| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Bioengineering Institute and Departments of 2Anatomy, 3Engineering Science, and 4Physiology, The University of Auckland, Auckland 1001, New Zealand; and 5Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139-4307
Submitted 23 June 2003 ; accepted in final form 2 October 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cardiac muscle; passive tissue compliance; stress-strain relations; stretch
However, the passive tension-length relationship is difficult to determine at the tissue level due to the complex three-dimensional geometry of the heart and the laminar microstructure of myocardium (6, 26, 35, 44). To reduce the anatomic and microstructural complexities present in a whole heart model, mechanical experiments are often conducted on smaller samples of myocardial tissue. Such experiments range from biaxial testing of myocardial sheets (10) to uniaxial testing of ventricular strips (33), papillary muscles (4, 16), cardiac trabeculae (42), isolated myocytes (5, 17), myofibrils (29), actin filaments (30), and individual titin molecules (25). Anisotropic, nonlinear, large-deformation material behavior has been observed together with strain history-dependent and strain rate-dependent phenomena.
It is well known that myocardium exhibits viscoelastic behavior because the stress response depends on strain history and strain rate. However, strain softening phenomena have also been observed in which the myocardial stress-strain relation depends on the maximum strain state obtained in an experimental protocol. This is typically manifested as a stiffer stress-strain relationship on the first loading cycle than in all subsequent cycles. On stretching to a new maximum strain level, the stress-strain relation follows the previously recorded curve up to the previous maximum strain level. Subsequent repetitions to the new maximum level follow a new, softer stress-strain relationship. Figure 1 illustrates a typical strain softening response. Unlike viscoelastic behavior, the tissue does not recover, even after several hours of rest.
|
Emery et al. (13, 14) measured the pressure-volume relation and strain softening behavior of whole rat hearts and found that the "preconditioning" behavior observed was better explained by a strain softening model than by viscoelastic effects. Dokos et al. (12) subjected blocks of pig ventricular myocardium to simple shear deformations and recorded strain softening behavior even at low levels of strain. Both these studies reported strain softening well within the physiological range. Linke et al. (27) found strain softening in isolated rabbit cardiac myofibrils but only at sarcomere length (SL) extensions above 2.7 µm, which is greater than the normal physiological SL extension. Because all of the collagen had been removed, this strain softening was attributed to titin being nonreversibly altered. In contrast to this result, Kellermayer et al. (25) measured reversible softening in single titin molecules, which was described as "mechanical fatigue." This softening occurred at low levels of stress, comparable with those reached by skeletal muscle when stretched to 2–3 µm SL.
Possible causes of myocardial strain softening include damage to elastic constitutive components (13, 14), fluid shifts, or rearrangement of microstructural components. However, studies that used the chemical phosphatase 2,3-butanedione monoxime (BDM) may also have been influenced by the effect of this drug on cross-bridge action and mechanical stiffness. The negative inotropic action of BDM on cardiac tissue was first demonstrated by Wiggins et al. (43) in 1980. BDM has been shown to protect human cardiac tissue from cutting injury (34), delay the onset of rigor in isolated chick and rabbit myocytes (22), and delay the onset of rigor as well as the depletion of ATP and creatine phosphate in ferret papillary muscle (19). Although BDM has been unanimously reported to inhibit contractile force in a dose-dependent manner, its effects on calcium transients, intracellular calcium concentration, and action potentials are less well defined (for a review, see Ref. 38). It appears to affect multiple sites in the excitation-contraction coupling pathway, with each site having a different and dose-dependent response. The effects of BDM also appear to be temperature, Ca2+ concentration, and species dependent. The effects of BDM on passive tissue properties have not been fully explored, although BDM has been reported to affect the passive pressure-volume relationship of in vivo porcine hearts (31). Because BDM is presumed to increase the proportion of cross-bridges in the weakly bound state in activated cardiac tissue (45), the presence of BDM in quiescent cardiac tissue may induce a similar increase in the proportion of weakly bound cross bridges and hence increase the observed strain softening.
The objective of the present study was to ascertain whether strain softening occurs in intact healthy rat cardiac trabeculae and, if so, whether high concentrations of BDM, as used in studies of Dokos et al. (12) (50 mM) and Emery et al. (13, 14) (30 mM), is an exacerbating factor. The use of isolated trabeculae, as opposed to whole hearts or blocks of ventricular tissue, allowed improved measurement of muscle length extension while minimizing damage due to specimen preparation and allowing an adequate diffusive supply of oxygen. The trabecula preparation is a thin, naturally occurring collection of axially aligned myocytes. Mounted between a force transducer and an actuator, it has been extensively used for the examination of myocardial mechanical properties since the seminal paper by ter Keurs et al. (42). We used these preparations from the rat right ventricle to answer the following questions: 1) Is strain softening behavior exhibited in viable intact cardiac trabeculae [cf. intact hearts (13, 14) and ventricular myocardial blocks (12)]? 2) Is strain softening influenced by the presence of BDM? and 3) Does BDM affect the passive force-length relationship of cardiac trabeculae?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All experiments were approved by the University of Auckland Animal Ethics Committee. Wistar rats (age 71 ± 28 days, weight 303 ± 38 g, means ± SD) were stunned and immediately decapitated. The heart was quickly excised and plunged into dissection solution at 0°C to induce arrest. After the aorta was cannulated, the coronary vasculature was perfused using the Langendorff technique.
The criteria for selection of right ventricular trabeculae were similar to those described by de Tombe and ter Keurs (9). Unbranched, geometrically uniform specimens were sought that ran freely from the right ventricular free wall to the atrioventricular ring and were relatively long (2.52 ± 0.87 mm, n = 10) and sufficiently thin (207 ± 41 µm) to ensure adequate oxygenation (37). The trabecula preparation, including blocks (
300 x 300 x 600 µm) of tissue at each end for mounting, was excised from the ventricular wall. When the preparation was transferred to the muscle bath, the transferral method avoided pulling the trabeculae through fluid menisci.
The superfusate was a modified Tyrode solution [containing (in mM) 141.6 NaCl, 5.97 KCl, 1 MgCl2, 1.18 NaH2PO4, 10 HEPES, 10 glucose, and 0.5 CaCl2]. The pH was adjusted to 7.4 by the addition of 1 M Tris. To form the dissection solution, the Tyrode solution was supplemented with 20 mM BDM. During experiments, preparations were superfused with the Tyrode solution in either the presence (+BDM, 50 mM) or absence (–BDM) of BDM. A concentration of 50 mM BDM was chosen to emulate the Dokos et al. study (12), in which strain softening was measured. Solutions were vigorously bubbled with 100% O2. All chemicals were purchased from Sigma Aldrich and were of analytic grade. All experiments were conducted at 26°C.
Muscle Bath and Superfusate Delivery System
The U-shaped muscle bath (2 x 2 x 10 mm, volume 40 µl) was mounted on a Peltier device (MI 1023 TAC, 9.2 W, Marlow Industries), which was, in turn, mounted on an X-Y-Z micromanipulator (MicroMech, Coherent). This system had a working temperature range of 10–50 ± 0.1°C. Surface tension constrained the superfusate within the open-topped and open-ended bath, while the superfusate flow was controlled by two peristaltic pumps (Minipuls III, Gilson). To allow electric field stimulation of the trabecula, two 100-µm-diameter platinum wire electrodes were situated either side of the muscle bath. An SD9 Stimulator (Grass Instruments) provided the stimulus.
Data Acquisition and Experimental Control
A Pentium 3-equivalent personal computer was used for data acquisition and experimental control. Data acquisition, motor control, and image processing were achieved using software custom written in LabView 5.1 (National Instruments), a data-acquisition card (PCI 6031E, National Instruments), a four-axis motor controller card (PCI 7344, National Instruments), and a digital frame-grabber card (PCI FG-3170 CameraLink, Silicon Imaging).
Muscle Attachment
An end-on muscle attachment system was constructed using glass hooks fashioned from thin glass tubing (500 µm inner diameter with a wall thickness of 50 µm, VitroCom). One hook was attached to the strain gauge, and the second hook was attached to the mechanical perturbation system. The trabecula was carefully maneuvered into the hooks, which were then separated until the trabecula was gently held. Care was taken not to stretch the muscle beyond resting length.
Force Measurement
Force was measured via a silicon beam strain gauge (model AE-801, SensoNor), which was attached to a well-damped, three-axis micromanipulator (Prior Scientific; Cambridge, UK), and amplified by a factor of 400. This system had a sensitivity of 1.262 V/N with 10 µN resolution.
Mechanical Perturbation System
Mechanical perturbation was supplied via a high-resolution linear actuator (M-227.25 DC-Mike, Physik Instrumente), which was controlled via the four-axis motor controller card. The first resonant frequency of the complete mechanical system was above 1 kHz.
Optical Systems
A Wild8 stereomicroscope (Leica), a color charge-coupled device camera (1CD-800P, Ikegami), and a monitor (PVM-1454QM Trinitron, Sony) aided in the manipulation, mounting, and monitoring of trabeculae during the course of the experiment.
A second microscope system obtained high-magnification images of sections of the trabecula, from which an estimation of SL was calculated. This microscope system was composed of the following components: a x50 objective lens (No. 101950, LWD MSP Plan, Olympus), an Infinitube in-line assembly (No. L54-590, Edmund Optics), a video camera (SI-3170 MegaCamera, Silicon Imaging), a frame grabber (FG-3170 CameraLink card, Silicon Imaging), and a controlling computer. The camera, in conjunction with the x50 objective lens and Infinitube, gave a 120 x 90 µm field of view. An automated X-Y-Z micromanipulator stage (M425 series, Newport) and the four-axis motor controller card were used to position the SL microscope objective lens.
A fast Fourier transform technique, similar to that validated by Dobesh et al. (11), was used to calculate the average SL within the field of view. The system was calibrated with 600 and 200 lines/mm diffraction gratings and gave a measured accuracy of 6 nm. This system was used to set the resting SL to 1.9 µm and to calculate the SL extension during experimental protocols.
Mounting of the Muscle
After the trabecula was mounted in the muscle bath, the resting SL was measured. If the SL was <1.9 µm, the muscle length (ML) was increased until SL reached 1.9 µm. If the SL was >1.9 µm, then ML was not reduced (to avoid preconditioning the muscle). The average resting SL was 1.97 ± 0.06 µm, whereas the longest SL induced during the 15% ML stretches was 2.28 µm.
If undue spontaneous activity was present when the specimen was at its operating temperature of 26°C, the muscle was excluded from analysis (9). Each trabecula was stimulated (rectangular, 4-ms duration, 0.2 Hz, 20% supramaximal voltage) until its force response stabilized (minimum of 10 min). After stabilization, the resting SL was remeasured; if the preparation had shortened due to the elicited twitches, then it was lengthened back to the prestimulation SL.
Experimental Protocols
The goals of this study were to ascertain the existence of strain softening in intact rat trabeculae in the presence and absence of 50 mM BDM. The mechanical perturbation protocol used to measure strain softening was similar to that used by Dokos et al. (12) on blocks of ventricular tissue. Each trabecula was subjected to six ML extension-and-release cycles (Cycles), where the extensions were 5%, 10%, and 15% ML (Stretch). After a 10-min rest interval, the 15% ML Stretch was repeated to test for recovery of stress. This series of six Cycles at three Stretches, plus repetition of the 15% Stretch, is referred to as one Set. Stretches were conducted at a constant velocity of 25 µm/s. Upon completion of a Set, the viability of the trabecula was assessed via twitch force. Provided that active muscle performance remained satisfactory, up to four Sets were then performed. Strain softening was assessed by comparing the stresses at a given ML extension: 1) between Stretches within Set 1, and 2) between corresponding Cycles (Cycles 1–6) of Set 1 and Set 2.
The effect of BDM on strain softening was examined by subjecting five trabeculae to +BDM in Set 1 and to –BDM in Set 2. Another five trabeculae were tested in the converse order. This allowed the effect of BDM to be determined independent of the order in which the tests were performed (denoted as Order below).
Stress was defined as force per (ellipsoidal) cross-sectional area (estimated from top and side dimensions). For the trabeculae used in this study, the peak twitch force measured at resting SL (1.97 ± 0.06 µm) was 25.9 ± 7.4 kPa (26°C, 0.5 mM Ca2+). Note that to avoid preconditioning, SL was not increased to give maximal twitch force.
Statistical Analysis
Data were subjected to repeated-measures ANOVA as appropriate for a 3 x 4 x 6 (Set x Stretch x Cycle) design. Differences among levels of statistically significant (P < 0.05) main effects or interactions were sought, post hoc, using appropriate sets of mutually orthogonal contrast coefficients. All analyses were performed using SAS software. Unless otherwise stated, all data are presented as means ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A typical example of the stresses induced upon subjecting a quiescent trabecula to its first Set of length perturbations, consisting of 5%, 10%, and 15% Stretches, is plotted in Fig. 2. Note the absence of irreversible strain softening. Similarly, strain softening is not apparent in Fig. 3, which compares the 15% Stretches between Set 1 (–BDM) and Set 2 (+BDM).
|
|
In Figs. 2 and 3, the observed hysteresis loops demonstrate energy dissipation consistent with the presence of viscoelastic material properties. A small degree of softening was observed between the first and subsequent Cycles of each Stretch (Figs. 2 and 3), which recovered before the next Stretch. This reversible softening can be understood in terms of viscoelasticity but not in terms of irreversible strain softening.
Figure 4A presents stress as a function of SL, illustrating that viscoelastic softening also occurs at the SL level. Figure 4B shows that SL faithfully tracks ML across the full range of extensions. The nearly linear form of the relationship indicates that end compliance is roughly proportional to midtrabecula compliance for these low extensions. Because SL data were acquired only 12 times per cycle, and no stress-SL plots showed any signs of strain softening (Fig. 4A), subsequent statistical analyses were undertaken using ML data, which were sampled more frequently. Note that across all the experiments used in this study, SLs between 1.9 and 2.28 µm were observed during the 15% ML extensions of viable trabeculae.
|
In the presence of strain softening, any given ML extension would induce a maximum stress the first time the muscle experiences that level of extension. Therefore, the first 5% ML cycle would have the steepest stress-strain relationship, with the following five cycles (at 5% ML extension) showing a more compliant stress-strain relationship. The same softening process would also be observed when the ML stretch was increased to 10% and again to 15%. Thereafter, strain softening would not be observed if the ML extensions were less than or equal to 15%.
Several methods were applied to detect strain softening in the stress-ML extension data. Initially, the stresses measured at 3.5% ML extension were compared across all six Cycles within all four Stretches of Set 1 and Set 2 using repeated-measures ANOVA. For Stretches of 10% and 15%, a similar analysis was conducted using the stresses measured at 7% ML extension. The ANOVA examined the statistical significance of the five main effects (Order, BDM, Cycle, Stretch, and Set) and all possible interactions on stress. The ANOVA results are summarized in Table 1. Figure 5 summarizes the Cycle x Set interactions for both the 3.5% and 7% ML extensions. Note that, for each extension, the statistical power to detect a significant difference (had one existed) was considerable, because the F ratio arising from the repeated-measures ANOVA has 5 and 35 degrees of freedom for the numerator and denominator, respectively. Muscle stiffness did not decrease between Set 1 and Set 2 for either extension, thereby indicating an absence of strain softening. Likewise, BDM had no effect at either extension. Furthermore, there was neither an Order main effect nor a BDM x Order interaction. These results preclude an effect of BDM on passive stress in any of our protocols. Not surprisingly, there was a significant effect of Cycle on stress (P < 0.0001), due to reversible viscoelastic softening. The interaction of Stretch and Cycle was also significant, indicating that Stretch size had an effect on the extent of reversible softening. No other interactions were statistically significant. Post hoc analysis revealed that for both ML extensions, stress was significantly larger during the first cycle, with no significant differences among Cycles 2–6. The value of stress in Cycle 1 (averaged across all Stretches) did not differ between Set 1 and Set 2 (Fig. 5). Thus no test showed the presence of strain softening.
|
|
Despite these null results, we devised and tested two additional indexes of strain softening. The first involved calculating the average stress difference between Cycle 1 and Cycle 6, Cycle 2 and Cycle 6, etc., for each value of Stretch. The second was to find the average stress difference between Cycle 1 of Set 1 and Cycle 1 of Set 2 for each of the 5%, 10%, and 15% ML Stretches. In both cases, the average difference values were then subjected to ANOVA, as above. In neither case (data not presented) did we obtain any evidence of strain softening in either the presence or absence of BDM.
In contrast to the lack of strain softening found in viable trabeculae, Fig. 6 shows evidence of strain softening behavior during one Set of stretches (5%, 10%, and 15%) of a trabecula that had become electrically nonresponsive. Such behavior was found in all trabeculae that ceased to respond to electrical stimulation.
|
|
To determine the effect of BDM on the diastolic compliance of passive cardiac trabeculae, the stresses measured at 3.5% and 7% ML extension were compared between Set 2 and Set 3, where the presence or absence of BDM was reversed between Sets (n = 9). Comparison of Set 2 and Set 3 removed any potential strain softening artifacts, because the maximum strain level of 15% had already been reached in Set 1.
With the use of the same techniques as those for the first index of strain softening (above), the stresses at 3.5% and 7% ML extension for both Set 2 and Set 3 were compared across Stretches (Fig. 7). ANOVA revealed no significant effect of BDM for either the 3.5% (P > 0.43) or 7% (P > 0.28) ML extensions (Table 2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To our knowledge, this is the first study that has attempted to measure irreversible strain softening in viable intact cardiac trabeculae of any species. The term strain softening is applied to materials that exhibit permanently increased compliance after the applied load, or stress, reaches a new maximum. Strain softening has been observed in intact cardiac tissue during both shear experiments on blocks of pig ventricular myocardium (12) and pressure inflation of whole rat hearts (13, 14). In both cases, strain softening was reported to be irreversible at physiologically relevant strains.
We found no evidence that strain softening occurs in intact rat cardiac trabeculae at physiologically relevant SL extensions (Table 1 and Figs. 2, 3, 4). Whereas a small viscoelastic stress relaxation effect was observed, this recovered fully between Sets (Fig. 3) as well as during the 30-s resting interval between Stretches (Fig. 2). We interpret this behavior as evidence of viscoelastic, reversible stress relaxation akin to the mechanical fatigue seen in single molecules of titin (25).
Possible causes of the previously reported strain softening at physiological strains include 1) BDM, 2) tissue microstructure, 3) mode of deformation, 4) fluid movement, 5) ischemic anoxia and nonviability, 6) cutting damage, or 7) strain rate. These are considered in turn below.
BDM. In the studies performed by Emery et al. (13, 14) and Dokos et al. (12), the tissues were perfused with high concentrations of BDM (30 mM, 30 and 50 mM, respectively) before mechanical testing began. This was to protect the cardiac tissue from cutting injury (34) and to inhibit active contractile processes (38). Both Emery et al. and Dokos et al. speculated that the use of BDM might have contributed to the observed strain softening behavior, because BDM is known to increase the proportion of cross bridges in the weakly bound state. Our experiments showed no influence of BDM on strain softening or passive compliance in viable trabeculae (Table 1 and Fig. 7). Hence, we discount any contributions from BDM to the strain softening that has been observed by others.
Tissue microstructure. Strain softening in response to applied loads that are larger than normally observed in vivo is due to either plastic deformation or damage of collagen and other extracellular structures. In trabeculae, wavy perimysial collagen fibers are shown to straighten progressively when extended from a resting SL of 1.85 to 2.3 µm (20). Extension of cardiac tissue beyond a SL of 2.3 µm may well damage perimysial collagen and other extracellular matrix components to which the perimysial fibers are tethered. If the extracellular matrix is removed, cardiac titin exhibits strain softening above a SL of 3 µm (17, 27–29). Because extending sarcomeres beyond a length of 2.3 µm is reported to affect trabeculae viability (8), we kept the SL below 2.3 µm to avoid overextension of myocytes and the extracellular matrix.
Because trabeculae consist of an axial alignment of cardiac myocytes, our results imply that strain softening is absent during physiological extensions of intact viable myocytes. Likewise, Granzier and Irving (17) performed axial extensions of skinned cardiac myocytes from the rat and found no evidence of strain softening between SLs of 1.85 and 2.4 µm. Upon axial extension, isolated rabbit (29) and rat (27) myofibrils have demonstrated strain softening, but only when the SL was extended above 3 µm. Because the skinning process removed the entire extracellular matrix and collagen, Linke et al. (28) attributed the resulting development of passive force and strain softening at large extension to titin.
When single titin molecules from the rat were subjected to repeated stretch-and-release protocols, a rightward shift in the lower force region of the length-stress curve was observed (25). This mechanical fatigue recovered after a 4-min rest and thus was not strain softening, which, by definition, is an irreversible process. Similar softening behaviors have been observed in skinned rat myocytes from which actin filaments were removed (21) and where rest periods as short as 10 s allowed the passive stress to recover. This stress relaxation of the skinned myocytes was once again attributed to mechanical fatigue of titin because the extracellular matrix and actin filaments had been removed. From the results of these studies, we preclude titin as a source of irreversible strain softening.
There are microstructural differences between ventricular tissue and cardiac trabeculae. Ventricular tissue has a complex three-dimensional microstructural architecture that varies with location and depth across the ventricular wall (6, 26, 35, 44). Although trabeculae are often treated as homologous to ventricular tissue (20, 23), they have a simpler microstructure, consisting of axially aligned myocytes and large parallel perimysial collagen fibers (20). More complex microstructures may show strain softening behavior.
Mode of deformation. The mode of deformation applied in the present study was certainly different to those in previous reports of shear deformation of ventricular blocks (12) or volume inflation of whole hearts (13, 14). It is possible that shear or torsion experiments in trabeculae could exhibit strain softening.
Fluid movement. In both whole hearts and tissue blocks, strain softening could be partly due to extrusion of fluid from the vasculature and tissue. However, fluid shifts do not explain strain softening in the shear tests of Dokos et al. (12) because positive shear strain softening did not soften the negative shear stress-strain relationship.
Ischemic anoxia and nonviability. All experiments that have studied strain softening have employed nonperfused tissues. In our case, however, we are confident that the superfused trabeculae were not hypoxic, because they were smaller than the critical diameter (300 µm) at which an anoxic core arises in quiescent preparations (7). Furthermore, trabecula viability, assessed as the twitch force response to electrical stimulation, was confirmed between Sets. In contrast, in both the Emery et al. (13, 14) and Dokos et al. (12) studies, hearts were briefly perfused with high-potassium salt solutions containing BDM designed to induce arrest; thereafter, tissues were neither perfused nor superfused. The resulting ischemia may have induced changes in passive tissue properties and allowed rigor cross bridges to form.
Although BDM has been reported to delay the onset of rigor (1, 19, 22) and to reduce final rigor force in both metabolically challenged cardiac tissue (40) and skinned skeletal muscle (15), it does not prevent rigor bridge formation (1, 19, 22). Strain softening at low strain levels could, therefore, be a by-product of ischemia. If the ischemic period has allowed rigor cross bridges to form, then even a small SL extension may damage tissue, resulting in strain softening. In the present study, only preparations that remained viable (i.e., responsive to electrical stimulation) were included in the data analysis. It was noted that trabecula that were electrically nonresponsive (Fig. 6) showed classic strain softening behavior, together with elevated stress levels, compared with results from viable trabeculae (Fig. 2).
Cutting damage. In the present study, we confined cutting damage to blocks of tissue located well beyond the regions of stress measurement. As Dokos et al. (12) studied the shear behavior of blocks dissected from the ventricular wall, cutting damage was unavoidable, although possibly ameliorated by the presence of BDM (34). Studies that used isolated whole hearts avoided this problem entirely.
Strain rate. The ML extension velocity used in present study was 25 µm/s, which induced an approximate strain rate of 1% ML/s. Whereas this strain rate is lower than that observed during normal physiological operation, it is slightly larger than those used in studies that have demonstrated strain softening. A strain rate of 0.7% initial length (L0)/s was applied during shear deformation of blocks of pig myocardium (12), whereas a strain rate of 0.3% was induced during inflation studies of rat hearts (13, 14). Because we used a strain rate higher than was used in any of these studies, the absence of strain softening cannot be attributed to our use of a low strain rate.
Effect of BDM on Passive Compliance
In the present study, there were no significant differences in trabeculae compliance in the presence or absence of BDM (Fig. 7). This lack of effect of BDM on compliance is consistent with the notion that there are relatively few force-bearing cross bridges in quiescent tissue. Similarly, BDM does not affect the passive force (29) or stiffness (28) generated by titin during a low extension stretch.
There is controversy in the literature concerning the presence of both weakly and strongly bound cross bridges in both the diastolic and quiescent states. The existence of weakly bound cross bridges has been investigated by measurement of instantaneous stiffness. With the use of this technique, studies have inferred both their presence (18, 36) and their absence (3, 8, 28). Similar controversy exists concerning the presence of strongly bound cross bridges during diastole. Studies using a variety of interventions, specimens, and indexes, in the absence of BDM, have provided evidence both for (31) and against (8, 24, 41) the existence of strongly bound, or force-bearing, cross bridges. Curiously, studies performed in the presence of BDM have shown comparably contradictory results (2, 19, 31, 39). Our results (Fig. 7 and Table 2) show a lack of effect of BDM on the compliance of intact rat cardiac trabeculae, thereby concurring with those studies that show a lack of measurable cross-bridge activity in quiescent tissue. This result, however, does not preclude the existence of cross-bridge activity during diastole, as such activity may cease during the long quiescent periods adopted during this study.
BDM has also been shown to increase diastolic pressure in isolated perfused whole hearts (32). A BDM-induced increase of slope of the end-diastolic pressure versus end-diastolic length relation (i.e., stiffness) has been reported in intact pig hearts (31). This was attributed to an increased tissue stiffness, or turgor ("garden hose effect"), which may have resulted from a BDM-induced dilation of the coronary vascular network. No corresponding increase of tension or stiffness is seen in papillary or trabecula preparations (Ref. 19 and present study), nor would any be expected due to this cause, because neither preparation was perfused.
Limitations
One reason for the lack of observable strain softening in the present study could have been accidental preconditioning of trabeculae during dissection or mounting. The utmost care was taken to avoid stretching the trabeculae during the dissection and mounting process, and the method of transferring the muscle from the dissection bath to the muscle bath via glass tubing was specifically designed to avoid the possibility of stretching. It is also possible that specimens could have moved or settled in the mounting hooks during Stretches. However, this would have been apparent by a sudden softening in the stress-strain relationship. As Fig. 6 demonstrates, strain softening was readily measurable by our apparatus in preparations that had stopped responding to electrical stimulation.
In conclusion, the present study has shown an absence of irreversible strain softening in viable, intact, rat cardiac trabeculae subjected to axial extensions within the physiological range regardless of the presence or absence of BDM. This absence of strain softening, in contrast to previously reported findings in other preparations (12–14), may be due to differences in tissue health, fluid movement, tissue microstructure, or the method of strain induction. The present study also found no effect of BDM on passive compliance, at least at the very low strain rates applied, consistent with studies on skinned cardiac myocytes, myofibrils, and titin molecules.
|
ACKNOWLEDGMENTS |
|---|
We acknowledge the generous financial support of the Royal Society of New Zealand (Marsden Fund) and the Health Research Council of New Zealand.
|
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 |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. F. Asnes, J. P. Marquez, E. L. Elson, and T. Wakatsuki Reconstitution of the Frank-Starling Mechanism in Engineered Heart Tissues Biophys. J., September 1, 2006; 91(5): 1800 - 1810. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Speich, L. Borgsmiller, C. Call, R. Mohr, and P. H. Ratz ROK-induced cross-link formation stiffens passive muscle: reversible strain-induced stress softening in rabbit detrusor Am J Physiol Cell Physiol, July 1, 2005; 289(1): C12 - C21. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Kirton, A. J. Taberner, P. M. F. Nielsen, A. A. Young, and D. S. Loiselle Effects of BDM, [Ca2+]o, and temperature on the dynamic stiffness of quiescent cardiac trabeculae from rat Am J Physiol Heart Circ Physiol, April 1, 2005; 288(4): H1662 - H1667. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
