|
|
||||||||
NEUROHUMORAL CONTROL OF CARDIOVASCULAR FUNCTION
1Center for Cardiovascular Research, Department of Physiology and Biophysics, University of Illinois at Chicago, Chicago, Illinois, and 2Department of Cardiology, Jeroen Bosch Hospital, 's-Hertogenbosch, The Netherlands
Submitted 18 December 2006 ; accepted in final form 14 March 2007
| ABSTRACT |
|---|
|
|
|---|
myofibrillar proteins; cross-bridges; sarcomere mechanics; laser diffraction; calcium; myosin ATPase; myofilaments
| METHODS |
|---|
|
|
|---|
Isolation and mounting of cardiac trabeculae. Right ventricular (RV) trabeculae were dissected as previously described (7, 10). Intact trabeculae were mounted in a glass-covered experimental chamber positioned on the stage of an inverted microscope (7); trabeculae were stimulated at 0.5 Hz. Skinned trabeculae were prepared by exposure to relaxing solution with 1% Triton X-100 at 4°C for at least 2 h (7, 9). Skinned trabeculae were then attached to aluminum T-clips and mounted in an experimental setup that allows for simultaneous measurement of calcium-activated myofilament force and ATP hydrolysis rate (9, 35, 36). Finally, following dissection of suitable trabeculae, the left ventricle (LV) was cut transversely at the base of the papillary muscles to measure its unstressed diameter. Residual LV and RV tissues were briefly blotted, weighed, flash-frozen, and stored at 80°C for later analyses by Northern and Western blot assays.
Solutions. The standard solution used for intact twitching trabeculae was composed of (in mM): 142.5 Na+, 5.0 K+, 127.5 Cl, 1.2 Mg2+, 2.0 H2PO4, 1.2 SO42, 21 HCO3, 10 D-glucose; [Ca2+] as indicated and equilibrated with 95% O2-5% CO2 (25.0 ± 0.1°C). For skinned fiber studies, three bathing solutions were used (20°C): a relaxing solution, a preactivating solution with low-calcium buffering capacity, and an activating solution, as previously described (9).
Measurement of twitch force and myofilament function.
Sarcomere length (SL) was measured by laser diffraction techniques (7). Muscle length was controlled by servomotor (Cambridge Technology); force was measured by semiconductor strain gauge (Sensonor). Stress, expressed as mN/mm2, was calculated as developed force normalized to cross-sectional area of the trabecula estimated from the width and thickness of the muscle; these were measured at
10-µm resolution either in the dissection dish (intact twitching trabeculae) or in the experimental apparatus (skinned trabeculae). On average, intact twitching trabeculae (n = 28) were 2.8 ± 0.1 mm in length, 275 ± 26 µm in width, and 127 ± 8 µm in thickness; skinned trabeculae (n = 28) were 1.7 ± 0.1 mm in length, 218 ± 11 µm in width, and 163 ± 9 µm in thickness. The rate of ATP hydrolysis was measured in skinned trabeculae over a range of contractile activation levels induced by varying concentrations of free calcium, as described previously (9).
Measurement protocols. Intact electrically stimulated trabeculae were stretched to maintain resting SL at 2.10 µm, while the extracellular calcium concentration ([Ca2+]o) was gradually increased to 1.5 mM followed by 1 h of stabilization. The active force-SL relationship was then measured, as previously described (10). Next, twitch force was measured over a range of [Ca2+]o at SL = 2.0 µm to evaluate the twitch force-[Ca2+]o relationship (7, 10). Skinned trabeculae were bathed in a series of solutions with varied activating calcium levels to compile the steady state contractile force-free [Ca2+] and force-ATPase relationships (9, 35, 36).
Northern blot mRNA analysis. Total RNA was isolated from RV tissues by acid guanidinium thiocyanate-phenol-chloroform extraction, separated on 1% formaldehyde-agarose gels (10 µg), and transferred to nylon membranes, according to standard protocols. The nylon membranes were successively hybridized with randomly primed 32P-labeled rat cDNA probes; human GAPDH (Y02278, Gibco), and rat 18S rRNA (Ambion) were used to control gel loading. All cDNA fragments were subcloned into cloning vectors, and their integrity was confirmed by direct sequencing. Digitized photographic films were analyzed using image processing software (NIH image). To allow for quantitative comparison between samples analyzed on separate gels, each gel also contained lanes loaded with RNA extract obtained from standard sample (a 7-wk-old rat ventricle; thus, an identical reference sample was included in each analysis). Linearity for each cDNA probe was confirmed in preliminary experiments.
Western blot protein analysis. Cardiac RV whole homogenates (20 µg total protein) were separated by SDS-PAGE and transferred onto polyvinylidene difluoride membranes using standard protocols. Total protein content was determined by the bicinchoninic acid method (Pierce). Membranes were incubated with an excess amount of antibody (in separate runs): monoclonal sarcoplasmic reticulum calcium pump (SERCA2; 2A7-A1, ABR), polyclonal sodium-calcium exchanger (NCX; pai11-13, Swant), monoclonal ryanodine receptor (RyR2; C3-33, ABR), or monoclonal phospholamban (PLBm; 2D-12, ABR); incubation with the appropriate horseradish peroxidase-conjugated second antibody was done according to the supplier's instructions. Next, the membranes were incubated in chemiluminescence substrate (Pierce), and digitized photographic films were analyzed using image processing software (NIH image). Again, an identical reference sample was included in each analysis; linearity of each antibody used was confirmed in preliminary experiments.
-
myosin heavy chain distribution analysis.
The isoform distribution of myosin heavy chain (MHC) was estimated using SDS-PAGE as described previously (36). Briefly, whole RV homogenates (510 µg total protein) were separated using a 6% resolving and 5% stacking gel at 45 mA for 6.5 h at 4°C. Gels were stained with Coomasie brilliant blue R-250 and digitally scanned. Images were analyzed by commercially available software (Kodak 1D) to measure relative densities of
-MHC and
-MHC.
Data analysis. Statistical significance was tested by one- or two-way ANOVA; P < 0.05 was considered significant. Data are expressed as means ± SE. Sigmoidal force-[Ca2+] relations were fit using a nonlinear fit procedure to a modified Hill equation:
![]() | (1) |
where F is peak twitch force (intact) or steady-state force (skinned), Fmax is the maximum saturated value F can attain, EC50 is the concentration of [Ca2+] at which F is 50% of Fmax, and H represents the slope of the force-[Ca2+] relation (the Hill coefficient). In skinned trabeculae, the ATP hydrolysis rate as a function of steady-state force over a range of free [Ca2+] was fit by linear regression, the slope of which represents tension-dependent myofilament ATP consumption rate (tension cost parameter; a direct measure of average cross-bridge cycling rate).
| RESULTS |
|---|
|
|
|---|
-myosin heavy chain), as well as an increase in "fetal genes" mRNA levels (skeletal
-actin,
-myosin heavy chain). Atrial natriuretic factor and brain natriuretic factor expression were also enhanced in both groups. Aging by itself, on the other hand, did not affect mRNA expression of these genes. In contrast, in either group, neither cardiac
-actin mRNA nor NCX mRNA levels were affected (results not shown). These qualitative Northern blot analysis results are consistent with the more quantitative results generally found during the development of HF. It is well established that cardiac dysfunction of diverse etiologies in the small rodent is associated with a shift in MHC expression to the beta isoform (8, 42), and this was also observed in the present study (cf. Fig. 2). Early HF resulted in a decrease in
-MHC mRNA, concomitant with an increase in
-MHC mRNA, and furthermore, these alterations were more pronounced in late HF. As expected, signals from a probe that hybridized with either
- or
-MHC mRNA showed no significant alteration during the development of HF. In addition, aging by itself was associated with a small, but significant increase in
-MHC mRNA, as has been observed previously (42).
|
|
|
1 mM) was not appreciably affected by the development of HF. Finally, at [Ca2+]o = 1.5 mM, the entire twitch force-sarcomere length relationship was markedly depressed both in the early phase (33%) and late phase (47%) of HF (results not shown). Thus, depressed myocardial function following myocardial infarction in both phases was associated not only with depressed maximum twitch force at saturating [Ca2+]o but also with a depression of the Frank-Starling relationship.
|
|
|
Relation of tension cost to
-myosin content.
HF and aging were associated with a shift in isomyosin expression from
- to
-MHC) (cf. Fig. 2). We have recently demonstrated that increased
-MHC content in rat myocardium correlates with reduced tension cost (35, 36). Therefore, the depression of Fmax and tension cost seen in the late HF group could be due merely to altered isomyosin composition. To investigate this possibility, we measured force and tension cost in a separate group of skinned trabeculae that were obtained from animals that were rendered hypothyroid (see METHODS), which resulted in the virtually exclusive synthesis of RV
-MHC (Fig. 5). Early HF was associated with a significant increase (16%) in RV
-MHC protein, which increased further to 39%
-MHC in end-stage HF (39%). Aging by itself was associated with a small (4%), but significant, increase in RV
-MHC content. Neither maximum force nor calcium sensitivity was affected by hypothyroidism, as reported by us previously (10, 36). Tension cost was significantly depressed in the late HF group to levels below that of the hypothyroidism group, despite the fact that the hearts still contained
60%
-MHC (Fig. 5, middle). This finding is further illustrated in Fig. 5, bottom, where tension cost is plotted as a function of
-MHC content for all the study groups. Relative to non-HF muscle groups (solid line), tension cost in relation to
-MHC content was
55% depressed in the late HF group.
|
| DISCUSSION |
|---|
|
|
|---|
Depressed twitch function. The twitch function of isolated trabeculae was depressed at both stages of HF, a result that is consistent with our previous observations in this model (7, 10). Depressed function manifested as a reduction in twitch force at saturating levels of Ca2+ in the bathing medium (Fig. 3, top). It is interesting to note that at an intermediate and, thus, more physiological concentration of Ca2+, twitch force was virtually identical between the control and HF groups. This was due to a significant increase in the sensitivity to extracellular Ca2+ at both stages of HF, but particularly at end-stage HF. Despite the clear defect in myocardial E-C coupling, physiological twitch force, and, as a consequence, we presume ventricular pump function, may well have been preserved in our HF rats. A similar observation has also been reported for another small rodent model of HF, the spontaneous hypertensive HF rat (32). The mechanisms that underlie this phenomenon cannot directly be determined from our study. HF was associated with a decrease in SERCA both at the mRNA (Fig. 1) and protein level (Fig. 4), as has been described previously in both human and animal HF (3, 26, 38), albeit not universal (8). In addition, reduced phosphorylation of phospholamban has been reported in HF (1, 31, 47). It is reasonable to conclude that the pool of calcium available for release by the SR is depressed in the HF myocyte. It has been established that action potential duration is prolonged in HF in general (6, 33), and this has indeed also been observed in the rat myocardial infarct model (46). Hence, in HF the source of calcium that activates the cardiac myofilament is shifted from the intracellular store to sarcolemmal calcium influx during the action potential, a process that apparently saturates at a relatively low extracellular calcium concentration. The mechanisms underlying this are unknown but may involve increased Ca2+ channel function. In addition, the duration of the calcium transient has also been shown to be prolonged in HF in general (6, 33), as well as in the rat MI HF model (46), suggesting that myofilament activation may approach a condition that is closer to steady state and thus higher levels of activation (4, 32, 37, 46). Although beneficial for maintaining pump function in HF, the sustained force development comes at a "cost" in the form of a high sensitivity toward extracellular calcium Ca2+ and a prolonged calcium transient, which may increase the incidence of ventricular arrhythmias (6, 33). Another mechanism that may underlie the increase in sensitivity toward extracellular Ca2+ is an increase in the sensitivity of the SR calcium release process triggered by calcium influx, as has been suggested by the Marks group (45). It is important to note that we found no indication of altered RyR mRNA or protein levels, which is consistent with previous reports (26). A change in NCX function has been reported in some (15, 24) but not all (33, 39) studies. In our study, we found no significant changes in NCX protein content in early and late HF. Regardless of the underlying mechanisms, the enhanced sensitivity of twitch force to extracellular Ca2+ may function as a cellular compensation aimed to maintain twitch function in the face of depressed myofilament function in the late HF myocyte.
Depressed myofilament function. Myofilament function was depressed at end-stage HF but not in the early stage of HF (Figs. 3 and 5). This manifested itself as a depression of Ca2+-saturated myofilament steady-state force development and a reduction in cross-bridge cycling rate. Myofilament Ca2+ sensitivity on the other hand, was not affected by HF in the present study on isolated skinned cardiac trabeculae (cf. Table 2), consistent with previous studies (13, 17).
Previous studies in single skinned myocytes in contrast, both by others and us have reported either increases or decreases in myofilament Ca2+ sensitivity both in human and experimental HF (5, 22, 44). The reasons for these different findings are not entirely clear; one explanation for these different findings may relate to the different experimental preparations that were studied (multicellular vs. single myocyte skinned myocardium). It is unlikely that depressed myofilament function was due to a reduction in the cross-sectional myofibrillar protein content, resulting from a relative increase in extracellular matrix protein (fibrosis). First, our measurement of cross-bridge cycling rate is not affected by this confounding factor, since ATP hydrolysis rate is measured as a function of myofilament force development. Second, we found no evidence of increased passive force development in our skinned muscle preparations, as may be expected from a significant increase in fibrosis. Finally, we also observed decreased myofilament force development in single skinned myocytes in this HF model (5), an approach that altogether eliminates the extracellular matrix.
Heart failure and cardiac hypertrophy in small rodents are associated with a shift in isomyosin synthesis from predominantly
-myosin toward
-myosin (42). Our study confirms this observation (Figs. 2 and 5). Such a change in isomyosin composition does not occur in larger mammals such as the human, at least not to the extent as observed in the small rodent models (27). However, it is unlikely that the depressed myofilament force development that we observed at end-stage HF was the result of altered isomyosin composition. That is, we have consistently found that an alteration in isomyosin composition in the rat does not affect myofilament force development and calcium responsiveness (10, 36), which is in line with studies by others (14, 21). A new finding in the present study is the reduction in cross-bridge cycling rate that accompanied the depression of myofilament force development at end-stage HF. It is well established that
-myosin is associated with a reduction in cross-bridge cycling (35, 36). Therefore, reduced cross-bridge cycling may merely have been the result of altered isomyosin composition in HF. This was clearly not the case; tension-cost, an index of cross-bridge cycling, was higher in muscles composed exclusively of
-myosin than that observed in end-stage HF myocardium, despite the fact that end-stage HF RV tissue was composed of a mixture of
- and
-myosin (Fig. 5). In fact, cross-bridge cycling rate in end-stage HF was only about half the rate that would be expected based solely on MHC content. We have previously shown a linear relationship between tension-cost and the
/
-myosin ratio in both rat and mouse myocardium (25, 36); a similar result was also reported in guinea-pig myocardium, albeit over a smaller range of
/
-myosin ratio (43). Likewise, human ventricular myocardium displays a lower tension cost than human atrial myocardium that is largely explained by the difference in myosin composition (29). It should be noted that the tension-cost parameter is measured during an isometric contraction, that is, under conditions in which the cross-bridges are under high strain and developing force. A linear relationship between tension-cost and
/
-myosin ratio implies independent cycling of cross-bridges, generating force and consuming ATP independently. Although measurement of thin filament sliding speed (an unloaded, zero-strain situation) has demonstrated a curvilinear relationship between sliding velocity and
/
-myosin ratio (19), there is still a general decline in cross-bridge cycling rate in proportion to an increase in
-myosin content, even under those conditions. However, since here, we measured isometric contractions, it is appropriate to compare tension-cost in relation to the
/
-myosin ratio. Reduced cross-bridge cycling rate has also been observed in human end-stage HF (12, 34). What could be the mechanism of reduced cross-bridge cycling rate in end-stage HF? A reduction in myofibrillar ATP hydrolysis rate has consistently been observed in human end-stage HF, but only for calcium-regulated actin-activated myosin ATPase activity (2, 30). Hence, those results suggest that reduced cross-bridge cycling rate in HF has at its basis an alteration in thin filament structure. Indeed, alterations in troponin-T isoform distribution have been reported in both human and animal models of hypertrophy and HF (8). Alternatively, posttranslational modification of thin-filament proteins, in particular, protein phosphorylation, may underlie this phenomenon (40). In particular, PKC may be involved in this pathology since this kinase is known to be activated in HF (8, 40). Furthermore, treatment of myocardial infarct rats with losartan, an antagonist of the angiotensin receptor that is coupled to PKC, prevented reduced function of isolated RV trabeculae in this HF model, at least at the early stage of HF (7). Finally, PKC-induced contractile protein phosphorylation is associated with decreased cross-bridge cycling rate (8, 40, 41). It has recently been suggested that PKC-
is upregulated in HF (18, 23, 40), and this kinase may mediate the changes in myofilament function that we observed in HF, particularly via phosphorylation of troponin in heart failure (5). However, it should be noted that the underlying molecular mechanisms that may be responsible for the reduced cross-bridge cycling rate observed in this model of HF could not be directly determined from our present study.
Limitations.
Heart failure was induced by infarction of the left ventricle. We adopted this small rodent model of HF because the RV of rats can provide thin and homogeneous cardiac trabeculae. This feature is essential to ensure both the metabolic stability of the preparation and to allow for the measurement and control of sarcomere length by laser diffraction techniques (7, 10). Furthermore, by studying trabeculae from the right ventricle, we ensured that only noninfarcted tissue was studied. Nevertheless, by studying the RV, important changes in the LV during the development of HF may have gone unnoticed in this study. The finding of depressed function of isolated right ventricular myocardium may seem surprising since congestive heart failure was induced by left ventricular infarction. An explanation may be that, in this HF model, both the right and left ventricle are exposed to increased mechanical load and circulating neurohormonal stimuli during the development of HF. We employed hypothyroidism to induce synthesis of
-myosin. Hypothyroidism potentially affects many aspects of cardiac myocyte biology. However, in the present study a hypothyroid model was used only for the purpose of comparing tension-cost between sham and HF myofilaments. Neither myofilament maximum force, calcium sensitivity, or level of cooperativity is altered between euthyroid and hypothyroid derived myocardium (10, 14, 21, 25, 36), strongly suggesting that, as far as the contractile proteins are concerned, hypothyroidism only affects myosin isoform expression and, thereby, cross-bridge cycling rate; we would have expected a change in these functional parameters should hypothyroidism have caused other changes in myofilament function apart from cross-bridge cycling rate under our conditions (23, 40). Finally, our study was performed at 25°C, at a relatively slow heart rate (0.5 Hz), and in the absence of catecholamine stimulation. Therefore, the contractile function that we measured in the isolated RV trabeculae may not reflect the in situ ventricular function that was present in the animals under physiological conditions.
In conclusion, we investigated myocardial contractile function in isolated cardiac trabeculae during the development of HF in the rat myocardial infarct heart failure model. Both early and late HF were associated with depressed twitch function, whereas myofilament function was depressed only at end-stage HF. Depressed myofilament function manifested itself as a depression of force development and cross-bridge cycling rate. Our results suggest a hierarchy of cellular events during the development of HF, starting with altered calcium homeostasis during the early phase followed by myofilament dysfunction at end-stage HF. We speculate that the latter may be coincident with the transition to overt end-stage decompensated HF in this model.
| GRANTS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. J. Duncker, N. M. Boontje, D. Merkus, A. Versteilen, J. Krysiak, G. Mearini, A. El-Armouche, V. J. de Beer, J. M.J. Lamers, L. Carrier, et al. Prevention of Myofilament Dysfunction by {beta}-Blocker Therapy in Postinfarct Remodeling Circ Heart Fail, May 1, 2009; 2(3): 233 - 242. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. S. Dhalla, H. K. Saini-Chohan, D. Rodriguez-Leyva, V. Elimban, M. R. Dent, and P. S. Tappia Subcellular remodelling may induce cardiac dysfunction in congestive heart failure Cardiovasc Res, February 15, 2009; 81(3): 429 - 438. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. M. Shioura, D. L. Geenen, and P. H. Goldspink Sex-related changes in cardiac function following myocardial infarction in mice Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2008; 295(2): R528 - R534. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. T. McCloskey, S. Turcato, G.-Y. Wang, L. Turnbull, B.-Q. Zhu, T. Bambino, A. P. Nguyen, D. H. Lovett, R. A. Nissenson, J. S. Karliner, et al. Expression of a Gi-coupled receptor in the heart causes impaired Ca2+ handling, myofilament injury, and dilated cardiomyopathy Am J Physiol Heart Circ Physiol, January 1, 2008; 294(1): H205 - H212. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. M. Shioura, D. L. Geenen, and P. H. Goldspink Assessment of cardiac function with the pressure-volume conductance system following myocardial infarction in mice Am J Physiol Heart Circ Physiol, November 1, 2007; 293(5): H2870 - H2877. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |