Triggered Propagated Contractions in Rat Cardiac Trabeculae
Inhibition by Octanol and Heptanol
We studied the role of Ca2+ diffusion through gap junctions (GJs) in triggering and propagation of damage-induced contractions in cardiac muscle (TPCs) by evaluating effects of the GJ blockers octanol and heptanol (O&H) on TPCs. TPCs were elicited in trabeculae from rat right ventricle superfused with Krebs-Henseleit solution at 20°C and 0.7 to 1.75 mmol/L [Ca2+]o. Force was measured with a silicon strain gauge; sarcomere length, by laser diffraction techniques. O&H (3 to 300 μmol/L) decreased force, propagation velocity, and triggering rate of TPCs in a dose-dependent manner. At 300 μmol/L, O&H decreased TPC force to 21.3% and 25.7%, propagation velocity to 15.4% and 13.0%, and triggering rate to 26.5% and 25.7%. At 300 μmol/L, O&H decreased twitch force to 79.0% and 77.8% and reduced time to 90% relaxation by 10% to 15%. Above 1 mmol/L, O&H abolished twitch force and TPCs. Image analysis of spread of the fluorescence profile of microinjected fura 2 salt revealed an effective diffusion coefficient for fura 2 of 21.0±3.3 μm2/s, which decreased to 12.6±1.5 and 7.07±0.7 μm2/s after 1 and 3 hours of exposure, respectively, to 100 μmol/L octanol, with a time constant of decline of 1.5±0.5 hours. These results are consistent with the hypothesis that propagation of TPCs is due to Ca2+-induced Ca2+ release mediated by Ca2+ diffusion from cell to cell through GJs. Reduction of propagation velocity reduces the number of activated sarcomeres in the TPC, which reduces TPC force. O&H slow triggering of TPCs, presumably by blocking Ca2+ diffusion from myocytes within damaged areas to adjacent normal cells.
Acute damage to cardiac muscle induces local contractions that follow the regular twitch and then travel from the damaged region into the adjacent myocardium.1 These so-called TPCs have been observed in rat and human cardiac trabeculae.1 2 3
We have previously shown that TPCs do occur in human myocardium after acute damage, even at body temperature.3 The occurrence of TPCs in acutely damaged myocardium is of possible clinical importance, because TPCs are accompanied by DADs,4 5 which are presumably due to activation of transient inward currents.6 The DADs can reach threshold potential, trigger an action potential, and induce triggered twitches. DADs have usually been related to nondriven rhythmic activity, such as in triggered arrhythmias in conditions of cellular Ca2+ overload due to digitalis toxicity.1 3 However, the observation that TPCs and DADs are generated in acutely damaged myocardium suggests an important new mechanism of initiation of arrhythmias: a single premature action potential induced by the DAD near damaged cells may induce reentry arrhythmias in vulnerable myocardium, such as that found in ischemic heart disease.
TPCs commonly arise in regions of trabeculae that are damaged (eg, regions damaged by the dissection procedure).2 Damage to the muscle always causes spontaneous activity in cells in the damaged region and also in adjacent cells. Spontaneous activity reduces twitch force development,7 so that during twitch contraction the damaged region is stretched by viable cells in the healthy region; subsequently, the damaged region is rapidly released during twitch relaxation. We have shown that triggering of TPCs is abolished if stretch and rapid mechanical release of the damaged part of the muscle by the preceding twitch are eliminated.2 Hence, a plausible mechanism of triggering of TPCs would be that the rapid mechanical release during twitch relaxation leads to Ca2+ dissociation from the myofilaments in the cells of the damaged region,8 which in turn would induce Ca2+ release from the SR of adjacent cells.
Propagation of TPCs along the undamaged part of the trabeculae was shown to occur at constant velocities that ranged from 0.1 to 15 mm/s.1 The observation that TPCs travel at constant velocity through undamaged muscle led us to postulate that they are caused by CICR from the SR mediated by diffusion of Ca2+ to adjacent SR.1 Model simulation of the propagation of TPCs on the basis of CICR mediated by diffusion of Ca2+ to adjacent SR within the cell and to SR in adjacent cells1 9 predicted propagation velocities that were in agreement with the observed data. Propagation of spontaneous localized contractions as well as waves of increased cytosolic free Ca2+ at velocities ranging from 0.05 usually to 0.2 mm/s in single cells10 11 12 13 14 15 16 17 18 and as high as 3 mm/s (M. Miura, unpublished data, 1996) has also been explained on the basis of CICR from SR mediated by diffusion of Ca2+ to adjacent SR.19 20 The observed spatial and temporal patterns of propagation of [Ca2+]i waves by Takamatsu and Wier15 are consistent with this hypothesis.
It is clear that both triggering and propagation of TPCs require diffusion of Ca2+ through GJs. Cardiac GJ channels share a basic architectural design with other transmembrane ion channels, yet they possess a set of unique properties, ie, low selectivity and a high permeability, allowing them to coordinate cellular function.21 22 GJs respond dynamically to a variety of regulatory factors, ie, Ca2+, pH, hormones, phosphorylation due to second messengers, and transjunctional voltage. GJs are sensitive to pharmacological agents such as octanol and heptanol, although these agents are not perfectly selective.21 22 23 Octanol and heptanol were first discovered to uncouple crayfish axons.24 The rapid and reversible uncoupling action of these alcohols on cardiac cells,25 26 27 presumably resulting from a direct interaction with the lipid bilayer in the membrane,21 28 is due to a decreased open probability of GJ channels28 rather than to a decreased conductance of a single GJ channel. Hence, a decreased open probability of the GJs would be expected to reduce the rate of diffusion of Ca2+ from cell to cell and, therefore, to reduce the propagation velocity and the rate of triggering of the TPCs. In the present study, we have tested the hypothesis that triggering and propagation of TPCs requires Ca2+ diffusion through GJs by investigating the effects of octanol and heptanol on TPCs.
Materials and Methods
Dissection and Mounting of the Preparation
Brown-Norway rats of either sex, 0.20 to 0.25 kg body weight, were anesthetized with diethyl ether. The hearts were rapidly removed and perfused through the aorta. Fourteen trabeculae with a slack length of 2.04±0.075 mm, width of 0.26±0.039 mm, and thickness of 0.13±0.013 mm (mean±SEM) running between the free wall of the right ventricle and the atrioventricular ring were dissected and mounted horizontally between a silicon strain-gauge force transducer (model AE 801, SensoNor) and a motor arm in a 0.5-mL bath. Both force transducer and motor arm were controlled by micromanipulators in order to position the muscle and to adjust sarcomere length.
During dissection and during the experiments, the preparation was superfused with KH solution bubbled with 95% O2/5% CO2. KH solution contained the following (mmol/L): NaCl 120, KCl 5 (20 KCl in the dissection solution), Na2SO4 1.2, NaH2PO4·2H2O 2, MgCl2·6H2O 1.2, NaHCO3 19, and glucose 10, along with various concentrations of CaCl2 as specified below. The pH in the solution at 20°C and 26°C was 7.38 and 7.45, respectively.
The procedures used in the present study have been described before1 ; in short, the trabeculae were stimulated through parallel platinum electrodes in the bath with rectangular pulses (5 milliseconds, twice the threshold) from a stimulator (model SD9, Grass Instrument Co) triggered by a computer (PC-AT, IBM). Laser diffraction techniques were used to measure sarcomere length in the illuminated area (He-Ne laser; cross section of the beam, 350 μm). An inverted microscope (Diaphot-TMD, Nikon Inc) and a video system (camera model WV3170 and recorder model AG 2400, Panasonic) were used to observe the trabeculae. The microscope stage together with the bath could be moved with respect to the laser beam so that sarcomere length could be measured at different sites along the trabecula; the distance between the selected sites could be measured by a linear potentiometer attached to the microscope stage. Force and sarcomere length were displayed on a storage oscilloscope (model V134, Hitachi), recorded with a chart recorder (model 2800S, Gould), and sampled by analog-digital converter (model DT 2801A, Data Translation Inc) installed in the computer. All recordings were stored on hard disk for later analysis. The computer program for data analysis could display the record of the last twitch of the conditioning stimulus train and a subsequent period on a graphics monitor (model CGA, IBM).
The muscles were initially allowed to stabilize in KH solution with [Ca2+]o of 0.7 mmol/L at 26°C in the bath at a sarcomere length of 2.15 μm, such that passive force was 5% of active twitch force. After equilibration, [Ca2+]o was changed to 0.35 mmol/L, and the temperature was lowered to 19°C to 21°C. The trabeculae were then stimulated with trains of 15 stimuli at a rate of 2 Hz, interspersed with 15-second rest intervals. This stimulation protocol has previously been shown to induce TPCs after the last stimulated twitch1 ; if TPCs were not elicited by this protocol, [Ca2+]o was increased in steps of 0.2 mmol/L. Usually, the [Ca2+]o required to elicit TPCs was 0.5 to 0.7 mmol/L. Measurements were started 15 minutes after beginning of the protocol, when we had verified that the TPCs were reproducible.
Data were obtained within the first 2 hours, since the properties of the TPCs appeared to be constant in this period.1 These experiments were performed at lowered temperature (20°C) in order to increase the likelihood of stable TPCs.1 TPCs can be observed after damage to cardiac muscle at body temperature but then disappear within 10 minutes.5 Before the measurement of the effect of octanol and heptanol, [Ca2+]o was increased to 0.7 to 1.75 mmol/L, which increased the velocity of propagation of the TPCs to ≈5 mm/s, so that accurate assessment of the effect of the drugs on properties of the TPCs was possible. The data were collected within 15 minutes after addition of octanol or heptanol at every concentration. Stock solution of octanol and heptanol (Sigma Chemical Co) were prepared in 95% ethanol and added directly to the KH solution to determine a cumulative concentration-response relation. The final concentration of ethanol in any solution never exceeded 0.05%.
The recordings of force and sarcomere length of TPCs stored on computer were used to measure twitch force, peak force of the last twitch in the stimulus train, force produced by the TPC, TPC propagation velocity, and triggering rate.
Propagation velocity was calculated from the interval between peak sarcomere shortening due to a TPC at two sites of the trabecula and the distance between the two sites.29 The time required for the underlying processes to trigger a propagated contraction is reflected by the latency of development of TPCs. The latency of the TPC can be represented by (1) time to peak TPC force: the time between the last stimulus and the peak of the TPC force; (2) time to onset of TPC force: the time between the beginning of the last stimulus and the beginning of the TPC force; (3) time from 75% twitch relaxation to onset of TPCs; and (4) calculated time: the calculated time of TPC start in the end region could be obtained from t−x/v (where t is time to peak sarcomere shortening during a TPC at site X, x is the distance between site X and the end region from which the TPC started, and v is propagation velocity of the TPC). Previous studies have shown that the calculated time of the start of the TPC in the damaged end region is closely correlated with time to peak TPC force,29 indicating that time to peak TPC force provides a reliable measure of TPC latency.1 2 3 However, both time to peak TPC force and time to onset of TPC force include the duration of the last twitch, which might be changed by octanol, heptanol, or other interventions. Moreover, the time to peak TPC force is affected by propagation velocity, which is influenced by the drugs. Therefore, we considered time from 75% relaxation of the last twitch to the onset of TPC force to be the most accurate for the evaluation of TPC triggering latency. In order to compare the behavior of the triggering process with the rate of propagation of TPCs, we used in this study the inverse of latency (1/latency), or triggering rate.
Fura 2 Diffusion in Trabeculae
Thirteen thin trabeculae (length, 2.4±0.41 mm; width, 200±57 μm; and thickness, 85±8.6 μm [mean±SD]) were studied in order to evaluate the rate of diffusion of fura 2 in the muscles before and after exposure to 100 μmol/L octanol. The muscles were mounted between a force transducer and a micromanipulator in a perfusion bath located on the stage of an inverted microscope (Nikon). The trabeculae were stimulated at 0.5 Hz and were superfused by the bicarbonate-buffered KH solution (26.3±0.2°C). During superfusion with KH solution, fura 2 potassium salt was microinjected iontophoretically into trabeculae using 10 nA of negative current for 10 minutes, as described previously,30 and was allowed to spread throughout the trabeculae. To study the effect of octanol on GJs, fura 2 potassium salt was microinjected 1 or 3 hours after the superfusion of KH solution containing 100 μmol/L octanol. Fluorescence of fura 2 from trabeculae excited by 360-nm wavelength light was projected onto an image-intensified CCD camera (model C330, General Scanning Inc) through a 510- to 560-nm bandpass filter (Nikon). The images were recorded on a VCR recorder (EV-S7000 NTSC, Sony) for 60 minutes every 10 minutes starting at the end of microinjection of fura 2. To eliminate the effects of both uneven intensity of the illumination and spatial differences in the sensitivity of the camera, fluorescence of a solution containing 1 μmol/L fura 2 potassium salt excited by 360-nm wavelength light was also recorded. All settings of the image-intensified CCD camera and microscope were kept constant during the recordings.
The fluorescence data of each video frame were digitized with an eight-bit analog-digital converter and stored in a frame buffer memory of 512×480 pixels (Coreco Inc). In the optical system used in this measurement, one pixel of memory corresponded to 2.87×2.87 μm in the image plane. For analysis of fura 2 diffusion, a sampling region of interest was defined along the long axis of the muscle image. We calculated the average intensity profile of fluorescence along the long axis of the muscle from each successive line of pixels across a sampling region of 512×21±8 pixels (depending on the width of trabeculae) along the muscle. To improve the ratio of signal to noise, we calculated an average intensity profile from 25 sequential frames. The error introduced by averaging over 25 frames was small, because the rate of diffusion of fura 2 in the muscle is slow (see “Results”). In order to correct for the intensity distribution of the illuminating light, we calculated in vitro standard intensity profiles of the fluorescence from the video frames of a solution containing 1 μmol/L fura 2. We divided the averaged intensity profiles from the trabeculae by the spatially corresponding values of the in vitro standard intensity profiles; the ratio was denoted as the normalized intensity profile of fluorescence along the muscle. The normalized intensity profile along the muscle from the video frames calculated in this way was obtained every 10 minutes. The normalized intensity profile exhibited a maximum in the cell in which fura 2 had been injected and appeared to be gaussian. Deq for fura 2 was calculated by fitting the profile between the positions, where the intensity equaled 73% and 27% of the maximal intensity, as follows31 :in which F is the fluorescence intensity, Deq is the effective diffusion coefficient (in μm2/s), K is a constant related to the amount of the fura 2 deposited in the cell, t is time, and x is the distance from the center of the fluorescence profile. Fitting of the data over this range avoided the peak of the distribution and noise on the far edge of the fluorescence distribution (the r values ranged from .95 to .99 for all fits). From the fitted data, Deq·t was calculated. This procedure was repeated for the profiles obtained every 10 minutes, and Deq was calculated from the time course of Deq·t. The same procedure was used for determining Deq of fura 2 in water using the video images of a droplet of fura 2 with a diameter of 300 μm that had been placed in the nonperfused muscle bath by use of a microelectrode. The contribution of convective fluid motion due to illumination of the fluid in this case was ignored.
Dose (x)–response (R) curves for twitch force, TPC force, TPC propagation velocity, and triggering rate were fitted using the following equation with n as the Hill coefficient:The EC50 of the group of muscles was obtained from the relationship calculated from the data combined from the individual muscles and from the average of the EC50 calculated for the individual muscles. Student's t test was used to compare the control parameters with those in the presence of octanol and heptanol.
Force and sarcomere length recordings during the last of a series of stimulated twitches and the subsequent TPC obtained in one trabecula (Fig 1⇓) show that octanol and heptanol had only a small effect on the twitch but strongly reduce TPC force, triggering rate, and propagation velocity. These effects, obtained in 14 trabeculae from 11 rats, are summarized in Fig 2⇓ and Fig 3⇓. The absolute values of force of the twitch and of the TPC as well as the triggering rate and the propagation velocity before addition of the drugs were similar in the muscles that were exposed to octanol or heptanol, as is shown in the Table⇓. The effects of octanol and heptanol appeared to be nearly identical, with respect to the twitch and with respect to TPC characteristics. Concentrations of the alcohols lower than 1 μmol/L did not show any effect on either twitch force or TPC characteristics.
Between 3 and 300 μmol/L, octanol and heptanol decreased TPC force, triggering rate, and propagation velocity in a concentration-dependent manner. Both alcohols reduced twitch force in this concentration range by only a small extent (≈20%), whereas the TPCs were virtually abolished at 300 μmol/L. Relaxation time of the twitch was decreased to 90.0±6.9% and 87.1±6.6%, respectively, which is similar to the effect of a reduction of the twitch force by lowering external [Ca2+].
Higher concentrations of octanol and heptanol also eliminated the twitch but with a 20-fold difference of the EC50 for the twitch compared with the TPCs (see Table). The EC50 values of octanol or heptanol for the effects on the TPCs and twitch calculated for the individual muscles were not significantly different from the EC50 values for the grouped data. The Hill coefficient for the effect of the alcohols on the twitch (range, 1.7 to 2.0) was significantly higher than the Hill coefficient for their effect on the TPCs (range, 0.59 and 0.83) (P<.05).
These results suggest that the effects of octanol and heptanol on TPCs are independent of their effects on mechanisms that determine twitch force. Because there is no relationship between either slowing of triggering or propagation of the TPCs and the force of the twitch, we also conclude that the effect is independent of the Ca2+ content of the SR. The differential effects of octanol and heptanol on the TPCs compared with the twitch clearly differ from the way in which interventions that directly modulate the Ca2+ content of the SR affect TPCs; eg, a varied Ca2+ load of the SR by varied [Ca2+]o or by Ca2+ channel blockers causes effects on the TPC that are proportional to the effects on the force of the twitch.1 32
In contrast to the large difference in the effect of octanol and heptanol on the twitch compared with the TPC, there was a striking similarity of the effect of the alcohols on the process of triggering of the TPC compared with the process of propagation. Fig 4⇓ shows that the rate of triggering of the TPC correlated closely with the propagation velocity at all concentrations of octanol and heptanol. A change in the propagation velocity of the TPCs as a result of the effect of the drugs was always accompanied by a proportional change of the rate of triggering, suggesting that triggering and propagation were determined by processes with a similar sensitivity to octanol and heptanol.
The use of ethanol to dissolve octanol and heptanol may have influenced these observations. In order to test whether the reduction of TPC force, triggering rate, and propagation velocity resulted partially from the effects of ethanol, we investigated effects of ethanol on twitch and TPCs in three additional trabeculae. Because the final ethanol concentration was ≈0.02% at 300 μmol/L octanol and heptanol, we tested the effects of ethanol at this concentration. Ethanol appeared to decrease twitch force by 10% but exerted no discernible effects on TPC characteristics. We assume that at lower concentrations of ethanol, which were used at lower concentrations of octanol and heptanol (<300 μmol/L), the effect of ethanol was even smaller; hence, the reduction of TPC force, triggering rate, and propagation velocity was only caused by the effects of octanol and heptanol.
Finally, we have tested whether octanol indeed reduced the diffusion from cell to cell. Assuming that fura 2 diffuses from cell to cell only through GJs, we evaluated whether octanol reduces Deq for fura 2 along the length of 13 trabeculae from as many rats. Deq in water was 130 μm2/s, which was 6.2-fold greater than Deq in trabeculae, which was 21.0±3.1 μm2/s (mean±SEM, n=5) during superfusion with KH solution. Deq was significantly decreased after 1 and 3 hours of exposure to 100 μmol/L octanol by 40% to 12.65±1.5 μm2/s (n=5) and by 67% to 7.07±0.66 μm2/s (n=5), respectively. The effect of exposure to octanol followed an exponential time course (r=.85), with a calculated average time constant of the decrease in Deq of 1.5 hours (range, 1.2 to 2.1 hours [SEM]). After 3 hours of exposure of the muscle to 100 μmol/L octanol, residual twitch force was 51±5% of the control value.
The finding that octanol and heptanol decrease both force and propagation velocity as well as the triggering rate of the TPC is new. As we will discuss in the following paragraphs, the effects of octanol and heptanol on the TPCs are consistent with the assumption that both triggering and propagation of TPCs require diffusion of Ca2+ ions from cell to cell through GJs. The inhibitory effects of octanol and heptanol can be explained on the basis of the ability of these drugs to block the GJs.
Effects of Octanol on the Diffusion Rate of Fura 2
GJs have a 1.5-nm pore, which allows passage of molecules with a molecular weight up to 2500 (including fura 2, with a molecular weight of 832), even though the passage of anions is more impeded than that of cations.33 Hence, we used Deq derived from the fura 2 concentration profiles as an indicator of the effect of octanol on GJs. Our results involving Deq are consistent with those previously reported. Deq in trabeculae was 6.2-fold lower than in KH solution (130 μm2/s) and 1.5-fold lower than in isolated myocytes,34 suggesting that GJs provide a noticeable diffusion barrier for fura 2. A reasonable approximation of the diffusion constraints for longitudinal diffusion of fura 2 in these ribbon-shaped trabeculae would be to assume a composite medium of diffusion,31 formed by a chain of myocytes separated by GJs. In the simplest form of such a composite diffusion medium, Deq (21 μm2/s; see “Results”) would be determined by Dmyocyte (for which Blatter and Wier34 reported a value of 32 μm2/s) in series with DGJ. It follows31 fromthat DGJ equals 61 μm2/s in the muscles under control circumstances. After 60 (to 90) minutes of exposure to 100 μmol octanol, Deq decreased to (10 to) 12.65 μm2/s, so that DGJ must have decreased to (14.5 to) 19.2 μm2/s. The low residual (25% to) 30% value of DGJ is consistent with what one would expect after exposure of the muscle to a compound that reduces the open probability of GJs.28 The magnitude and rate of development of the effect of octanol is comparable to that previously found in studies involving isolated dog Purkinje fibers.35 The rate of onset of the effect of octanol in Purkinje fibers35 and in trabeculae is slow and contrasts with the rapid effect that has been observed in isolated myocytes,25 26 27 suggesting that the accessibility of the GJs is less in the multicellular trabecula than in isolated myocytes.
Effects of Octanol and Heptanol on Twitch Force
High concentrations of octanol and heptanol (>1 mmol/L) decreased twitch force. This is consistent with studies reporting that octanol and heptanol influence nonjunctional channels, for example, by enhancing Na+ channel inactivation28 36 37 38 or by a decrease of Ca2+ current, which has been observed in rat and guinea pig cardiac cells, as well as a decrease of inwardly rectifying background K+ current and delayed rectifier K+ current.28 37 A reduction of the Na+ current or Ca2+ current can readily explain that high concentrations of both octanol and heptanol decreased twitch force.
It is unlikely that GJ blockade by octanol and heptanol has contributed to the observed reduction of twitch force, because the stimuli to the muscles were always delivered via parallel platinum electrodes and because the stimulus pulses were twice threshold level, so that the activation of individual myocytes in the muscles was synchronous and did not require coupling via GJs. In addition, it should be noted that part of the 20% decrease of twitch force may have resulted from the effect of the ethanol used as a solvent for the octanol and heptanol.
Reversibility of the Effects of Octanol and Heptanol
It has been argued that damage to a group of myocytes would be self limiting because the local damage causes an increase of [Ca2+]i and thereby induces closure of GJs, preventing further Ca2+ diffusion to adjacent cells.39 In the present study, we performed the experiments in the first 90 minutes after the start of the protocol, ie, during a period in which the TPCs exhibited a stable behavior. The process of “healing over” of GJs over this time of experimentation seems, therefore, to have been negligible under the conditions of our experiments (in particular, at the temperature at which these experiments have been performed [20°C]). It seems not likely either that healing over has been accelerated by octanol or heptanol, because the effects of the two alcohols on the TPCs was reversible after the muscle had been exposed for 30 minutes to the alcohols.
Effects of Octanol and Heptanol on Triggering of TPCs
It has been shown that a rapid mechanical release of a contracting muscle causes Ca2+ dissociation from the myofilaments.8 We have previously proposed that this may happen in the damaged region of muscle and then trigger CICR.1 2 In order for the Ca2+, which dissociates from the myofilaments, to reach the SR of adjacent cells in the course of the triggering process, it is necessary that the trigger Ca2+ diffuses through GJs. It is also likely that propagation of TPCs is based on CICR mediated by Ca2+ diffusion from cell to cell. Therefore, both triggering and propagation require diffusion of Ca2+ through GJs and would be sensitive to GJ blockers.
The present study shows that triggering of TPCs occurs more slowly in the presence of octanol and heptanol at concentrations that have no significant effect on the amplitude of the twitch. This observation implies that the reduction of the currents that influence twitch force was minimal. In addition, there was no relationship between the extent of slowing of triggering and propagation of the TPCs and the reduction of the twitch, suggesting that the slowing of the TPCs was not related to a decrease in the Ca2+ content of the SR. Consequently, it is likely that the effects of the alcohols were caused by another mechanism of action. We are not aware of studies that have investigated the effect of octanol and heptanol on the rates of binding and dissociation of Ca2+ to and from the myofilaments, but such effects must have been small in our experiments because of the negligible effects on twitch force. It is important to note that we have not observed a decrease of the rate of relaxation of the twitch associated with octanol or heptanol. Hence, it is not likely either that octanol and heptanol delayed triggering of TPCs by slowing the rate of dissociation of Ca2+ from troponin C during the relaxation phase of the twitch. Such a change in the rate of Ca2+ dissociation from troponin C would be expected to lead to a change in the rate of relaxation of the twitch. The main mechanism by means of which octanol and heptanol can have delayed triggering of the TPC is therefore probably a result of closure of the GJs.
The present results suggest a substantial difference in sensitivity of the nonjunctional sarcolemmal ion channels compared with the GJs for octanol and heptanol, both in terms of the 20-fold difference of the EC50 for the twitch versus the EC50 for the TPC properties and in terms of the threefold higher Hill coefficient for the twitch compared with the Hill coefficient for the effects on the TPCs. The slow time course of the effect of octanol cannot have been the cause of an apparent difference in octanol sensitivity of the twitch compared with the TPCs, because the decrease of relative twitch force was substantially less (49%) after 3 hours of exposure to octanol than the decrease of the rate of triggering and propagation of the TPCs after 90 minutes (75%). This conclusion contrasts with the suggestion from one previous study28 that the sensitivity was similar; this contrast is possibly due to the large technical differences between the present study and the previous report.
Effects of Octanol and Heptanol on Propagation of TPCs
Mathematical simulation of CICR suggests that the rate-limiting step for propagation of Ca2+ waves in a model of the cell is the rate of Ca2+ release from the SR, which depends on the Ca2+ content of the SR.9 The experimentally observed propagation velocity of the TPCs can be attained if one assumes that Ca2+ release from the SR takes place within a millisecond, which is compatible with the time course Ca2+ sparks40 as observed by confocal scanning microscopy using the indicator fluo-3.40 In the study of mathematical simulation of propagation of Ca2+ waves,9 the effect of variation of the diffusion constant for Ca2+ was not investigated; hence, the importance of the conductance of GJs was not evaluated.9 It is conceivable that GJs limit the diffusion of Ca2+ and thereby reduce the propagation velocity of Ca2+ waves to a level below that imposed by physicochemical diffusion within a single cell. Takamatsu et al32 have observed that in a preparation consisting of two coupled cells, continuous propagation of Ca2+ wave from cell to cell did occur without a noticeable delay or change in the velocity at cell junction, suggesting that the diffusion limitation by GJs was minimal. On the other hand, Ca2+ waves and sarcomere shortening waves in isolated trabeculae are often limited to individual cells and do not always cross the boundaries between cells (H.E.D.J. ter Keurs, unpublished data, 1976 to 1995), suggesting that GJs can indeed limit Ca2+ diffusion in multicellular preparations such as the trabeculae studied here.
Octanol reduces the diffusion of fura 2 from cell to cell in these muscles considerably. Both this effect and the effects of octanol and heptanol on the propagation velocity of the TPCs are consistent with the proposed mechanism of action of the alcohols on the GJs. Octanol and heptanol have been shown to decrease the open probability of the GJs,28 so that the permeability for all molecular species that pass the channel (including Ca2+ ions and fura 2) is reduced. The decrease of DGJ to 25% to 30% of the control value after a 60- to 90-minute exposure to 100 μmol/L octanol may well explain the effect of octanol and heptanol on the TPCs, ie, a reduction of the rate of triggering and of propagation to 25% to 30% of control value (see Figs 2 and 3⇑⇑), because the diffusion of Ca2+ ions would be three to four times slower than in the control conditions. Even at the lower concentrations of the alcohols, the effect on the GJs was probably great enough to explain the slowing of the TPCs, although a precise correlation between TPC properties requires measurement of both DGJ and TPC properties at all concentrations of the alcohols after similar periods of exposure to the alcohols.
We have studied the effects of both alcohols on TPCs using increasing concentrations of the alcohols in order to obtain a cumulative dose-response curve; the exposure to each concentration of the alcohols was ≈15 minutes. This led to a maximal exposure time of the trabeculae to the alcohols of 60 to 90 minutes, depending on the number of doses studied (see time indication in Figs 2 and 3⇑⇑). This period was necessarily limited because of the limited time over which the TPCs behave in a stable fashion. As a result of the relatively slow decline of fura 2 diffusion observed with octanol, the effect of each concentration of the alcohols on the TPCs could not have reached steady state. Therefore, the dose-response curve of the effect as observed in these experiments is probably shifted rightward, and the EC50 for the alcohols may have been overestimated. The observation (not shown) that twitch force decreased to 49±5% after 3 hours of exposure to 100 μmol/L octanol is in agreement with this interpretation. Also, this observation confirms the differential effect of the alcohol on the TPCs compared with the twitch, because the reduction (≈80%) of force and velocity of the TPC was already greater after 20 minutes. We consider the interpretation that the effect of octanol is relatively slow more plausible than the assumption that the effect of the alcohol on fura 2 diffusion may be different from its effect on Ca2+ diffusion.
Effects of Octanol and Heptanol on TPC Force
Our results show that TPC force was decreased both by octanol and by heptanol. This can be explained on the basis of two mechanisms that are not mutually exclusive. First, a decrease of propagation velocity reduces the number of sarcomeres activated along the trabeculae at any moment in time, which means that the active sarcomeres shorten at the expense of stretch of a long inactive and compliant muscle segment; therefore, the concomitant force must decrease. TPC force always follows the changes in propagation velocity. Second, Fabiato41 has shown that CICR from the SR is a graded Ca2+-dependent process. Blocking of GJs by octanol and heptanol is expected to reduce Ca2+ diffusion to adjacent cells, and the relatively smaller number of Ca2+ ions arriving in the adjacent cell would trigger less Ca2+ release from the SR, which would lead to a smaller contraction.
The detailed mechanism of Ca2+ release from the SR that produces the propagation is not known. It has been proven neither that Ca2+ release from the SR is triggered by a transient rise of cytoplasmic Ca2+ during a TPC nor that Ca2+ release from the SR during a TPC occurs in the same way as in normal excitation-contraction coupling. The present experiments cannot resolve this question. However, an alternative mechanism explaining TPCs (ie, that Ca2+ release only seems to propagate because Ca2+ release occurs progressively later at a greater distance from the damaged end region because of a gradient of SR Ca2+ overload along the muscle) can be ruled out, because the effects of octanol and heptanol on TPCs appeared to be dissociated from the effect of the alcohols on the twitch.
It has been shown that spontaneous oscillations can propagate in Triton X-100–skinned myofibrils without SR, suggesting that propagation of the oscillation is mechanical and stretch-related.42 43 It is unlikely that this mechanism plays a role in TPCs, since stretch of the muscle failed to change the velocity of TPCs.2 Also, our present study that octanol and heptanol decrease propagation velocity cannot be explained by the hypothesis that stretch induces propagation of TPCs at the myofilament level.
Since the propagation of TPCs was wavelike, the contractions along the trabeculae are nonuniform. The local contraction of a TPC is expected to stretch adjacent cells along the trabeculae if muscle length is constant during the TPC. The ensuing membrane stretch may activate stretch-activated channels via the cytoskeleton attached to the sarcolemma.44 45 Activation of stretch-activated channels allows Ca2+ influx,46 47 which may induce Ca2+ release from the SR. Such a stretch-dependent process might contribute to the spontaneous local contraction along the trabeculae and, therefore, cause or modulate the propagation of TPCs. However, this hypothesis has been ruled out by the observation that gadolinium, a stretch-activated channel blocker, does not specifically decrease the triggering or propagation of TPCs.48 The results of the present study do not suggest the involvement of stretch-activated channels in TPCs, because to our knowledge no effect of octanol and heptanol on these channels is known.
The present study shows that octanol and heptanol reduce propagation velocity, triggering rate, and force of the triggered propagated contractions at concentrations in which octanol and heptanol do not affect twitch force. The observation that the effects of octanol and heptanol on TPCs are independent of their effects on mechanisms that determine twitch force is novel and suggests that they are caused by uncoupling of cells due to closure of the GJs. The similarity of the effect of octanol and heptanol on the rate of triggering and on the propagation velocity suggests that both triggering and propagation depend on diffusion of Ca2+ ions through the GJs. The observed reduction of the fura 2 diffusion rate by octanol supports the hypothesis that the effect of this alcohol is mediated by an increase of the resistivity of the GJs. These observations are important, because they shed light on the mechanisms involved in triggering and propagation of the TPCs, which are induced by damage and by themselves may lead to arrhythmias.
Selected Abbreviations and Acronyms
|CICR||=||Ca2+-induced Ca2+ release|
|Deq||=||effective diffusion coefficient|
|DGJ||=||gap junctional diffusion coefficient|
|Dmyocyte||=||myocyte diffusion coefficient|
|TPC||=||triggered propagated contraction|
This study was supported by grants from the Alberta Heart and Stroke Foundation. Dr ter Keurs is a Medical Scientist of the Alberta Heritage Foundation for Medical Research (AHFMR); Y.M. Zhang holds an AHFMR studentship. Dr Miura holds a postdoctoral research fellowship from Merck Frosst Canada Inc.
Reprint requests to Henk E.D.J. ter Keurs, MD, PhD, Department of Medicine, Health Science Centre, 3330 Hospital Dr NW, Calgary, Alberta, Canada T2N 4N1. E-mail firstname.lastname@example.org.
- Received February 9, 1996.
- Accepted September 11, 1996.
Mulder BJM, de Tombe PP, ter Keurs HEDJ. Spontaneous and propagated contractions in rat cardiac trabeculae. J Gen Physiol. 1989;93:943-961.
Daniels MCG, ter Keurs HEDJ. Spontaneous contractions in rat cardiac trabeculae: trigger mechanism and propagation velocity. J Gen Physiol. 1990;95:1123-1137.
Daniels MCG, Kieser T, ter Keurs HEDJ. Triggered propagated contractions in human atrial trabeculae. Cardiovasc Res. 1993;27:1831-1835.
Fedida D, Sethi S, Mulder BJM, ter Keurs HEDJ. An ultra-compliant glass microelectrode for intracellular recording. Am J Physiol. 1990;258:C164-C170.
Daniels MCG, Fedida D, Lamont C, ter Keurs HEDJ. Role of the sarcolemma in triggered propagated contractions in rat cardiac trabeculae. Circ Res. 1991;68:1408-1421.
Berlin JR, Cannell MB, Lederer WJ. Cellular origins of the transient inward current in cardiac myocytes: Role of fluctuations and waves of elevated intracellular calcium. Circ Res. 1989;65:115-126.
Kort AA, Lakatta EG. Calcium-dependent mechanical oscillations occur spontaneously in unstimulated mammalian cardiac tissues. Circ Res. 1984;54:396-404.
Housmans PR, Lee NKM, Blinks JR. Active shortening retards the decline of the intracellular calcium transient in mammalian heart muscle. Science. 1983;221:159-161.
Backx PHM, de Tombe PP, van Deen JHK, Mulder BJM, ter Keurs HEDJ. A model of propagating calcium-induced calcium release mediated by calcium diffusion. J Gen Physiol. 1989;93:963-977.
Kort AA, Capogrossi MC, Lakatta EG. Frequency, amplitude, and propagation velocity of spontaneous calcium-dependent contractile waves in intact adult rat cardiac muscle and isolated myocytes. Circ Res. 1985;57:844-855.
Golovina VA, Rozenshtraukh LV, Solov'ev BS, Undrovinas AI, Chernaya GG. Wavelike spontaneous contractions of isolated cardiomyocytes. Biophysics. 1986;31:311-318.
Rieser G, Sabbadini R, Paolini P, Fry M, Inesi G. Sarcomere motion in isolated cardiac cells. Am J Physiol. 1979;236:C70-C77.
Wier WG, Cannell MB, Berlin JR, Marban E, Lederer WJ. Cellular and subcellular heterogeneity of intracellular calcium concentration in single heart cells revealed by fura-2. Science. 1987;235:325-328.
Takamatsu T, Wier WG. Calcium waves in mammalian heart: quantification of origin, magnitude, waveform and velocity. FASEB J. 1990;4:1519-1525.
Ishide N, Urayama T, Inoue K, Komaru T, Takishima T. Propagation and collision characteristics of calcium waves in rat myocytes. Am J Physiol. 1990;259:H940-H950.
Regirer SA, Tsaturyan AK, Chernaya GG. Mathematical model of propagation of activation waves in an isolated cardiomyocyte. Biophysics. 1986;31:725-730.
Spray DC, Burt JM. Structure-activity relations of the cardiac gap junction channel. Am J Physiol. 1990;258:C195-C205.
Spray DC, White RL, Mazet F, Bennett MV. Regulation of gap junctional conductance. Am J Physiol. 1985;248:H753-H764.
Veenstra RD, DeHaan RL. Cardiac gap junction channel activity in embryonic chick ventricle cells. Am J Physiol. 1988;254:H170-H180.
Burt JM, Spray DC. Single-channel events and gating behavior of the cardiac gap junction channel. Proc Natl Acad Sci U S A. 1988;85:3431-3434.
White RL, Spray DC, Campos dCA, Wittenberg BA, Bennett MV. Some electrical and pharmacological properties of gap junctions between adult ventricular myocytes. Am J Physiol. 1985;249:C447-C455.
Takens-kwak BR, Jongsma HJ, Rook MB, Van Ginneken ACG. Mechanism of heptanol-induced uncoupling of cardiac gap junctions: a perforated patch-clamp study. Am J Physiol. 1992;262:C1531-C1538.
Daniels MCG. Mechanism of Triggered Arrhythmias in Damaged Myocardium. Utrecht, the Netherlands: University of Utrecht; 1991. Thesis.
Backx PHM, ter Keurs HEDJ. Fluorescent properties of rat cardiac trabeculae microinjected with fura-2 salt. Am J Physiol. 1993;264:H1098-H1110.
Crank J. The Mathematics of Diffusion. Bristol, England: Oxford University Press; 1979.
Takamatsu T, Minamikawa T, Kawachi H, Fujita S. Imaging of calcium wave propagation in guinea-pig ventricular cell pairs by confocal laser scanning microscopy. Cell Biol. 1991;16:341-346.
Loewenstein WR. Junctional intercellular communication: the cell-to-cell membrane channel. Physiol Rev. 1981;61:829-913.
Joyner RW, Overholt ED. Effects of octanol on canine subendocardial Purkinje-to-ventricular transmission. Am J Physiol. 1985;249:H1228-H1231.
Niggli E, Rudisuli A, Maurer P, Weingart R. Effects of general anesthetics on current flow across membranes in guinea pig myocytes. Am J Physiol. 1989;256:C273-C281.
De Mello WC. Cell coupling and healing-over in cardiac muscle. In: Sperelakis N, ed. Physiology and Pathophysiology of the Heart. Norwell, Mass: Kluwer Academic Publishers; 1989:541-549.
Cheng H, Lederer WJ, Cannell MB. Calcium sparks: elementary events underlying excitation-contraction coupling in heart muscle. Science. 1993;262:740-744.
Fabiato A. Time and calcium dependence of activation and inactivation of calcium-induced release of calcium from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. J Gen Physiol. 1985;85:247-289.
Fabiato A, Fabiato F. Myofilament-generated tension oscillations during partial calcium activation and activation dependence of the sarcomere length-tension relation of skinned cardiac cells. J Gen Physiol. 1978;72:667-699.
Linke WA, Bartoo ML, Pollack GH. Spontaneous sarcomeric oscillations at intermediate activation levels in single isolated cardiac myofibrils. Circ Res. 1993;73:724-734.
Yang X, Sachs F. Mechanically sensitive cation channels. In: Siemen D, Hescheler J, eds. Nonselective Cation Channels: Pharmacology, Physiology and Biophysics. Basel, Switzerland: Birkhauser Verlag; 1993:79-92.
Isenberg G. Nonselective cation channels as regulatory components of cells from various tissues. In: Siemen D, Hescheler J, eds. Nonselective Cation Channels: Pharmacology, Physiology and Biophysics. Basel, Switzerland: Birkhauser Verlag; 1993:247-260.