Am J Physiol Heart Circ Physiol 291: H1003-H1014, 2006.
First published May 5, 2006; doi:10.1152/ajpheart.00132.2006
0363-6135/06 $8.00
INVITED REVIEW
CALL FOR PAPERS
Cytoskeletal Networks and the Regulation of Cardiac Contractility
Cytoskeletal networks and the regulation of cardiac contractility: microtubules, hypertrophy, and cardiac dysfunction
George Cooper, IV
Gazes Cardiac Research Institute, Cardiology Division, Medical University of South Carolina, and Department of Veterans Affairs Medical Center, Charleston, South Carolina
 |
ABSTRACT
|
|---|
The cytoskeleton as classically defined for eukaryotic cells consists of three systems of protein filaments: the microtubules, the intermediate filaments, and the microfilaments. In mature striated muscle such as the heart of the adult mammal, these three types of cytoskeletal filaments are superimposed spatially on the myofilaments, a specialized system of contractile protein filaments. Each of these systems of protein filaments has the potential to respond in an adaptive or maladaptive manner during load-induced hypertrophic cardiac growth. However, the extent to which such hypertrophy is compensatory is also critically dependent on the type of hemodynamic overload that serves as the hypertrophic stimulus. Thus cardiac hypertrophy is not intrinsically maladaptive; rather, it is the nature of the inducing load rather than hypertrophy itself that is responsible, through effects on structural and/or regulatory proteins, for the frequent deterioration of initially compensatory hypertrophy into the congestive heart failure state. As one example reviewed here of this load specificity of maladaptation, increased microtubule network density is a persistent feature of severely pressure-overloaded, hypertrophied, and failing myocardium that imposes a primarily viscous load on active myofilaments during contraction.
myocardial contraction; cytoskeleton
BUILDING ON PIONEERING WORK of physiologists such as A. V. Hill and A. F. Huxley that gave us our initial insights into the mechanisms of contraction of skeletal muscle, there was an explosion of knowledge defining cardiac muscle contraction and its control from the National Heart Institute (Bethesda, MD) beginning in the middle of the last century. This knowledge base made it possible, for the first time, for physiologists and physicians at the National Heart Institute, such as S. J. Sarnoff, E. Braunwald, and E. H. Sonnenblick, to ask whether there are settings in which heart failure is caused by primary myocardial defects. For hemodynamic overloading of the heart, this question was answered in the affirmative, since a sustained myocardial pressure overload caused progressive decrements in contractile performance per unit mass of hypertrophied myocardium (1).
My own much smaller part in this story, which has led to the discovery reviewed here of the role of microtubules in cardiac dysfunction, begins at this point. The contributions of the many other investigators who have been involved in this science have been reviewed along the way (8–10), and their contributions to the microtubule aspect of this story are the subject of a recent review (11) and of an upcoming book chapter (12). Also found there (11) is a fully developed discussion of the early controversy generated by the microtubule hypothesis, especially in terms of the specific pathological settings wherein cardiac microtubule network densification is found.
Thus my intent here is not to provide yet another, and redundant comprehensive review of this subject area. Instead, to best describe the excitement of this adventure as it evolved, I will tell this story as a personal account from the point of view of my collaborators and myself.
|
MYOCARDIAL MECHANICS AND ENERGETICS
|
|---|
Because it was now established that myocardial pressure overloading causes clear decrements in contractile function (1), the next logical question was whether this is a universal consequence of cardiac hemodynamic overloading or, rather, whether intrinsic cardiac defects appear only in more specific abnormal hemodynamic settings. Clinical experience certainly suggested that the latter would be the case, and the first studies in which I was involved bore this out (5–7). Figure 1 describes the intrinsic contractile properties of isolated right ventricular (RV) papillary muscles in terms of the classic force-velocity and length-tension relationships. Pressure overloading caused clear abnormalities in both of these relationships, but an equivalent degree and duration of hypertrophy caused by volume overloading did not result in any detectable contractile abnormality. Thus it was indeed the nature of the inducing stress, rather than the hypertrophy process itself, that was responsible for the contractile defects caused by myocardial hemodynamic overloading. A second important fact brought out by these early studies, as also shown in Fig. 1, is that the myocardial changes responsible for contractile dysfunction after pressure overloading are not fixed structural or biochemical defects, because removal of the pressure overload restored contractile dysfunction to normal.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1. Force-velocity (left) and length-tension (right) relationships for right ventricular (RV) papillary muscles from 4 groups of cats: control, pressure overload, volume overload, and pressure unload. Pressure overload was induced by pulmonary artery banding, volume overload by atrial septum resection, and pressure unloading by removal of the pulmonary artery band after hypertrophy was induced. In each case, 4–6 wk elapsed before the next intervention or final study. RV-to-body weight ratio (g/kg) was 0.59 ± 0.03 for control, 1.05 ± 0.06 for pressure overload, 0.97 ± 0.03 for volume overload, and 0.56 ± 0.02 for pressure unload. For the length-tension relationship, the upper set of curves describes developed tension, and the lower set describes passive tension. LMax, that muscle length at which maximum active tension is generated. (Adapted from Refs. 5–7.)
|
|
As part of these early studies of cardiac muscle mechanics, simultaneous measurements were made of myocardial energetics as quantified by myocardial O2 consumption. In this way, energy utilization could be related directly to mechanical output. Because a reduction in cardiac mechanical performance is ordinarily associated with a corresponding reduction in energy utilization, the finding in Fig. 2 was rather curious. That is, for pressure-overload RV hypertrophy, where the rate of myocardial shortening and the extent of myocardial tension generation were greatly reduced during contractions at LMax when the length-tension relationship given in Fig. 1 was being defined, there was an unexpected increase rather than the expected decrease in myocardial O2 consumption. As also shown in Fig. 2, this energetic abnormality reverted to normal in parallel with the contractile abnormality when the pressure overload was removed. This reversible energetic abnormality, seen in pressure overload, but not volume overload, hypertrophy, puzzled me and others for a good while. However, it later came to provide, as will be seen in what follows, an essential hint that would open the microtubule portion of this story.
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 2. Relation between myocardial O2 consumption (M O2) and developed tension at LMax point for papillary muscles from control, pressure-overloaded, volume-overloaded, and pressure-unloaded cats described in Fig. 1. *P < 0.001 vs. control (by 1-way ANOVA followed by Scheffé's S procedure). (Adapted from Refs. 5–7.)
|
|
|
CARDIAC MUSCLE CELL MECHANICS
|
|---|
Up to this juncture, work on cardiac hypertrophy had employed experimental models ranging in complexity from intact animals to isolated myocardium. Development of techniques that allowed for the reproducible isolation of viable cardiocytes from the hearts of adult mammals provided the opportunity (8) to take the search for the etiology of the contractile dysfunction of myocardium hypertrophying in response to a pressure overload to the muscle cell itself. This was important because there was no reason to assume that such contractile defects were not, instead, caused in whole or in part by the structural and/or biochemical environment of the cardiocyte.
Two further steps were required, however, before this search could begin: 1) development of methods that allowed afterloaded contractions of isolated cardiocytes to be measured at the level of sarcomere motion within these cells and 2) use of these methods to determine whether contractile abnormalities defined at the tissue level were reproduced in cardiocytes isolated from the original models (5–7) of RV pressure-overload and volume-overload hypertrophy. Figure 3 illustrates the application of the methods that we developed (17, 20) to cardiocytes from the two relevant models of cardiac hypertrophy. In essence, the data that originally described isotonic contractile dysfunction in papillary muscles from the pressure-overloaded RV (Fig. 1) were reproduced in cardiocytes from this myocardium (Fig. 3). Furthermore, Fig. 4 shows that even in a different RV volume-overload model that was of such severe extent and prolonged duration that it caused an 
50% mortality rate from congestive heart failure, normal cardiocyte sarcomere motion was retained (15).
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 3. Sarcomere force-velocity (top) and force-shortening (bottom) relationships for RV cardiocytes from the same feline pressure-overload model characterized in Figs. 1 and 2. RV-to-body weight ratio (g/kg) was 0.58 ± 0.10 for control and 1.20 ± 0.10 for pressure overload. Sarcomere motion was measured at 37°C, 2.5 mM Ca2+, and 0.25-Hz stimulation frequency at a sampling rate of 1 kHz via our laser diffraction system; superfusate viscosity was used as a surrogate for the external force used in Fig. 1. In contrast to Fig. 1, this is a semilogarithmic plot, such that these hyperbolic relations are linearized. (Adapted from Refs. 17 and 20.)
|
|
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 4. Sarcomere motion sampled at a frequency of 1 kHz during contractions of single RV cardiocytes. Left, a cell from a control dog; right, a dog with severe tricuspid regurgitation for 3.5 yr and overt clinical RV failure. Sarcomere mechanics were averaged for 10 contractions of each cardiocyte studied at 37°C, 2.5 mM Ca2+, and 0.25-Hz stimulation frequency. (Adapted from Ref. 15.)
|
|
|
MICROTUBULE HYPOTHESIS
|
|---|
The stage was now set to search for the intracellular impediment to contraction that was being generated during hypertrophic growth in response to a pressure overload. Extensive work in many laboratories (8, 10) defining a variety of biochemical defects in hypertrophied myocardium made it quite unlikely that a single abnormality uniquely responsible for contractile dysfunction would be found, and this would become even less likely as compensated hypertrophy degenerated into congestive heart failure. To proceed effectively, however, it was necessary to focus this search in terms of a specific, testable hypothesis that could account for the progressive decrement in contractile output per unit mass of pressure-hypertrophied myocardium that forms the myocardial basis for heart failure in this entity.
In generating this hypothesis, I first took the hint provided by the data given in Fig. 2. That is, rather than a biochemical defect, might a structural impediment to myofibrillar shortening appear during pressure-overload but not volume-overload hypertrophy? This could act as a viscous damper, such that part of the energy used for cross-bridge cycling would be dissipated within this damper, as suggested by Figs. 1 and 2, rather than generate active tension and external work. A serious problem with this idea, however, was that there are neither qualitative nor quantitative differences among normal and hypertrophied cardiocytes from our models of pressure-overloaded versus volume-overloaded RVs when defined in terms of standard ultrastructure (8). Therefore, any posited structural defect would have to 1) appear or increase in response to load, 2) discriminate between the stimuli of stress, or pressure loading versus strain, or volume loading, 3) not be obvious ultrastructurally, and 4) have the potential to interfere with sarcomere motion.
The hypothesis that evolved from all of this was that microtubule network densification might be such a culprit structural defect. In terms of the four criteria stated above, this was because 1) microtubules are known to form along cellular stress axes, 2) for a linear steady-state polymer such as the 
,
-tubulin heterodimer-microtubule system, thermodynamics would cause the subunit concentration for assembly to be lowered and polymer stability to be enhanced by an extending stress force, 3) microtubules are, in fact, not obvious by standard ultrastructural analysis in mature striated muscle, and 4) the intimate cardiac myofibrillar investment by microtubules implied that in excess they might well interfere with sarcomere motion.
Initial test of the hypothesis.
This hypothesis was tested (29, 30) in the same RV volume- and pressure-overload models that were used to generate the data shown in Figs. 1–3. The results are given in Fig. 5 for volume-overload hypertrophy and in Fig. 6 for pressure-overload hypertrophy. In the cats with volume-overload RV hypertrophy, there was no difference between the RV and the normally loaded left ventricle (LV) from the same animal in terms of cardiocyte microtubule network density, free or polymerized myocardial tubulin, or the extent and velocity of cardiocyte sarcomere shortening before or after microtubule depolymerization. In the cats with pressure-overload RV hypertrophy, cardiocyte microtubule network density and myocardial free and polymerized tubulin were greater in the RVs than in the normally loaded LVs from the same animals. As expected from Figs. 1–3, there were baseline decrements in the extent and velocity of RV cardiocyte sarcomere shortening, but the remarkable finding was that sarcomere motion in these same cardiocytes was restored almost to normal by microtubule depolymerization. The hypothesis was therefore borne out. Given that 20 years had passed between my initial studies of these models as a postdoctoral fellow (5, 6) and the above-mentioned study (29), this was quite a day in the laboratory!
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 6. Pressure-overload hypertrophy via pulmonary artery banding (PA band). Format is identical to that of Fig. 5. Samples are from cats subjected to RV pressure overload via pulmonary artery banding 2 wk earlier. RV-to-body weight ratio (g/kg) was 0.59 ± 0.04 for control and 0.82 ± 0.02 for pressure overload. Resting sarcomere length was 1.93 ± 0.02 and 1.93 ± 0.02 µm for LV and RV cells, respectively (P = not significant). For D and E, statistical comparisons were by 2-way ANOVA and a means comparison contrast; n, number of cells. *P < 0.001 vs. LV value at matched time points;  P < 0.001 vs. time 0 value within a group. (Adapted from Refs. 29 and 30.)
|
|
Does this apply to the hypertrophied and failing RV?
Having established a role for excess microtubules in contractile dysfunction of compensated pressure-overload hypertrophy, the next two logical questions were as follows: 1) Do increased microtubule network density and the associated contractile defects persist during and, thus, potentially contribute to the eventual development of RV failure in this model when the pressure overload is severe? 2) Are the more marked contractile defects present at this later stage also reversible with microtubule depolymerization? Excess microtubules were indeed found in pressure-overloaded failing RVs (25), and the answers to these questions in terms of sarcomere mechanics are given in Fig. 7. In RV pressure-overload hypertrophy alone and in severe RV pressure-overload hypertrophy with failure, cardiocyte sarcomere mechanics were normalized by microtubule depolymerization.
Does this apply to LV pressure-overload hypertrophy?
Because my ultimate goal was to gain insight into the causes of human heart failure, the next step was to move these studies, via a collaborative effort with B. A. Carabello and M. R. Zile, into the clinically more important LV in a large animal model, where cellular and myocardial function could be characterized in the same animals (27). Figure 8 shows that when canine LV afterload was increased in stages at 2-wk intervals, the animals self-segregated into two groups: one developed very extensive but compensated LV hypertrophy, and the other ceased to hypertrophy in response to increasing load and developed LV failure. Myocardial and cardiocyte samples obtained from these LVs at the times indicated by "Biopsy" and "Final" in Fig. 8 allowed us to ask whether microtubules had a role in the development of the contractile abnormalities that are shown here. Figure 9 shows that, for a dog from the hypertrophy group of Fig. 8, there was no increase in cardiocyte microtubule network density at final study or in free, polymerized, or total myocardial tubulin either at the time of biopsy or at final study. This dog had very extensive but well-compensated LV hypertrophy and normal LV wall stress. In striking contrast, Fig. 9 (right) shows that for a dog from the failure group of Fig. 8, there was a major increase in cardiocyte microtubule network density at final study. At the time of biopsy, when LV wall stress remained normal, free, polymerized, and total myocardial tubulin were also normal. However, at final study, when LV wall stress was quite high and LV contractile function was quite depressed, there were very significant increases in free, polymerized, and total tubulin. Figure 10 shows the functional consequences of these changes on the cellular level in the failure group defined in Figs. 8 and 9. At the time of biopsy, when LV wall stress and function remained within the normal range, LV cardiocyte sarcomere mechanics remained normal and unaffected by microtubule depolymerization. At final study, when LV wall stress and function were very abnormal, LV cardiocyte sarcomere mechanics were also very abnormal. However, as in the pressure-overloaded RV, microtubule depolymerization normalized cellular contractile function. Thus the hypothesis was now borne out in the LV, with a specific association with decompensated, high wall-stress pressure-overload cardiac hypertrophy.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 10. LV pressure-overload hypertrophy and failure: summary data for LV cardiocytes from control dogs and aortic-banded dogs with LV failure for which data are shown in Figs. 8 and 9. Top: maximum velocity of sarcomere shortening at indicated times after addition of 1 µM colchicine to LV cardiocytes from control dogs, from failure dogs at biopsy, or from the same and additional failure dogs at final study. All cells were sampled sequentially at each indicated time after drug exposure. Bottom: maximum extent of sarcomere shortening for cells shown at top. Statistical comparisons, which considered all cells from a given group together, were by 2-way ANOVA and a means comparison contrast; n, number of cells. *P < 0.01 for difference from the control group at matched time points; P < 0.01 for difference from the initial time 0 value within a group. (Adapted from Ref. 27.)
|
|
Does this apply to the LV in vivo?
Up to this point, the association of the microtubule hypothesis with cardiac dysfunction in the intact animal had been inferential. To move this story further toward clinical relevance, the hypothesis was now tested directly, again via a collaborative effort with B. A. Carabello and M. R. Zile, in the intact dog with a surgical model of aortic stenosis (19). Figure 11 compares LV contractile function and myocardial free and polymerized tubulin in a dog with severe aortic stenosis with these same variables in a normal dog. At baseline (see 1 in both panels), the ratio of free to polymerized LV tubulin in the immunoblots shifts toward polymerized tubulin in the aortic stenosis versus the control dog, and the aortic stenosis dog also shows a clear increase in LV wall stress and a clear decrease in LV contractile function. The immunoblots then show that intravenous colchicine caused LV microtubule depolymerization in both dogs. However, although colchicine had a negative effect on contractile function in the normal control dog, any such negative effect in the dog with aortic stenosis was overridden by the ameliorative effect of microtubule removal on LV contractile function in the intact animal.
Does this apply to clinical LV pressure-overload hypertrophy?
To take the final step on the path to clinical relevance, the hypothesis was tested via a collaboration with M. R. Zile and D. C. Miller in the LVs of patients with aortic stenosis (33). Figure 12 (top) shows the four groups of patients that were defined on the basis of LV function. The immunoblot shows for such patients that there is an increase in LV polymerized tubulin of a minor degree with borderline LV dysfunction but of major degree with definite LV dysfunction. The bottom panel of Fig. 12 shows that there is a clear inverse relationship between myocardial microtubules and myocardial function in the LVs of these patients. While we could not manipulate myocardial microtubules directly in this clinical setting to assign a causal relationship, the association shown here suggested that, as in animal models of human disease, microtubule network densification may well contribute to the development of heart failure in some patients with pressure-overload hypertrophy.
|
IS THERE A SPECIFIC LINK BETWEEN MICROTUBULE NETWORK DENSITY AND MYOCARDIAL CONTRACTILE FUNCTION?
|
|---|
Having followed this path of microtubule network densification versus myocardial contractile dsyfunction from the sarcomeres of pressure-overloaded cardiocytes to the ventricles of pressure-overloaded hearts in patients, it is useful to ask whether the association shown thus far is specific. Ideally, I wanted to satisfy by analogy of microtubules to bacteria the formal cause-and-effect test given by Koch's postulates for linking a bacterium to a disease (18): 1) the bacterium must be present in diseased, but not healthy, animals; 2) it must be isolated from the host and propagated as a unique entity; 3) it must cause the disease when introduced into a healthy animal; and 4) it must then be found in that infected animal. Thus, in the initial study of cardiocytes from the pressure-overloaded feline RV (29), controls consisted of nonhypertrophied cardiocytes from the normally loaded LV from the same animal and equivalently hypertrophied cardiocytes from the volume-overloaded feline RV, which was known to exhibit normal contractile function (5). In neither control were the linked cytoskeletal and contractile defects present. To avoid making the conclusions dependent on the use of a single modality such as colchicine for microtubule depolymerization and because any chemical has unintended and likely unrecognized secondary effects, microtubules were also depolymerized via the physical means of low temperature, which would be anticipated to have secondary effects very different from those of a chemical agent. The results replicated those seen with colchicine. To determine whether excess microtubule network density was itself responsible for the contractile dysfunction seen in cardiocytes hypertrophying in response to a pressure overload, excess microtubule polymerization was caused to occur in normal cells by two independent means. First, taxol was used (30) to hyperpolymerize the microtubules of normal cardiocytes; the contractile dysfunction characteristic of pressure overload-hypertrophied cardiocytes was reproduced. Second, and for the same reason that a physical, as well as a chemical, agent was used to depolymerize microtubules, deuterium oxide was used to hyperpolymerize the microtubules (30) and found to replicate the effects of taxol. Therefore, by analogy with Koch's first and second postulates, the microtubule increase is absent in even extensive cardiac hypertrophy so long as contractile function remains normal. Instead, increased microtubules develop solely in dysfunctional hypertrophied muscle and only when such dysfunction appears, at which time microtubule depolymerization restores normal contractile function. By analogy with Koch's third and fourth postulates, chemical and physical agents that increase microtubule network density independent of hemodynamic input were shown to reproduce the linked contractile and cytoskeletal abnormalities seen in severe pressure-overload cardiac hypertrophy. In our other studies at the levels of RV cardiac tissue in vitro, LV cardiac cells in vitro, and myocardium in vivo, these controls were replicated, to the extent permitted by each preparation, with equivalent results (11).
Nonetheless, apart from concerns about unintended cardiac inotropic effects of microtubule depolymerization agents, there are other concerns based on the normal roles of microtubules in interphase cells such as the cardiocyte. For instance, microtubule-based transport is important for a number of intracellular entities, and we have shown this to be true for activated -adrenergic receptors (2, 3). Thus microtubule depolymerization could cause an unintended inotropic effect via altered -adrenergic receptor activity. However, this concern was not applicable to our in vitro studies where -adrenergic input was absent, and it was obviated in our in vivo work by the use of -receptor blockade.
Another possibility, whenever an experimental intervention alters cardiac inotropic state, is that there might have been unrecognized effects on Ca2+ metabolism. In this work, we have found that microtubule depolymerization increases neither resting nor peak activated Ca2+ levels in normal or hypertrophied cardiocytes (30), and colchicine increases neither cAMP levels, peak activated Ca2+ levels, nor the rate of rise or fall of intracellular Ca2+ in normal or hypertrophied myocardium (32). Nevertheless, it has been suggested that colchicine increases Ca2+ current density and the intracellular Ca2+ transient in adult cardiocytes (13) via ,-tubulin heterodimers acting as a functional analog of G proteins to activate adenyl cyclase when the free tubulin concentration is increased by microtubule depolymerization. Therefore, the study shown in Fig. 13 was done (16) to test this idea. When vincristine, rather than colchicine or hypothermia, was used to depolymerize cardiocyte microtubules, the effect of restoring the contractile dysfunction of pressure-overloaded hypertrophied myocardium to normal was identical. However, as shown by the immunoblots, this occurred despite the vincristine-induced reduction in the ,-tubulin heterodimer concentration to essentially zero. Thus this concern is not tenable.
The most definitive solution, however, would be to close the loop and address the microtubule hypothesis directly in a context apart from hemodynamic alterations in vivo or microtubule alterations in vitro. On the basis of some of our other work defining the basic causes of microtubule network densification in cardiac hypertrophy (21, 23, 28), I very recently tested, via a collaboration with F. Cabral and D. R. Menick, the effects of 1-tubulin mutants that had been selected for their effects on microtubule stability and then expressed in the hearts of transgenic mice (4). We found that when intrinsic microtubule stability was increased as an isolated variable, the contractile defects characteristic of pressure-overloaded hypertrophied myocardium were reproduced. In contrast, Fig. 14 shows that when intrinsic microtubule stability was decreased as an isolated variable, the contractile defects characteristic of pressure-overloaded hypertrophied myocardium were largely prevented. Thus cardiocyte microtubule network density, as a single isolated variable, is inversely related to cardiac contractile function, and I think that it is fair to say that the microtubule hypothesis is thus shown directly to be valid.
|
MATERIAL PROPERTIES OF THE MICROTUBULE NETWORK
|
|---|
In normal cardiocytes, although very recent data suggest that microtubules may contribute to longitudinal shear stiffness (22), the normal microtubule network actually has very little effect on cytoskeletal material properties in terms of influencing sarcomere (29), cardiocyte (31), myocardial (14), or cardiac (19) mechanics. What then is the precise physical nature of the impediment to contraction produced by microtubule network densification? I addressed this question in a collaboration with N. Wang and M. R. Zile by directly probing the viscoelastic properties of the tensegrity network of physically interrelated cytoskeletal filaments in living cardiocytes (26) and then confirming these findings at the level of whole cell and tissue mechanics (14, 31). Figure 15 (top) shows the appearance of beads that are attached via integrin receptors to the cardiocyte cytoskeleton. The middle panel of Fig. 15 shows the results of twisting and then releasing these beads in a magnetic field. The bottom panel shows that in the hypertrophied cardiocyte there is an increase in cytoskeletal stiffness and apparent viscosity. In the summary data from this study (26), cytoskeletal stiffness doubled from 8.53 ± 0.77 dyn/cm2 in control cardiocytes to 16.46 ± 1.32 dyn/cm2 in hypertrophied RV cardiocytes, but there was a much larger (fourfold) increase in cytoskeletal apparent viscosity from 20.97 ± 1.92 poise in control cardiocytes to 87.85 ± 6.95 poise in hypertrophied RV cardiocytes.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 15. Viscosity and stiffness of cardiocyte extramyofilament cytoskeleton. Top: modulation contrast micrograph of a feline cardiocyte with integrin-attached Arg-Gly-Asp peptide-coated ferromagnetic microbeads. Middle: data obtained in magnetic twisting cytometer from normal microbead-coated cardiocytes. Relaxation curve ("Relaxation") represents spontaneous remanent field decay without a twisting field. Arrows, times of twisting field application ("Twist On") and removal ("Twist Off"). Cytoskeletal stiffness is inversely related to decrease in remanent field after twisting field is applied, and cytoskeletal apparent viscosity is inversely related to slope of remanent field recovery after twisting field is removed. Bottom: effect of microtubule depolymerization on cardiocyte cytoskeletal viscosity and stiffness. Magnetometry data were obtained from control RV cardiocytes at baseline and from pressure-hypertrophied RV cardiocytes at baseline and 1 h after addition of 1 µM colchicine to cardiocyte superfusate. Cytoskeletal apparent viscosity and stiffness are normalized by microtubule depolymerization in hypertrophied cells. (Adapted from Ref. 26.)
|
|
In real physical terms, this does not represent a change in intracellular fluid viscosity. Rather, the changes in apparent viscosity observed in hypertrophied cardiocytes appear to represent structural damping attributable to the microtubule component of the cytoskeleton. In these terms, apparent viscosity might well reflect a process of intracellular frictional dissipation that impedes cellular shortening. Hence, I would expect this impeding effect, as observed here experimentally, to become more pronounced at higher rates of sarcomere motion, and I would also expect on this basis the loss of energetic efficiency during contraction seen in my earliest work as a postdoctoral fellow (Figs. 1 and 2) that initiated these 30 years of work on the causes and consequences of cardiac hypertrophy.
|
FUTURE DIRECTIONS
|
|---|
As indicated earlier, microtubules are critically important to a number of aspects of cellular homeostasis. In fact, their mechanical effects on striated muscle contraction probably have nothing at all to do with their normal role in the life of the cardiocyte or any other cell. In then thinking about normal microtubule functions that might be altered in important ways in the hypertrophied cardiocyte, their role in intracellular transport is especially intriguing. Briefly, the cytoplasm of the adult cardiocyte is an extremely diffusion-restricted space, such that macromolecules, particles, vesicles, and organelles cannot reach their proper intracellular locations via simple diffusion. Instead, active transport is required for this purpose, and microtubules supply this function: the dynein family of motor proteins moves these cargoes along microtubules toward the cell center, and the kinesin family of motor proteins moves these cargoes along microtubules from the nuclei toward the cell periphery. In cardiac hypertrophy, the microtubules not only proliferate but also become heavily decorated by microtubule-associated protein-4, the predominant cardiac microtubule-associated protein (23, 28). One consequence of this is that the normal microtubule transport function is sterically inhibited by the presence of microtubule-associated protein-4 on these transport tracks, and as one example that may help explain -adrenergic receptor downregulation in cardiac hypertrophy and failure, we have shown that activated G protein-coupled receptors fail to be recycled properly to the cell membrane (2, 3).
Of more general interest is whether the microtubule network densification that is specifically associated with high-wall-stress pressure-overload cardiac hypertrophy may itself contribute via a positive-feedback loop to heart failure. This would occur via an inhibition of the microtubule-based, kinesin-mediated mRNA complex transport function, which is essential for the synthesis and assembly of structural proteins. That is, the LVs of the failure group in Fig. 8 first showed typical microtubule changes at 4 wk of pressure loading, which is also the time at which compensatory growth essentially ceased while LV wall stress was beginning to increase. These dogs uniformly went on to develop heart failure. In contrast, the LVs of the hypertrophy group retained normal microtubules and continued to grow in response to increasing load, such that LV wall stress did not increase. These dogs never developed systolic heart failure. Is it possible that the integrity of cardiocyte microtubule transport may influence the extent to which the cardiac hypertrophic response to load is compensatory? I do not know, but our early data (24) suggest that this may well prove to be the case.
|
GRANTS
|
|---|
The studies from my laboratory and the laboratories of my colleagues at the Gazes Institute were supported by National Heart, Lung, and Blood Institute Program Project Grant HL-48788 and by Merit Awards from the Research Service of the Department of Veterans Affairs.
|
ACKNOWLEDGMENTS
|
|---|
I owe an enormous debt of gratitude to the many students, postdoctoral fellows, and faculty, here and elsewhere, who have been responsible for so much of the science described in this review.
|
FOOTNOTES
|
|---|
Address for reprint requests and other correspondence: G. Cooper IV, Gazes Cardiac Research Institute, PO Box 250773, Medical Univ. of South Carolina, 114 Doughty St., Charleston, SC 29403 (e-mail: cooperge{at}musc.edu)
|
REFERENCES
|
|---|
- Braunwald E, Ross J, and Sonnenblick EH. Mechanisms of Contraction of the Normal and Failing Heart (2nd ed.). Boston, MA: Little, Brown, 1976.
- Cheng G, Iijima Y, Ishibashi Y, Kuppuswamy D, and Cooper G. Inhibition of G protein-coupled receptor trafficking in neuroblastoma cells by MAP4 decoration of microtubules. Am J Physiol Heart Circ Physiol 283: H2379–H2388, 2002.[Abstract/Free Full Text]
- Cheng G, Qiao F, Gallien TN, Kuppuswamy D, and Cooper G. Inhibition of -adrenergic receptor trafficking in adult cardiocytes by MAP4 decoration of microtubules. Am J Physiol Heart Circ Physiol 288: H1193–H1202, 2005.[Abstract/Free Full Text]
- Cheng G, Takahashi M, Shiraishi H, Baicu CF, Cabral F, Zile MR, Menick DR, and Cooper G. A direct test of the cardiac microtubule hypothesis (Abstract). Circulation 112, Suppl II: II-302, 2005.
- Cooper G, Puga F, Zujko K, Harrison CE, and Coleman HN. Normal myocardial function and energetics in volume-overload hypertrophy in the cat. Circ Res 32: 140–148, 1973.[Abstract/Free Full Text]
- Cooper G, Satava RM, Harrison CE, and Coleman HN. Mechanism for the abnormal energetics of pressure-induced hypertrophy of cat myocardium. Circ Res 33: 213–223, 1973.[Abstract/Free Full Text]
- Cooper G, Satava RM, Harrison CE, and Coleman HN. Normal myocardial function and energetics after reversing pressure-overload hypertrophy. Am J Physiol 226: 1158–1165, 1974.[Free Full Text]
- Cooper G. Cardiocyte adaptation to chronically altered load. Annu Rev Physiol 49: 501–518, 1987.[CrossRef][Web of Science][Medline]
- Cooper G. Load and length regulation of cardiac energetics. Annu Rev Physiol 52: 505–522, 1990.[CrossRef][Web of Science][Medline]
- Cooper G. Basic mechanisms of myocardial hypertrophy. Annu Rev Med 48: 13–23, 1997.[CrossRef][Web of Science][Medline]
- Cooper G. Cardiocyte cytoskeleton in hypertrophied myocardium. Heart Fail Rev 5: 187–201, 2000.[CrossRef][Medline]
- Cooper G. Cytoskeletal abnormalities in cardiocytes during heart failure. In: Congestive Heart Failure (3rd ed.), edited by Hosenpud JD and Greenberg BH. Philadelphia, PA: Lippincott Williams & Wilkins, 2006.
- Gómez AM, Kerfant BG, and Vassort G. Microtubule disruption modulates Ca2+ signalling in rat cardiac myocytes. Circ Res 86: 30–36, 2000.[Abstract/Free Full Text]
- Harris TS, Baicu CF, Conrad CH, Koide M, Buckley JM, Barnes M, Cooper G, and Zile MR. Constitutive properties of hypertrophied myocardium: cellular contribution to changes in myocardial stiffness. Am J Physiol Heart Circ Physiol 282: H2173–H2182, 2002.[Abstract/Free Full Text]
- Ishibashi Y, Rembert JC, Carabello BA, Nemoto S, Hamawaki M, Zile MR, Greenfield JC, and Cooper G. Normal myocardial function in severe right ventricular volume-overload hypertrophy. Am J Physiol Heart Circ Physiol 280: H11–H16, 2001.[Abstract/Free Full Text]
- Ishibashi Y, Takahashi M, Isomatsu Y, Qiao F, Iijima Y, Shiraishi H, Simsic JM, Baicu CF, Robbins J, Zile MR, and Cooper G. Role of microtubules vs. myosin heavy chain isoforms in contractile dysfunction of hypertrophied murine cardiocytes. Am J Physiol Heart Circ Physiol 285: H1270–H1285, 2003.[Abstract/Free Full Text]
- Kent RL, Mann DL, Urabe Y, Hisano R, Hewett KW, Loughnane M, and Cooper G. Contractile function of isolated feline cardiocytes in response to viscous loading. Am J Physiol Heart Circ Physiol 257: H1717–H1727, 1989.[Abstract/Free Full Text]
- Koch R. The etiology of tuberculosis. Am Rev Tuberc 25: 285–323, 1932.
- Koide M, Hamawaki M, Narishige T, Sato H, Nemoto S, DeFreyte G, Zile MR, Cooper G, and Carabello BA. Microtubule depolymerization normalizes in vivo myocardial contractile function in dogs with pressure-overload left ventricular hypertrophy. Circulation 102: 1045–1052, 2000.[Abstract/Free Full Text]
- Mann DL, Urabe Y, Kent RL, Vinciguerra S, and Cooper G. Cellular versus myocardial basis for the contractile dysfunction of hypertrophied myocardium. Circ Res 68: 402–415, 1991.[Abstract/Free Full Text]
- Narishige T, Blade KL, Ishibashi Y, Nagai T, Hamawaki M, Menick DR, Kuppuswamy D, and Cooper G. Cardiac hypertrophic and developmental regulation of the -tubulin multigene family. J Biol Chem 274: 9692–9697, 1999.[Abstract/Free Full Text]
- Nishimura S, Nagai S, Katoh M, Yamashita H, Saeki Y, Okada J, Hisada T, Nagai R, and Sugiura S. Microtubules modulate the stiffness of cardiomyocytes against shear stress. Circ Res 98: 81–87, 2006.[Abstract/Free Full Text]
- Sato H, Nagai T, Kuppuswamy D, Narishige T, Koide M, Menick DR, and Cooper G. Microtubule stabilization in pressure overload cardiac hypertrophy. J Cell Biol 139: 963–973, 1997.[Abstract/Free Full Text]
- Scholz D, Garnovskaya M, Huettelmaier S, Goetze B, Kiebler M, and Cooper G. Microtubule-dependent distribution of cardiocyte mRNA: inhibition by MAP4 decoration of microtubules (Abstract). Circulation 112, Suppl II: II-99, 2005.
- Tagawa H, Koide M, Sato H, and Cooper G. Cytoskeletal role in the contractile dysfunction of cardiocytes from hypertrophied and failing right ventricular myocardium. Proc Assoc Am Physicians 108: 218–229, 1996.[Web of Science][Medline]
- Tagawa H, Wang N, Narishige T, Ingber DE, Zile MR, and Cooper G. Cytoskeletal mechanics in pressure-overload cardiac hypertrophy. Circ Res 80: 281–289, 1997.[Abstract/Free Full Text]
- Tagawa H, Koide M, Sato H, Zile MR, Carabello BA, and Cooper G. Cytoskeletal role in the transition from compensated to decompensated hypertrophy during adult canine left ventricular pressure overloading. Circ Res 82: 751–761, 1998.[Abstract/Free Full Text]
- Takahashi M, Shiraishi H, Ishibashi Y, Blade KL, McDermott PJ, Menick DR, Kuppuswamy D, and Cooper G. Phenotypic consequences of 1-tubulin expression and MAP4 decoration of microtubules in adult cardiocytes. Am J Physiol Heart Circ Physiol 285: H2072–H2083, 2003.[Abstract/Free Full Text]
- Tsutsui H, Ishihara K, and Cooper G. Cytoskeletal role in the contractile dysfunction of hypertrophied myocardium. Science 260: 682–687, 1993.[Abstract/Free Full Text]
- Tsutsui H, Tagawa H, Kent RL, McCollam PL, Ishihara K, Nagatsu M, and Cooper G. Role of microtubules in contractile dysfunction of hypertrophied cardiocytes. Circulation 90: 533–555, 1994.[Abstract/Free Full Text]
- Zile MR, Richardson K, Cowles MK, Buckley JM, Koide M, Cowles BA, Gharpuray V, and Cooper G. Constitutive properties of adult mammalian cardiac muscle cells. Circulation 98: 567–579.
- Zile MR, Koide M, Sato H, Ishiguro Y, Conrad CH, Buckley JM, Morgan JP, and Cooper G. Role of microtubules in the contractile dysfunction of hypertrophied myocardium. J Am Coll Cardiol 33: 250–260, 1999.[Abstract/Free Full Text]
- Zile MR, Green GR, Schuyler GT, Aurigemma GP, Miller DC, and Cooper G. Cardiocyte cytoskeleton in patients with left ventricular pressure overload hypertrophy. J Am Coll Cardiol 37: 1080–1084, 2001.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
K. G. Young, S. F. Thurston, S. Copeland, C. Smallwood, and J. W. Copeland
INF1 Is a Novel Microtubule-associated Formin
Mol. Biol. Cell,
December 1, 2008;
19(12):
5168 - 5180.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. Joubert, J. R. Wilding, D. Fortin, V. Domergue-Dupont, M. Novotova, R. Ventura-Clapier, and V. Veksler
Local energetic regulation of sarcoplasmic and myosin ATPase is differently impaired in rats with heart failure
J. Physiol.,
November 1, 2008;
586(21):
5181 - 5192.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Davis, M. V. Westfall, D. Townsend, M. Blankinship, T. J. Herron, G. Guerrero-Serna, W. Wang, E. Devaney, and J. M. Metzger
Designing Heart Performance by Gene Transfer
Physiol Rev,
October 1, 2008;
88(4):
1567 - 1651.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Cheng, M. R. Zile, M. Takahashi, C. F. Baicu, D. D. Bonnema, F. Cabral, D. R. Menick, and G. Cooper 4th
A direct test of the hypothesis that increased microtubule network density contributes to contractile dysfunction of the hypertrophied heart
Am J Physiol Heart Circ Physiol,
May 1, 2008;
294(5):
H2231 - H2241.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Scholz, C. F. Baicu, W. J. Tuxworth, L. Xu, H. Kasiganesan, D. R. Menick, and G. Cooper IV
Microtubule-dependent distribution of mRNA in adult cardiocytes
Am J Physiol Heart Circ Physiol,
March 1, 2008;
294(3):
H1135 - H1144.
[Abstract]
[Full Text]
[PDF]
|
|
|
Copyright © 2006 by the American Physiological Society.
TOP
ABSTRACT
MYOCARDIAL MECHANICS AND...
P < 0.01 vs. RV hypertrophy at matched time points. (Adapted from Ref. 
