Mechanisms of Myotropes In Human Heart Failure

1) SPECIFIC AIMS: Mechanisms of Myotropes In Human Heart Failure Despite significant research efforts, therapeutic treatments for heart failure have been limited. Within the past decade, however, there has been a positive focus on developing “myotropes” to improve cardiac dysfunction; myotropes are small molecule therapeutics targeting cardiac myofilament proteins. These myotropes also represent exciting new “tools” to interrogate and better understand cardiac muscle function at the molecular and cellular levels. In healthy hearts, myofilaments become more sensitive to Ca2+ as the myocardium is stretched. This effect is known as length-dependent activation and is an important cellular-level component of the Frank–Starling mechanism. The Frank-Starling mechanism is impaired in patients who have heart failure. Most candidates for transplant are described as having ischemic heart failure (that is, heart failure after an infarction) or non-ischemic heart failure (everything else). Few studies have measured length-dependent activation in the myocardium from failing human hearts, and we recently showed that ischemic and non-ischemic heart failure results in different length-dependent activation responses at physiological temperature (37°C). Thus, it seems critically important to characterize, and potentially adjust, sarcomere-targeted therapeutics depending upon the etiology of heart failure. This project aims to advance understanding of mechanisms underpinning the Frank-Starling mechanism in human heart failure, and how they are affected by newly developed cardiac myotropes. We will focus on the effect(s) of each myotrope on length-dependent activation in permeabilized myocardial strips from donors and heart failure patients. Dr. Tanner (PI) has been collaborating with Dr. Kenneth S. Campbell (co-I) to make biophysical and biomechanical measurements using human myocardial samples from organ donors and transplant recipients for roughly 5 years. Throughout these 5 years Dr. Campbell has made significant progress on multi-scale computational models of muscle contraction, from the cross-bridges to the whole circulatory system. We hypothesize that these cardiac myotropes affect length-dependent contraction differently between ischemic and non-ischemic heart failure. To test this, we will make biophysical and biochemical experiments in Dr. Tanner’s lab, which will be used to drive XXX modeling from Dr. Campbell’s lab to extend molecular and cellular level data to organ and circulatory system function. (Fig 1). Aim 1: Measure the effect of XX and XX myotropes on contractile function in permeabilized human myocardium as muscle length changes. Biophysical and biochemical measurements will test how XX and XX (a recently developed pharmaceutical that binds myosin) modulates length-dependent activation in myocardial strips from organ donors and failing human hearts (ischemic vs. non-ischemic). Aim 2: Use computer modeling to predict how the length-dependent contraction ± XXX and YYY myotrope can impact hemodynamics in donors and heart-failure patients. In concert with novel measurements of length-dependent forces and cross-bridge kinetics (Aim 1), we will develop multiscale computational algorithms to illustrate how strategic manipulation of the cross-bridge activity impacts contractility, ventricular pressure-volume loops, and circulatory dynamics. 2) SIGNIFICANCE AND PRELIMINARY RESULTS 2.1 Clinical importance of heart failure and length-dependent activation Diseases caused by reduced or dysregulated contractile function are a major clinical problem. About half of the 6 million Americans who have heart failure exhibit depressed contractile function (1). Another 700,000 Americans have inherited genetic mutations that have been linked to myopathies ( 2, 3). Treatment options for most of these patients remain limited. For example, the clinical guidelines for heart failure ( 4) recommend standardized therapies (primarily β-blockers and ACE inhibitors), which were developed over 30 years ago and produce a 5-year survival rate of only 50% (1). Lack of cardiac reserve is a characteristic feature of heart failure that often reflects impairment of the Frank-Starling mechanism. Patients struggle to elevate their cardiac output on demand because the relationship between stroke volume and ventricular filling is compromised. This in turn links the clinical symptoms of heart failure to length-dependent activation, a cellular-level response that is inherent to the thick and thin filaments. Biophysicists define length-dependent activation as the increases in maximal force and Ca2+ sensitivity of contraction that are produced by myocardial stretch. Throughout this proposal, Ca2+ sensitivity will be quantified as pCa50, i.e. the pCa value (= -log10[Ca2+]) required to produce half- maximal force. Both functional effects are important, but myocytes are only partially activated during a normal heartbeat, so length-dependent changes Ca2+-sensitivity are likely to have greater clinical significance. In part, this is because myofilaments are highly cooperative, whereby even subtle changes in Ca2+-sensitivity can produce important changes in function. Length-dependent changes in Ca2+ sensitivity are an important component of the Frank-Starling mechanism, which is frequently impaired in Heart failure status, p = 0.472 Length, p < 0.001 patients who have heart failure. Thus, it is our Heart failure status*Length, p = 0.137 rationale that that targeting the mechanisms that regulate Ca2+ sensitivity may ultimately become a p < 0.001 p = 0.003 useful therapeutic pathway to treat heart failure. 2.2 Pilot data from human myocardium 5.7 We measured length-dependent activation in 98 pCa50 myocardial preparations from 12 heart failure patients who received cardiac transplants, and 60 5.2 preparations from 6 organ donors (Fig 2). Samples from the organ donors are the closest to µm µm µm “healthy controls” that can be attained with human µm µm µm myocardium, and skinned myocardial strips exhibit a robust length-dependent increase in Ca2+ 1.9 1.9 1.9 sensitivity. The results from failing myocardium 2.3 2.3 2.3 are more complex. Samples from patients who received a transplant subsequent to a myocardial Non- Ischemic Non-ischemic infarction (ischemic heart failure) showed a length- failing heart failure heart failure dependent increase in Ca2+ sensitivity, which was comparable to that measured in donor samples. Fig 2: Length-dependent increase in Ca2+- sensitivity of contraction (pCa50) is eliminated in non-ischemic heart failure. Symbols show individual preparations. Statistical tests are implemented using linear mixed models (described Section 3.1.3). Samples from patients with non-ischemic DYNAMICALLY COUPLED Actin Noff heart failure did not exhibit length-dependent FILAMENT SYSTEMS changes in Ca2+ sensitivity. Thus, length- JCa2+ off dependent activation may vary with the type of Thin filament Tropomyosin Ca2+ heart failure in patients. Ca2+ Jon These pilot data suggest that patients with Nunbound ischemic and non-ischemic heart failure may regulation have different cellular-level contractile Ca2+ phenotypes. One interpretation is that Thick filament Troponin myocardium from patients with ischemic regulation Nbound disease exhibits relatively normal length- dependent activation (i.e. these patients J1 J3(x) develop heart failure because their ventricle is compromised by a scar). In contrast, patients J2 J4(x) xps with non-ischemic heart failure have inherent x cellular-level contractile dysfunction. MOFF OFF Myosin MFG(x) 2.3 New appreciation for dynamic coupling between thick and thin filaments ON Myosin Force-generating Myosin Fig 3. Schematic showing dynamic coupling between thick and thin filament regulation. This dynamic coupling implies that any modification to thin filament function will in turn change the status of thick filament regulation, and vice versa. Modified from Campbell et al. ( Understanding of myofilament regulatory function has been transformed in the last few years via the discovery that myosin heads on thick filaments switch back and forth between OFF (also called super-relaxed, or interacting heads motif) and ON states (also called disordered relaxed) ( 5, 6). Heads in the OFF state are unable to bind actin (Fig 3), while those in the ON state can form cross-bridges by attaching to actin ( 7- 9). OFF-ON transitions are highly-dynamic; rates can exceed 100 s-1 ( 10- 12). In addition, these transitions are regulated by (at least): i) phosphorylation of regulatory light chain ( 13, 14), ii) interactions with myosin binding protein-C ( 15, 16), and iii) force ( 13, 17- 20). Consistently, our recent modeling showed that the OFF-ON equilibrium becomes biased towards the ON state as thick filament stress increases ( 20) (Fig. 4). This new thick-filament regulatory pathway is tremendously significant, as it unveils how force generation impacts the OFF-ON equilibrium couple the myofilaments dynamically. For example, a genetic mutation that enhances troponin C’s affinity for Ca2+ will increase the number of available binding sites on actin, and thus, increase the number of attached cross-bridges. This increased cross-bridge binding augments contractile force, which in turn pulls new myosins from OFF to ON, initiating a complex feedback loop. 2.4 Integrative and predictive contributions of computer modeling Dynamic coupling complicates the interpretation of contractile measurements. Prior to the discovery of force-dependent OFF-ON transitions, length-dependent changes in Ca2+ sensitivity were typically thought to reflect changes in thin filament function and/or the binding kinetics of individual myosin heads. Now, an additional possibility is that when cells are stretched, passive force pulls myosin heads from OFF to ON. This will increase the flux through J3 in Fig 3, thereby augmenting myosin binding without changing the attachment kinetics of individual heads. We developed a mathematical model of dynamically coupled filaments using MyoSim ( 21) and deployed our software to analyze published measurements of length-dependent activation in rat myocardium ( 13). Our calculations showed that a model with a force-dependent OFF-to-ON transition fitted the length-dependent change in Ca2+ sensitivity better (p<0.001) than a model without force- dependent OFF-to-ON recruitment (Fig 4). This modeling provides statistical justification for mechanical regulation of the thick filament OFF-ON equilibrium ( 20). Next, we used this model to screen for potential molecular mechanisms that could explain the loss of length-dependent Ca2+-sensitivity in myocardium from non-ischemic heart failure patients (Fig 2). The only perturbation that reproduced the data in our simulations was destabilizing the OFF state for non-ischemic heart failure. Computational simulations illustrate how length- dependent changes in Ca2+-sensitivity (ΔpCa50) decrease as the OFF-to-ON rate accelerates (Fig. 5). Altogether, these novel insights (Figs 2-5) suggest that dysfunctional regulation of the myosin OFF state compromises the Frank-Starling response in non-ischemic heart failure. 60 Stress (kN m-2) 0 6.2 5.7 pCa 5.2 4.7 1.00 1.00 Non Fraction of heads in 0.00 5.7 5.2 4.7 state MOFF 5.7 5.2 4.7 6.2 pCa pCa 0.00 6.2 0.20 5.7 5.2 4.7 0.60 5.7 5.2 4.7 pCa pCa Fraction Fraction of heads in of heads in state MON state MFG 0.00 0.00 6.2 6.2 Fig 4. Simulations of tension-pCa curves measured at short and long lengths. Experimental data from Kamporuakis et al. ( 13). Computational predictions illustrate the force- dependent myosin OFF-ON regulatory pathway as cardiac muscle is stretched. Modified from Campbell