| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
MS 7596 Received 19 November 1997; accepted after revision 29 May 1998.
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Myocardial performance in mammals depends upon factors that include the architecture of the ventricles, after-load against which the heart must work, and the contractile state of the myocardium. The capacity of the heart to perform external work is essential for moving blood throughout the circulatory system but little is known about the mechanisms that regulate this important physiological variable. In this study we present results from a single myocyte preparation in which we have investigated factors regulating the rate of work production, or power output, of mammalian myocardium.
The velocity of muscle shortening is a primary determinant of myocardial power output, i.e. the product of force and velocity, and is thought to be limited by the rate of detachment of myosin cross-bridges from actin. Several factors that affect cross-bridge detachment, and thus the velocity of shortening, have been investigated by characterizing the force-velocity relationships of a variety of cardiac muscle preparations. Force-velocity relationships from both intact (Sonnenblick, 1962; De Clerk et al. 1977; Chiu et al. 1987; de Tombe & ter Keurs, 1990) and skinned (Maughan et al. 1978; Pagani & Julian, 1984; Sweitzer & Moss, 1993) myocardial preparations exhibit the expected decrease in shortening velocity as the load on the muscle is increased. Furthermore, the time course of muscle shortening against constant load has been observed to be curvilinear in cardiac muscle preparations, continually slowing as shortening proceeds (Maughan et al. 1978; Brenner, 1980, 1986; Daniels et al. 1984; Chiu et al. 1987; Sweitzer & Moss, 1993).
One objective of the present study was to assess the time course (i.e. linear or non-linear behaviour) of isotonic shortening using a skinned single myocyte preparation having low end compliance at the points of attachment to the apparatus. This preparation offers several advantages for such a study. First, skinning allows experimental control of the level of Ca2+ activation during isotonic shortening. Second, the use of single myocytes allows precise characterization of contractile and regulatory protein content, as well as important physical variables (for example, sarcomere length) during activation. In this way, accurate myofibrillar protein structure-function analysis can be achieved. Third, skinned myocytes are free of extracellular viscoelastic elements that may give rise to restoring forces during shortening and contribute to non-linear length traces previously observed during shortening of myocardial preparations under force clamp conditions. Last, single myocytes reduce the diffusion distances for ATP and ADP, which decreases the likelihood of concentration gradients arising between the muscle core and surrounding solution.
The second major objective of this study was to assess the load dependence of power output in skinned cardiac myocyte preparations. Changes in load and/or contractility are known to affect cardiac power output during normal physiological phenomena such as exercise and pathological conditions such as hypertension and heart failure. Since Ca2+ is a likely regulator of myocardial power output and intracellular Ca2+ varies on a beat-to-beat basis, we also assessed the effects of Ca2+ on load dependence of power output.
The words force and load are used interchangeably throughout the manuscript, and tension refers to force per muscle cross-section.
This work has been described in part in abstract form (McDonald et al. 1996a, b).
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cardiac myocyte preparation
Myocytes were obtained by mechanical disruption of hearts from Sprague-Dawley rats. Rats were placed in a small air-tight chamber, anaesthetized by inhalation of methoxyflurane, and their hearts were quickly excised and placed in ice-cold Ca2+-free, high-K+ modified Ringer solution. The hearts were gently squeezed to remove blood from the ventricles and then cut into 2-3 mm pieces. These pieces were placed in 
30 ml of cold relaxing solution and further disrupted for 5 s using a Waring blender. The resulting suspension of cells and cell fragments were centrifuged for 90 s at 165 g. The myocytes were subsequently skinned by suspending the pellet for 4 min in 0·3 % ultrapure Triton X-100 (Pierce Chemical Co.) in relaxing solution. The skinning process also disrupted the sarcoplasmic reticulum and improved the contrast of the striation pattern observed by light microscopy. The pellet was washed twice with cold relaxing solution, and the skinned cells were then resuspended in 10-20 ml of relaxing solution and kept on ice during the day of the experiment.
Experimental apparatus
The experimental apparatus for physiological measurements on myocyte preparations was modified from one previously described in detail for skeletal muscle fibres (Moss, 1979). Briefly, skinned cardiac myocytes were attached between a force transducer and torque motor by placing the ends of the myocyte preparation into stainless steel troughs (25 gauge). The ends were secured by overlaying a 0·5 mm length of 4-0 monofilament nylon suture (Ethicon, Inc.) onto each end of the myocyte, and then tying the suture into the troughs with two loops of 10-0 monofilament (Ethicon, Inc) (see Fig. 1). The myocyte preparations remained submerged in relaxing solution during the attachment procedure, which was performed under a stereomicroscope (× 80 magnification) using finely shaped forceps. Preparations having a slack length of 120-200 µm were used, since this provided sufficient cell length to grasp the ends with forceps for placement and proper alignment within the troughs. Approximately 20-30 µm of each end of the myocyte preparation was placed in the trough and was therefore covered with 4-0 suture. Given that rat single myocytes do not usually exceed 120 µm in length (Brady, 1991), the myocyte preparations used in these experiments were in most cases two myocytes connected end-to-end by an intercalated disc. Given the overall low compliance of these preparations, we conclude that the structural integrity of the intercalated discs was maintained during maximal contractile stress.
Mechanical measurements were performed using a capacitance-gauge transducer (Model 403, sensitivity of 20 mV mg-1 and resonant frequency of 600 Hz, Cambridge Technology, Inc., Cambridge, MA, USA). Length changes during mechanical measurements were introduced at one end of the preparation using a DC torque motor (Model 300, Cambridge Technology) driven by voltage commands from a personal computer via a 12 bit D/A converter. Force and length signals were digitized at 1 kHz using a 12-bit A/D converter and each was displayed and stored on a personal computer using customized software based on LabView for Windows (National Instruments Corp., Austin, TX, USA). The entire mechanical apparatus was mounted on a pneumatic vibration isolation table having a cut-off frequency of 1 Hz.
Images of the skinned myocyte preparations were continuously recorded on video tape using a Panasonic video camera (WV-B1600) and a JVC VHS recorder (HR-s6600u). Videomicroscopy was done using a × 40 objective lens (Olympus UWD 40) and a × 8 intermediate lens. After each experiment, the video tape was played back to allow measurement of sarcomere length of the myocyte while relaxed and during activation.
Sarcomere length monitoring
In some experiments, sarcomere length was measured in real time during isotonic shortening using a laser diffraction system similar to the one previously described in detail by de Tombe & ter Keurs (1990) and used in our laboratory (Wolff et al. 1995; McDonald et al. 1997). Skinned myocyte preparations were illuminated with a helium-neon laser beam (Melles-Griot Model 05-LHP151, 5 mW output, 632·8 nm wavelength) perpendicular to the long axis of the myocyte preparation, and the position of the first-order diffraction line was monitored with a 512-element photodiode array (Reticon Model RC 105) that was scanned electronically every 0·5 ms. A glass coverslip was placed over the chamber containing the myocyte segment to eliminate scattering of the first-order line by the fluid meniscus of the activating solutions. The median position of the first-order intensity distribution was converted into a voltage proportional to sarcomere length using a non-linear amplifier (Biomedical Technical Support Centre, University of Calgary). The transfer function of the amplifier was adjusted and calibrated using glass diffraction gratings of known spacing.
Solutions
Compositions of relaxing and activating solutions used in mechanical measurements were as follows (mM): 7 EGTA, 1 free Mg2+, 20 imidazole, 4 MgATP, and 14·5 creatine phosphate (pH 7·0), with various Ca2+ concentrations between 10-9 M (relaxing solution) and 10-4·5 M (maximal Ca2+ activating solution), and sufficient KCl to adjust ionic strength to 180 mM. The final concentrations of each metal, ligand and metal-ligand complex at 12°C were determined with the computer program of Fabiato (1988). Preceding each activation, myocyte preparations were immersed for 30 s in a solution of reduced Ca2+-EGTA buffering capacity, which was identical to normal relaxing solution except that EGTA was reduced to 0·5 mM and 6·5 mM HDTA was added. This protocol resulted in more rapid development of steady-state force during subsequent activation and helped preserve the striation pattern during activation. Relaxing solution in which the ventricles were mechanically disrupted and myocytes resuspended contained (mM): 2 EGTA, 1 MgCl2, 4 ATP, 10 imidazole, and 100 KCl (pH 7·0). Modified Ringer solution contained (mM): 140 NaCl, 15 KCl, 1·2 MgCl2, 2·0 NaH2PO4, 5 sodium acetate, 10 glucose, and 10 Hepes (pH 7·4).
Force-velocity and power-load measurements
All mechanical measurements were made at 12°C. For force-velocity and power-load measurements, a servo-system incorporating integrative gain feedback (Cambridge Technology) was used to control the load on the myocyte preparation (Sweitzer & Moss, 1993). Muscle force was controlled by comparing the force transducer output with a command signal generated by the computer via a D/A board (National Instruments). The protocol for measuring force-velocity curves involved first transferring the myocyte preparation into pre-activating solution (30 s) and then into activating solution. Once steady-state force developed, the computer switched the comparator circuit from length control to force control by applying a 5 V logic pulse. The myocyte preparation was then rapidly stepped (< 5 ms) to a force less than steady state. Force was maintained constant at the specified load for a limited period of time (100-250 ms) during which the length change was continuously monitored. Following the force clamp, the myocyte preparation was slackened to reduce force to near zero to allow estimation of the relative load sustained during isotonic shortening, after which the myocyte preparation was re-extended to its initial length. Because of the short lengths of these myocyte preparations, the rapid length change introduced after isotonic shortening did not always slacken the preparation to yield a baseline force value. This resulted in an underestimation of peak force in many cases and thus of relative force during loaded contractions. To obtain more accurate estimates of relative forces borne during loaded contractions, relative forces were calculated by interpolating peak force from activations performed before and after the series of loaded contractions.
To test the effects of variable Ca2+ activations on force-velocity and power-load curves, a series of isotonic contractions was first performed in a myocyte preparation during maximal activation (pCa 4·5). The myocyte preparation was then activated in a series of solutions in order to determine the pCa that yielded half-maximal force. The myocyte was then transferred into half-maximal activating solution and a series of isotonic shortening contractions was performed. Half-maximal activations were done in solutions ranging in pCa between 6·0 and 5·6.
As many as twenty force clamps were performed on a given cell for a given condition. Typically, myocyte preparations could be kept in activating solution throughout the entire series of force clamps without significant loss of force. If maximal force fell below 80 % during the series of force clamps, the preparation was discarded, but in the majority of cases, force declined less than 10 %.
Data analysis
For length traces that appeared linear, i.e. during isotonic shortening of maximally activated myocytes, velocities were determined using least-squares linear regression analysis. For records of isotonic shortening that were curvilinear, the records were well fitted by a single exponential of the following form:
L = Ae-k(t) + C, (1)
where, L is cell length at time t, A is initial length, k is the rate constant of decay, and C is the length at which velocity became zero. Velocity of shortening at any given time, t, was determined as the slope of the tangent to the fitted curve at that time point. For this study, velocities of shortening were calculated at t = 0 and at t = 100 ms.
Hyperbolic force-velocity curves were fitted to the relative force-velocity data using the Hill equation (Hill, 1938):
(P + a)(V + b) = (Po + a)b, (2)
where, P is force during shortening at velocity V; Po is the peak isometric force; and a and b are constants with dimensions of force and velocity, respectively. Power-load curves were obtained by multiplying force by velocity at each load on the force-velocity curve. The optimum force for mechanical power output (Fopt) was calculated using the equation (Woledge et al. 1985):
Fopt = (a2 + aPo)½ - a. (3)
Curve fitting was performed using a customized program written in Qbasic (Microsoft Corp., Redmond, WA, USA), as well as commercial software (Systat and Sigmaplot from SPSS/Jandel, Chicago, IL, USA).
To determine whether there were significant effects on velocity or power at a given load due to variations in Ca2+ concentration, paired t tests were performed. P < 0·05 was chosen as indicating significance. All values are expressed as means ±
Skinned myocyte preparation
In these studies single myocyte-sized preparations were attached to the mechanical apparatus using a modification of a method used previously on skinned skeletal muscle fibres (see Methods for details). Following attachment to the apparatus, myocytes (n = 13) were adjusted to a relaxed sarcomere length of
Table 1. Myocyte dimensional, force-velocity and power output characteristics during maximal activations (pCa 4·5)
The attachments to the ends of the preparations were of very low compliance, i.e. sarcomere lengths in the middle of the myocyte preparations decreased by less than 3 % in the transition from rest to full activation. Additionally, as shown in Fig. 1, following a rapid length change, force fell to zero when the preparation was slackened and after the myocyte actively shortened to take up the slack, force returned to approximately pre-slack levels of force. This result indicates that sarcomeres remained at lengths within the plateau of the length-tension relationship both before and after imposition of slack, as would be expected for a length change of this magnitude if end compliance was low. Compliance at the attachment points was also estimated by examining myocyte length traces during load clamps. As illustrated in Fig. 2A, there was an initial rapid phase of series elastic recoil indicated by the sudden decrease in length just prior to isotonic shortening. Figure 2B is a plot of the amount of series elastic recoil before isotonic shortening versus relative load for all load clamps that were performed on maximally activated myocyte preparations. This relationship appears to have two phases, a linear phase from approximately 1·0 to 0·5 Po and a curvilinear phase at forces < 0·5 Po. Series elastic recoil was estimated from the linear phase only, since the curvilinear phase probably includes cross-bridge turnover, i.e. during longer length steps it is likely that cross-bridges detach and re-attach due to the relatively long time required for these steps. We assume that for shorter length steps, force declines with shortening of elastic elements within the myofilaments as well as elastic elements in series with the myofilaments, i.e. end compliance. Extrapolation of the linear phase yielded an x-intercept of 2·5 % of myocyte preparation length, which corresponds to
A, relaxed myocyte (pCa 9·0); B, during maximal activation (pCa 4·5). As illustrated in the drawing (C, modified from Fig. 3, Metzger, Greaser & Moss, 1989), the myocyte preparation was mounted between a force transducer and a position motor. The ends of the myocyte were secured by overlaying a 0·5 mm piece of 4-0 suture, which were tied into troughs using loops of 10-0 suture. The length of myocyte between the two troughs in this preparation is 160 µm. D shows force traces during maximal activation of this myocyte preparation following a rapid change (
A, representative myocyte length (ML) trace and force clamp during isotonic shortening. The length change (indicated by the asterisk) associated with reducing force to the level shown was estimated by extrapolating the fitted straight line to the onset of the force clamp. The oscillation in the length trace prior to isotonic shortening is due to initial overshoot in the force clamp induced by the feedback circuit. B, plot of instantaneous length change versus relative load at the end of the length step. The relationship exhibits two phases, a linear phase for relative forces (P/Po) between 1·0 and 0·5 Po and a curvilinear phase for forces < 0·5 Po. The linear phase was used to estimate series elasticity of the myocyte preparations. Extrapolation of the linear phase approximates a series elasticity of 2·5 % myocyte preparation length.
Force-velocity and power-load curves
Force-velocity relationships were determined in skinned myocyte preparations by measuring shortening velocities during force clamps imposed by a servo-controlled feedback system (Sweitzer & Moss, 1993). Figure 3 presents representative length traces during isotonic shortening of a skinned cardiac myocyte preparation during maximal activation (pCa 4·5). Panel A presents length (upper) and force (lower) traces during an isotonic clamp of the myocyte preparation to 0·22 Po; the inset is an enlargement of the length trace during isotonic shortening. A straight line was well fitted through the length trace (correlation coefficient (r2) = 0·999) and thus, shortening velocity appeared to remain constant as the myocyte preparation shortened by a total of
A shows length (top) and force (bottom) traces during an isotonic clamp to 0·22 of peak force (Po); the inset shows an enlargement of the length trace during isotonic shortening. The straight line fit is superimposed on the length trace. B shows a load clamp to 0·75 Po in the same myocyte. In A and B, the myocyte preparation was rapidly slackened after the period of isotonic shortening and after 20 ms the preparation was rapidly relengthened to its original length.
Myocyte shortening at the level of the sarcomere was also approximately linear during force clamps imposed in skinned myocyte preparations. Figure 4 shows examples of myocyte length (ML), sarcomere length (SL) and force traces during a single isotonic clamp. Sarcomere length during isotonic shortening was monitored using a laser diffraction system. Myocyte shortening and sarcomere shortening are shown in Fig. 4B for a range of load clamps applied to this myocyte preparation. ML and SL shortening velocities were very similar for this myocyte preparation, although there was a tendency for ML shortening to overestimate SL shortening at low loads. Consequently, maximal ML shortening velocity appears to slightly overestimate maximal shortening rate at the level of the sarcomere. Overall, SL shortening was measured in three myocyte preparations, all during maximal activation. Shortening velocities were similar for both SL and ML, with the SL: ML shortening velocity ratio being 1·01 ± 0·12 (mean ±
A, both ML and SL exhibited linear shortening time courses during the load clamp. The regression coefficients for ML and SL were 0·996 and 0·985, respectively. B shows force-velocity points for both whole preparation shortening (
Force-velocity relationships were plotted from the raw data, and power-load curves were then constructed by multiplying shortening velocity by force for each load. A summary of the force-velocity and power-load curves is shown in Fig. 5 for thirteen maximally activated myocyte preparations. Shortening velocities during force clamps of these myocyte preparations yielded a mean hyperbolic force-velocity relationship that extrapolated to a maximal velocity (Vmax) of
Data were obtained from 13 skinned rat cardiac myocyte preparations during maximal activation. Data points in both force-velocity and power-load curves are means ±
V/Vmax = (1 - P/Po)/1 + (P/Po)(Po/a),
where Vmax and Po are maximum velocity of shortening and isometric tension, respectively, and a is a force constant.
Ca2+ dependence of force-velocity and power-load curves
Since myoplasmic Ca2+ levels can vary on a beat-to-beat basis in living heart and since Vmax varies with activation levels in both living and skinned myocardium, we hypothesized that Ca2+ would have direct effects on the power-output characteristics of myocardium. To test this idea power-load curves were characterized during half-maximal Ca2+ activations (half-maximal activations were produced using solutions with pCa ranging from 6·0 to 5·6). Reducing the level of activator Ca2+ resulted in several alterations in force-velocity and power output properties. While maximally activated myocyte preparations appeared to shorten at constant velocity during force clamps, this was not the case for half-maximally activated myocyte preparations in which shortening at low loads (< 25 % Po) was clearly curvilinear, i.e. there was continual slowing as shortening proceeded (Fig. 6).
The myocyte preparation shortened at a constant velocity during the entire load clamp during maximal activation (i.e. pCa 4·5). When [Ca2+] was reduced to yield ~50 % Po (i.e. pCa 5·8), the length trace was curvilinear during isotonic shortening. The length trace was fitted using a single exponential equation. The dashed line illustrates the deviation from linearity of the length trace. The dashed line was best-fitted by eye for ~50 ms of the length trace near the beginning of the load clamp.
Force-velocity and power output characteristics are presented in Table 2 for both maximal and submaximal Ca2+ activations in seven paired myocyte preparations. Figure 7 summarizes the features of the force-velocity and power-load curves at low levels of Ca2+ activation for these seven myocyte preparations. Shortening velocities during maximal activations were obtained by least-squares regression of straight lines through the length traces, while shortening velocities at submaximal activations were obtained by fitting single exponential curves through the length traces (see Fig. 6). Velocities during the linear phase at maximal activation were faster than initial shortening velocities at submaximal activations, and this difference was greater when velocities at submaximal activations were measured 100 ms after the onset of shortening. Vmax values extrapolated from the force-velocity curves of seven myocyte preparations were 1·15 ± 0·26 ML s-1 during maximal activation and slowed to 0·66 ± 0·22 and 0·56 ± 0·25 ML s-1 during submaximal activations when velocities were measured at 0 and 100 ms, respectively.
Ca2+ concentration also had significant effects on power- load curves. Cumulative plots in Fig. 7 show that maximum relative power output decreased when activator Ca2+ was reduced and velocities were measured after 100 ms shortening. Also, the force optimum for power output (Fopt) shifted to greater relative loads during submaximal Ca2+ activations compared with maximal Ca2+ activations (0·29 ± 0·04 versus 0·34 ± 0·04, P < 0·05, see Table 2). Since a given after-load presents a greater relative load to the heart at lower levels of myocardial activation, this shift in the force optimum implies that cross-bridge interaction properties are altered at low Ca2+ in a way that enhances the work rate of the heart for a given after-load (see Figs 7 and 8).
Table 2. Myocyte force-velocity and power output characteristics during maximal and submaximal Ca2+ activations
Plotted values are pooled results from 7 myocyte preparations. In A and B, the load at each step was expressed as a fraction of isometric tension for each myocyte preparation. In C, the load was normalized to force during maximal activation (i.e. pCa 4·5). For submaximal activations, length traces during load clamps were well fitted by a single exponential equation. Initial velocities of shortening at time 0 ms (i.e. immediately after the onset of the load clamp) are plotted in A. Shortening velocities at 100 ms following initiation of the load clamp are also plotted in A and these velocities were used to calculate power outputs, which are plotted in B and C.
Skinned myocyte preparation
Elucidation of structure-function relations for myofibrillar proteins in striated muscles has been enhanced by the use of single cell preparations. The use of single skeletal muscle fibres or cardiac myocytes minimizes the heterogeneities in protein content that are typical of multicellular muscle preparations. This is advantageous because inhomogeneities in protein content among the cells of a multicellular preparation will complicate the interpretation of experimental results, since it is then difficult to assign function to a single protein or a particular mix of proteins. Typically, single cells provide better uniformity of striation patterns during activations, allowing more accurate measurements of sarcomere length. Furthermore, skinning of single cells allows steady activation of the myofibrils with known concentrations of Ca2+, thus providing experimental control of the level of activation. In the present study, refinement of a procedure used earlier on skeletal muscle fibres has resulted in low compliance attachments of cardiac myocyte-sized preparations and maintenance of sarcomere spacing during mechanical measurements. The preparations developed
Isotonic shortening in myocardial preparations
Force-velocity relationships have been measured by several investigators in both skinned (Maughan et al. 1978; Brenner, 1980, 1986; Pagani & Julian, 1984) and intact (e.g. Sonnenblick, 1962; Brutsaert et al. 1973; De Clerk et al. 1977; Chiu et al. 1982, 1987; Daniels et al. 1984; de Tombe & ter Keurs, 1990, 1992) multicellular preparations of cardiac muscle. These measurements were made by clamping forces to loads less than maximal, resulting in shortening of the muscle at velocities which varied inversely with the size of the load. The Vmax values measured in this myocyte preparation (1·50 ± 0·62 ML s-1 at 12°C) are similar to those reported by Sweitzer & Moss (1993) (2·83 ML s-1, 15°C, rat skinned myocytes). The Vmax values in both studies using myocyte preparations are somewhat greater than Vmax values measured in living cardiac muscle when our cooler experimental temperatures are taken into account (Q10 = 4·6, de Tombe & ter Keurs, 1990). Vmax values previously reported from bundles of living myocardium include 2·34 ML s-1 (26°C, cat papillary muscles) (Chiu et al. 1987), 4·0 ML s-1 (25°C, rabbit papillary muscles) (Pagani & Julian, 1984), and
In previous experiments using multicellular preparations, muscle length traces during isotonic (constant force) shortening were observed to be curvilinear. The factors involved in the progressive slowing of velocity during shortening are unknown but may include non-cross-bridge-derived structural components such as (1) a passive extracellular elasticity that first facilitates shortening by unloading the contractile elements but then gradually slows shortening as the passive elements shorten and load is transferred to the cross-bridges (Brady, 1991), (2) viscoelasticity associated with compliant end attachments to the muscle (Seow & Ford, 1992; de Tombe & ter Keurs, 1990, 1992), i.e. in these earlier studies muscle length traces were curvilinear but sarcomere length traces appeared linear during isotonic shortening, and (3) an internal viscous load resulting from muscle structural proteins such as titin or deformations of the cytoskeleton (Brenner, 1980; de Tombe & ter Keurs, 1992; Helmes et al. 1996) Alternatively, curvature of length traces may originate from a cross-bridge-dependent internal load arising from cross-bridges that detach slowly during shortening even during maximal activations. For example, in large multicellular skinned preparations, substantial concentrations gradients of ADP and ATP may arise between the muscle core and surrounding solution. Non-uniform accumulation of ADP would tend to slow cross-bridge detachment, but only in populations of cross-bridges in the muscle core. During isotonic shortening, these slowly detaching cross-bridges may remain attached to a point where they are compressed and bear a load that opposes further shortening (for additional explanation of cross-bridge-dependent internal loads, see our discussion regarding submaximal Ca2+ activations).
Experiments in the present study were performed in part to assess the nature of the load that causes slowing of shortening velocity with time. Force clamps were imposed on single myocyte-sized preparations that were skinned to minimize extracellular elements that might contribute to load bearing or give rise to restoring forces during shortening. These myocyte preparations exhibited very low end compliance, which has previously been shown to contribute to curvilinear shortening traces in skeletal muscle (Seow & Ford, 1992). In the preparations used here, isotonic shortening velocity appeared to be constant throughout shortening during maximal activations. These observations suggest that curvilinear shortening, previously reported in maximally activated multicellular cardiac muscle preparations, results in large part from extracellular viscoelastic elements, high end compliance, or perhaps ATP and ADP concentration gradients that are not present in the much smaller myocyte preparation. Interestingly, several studies using single skinned skeletal muscle fibres also report approximately linear length traces during isotonic shortening while maximally activated (Julian, 1971; Julian & Moss, 1981; Moss, 1982).
The amount of activator Ca2+ that enters the myoplasm during excitation-contraction coupling is an important beat-to-beat regulator of ventricular stroke volume. Thus, we also examined whether reductions in Ca2+-activated force altered force-velocity and power-load curves in myocyte preparations. We observed that shortening velocities were slower at virtually all relative loads when activator Ca2+ was reduced to yield half-maximal force. This result implies that there is a Ca2+-dependent mechanism which slows cross-bridge cycling kinetics in myocardium. For example, slower cycling may arise from reductions in the level of thin filament activation associated with lower Ca2+ concentrations and fewer cross-bridges bound to the thin filament. Our results also indicate that both absolute and relative (i.e. normalized to maximal force) power outputs are significantly reduced when activator Ca2+ is adjusted to levels that yield
A, force-velocity curves are shown for a myocyte during maximal and submaximal Ca2+ activation. An elevation in [Ca2+] doubles the maximum force-generating capacity (i.e. force increases from 0·5 to 1·0 P4·5) and increases the curvature of the force-velocity relationship. Since force increased 2-fold, an after-load that was 40 % of isometric force at low [Ca2+] force becomes only 20 % of isometric force produced at high [Ca2+]. B shows power-load curves normalized to low [Ca2+] force and high [Ca2+] force. The force optimal for power output is 0·40 of low [Ca2+] isometric force and 0·20 of high [Ca2+] isometric force as a result of alterations in the shape of the force-velocity curves. C also shows power-load curves at low and high [Ca2+], but in this case normalized only to high [Ca2+] force. This illustrates that power output is optimal at the same absolute load during both low and high [Ca2+] activations. Thus, when myocardium is working against the same after-load, changes in force-velocity properties in response to varying [Ca2+] have the effect of maintaining power output near optimal levels despite altered contractility.
Reducing the level of activator Ca2+ also had a marked effect on the time course of myocyte shortening during force clamps, i.e. myocyte shortening traces at low loads were curvilinear in submaximally activated myocyte preparations. (Sarcomere length shortening was not measured during submaximal activations.) Such a phenomenon could contribute to curvilinear shortening records observed previously in living cardiac muscle, since the level of activation during a twitch might never be maximal. One possible cause of curvilinear length traces during submaximal activations of cardiac myocytes is the presence of a fixed internal load (de Tombe & ter Keurs, 1992; Helmes et al. 1996), presumably due to passive structural elements. If there is a fixed internal load, it is likely to be small since length traces during force clamps were well fitted by straight lines in maximally activated myocytes. However, the presence of a fixed internal load may become significant in high pCa solutions where forces are submaximal and the internal load is therefore a greater proportion of active force. It has been hypothesized that the myofibrillar protein titin (de Tombe & ter Keurs, 1992; Helmes et al. 1996) or deformations of the cytoskeleton (Brenner, 1980) are sources of an internal viscoelastic load in myocardium that progressively slows shortening velocity. Further experiments are required to assess whether either of these two factors influence sarcomere shortening velocity in rat myocardium.
Another plausible mechanism for curvilinear shortening during submaximal activation is an internal load that varies as a function of the level of thin filament activation. Our laboratory (Moss, 1992) has proposed that long-lived or slowly detaching cross-bridges give rise to an activation-dependent internal load in skeletal muscles, presumably as a result of reduced levels of thin filament activation. According to this idea, at the onset of myocyte shortening these long-lived cross-bridges, like normal cross-bridges, would bear only a small compressive load opposing shortening. However, with further shortening these cross-bridges would remain attached to the thin filament and would eventually be strained to bear a considerable load that opposes contraction and thereby progressively slows shortening velocity. This model is consistent with the findings of the present study in which curvilinear shortening in cardiac myocytes was observed only during submaximal activations and only at low loads where there would be sufficient shortening to manifest a load opposing filament sliding. Based on shortening records we estimate that a shortening of
While Ca2+ dependence of power output has not previously been directly measured in skinned cardiac muscle, the effects of Ca2+ on isotonic shortening velocity has been assessed in several studies, although the nature of these effects has been variable. Lowering [Ca2+] has been reported to both slow (Maughan et al. 1978; Brenner, 1980) and accelerate isotonic shortening velocities (Sweitzer & Moss, 1993) during force clamps. A potential source of variability in these results is the curvilinear nature of the shortening traces. For instance, if velocity is measured late in the length trace, then increases in curvature would result in slower shortening velocity and may be interpreted as reduced rate of cross-bridge cycling. As suggested by Brenner (1986), changes in curvature might be caused by non-cross-bridge components, so that only the initial speed of shortening directly reflects kinetics of cross-bridge cycling. However, even when the initial speed of shortening is examined, lowered [Ca2+] has been reported to have variable effects on isotonic shortening velocities (Maughan et al. 1978; Brenner, 1980; Sweitzer & Moss, 1993).
The effects of maximum power output following changes in activator [Ca2+] have been examined in single skinned psoas fibres from rabbit (Ford et al. 1991). Normalized maximum power output was unchanged when peak force was lowered to 50 % by reducing [Ca2+]. This result suggests that maximum power (normalized to maximum force) does not vary with [Ca2+], at least when changes in force generating capability are taken into account. Our findings from cardiac myocyte preparations differ from these earlier results in skeletal muscle (Ford et al. 1991) and those in cardiac muscle which show faster (Sweitzer & Moss, 1993) isotonic shortening velocity due to decreases in [Ca2+]. The reason(s) for these differences may involve the elimination of passive viscoelastic elements and/or lower end compliance of the myocyte preparation used in this study. In any case, the myofibrillar mechanisms by which Ca2+ alters loaded shortening velocity, power output and the load optimum for maximum power output in cardiac myocytes remain to be elucidated.
This work was supported by National Institutes of Health grant R01 HL57852 (K. S. M).
Corresponding author
K. S. McDonald: Department of Physiology, School of Medicine, University of Missouri, Columbia, MO 65212, USA.
Email: kerry-mcdonald{at}muccmail.missouri.edu
This article has been cited by other articles:
RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References
2·35 µm, corresponding to mean (± 10 % of maximal force, which was determined in a separate population of myocyte preparations.
Cell length (µm) Cell width (µm) SL (µm) Maximal force (µN) Maximal tension (kN m-2) Vmax (ML s-1) a/Po Force optimum (P/Po) Power output (µW mg-1) 164 ± 29 21 ± 3 2·30 ± 0·08 7·8 ± 2·4 26 ± 11 1·50 ± 0·62 0·19 ± 0·11 0·28 ± 0·05 2·65 ± 1·17 29 nm per half-sarcomere. Since the elasticity of the myofilaments in skeletal muscle is estimated to be 5·4 nm per half-sarcomere (Higuchi et al. 1995), average series end compliance is estimated at 24 nm per half-sarcomere, which corresponds to 2 % of myocyte preparation length. Taken together, these findings indicate that this cardiac myocyte preparation has low end compliance and exhibits excellent mechanical integrity for studying contractile properties of skinned cardiac myocytes.
View larger version
[in this window]
[in a new window]
Figure 1. Photomicrographs of a single rat skinned cardiac myocyte preparation
) in myocyte length. The maximal force generated by this preparation was 9·5 µN.
View larger version
[in this window]
[in a new window]
Figure 2. Estimation of myocyte preparation end compliance
14 µm or 6·7 % of its initial length. Panel B shows a force clamp to 0·75 Po applied to the same myocyte preparation, which yielded an r2 value of 0·969 for the regression line fitted to the length trace.
View larger version
[in this window]
[in a new window]
Figure 3. Length traces during isotonic shortening in a maximally activated myocyte preparation
View larger version
[in this window]
[in a new window]
Figure 4. Myocyte length (ML), sarcomere length (SL), and tension during a load clamp
) and sarcomere shortening (
) during the same load clamps, indicating that changes in myocyte preparation length during isotonic shortening closely reflect shortening at the level of the sarcomere.
2 ML s-1 (Fig. 5). As summarized in Table 1, the mean Vmax for these thirteen myocyte preparations was 1·50 ± 0·62 ML s-1. The mean maximum relative power output was 0·10 ± 0·03, and when expressed in absolute units was 2·65 ± 1·17 µW mg-1, a value similar to that previously reported for myocardium (Chiu et al. 1987) when adjusted for temperature (assuming a Q10 temperature coefficient of 4·6; de Tombe & ter Keurs, 1990). The load at which power output was optimum was 28 ± 5 % of the maximum isometric force produced by these preparations.
View larger version
[in this window]
[in a new window]
Figure 5. Cumulative force-velocity (
) and power output () relationships
View larger version
[in this window]
[in a new window]
Figure 6. Effects of Ca2+ on shortening during isotonic load clamps in a rat skinned cardiac myocyte preparation
Relative force (P/Po) Vmax (ML s-1) a/Po Force optimum (P/Po) Power output (µW mg-1) 1·00 1·15 ± 0·26 0·20 ± 0·12 0·29 ± 0·04 2·47 ± 0·13 0·49 ± 0·04 * 0·56 ± 0·25 * 0·37 ± 0·17 * 0·34 ± 0·04 * 0·84 ± 0·56 *
View larger version
[in this window]
[in a new window]
Figure 7. Plots of force-velocity and power-load data obtained during maximal (pCa 4·5) and submaximal activations (P/Po = 0·49 ± 0·04)
DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References
8 µN maximal force at 12°C, which corresponds to 26 kN m-2 when normalized to myocyte cross-sectional area. These values are similar to tensions reported in other single myocyte preparations. Our laboratory and others have previously reported tensions ranging from 22 to 43 kN m-2 at 15°C for single myocytes attached using adhesives (Sweitzer & Moss, 1990, 1993; Strang et al. 1994; Hofmann & Lange, 1994; McDonald & Moss, 1995). Fabiato & Fabiato (1975) reported tensions of 80 kN m-2 at 23°C in rat myocardial preparations of approximately half the width of the myocyte preparations in this study. Maximal tensions reported in skinned cardiac trabeculae preparations include 100 kN m-2 at 15°C (Wolff et al. 1995), 63 kN m-2 at 15°C (Janssen & de Tombe, 1997), 58 kN m-2 at 21°C (Kentish & Stienen, 1994) and 100 kN m-2 at 22°C (Gao et al. 1994). The reasons for the differences in tension measurements are uncertain, but may include inaccurate estimates of preparation cross-sectional area, differences in solutions, and/or variable myofibrillar densities between preparations.
6 ML s-1 (25°C, rat trabeculae muscles) (de Tombe & ter Keurs, 1990). The greater Vmax values in skinned muscle preparations may be due in part to the swelling of the myofibrillar lattice which is a consequence of muscle permeabilization, a phenomenon that has been reported to reduce muscle stiffness during shortening (Goldman & Simmons, 1986). Additional factors to explain the different Vmax values are differences in species, experimental solutions and conditions, and perhaps higher levels of Ca2+ activations in skinned preparations than can be obtained in living myocardial preparations.
50 % of peak force. In addition, the force optimum for power output shifted to higher relative loads when activator Ca2+ was reduced. Physiologically, this phenomenon may help to maintain near-optimal myocardial efficiency during contractions against similar absolute after-loads even when the level of activation is reduced (see Fig. 8).
View larger version
[in this window]
[in a new window]
Figure 8. Schematic representation of the effect of contractility on force-velocity curves and its implications for myocardial performance
50 nm per half-sarcomere is required before these long-lasting cross-bridges produce a load that is sufficient to oppose shortening and cause myocyte length traces to deviate from linearity.
REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References
Acknowledgements
Brady, A. J. (1991). Mechanical properties of isolated cardiac myocytes. Biological Reviews 71, 413-428.
Brenner, B. (1980). Calcium activation and maximum unloaded shortening velocity. Investigations on glycerinated skeletal and heart muscle preparations. Basic Research in Cardiology 75, 40-46.
[Medline]
Brenner, B. (1986). The necessity of using two parameters to describe isotonic shortening velocity of muscle tissues: The effect of various interventions upon initial shortening velocity (vi) and curvature (b). Basic Research in Cardiology 81, 54-69.
[Medline]
Brutsaert, D. L., Claes, V. A. & Goethals, M. A. (1973). Effect of calcium on force-velocity relations of heart muscle of the cat. Circulation Research 32, 385-392
[Medline]
Chiu, Y. C., Ballou, E. W. & Ford, L. E. (1982). Internal viscoelastic loading in cat papillary muscle. Biophysical Journal 40, 109-120
[Abstract]
Chiu, Y. C., Ballou, E. W. & Ford, L. E. (1987). Force, velocity, and power changes during normal and potentiated contractions of cat papillary muscle. Circulation Research 60, 446-458
[Abstract]
Daniels, M., Noble, M. I. M., ter Keurs, H. E. D. J. & Wohlfart, B. (1984). Velocity of sarcomere shortening in rat cardiac muscle: Relationship to force, sarcomere length, calcium and time. The Journal of Physiology 355, 367-381
[Abstract]
De Clerk, N. M., Claes, V. A. & Brutsaert, D. I. (1977). Force velocity relations of single cardiac muscle cells. Journal of General Physiology 69, 221-241.
[Abstract]
de Tombe, P. P. & ter Keurs, H. E. D. J. (1990). Force and velocity of sarcomere shortening in trabeculae from rat heart: Effects of temperature. Circulation Research 66, 1239-1254
[Abstract]
de Tombe, P. P. & ter Keurs, H. E. D. J. (1992). An internal viscous element limits unloaded velocity of sarcomere shortening in rat myocardium. The Journal of Physiology 454, 619-642
[Abstract]
Fabiato, A. (1988). Computer programs for calculating total from specified free or free from specified total ionic concentrations in aqueous solutions containing multiple metals and ligands. Methods in Enzymology 157, 378-417
[Medline]
Fabiato, A. & Fabiato, F. (1975). Contractions induced by a calcium-triggered release of calcium from sarcoplasmic reticulum of single skinned cardiac cells. The Journal of Physiology 249, 469-496
[Abstract]
Ford, L. E., Nakagawa, K., Desper, J. & Seow, C. Y. (1991). Effect of osmotic compression force-velocity properties of glycerinated rabbit skeletal muscle cells. Journal of General Physiology 97, 73-88
[Abstract]
Gao, W. D., Backx, P. H., Azan-Backx, M. D. & Marban, E. (1994). Myofilament Ca2+-sensitivity in intact versus skinned rat ventricular muscle. Circulation Research 74, 408-415
[Abstract]
Goldman, Y. E. & Simmons, R. M. (1986). The stiffness of frog skinned muscle fibres at altered lateral filament spacing. The Journal of Physiology 378, 175-194
[Abstract]
Helmes, M., Trombitàs, K. & Granzier, H. (1996). Titin develops restoring force in cardiac myocytes. Circulation Research 79, 619-626
[Abstract/Full Text]
Higuchi, H., Yanagida, T. & Goldman, Y. E. (1995). Compliance of thin filaments in skinned fibers of rabbit skeletal muscle. Biophysical Journal 69, 1000-1010
[Abstract]
Hill, A. V. (1938). The heat of shortening and the dynamic constants of muscle. Proceedings of the Royal Society B 126, 136-195.
Hofmann, P. A. & Lange, J. H. (1994). Effects of phosphorylation of troponin I and C protein of isometric tension and velocity of unloaded shortening in skinned single cardiac myocytes from rats. Circulation Research 74, 718-726
[Abstract]
Janssen, P. M. L & de Tombe, P. P. (1997). Protein kinase A does not alter unloaded velocity of sarcomere shortening in skinned rat cardiac trabeculae. American Journal of Physiology 275, H2415-2422.
Julian, F. J. (1971). The effect of calcium on the force-velocity relation of briefly glycerinated frog muscle fibres. The Journal of Physiology 218, 117-145
[Medline]
Julian, F. J. & Moss, R. L. (1981). Effects of calcium and ionic strength on shortening velocity and tension development in frog skinned muscle fibres. The Journal of Physiology 311, 179-199
[Abstract]
Kentish, J. C. & Stienen, G. J. M. (1994). Differential effects of length on maximum force production and myofibrillar ATPase activity in rat skinned cardiac muscle. The Journal of Physiology 475, 175-184
[Abstract]
McDonald, K. S. & Moss, R. L. (1995). Osmotic compression of single cardiac myocytes eliminates the reduction in Ca2+ sensitivity of tension at short sarcomere length. Circulation Research 77, 199-205
[Abstract/Full Text]
McDonald, K. S., Wolff, M. R. & Moss, R. L. (1996a). Length traces during isotonic shortening are linear in single myocyte-sized preparations. Biophysical Journal 70, A47.
McDonald, K. S., Wolff, M. R. & Moss, R. L. (1996b). Effects of calcium on power output of skinned cardiac myocytes. Journal of Molecular and Cellular Cardiology 28, A139.
McDonald, K. S., Wolff, M. R. & Moss, R. L. (1997). Sarcomere length dependence of the rate of force redevelopment and Ca2+ sensitivity of tension in fast and slow skinned skeletal muscle fibres. The Journal of Physiology 501, 607-621
[Abstract]
Maughan, D. W., Low, E. S. & Alpert, N. R. (1978). Isometric force development, isotonic shortening, and elasticity measurements from Ca2+-activated ventricular muscle of the guinea pig. Journal of General Physiology 71, 431-451
[Medline]
Metzger, J. M., Greaser, M. L. & Moss, R. L. (1989). Variations in cross-bridge attachment rate and tension with phosphorylation of myosin in mammalian skinned skeletal muscle fibers. Journal of General Physiology 83, 855-883.
Moss, R. L. (1979). Sarcomere length-tension relations of frog skinned muscle fibres during calcium activation at short lengths. The Journal of Physiology 292, 177-202
[Abstract]
Moss, R. L. (1982). The effect of calcium on the maximum velocity of shortening in skinned skeletal muscle fibers of the rabbit. Journal of Muscle Research and Cell Motility 3, 295-311
[Medline]
Moss, R. L. (1992). Ca2+ regulation of mechanical properties of striated muscle: Mechanistic studies using extraction and replacement of regulatory proteins. Circulation Research 70, 865-884
[Abstract]
Pagani, E. D. & Julian, F. J. (1984). Rabbit papillary muscle myosin isozymes and the velocity of muscle shortening. Circulation Research 54, 586-594
[Abstract]
Seow, C. Y. & Ford, L. E. (1992). Contribution of damped passive recoil to the measured shortening velocity of skinned rabbit and sheep muscle fibres. Journal of Muscle Research and Cell Motility 13, 295-307
[Medline]
Sonnenblick, E. H. (1962). Force-velocity relations in mammalian heart muscle. American Journal of Physiology 202, 931-939.
Strang, K. T., Sweitzer, N. K., Greaser, M. L. & Moss, R. L. (1994).
-Adrenergic receptor stimulation increases unloaded shortening velocity (Vo) of rat skinned single ventricular myocytes. Circulation Research 74, 542-549[Abstract]
Sweitzer, N. K. & Moss, R. L. (1990). The effect of altered temperature on Ca2+-sensitive force in permeabilized myocardium and skeletal muscle: evidence for force dependence of thin filament activation. Journal of General Physiology 96, 1221-1245
[Abstract]
Sweitzer, N. K. & Moss, R. L. (1993). Determinants of loaded shortening velocity in single cardiac myocytes permeabilized with
-hemolysin. Circulation Research 73, 1150-1162[Abstract]
Woledge, R. C., Curtin, N. A. & Homsher, E. (1985). Energetic Aspects of Muscle Contraction, pp. 47-71. Academic Press, London.
Wolff, M. R., McDonald, K. S. & Moss, R. L. (1995). Rate of tension development in cardiac muscle varies with level of activator calcium. Circulation Research 76, 154-160
[Abstract/Full Text]
J. G. Jacot, A. D. McCulloch, and J. H. Omens
Substrate Stiffness Affects the Functional Maturation of Neonatal Rat Ventricular Myocytes
Biophys. J.,
October 1, 2008;
95(7):
3479 - 3487.
[Abstract]
[Full Text]
[PDF]
S. A. Niederer, P. J. Hunter, and N. P. Smith
A Quantitative Analysis of Cardiac Myocyte Relaxation: A Simulation Study
Biophys. J.,
March 1, 2006;
90(5):
1697 - 1722.
[Abstract]
[Full Text]
[PDF]
F. S. Korte, T. J. Herron, M. J. Rovetto, and K. S. McDonald
Power output is linearly related to MyHC content in rat skinned myocytes and isolated working hearts
Am J Physiol Heart Circ Physiol,
August 1, 2005;
289(2):
H801 - H812.
[Abstract]
[Full Text]
[PDF]
R. S. Haworth, F. Cuello, T. J. Herron, G. Franzen, J. C. Kentish, M. Gautel, and M. Avkiran
Protein Kinase D Is a Novel Mediator of Cardiac Troponin I Phosphorylation and Regulates Myofilament Function
Circ. Res.,
November 26, 2004;
95(11):
1091 - 1099.
[Abstract]
[Full Text]
[PDF]
K. S. Campbell and R. L. Moss
SLControl: PC-based data acquisition and analysis for muscle mechanics
Am J Physiol Heart Circ Physiol,
December 1, 2003;
285(6):
H2857 - H2864.
[Abstract]
[Full Text]
[PDF]
F. S. Korte, K. S. McDonald, S. P. Harris, and R. L. Moss
Loaded Shortening, Power Output, and Rate of Force Redevelopment Are Increased With Knockout of Cardiac Myosin Binding Protein-C
Circ. Res.,
October 17, 2003;
93(8):
752 - 758.
[Abstract]
[Full Text]
[PDF]
S.-i. Yasuda, S. Sugiura, H. Yamashita, S. Nishimura, Y. Saeki, S.-i. Momomura, K. Katoh, R. Nagai, and H. Sugi
Unloaded shortening increases peak of Ca2+ transients but accelerates their decay in rat single cardiac myocytes
Am J Physiol Heart Circ Physiol,
July 11, 2003;
285(2):
H470 - H475.
[Abstract]
[Full Text]
[PDF]
G. M. Diffee and E. Chung
Altered single cell force-velocity and power properties in exercise-trained rat myocardium
J Appl Physiol,
May 1, 2003;
94(5):
1941 - 1948.
[Abstract]
[Full Text]
[PDF]
T. J. Herron, F. S. Korte, and K. S. McDonald
Loaded shortening and power output in cardiac myocytes are dependent on myosin heavy chain isoform expression
Am J Physiol Heart Circ Physiol,
September 1, 2001;
281(3):
H1217 - H1222.
[Abstract]
[Full Text]
[PDF]
G. M. Diffee, E. A. Seversen, and M. M. Titus
Exercise training increases the Ca2+ sensitivity of tension in rat cardiac myocytes
J Appl Physiol,
July 1, 2001;
91(1):
309 - 315.
[Abstract]
[Full Text]
[PDF]
R. L. Moss
Plasticity in the Dynamics of Myocardial Contraction : Ca2+, Crossbridge Kinetics, or Molecular Cooperation
Circ. Res.,
April 16, 1999;
84(7):
862 - 865.
[Full Text]
[PDF]
T. J. Herron, F. S. Korte, and K. S. McDonald
Power Output Is Increased After Phosphorylation of Myofibrillar Proteins in Rat Skinned Cardiac Myocytes
Circ. Res.,
December 7, 2001;
89(12):
1184 - 1190.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Full Text (PDF)
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Citing Articles
Citing Articles via HighWire
Citing Articles via Google Scholar
Google Scholar
Articles by McDonald, K. S.
Articles by Moss, R. L.
Search for Related Content
PubMed
PubMed Citation
Articles by McDonald, K. S.
Articles by Moss, R. L.
HOME