Impaired Myofilament Contraction Drives Right Ventricular Failure Secondary to Pressure Overload: Model Simulations, Experimental Validation, and Treatment Predictions

Abstract

Introduction: Pulmonary hypertension (PH) causes pressure overload leading to right ventricular failure (RVF). Myocardial structure and myocyte mechanics are altered in RVF but the direct impact of these cellular level factors on organ level function remain unclear. A computational model of the cardiovascular system that integrates cellular function into whole organ function has recently been developed. This model is a useful tool for investigating how changes in myocyte structure and mechanics contribute to organ function. We use this model to determine how measured changes in myocyte and myocardial mechanics contribute to RVF at the organ level and predict the impact of myocyte-targeted therapy. Methods: A multiscale computational framework was tuned to model PH due to bleomycin exposure in mice. Pressure overload was modeled by increasing the pulmonary vascular resistance (PVR) and decreasing pulmonary artery compliance (CPA). Myocardial fibrosis and the impairment of myocyte maximum force generation (Fmax) were simulated by increasing the collagen content (↑PVR + ↓CPA + fibrosis) and decreasing Fmax (↑PVR + ↓CPA + fibrosis + ↓Fmax). A61603 (A6), a selective α_1A-subtype adrenergic receptor agonist, shown to improve Fmax was simulated to explore targeting myocyte generated Fmax in PH. Results: Increased afterload (RV systolic pressure and arterial elastance) in simulations matched experimental results for bleomycin exposure. Pressure overload alone (↑PVR + ↓CPA) caused decreased RV ejection fraction (EF) similar to experimental findings but preservation of cardiac output (CO). Myocardial fibrosis in the setting of pressure overload (↑PVR + ↓PAC + fibrosis) had minimal impact compared to pressure overload alone. Including impaired myocyte function (↑PVR + ↓PAC + fibrosis + ↓Fmax) reduced CO, similar to experiment, and impaired EF. Simulations predicted that A6 treatment preserves EF and CO despite maintained RV pressure overload. Conclusion: Multiscale computational modeling enabled prediction of the contribution of cellular level changes to whole organ function. Impaired Fmax is a key feature that directly contributes to RVF. Simulations further demonstrate the therapeutic benefit of targeting Fmax, which warrants additional study. Future work should incorporate growth and remodeling into the computational model to enable prediction of the multiscale drivers of the transition from dysfunction to failure.

Keywords: computational modeling; fibrosis; myocyte force generation; myoycte mechanics; pulmonary hypertension; right ventricle; right ventricular failure.

FIGURE 1

Schematic of multiscale computational model adapted from Tewari et al. (2016a,b). Myocyte mechanics model: F_pas: passive force, including contributions of intracellular titin and extracellular collagen, F_act: active force, incorporating actin-myosin cross-bridging kinetics and maximum myocyte calcium-activated force (Fmax), μ: viscous force, F_SE: corrective factor to account for sarcomere length. Biventricular heart and lumped parameter circulation models: RV: right ventricle, LV: left ventricle, C: compliance, PA: pulmonary artery, PV: pulmonary veins, SV: systemic veins, Ao: aorta, PVR: pulmonary vascular resistance, SVR: systemic vascular resistance.

FIGURE 2

Time varying atrial compliance (C_atria) for a physiologically realistic rodent heart rate of 420 bpm.

FIGURE 3

Measured and simulated increases in right ventricular afterload due to bleomycin exposure. (A,B) Predicted increases in RVSP and E_a in simulations of pressure overload alone (↑PVR + ↓CPA), with fibrosis (Fibrosis), and decreased myocyte maximum force generation (↓Fmax) match increases compared to control found in experimental measurements in mice exposed to bleomycin from (Hemnes et al., 2008). RVSP: right ventricular systolic pressure, PVR: pulmonary vascular resistance, CPA: pulmonary artery compliance.

FIGURE 4

Measured and simulated decreases in right ventricular function due to bleomycin exposure. (A) Predicted decreases in EF with pressure overload alone (↑PVR + ↓CPA), fibrosis (Fibrosis), and decreased myocyte maximum force generation (↓Fmax) match decreases compared to control found in experimental measurements in mice exposed to bleomycin from Hemnes et al. (2008). (B) Predicted decreases in cardiac output (CO) with pressure overload alone and with fibrosis are more moderate than the decrease in CO compared to control found experimentally; however, including impaired myocyte force generation predicts decreases in CO that better match experiments (Hemnes et al., 2008).

FIGURE 5

Simulated decreases in contractility, ventricular-vascular coupling, and diastolic function. (A) Predicted decrease in RV contractility end-systolic elastance (E_es) in the setting of pressure overload alone (↑PVR + ↓CPA) is moderate with limited additional impact of fibrosis (Fibrosis); the predicted decrease with decreased myocyte maximum force generation (↓Fmax) is substantial. (B) Predicted decrease in ventricular-vascular coupling (E_es/E_a) in the setting of pressure overload alone is dramatic; additional decreases with fibrosis and decreased myocyte maximum force generation are limited. (C) Predicted decrease in RV compliance in the setting of pressure overload alone is substantial without further impairments with fibrosis and reduced myocyte maximum force generation.

FIGURE 6

Predicted relationship between right ventricular function and maximum myocyte force generation (Fmax) in the context of normal afterload. (A) CO and (B) EF are predicted to be linearly dependent on Fmax for baseline pulmonary vascular resistance and pulmonary artery compliance values.

FIGURE 7

Measured prevention and simulated rescue of RVF by A61603 (A6). (A,B) Simulation of improved myocyte maximum force generation due to A6 rescue of RVF (RVF + A6) does not impact degree of pressure overload. Simulated A6 rescue of RVF also results in (C) increased ejection fraction and (D) improved CO that show the same trends as experimental measurements of A6 prevention of RVF (Cowley et al., 2017). (E,F) Simulations further predict A6 rescue of (E) RV contractility (E_es) and (F) ventricular-vascular coupling (E_es/E_a).