heart rate as assessed by MRI.

Echocardiography assessment of the RV pressure gradient.

Echocardiography was performed on conscious, gently restrained mice using an Acuson S2000 (Siemens) with a 14L5 SP linear array probe. Color Doppler on two-dimensional imaging was used to detect the presence of the stenosis. Pulse wave Doppler was used to evaluate the stenosis and determine the pressure gradient.

Voluntary exercise.

After mice had been trained for 2 nights, exercise wheels were placed overnight in a cage containing a single mouse. Electronic monitors recorded distance run and total run time.

Serum cardiac troponin I.

Mice were anesthetized with isoflurane and heparinized (100 units ip). A midline thoracotomy was performed, and blood was collected by apical left ventricular puncture with a 20-gauge needle. Whole blood, 1−2 drops, was assayed immediately for cardiac troponin I (cTnI) concentration (in ng/ml) using VetScan i-STAT cTnI cartridges (Abaxis).

Fibrosis.

Freshly isolated RV free wall samples were fixed in phosphate-buffered 4% paraformaldehyde (Fisher Scientific, Fair Lawn, NJ) at 4°C for at least 24 h. Paraffin-embedded 5-µm-thick sections were stained with hematoxylin and eosin and picrosirius red as previously described (5). Images of stained cross sections (3 images/animal) were quantified using ImageJ (National Institutes of Health, Bethesda, MD) thresholding analysis to quantify collagen-positive areas.

Ex vivo contraction of demembranated RV cardiac trabeculae.

Trabeculae from mouse hearts were demembranated, and ex vivo force development was measured in activating solutions of various Ca²⁺ concentrations ([Ca²⁺]) as recently described (6). Measurements of muscle force were normalized to muscle cross-sectional area. For each experiment, the relationship between developed force (F; active minus passive force) versus [Ca²⁺] was fit to the following Hill equation: $\text{F}={\text{F}}_{\text{max}}\hspace{0.17em}\times \hspace{0.17em}{\left[{\text{Ca}}^{2+}\right]}^n{}^{{}_{\text{H}}}/\left({\left[{\text{Ca}}^{2+}\right]}^n{}^{{}_{\text{H}}}+{\text{EC}}_{50}{{}^n}^{{}_{\text{H}}}\right)$, where F_max is the maximum Ca²⁺-activated force, EC₅₀ is the [Ca²⁺] at which F is 50% of F_max, and n_H is the Hill coefficient reflecting the slope of the Ca²⁺-force relationship at EC₅₀.

Reactive oxygen species assay.

RV free wall samples were homogenized at 1:10 (wt/vol) in 4°C standard relaxation buffer [containing (in mM) 10 imidazole, 75 KCl, 2 MgCl₂, 2 EDTA, and 1 NaN₃, pH 7.2] (25). Samples were centrifuged for 15 min at 10,000 g at 4°C, and the supernatant was collected and stored at −80°C for use in the reactive oxygen species (ROS) assay. The pellets were homogenized in 4°C standard relaxation buffer with 1% (vol/vol) Triton X-100. Samples were spun for 15 min at 10,000 g at 4°C, and the supernatant was collected and stored at −80°C for Western blot analyses (see below). The previous step was repeated but this time the supernatant was discarded. The pellet was homogenized with standard relaxation buffer without Triton X-100 and spun at 1,000 g for 10 min at 4°C, and the supernatant was discarded. The pellet of enriched myofibrillar proteins was solubilized in Laemmli sample buffer (Sigma), boiled for 5 min, spun clarified, and then stored at −80°C for analyses of carbonylated myofibrillar proteins by Western blot (see below). Protease and phosphatase inhibitors (Cell Signaling) were present during each of the steps described above.

The fluorescent dye dihydrorhodamine 123 (DHR123) was used as a ROS probe (14, 15). The reaction was initiated by adding 15 μg tissue homogenate to 10 μM DHR123, which was incubated at 37°C in the dark for 30 min. Fluorescence was measuring using the blue optical kit (490-nm excitation and 510- to 57-nm emission) of the GloMax-Multi Detection System (Promega).

Western blot analysis.

Quantification of protein concentrations was performed using the 660-nm protein assay (Thermo Scientific). Samples were prepared for immunoblot analysis by diluting equal amounts of protein in Laemmli sample buffer (Bio-Rad) followed by boiling the samples for 5 min. Equal amounts of protein were separated on 4–20% Criterion TGX SDS-PAGE gels and then transferred to polyvinylidine difluoride or nitrocellulose membranes. Membranes were blocked in 5% nonfat milk in Tris-buffered saline-Tween (0.05%) for 1 h and then incubated overnight at 4°C with the primary antibody for glutathione peroxidase 1 (GPx1; 1:250, AF3798, Novus Biologicals) or Bcl-2 (1:1,000, AAP-07, StressGen). For total ERK1/2 (1:1,000, no. 9102, Cell Signaling) and phospho-ERK1/2 (1:1,000, no. 4370, Cell Signaling; Thr²⁰²/Tyr²⁰⁴-p44/42 MAPK), the primary antibodies were incubated in 5% BSA-Tween (0.05%). The membrane was then washed with Tris-buffered saline-Tween and incubated with horseradish peroxidase-conjugated anti-rabbit IgG antibody (no. 111-035-144, Jackson ImmunoResearch) in 5% nonfat skim milk for 1 h at room temperature. The immunoreactive bands were developed using Clarity Western ECL substrate (Bio-Rad), and the membranes were exposed to light-sensitive film or captured using the ChemiDoc XRS system (Bio-Rad). Images were analyzed using ImageJ. To ensure equal protein loading and transfer, membranes were stained with 0.2% India ink solution in PBS (Alfa Aesar) (24) or with the MemCode Reversible Protein Stain Kit (Thermo Scientific). The membrane was scanned, and individual protein bands were quantified using ImageJ.

GPx1 and Bcl-2 were expressed relative to the sum of the protein bands per lane. Phospho-ERK and total ERK consisted of a pair of bands (44 and 42 kDa). The intensities of both bands were combined, and phospho-ERK1/2 was expressed relative to total ERK1/2. For each of these three assays, all samples were contained within the same blot. Data were normalized to the mean value of data in the sham group.

Analysis of carbonylated myofibrillar proteins.

Quantification of protein concentration was performed using the 660-nm protein assay (Thermo Scientific). Carbonylated myofibrillar proteins were detected using the OxyBlot kit (EMD Millipore) according to the manufacture’s protocol. Protein carbonyl groups were derivatized by reaction with 2,4-dinitrophenylhydrazane for 15 min, neutralized, and loaded onto 12% SDS-PAGE gels for Western blot analysis. Protein was transferred to polyvinylidene fluoride membranes. Membranes were stained with the MemCode Reversible Protein Stain Kit (Thermo Scientific) to ensure equal protein loading and transfer. Membranes were blocked with 5% skim milk and incubated with the anti-dinitrophenyl primary antibody (1:150) overnight at 4°C and then with goat anti-rabbit secondary antibody (1:150) in 5% nonfat skim milk for 1 h at room temperature. The immunoreactive bands were developed and analyzed according to the procedures described above. Carbonylated protein bands were expressed relative to the sum of the protein bands per lane. To confirm the identity of the carbonylated bands, the membrane was stripped with Restore Western blot stripping buffer (Thermo Scientific) and reprobed with primary antibodies for actin (1:1,000, sc-53141, Santa Cruz Biotechnology) and tropomyosin (1:500, sc-58868, Santa Cruz Biotechnology) according to the procedures described above. Secondary antibody was a mouse IgGκ light chain-binding protein (sc-516102, Santa Cruz Biotechnology). This assay was run using three different gels, each containing samples from all three groups. All samples were derived at the same time, and the three Western blots were processed together. For each blot, data were normalized to the mean value of data in the sham group, and then data from all three blots were pooled.

Statistical analysis.

Data are presented as means ± SE. Statistical tests (paired or unpaired t-tests and one-way or two-way ANOVA with post hoc analysis using Dunnett’s test for multiple comparisons) were performed using Prism 7 software (GraphPad Software, La Jolla, CA) with a significance level set at P < 0.05.

RESULTS

PAC model of RV failure.

Figure 1 shows cross-sectional images of mouse hearts obtained at end systole using cardiac MRI 4 wk after PAC or sham surgery. PAC caused marked dilation of the RV and flattening and displacement of the septum toward the left ventricle.

Fig. 1.

Effect of pulmonary artery constriction (PAC) on right ventricular (RV) structure. A: axial images obtained at the midventricle level at the end of systole using cine cardiac MRI. Images are from hearts 4 wk after sham or PAC surgery. The outline of the RV chamber is indicated. Four weeks after PAC, the RV was markedly dilated and the left ventricular (LV) septum was flattened and displaced toward the LV. Scale bar = 5 mm. B: RV weight referenced to body weight for sham-operated mice and mice subjected to PAC for 4 wk with chronic treatment with either vehicle saline (Veh) or A61603 (10 ng·kg⁻¹·day⁻¹) for the final 2 wk. RV weight was increased by PAC (PAC + Veh: n = 13 vs. sham: n = 17, ****P < 0.0001 by ANOVA with Dunnett’s multiple-comparisons test). RV weight after PAC was not affected by A61603 treatment (PAC + A61603: n = 16 vs. PAC + Veh: n = 13, ^nsP = 0.98 by ANOVA, where ns is not significant). C: RV fibrosis, as measured by picrosirius red staining, was increased by PAC (PAC + Veh: n = 9 vs. sham: n = 7, ****P < 0.0001 by ANOVA) but not significantly changed by A61603 treatment (PAC + A61603: n = 8 vs. PAC + Veh: n = 9, ^nsP = 0.19 by ANOVA).

For mice subjected to PAC, pulse wave Doppler echocardiography assessment revealed the presence of a pressure gradient of 46 ± 1 mmHg (n = 26) across the constriction 1 wk after surgery. No pressure gradient was detectable in sham-operated control mice.

PAC resulted in RV hypertrophy as evidenced by doubling of RV free wall weight measured at necropsy (Fig. 1B) and marked RV fibrosis (Fig. 1C). Chronic treatment with the α_1A-AR agonist A61603, the therapeutic intervention tested in this study, had no effect on the levels of RV hypertrophy or RV fibrosis.

Chronic A61603 treatment reversed RV dysfunction.

Figure 2A shows cardiac MRI assessment of in vivo RV function from measures of RV FS. Two weeks after surgery, the RV FS for sham control mice (52 ± 2%, n = 13) was not changed from the initial level, but for mice subjected to PAC, RV FS was reduced by half (to 25 ± 2%, n = 27, P < 10⁻¹¹ by an unpaired t-test). Importantly, for mice subjected to PAC for 2 wk, subsequent chronic treatment with A61603 for a further 2 wk resulted in a substantial reversal of RV dysfunction (RV FS increased to 41 ± 2%, n = 17, P < 10⁻⁷ by a paired t-test).

Fig. 2.

A61603 treatment increased right ventricular (RV) function in vivo measured by MRI. A: right ventricular (RV) fractional shortening (FS) was measured repeatedly in the same animals with time after sham or pulmonary artery constriction (PAC) surgery. Chronic treatment with either vehicle saline (Veh) or A61603 was started at 2 wk. The decrease of RV FS with PAC was significantly reversed by A61603 treatment (****P < 0.0001 by a paired t-test). Inset: there was no difference in heart rate (HR; in beats/min) among treatment groups during MRI assessments 4 wk after surgery. B: RV chamber area at end systole was increased by PAC (PAC + Veh: n = 9 vs. sham: n = 13, ****P < 0.0001 by ANOVA), and this increase was partly reversed by A61603 treatment (PAC + A61603: n = 15 vs. PAC + Veh: n = 9, **P < 0.01 by ANOVA). C: RV chamber area at end diastole was increased by PAC (PAC + Veh: n = 9 vs. sham: n = 13, ***P < 0.0001 by ANOVA), but this increase was not reversed by A61603 treatment (PAC + A61603: n = 15 vs. PAC + Veh: n = 9, ^nsP = 0.76 by ANOVA, where ns is not significant).

Heart rate, obtained during MRI assessment of RV FS, did not differ among groups (Fig. 2A, inset). Thus, the increase of RV FS after A61603 treatment did not involve a heart rate effect.

The decrease in RV FS after PAC was associated with an increased cross-sectional area of the RV chamber in both systole and diastole (Fig. 2, B and C). The improvement of RV FS mediated by A61603 was due to a reduction in RV chamber area during systole (Fig. 2B), without an effect on the diastolic RV chamber area (Fig. 2C). Thus, an improvement in systolic contraction was responsible for the improved RV FS with A61603 treatment.

In a subset of sham-operated animals, RV FS after 2 wk (53 ± 4.6%, n = 3) was not changed after chronic treatment with A61603 for a further 2 wk (52.3 ± 3.2%, n = 3, P = 0.7 by a paired t-test), indicating that in the absence of disease, low-dose chronic A61603 treatment did not increase RV contraction.

Chronic A61603 treatment reversed RVF.

Estimated RV cardiac output was decreased by 30 ± 6% (P < 0.01 by an unpaired t-test) 2 wk after PAC. This decrease was reversed by subsequent treatment with A61603 for 2 wk (P < 10⁻⁶ by a paired t-test). PAC resulted in increased liver weight (Fig. 3A), a sign of RV failure, that was normalized after chronic A61603 treatment. PAC also resulted in an increased serum cTnI level, an indicator of cardiac injury (Fig. 3B), that was substantially reversed by chronic A61603 treatment. For all animals, there was a strong inverse correlation between RV FS and serum cTnI (Fig. 3C).

Fig. 3.

A61603 treatment reversed right ventricular (RV) failure. A: liver weight referenced to body weight for sham-operated mice and 4 wk after pulmonary artery constriction (PAC) with either vehicle saline (Veh) or A61603 treatment for the final 2 wk. Liver weight was increased by PAC (PAC + Veh: n = 12 vs. sham: n = 17, ***P < 0.001 by ANOVA), and this increase was partly reversed by A61603 treatment (PAC + A61603: n = 16 vs. PAC + Veh: n = 12, *P < 0.05 by ANOVA). B: serum level of cardiac troponin I (cTnI) was increased after PAC (PAC + Veh: n = 10 vs. sham: n = 11, ****P < 0.0001 by ANOVA), and this increase was partly reversed by A61603 treatment (PAC + A61603: n = 10 vs. PAC + Veh: n = 10, **P < 0.01 by ANOVA). C: there was a highly significant inverse relationship between values of RV FS and serum cTnI measured in the same animals.

Chronic A61603 treatment reversed exercise dysfunction.

Measurements obtained at the end of the study period showed that voluntary exercise wheel running was reduced in the PAC plus vehicle treatment group, as evidenced by a lower distance run per day and less duration of running per day relative to the sham group (Fig. 4, A and B). Moreover, exercise capacity was improved by chronic treatment with A61603, as evidenced by a greater distance run per day and greater duration of running per day relative to the vehicle-treated group. Consistent with this, a paired analysis using measurements obtained before versus after chronic treatment with A61603 showed that the distance run per day 2 wk after PAC (4.5 ± 0.6 km/day, n = 17) was increased by subsequent chronic treatment with A61603 for a further 2 wk (to 6 ± 0.4 km/day, n = 17, P < 0.01 by a paired t-test). However, for the vehicle-treated PAC group, the distance run per day 2 wk after PAC (4 ± 0.6 km/day, n = 10) was not changed after treatment with saline for a further 2 wk (4 ± 0.6 km/day, n = 10, P = 0.9 by a paired t-test).

Fig. 4.

A61603 treatment rescued voluntary exercise capacity. A and B: distance run per day (A) and time running per day (B) were both decreased by pulmonary artery constriction (PAC) [PAC + vehicle (Veh): n = 10 vs. sham: n = 12, ****P < 0.0001 by ANOVA], and these decreases were partly reversed by A61603 treatment (PAC + A61603: n = 16 vs. PAC + Veh: n = 10, *P < 0.05 by ANOVA). C: there was a highly significant positive relationship between distance run per day and RV FS measured in the same animals.

For all animals, there was a significant positive correlation between the distance run per day and RV FS (Fig. 4C), stressing the functional significance of changes in RV FS as assessed by MRI.

Chronic A61603 treatment increased force development.

We measured force development per unit area of ex vivo myocardium using demembranated RV trabeculae. Figure 5A shows that relative to sham controls, 4 wk after PAC force development was substantially reduced at almost all levels of Ca²⁺ activation (P < 0.0001). However, for animals subjected to PAC, chronic A61603 treatment for 2 wk was associated with increased force development (P < 0.05 to 0.01).

Fig. 5.

A61603 treatment increased force development. A: force development in demembranated right ventricular trabeculae samples at various levels of activator Ca^{2 +} concentration ([Ca²⁺]; expressed as pCa (−log [Ca²⁺]). Force development was appreciably decreased by pulmonary artery constriction (PAC) [PAC + vehicle (Veh): n = 5 vs. sham: n = 4, #P < 0.0001 by two-way ANOVA], and this decrease was partly reversed by A61603 treatment (PAC + A61603: n = 6 vs. PAC + Veh: n = 5, *P < 0.05 and **P < 0.01 by two-way ANOVA). B: significant relationship between maximum Ca²⁺-activated force (F_max) and myofilament Ca²⁺ sensitivity (EC₅₀ Hill parameter). C: significant relationship between RV FS and F_max measured in the same animals.

For all trabeculae studied, the maximum force obtained at the highest level of activation (F_max) was significantly correlated with myofilament Ca²⁺ sensitivity (EC₅₀ parameter of the Hill equation; Fig. 5B). Thus, decreased F_max after PAC was associated with decreased Ca²⁺-sensitivity (increased EC₅₀), and these changes were partially reversed by A61603 treatment. The slope of the force-Ca²⁺ relationship (n_H) was not different among groups (not shown).

F_max was significantly correlated with RV FS measured in the same animals (Fig. 5C) suggesting that F_max was a determinant of RV FS in RVF induced by PAC and that recovery of F_max after chronic A61603 treatment contributed to the recovery of RV FS.

The passive force measured in relaxing solution was not different between sham versus PAC (3.5 ± 0.9 vs. 2 ± 0.5 mN/mm², P = 0.6) and was not changed by A61603 treatment (3.6 ± 1.5 mN/mm², P = 0.5).

Prosurvival, antiapoptosis, and antioxidant effects of chronic A61603 treatment.

Figure 6A shows that phosphorylation of the prosurvival signaling kinase ERK was significantly decreased in RV free wall myocardium after PAC; however, chronic treatment with A61603 normalized the ERK phosphorylation level.

Fig. 6.

A61603 increased ERK phosphorylation and abundance of Bcl-2 and glutathione peroxidase 1 (GPx1). Examples of Western blots and summary data (normalized to the sham group) are shown. A: phosphorylated ERK (p-ERK) relative to total ERK (T-ERK). p-ERK was decreased by pulmonary artery constriction (PAC) [PAC + vehicle (Veh): n = 7 vs. sham: n = 8, *P < 0.05 by ANOVA], and this decrease was reversed by A61603 treatment (PAC + A61603: n = 9 vs. PAC + Veh: n = 7, *P < 0.05 by ANOVA). B: Bcl-2 was increased by PAC (PAC + Veh: n = 8 vs. sham: n = 4, **P < 0.01 by ANOVA) and further increased by A61603 treatment (PAC + A61603: n = 8 vs. PAC + Veh: n = 4, **P < 0.01 by ANOVA). C: GPx1 was not significantly changed after PAC (PAC + Veh: n = 8 vs. sham: n = 4, ^nsP = 0.13 by ANOVA, where ns is not significant) but was increased by A61603 treatment (PAC + A61603: n = 8 vs. PAC + Veh: n = 4, **P < 0.01 by ANOVA).

Chronic A61603 treatment increased the level of the antiapoptosis factor Bcl-2 (Fig. 6B) and increased the level of the antioxidant GPx1 (Fig. 6C). Other antioxidant defenses (catalase, SOD1, and SOD2) were not increased by A61603 (data not shown).

Chronic A61603 treatment reduced oxidative stress and carbonylation of myofilaments.

Generation of ROS by RV free wall homogenates was assessed using the fluorescent probe DHR123 (Fig. 7A). DHR123 fluorescence was higher in the PAC group relative to the sham group. However, DHR123 fluorescence was normalized after chronic treatment with A61603.

Fig. 7.

A61603 reduced reactive oxygen species (ROS) formation and myofilament carbonylation. A: ROS formation was assessed by dihydrorhodamine 123 (DHR123) fluorescence in right ventricular (RV) homogenates (normalized to the sham group). ROS formation was increased by pulmonary artery constriction (PAC) [PAC + vehicle (Veh): n = 10 vs. sham: n = 10, **P < 0.01 by ANOVA], and this increase was reversed by A61603 treatment (PAC + A61603: n = 12 vs. PAC + Veh: n = 10, *P < 0.05 by ANOVA). B: examples of Western blots (noncontiguous) to assay carbonylated proteins in RV myofilaments. Multiple bands are visible, and individual contractile proteins were identified based on molecular weight. Actin and tropomyosin (Tm) were positively identified by reprobing the blots. The abundance of carbonylated proteins was increased in PAC, and this increase was reversed by A61603 treatment. C and D: summary data normalized to the sham group. The total level of protein carbonylation (sum of all bands per lane) and carbonylation level of actin were increased by PAC (PAC + Veh: n = 7 vs. sham: n = 7, **P < 0.01 by ANOVA), and these increases were reversed by A61603 treatment (PAC + A61603: n = 10 vs. PAC + Veh: n = 7, *P < 0.05 by ANOVA).

The insoluble fraction of RV homogenates, containing muscle myofilaments, was evaluated for increased protein carbonylation, a protein modification associated with oxidative stress (7). Western blot analysis to detect protein carbonylation (Fig. 7B) showed that the abundance of protein carbonyls was considerably increased in the PAC group relative to the sham control group. Multiple bands were evident with molecular weights consistent with myosin, desmin, actin, and tropomyosin. The identities of actin and Tm were confirmed by reprobing gels with antibodies for actin and tropomyosin (data not shown). For each animal, the total level of myofilament carbonylation was determined from the sum of the intensities measured for each band (Fig. 7C). Relative to the sham control group, the carbonylation level of myofilaments was significantly increased after PAC. However, after a 2-wk treatment with A61603, myofilament carbonylation was normalized. Similar results were obtained when each band was considered individually. For example, actin carbonylation (Fig. 7D) was elevated in PAC relative to sham and normalized by chronic A61603 treatment.

DISCUSSION

The major finding of this study is that RVF was substantially reversed by chronic treatment for 2 wk with the α_1A-AR agonist A61603. Using a clinically relevant disease recovery study design, we first induced RVF in mice by PAC. After RVF was established 2 wk after PAC, we then started chronic treatment with A61603. Two weeks after PAC, in vivo RV FS and measures of voluntary exercise capacity were substantially reduced compared with sham-operated control animals. However, subsequent treatment with A61603 for a further 2 wk resulted in substantial recovery of RV function and voluntary exercise capacity. Animals that were not treated with A61603 showed no recovery. As there are no therapies available for treating human patients with RVF (13), the significance of this study is the finding that the α_1A-AR might be a therapeutic target to treat RVF.

PAC model of RVF.

In this study, the presence of RVF was evidenced by decreased RV FS, decreased estimated RV cardiac output, liver congestion (increased liver weight), and indicators of cardiomyocyte necrosis (increased serum level of cTnI and RV fibrosis). Echocardiography revealed the presence of a significant pressure gradient across the constriction in the pulmonary artery. RV hypertrophy was evident by increased RV weight/body weight. Moreover, cardiac MRI assessment showed marked RV dilation and flattening of the septum.

Reversal of RVF by α_1A-AR therapy.

Two weeks after PAC, RV FS declined to ≈50% of the value for the sham group. Chronic treatment with A61603 for a further 2 wk resulted in appreciable recovery of RV FS to ≈80% of the value for the sham-operated control group. In contrast, for the vehicle-treated (i.e., untreated) control group, there was no recovery.

The recovery of RV FS after chronic treatment of PAC with A61603 was due to improved systolic contraction. Consistent with this, the decreased force development by cardiac trabeculae in the PAC group was significantly reversed by chronic treatment with A61603.

Recovery of RV FS with chronic A61603 treatment was associated with normalization of both estimated RV cardiac output and liver weight, consistent with reversal of RVF. Furthermore, exercise capacity was improved indicating functional improvement at the level of the whole organism.

Chronic A61603 treatment caused a reduction in serum levels of cTnI, suggesting that A61603 treatment reduced cardiomyocyte necrosis due to PAC.

Some effects of RV pressure overload were not reversed by A61603 treatment. Specifically, RV dilation, as evidenced by increased RV diastolic chamber area, and myocardial fibrosis were not reversed.

Mechanism of RVF recovery induced by A61603.

RV FS of sham-operated animals was not increased by chronic A61603 treatment, indicating that the low dose of A61603 used did not result in a direct inotropic effect. This also suggests that after PAC, the recovery of RV FS resulting from chronic A61603 treatment was not due to a direct inotropic action of A61603.

In mice subjected to PAC, chronic A61603 treatment increased phosphorylation of the signaling kinase ERK. A previous study has identified an α_1A-AR-ERK survival signaling pathway (9). ERK activation by α_1A-ARs leads to increased survival of cardiac myocytes challenged with norepinephrine, doxorubicin, or H₂O₂ (9). Recent studies have found that ERK activation plays a role in α_1A-AR-mediated protection against doxorubicin cardiotoxicity (2, 17).

Chronic A61603 treatment also increased expression of the antiapoptosis factor Bcl-2. Previous studies have reported that Bcl-2 prevents apoptosis in cardiac myocytes (12) and protects myocyte mitochondrial function after ischemia-reperfusion injury (4, 28).

The cardioprotective effects mediated by increased ERK activation and increased Bcl-2 abundance could contribute to the beneficial effects of chronic A61603 treatment.

After PAC, chronic A61603 treatment increased the abundance of the cellular antioxidant GPx1. Increased GPx1 could decrease cellular H₂O₂ and thereby limit formation of the toxic radical ^.OH⁻. Consistent with these effects, chronic A61603 treatment resulted in lower generation of reactive oxidants by RV homogenates as detected by the fluorescent probe DHR123. Oxidative stress results in multiple pathological effects in the heart (27), including apoptosis, and oxidant-mediated damage to multiple proteins involved in both excitation-contraction coupling (29), and myofilament contraction (26). Decreased oxidant generation with chronic A61603 treatment could have a beneficial effect on contraction by reducing oxidant-mediated damage to myocytes. Indeed, we found that the high level of oxidant-mediated carbonylation of myofilament proteins after PAC was normalized by chronic A61603 treatment. Previous studies have found that carbonylation of actin and tropomyosin was associated with impaired myofilament force (3). In agreement with this, we found increased myofilament carbonylation after PAC and decreased myocardial force. Moreover, we found that A61603 treatment normalized myofilament carbonylation and increased force development. Taken together, our findings suggest that after PAC, chronic treatment with A61603 resulted in lowering of oxidative stress and, consequently, less carbonylation of myofilament proteins and greater myofilament force development.

The presence of considerable myocardial fibrosis could complicate the interpretation of force measurements assessed per unit area of RV trabeculae because some area fraction of trabeculae could be occupied by replacement fibrosis. However, although myocardial fibrosis of the RV free wall was increased in the PAC group, cardiac trabeculae did not appear to show appreciable fibrosis and the passive force level measured was not increased. Thus, the decrease in force development of RV trabeculae after PAC is likely due to decreased myofilament function, rather than excessive fibrosis. After PAC, A61603 treatment was associated with increased force development, again without an effect on passive force or change in the level of fibrosis. This suggests that after PAC plus A61603 treatment, the increased force development measured in RV trabeculae was due to increased myofilament function.

Myofilament force might be an important determinant of RVF and therapeutic reversal of RVF by A61603. Our recent study using a multiscale computational model of the cardiovascular system suggested that decreased myofilament function plays a key role in the development of RVF (19). Furthermore, model simulations predicted that increased myofilament force after A61603 treatment would preserve RV function despite continued pressure overload (20).

α_1A-AR as a therapeutic target to treat heart failure.

In heart failure, α₁-AR signaling has been linked to powerful cardioprotective effects in experimental and clinical settings (for reviews, see Refs. 1, 11, and 18). For example, stimulation of α_1A-ARs protects isolated cardiac myocytes against necrosis induced by oxidative stress, norepinephrine, or doxorubicin (9). Chronic A61603 treatment protects against necrosis and improves function in mouse heart failure models based on doxorubicin toxicity (2, 17) or transaortic constriction (16). Moreover, we recently reported that chronic A61603 treatment protects against development of RVF due to pulmonary fibrosis (6). This previous study of RVF used a disease prevention design. The present study used a clinically relevant disease reversal design to show that chronic A61603 treatment could reverse established RVF.

Recently, we showed that in the failing human RV, the α_1A-AR remains functional and mediates a robust inotropic response (10). The presence of functional α_1A-ARs in failing human RV suggests that the α_1A-AR might be a therapeutic target to support cardioprotective effects in patients with heart failure (10).

Limitations.

In addition to the changes that we observed, RV failure and the response to therapy likely involve multiple changes in the abundance and posttranslational modification of regulatory and contractile proteins, along with changes in excitation-contraction coupling and energy metabolism. The effects of all such changes on the extent of RV failure and recovery remains to be determined.

Conclusions.

There are currently no effective therapies to treat patients with RVF. In a mouse model of RVF, we found that chronic treatment with A61603 reversed RVF resulting in recovery of RV function and voluntary exercise capacity. This finding suggests that the α_1A-AR is a potential therapeutic target to treat RVF.

GRANTS

This work was supported by Department of Veterans Affairs Merit Review Awards I01BX000740 (to A. J. Baker) and I01BX001970 (to P. C. Simpson) and Shared Equipment Grant 1IS1BX003101-01 (to A. J. Baker); National Heart, Lung, and Blood Institute Grant HL-31113 (to P. C. Simpson); the University of California-San Franciso Resource Allocation Program grant (to A. J. Baker); and American Heart Association Postdoctoral Fellowship 16POST30970031 (to P. M. Cowley) and Grant-in-Aid 15GRNT25550041 (to A. J. Baker).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.J.B conceived and designed research; P.M.C., G.W., P.M.S., A.R., N.P.R., P.D., and D.H.L. performed experiments; P.M.C., G.W., P.M.S., A.R., N.P.R., P.D., D.H.L., and A.J.B. analyzed data; P.M.C., P.M.S., D.H.L., P.C.S., and A.J.B. interpreted results of experiments; P.M.C., P.M.S., and A.J.B. prepared figures; P.M.C. and A.J.B. drafted manuscript; P.M.C., G.W., P.M.S., D.H.L., P.C.S., and A.J.B. edited and revised manuscript; P.M.C., G.W., P.M.S., A.R., N.P.R., P.D., D.H.L., P.C.S., and A.J.B. approved final version of manuscript.