| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cellular Biology |
From the Division of Experimental Cardiology (V.B., K.R.S.), Division of Cardiac Imaging (P.C., C.D, B.B.), Laboratory of Pathology (A.V.L., E.V.), and Department of Nuclear Medicine (L.M), University Hospital Gasthuisberg and University of Leuven, Belgium; Laboratory for Physiology (J.v.d.V., G.S.), VU University Medical Center, Amsterdam, Netherlands.
Correspondence to Karin R. Sipido, MD, PhD, Division of Experimental Cardiology, KUL, Campus Gasthuisberg O/N 7th floor, Herestraat 49, B-3000 Leuven, Belgium. E-mail Karin.Sipido{at}med.kuleuven.ac.be
| Abstract |
|---|
|
|
|---|
Key Words: hibernation contractile function myocytes calcium
| Introduction |
|---|
|
|
|---|
The prevalence of hibernating myocardium in the population with CAD is high, and it is a challenge for treatment.7,8 Indeed, without revascularization, mortality in CAD patients with hibernating myocardium is higher than in CAD patients without viable myocardium.9 The mechanisms underlying the reduced contractile function and increased mortality remain incompletely understood. Both changes in the local environment in vivo and intrinsic remodelling of the myocytes are likely to contribute.
Studies on human biopsies of hibernating myocardium from patients undergoing bypass surgery have focused on the structural changes and altered gene regulation. Hibernating myocardium contains a variable number of cells with altered ultrastructure, with loss of myofibrillar proteins, swollen mitochondria and glycogen accumulation.1012 These changes have been interpreted as cell de-differentiation with altered gene expression and reexpression of fetal structural genes10,13 but also as signs of degeneration.11 Apoptosis and compensatory hypertrophy are also observed, although there is some controversy about the extent and significance of apoptosis.10,1315 There are no functional studies in vitro on human tissue or cells from hibernating human myocardium.
Despite the difficulties involved in exactly reproducing hibernation in humans, a number of valid animal models have been developed (reviewed in16). Canty, Fallavollita and coworkers have characterized a pig model in which coronary stenosis develops slowly during the animal growth over several months.17,18 SERCA, phospholamban and RyR2 mRNA and protein expression were reduced, but troponin I (TnI) was not degraded.19,20 These authors also observed apoptosis and compensatory cellular hypertrophy.21 The functional characteristics of the hibernating myocardium in those models have been studied in vivo, but there are no in vitro data on isolated tissues or cells that characterize the functional remodeling.
We recently described a closed-chest model of hibernation in the pig where we isolated myocytes to study function in vitro.22 We could thus demonstrate that the decreased function in vivo is at least partly related to intrinsic remodeling of the myocytes. Contraction of the myocytes was blunted, the L-type Ca2+ current, ICaL, and Ca2+ release from the sarcoplasmic reticulum were reduced. Yet contraction amplitude of the intact cells could not be restored by increasing available Ca2+, suggesting there was additional impairment of myofilament function. We therefore in the present study use myocyte preparations devoid of membranes to characterize directly the myofilament function. We observe a reduced maximal force development that is not related to protein isoform shift or phosphorylation but results at least to a large extent from a reduction of myofilament density and ultrastructural reorganization.
| Materials and Methods |
|---|
|
|
|---|
Myocyte Isolation and Skinning
Single skinned myocytes were prepared from the frozen tissue samples as described previously23 and detailed in the online data supplement available at http://circres.ahajournals.org; half of the sample was used for myocyte preparation and half of the sample for protein studies.
Measurement of Isometric Force
In skinned myocytes, the myofilaments are directly activated by the [Ca2+] of the bathing solution. The experiments were performed as described previously,23 and detailed in the online supplement. Briefly, isometric force was measured in the skinned myocytes attached between a piezoelectric motor and a force transducer (Figure 1A, B) at 15°C at a sarcomere length set at 2.2 µm. The isometric force developed was normalized to the cross-sectional area of the myocyte. ForcepCa relations were fit with a modified Hill equation:
|
F/F0=[Ca2+]nH/(Ca50nH+[Ca2+]nH)
Rate of force redevelopment (Ktr) was measured during steady state activation, at different pCa; sampling rate was at 1 kHz. After full force development, the myocyte length was suddenly reduced by 20% for 3 ms. The restretching of the preparation to its original length induces force redevelopment to the maximal force (Figure 1C). The passive force was measured in relaxing solution.
Protein Expression and Phosphorylation
Tissue homogenates were prepared from freeze-dried tissue, obtained from the same samples used for myocyte isolation, and diluted in sample buffer to a concentration of 1 µg/µL (dry weight/volume). Protein concentration was determined by using a Bio-Rad RC DC protein assay, calibrated with bovine serum albumin.24
The procedure for 2-dimensional polyacrylamide gel electrophoresis was as used in25 and is detailed in the online data supplement. The protocols for specific Western blots and for measuring the degree of phosphorylation can likewise be found in the online supplement.
In a previous study26 the phosphorylation status in a tissue homogenate was compared with that in mechanically isolated Triton-treated cardiomyocytes prepared from the same tissue. The 2D-gels did not reveal differences in the phosphorylation status of myofilament proteins between the 2 sample types.
Histological Data
For light microscopy, the samples were fixed in 4% paraformaldehyde, embedded in paraffin and sectioned at 4 µm thickness. For electron microscopy, tissue samples were minced into 1 mm3 pieces and fixed in 2.5% glutaraldehyde. The further procedures are detailed in the online data supplement available at http://circres.ahajournals.org.
The methods for determining volume density of myofilaments and mitochondria and for glycogen quantification are detailed in the online supplement.
Statistical Data
All the data are shown as mean±SEM. According to the type of experiments, we used either paired, unpaired t-test or 2-way ANOVA followed by post-hoc Bonferroni test. The data were considered significant for P<0.05.
| Results |
|---|
|
|
|---|
|
The rate of force redevelopment (Ktr) was determined at pCa between 4.5 and 5.6. Ktr decreases with a decrease in [Ca2+]o, but this relation was comparable between CTRL and HIB, suggesting that the cross-bridges kinetics are not altered in HIB (Figure 2D). The relation between Ktr and active force was linear in CTRL and HIB, approximately parallel, showing a leftward shift in HIB (not shown) as expected from a proportional decline in force generating capacity of the cells.
The isolation procedure of the current study was different from our previous study where the myocytes were enzymatically isolated. To test for a selection bias and to check we were studying a similar population of myocytes, we also performed the skinning procedure on myocytes obtained by enzymatic isolation. We compared tension-pCa curves of myocytes obtained with the 2 different procedures and found no differences (see online supplement). Therefore, the results we obtained from skinned preparations can be related to the previous data we obtained in enzymatically isolated cells.
Is the Decrease in Maximal Force Development Because of a Change in Phosphorylation Status of the Myofibrillar Proteins?
We looked for potential changes in the phosphorylation status of myosin light chain 1 and 2 (MLC-1 and MLC-2) and troponin T (TnT), by analyzing 2-dimensional gel electrophoresis of homogenates. These were prepared from the same biopsies that were used for preparation of skinned myocytes; each frozen sample was divided in 2 parts, 1 for myocyte preparation and the other for protein studies. Figure 3A is an example of a 2D-gel from a CTRL sample. The data are summarized in Figure 3B. We found no evidence for changes in MLC-1, MLC-2 and TnT phosphorylation in HIB. We excluded degradation in the samples by measuring the ratio of MLC-1/MLC-2 and TnT/actin, which were unchanged (Figure 3C).
|
We further investigated TnI phosphorylation by using specific antibodies against total TnI and dephosphorylated TnI. The results are expressed as the ratio of dephosphorylated TnI to total TnI. As summarized Figure 4A, we found no difference between CTRL and HIB (ratio 22B11/16A11=0.44±0.10 in 4 CTRL versus 0.46±0.14 in 4 HIB).
|
With the Pro-Q Diamond staining method (see online supplement) we evaluated the degree of phosphorylation of myosin binding protein C (MyBP-C), reported to be altered during ischemia/reperfusion.27 The values for phosphorylated MyBP-C did not differ between CTRL (n=5; 0.076±0.004) and HIB (n=5; 0.071±0.007) samples (Figure 4B). With the same approach we also could not detect differences in total TnI phosphorylation (HIB 0.04±0.01 versus CTRL 0.06±0.02, n=5 each). In separate samples we checked total titin phosphorylation and saw no difference (HIB, n=2; 0.33±0.01 versus CTRL, n=2; 0.32±0.01).
We also looked for TnI degradation by immunoblot and quantification of the 25 kDa band of degraded TnI, as described before.22 In the present samples we again could not detect a significant increase in TnI degradation (data not shown).
To evaluate the cytoskeletal component we measured expression of desmin,
-actinin in relation to actin. We found no difference between HIB and CTRL (Figure 4C). In separate gels we also evaluated titin isoform expression and found that the ratio of NB2A:NB2 was close to 1 and not different between groups (data in online supplement).
For a potential shift in isoform of myosin heavy chain we analyzed 1D silver-stained protein gels. In both groups we observed only one band, corresponding to the ß-isoform of the myosin heavy-chain. From the same gels, the TnC/actin ratios were measured and found similar for CTRL and HIB (Figure 4D).
Is There a Global Reduction of Myofibrillar Protein Density?
For the 2D-gel electrophoresis and immunoblots, equal amounts of protein were loaded onto the gels. This approach would not allow detection of a global reduction of myofibrillar protein density in the myocytes. As myofibrillar proteins make up the majority of total myocyte protein, normalization for total protein concentration of the tissue samples could obscure potential differences. We therefore measured total protein content per volume, as an indicator for the density of myofibrillar protein in the tissue (Figure 5A). Total protein concentration was significantly reduced in HIB compared with CTRL (P<0.05). In another approach we examined the relation between total protein concentration in homogenates and the force developed by the myocytes isolated from the same hearts. This relation showed that the low force developed in HIB corresponded with low protein content (Figure 5B).
|
Ultrastructural Changes
The data above suggest that there might be a global reduction of myofibrillar protein density, and we therefore examined the tissue for ultrastructural changes. In the light microscopy sections with PAS staining, we saw in tissues from HIB a large fraction of cells that had clear accumulations of glycogen, whereas this was seen to a much lesser extent in CTRL (Figure 6A).
|
Studies of thin sections with electron microscopy confirmed the glycogen accumulation and showed in addition alterations in mitochondrial structure (Figure 7A). Quantification revealed a decrease in myofilament density in HIB (57.1±0.9% of surface area in HIB versus 69.9±1.9% in CTRL, P<0.05) and an increase of the area occupied by mitochondria (34.4±0.5% of surface area in HIB versus 29.7±1.2% in CTRL, P<0.05) (Figure 7B). The above data are consistent with the hypothesis that the amount of myofilaments per unit of volume is decreased. We also quantified glycogen accumulation using a semiquantitative scoring (see online supplement) confirming the higher glycogen content in HIB.
|
In our previous analysis of enzymatically isolated single cells, we noticed that cell length and width were significantly increased.22 We also measured cell width in the PAS stained sections and it was significantly larger in HIB (25.4±0.7 µm versus 22.1±0.7 µm in CTRL, P<0.05, Figure 6B). In the isolated skinned myocytes, we also found an increase in primary cell dimensions (Figure 8A), statistically significant for length, width, thickness (P<0.05), and also for the calculated cross-sectional area (Figure 8B).
|
| Discussion |
|---|
|
|
|---|
The latter was a somewhat surprising finding as we had expected alterations in sensitivity, potentially result from changes in phosphorylation of TnI, as previously described.24,26 The 2D-gel electrophoresis and specific immunoblots for TnI phosphorylation however did not reveal changes in phosphorylation, consistent with the functional data. There was no evident shift in isoform expression either and changes in MyBP-C27 were also not observed.
A potential explanation for the reduction of maximal force could be degradation of TnI.28 As we had previously reported, we could not detect an increase in TnI degradation, but it must be stressed that the immunoblot approach may not be sensitive enough to pick up a modest increase that may nevertheless be of functional relevance. The recent development of assays to directly quantify oxidation of myofibrillar protein29,30 also warrants further study of the potential role of oxidative modifications.
Because passive and restoring forces are mainly due to titin31 our findings of unaltered passive force and rate of force redevelopment in HIB, suggested that titin function was preserved. Protein analysis confirmed the absence of an isoform switch or changes in phosphorylation. This is in line with recent studies linking isoform switches in titin to changes in passive force in cardiomyofibrils and single skinned cardiomyocytes,32,33 and studies showing that PKA-mediated phosphorylation of titin alters passive force.34,35
Our current data suggest that the reduced force development is mainly the result of a decrease in myofilament volume density. The loss of force development per cross-sectional area in individual myocytes is around 35%. With the current techniques it is difficult to quantify the reduction in myofibrillar protein density and relate it to the reduction in force development, but we have some indicative measurements. The difference in protein concentration of homogenates is 23% whereas the quantification of the electron micrographs indicates a reduction of density of myofilaments of 19%. Additional factors are therefore likely to contribute to the reduced contractile function. Potential factors are a small increase in TnI degradation or loss of TnT, which could escape detection in the immunoblotting but could have major effects on a background of reduced myofilament density, and oxidative modifications of proteins.
A recurrent question and discussion point with regard to the use of isolated cells is whether the myocytes are representative for the entire myocardium. It is indeed a limitation of the technique that only a small number of myocytes can be studied per sample, and there is the possibility that cells from the hibernating myocardium with more extensive changes might be less resistant to the isolation procedure. We looked at the relative yield of myocytes during isolation, considering that a selective loss of the more damaged cells would lower the ratio of viable to total number of myocytes. As we could not observe differences between HIB and CTRL, we assume that our isolation is not biased in this way. We cannot examine glycogen loading in the same cells as used for functional studies, but in two HIB hearts we stained a sample of isolated cells and could see abundant glycogen in a number of cells (data not shown), suggesting that the isolation is not skewed toward less affected myocytes.
The histology data are consistent with previous data in humans,3 and those data predicted a loss of contractile function. However, our data are the first to actually measure force development in isolated myocytes.
The Role of Hypertrophy and Reduced Myofilament Density
We previously reported that the myocytes isolated from HIB were hypertrophied, in line with histological studies on tissue from human hearts and from animal models. The cells used in the current study again were larger than their controls. The increase in cell diameter may actually partially compensate for the reduction in myofilament density, as the amount of force per cell is not significantly smaller in HIB. It is conceivable that the synthesis of contractile proteins is deregulated in relation to the total increase in cell size.
Glycogen loading and alterations in mitochondria occupying a larger fraction of the cytosol may also lead to reduced myofilament density. These alterations are likely to result from activation of a complex signaling, including activation of the hypertrophic response through increased loading by the surrounding myocardium. A recent study linked activation of the p38-MAPK axis36 to increased glucose uptake, a factor contributing to glycogen loading. Glycogen synthesis is further promoted by a reduction in the GSK-3ß activity.37 As the flow reserve is significantly reduced in hibernating myocardium, intermittent episodes of acute ischemia may in addition provide a unique stimulus for remodelling, aimed at enhanced survival.38,39
Relation to Previous Findings
Our observation of a reduction in maximal force is in line with the data of Heusch and colleagues who measured a reduction in Ca2+ responsiveness of contraction in situ in their short-term model for hibernation.40 Such changes were not investigated in the chronic hibernation model, but there alterations in Ca2+ handling, namely a reduction in SERCA and RyR were seen.19 In our model we reported a reduction in Ca2+ current density.22 This altered Ca2+ handling will further exacerbate the myofilament defect. Furthermore, subcellular heterogeneity in Ca2+ handling may be present and potential alterations in T-tubule density could contribute to enhanced dyssynchrony.41,42
Relation to In Vivo Function
In vivo, the rate of pressure development in the HIB animals was significantly reduced.22 Analysis of the local wall deformation by strain rate indicated that the rate of wall thickening was reduced. Our measurements of the rate of force redevelopment on relengthening could not detect any differences between HIB and CTRL, indicating that the intrinsic crossbridge kinetics were not different. In intact cells however, the rate of shortening was reduced.22 This suggests that the slowing of contraction is related to the changes in excitation-contraction coupling. In vivo, enhanced loading imposed by the surrounding normal myocardium is also an important factor which may further exacerbate the extent of contractile dysfunction of the hibernating myocardium.
Limitations of the Study
In our study we examined the region of interest only and compared with the equivalent region in normal controls. With this approach we avoided confounding factors such as regional differences in function and the presence of remodelling in the remote myocardium. Indeed, whereas the observed differences can explain changes in local contractile function, the overall changes in contractile function of the left ventricle may also be determined by changes in the remote myocardium.12
| Conclusions and Perspectives |
|---|
|
|
|---|
| Acknowledgments |
|---|
This study was supported by grants to K.R.S from the FWO, the Fund for Scientific Research Flanders (G. 0166.03N), and the European Union (LSHM-CT-2005018833, EUGeneHeart). V.B. is the recipient of a ISHR (International Society for Heart Research) Research Fellowship.
Disclosures
None.
| Footnotes |
|---|
Original received February 19, 2006; resubmission received November 30, 2006; accepted January 4, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W. Streb, M. Marciniak, P. Claus, A. Marciniak, M. McLaughlin, J. D'hooge, F. E. Rademakers, B. Bijnens, and G. R. Sutherland Full or pressure limited reperfusion of an acute myocardial infarct results in a different wall thickness and deformation of the distal myocardium - implications for clinical reperfusion strategies Eur J Echocardiogr, July 1, 2008; 9(4): 458 - 465. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. F. Kelly, W. Sluiter, and E. O. McFalls Hibernating Myocardium: Is the Program to Survive a Pathway to Failure? Circ. Res., January 4, 2008; 102(1): 3 - 5. [Full Text] [PDF] |
||||
![]() |
B. Page, R. Young, V. Iyer, G. Suzuki, M. Lis, L. Korotchkina, M. S. Patel, K. M. Blumenthal, J. A. Fallavollita, and J. M. Canty Jr Persistent Regional Downregulation in Mitochondrial Enzymes and Upregulation of Stress Proteins in Swine With Chronic Hibernating Myocardium Circ. Res., January 4, 2008; 102(1): 103 - 112. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Research Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |