| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-adrenergic stimulation and uncoupling
1Department of Physiology, Texas Tech University Health Sciences Center, Lubbock, Texas 79430; and 2Institut Non-Linéaire de Nice, Centre National de la Recherche Scientifique, F-06560 Valbonne, France
Submitted 24 April 2003 ; accepted in final form 25 July 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
zone; cardiac arrhythmia; functional border zone; ischemia-reperfusion injury
-adrenergic stimulation, Ca2+ overload, or reperfusion-like conditions (10). A recent theoretical study (26) has also pointed to an essential role of intermediate uncoupling in propagation of triggered activity from the ischemic area. Thus the conditions that favor origination of ectopic activity have been largely understood. At the same time, the process that allows ectopic activity to progress into a large-scale arrhythmia remains elusive. Our study attempts to give some insights into this process.
To visualize how the activity from individual ectopic sources propagates, we have used a recently developed experimental model, which allows one to create a localized area of injury (termed I-zone) and to monitor the development of arrhythmias within a two-dimensional (2D) network of cardiac cells (4). To create an injury we have employed a local application of a -adrenergic agonist isoproterenol (Iso) and a gap-junctional uncoupler heptanol (Hept). The use of Hept in our experiments allowed us to mimic progressive uncoupling known to occur during ischemia. In vivo, this process has been linked to a variety of factors, including increases in intracellular Ca2+ (
) and H+ concentrations (11, 17), accumulation of amphipathic lipid metabolites (24, 38), and changes in phosphorylation levels of connexins (6). The use of Iso, on the other hand, allowed us to simulate the massive release of endogenous catecholamines known to occur during myocardial ischemia (20). Notably, the levels of interstitial norepinephrine have been shown to increase progressively in the ischemic areas, whereas dialysate from nonischemic regions showed no significant changes (2). Such a local surge of catecholamines resembles the conditions of our experimental model, where a small area of the myocyte network is perfused with Iso. Besides mimicking endogenous catecholamine release, Iso application also served as a general means to increase the spontaneous depolarization rate of a myocyte, which in vivo was linked to multiple factors, including Ca2+ overload, injury currents, and inhibition of potassium conductance (10).
In our previous report (3), we observed ectopic arrhythmias during washout of the ischemia-like solution. The ischemic solution in that study included Hept, elevated potassium, low pH, and other components of the ischemic milieu. The process, which led from individual ectopic events to the full-scale arrhythmia, originated mostly at the border of the I-zone and was transient as it occurred during just the first few seconds of reperfusion. Therefore, characterization and modeling of the ectopic process progression was hardly feasible. To overcome these limitations, we decided to substitute a short-lived increase in cell excitability associated with reperfusion with a more reliable experimental intervention. Indeed, a combined application of Iso and Hept to the I-zone has allowed us to extend, both in time and space, the fleeting condition that appears to "breed" ectopic arrhythmias. We then characterized temporal features of the ectopic process, including its individual phases and modeled such behavior theoretically. Together, our data have shown that combination of low coupling, increased excitability, and spatial heterogeneity leads to the development of ectopic waves confined to the injured network. This transient condition served as an essential step for the ectopic activity to mature before escaping into the surrounding control network.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
Experimental chamber. We used a custom-made experimental chamber that allows one to perfuse a small area of cell network with a solution of interest, while observing events under the microscope. Hydrodynamics and temporal characteristics of flow, as well as the chamber's detailed design, have been described earlier (4). In brief, the chamber consisted of a stainless steel holder for mounting a glass coverslip and a Plexiglas base that contained two inlets and one outlet (Fig. 1A). Two platinum electrodes embedded in the top of the chamber were used to stimulate a small cluster of myocytes immediately below and the excitation then spread through the rest of network with 10–17 cm/s propagation velocity. To determine the excitation threshold, we applied monophasic 1.2-ms pulses starting at 0.4 V/cm and then increased the stimuli in 0.1-V increments until each pacing pulse was followed by a Ca2+ transient (CaT) wave. The monolayer, which behaved as a syncytium, was then continuously paced at 0.2–0.5 Hz by a voltage 20% higher than the excitation threshold (average threshold values were 0.8 V/cm). Experiments were conducted at 25°C as justified previously (3, 4).
|
Recording of ectopic waves. To visualize the propagation of ectopic activity, we recorded CaT instead of membrane potentials. Voltage-sensitive dyes were not employed because these indicators exhibit a <10% change of emission intensity in response to an action potential (as compared with a 10-fold increase in fluo 4 intensity associated with CaT). Such limited dynamic range renders potentiometric indicators unsuitable for experiments that require low magnification objectives to visualize the unpredictable occurrence of ectopic foci in a 2D multicellular network because these objectives collect only a limited amount of emitted light due to their low numerical apertures. In addition, rapid photobleaching of potentiometric dyes together with the toxicity of their oxidation products (29) makes them unsuitable for the continuous acquisition (>30 min) used in our studies. Thus we employed CaT recordings, which have been used successfully to monitor impulse propagation both in cell culture (3, 8) and on the whole heart level (21).
Monitoring CaTs. Cells plated on laminin-covered coverslips were loaded with 5 µM Fluo-4AM for 1 h. Each spontaneous or paced action potential was associated with an increase in recorded as a CaT. Notably, the motion artifacts are negligible when one acquires a signal from a large field of cells, such as the regions of interests used in our study. Fluo-4 was excited at 488 nm, and the fluorescence was acquired at wavelengths of >515 nm. Experiments were conducted in two modes: x-y and x-t (linescan). The two modes complemented each other, aiding in one's understanding of the events associated with ectopic activity. The x-y mode (128 x 128 pixels, 166 ms/frame) was used to monitor the developing 2D pattern of propagating CaT waves. The linescan mode allowed the recording of CaT from individual cells with a 6-ms temporal resolution (typical linescan consisted of 1,024 lines, 6 ms/line). Figure 1B illustrates the two modes of data acquisition.
Acquisition system. Cells were imaged with the use of a Bio-Rad laser scanning confocal system (model MRC-1024 interfaced to an Olympus IX-70 inverted microscope) with low-power magnification objective (Olympus PlanApo 4x/0.16 numerical aperture) to capture the injury and control zones simultaneously (Fig. 1B). Notably, low-magnification objectives have large depths of field, and use of confocal imaging system removes interfering signals from the plastic part of the chamber, displaying only a thin optical section from the focal plane where the cells are found. Another advantage of the employed imaging system was minimization of the light exposure (as a result of sequential, point-by-point laser scanning of the sample), which allowed us to continuously acquire data (for up to 2 h) without significant cell damage and probe bleaching (3).
Experimental protocols. All solutions were equilibrated with atmospheric oxygen. Initially, cells were superfused for 10–20 min with a control Tyrode solution supplemented with 10 mM HEPES pH 7.3 (300 mosM). To simulate conditions that promote ectopic activity, cells within the I-zone were superfused with Tyrode containing Hept and Iso. A typical protocol consisted of 10 min of perfusion with Tyrode, followed by a 5-min perfusion of the I-zone cells with Tyrode, containing either 5 µM Iso, 2 mM Hept, or 5 µM Iso + 2 mM Hept. Washout of these solutions from the I-zone by the control Tyrode constituted reperfusion. Pacing was applied continuously throughout the entire duration of the experiment [pacing electrodes were located in the control Tyrode zone (C-zone) as shown in Fig. 1B].
Numerical study of cardiac network behavior. Experiment-derived conclusions regarding the role of macroscopic parameters (coupling and cell excitability) have been confirmed using a generic FitzHugh-Nagumo (FHN) model of cardiac cell. FHN describes the qualitative aspects of excitation and propagation while lumping together the various slow currents (15). Specifically, the FHN model describes the state of a cell by a variable, e, related to the membrane potential, in such a way that e = 0 corresponds to the resting state, and e = 1 corresponds to the fully depolarized state, and by the variable 
, representing the total slow current. These variables conform to the following differential equations
 | (1) |
 | (2) |
is small ( = 0.01), reflecting the fact that the dynamics of is much slower than the dynamics of e. de/dt is the time derivative. The function f(e) has an inverted N shape, and in this study we represented it by the cubic function f(e) = e (1 – e) (e – 0.1). The parameter 
in Eq. 2 is related to an ectopic activity of the cell (e.g., automaticity) (15). A small value of implies that the cell is nonoscillatory: a linearly stable quiescent state exists around e = 0. A perturbation of amplitude larger than the excitation threshold, however, triggers an action potential. The smaller the , the larger the excitation threshold, therefore the more difficult it is to drive the system away from the quiescent state. For larger values of , the quiescent stable state around e = 0 disappears and the cell becomes spontaneously oscillatory.
We consider an idealized situation, where cells are located on a square lattice, so the variables are labeled by two integers, i and j, referring to the rows and columns of the lattice. The coupling term in Eq. 1 describes the electrical conduction between neighboring cells as
 | (3) |
, thus driving the cell toward an oscillatory state. For each cell, the value of is assumed to respond as a function of time according to
 | (4) |
0(i; j) is randomly distributed. The results reported here are obtained with a Gaussian distribution of 0 = 0.11 ± 0.05. This corresponds to a distribution of periods of the oscillating cells in the range of 9.7–13.6 s. During the application of Iso-Hept solution, the function f grows from 0 to 1 in a time of order 300 s. On washout, the value of f returns to 0 with a characteristic time of 150 s. To mimic the pacing applied experimentally, we added numerically an external periodic current at the edge of the C-zone, which generated a train of periodic waves. The system was studied numerically with the use of a finite difference method. We used either a Crank-Nicholson or a second-order Runge-Kutta scheme. These schemes are second order in space and time (27). The time step was chosen by making sure that the numerical results did not change when diminishing the time step by a factor two.
Data analysis. Each of the experimental protocols was conducted at least seven times. Presented figures and graphs are typical results of corresponding scenarios. Quantitative results are expressed as means ± SD. Data and images were plotted using Microcal Origin version 6.0 and NIH Image software (Scion).
Chemicals. Collagenase type II was obtained from Worthington (Freehold, NJ). Culture medium and porcine trypsin were obtained from GIBCO-BRL (Grand Island, NY). Fluo-4-AM was purchased from Molecular Probes (Eugene, OR). Fetal bovine serum (FBS), Tyrode solution, and all other chemicals were purchased from Sigma (St. Louis, MO).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
and spontaneous contractions is negligible (5, 16). In our experiments, the application of 2 mM Hept caused the rapid disappearance of CaT in the I-zone (24.1 ± 6.3 s, n = 7). Cells remained silent for the duration of Hept perfusion (up to 30 min; data not shown). The effect of Hept on CaT was reversible because the frequency and amplitude of CaT returned to the control levels after the removal of Hept (Fig. 2). Whereas Hept reversibly suppressed CaTs within I-zone, no disturbances of the rhythm in the control area were recorded either during Hept application or its washout.
|
Figure 3 illustrates that the absence of CaT in the I-zone during Hept application also indicates a failure of action potential propagation, not just suppression of CaTs. The experiment was performed in a mechanically partitioned cell monolayer. It shows that application of Hept to the I-zone prevents passage of the paced CaT to the adjacent part of the control network. If electrical activity were to be present in the I-zone during Hept application, it would lead to the resumption the pacing CaT pattern on the other side of the control network, which is contrary to the observed data. The substitution of Hept with another gap-junctional uncoupler, palmitoleic acid (28), led to similar results. Specifically, local perfusion with 20 µM palmitoleic acid inhibited propagation of the paced activity to the I-zone, followed by the restoration of CaT pattern on uncoupler removal (data not shown).
|
Local perfusion with Iso-containing Tyrode solution. Short-term application of 5 µM Iso to the I-zone led to a small increase in CaT amplitude (12.6 ± 5.0% of the initial value). We have also observed shortening of CaT due to a 17% decrease in the half-time (t
) of CaT decay (from 0.32 ± 0.06 to 0.28 ± 0.07 s). Both effects are in accordance with previous studies, which tested the effect of Iso on isolated myocytes from neonatal rats (19). Aside from these effects on individual CaT, perfusion with Iso did not alter the general pattern of CaT propagation across the myocyte network. Specifically, with the exemption of an occasional double CaT, due to known capacity of Iso to induce early afterdepolarizations (39), no arrhythmias were detected during either perfusion or on washout of 5 µM Iso.
General sequence of events during local perfusion with Iso-Hept-containing Tyrode solution. The combined application of Iso-Hept resulted in the behavior of a network that was markedly different from the experiments when either Hept or Iso alone were present. It led to a prominent arrhythmia-like response in both the I-zone and the control area (Fig. 4). Although the process was essentially continuous, we have divided it into phases for quantification and description purposes. Specifically, based on the results acquired from both x-y and linescan experiments (Figs. 4, 5, 6, 7) events were classified into the five subsequent phases: phase I, slowing of CaT wave propagation inside the I-zone; phase II, period of quiescence; phase III, appearance of ectopics and ectopic clusters; and phase IV, evolution of small ectopic clusters into local CaT waves confined to the I-zone. Phase V was assigned to the reperfusion and was associated with arrhythmias encompassing both the I- and C-zones. The occurrence and duration of individual phases as well as representative traces acquired from two regions positioned inside the I- and C-zones are shown in Fig. 4. Visual representation of the events as selected sequential x-y frames from individual phases can be seen in Fig. 5 (an online movie file further illustrates this process. See http://ajpheart.physiology.org/cgi/content/full/00381.2003/DC1). A compressed linescan image (complied from 50 sequential 6-s linescans from a representative experiment) allows one to observe the entire process as a continuously evolving sequence of events (Fig. 6). The detailed account of the events associated with each phase is presented in Fig. 7 and is discussed below.
|
|
|
|
Phase I. Decreasing conduction inside the I-zone. Linescan images of a monolayer acquired at different times after the addition of Iso-Hept and traces from individual sites are shown in Fig. 7A. At the macroscopic level (calculated over 1 mm distance), velocity of the propagating CaT wave displayed a smooth exponential decay and reached as low as 1 mm/s. Slowing was quickly followed by a propagation failure. On the level of individual cells, the amplitude of CaT abruptly declined to 
50% of initial CaT value, which was followed by a flat trace. Together, the events during phase I were similar to the behavior of the network during the initial stages of Hept experiments (Fig. 2) and lasted 27 ± 5 s.
Phase II. Period of quiescence. Phase I was followed by a short period when no activity was observed within the I-zone (Fig. 7A). This quiescent period lasted from 5 to 120 s. As shown in Fig. 2, Hept diminishes gap-junctional conductance almost immediately [it takes 10 s to establish the flow within the I-zone (4)], whereas it takes >1 min for Iso to affect its intracellular targets (19; and our unpublished data). Therefore, during phase II, the wave of excitation caused by the external pacing stimuli fails to penetrate into the I-zone due to an increased intercellular resistance, whereas the excitability of I-zone myocytes has not reached the critical value required for the spontaneous generation of ectopic activity.
Phase III. Appearance of individual CaT and formation of small ectopic clusters. Asynchronous CaT began to appear within I-zone 50 s after Iso-Hept perfusion was initiated (Fig. 7B). The amplitude of CaT was 50% of their original control values, but CaT temporal features (e.g., duration, t to peak, and t of decay) remained the same. Approximately 100 s into the Iso-Hept perfusion, CaT from individual cells began to trigger CaT in adjacent myocytes (Fig. 7B). These events appeared as small asymmetric clusters in which activity was propagated with 0.4–1 mm/s velocity (at the level of few cells the macroscopic velocity of CaT propagation can be only estimated due to its saltatory pattern). Notably, the linescan mode may, to some extent, exaggerate the "saltatory" pattern of cell-to-cell propagation due to the nonlinear arrangement of the cells (as connecting myocytes can lie above or beneath the line). The phase III duration ranged from 5 to 30 s.
Phase IV. CaT waves confined to I-zone. Progression of ectopic clusters into CaT waves leads to a complex spatiotemporal pattern of colliding CaT waves (Fig. 7C) which is reflected as highly irregular beats on the I-zone trace (Fig. 4). The linear velocity of these waves ranged between 0.4 and 1 mm/s and did not increase significantly during continuous exposure to Iso-Hept. The complex pattern of activity in the I-zone caused no substantial rhythm disturbance to the surrounding control network (Fig. 4). Multiple CaT waves remained confined to the I-zone and failed to trigger CaT in the control area (Fig. 7C, inset). Therefore two patterns of activity existed independently: across the C-zone, fast, paced CaT waves propagating with a single front; inside the I-zone, disorganized regimes of slow CaT waves with multiple wavebreaks. Phase IV persisted as long as the I-zone was perfused with Iso-Hept solution.
Phase V. Events during washout of Iso-Hept containing Tyrode solution. Washout of Iso-Hept led to tachyarrhythmias encompassing the entire myocyte network (Fig. 7D). Approximately 10 s after the flow of Iso-Hept was terminated, the CaT waves from the control area began to invade the I-zone (Fig. 7D). At the same time, some of the CaT waves originating in the I-zone escaped into the control area. Conduction velocity rapidly rose and the frequency of escaping CaT waves increased. The inhomogeneity on the cellular level led to the escape of the ectopic activity from several areas of the border. If the remaining I-zone was still in a recovery phase, such a scenario often led to a transient reentry-like pattern around the I-zone. Altogether, the mean frequency of CaT increased (from original pacing frequency of 0.4–2.5 Hz) and tachyarrhythmia encompassing both zones was recorded. It typically lasted from 50 to 100 s and was followed by a gradual return to the pacing frequency.
Numerical studies. We hypothesized that the above-described sequence of events is not limited to a specific response of cultured neonatal cardiomyocytes to a local application of Iso-Hept and may occur when part of a cardiac cell network is subjected to conditions that lead to a decreased cell-to-cell coupling and increased excitability (allowing individual ventricular-like cells to gain automaticity to various degrees). To test this hypothesis we have chosen a generic FHN model that does not tailor ionic currents to specific values from different species but operates with lump changes in coupling and excitability (see DISCUSSION for justification). Specifically, we employed a 20 x 40 array of FHN-based cardiac cells and modeled effects of Iso and Hept by changing with time the parameter and the coupling coefficient D, respectively (Fig. 8). These changes induced oscillatory behavior within the I-zone (Fig. 8B) and triggered arrhythmogenic process, which proceeded through experimentally observed phases (Fig. 8, C and D). The events in the I-zone included propagation slowing (e.g., Fig. 8D, 30 s), period of quiescence (Fig. 8, 90 s), individual cells firing, and expansion of clusters (Fig. 8D, 174 s). It was followed by the development of slow ectopic waves confined to the I-zone (Fig. 8D, 294 and 324 s). Reperfusion led to the escape of ectopic waves into the control network (Fig. 8D, 432 s). Similar to the experimental findings, paced activity failed to enter the ectopic area during phases III and IV and the ectopic activity propagated within the I-zone as slow, phasic waves (15). Because we were considering a random distribution of 0 values, several simulations using different patterns of 0 were performed. The exact patterns of activity depended on the precise distribution of 0 in each case, but the general behavior was in good qualitative agreement with the experimental findings (compare Fig. 4 with Fig. 8B, Fig. 6 with Fig. 8D and Fig. 5 with Fig. 8C).
|
Although a disturbance of the control rhythm is clearly present in simulation traces (Fig. 8B), the tachyarrhythmia was much less pronounced in the simulation trace compared with the one obtained experimentally (Fig. 4). One must remember, however, that our modeling studies examined the effects of a small (20 x 20-cell wide) I-zone region on an equally small control area. The size of the experimental I-zone is much larger and covers thousands of cells. Thus multiple ectopic waves escape from the various regions of the border, and their interference creates a more complex tachyarrhythmic pattern in the control network.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
When both Iso and Hept were applied to the I-zone (Fig. 4, 5, 6, 7), it led to the formation of ectopic waves within the I-zone and reperfusion arrhythmias. Although we divided the observed phenomenon into five phases to assist quantification and description of underlying events, the process was essentially continuous. Notably, it appeared that several protocols tested using our model, including Iso-Hept treatment, washout of ischemia-like solution, multiple ischemic episodes (3), proceeded through similar phases or steps that led to arrhythmia development.
To simulate network behavior we had to choose among a variety of established theoretical models, starting with a generic FHN model, which effectively lumps together the various slow currents (15), to a more detailed Luo-Rudy model, which consider specific ionic currents and their kinetics (22), or even more recent ones that include local factors, such as dyadic space (12). Our reasons for employing FHN are three-fold. First, our main purpose was to understand whether an experimentally observed sequence of events could be a result of concurrent changes in cell excitability and coupling under conditions of cell heterogeneity. We thus did not intend to use modeling to identify contributions from particular ionic currents (which is an arguably important but different issue). On the contrary, the fact that lumped changes in excitability and coupling (under condition of cell heterogeneity) are sufficient to describe the progression of ectopic activity from an injury area to the surrounding control network, suggest that this phenomenon may occur in a broad range of excitable tissues and is not limited to a network of neonatal rat cells. Second, the FHN model allowed us to simulate the behavior of a 2D lattice of heterogeneous cells. The use of detailed models would make our studies computationally unnecessarily costly without answering our original question. Finally, because the exact values, relative abundance of particular channels, and Iso-induced changes to specific ionic currents (a variety of currents are known to be affected by Iso and their relative contribution to increased cells excitability remain controversial) had to be derived from disparate data on different species and would have been entirely arbitrary, the likelihood of providing physiologically relevant insights using more detailed models of cardiac cells would have been negligible at this stage.
The computational approach has confirmed that the observed scenario can be reproduced by assuming a change in two macroscopic parameters: cell-to-cell coupling and depolarization rate, the latter being random. We emphasize the importance of the random nature of the activity. Specifically, the patterns shown in Fig. 8 correspond to a random Gaussian distribution of "slopes," whereas the system of identical cells does not generate waves of activity, but instead oscillates uniformly in the I-zone (data not shown). The problem of coupled oscillatory waves has been investigated in the past, in particular in the context of pacemaker activity in the sinoatrial node (25). In the case of well-coupled cells, where the heterogeneities in the period of oscillation are small, the dynamics of the system can be understood by using standard mathematical methods (15): the system oscillates as a whole; all cells are synchronized. In a weakly coupled case considered in this study, nontrivial spatiotemporal patterns of activities corresponding to slow waves of synchronization ("phase waves," see Ref. 15) are observed. In larger systems, as it happens in the experiment it leads to an apparently chaotic regime (Fig. 4). Importantly, disorganized regimes appear only when the dispersion of periods is large enough and the coupling between cells is small enough. Increasing the coupling, and/or decreasing the dispersion of frequencies of the cells, results in a better synchronization and in more regular regimes. Decreasing the number of oscillatory cells also results in more ordered regimes. A consequence of the low coupling between the cells in the I-zone is that the phase waves are slow, as observed experimentally and numerically. The numerically observed wave velocity drops by a factor 30 when the diffusion constant is reduced by a factor of 1,000, consistent with general results (15), and in qualitative agreement with the experimental data (Fig. 7A).
Our protocol generates a network of weakly coupled, pacemaker-like cells (I-zone) surrounded by the ventricular-type cells paced by the external source (C-zone). Experimental traces from the I-zone and the C-zone reveals an independent pattern of activity inside each zone during phases III and IV. Simulations using the FHN model have also shown that when the diffusion coefficients in the I-zone and the C-zone are sufficiently different (by >2 orders of magnitude), the I-zone waves are unable to trigger activity on the ventricular side and vice versa.
For individual ectopic foci, this situation and its relationship to coupling are often referred as protected loci or exit/entrance block (30, 35). The novelty of our results derives from a combination of an entrance/exit block concept (failure of CaT waves to enter and exit the ectopic area) with two other factors: network heterogeneity and spatial (2D) arrangement of many ectopic cells. Together these conditions led to multiple ectopic waves confined to the affected area.
The large discrepancy between the coupling constants in the two zones is responsible for the lack of propagation from one zone to the other. Making the two coupling constants closer renders propagation to and from the I-zone possible. During the reperfusion phase this effect is observed and is at the origin of the extra beats observed in Fig. 8 in the C-zone in the time interval t = 400–550 s.
Our experiments and simulations dealt with a 2D network of cardiac cells. Myocardium, on the other hand, is a 3D structure, and further studies will be required to extend our conclusions to the 3D case. We suggest, however, that in vivo ectopic events can also occur in a 2D-like environment. Specifically, we hypothesize that an ectopically active layer of cells can be formed from reversibly injured ventricular cardiomyocytes sandwiched between healthy tissue and an irreversibly injured ischemic core. This layer of cells is often referred to as a functional border of the infarcted area (23). Catecholamine surge and/or calcium overload due to microreperfusion (metabolic vasodilatation of neighboring coronary vessel, see Ref. 3) allow the functional border's cells to gain automaticity, i.e., become ectopically active]. Ischemic environment impairs cell-to-cell coupling. We hypothesize that a combination of intrinsic cell heterogeneity, low coupling, and gained automaticity may convert at least a small part of this border cell layer into a semi-2D functional network that exhibits phase IV behavior. We named this hypothetical layer an ectopic surface.
If an ectopic surface does exist in vivo, it may affect a relationship between frequency of ectopic loci and its ability to cause arrhythmia in the control network. In contrast to a classic scenario, frequencies of individual ectopic foci do not have to exceed external pacing frequency or be time matched to avoid refractoriness of the surrounding tissue. Indeed, the existence of the ectopic surface implies an essentially constant source of activation for the control network due to an added spatial component (e.g., ectopic waves "whirl" continuously within the ectopic surface). During ischemia-like conditions, the coupling gradient prevents exit of ectopic waves into the surrounding tissue. Reperfusion brings recovery of gap-junctional conductance across the border. Inevitably, ectopic waves propagating alongside the border escape at the first spot where conditions of exit block (e.g., uncoupling and refractoriness) are relieved. Thus, in the presence of an ectopic surface, arrhythmogenesis is not directly related to the frequency of individual ectopic cells and their paced counterparts. The same reasoning implies that the existence of an ectopic surface will substantially widen the "vulnerability window" required for reentry formation (32).
In conclusion, our experimental and theoretical studies have suggested a sequence of events that may allow for the ectopic activity to develop into an arrhythmia. Specifically, our data indicate the following: 1) local part of a ventricular network gains pacemaker-like properties in the presence of -adrenergic stimulation; 2) because of the intrinsic cell heterogeneity and low coupling, induced by gap junction uncoupler, slow ectopic waves develop within the injured zone; 3) the ventricular (paced) and injured (ectopically active) part of the network exhibit independent patterns of activity; and 4) on recovery of cell-to-cell coupling between the control and injured area, ectopic waves exit at multiple sites of the injured area's border, creating a disorganized wave pattern and arrhythmia in the control network.
|
DISCLOSURES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
2-Adrenergic receptor actions in neonatal and adult rat ventricular myocytes. Circ Res 76: 40–52, 1995.-adrenergic stimulation in rat ventricular myocytes. J Physiol 533: 837–848, 2001.This article has been cited by other articles:
|
|
 |
|
  M. Kay, L. Swift, B. Martell, A. Arutunyan, and N. Sarvazyan Locations of ectopic beats coincide with spatial gradients of NADH in a regional model of low-flow reperfusion Am J Physiol Heart Circ Physiol, May 1, 2008; 294(5): H2400 - H2405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Agladze, M. W. Kay, V. Krinsky, and N. Sarvazyan Interaction between spiral and paced waves in cardiac tissue Am J Physiol Heart Circ Physiol, July 1, 2007; 293(1): H503 - H513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Arora, J. Ng, J. Ulphani, I. Mylonas, H. Subacius, G. Shade, D. Gordon, A. Morris, X. He, Y. Lu, et al. Unique Autonomic Profile of the Pulmonary Veins and Posterior Left Atrium J. Am. Coll. Cardiol., March 27, 2007; 49(12): 1340 - 1348. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. E Steinberg, L. Glass, A. Shrier, and G. Bub The role of heterogeneities and intercellular coupling in wave propagation in cardiac tissue Phil Trans R Soc A, May 15, 2006; 364(1842): 1299 - 1311. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
