Static and Dynamic Properties of the HCM Myocardium

Hypertrophic cardiomyopathy (HCM) is a complex and relatively common genetic cardiac disease associated with significant genotypic and phenotypic heterogeneity and affects about one in 500 individuals of the general population [1-2]. HCM is typically associated with a portion of the myocardium being hypertrophied - usually of the left ventricle in the absence of loading conditions such as hypertension [3]. While the occurrence of HCM is a significant cause of unexpected (sudden) cardiac death in all age groups, it is best known as a leading cause of sudden cardiac death in young athletes [4]. HCM is associated with more than 550 mutations in 10 genes that code for cardiac sarcomeric proteins [5]. Five genes account for greater than 90% of the HCM mutations so far discovered [5]. These genes include β-myosin heavy chain (β-MyHC, encoded by the MYH7 gene), β-myosin binding protein C (MyBP-C encoded by the MYBPC3 gene), cardiac troponin T (cTnT, encoded by the TNNT2 gene), cardiac troponin I (cTnI, encoded by the TNNI3 gene), and α-tropomyosin (α-Tm, encoded by the TPM1 gene).

The clinical course of HCM is highly variable with some patients being asymptomatic or mildly symptomatic while others undergo dyspnea, syncope, and even sudden cardiac death. This dyspnea has been suggested to be largely due to increased stiffness of the left ventricle (LV), which impairs ventricular filling resulting in elevated pressures in the left atrium and left ventricle. Myocyte hypertrophy, disarray and interstitial fibrosis have also been proposed as main contributors to HCM induced diastolic and systolic dysfunction [6]. However, the link between the sarcomeric mutations as well as the actual molecular mechanism of pathological progression remain unclear. The paper by Hoskins et al. in this issue of Journal of Molecular and Cellular Cardiology investigated the steady-state and dynamic characteristics of passive stiffness and active force production in the myocytes from HCM and non-diseased hearts to determine the extent to which dysfunction at the myocyte level contributes to impairment of cardiac performance [7]. Myocytes from heart samples of patients with mutations in the two main sarcomeric proteins associated with HCM, β-MyHC (R719Q) and MyBP-C (R502W and T2604A + C deletion at 2605), as well as HCM hearts not associated with any mutations (three patients) were compared with myocytes from non-diseased hearts [7]. Although the number of HCM samples from patients with sarcomeric mutations were small (two patients with MYBPC3 mutations and one with a MYH7 mutation) it is important to be aware of the difficulty in getting these types of samples. All of the HCM samples were compared against four non-diseased donor hearts. Under pathogenic conditions, like HCM, the heart shows increased myocardial stiffness compared to non-diseased hearts [6]. Hoskins et al. investigated two types of passive force of myocytes: elastic (time-independent) and viscoelastic (time-dependent). Interestingly, and perhaps unexpectedly, Hoskins et al. found that the passive elastic and viscous properties of the myocytes were similar between HCM patients and were not significantly different from those of donor myocytes, suggesting that the increased wall stiffness was not due to changes in myocyte passive stiffness. While the peak force showed a trend to be higher for HCM myocytes it was not significantly different from the donor myocytes [7]. These results provide the first evidence that diastolic dysfunction in at least some HCM myocytes do not originate from the stiffness of the sarcomere.

Measurement of passive viscoelasticity in cardiac muscle started 28 years ago when Noble measured the force of cat papillary muscle during diastole at different stretch velocities [8]. Five years later, Chiu et al. observed viscous resistance to both stretch and active shortening in cat papillary muscle [9-10]. Viscoelasticity, which manifests itself in stress relaxation (force decay at constant length) following a stretch and in force hysteresis during a stretch-release cycle [11], is thought to reflect passive stiffness of the myocardium, and thus is an important measurement of diastolic function. Cardiac muscle behaves as a viscoelastic material, resulting in the filling of the ventricle which generates a resisting force (stress) when myocardial length is increased (strain). Changes in myocardial stiffness can be assessed by examination of the myocardial stress, strain, and strain-rate relationships during diastole [12].

Titin (also known as connectin) is known to be the main factor involved in determining both elasticity and viscoelasticity of myofibrils [13]. Titin is the largest known protein and extends from the Z-line to M-line regions of the sarcomere [14]. While part of titin is inextensible (Z line and A-band regions), the majority of the I-band region of titin functions as a molecular spring that develops passive force [15]. Cardiac titin exists in two main isoforms termed N2B and N2BA, and its isoform expression and relative phosphorylation has been shown to be altered in human heart failure (HF) myocytes when compared to donor myocytes. In heart failure myocytes, the N2BA titin (compliant isoform) increases relative to the N2B (stiff isoform) and the total phosphorylation of the stiff N2B titin isoform is significantly lower in HF resulting in raised passive force of human HF cardiomyocytes [16]. Hoskins et al. investigated titin and found that, unlike what was found in a previous animal model of overload hypertrophy [17], no differences in titin isoform content, amount, or extent of degradation was seen between the groups [7]. Measurement of titin phosphorylation is currently impractical since phosphorylation by PKCα increases stiffness [18] while phosphorylation by PKA or PKG decreases stiffness [19]; furthermore, these phosphorylation sites are not yet fully characterized. Considering that the passive viscoelastic properties of titin largely define the properties of the entire sarcomere, no change in the ratio of N2B to N2BA as well as no change in overall amount of N2B correlates with the results of the mechanical fiber experiments [7] which suggest that the stiffness of the sarcomere is not changed in the HCM hearts investigated.

While titin is the main protein involved in the viscosity of cardiac muscle [13], changes in viscous properties have been attributed to increased microtubule density in pressure-overload cardiac hypertrophy [20-21], suggesting that other factors may also contribute to the viscoelasticity of muscle. Other studies also suggest that Ca²⁺ [13, 22], collagen [23], and interstitial fibrosis [24] may be associated with passive myocardial tension. Two other proteins, merosin and alpha7-integrin (which both occur in cardiac muscle), have been suggested to be involved in passive skeletal muscle tension [25-26]. Collagen is the main structural component of connective tissue in the heart and experiments using a furazolidone-induced model of DCM in turkey poults trabecular muscle suggest that the increase in passive myocardial stiffness observed was primarily collagen-based [23]. It may be that collagen is partly involved in passive force in HCM myocytes since collagen IV fibers in HCM myocytes were more abundant and less associated with costamere structures than donor myocytes [7]. However, more detailed experiments on collagen in HCM hearts is needed out as some experimental data suggest that the higher diastolic stiffness in the failing myocardium may not be related to increased interstitial collagen [27].

The paper by Hoskins et al. is one of only two papers that have significantly investigated the contractile properties of myocytes from HCM patients with known sarcomeric mutations. A previous study by Van Dijk et al. compared myocyte force measurements in human samples from two MyBP-C truncation-mutation carriers (c.2373dupG, n=7 and c.2864_2865delCT, n=4) to myoctes from nonfailing donors (n=13) [28]. This group observed decreased maximal force of myocyte contraction (Table 1) and a non-significant trend of increased force in mutant fibers during passive measurements. To build on these findings, Hoskins et al. investigated several related aspects of myocyte performance besides passive force measurements: Ca²⁺ sensitivity of force, myocyte structure, collagen localization, contractile protein and phosphoprotein expression, maximum Ca²⁺ activated force, and cross-bridge cycling kinetics [7]. Hoskins et al. also found that skinned cardiac myocytes showed a large decrease in maximal force per cross-sectional area of the myocytes in samples from HCM hearts (irrespective of the mutation) relative to donor hearts (Table 1). This decreased maximum force is consistent with a transgenic mouse model of the thick filament protein myosin light chain 2 (MLC2) [29] (Table 1). This impaired maximal force production links systolic dysfunction to deficiency at the level of the sarcomere. The possibility of impaired force production due to myocyte structural deficiencies was examined by staining for myofibrillar components and videomicroscopy inspection [7]. While there was no difference in quantity of myofibrils there did seem to be structural implications which could be contributing to systolic impairment and warrant further investigation. Cross-bridge analysis revealed that a faster rate of cross-bridge detachment is a likely contributor to the force development deficiency [7]. The cross-bridge cycling kinetics during maximal Ca²⁺-activation as indexed by k_tr (rate constant of force redevelopment) was about 10% faster in HCM myocytes [7].

Table 1.

Effect of HCM on Ca²⁺ sensitivity, Maximal Force and Myofibrillar protein Changes

Parameter	HCM	Donor (control) heart	Significance	Ref.
Ca2+ sensitivity of force development (pCa50)	6.4±0.05 (human) 5.62±0.04 (human) 5.30±0.04 (mouse MLC2 D166V Tg)	6.09±0.02 (human) 5.54±0.02 (human) 5.02±0.03 (mouse MLC2 D166V Tg)	P<0.01 P<0.05 P<0.01	[7] [28] [29]
Maximal Ca2+ activated force	14±1 kN/m2(human) 20.2±2.7 kN/m2(human) 61.9±3.0 kN/m2 (mouse MLC2 D166V Tg)	23±3 kN/m2 (human) 34.5±1.1 kN/m2(human) 42.2±2.8 kN/m2 (mouse MLC2 D166V Tg)	P<0.001 P<0.001 P<0.001	[7] [28] [29]
cTnI Ser23/24 phosphorylation (relative to donor heart)	Decreased Decreased by 84±5%	–	NL P<0.0001	[7] [28]
MyBPC Ser282 Phosphorylation (relative to donor heart)	Decreased No change*	–	NL NS	[7] [28]
Titin isoform ratio (relative to donor heart)	No change	–	NS	[7]
Titin protein level (relative to donor heart)	No change	–	NS	[7]
Resting SL	1.73±0.01μM	1.78±0.01μM	P<0.001	[7]
nH	3.73±0.56	2.95±0.12	NS	[7]

NS, not significant; NL, not listed; n_H, hill coefficient; SL, sarcomere length.

^*Phosphorylation of MyBP-C normalized to its own protein level was similar between MyBP-C mutant and donor myocardium but phosphorylation of MyBP-C normalized to α-actinin was reduced by 47±7% in the MyBP-C mutant compared with donor myocardium (P<0.0001).

Several functional studies using both human recombinant proteins in reconstituted porcine fibers and skinned fibers from animal models have indicated increased Ca²⁺ sensitivity of contraction and variable maximal force production (but mainly decreased) for many sarcomeric protein mutations were associated with HCM [30-31]. Similar increases in sensitivity to Ca²⁺ were observed in both recent human HCM studies [7, 28] (Table 1). This increased Ca²⁺ sensitivity observed in HCM myocytes has also been found in myocytes from end-stage failing human myocardium (idiopathic dilated cardiomyopathy) [32]. To elucidate the source of the increase in Ca²⁺ sensitivity, Hoskins et al. determined the level of phosphorylation of both cTnI (at ser-23/24) and MyBP-C (at ser-282) [7]. Their results suggest that impairment of PKA regulated sarcomere phosphorylation may be at least partly responsible for the increased sensitivity to Ca²⁺. These results were in part corroborated by van Dijk et al., who also found decreased phosphorylation of cTnI, but that group showed no difference in MyBP-C phosphorylation [28]. Reduced cTnI and MyBP-C phosphorylations have been observed in animal models of cardiac hypertrophy and failure and in humans with end-stage idiopathic and ischemic cardiomyopathy [33-34]. Furthermore, a HCM mutation in cTnI (R21C) has been shown to directly affect the rate of phosphorylation of ser-23/24 on cTnI [35]. The differences observed between the two human HCM studies may be due to differences in how MyBP-C phosphorylation is determined. To further help understand the role of phosphorylation and sarcomeric protein levels in cardiac dysfunction we have to be careful not to rely solely on antibodies. Studies have shown that some phosphoantibodies do not accurately report the phosphorylation levels [36]. Both van Dijk et al. and Hoskins et al. base a significant amount of their conclusions on the phosphorylation status of cTnI and MyBP-C. While the cTnI phosphoantibody has been extensively validated and was obtained from Cell Signaling Technology for both studies, the anti phosphoMyBP-C was obtained from different non-commercial sources and has not been widely tested.

Van Dijk et al. showed that although the truncated MyBP-C that should be expressed by these mutant carriers were not detected by western blots, the total protein expression of MyBP-C was significantly reduced in hearts from mutant carriers by 33±5% [28]. They further observed reduced phosphorylation of desmin, cTnT, and MLC2 in MYBPC3 mutant carriers when compared with donor samples [28]. Treatment with exogenous protein kinase A (PKA, which mimics β-adrenergic stimulation) eliminated the increase in Ca²⁺ sensitivity between the MyBP-C mutant and donor samples [28]. These results suggest that enhanced Ca²⁺ sensitivity in MyBP-C mutant carriers is due to hypophosphorylation of cTnI secondary to mutation-induced dysfunction [28]. The decrease in Ca²⁺ sensitivity to levels similar to donor myocytes observed in HCM myocytes after PKA treatment is interesting since it would be expected that the 33% decrease in MyBP-C (independent of phosphorylation) would result in increased Ca²⁺ sensitivity [37-38]. Removal of ~30% to 70% of MyBP-C from rat cardiomyocytes has previously been shown to increase Ca²⁺ sensitivity [37-38]. Hoskins et al. found that treatment of unrestrained skinned myocytes with PKA decreased myofibrillar Ca²⁺ sensitivity but only to levels in-between the original HCM myocyte and the donor Ca²⁺ sensitivity [7]. The increased Ca²⁺ sensitivity observed in HCM myocytes (relative to donor myocytes) after PKA treatment may be due to the influence of the primary mutation or changes in the non-PKA dependent phosphorylation of other proteins. In both van Dijk et al. and Hoskins et al. studies, PKA treatment of MyBP-C mutant samples did not correct the reduced maximal force [7, 28].

As the number of investigated heart samples of known mutations increase, the unknown role of differences in medications, age, and gender on the biophysical parameters measured could be better determined. It is also important to determine if there are differences in the muscle parameters measured in tissue from LV versus the septum from the same heart. In both studies by Hoskins et al. and van Dijk et al. HCM tissue was obtained from septal tissue while donor tissue was obtained from the LV free wall [7] [28]. It is not known if the amount, isoform pattern and phosphorylation status of titin in different regions of the same heart are the same. Besides possible differences between the septum and the LV, another concern is the nonuniformity of the normal adult LV [39]. Stretch-dependent Ca²⁺ sensitization of skinned rat myocytes increases from sub-epicardium to sub-endocardium and correlates with an increase in passive tension. This increase in passive tension is not associated with changes in titin isoforms, changes in cTnI levels, changes in titin or cTnI phosphorylation levels but could be associated with phosphorylation of ventricular MLC2 isoform (VLC2b) in the sub-endocardium [39]. However, MLC-2 (RLC) phosphorylation level was not significantly different between human HCM and donor myocardium [7]. HCM is associated with marked regional differences in wall thickness and is characterized by regions of local fiber disarray which vary in size and are typically localized within hypertrophic regions [40-42]. In a study to investigate the regional and global LV function and wall thickness in patients with HCM using deformation imaging it was found that the heterogeneity of regional LV systolic function was partly attributable to the heterogeneity of LV hypertrophy. This suggests a link to impaired global LV relaxation in HCM [43]. Other studies also suggest a regional nature of myocardial involvement in HCM [42, 44]. Hence, the location of the tissue studied in the heart may be more important than we currently realize.

The experiments by Hoskins et al. provide evidence against the passive properties of the sarcomere as a cause of diastolic dysfunction. These studies also suggest that while the mutations as well as phosphorylation of cTnI and MyBP-C are important in HCM, other unknown factors are also involved [7]. Understanding the molecular mechanism(s) behind the decreased maximal force production of HCM myofibrils will lead to a better understanding of the pathogenesis of systolic dysfunction. The data regarding the contractile protein phosphorylation status as well as the impact of PKA strongly suggest higher activation of PKA in the donor samples than in HCM samples. Determination of PKA activity in HCM and donor samples will help to better define the role of PKA in HCM.

In summary, while the Hoskins et al. paper advances our knowledge about the passive properties of HCM myocytes [7], we are still in the infant stage of understanding the passive properties of HCM muscle. Although titin is a major contributor to passive tension, recent results suggest that it is unlikely that titin can account for all the features of the resting tension response in cardiac muscle. The observations by Hoskins et al. and van Dijk et al. suggest that HCM pathogenesis is associated with increased Ca²⁺ sensitivity (diastolic impairment) and decreased maximal force of contraction (systolic impairment). The increase in Ca²⁺ sensitivity is a sarcomeric contribution to diastolic dysfunction and the occurrence of these functional changes which were independent of the gene mutation suggests that they are secondary consequences of the underlying genetic differences between patients. As more laboratories gain access to human HCM hearts the dissection of the signaling pathways and components of active and passive tension will proceed at a significant rate.