OUP user menu

Effect of protein kinase A on calcium sensitivity of force and its sarcomere length dependence in human cardiomyocytes

J. van der Velden , J.W. de Jong , V.J. Owen , P.B.J. Burton , G.J.M. Stienen
DOI: http://dx.doi.org/10.1016/S0008-6363(00)00050-X 487-495 First published online: 1 June 2000

Abstract

Objective: We investigated whether the Frank–Starling mechanism is absent or preserved in end-stage failing human myocardium and if phosphorylation of contractile proteins modulates its magnitude through the sarcomere length-dependence of calcium sensitivity of isometric force development. Methods: The effect of phosphorylation of troponin I and C-protein by the catalytic subunit of protein kinase A (3 μg/ml; 40 min at 20°C) was studied in single Triton-skinned human cardiomyocytes isolated from donor and end-stage failing left ventricular myocardium at sarcomere lengths measured at rest of 1.8, 2.0 and 2.2 μm. Isometric force development was studied at various free-calcium concentrations before and after protein kinase A incubation at 15°C (pH 7.1). Results: Maximal isometric tension at 2.2 μm amounted to 39.6±10.4 and 33.7±3.5 kN/m2 in donor and end-stage failing cardiomyocytes, respectively. The midpoints of the calcium sensitivity curves (pCa50) of donor and end-stage failing hearts differed markedly at all sarcomere lengths (mean ΔpCa50=0.22). A reduction in sarcomere length from 2.2 to 1.8 μm caused reductions in maximum isometric force to 64% and 65% and in pCa50 by 0.10 and 0.08 pCa units in donor and failing cardiomyocytes, respectively. In donor tissue, the effect of protein kinase A treatment was rather small, while in end-stage failing myocardium it was much larger (ΔpCa50=0.24) irrespective of sarcomere length. Conclusions: The data obtained indicate that the Frank–Starling mechanism is preserved in end-stage failing myocardium and suggest that sarcomere length dependence of calcium sensitivity and the effects of phosphorylation of troponin I and C-protein are independent.

Keywords
  • Heart failure
  • Contractile apparatus
  • Contractile function
  • Myocytes
  • Protein kinases
  • Protein phosphorylation

Time for primary review 21 days.

1 Introduction

The Frank–Starling law of the heart, which is based on the relations between sarcomere length, force development and its calcium sensitivity, indicates that an increase in diastolic filling causes an increase in peak systolic pressure. Conflicting results have been reported about the existence of the Frank–Starling mechanism in end-stage failing myocardium. According to Schwinger et al. [1] the Frank–Starling mechanism is absent in the end-stage failing heart, while Holubarsch et al. [2] and Vahl et al. [3] found that it was preserved in failing human myocardium. In the non-failing heart, several compensatory mechanisms exist to meet higher demands. An increase in heart rate will increase force development by increasing intracellular calcium [4], while activation of the sympathic nervous system increases pump function via β-adrenergic receptor stimulation. However, under pathological conditions these processes are impaired [4–6]. Hence, the Frank–Starling mechanism would be one of the few processes remaining to compensate for increases in workload.

Cardiac muscle normally functions between sarcomere lengths ranging from about 1.7 to 2.3 μm. Within this range, an increase in length results in an increase in force, which is due to the change in the overlap of actin and myosin as well as an increase in calcium sensitivity of force development. The increase in calcium sensitivity of force with increasing sarcomere length most probably results from an increase in calcium binding to troponin C [7,8]. However, the molecular basis of this length-dependent alteration in the affinity of troponin C for calcium is not completely clear. Evidence suggests that interfilament lattice spacing could be an important determinant of calcium sensitivity [9–11]. Because volume is constant, an increase in length is associated with a decrease in interfilament lattice spacing. The decrease in interfilament lattice spacing probably promotes formation of strongly-bound cross-bridges [12,13] which may enhance force production directly or indirectly via cooperative activation of the thin filament [14].

In addition, thin filament proteins other than troponin C (e.g. troponin I, troponin T, tropomyosin) influence cross-bridge binding or troponin C calcium binding and in this way might modify the length-dependence of activation [15]. Phosphorylation of troponin I and C-protein by the catalytic subunit of protein kinase A (PKA) causes a decrease in calcium sensitivity of isometric force production [16–18]. Phosphorylation of troponin I accelerates calcium removal from troponin C [19]. Furthermore, phosphorylation of troponin I and C-protein by isoproterenol may enhance the length-dependent change in troponin C calcium affinity [20]. If phosphorylation of troponin I and C-protein is involved in the length-dependence of calcium sensitivity, an explanation for the observed discrepancies concerning the Frank–Starling mechanism in failing human myocardium [1–3] might reside in different phosphorylation levels of the contractile proteins in failing human hearts studied by the different groups.

In the present study, we therefore investigated the effect of PKA incubation on force development in single human myocytes from donor and end-stage failing hearts at different sarcomere lengths (1.8, 2.0 and 2.2 μm). This approach allowed us to study intrinsic differences in length-dependent force development of the myofibrils and to establish if phosphorylation of troponin I and C-protein are involved in the length-dependence of calcium sensitivity. Cardiomyocytes were mechanically isolated from frozen human left ventricular tissue and permeabilised with Triton X-100. Permeabilised cardiomyocytes have the advantage that sarcomere length, isometric force development and its calcium sensitivity can be measured without interference of extracellular matrix components. Alterations in the extracellular matrix (e.g. collagen) that occur during heart failure may affect contractile properties of the myocardium. Furthermore, by removing all membranous structures (sarcoplasmic reticulum, mitochondria), myofibrillar contractile properties can be measured under standardized conditions (i.e. composition of the intracellular medium, sarcomere length) without disturbing factors present in the intact heart (i.e. hormonal factors, variable calcium concentrations).

2 Methods

2.1 Biopsies

Left ventricular biopsies were taken during heart transplantation surgery from three explanted failing hearts and from three donor hearts which were not used for transplantation for technical reasons. The patients were classified according to the New York Heart Association (NYHA) as NYHA class IV (end-stage heart failure). The characteristics of patients and donors are given in Table 1. The samples were obtained with approval of the local Ethical Committees. The investigation conforms with the principles outlined in the Declaration of Helsinki [21]. After excision the biopsies were transferred in cardioplegic solution and upon arrival in the laboratory, frozen and stored in liquid nitrogen for up to 7 years. Prior to the experiments, the tissue was stored at −80°C for up to one month.

View this table:
Table 1

Characteristics of donor and end-stage failing heartsa

PatientSexAgeMedication
Donor 1Female26
Donor 2Male23
Donor 3Male44PI
Patient 1Male57AC,ACE,D,N, PI
Patient 2Male49
Patient 3Male59ACE,D,PI
  • a Abbreviations: AC=anticoagulant, ACE=ACE inhibitor, D=diuretics, N=nitrates, PI=positive inotrope.

2.2 Myocyte isolation and experimental set-up

The myocytes were mechanically isolated and mounted in the experimental set-up as described previously [22]. Briefly, a small piece of tissue (∼10 mg) was defrosted and kept on ice in 2 ml isolating solution which contained (mmol/l): free Mg2+ 1, KCl 145, EGTA 2, ATP 4, imidazole 10; pH adjusted to 7.0 with KOH. Subsequently, the tissue was cut into small pieces, put into a glass cylindrical beaker and mechanically disrupted with a Teflon piston within 5 to 10 s, using a tissue homogenizer set at 900 r.p.m. In this way, a suspension of small clumps of myocytes, single myocyte-sized preparations and cell fragments was obtained. To remove all membranous structures the cells were incubated in isolating solution containing 0.3% Triton X-100 for 5 min. Triton removes all soluble and membrane-bound kinases and phosphatases and thereby arrests the phosphorylation state of myofibrillar proteins. Thereafter, the myocytes were washed twice in isolating solution to remove Triton and kept at 0°C up to 24 h. Single myocyte-sized preparations were selected for mechanical measurements on the basis of size (100–150 μm long by 15–30 μm in diameter) and uniformity of striation pattern. The single myocyte was attached with silicon adhesive to thin stainless steel needles extending from a force transducer and a fixed bar while viewed (320× magnification) by an inverted microscope. Average sarcomere length was determined by means of a spatial Fourier transform [23] and adjusted to a sarcomere length of 1.8, 2.0 or 2.2 μm. Length and diameters of the cardiomyocytes were determined at a sarcomere length of 2.2 μm. Cross-sectional area was calculated assuming an elliptical cross-section.

2.3 Solutions

The relaxing and activating solutions used, contained, respectively (in mmol/l): MgCl2 6.42 and 6.28, Na2ATP 5.94 and 6.03, EGTA 5 and 0, Ca–EGTA 0 and 5. Ca–EGTA was made by dissolving equimolar amounts of CaCO3 and EGTA. In addition both solutions contained 14.5 mmol/l creatine phosphate and 60 mmol/l N,N-bis[2-hydroxyethyl]-2-aminoethane-sulphonic acid (pH adjusted to 7.1 with KOH). The ionic strength of the solutions was adjusted to 200 mmol/l with potassium propionate. The pCa (−log10[Ca2+]) of the relaxing solution and activating solution were, respectively, 9 and 4.5. The composition of the solutions was calculated by means of a computer program similar to that described by Fabiato [24]. The calculated free Mg2+ and MgATP concentrations were 1 and 5 mmol/l, respectively. Solutions with lower [Ca2+] were obtained by appropriate mixing of the activating and relaxing solutions, assuming an apparent stability constant of the Ca–EGTA complex of 106.45.

2.4 Force measurements

After curing of the glue for 50 min, the myocyte was transferred from the isolating solution on the mounting area to a small temperature-controlled well (volume 80 μl) containing relaxing solution. Isometric force was measured after the preparation was transferred, by moving the stage of the inverted microscope, to a temperature-controlled well containing activating solution. When steady tension was reached, the myocyte was returned to the relaxing solution. During the activation–relaxation cycles, the myocyte could be inspected through the glass bottom of the wells. All force measurements were performed at 15°C. The experimental temperature of 15°C was chosen to ensure stability of the preparation throughout the experiment. Moreover, at higher temperatures sarcomere uniformity during maximal force development is not well preserved [25]. Although maximal force is not very temperature-sensitive [26], caution should be exerted in extrapolation of the values found in this study to cardiac performance at body temperature. After the first activation at saturating (maximal) [Ca2+], sarcomere length was readjusted to the desired length (1.8, 2.0 or 2.2 μm), if necessary. The second measurement at pCa 4.5 was used to calculate maximal force per cross-sectional area. The next four to five measurements were carried out at submaximal [Ca2+] (pCa>4.5), followed by a control measurement at pCa 4.5. In two donor cardiomyocytes and four heart failure cardiomyocytes, a force–pCa curve was obtained at all sarcomere lengths (1.8, 2.0 and 2.2 μm), while in the other myocytes force measurements were performed at one or two different sarcomere lengths. After an initial series of force–pCa curves at different sarcomere lengths, the myocyte was incubated in relaxing solution containing the catalytic subunit of PKA (3 μg/ml: Sigma batch no. 35H9522) and 6 mM dithiothreitol for 40 min at 20°C. After PKA treatment, a second series of force–pCa curves was recorded. These force–pCa curves were obtained at sarcomere lengths of 1.8, 2.0 and 2.2 μm in, respectively, 7, 6 and 7 myocytes. On average, force decline in maximal force development after the first force–pCa curve measured at 2.2 μm for all cardiomyocytes (donor and heart failure) amounted to 2%. At the end of the experiment maximal force at 2.2 μm was on average reduced to 91±2%. Force decline during the entire experiment did not differ between donor and heart failure cardiomyocytes.

Passive force was recorded at a sarcomere length of 2.2 μm. It was obtained from the difference in force found when the myocyte was stretched from slack length corresponding to sarcomere length ∼1.7 to 2.2 μm.

2.5 Data analysis

The force–pCa relation was fit by a non-linear fit procedure to a modified Hill equation: Embedded Image where F is steady-state force. F0 denotes the steady isometric force at saturating Ca2+ concentration, nH represents the steepness of the relationship, and Ca50 (or pCa50) represents the [Ca2+] at which force=0.5×F0, i.e. the midpoint of the relation. Fmax, i.e. F0 at 2.2 μm sarcomere length, was determined from the second maximal activation at pCa 4.5.

Data points are given as means ±S.E.M. of n experiments. Differences were tested by means of Student's t-test.

3 Results

Isometric force and its calcium sensitivity were determined at different sarcomere lengths in eight myocytes isolated from three end-stage failing hearts and eleven myocytes from three donor hearts. Fig. 1 shows a ventricular cardiomyocyte from donor 2 when viewed in the mounting area at a sarcomere length of 1.98 μm (A) and 2.19 μm (B) in relaxing solution (pCa 9.0). Corresponding power spectra calculated from most of the cell visible by means of the spatial Fourier transform are shown in panels C and D. The half width of the power spectra is a measure of sarcomere non-uniformity within the preparation and corresponded to 0.02 μm in C and D. Average values for length, diameters and maximal force per cross-sectional area at a sarcomere length of 2.2 μm of failing and donor cardiomyocytes are given in Table 2. No significant differences were present between length, diameters and maximal isometric tension at 2.2 μm of the two groups. Passive force per cross-sectional area at a sarcomere length of 2.2 μm was low in both groups, albeit slightly higher in failing (2.4±0.3 kN/m2; n=8) compared to donor (1.3±0.2 kN/m2; n=6) myocytes (P<0.05).

Fig. 1

Cardiomyocyte from donor 2 at a sarcomere length of 1.98 μm (A) and 2.19 μm (B) in relaxing solution. Corresponding power spectra obtained with the spatial Fourier transform are shown in (C) and (D). Power is normalized to the peak value. The half width of the power spectra is indicated by the dotted lines. Sarcomere length at rest in failing and donor cardiomyocytes amounted to 2.19±0.01 and 2.21±0.01 μm, 2.01±0.01 and 2.00±0.01 μm, 1.83±0.02 and 1.81±0.01 μm, respectively.

View this table:
Table 2

Length between attachments, diameters and maximal tension at 2.2 μm

PreparationsLength (μm)Width (μm)Depth (μm)Maximal tension (kN/m2)
Donor (n=6)80.4±4.324.5±1.416.8±0.739.6±10.4
Heart failure (n=8)76.0±5.128.1±2.614.1±1.433.7±3.5

3.1 Force measurements at different sarcomere lengths

Recordings of force production obtained during a contraction–relaxation cycle at sarcomere lengths of 1.8 and 2.2 μm measured at saturating and at submaximal [Ca2+] (pCa 4.5 and 5.0, respectively) are shown in Fig. 2. Force development was lower at 1.8 μm sarcomere length both at maximal and submaximal [Ca2+] compared to 2.2 μm. The sarcomere length-dependence of maximal isometric force and its calcium sensitivity were determined by exposing the cardiomyocytes to different [Ca2+] and measuring isometric force development at sarcomere lengths of 1.8, 2.0 and 2.2 μm. The force–pCa relationships obtained in myocytes from donor and end-stage failing hearts are shown in Fig. 3. Maximal isometric force development increased substantially with increasing sarcomere length (Table 3). Calcium responsiveness of force also increased with increasing sarcomere length in donor and failing cardiomyocytes. In both groups an increase in sarcomere length from 1.8 to 2.2 μm caused an increase in pCa50 by approximately 0.1 unit as illustrated by the dotted lines in Fig. 3. The steepness (nH) of the force–pCa relations shown in Fig. 3 did not differ between sarcomere lengths. The average values for pCa50 and nH are summarized in Table 3.

Fig. 3

Calcium sensitivity of force at different sarcomere lengths (1.8, 2.0 and 2.2 μm) in donor (A) and failing (B) cardiomyocytes. Force at submaximal and at maximal [Ca2+] at a sarcomere length of 1.8 and 2.0 μm is normalized to the control force found at a sarcomere length of 2.2 μm in the same cardiomyocyte at saturating [Ca2+] (pCa 4.5). The force–pCa relations were fitted to a modified Hill equation. In both groups an increase in sarcomere length from 1.8 to 2.2 μm caused an increase in the pCa50 value by approximately 0.1 pCa unit, as indicated by the dotted lines. n=number of cardiomyocytes. *P<0.05 2.0 μm vs. 2.2 μm. #P<0.05 1.8 μm vs. 2.2 μm.

Fig. 2

Recordings of isometric contractions at sarcomere lengths of 1.8 (dashed recording) and 2.2 μm (continuous recording) during maximal (pCa 4.5) and submaximal activation (pCa 5.0). Force development was lower at 1.8 μm compared to 2.2 μm both at maximal and submaximal [Ca2+]. The abrupt changes in force mark the transitions of the preparation through the interface between solution and air. The dashed horizontal lines indicate the passive force level, determined immediately after the preparation was transferred to the activating solution, i.e. before active force developed. The differences in baselines of the force signal in relaxing and activating solution are due to the effect of surface tension on the output of the force transducer.

View this table:
Table 3

Maximal force, pCa50 and nH valuesa

PreparationParameter1.8 μm2.0 μm2.2 μm
DonorF00.64±0.12b (n=2)0.77±0.02b (n=4)1.0 (n=5)
pCa505.04±0.02b (n=5)5.12±0.01 (n=7)5.14±0.03 (n=5)
nH3.76±0.28 (n=5)3.72±0.35 (n=7)3.65±0.44 (n=5)
Heart failureF00.65±0.03b (n=7)0.77±0.05b (n=4)1.0 (n=8)
pCa505.28±0.02bc (n=7)5.31±0.03c (n=4)5.36±0.02c (n=8)
nH3.07±0.19 (n=7)3.09±0.43 (n=4)3.44±0.28 (n=8)
  • a Maximal force (F0) at a sarcomere length of 1.8 and 2.0 μm is normalized to the maximal control force found at a sarcomere length of 2.2 μm obtained in the same myocyte. n=number of cardiomyocytes.

  • b P<0.05 vs. 2.2 μm.

  • c P<0.01 vs. donor.

3.2 Donor versus failing cardiomyocytes

As can be seen in Table 3 calcium responsiveness of force was significantly higher in end-stage failing cardiomyocytes than in donor myocytes (P<0.01). At all sarcomere lengths the force–pCa curve of heart failure myocardium was shifted to the left compared to donor myocardium (mean ΔpCa50=0.22). The difference in calcium sensitivity at a sarcomere length of 2.2 μm is illustrated in Fig. 4. The steepness of the force–pCa curves, nH, did not differ significantly between donor and failing myocardium.

Fig. 4

Force–pCa relations of donor and failing cardiomyocytes at a sarcomere length of 2.2 μm. Calcium sensitivity was significantly higher in failing than in donor cardiomyocytes. The relations were fitted to a modified Hill equation. n=number of myocytes. *P<0.05 vs. donor.

3.3 Effect of PKA on calcium sensitivity

To investigate if phosphorylation of the contractile proteins troponin I and C-protein are involved in the Frank–Starling mechanism, force–pCa relations were measured before and after PKA treatment at different sarcomere lengths. Recordings of force production before and after PKA treatment in a failing cardiomyocyte at a sarcomere length of 2.2 μm are shown in Fig. 5. PKA treatment did not significantly change maximal isometric force development at saturating [Ca2+] (pCa 4.5) in both donor and in failing cardiomyocytes. At submaximal [Ca2+], however, isometric force development after PKA treatment was significantly decreased (see pCa 5.2 in Fig. 5). A small non-significant decline in maximal force (6±2%) was seen which might be attributable to the duration of the incubations performed at 20°C.

Fig. 5

Recordings of isometric force development at a sarcomere length of 2.2 μm before (continuous recording) and after PKA treatment (dashed recording) during maximal (pCa 4.5) and submaximal activation (pCa 5.2). Force at submaximal [Ca2+] decreased substantially after PKA treatment, while only a small decline in maximal force development was observed. The dashed horizontal lines denote the passive force level.

The decrease in force at submaximal [Ca2+] reflects a decreased calcium sensitivity of force after PKA treatment. The PKA-induced shift in pCa50 in cardiomyocytes from failing hearts was almost identical at all sarcomere lengths (0.24–0.25 pCa unit, Fig. 6A). On the other hand, in donor cardiomyocytes PKA treatment caused only a minor non-significant shift in pCa50 (Fig. 6B), but the sarcomere length-dependent change in calcium sensitivity was preserved after incubation in PKA. Fig. 6B clearly illustrates that the difference in calcium sensitivity of force between donor and failing cardiomyocytes is abolished after PKA treatment. The values for pCa50 and nH at various sarcomere lengths obtained after PKA treatment are summarized in Table 4.

Fig. 6

Calcium sensitivity of force before and after PKA treatment at different sarcomere lengths in failing cardiomyocytes (A) and at a sarcomere length of 2.2 μm in donor (D) and heart failure (HF) cardiomyocytes (B). Force at submaximal [Ca2+] was normalized to the control force found at maximal activation. The force–pCa relations were fitted to a modified Hill equation. PKA treatment significantly decreased calcium sensitivity in failing myocytes at all sarcomere lengths (A). After PKA treatment calcium responsiveness of donor and failing myocytes was similar (B). n=number of cardiomyocytes.

View this table:
Table 4

pCa50 and nH values after PKA treatmenta

PreparationParameter1.8 μm2.0 μm2.2 μm
DonorpCa505.03±0.02b (n=3)5.08±0.02 (n=4)5.14±0.03 (n=2)
nH2.65±0.08 (n=3)2.58±0.22 (n=4)3.12±0.11 (n=2)
Heart failurepCa505.06±0.03 (n=4)5.11±0.06 (n=2)5.12±0.03 (n=5)
nH3.85±0.45 (n=4)4.04±0.48 (n=2)3.61±0.36 (n=5)
  • a The data from each preparation were fitted to the Hill equation as shown in Fig. 6. Mean values were obtained by averaging the Hill parameters (pCa50 and nH) of all individual cardiomyocytes. n=number of cardiomyocytes.

  • b P<0.05 vs. 2.2 μm.

4 Discussion

To study the force–length relation in human myocardium we have applied a previously developed technique using mechanically isolated single cardiomyocytes [25]. Control of the sarcomere length is of major concern when studying myofibrillar calcium responsiveness. A small decrease in sarcomere length of 0.07±0.01 μm, i.e. about 3% of the length of the preparation between the attachments was observed previously during maximal activation at a sarcomere length of 2.2 μm [27], which mainly originated from the compliance of the mechanical parts of the set-up [22]. This indicates that the sarcomere lengths during the activations in this study will be somewhat less than the nominal values determined in relaxing solution. However, the differences are similar in donor and failing cardiomyocytes.

It should be noted that the cardiomyocytes swell by removing the membranes, thereby increasing the interfilament lattice spacing. Since interfilament lattice spacing is an important determinant of calcium sensitivity [9–11], the values obtained for calcium sensitivity of force in this study may differ from those found in intact tissue. However, since donor and failing cardiomyocytes were treated equally, we do not expect that swelling interferes with the conclusions presented in this study.

In accordance with previous studies on human tissue no difference was observed between maximal isometric tension of donor and end-stage failing cardiomyocytes, whereas calcium sensitivity was significantly increased in failing myocardium compared to donor hearts [1,18,28]. Wolff et al. [18] suggested that the increased calcium responsiveness observed in failing myocardial tissue may be due to a reduction of the β-adrenergically mediated phosphorylation of troponin I and C-protein. In agreement with their study, our results indicate that PKA treatment of cardiomyocytes abolished the difference in calcium sensitivity of force between donor and failing myocytes at all sarcomere lengths. It should be noted that minor residual differences in calcium responsiveness due to changes in isoform composition (re-expression fetal troponin T isoforms [29–32], expression atrial light chain-1 [28]) and phosphorylation levels of other contractile proteins cannot be completely ruled out.

An increase in sarcomere length in cardiomyocytes from donor tissue resulted in an increase in maximal force development as well as in its calcium sensitivity. The increase in calcium sensitivity of force observed in human cardiomyocytes is consistent with previous findings in single cardiomyocytes from rat [10,14]. In addition, in agreement with these studies, sarcomere length did not affect the steepness of the force–pCa relationships.

Similar increases in maximal force and its calcium sensitivity were observed in cardiomyocytes isolated from end-stage failing myocardium. Hence, our results are in accordance with previous observations by Holubarsch et al. [2] and Vahl et al. [3] who found the Frank–Starling mechanism to be present in failing human myocardium. The change in calcium sensitivity in response to a change in sarcomere length was also found to be preserved in hypertrophied rat hearts using similar force measurements in enzymatically isolated single cardiomyocytes [23]. Schwinger et al. [1] did not observe an increase in force development after an increase of sarcomere length in failing myocardium, which suggested complete absence of the length-dependent force development in end-stage heart failure.

We hypothesized that a possible explanation for this discrepancy may be different phosphorylation levels of contractile proteins in failing myocardial tissue, since the phosphorylation level of contractile proteins affects calcium sensitivity of force development in myofibrils [16–18,33–35]. The phosphorylation level of contractile proteins may be altered in heart failure due to β-adrenergic desensitization resulting in decreased phosphorylation via PKA as discussed above or protein kinase C (PKC) [33,34]. The absence of desensitization in donor hearts might provide an explanation for the minor effect of PKA in donor tissue. Increased PKC activity has been observed in failing human myocardium [36]. As a consequence of altered PKA and PKC activity, resulting in variable phosphorylation levels of contractile proteins, the Frank–Starling mechanism may be masked in the end-stage failing myocardium. Based on our findings it appears unlikely that the different observations concerning the Frank–Starling mechanism in failing human myocardium [1–3] are due to differences in phosphorylation levels of troponin I and C-protein.

Komukai and Kurihara [20] observed that isoproterenol treatment of ferret papillary muscles enhanced the length-dependent change in calcium sensitivity of force. This observation suggests that phosphorylation of troponin I and C-protein may be involved in the length-dependent alteration in the affinity of troponin C for calcium. However, in our study, the length-dependent shift in calcium sensitivity was unchanged in both donor and failing cardiomyocytes after PKA incubation, indicating that PKA-mediated phosphorylation is not involved in the sarcomere length-dependent force development of the heart.

It is tempting to speculate that existence of the Frank–Starling mechanism may depend on variable, regional expression of fetal troponin T isoforms. Akella et al. [37] observed a larger increase in calcium sensitivity of force with an increase in sarcomere length from 1.9 to 2.4 μm in diabetic rat hearts re-expressing fetal troponin T than in control hearts. They envisaged that the modified troponin T could be incorporated preferentially in the part of the thin filament closest to the Z-line and that consequently, at longer sarcomere length, the unmodified complex would retain control of cross-bridge interaction. This would suggest that alterations in the troponin T isoform composition might influence the length-dependency of force development. In human end-stage failing hearts, Anderson et al. [29] observed re-expression of the fetal troponin T4 isoform in all cases, while others found increased troponin T4 levels in only half [30] or even 10% [31] of their failing samples. In contrast, Townsend [32] et al. observed an increase in fetal troponin T1, while Mesnard et al. [30] found decreased troponin T1 protein expression in failing ventricles. The ventricular samples used in our study proved to be uniform in troponin T composition by one-dimensional gel electrophoresis (data not shown). Hence, possible heterogeneity in troponin T isoform composition among failing human hearts and in its distribution inside the sarcomere might provide an explanation for the discrepancies concerning the length-dependent force development in human end-stage failing myocardium. It should be noted, however, that the findings of Akella et al. [37] would imply an enhancement of the Frank–Starling effect. Furthermore, the heterogeneous expression of troponin T might be a transient phenomenon in humans.

Some caution should be exerted in the interpretation of our findings. One donor and two heart failure patients received positive inotropic support (Table 1), which may have altered the phosphorylation status of myocardial contractile proteins. Therefore, effects of medication on our results cannot be completely ruled out. Furthermore, because donor tissue is scarce, the number of experiments on donor hearts is limited. In addition, other limitations inherent in human myocardial studies (variability in age, sex) apply. Nevertheless, this study shows clear differences between donor and failing heart, while the effects of PKA were studied within the same preparations.

In summary, our observations on donor and end-stage failing human cardiomyocytes indicate that the Frank–Starling mechanism is preserved in end-stage heart failure, suggesting that loss of the Frank–Starling mechanism does not contribute to the decreased cardiac performance in congestive heart failure. Other cellular mechanisms which may contribute to the dysfunction of failing myocardium involve alterations in calcium handling, contractile protein composition and extracellular matrix [38]. Furthermore, our results suggest that the length-dependent alterations in force development are independent of the degree of phosphorylation of troponin I and C-protein.

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
View Abstract