Cardiac Function and Modulation of Sarcomeric Function by Length

Transcription

1 Cardiovascular Research Advance Access published December 12, 2007 Cardiac Function and Modulation of Sarcomeric Function by Length Laurin M. Hanft, F. Steven Korte, and Kerry S. McDonald Department of Medical Pharmacology & Physiology School of Medicine, University of Missouri, Columbia, MO, USA Running Title: Sarcomere length dependence of cardiac function Address correspondence to: Kerry S. McDonald, Ph.D. Associate Professor Department of Medical Pharmacology & Physiology MA 415 Medical Sciences Building School of Medicine University of Missouri Columbia, MO Phone: (573) Fax: (573) Time for primary review: 34 Published on behalf of the European Society of Cardiology. All rights reserved. The Author For permissions please 1

2 Abstract The Frank-Starling relationship provides beat-to-beat regulation of ventricular function by matching ventricular input and output. This review addresses the subcellular mechanisms by which the ventricle adjusts its output (i.e. stroke volume) by changes in end-diastolic volume. The subcellular processes are placed in the context of the four phases of the cardiac cycle with emphasis on the sarcomeric properties that mediate the number of force-generating crossbridges recruited during pressure development. Additional mechanistic insight is provided regarding the factors the regulate myocyte loaded shortening speeds, which are paramount for dictating ejection volume. Emphasis is placed on the interplay between cross-bridge-induced cooperative activation of the thin filament and cooperative deactivation of the thin filament induced by muscle shortening. The balance of these two properties seems to determine systolic hemodynamics, and how this balance is modulated by sarcomere length, in part, underlies the Frank-Starling relationship. Introduction Human cardiac output is remarkably fine-tuned by ventricular filling volume, whereby an increase in end diastolic volume leads to an appropriate increase in stroke volume 1. This control feature of the cardiac pump provides beat-to-beat tuning of systemic blood supply to the demands of the peripheral tissues and is important for equilibrating the output between right and left ventricles over time. Any malfunction of this control system will invariably lead to mismatches between supply and demand of peripheral energy, which could severely compromise organ system function and over time could result in fluid buildup such as pulmonary edema (which occurs with congestive heart failure). The relationship between ventricular filling and ventricular output was first described in the early 20 th century by Otto 2

3 Frank in frog ventricles 2 and Ernest Starling in mammalian heart-lung preparations 3, and has become known as the Frank-Starling relationship. The Frank-Starling relationship has been central in the development of ideas about the control of cardiac output over the last half century. The Frank-Starling relationship (i.e., ventricular function curves) has been experimentally determined to be modulated by a number of factors including autonomic innervation, humoral factors, exercise, coronary ischemia, anemia, shunts, toxic shock, and anesthesia (Figure 1). This has allowed for translation of the Frank-Starling relationship to the clinic as a reference indicator of the physiological and pathophysiological features of the ventricle. Importantly, the Frank-Starling relationship may become greatly depressed (in some cases nearly ten-fold) in late stage cardiac failure 4-6. So just how do the ventricles appropriately adjust their output to end-diastolic volume and what leads to a diminishment of this control system compared to a normal, healthy heart? This review will address this question by examining the left ventricular cardiac cycle and defining the sub-cellular process thought to mediate each phase. This is to provide helpful context for how filling and its consequent changes in myocyte length alter ventricular output and what processes may fall into disarray in failing myocardium. Myofibrillar regulation of the cardiac cycle The first phase of the cardiac cycle is ventricular filling whereby pressure in the left atrium exceeds pressure in the left ventricle and blood enters the ventricle through the atrioventricular valve. During ventricular filling the myoplasmic [Ca 2+ ] is low (~10-7 M) and force generating interactions are minimal between myosin on the thick filament and actin on the thin filaments. Force generating transitions are inhibited in the absence of Ca 2+ due to the relative position of tropomyosin, which tends to be trapped by troponin in a position that covers 3

4 the myosin binding domains on actin monomers 7. Since thin filaments are confined to the blocked state in the absence of Ca 2+, individual myocytes are considered relaxed and are passively strained during filling to an end diastolic length. Passive lengthening of myocytes over the working sarcomere length range is modulated primarily by titin with contributions from extracellular matrix proteins and the pericardium. Titin is a 3-4 MDa protein that spans the full half sarcomere from Z-line to M-line (for review of titin see 8 (Figure 2)). Titin molecules contain extensible and non-extensible regions, where the non-extensible regions provide a scaffold for associated proteins and serve as a filamentous anchor for attachment to the Z lines and M-lines within the sarcomere. The extensible regions span the I-band region of the sarcomere and play a prominent role in passive resistance over the working range of cardiac sarcomeres (i.e., ~1.8 μm to 2.3 μm 9-11 ) 12, 13. The extensible regions of titin contain a PEVK region (rich in proline, glutamate, valine and lysine), tandem immunoglobulin (Ig) repeats, and an adjustable N2B region 8, 14. These three regions form the molecular spring region whereby their straightening contributes to a passive force. Together these titin regions, in response to volume filling, help establish the sarcomere length of healthy cardiac myocytes just prior to electrical excitation of the ventricles (Figure 2). The second phase of the cardiac cycle commences after electrical excitation of the myocytes, whereby myoplasmic [Ca 2+ ] increases from ~10-7 M to approximately 10-6 to 10-5 M. The increase in myoplasmic Ca 2+ results in its binding to the low affinity Ca 2+ binding site on the amino terminus of cardiac troponin C. This binding triggers an allosteric signaling cascade that releases cardiac troponin I from actin and untraps tropomyosin such that it translates toward the groove between actin filaments, which exposes myosin binding sites on actin 15, 16. This defines the transition of the thin filament from the blocked state to the closed state 17, 4

5 18. In the closed state the position of tropomyosin is thought to allow for increased numbers of weakly bound cross-bridges to interact with actin as well as provision of a population of actin monomers that are made available for strongly bound (high stiffness), non-force generating cross-bridges 19. Strong cross-bridge binding initiates the thin filament transition from the closed state to the open state by translating tropomyosin further into the groove between actin filaments 15. It is only in the open state that strongly bound cross-bridges are thought to make the transition from strongly bound non-force generating to strongly bound force-generating. Thus, in this model, full activation of thin filament regions only occurs in the presence of both Ca 2+ and strongly bound cross-bridges, wherein Ca 2+ first allows cross-bridges to bind and these in turn promote stereospecific binding. Importantly, strongly bound crossbridges likely influence additional cross-bridge binding by cooperative spread of the tropomyosin positional shift to neighboring tropomyosins and the propensity of strongly bound cross-bridges to maintain tropomyosin position in the open state. Structural data implies that strongly bound cross-bridges influence tropomyosin molecules a minimum of three regulatory units (where 1 regulatory unity spans ~38 nm and is composed of 7 actins, 1 troponin complex, and 1 tropomyosin) for a total distance of ~115 nm centered from the individual strongly bound cross-bridge 15, 20. These structural studies are consistent with skinned muscle fiber experiments that also suggested the cooperative spread of activation by a strongly bound crossbridge to be ~3 regulatory units (i.e., ~115 nm) 21. Interestingly, a calculation of the number of cross-bridges that interact with thin filaments during activation yields ~1 cross-bridge bound per regulatory unit (for detailed calculation see Gordon et al. 22 ). This calculation assumed 300 myosins per thick filament, 0.75 μm of each thin filament is in the overlap region, and 30% of all cross-bridges interact during a maximal isometric activation 22, 23. Interestingly, since the fraction of interacting cross-bridges would fall during submaximal Ca 2+ activation and further 5

6 during sarcomere shortening as cross-bridge detachment is accelerated 24 ; this would again yield approximately one strongly bound cross-bridge per three tropomyosins (i.e., a total of ~115 nm along the thin filament). While this calculation is applicable for both cardiac and skeletal mammalian myofibrils, the single fiber experimental analysis was performed using fast-twitch skeletal muscle fibers; moreover, the structural analysis was performed with skeletal muscle proteins in the presence of rigor cross-bridges. Similar structural studies have not been performed with cardiac muscle proteins or in the context of cycling cross-bridges. However, experiments investigating the cooperative spread of thin filament activation have been examined using skinned cardiac muscle preparations. It appears that Ca 2+ binding to cardiac troponin C is insufficient to completely activate one regulatory unit 25. However, strong binding cross-bridges appear to yield a greater degree of cooperative activation in cardiac muscle compared to skeletal muscle 26, 27. Along these lines, cooperative activation of steadystate force has been found to be greater in cardiac and fast-twitch skeletal muscle compared to slow-twitch skeletal muscle 28, 29. Nonetheless, the exact distance of lateral spread of thin filament activation induced by a cycling cross-bridge in cardiac muscle remains to be elucidated. Certainly, both Ca 2+ and cross-bridge induced cooperative activation of the thin filament are important determinants of the number of force-generating cross-bridges formed in response to the Ca 2+ -bound troponin C signal. Another important determinant of the number of force generating cross-bridges appears to be the lateral spacing between thick and thin filaments, which dictates the proximity of cross bridges to actin monomers (i.e. it determines the effective concentration of cross-bridges likely to undergo force generating transitions). Intact cardiac myocytes exhibit constant volume behavior, thus, increased myocyte length will reduce interfilament lattice spacing 30-33, which will increase the effective concentration of myosin cross-bridges and yield increased Ca 2+ sensitivity of myofibrillar force at least over an 6

7 optimal range of spacing (for review see Fuchs & Martyn 34 ). Ca 2+ sensitivity of force is thought to be paramount in determining the number of force generating cross-bridges during a heartbeat as decreases in length clearly decrease twitch force without alterations in the intracellular Ca 2+ transient in intact mammalian ventricular preparations 35. The changes in lattice spacing with length are thought to be mediated in part by titin, whose connections to the thin and thick filaments yields a radial force that physically pulls the filaments together and overcomes the electrostatic repulsion between the filaments A third factor that determines the number of force generating cross-bridges during phase two is the intrinsic rate of crossbridge transition from the non-force generating to force generating state, which is mainly dictated by the inherent properties of myosin heavy chain (MyHC) isoform. There are two myosin heavy chain isoforms in mammalian myocardium α-myhc and β-myhc and α-myhc has been shown to display two-to-three fold faster force generating transitions than β-myhc Thus, during phase two of the cardiac cycle, the rate and number of force generating crossbridges are determined by (i) the rate and extent of cross-bridge induced cooperative activation, (ii) lattice spacing, and (iii) the inherent rate of force generating transitions. The rate of force generation is extremely important in determining the systolic time in phase two, which yields a greater or lesser fraction of systolic time for ejection. The number of force generating cross- bridges is crucial for determining where on the force-velocity relationship the ensemble of cross-bridges will work (Figure 3), whereby an increase in the number of force generating cross-bridges will make a given afterload less of a relative load, which will increase the rate and extent of shortening during the third phase of the cardiac cycle, i.e., ejection. The ejection phase of the cardiac cycle begins once left ventricular pressure rises sufficiently to open the aortic valve. The volume of blood ejected is largely determined by the 7

8 extent of myocyte shortening, which depends on two main factors, (i) the number of force generating cross-bridges and (ii) the intrinsic rate of cross-bridge cycling. As mentioned above, the number of force generating cross-bridges is determined by the myoplasmic [Ca 2+ ] and the cooperative activation of the thin filament by strong binding cross-bridges, which is mediated, in part, by the spacing between myofilaments. The number of strongly bound cross-bridges will dictate where on the force-velocity curve the ensemble of cross-bridges will perform. More force-generating cross-bridges yield less force per cross-bridge allowing faster cycling per cross-bridge that results in more shortening and, ultimately, more stroke volume. It is important to understand that thin filament activation levels are highly dynamic and change as a function of time and in response to sarcomere shortening. Once myocyte shortening commences, in fact, there appears to be a stimulus to deactivate the thin filament. This results from the actuality that as the thin filament slides toward the middle of the sarcomere there is less strain per cross-bridge and, thus, less force per cross-bridge 24. In addition, reduced strain per cross-bridge results in faster cross-bridge detachment so there are fewer strong binding cross-bridges to maintain the open state via cross-bridge induced cooperative activation of the thin filament (Figure 4). Furthermore, sarcomere shortening increases interfilament spacing, providing an additional stimulus to reduce the number of strongly bound crossbridges by decreasing the probability of a detached cross-bridge from re-attaching to the thin filament as it slides past the thick filament. Overall, these processes would act to progressively decrease the number of force-generating cross-bridges, resulting in shortening-induced cooperative deactivation of the thin filament. Shortening-induced cooperative deactivation has been observed consistently during shortening of isolated striated muscle preparations This mechanism may, in fact, provide an important myofibrillar brake system, which could be important in preventing excessive pressure development in the face of diminishing chamber 8

9 size during ejection. In addition, shortening-induced cooperative deactivation is likely a major factor for termination of phase three and, thus, control of ventricular relaxation. When pressure in the aorta exceeds pressure in the left ventricle the semilunar valve will close and phase four (isovolumic relaxation) begins. The factors that determine myocyte relaxation are the underpinnings for the onset and duration of phase four. Certainly, the removal of sarcoplasmic Ca 2+ is an important determinant of myofibrillar relaxation, however, its role may be more permissive, at least in healthy myocytes. This idea is derived from the findings that the fall in thin filament bound Ca 2+ seems to precede the fall in force in a cardiac muscle isometric twitch and ventricular models suggest that the transient of Ca 2+ bound to thin filaments is virtually complete while the ventricle is still in the ejection phase 55, 56. In addition, recent studies have demonstrated that myofibrillar relaxation rates are considerably slower than rates of Ca 2+ dissociation from cardiac troponin C in human cardiac myofibrils 57, 58 ). These findings implicate that myofibrillar processes elicited by shortening-induced cooperative deactivation of the thin filament are most important in governing ventricular relaxation in the healthy heart. It is, however, not out of the question that some Ca 2+ remains bound to cardiac troponin C for the duration of ejection and, thus, contributes to the rate of relaxation. For instance, in intact myocardium a relatively small amount of Ca 2+ is still being removed from the sarcoplasm even during the latter phases of relaxation 54, which would tend to prolong relaxation. Effect of length on sarcomeric function during the cardiac cycle Within the context of the sub-cellular mechanisms that underlie the four phases of the cardiac cycle, it becomes more apparent how changes in filling volume alter ventricular output. 9

10 An increase in filling volume (i.e., preload) during phase one will stretch the myocytes causing an increase in sarcomere length within the myofibrils (Figure 2). As sarcomere length is increased from ~1.8 μm to 2.3 μm there is an increase in force generating capacity of striated muscle due, in part, to more optimal overlap of thick and thin filaments. This is referred to as the ascending limb of the length-tension relationship, first defined by Ramsey and Street in and further defined in 1966 in the classic study by Gordon, Huxley, & Julian 60 using frog fast-twitch skeletal muscle fibers. It is important to point out that cardiac myocytes exhibit a similar (maybe even shallower) ascending limb of the length-tension relationship when maximal Ca 2+ activations in permeabilized cardiac myocytes are compared to tetanic forces in fast-twitch skeletal muscle fibers The ascending limb of the length-tension relationship, however, becomes much steeper for twitch contractions (in both skeletal muscle 64 and cardiac muscle 63 ) and twitch contractions better simulate the conditions during a cardiac cycle. The sub-cellular basis for the steep length-twitch tension relationship over the ascending limb is the increased myofibrillar Ca 2+ sensitivity of force with increased sarcomere length 35, 63, 65. The sub-cellular basis for the changes in Ca 2+ sensitivity of force with length has been an area of considerable research over the last 25 years (for review see 34 ). A plausible biophysical mechanism for greater Ca 2+ sensitivity at longer lengths is that stretching of myocytes yields longitudinal and radial forces that reduce the interfilament lattice spacing to increase the effective concentration of myosin cross-bridges available for binding to Ca 2+ activated thin filaments 36, 37, (Figure 2). Here the sarcomeres sense an increase in length by physical changes in interfilament lattice spacing induced, in part, by titin strain. However, while there is consistent evidence that Ca 2+ sensitivity of force increases with compression of the interfilament lattice 33, 37, 67, 68, 71 direct examination of interfilament lattice spacing by X-ray diffraction has not always correlated well with length dependent changes in force production 70, 10

11 72. This may arise from interfilament spacings in experimental myocardial preparations that deviate from the interfilament lattice spacings that are optimal for thin filament responsiveness as suggested by Martyn and Fuchs 34. Alternatively, it may indicate a separate, non-mutually exclusive mechanism by which the sarcomeres sense length changes. One possibility is that the strain on titin may more directly correlate with the position of cross-bridges so as to influence probability of binding 33. For instance, if increased length yields more cross-bridges in proximity that favors weak binding this would tend to shift both the thin filament and crossbridge state equilibriums toward positions that favor more strongly bound cross-bridges, which, in turn, would further activate the thin filaments. This is consistent with the X-ray diffraction measurements that showed temperature reduction increased proximity of myosin cross-bridges toward the thin filament (as indicated by the increased I 1,1 /I 1,0 intensity ratio), which resulted in greater Ca 2+ sensitivity of force and reduced length dependence of Ca 2+ sensitivity of force 73, 74. The length change may also be sensed by thin filament proteins such as cardiac troponin I since transgenic replacement of cardiac troponin I with slow skeletal troponin I resulted in reduced length dependence of Ca 2+ sensitivity of force 72. Additionally, Ca 2+ binding to cardiac troponin C appears to be increased at longer sarcomere lengths 71, 75, 76, which would contribute to increased levels of thin filament activation and force-generating cross-bridges 77. Increased Ca 2+ binding with increased length appears to be most responsive to the presence of more strongly-bound force generating cross-bridges rather than changes in sarcomere length per se since the effect was eliminated when force was prevented by vanadate 75, 78. In fact, Ca 2+ induced force generation has been shown to alter the linear dichroism of cardiac troponin C, labeled near its N-terminus by rhodamine, which suggests cycling cross-bridges alter the N- terminal structure of cardiac troponin C perhaps changing its Ca 2+ binding affinity. In summary, increased sarcomere length (in response to greater filling volume) results in greater 11

12 Ca 2+ sensitivity of myofibrillar force, which likely is induced by altered interfilament lattice spacing and/or other length-induced allosteric changes along the myofilaments that increase the probability of thin filament activation. As mentioned above, the duration of phase two is another factor besides recruitment of force generating cross-bridges that will have an impact on stroke volume. Increased sarcomere length does not appear to speed the rate of force development in cardiac muscle at least at a given activator [Ca 2+ ]. In fact, when Ca 2+ activation is matched to yield the same relative force at long and short sarcomere length, force development rates are actually faster at short sarcomere length in both α-myhc and β-myhc cardiac preparations 42, 79, 80. This appears to differ from skeletal muscle fiber preparations where rates of force development seem to be slower at short sarcomere length even when Ca 2+ activated force levels are matched 28. The sub-cellular mechanism for faster rates of force development at short sarcomere length in cardiac myocytes is unknown but physiologically may serve as a compensatory mechanism to prevent too much of a fall in output due to the relatively large drop in number of force generating cross-bridges that occurs at short sarcomere length. Once phase 3 (i.e., ejection) begins the amount of blood expelled will be determined by the rate of loaded shortening. Sarcomere length has a major influence on this rate by first determining the number of force generating cross-bridges, which dictates the relative load that the sarcomeres will work against (i.e., afterload). As mentioned above, with an increase in preload each given afterload becomes a lower relative load due to greater isometric force producing capabilities (i.e. the force-velocity curves shift upward) (Figure 3). Thus, loaded shortening velocity is increased as there are more cross-bridges to work against a load, yielding less load per cross-bridge and therefore faster loaded cross-bridge cycling. This is observed 12

13 consistently in isolated muscle preparations regardless of whether the measurements are made during afterloaded contractions in intact cardiac muscle preparations 81, during imposition of load clamps in intact cardiac muscle preparations 48, or during load clamps in permeabilized single myocyte preparations 42. It also appears that maximum velocity of shortening is sarcomere length dependent in cardiac muscle preparations, at least during submaximal Ca 2+ activations. For comparative purposes, in mammalian fast-twitch skeletal muscle the maximum velocity of shortening was found to be independent of sarcomere length at least over the sarcomere length range of 1.8 μm to 2.7 μm based upon slack test analysis 82. At sarcomere lengths less than 1.8 µm maximum velocity of shortening decreased linearly with further decreases in sarcomere length, supposedly due to increased internal loads associated with thin filament overlapping with cross-bridges on the opposite side of the sarcomere and thick filaments abutting against Z discs 82. As an aside, it is interesting that fiber shortening speed still remains ~50% of maximum at sarcomere lengths of ~1.5 μm 60, 82, implicating considerable thin filament sliding capacity even at very short sarcomere length (perhaps allowed for by thick filament buckling). The maximum velocity of shortening shows a similar sarcomere length dependence in intact rat cardiac muscle preparations during near maximal Ca 2+ activations 83. However, during submaximal Ca 2+ activations maximum shortening velocity was highly dependent on sarcomere length over the SL range of μm as measured directly by calibrated laser light diffraction 83. A similar finding was observed in ferret papillary muscle preparations using an electronic feedback system to track muscle segment length 43. The exact mechanism for length dependence of unloaded shortening remains unclear but may arise from a fixed internal load that cycling cross-bridges must work against. The source of the internal load has been suggested to arise from titin molecules undergoing compression and buckling 83 and/or extracellular components such as collagen. One 13

14 observation that persists in these experiments is the increased curvature of most length traces (whether from sarcomere length, segment length, or muscle length recordings) during lightly loaded contractions 43, 48, 83. This is also routinely observed in permeablized cardiac and skeletal muscle cell preparations regardless of whether preparation length or sarcomere length are monitored 44-47, 50, 84. This result implicates either a passive internal load that resides in the sarcomeres and/or loss of cross-bridges during shortening due to shortening-induced cooperative deactivation. Evidence for the latter comes from the experiments in skeletal muscle fibers whereby reduction of strong binding cross-bridges by a repetitive isotonic shortening protocol eliminated the initial fast component of shortening 46, and the finding that fiber stiffness (which is thought to reflect the number of strong binding cross-bridges) fell coincident with the changes in curvature of the fiber length trace 47. Similar experiments have not been performed in cardiac myocyte preparations. Regarding shortening at greater relative loads (i.e., at ~30% of isometric force) where power output is optimal and striated muscle is most efficient and thought to work in vivo 85, 86, there is considerable sarcomere length dependence of loaded shortening at a given absolute load and this remains even when force velocity curves are normalized for differences in isometric force, i.e., loaded shortening velocity was slower at short sarcomere length 42. A possible mechanism underlying the decreased loaded shortening and power output at short sarcomere length could be the associated decrease in thin filament activation levels perhaps due to increased interfilament lattice spacing, which reduces the probability of cross-bridge binding. In fact, when Ca 2+ activated force and presumably thin filament activation levels were matched at short sarcomere length to those at long SL (by increasing the activator [Ca 2+ ]) short sarcomere length actually yielded faster loaded shortening velocities and greater peak normalized power output, at least in adult rat myocyte preparations that contained mostly α-myhc 42. This suggests a myofibrillar mechanism that tends to speed 14

15 loaded cross-bridge cycling to minimize the fall of power at short sarcomere length. Interestingly, the presence of 2% dextran also resulted in faster loaded shortening than at long sarcomere length, again implicating a myofibrillar mechanism that leads to faster loaded crossbridge cycling at short sarcomere length 42. A potential mechanism for the speeding of loaded shortening at short sarcomere length (when force is matched) is that as sarcomere length is shortened, titin becomes less taut. This may reduce the impedance on the cross-bridge (which may be mediated by titin s interaction with myosin binding protein C (MyBP-C) on the thick filament (Figure 2), which in turn increases myosin head radial and azimuthal mobility 42. Ultimately, this may lead to faster cross-bridge cycling directly or indirectly by creating more flexible cross-bridges that are more likely to maintain thin filament activation. However, this mechanism alone cannot overcome the decrease in the number of cross-bridges induced by increased lattice spacing at short sarcomere length, which is why loaded shortening is slower at short sarcomere length when activator [Ca 2+ ] is the same between long and short SL. These mechanistic ideas are consistent with findings in cardiac myocyte preparations from MyBP-C ablated mice, where loaded shortening was elevated for any given load but especially at high loads 87. This response (i.e., faster loaded shortening at short sarcomere length), however, may be MyHC dependent as loaded shortening remained slower even when Ca 2+ activated force levels were matched in β-myhc myocyte preparations 42. Interestingly though, compression of the lattice caused faster loaded shortening in β-myhc myocytes (similar to that observed in α- MyHC myocytes) 42. This effect implicates lattice spacing as a key determinant in transducing the sarcomere length response of loaded shortening in cardiac myocytes. The mechanisms underlying the MyHC dependence of sarcomere length on loaded shortening is unclear but may be due to less length-responsive flexibility of the myosin cross-bridges in response to length changes in β-myhc myocytes but more responsiveness to changes in lattice spacing. These 15

16 ideas will require careful examination by X-ray diffraction studies. The greater sarcomere length dependence of loaded shortening in β-myhc cardiac myocytes implicates a steeper Frank-Starling relationship in left ventricles that contain β-myhc, an observation that has been borne out experimentally 40, 88. The sarcomere length dependence of loaded shortening in intact cardiac muscle preparations, and the extent of cross-bridge induced cooperative deactivation that occurs in intact preparations remain to be determined. In addition, the effect of sarcomere length on rates of relaxation remains an understudied area. Changes in length dependent sarcomeric function with heart failure Heart failure by definition is a downward and rightward shift of the Frank-Starling relationship (i.e., depressed ventricular function curves) compared to the normal, healthy heart 89 (Figure 1). There are a number of changes to phase one of the cardiac cycle that occur in heart failure. A relatively common occurrence, especially in hypertensive heart disease, is an increase in fibrosis of cardiac tissue 90. This will increase the stiffness of the myocardial wall, which means higher diastolic filling pressures are required to fill the ventricle. Increased stiffness also was seen in the spontaneous hypertensive rat (SHR) model, which was attributed to a decrease in the N2BA (less stiff) isoforms of titin 91. Similarly, increases in passive stiffness were seen in human papillary muscle from failing hearts 92, in trabeculae from human patients exhibiting dilated cardiomyopathy 93, and in diastolic heart failure patients where the increased stiffness was attributed to decreased phosphorylation of titin 94. If increased stiffness compromises ventricular filling then end-diastolic sarcomere length may be shorter causing the heart to work lower on the ventricular function curve. In contrast, there is evidence for reduced stiffness (i.e., increased compliance) in human patients with ischemia induced left sided heart failure, which was attributed to an increase in the N2BA (more compliant) isoform of titin 95. A 16

17 more compliant isoform of titin in face of greater filling pressure (that arise with pulmonary congestion) may result in a markedly diminished left ventricular reserve capacity (i.e., inability for sarcomere length to be further elongated or interfilament spacings that are compacted below optimal such that cross-bridge function is impaired 34 ). Regardless of whether ventricular stiffness rises or falls in disease states, any independent modulation of stiffness between the right and left ventricles could undermine a key mission of the Frank-Starling mechanism, which is to match right and left ventricular output, whereby a mismatch may predispose the system to edema and congestion. In the context of phase two of the cardiac cycle, myofilaments from failing human hearts have been reported to exhibit an increased Ca 2+ sensitivity of force 4, 96, 97, which was linked to reduced levels of PKA-induced phosphorylation of myofilament proteins 96. This is consistent with down-regulated β-adrenergic responsiveness associated with failing hearts 98. Increased Ca 2+ sensitivity of force, in and of itself, may yield reduced length dependent response of isometric force development as has often been observed 69, 80, 99. This was directly observed in skinned human papillary muscle strips from heart failure patients, which exhibited decreased length dependent Ca 2+ sensitivity of force 4, 100. This, however, is not a consistent finding as intact human myocardium, stretched from 90% to 100% of muscle length (that yields maximum isometric force), had similar force responses between failing and non-failing myocardium 5. Interestingly, though, in this same study, the left ventricular end diastolic volume-end systolic pressure relationship was depressed considerably in dilated cardiomyopathic hearts compared to donor hearts 5. This may implicate changes to elements in parallel with the sarcomeres or changes in the kinetics of cross-bridge interactions. For instance, PKA induced phosphorylation of myofilaments has been reported to increase rates of 17

18 force development 101 and stretch activation 102, which would speed pressure development and thus phase two in normal hearts compared to diseased hearts that have reduced PKA phosphorylation of myofibrils. In addition, heart failure has been correlated with increased PKC activity 103 and PKC induced phosphorylation has been shown to depress twitch force and kinetics of force development 104, 105. The effect of heart failure on myocyte loaded shortening remains unclear. However, here again, PKA-induced phosphorylation has been found to speed loaded shortening in cardiac myocytes 106, while PKC-induced phosphorylation reduced loaded shortening and attenuated the PKA-induced augmentation of power output 107. In addition human heart failure is associated with a small shift in MyHC content (~85% β-myhc in donor hearts to 100% β-myhc in failing hearts) 108 and this shift in MyHC is known to result in significantly slower rates of force development 39, 41, and loaded shortening, and decreased power output 109, 110. The effects of heart failure on myofibrillar relaxation rates have not been directly studied but there are a number of studies indicating that Ca 2+ handling kinetics are slowed with heart failure 111 and a prolonged Ca 2+ transient has been shown to circumvent the sarcomeric role in determining relaxation 52 and inevitably yields defects in relaxation. In summary, the sub-cellular processes that underlie the cardiac cycle involve, to large extent, myofibrillar responses to changes in sarcoplasmic [Ca 2+ ]. The beat-to-beat regulation of ejection is largely determined by the number of strongly bound cross-bridges that interact during systole, which is determined by cooperative activation of the thin filament during phase two (isovolumic contraction) of the cardiac cycle. Importantly, the degree of cooperative activation is greatly pre-determined by the degree of sarcomeric stretch attained during phase one (ventricular filling). During ejection, the number of strongly bound cross-bridges dictates the degree of shortening but is controlled by the balance of cooperative activation (by strong 18

19 binding cross-bridges) and cooperative deactivation induced by shortening (which lowers the number of strongly bound cross-bridges). It remains to be clarified just how sarcomere length directly modulates shortening under physiological loads and how it affects the extent of cooperative deactivation. These questions are at the heart of the matter by which chronic adaptations (such as exercise and heart failure) lead to hyper- or hypo-effective Frank-Starling relationships. Funding National Institutes of Health (AR48523 to L.M.H., HL & HL71550 to K.S.M). Conflict of Interest None declared. 19

20 References [1] Lakatta EG. Length modulation of muscle performance: Frank-Starling Law of the Heart. In: Fozzard HA, ed. The Heart and Cardiovascular System. New York: Raven Press, Ltd, 1992: [2] Frank O. Zur dynamik des herzmuskels Z Biol 1895;32: [3] Patterson S, Starling EH. On the mechanical factors which determine the output of the ventricles. J Physiol 1914;48: [4] Schwinger RH, Bohm M, Koch A, Schmidt U, Morano I, Eissner HJ, et al. The failing human heart is unable to use the Frank-Starling mechanism. Circ Res 1994;74: [5] Holubarsch C, Ruf T, Goldstein DJ, Ashton RC, Nickl W, Pieske B, et al. Existence of the Frank-Starling mechanism in the failing human heart: investigations on the organ, tissue, and sarcomere levels. Circ 1996;94: [6] Jacob R, Dierberger B, Kissling G. Functional significance of the Frank-Starling mechanism under physiological and pathophysiological conditions. Eur Heart J :7-14. [7] Pirani A, Vinogradova MV, Curmi PMG, King WA, Fletterick RJ, Craig R, et al. An Atomic Model of the Thin Filament in the Relaxed and Ca2+-Activated States. J Mol Biol 2006;357: [8] Granzier HL, Labeit S. The giant protein titin: a major player in myocardial mechanics, signaling, and disease. Circ Res 2004;94: [9] Pollack GH, Huntsman LL. Sarcomere length-active force relations in living mammalian cardiac muscle. Am J Physiol 1974;227: [10] Grimm AF, Lin H-L, Grimm BR. Left ventricular free wall and intraventricular pressure-sarcomere length distributions. Am J Physiol 1980;239:H101-H

21 [11] Guccione JM, O'Dell WG, McCulloch AD, Hunter WC. Anterior and posterior left ventricular sarcomere lengths behave similarly during ejection. Am J Physiol 1997;272:H469- H477. [12] Granzier HL, Irving TC. Passive tension in cardiac muscle: Contribution of collagen, titin, microtubules, and intermediate filaments. Biophys J 1995;68: [13] Wu Y, Cazorla O, Labeit D, Labeit S, Granzier H. Changes in titin and collagen underlie diastolic stiffness diversity of cardiac muscle. J Mol Cell Cardiol 2000;32: [14] Helmes M, Trombitas K, Centner T, Kellermayer M, Labeit S, Linke WA, et al. Mechanically Driven Contour-Length Adjustment in Rat Cardiac Titin's Unique N2B Sequence : Titin Is an Adjustable Spring. Circ Res 1999;84: [15] Vibert P, Craig R, Lehman W. Steric-model for activation of muscle thin filaments. J Mol Biol 1997;266:8-14. [16] Xu C, Craig R, Tobacman L, Horowitz R, Lehman W. Tropomyosin positions in regulated thin filaments revealed by cryoelectron microscopy. Biophys J 1999;77: [17] McKillop DF, Geeves MA. Regulation of the interaction between actin and myosin subfragment 1: evidence for three states of the thin filament. Biophys J 1993;65: [18] Solaro RJ, Van Eyk J. Altered interactions among thin filament proteins modulate cardiac function. J Mol Cell Cardiol 1996;28: [19] Wahr PA, Metzger JM. Role of Ca 2+ and cross-bridges in skeletal muscle thin filament activation probed by Ca 2+ sensitizers. Biophys J 1999;76: [20] Phillips GN, Fillers JP, Cohen C. Tropomyosin crystal structure and muscle regulation. J Mol Biol 1986;192:

22 [21] Moss RL, Allen JD, Greaser ML. Effects of partial extraction of troponin complex upon the tension-pca relation in rabbit skeletal muscle. Further evidence that tension development involves cooperative effects within the thin filament. J Gen Physiol 1986;87: [22] Gordon AM, Homsher E, Regnier M. Regulation of contraction in striated muscle. Physiol Rev 2000;80: [23] de Tombe PP, Stienen GJM. Impact of temperature on cross-bridge cycling kinetics in rat myocardium. J Physiol 2007;584: [24] Ford LE, Huxley AF, Simmons RM. Tension transients during steady shortening of frog muscle fibres. J Physiol 1985;361: [25] Gillis TE, Martyn DA, Rivera AJ, Regnier M. Investigation of thin filament nearneighbor regulatory unit interactions during force development in skinned cardiac and skeletal muscle. J Physiol 2007;580: [26] Fitzsimons DP, Patel JR, Moss RL. Cross-bridge interaction kinetics in rat myocardium are accelerated by strong binding of myosin to the thin filament. J Physiol 2001;530: [27] Metzger JM. Myosin binding-induced cooperative activation of the thin filament in cardiac myocytes and skeletal muscle fibers. Biophys J 1995;68: [28] McDonald KS, Wolff MR, Moss RL. Sarcomere length dependence of the rate of tension redevelopment and submaximal tension in rat and rabbit skinned skeletal muscle fibers. J Physiol 1997;501: [29] Konhilas JP, Irving TC, de Tombe PP. Length-dependent activation in three striated muscle types of the rat. J Physiol 2002;544: [30] Matsubara I, Millman BM. X-ray diffraction patterns from mammalian heart muscle. J Mol Biol 1974;82: [31] Millman BM. The filament lattice of striated muscle. Physiol Rev 1998;78:

23 [32] Yagi N, Shimizu J, Mohri S, Araki Ji, Nakamura K, Okuyama H, et al. X-ray Diffraction from a Left Ventricular Wall of Rat Heart. Biophys J 2004;86: [33] Farman GP, Allen EJ, Gore D, Irving TC, de Tombe PP. Interfilament Spacing Is Preserved during Sarcomere Length Isometric Contractions in Rat Cardiac Trabeculae. Biophys J 2007;92:L [34] Fuchs F, Martyn DA. Length-dependent Ca 2+ activation in cardiac muscle: some remaining questions. J Muscle Res Cell Motil 2005;26: [35] Allen DG, Kurihara S. The effects of muscle length on intracellular calcium transients in mammalian cardiac muscle. J Physiol 1982;327: [36] Cazorla O, Vassort G, Garnier D, J.Y. LG. Length modulation of active force in rat cardiac myocytes: is titin the sensor? J Mol Cell Cardiol 1999;31: [37] Cazorla O, Wu Y, Irving TC, Granzier H. Titin-based modulation of calcium sensitivity of active tension in mouse skinned cardiac myocytes. Circ Res 2001;88: [38] Fukuda N, Granzier HL. Titin/connectin-based modulation of the Frank-Starling mechanism of the heart. J Muscle Res Cell Motil 2005;26: [39] Fitzsimons DP, Patel JR, Moss RL. Role of myosin heavy chain composition in kinetics of force development and relaxation in rat myocardium. J Physiol 1998;513: [40] Fitzsimons DP, Patel JR, Moss RL. Aging-dependent depression in the kinetics of force development in rat skinned myocardium. Am J Physiol 1999;276:H [41] Rundell VLM, Manaves V, Martin AF, detombe PP. Impact of beta-myosin heavy chain isoform expression on cross-bridge cycling kinetics. Am J Physiol 2005;288:H896-H903. [42] Korte FS, McDonald KS. Sarcomere length dependence of rat skinned cardiac myocyte mechanical properties: dependence on myosin heavy chain. J Physiol 2007;581:

24 [43] Martyn DA, Rondinone JF, Huntsman LL. Myocardial segment velocity at a low load: time, length, and calcium dependence. Am J Physiol 1983;13:H708-H714. [44] Brenner B. The necessity of using two parameters to describe isotonic shortening velocity of muscle tissues: the effect of various interventions upon initial shortening velocity (v i ) and curvature (b). Basic Res Cardiol 1986;81: [45] Moss RL. The effect of calcium on the maximum velocity of shortening in skinned skeletal muscle fibers of the rabbit. J Muscle Res Cell Motil 1982;3: [46] Iwamoto I. Thin filament cooperativity as a major determinant of shortening velocity in skeletal muscle fibers. Biophys J 1998;74: [47] McDonald KS. Ca 2+ dependence of loaded shortening in rat skinned cardiac myocytes and skeletal muscle fibers. J Physiol 2000;525: [48] Chiu YC, Ballou EW, Ford LE. Force, velocity, and power changes during normal and potentiated contractions of cat papillary muscle. Circ Res 1987;60: [49] Sweitzer NK, Moss RL. Determinants of loaded shortening velocity in single cardiac myocytes permeabilized with alpha-hemolysin. Circ Res 1993;73: [50] Maughan DW, Low ES, Alpert NR. Isometric force development, isotonic shortening, and elasticity measurements from Ca 2+ -activated ventricular muscle of the guinea pig. J Gen Physiol 1978;71: [51] Brutsaert DL, Sys SU. Relaxation and diastole of the heart. Physiol Rev 1989;69: [52] Peterson JN, Hunter WC, Berman MR. Estimated time course of Ca 2+ bound to troponin C during relaxation in isolated cardiac muscle. Am J Physiol 1991;260:H1013-H1024. [53] Backx PH, Gao WD, Azow-Backx MD, Marban E. The relationship between contractile force and intracellular [Ca 2+ ] in intact cardiac trabeculae. J Gen Physiol 1995;105:

25 [54] Janssen PML, Stull LB, Marban E. Myofilament properties comprise the rate-limiting step for cardiac relaxation at body temperature in the rat. Am J Physiol 2002;282:H [55] Burkhoff D. Explaining load dependence of ventricular contractile properties with a model of excitation-contraction coupling. J Mol Cell Cardiol 1994;26: [56] Hinken AC, Solaro RJ. A dominant role of cardiac molecular motors in the intrinsic regulation of ventricular ejection and relaxation. Physiol 2007;22: [57] Piroddi N BA, Scellini B, Tesi C, Giunti G, Cerbai E, Mugelli A, Poggesi C. Tension generation and relaxation in single myofibrils from human atrial and ventricular myocardium. Pflug Arch 2007; [58] Solzin J, Iorga B, Sierakowski E, Gomez Alcazar DP, Ruess DF, Kubacki T, et al. Kinetic Mechanism of the Ca2+-Dependent Switch-On and Switch-Off of Cardiac Troponin in Myofibrils. Biophys J 2007;93: [59] Ramsey RW, Street SF. The isometric length-tension diagram of isolated skeletal muscle fibre of the frog. J Cell Comp Physiol 1940;15: [60] Gordon AM, Huxley AF, Julian FJ. The variation in isometric tension with sarcomere length in vertebrate muscle fibers. J Physiol 1966;184: [61] Fabiato A, Fabiato F. Dependence of the contractile activation of skinned cardiac cells on the sarcomere length. Nature 1978;256: [62] Allen DG, Kentish JC. The cellular basis of the length-tension relation in cardiac muscle. J Mol Cell Cardiol 1985;17: [63] Kentish JC, ter Keurs HEDJ, Ricciardi L, Bucx JJJ, Noble MIM. Comparison between the sarcomere length-force relations of intact and skinned trabeculae from rat right ventricle. Influence of calcium concentrations on these relations. Circ Res 1986;58:

26 [64] Close RI. The relations between sarcomere length and characteristics of isometric twitch contractions of frog sartorius muscle. J Physiol 1972;220: [65] Hibberd MJ, Jewell BR. Calcium and length dependent force production in rat ventricular muscle. J Physiol 1982;329: [66] Moss RL, Swinford AE, Greaser ML. Alterations in the Ca2+ sensitivity of tension development by single skeletal muscle fibers at stretched lengths. Biophys J 1983;43: [67] Harrrison SM, Lamont C, Miller DJ. Hysteresis and the length dependence of calcium sensitivity in chemically-skinned rat cardiac muscle. J Physiol 1988;401: [68] McDonald KS, Moss RL. Osmotic compression of single cardiac myocytes reverses the reduction in Ca 2+ sensitivity of tension observed at short sarcomere length. Circ Res 1995;77: [69] Fitzsimons DP, Moss RL. Strong binding of myosin modulates length-dependent Ca 2+ activation of rat ventricular myocytes. Circ Res 1998;83: [70] Konhilas JP, Irving TC, detombe PP. Myofilament calcium sensitivity in skinned rat cardiac trabeculae: role of interfilament spacing. Circ Res 2002;90: [71] Wang YP, Fuchs F. Osmotic compression of skinned cardiac and skeletal muscle bundles: effects on force generation, calcium sensitivity, and calcium binding. J Mol Cell Cardiol 1995;27: [72] Konhilas JP, Irving TC, Wolska BM, Jweied EE, Martin AF, Solaro RJ, et al. Troponin I in the murine myocardium: influence on length-dependent activation and interfilament spacing. J Physiol 2003;547: [73] Martyn DA, Adhikari BB, Regnier M, Gu J, Xu S, Yu LC. Response of equatorial X-ray reflections and stiffness to altered sarcomere length and myofilament lattice spacing in relaxed cardiac muscle. Biophys J 2004;86: