Remodeling in Cells From Different Regions of the Reentrant Circuit During Ventricular Tachycardia
http://www.100md.com
《循环学杂志》
the Department of Pharmacology, Center for Molecular Therapeutics, Columbia University (S.B., W.D., C.C., P.A.B.), and the Department of Computer Systems, NYC College of Technology, City University of New York (C.C.), New York, NY.
Abstract
Background— Anisotropic reentrant excitation occurs in the remodeled substrate of the epicardial border zone (EBZ) of the 5-day infarcted canine heart. Reentry is stabilized because of the formation of functional lines of block. We hypothesized that regional differences of ionic currents in cells of the EBZ form these lines of block. Therefore, we first mapped reentrant circuits of sustained tachycardias, then dispersed cells (infarct zone cells, IZs) from the central common pathway of the circuit (IZc) as well as from the other side of the line of block (outer pathway, IZo) for study.
Methods and Results— We mapped reentrant circuits in the EBZ of infarcted hearts during sustained ventricular tachycardias (>30 seconds, n=17 episodes, cycle lengths=218±7.9 ms). INa density was reduced in both IZc and IZo, and the kinetic properties of IZc INa were markedly altered versus IZo. Structural remodeling of the sodium channel protein Nav1.5 occurred in IZs, with cell surface localization differing from normal cells. Both IZc and IZo have similar but reduced ICaL, whereas IZc showed changes in Ca2+ current kinetics with an acceleration of current decay. Computer simulations of the 2D EBZ showed that incorporating only differences between INa in IZc and IZo prevented stability of the reentrant circuit. Incorporating only differences between ICaL in the IZc and IZo cells also prevented stability of the circuit. However, incorporating both INa and ICaL current differences stabilized the simulated reentrant circuit, and lines of block formed between the 2 distinct regions.
Conclusions— Despite differences in INa and ICaL properties in cells of the center and outer pathways of a reentrant circuit, the resulting changes in effective refractory periods tend to stabilize reentry in this remodeled substrate.
Key Words: myocardial infarction ion channels reentry arrhythmia electrophysiology
Introduction
Structural and electrical nonuniform anisotropy underlies reentrant excitation and ventricular tachycardias (VTs) in the epicardial border zone (EBZ) of 5-day infarcted canine heart.1,2 This substrate is highly remodeled compared with normal noninfarcted epicardium.3 For example, Na+, Ca2+, and K+ currents in cells isolated from the EBZ of 5-day infarcted heart (IZs) have altered current amplitudes as well as critical changes in kinetics.4–7 Furthermore, studies have shown a different pharmacological response of the center and outer pathway tissues, which again suggests differences in currents between cells of the center and outer pathways. Despite results from experiments in normal tissue and computer simulations8,9 that indicate that a nonuniform substrate causes drift and self-termination of reentrant waves, VTs induced in the EBZ are sustained (>30 seconds).
To understand this paradox, we measured regional differences of ionic currents (Na+, Ca2+ and K+) in cells of the EBZ and studied how they contribute to the perpetuation of reentrant tachycardias. Accordingly, we first mapped reen-
trant circuits in the EBZ of 5-day infarcted hearts with sustained tachycardias. Then, we dispersed cells from regions of the EBZ in the central common pathway of the circuit (IZc), as well as cells from regions directly adjacent to the central path but on the other side of the line of functional block (outer pathway, IZo). Finally, we formulated computer models of the IZc and IZo cells to study the contribution of the remodeled ion channel activity to stability of reentry.
Clinical Perspective p 2396
Methods
Healthy mongrel male dogs (12 to 15 kg, 2 to 3 years old) were used in these studies. Under isoflurane anesthesia (30 mg/kg) and sterile conditions, myocardial infarction was produced by a 2-step total occlusion of the left coronary artery using the Harris procedure.10 Dogs were treated with lidocaine (2 mg/kg IV) if multiple ventricular beats occurred at the time of the surgical procedure. Five days after surgery, mapping experiments were performed with animals under sodium pentobarbital (30 mg/kg IV) anesthesia.
We mapped reentrant circuits in the EBZ of 5-day infarcted hearts during sustained VTs (>30 seconds (cycle lengths [CLs]=218±7.9 ms, n=17 episodes in 17 animals) induced by programmed stimulation.11 After at least 2 sequential beats of the VT had been mapped and the location of the reentrant pathway determined (Figure 1), the location of the line of block that formed during the tachycardia was carefully marked on a transparent electrode grid. The size and location of the central and outer pathways were noted, and the heart was removed. Thin sections of EBZ were then made from the center and outer paths (Figure 1A). Each was subjected to our cell dispersion protocol.7 Single ventricular myocytes were enzymatically dispersed from both the central common path (IZc) and the outer path (IZo), as well as from normal noninfarcted epicardium (NZ).
Electrophysiology
For study of sodium currents, an aliquot of cells was transferred onto a polylysine-coated glass coverslip placed at the bottom of a chamber that had been mounted on the stage of a Nikon inverted microscope (Nikon Diaphot). Myocytes were continuously superfused (2 to 3 mL/min) with Tyrode’s solution containing (in mmol/L) NaCl 137, NaHCO3 24, NaH2PO4 1.8, MgCl2 0.5, CaCl2 2.0, KCl 4.0, and dextrose 5.5 (pH 7.4). The solution was bubbled with 5% CO2/95% O2. Whole-cell INa was recorded using the patch-clamp technique and recording conditions (5 mmol/L Nao, etc) as previously described4,12,13 (see Data Supplement at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.534784/DC1).
For Ca2+ currents, pipette resistances were 1 to 2 M when filled with an internal solution of (in mmol/L) CsOH 125, aspartic acid 125, tetraethylammonium chloride 20, HEPES 10, Mg-ATP 5, EGTA 10, and phosphocreatine 3.6 (pH adjusted to 7.3 with CsOH). After rupture, 10 minutes was allowed for intracellular dialysis to begin before switching to the sodium-free recording solution containing (in mmol/L) CaCl2 5.0, MgCl2 0.5, tetraethylammonium chloride 140, HEPES 12, dextrose 10, and 4-aminopyridine 2 (pH adjusted to 7.3 with CsOH) at a temperature of 35°C as in Reference 5 (Data Supplement).
For K+ currents, patch pipette resistances equaled 1 to 2 M when filled with the internal solution (in mmol/L): KCl 140, MgCl2 1, EGTA 10, MgATP 5, creatine phosphate 5, GTP 0.2, and HEPES 10, pH adjusted to 7.2 with KOH. Cells were superfused with a Na+-free solution containing (in mmol/L) N-methyl-D-glucamine (NMDG)-Cl 144, KCl 5.4, MgCl2 1, CaCl2 2.5, CdCl2 0.5, and HEPES 10, pH 7.4, 30°C to 31°C for study (Data Supplement).
Computer Simulations
Data from experiments were used to formulate ionic models of the IZc and IZo action potential (AP) (Data Supplement).
Effective refractory periods (ERPs) and action potential durations (APDs) were calculated in 1D simulations as described by Cabo and Boyden.14 The dynamics of reentry were studied in 2D arrays. A 2D sheet (5x2 cm2) was divided into 2 vertical regions of equal area to represent the center and outer pathways. To initiate a clockwise wave, we electrically isolated the central and outer regions and initiated a propagating planar wave by external stimulation along the top boundary of the outer region. Once the wave was created, the electrical connection between the central and outer regions was established, and a spiral rotating reentering wave was initiated. The spatial resolution was 100 μm in the direction of the fiber orientation and 20 μm in the direction transverse to the fibers. The intracellular resistivity in the longitudinal direction was 0.5 k · cm and in the transverse direction was 10 k · cm to simulate the reduced side-to-side coupling between IZ cell pairs as reported by Yao et al.15
We completed 4 sets of simulations. In the first, to simulate reentry in a substrate with identical cell APs, we used our generic model14 of an IZ at nodes in both the center and outer regions. In the second, all ionic currents except INa as well as changes in ionic concentrations and intracellular calcium handling were formulated as reported in the model of an IZ.14 In the third set, all ionic currents except ICaL were formulated as reported for IZ.14 In the fourth set, we combined regional changes in INa and ICaL and K+ currents to formulate complete cell models of IZc and IZo.
Immunocytochemistry
Some cells dispersed for the electrical studies described above were plated on laminin-coated glass chamber slides and then fixed with 4% paraformaldehyde for 30 minutes, rinsed with PBS (Sigma), blocked in 2% avidin-PBS and rinsed in PBS, then blocked in 2% biotin-PBS and rinsed in PBS. Cells were then stained for Nav1.5 (Data Supplement).
Statistics
All values are represented as mean±SEM. n is the number of cells, and N is the number of animals. A value of P<0.05 was considered significant. For multiple comparisons (eg, see Table 1), ANOVA was used to determine whether sample mean values between groups were significantly different. If so, a modified t test with Bonferroni correction was used (Sigmastat, Jandel Scientific). Fisher’s exact test was used to determine the change in number of cells with specific location of Na+ channel antibody (see Figure 3B below).
Results
Sodium Currents
Compared with NZs, average peak INa densities of IZc and IZo were significantly smaller, but IZo peak INa did not differ from that of IZc (Vh=–100 mV) (Figure 2, A and B). Average current density-voltage relations of IZc and IZo also differed from that of NZs (Figure 2C). Current tracings of INa in a NZ, IZc, and IZo are shown in Figure 2B. Note that in the IZo, the time constants of current decay and times to peak of INa were greater than those of IZc (P<0.05) (Table 1). I/Imax curves of IZc showed a significant negative shift (&8 mV) compared with that of NZs (Table 2). IZo V0.5 values did not differ from the NZ value. The average activation curve of IZc also showed a significant negative shift (&5 mV) compared with that of NZs (P<0.05). This differed significantly from IZo (Table 2). Thus, a negative shift in both activation and I/Imax curves was observed in IZc. INa recovery from inactivation in IZc and IZo was slowed compared with that of NZs; however, values in IZc and IZo did not differ (Figure I, Data Supplement). Finally, clamp protocols designed to test closed-state inactivation were completed as before16 and showed that the time course of closed-state inactivation was markedly accelerated in IZc (Table 3). In sum, although INa density was reduced in both IZc and IZo, the kinetic properties of IZc INa were markedly altered.
Immunocytochemistry: Nav1.5 in Canine Epicardial Cells
Previous reports in mouse ventricular cells from normal hearts have suggested that the cardiac Na+ channel isoform, Nav1.5, localizes primarily to gap junction regions.17 Because other proteins also localize to this area in normal canine epicardial cells,18,19 we hypothesized that like Cx43 and Kv1.5 proteins, the cardiac Na+ channel protein may redistribute around the cell membrane (lateralize) in IZs. Thus, we completed a series of immunostaining experiments in IZc and IZo and compared their Nav1.5 staining pattern with that of NZs. We found that Nav1.5 staining in NZs was uniform, with staining all along the sarcolemma (SL) in all optical sections as well as in the gap junction region. There was no staining of T-tubular membranes. SL staining of an NZ showed a specific pattern, with staining being separated every 1.9 μm by a small region of no staining (because of T-tubular invaginations) (Figure 3A, inset). Most NZs studied showed this staining pattern (Figure 3B), whereas 20% showed a gap staining pattern accompanied by nonuniform staining of the SL. In contrast, most cells from the EBZ (62% and 68% IZc and IZo, respectively) showed Nav1.5 gap staining with variable, nonuniform SL staining (Figure 3, A and B) (P<0.05 each IZ group versus NZ). However, there was no difference in the percentage of cells with this abnormal staining pattern in IZc versus IZo.
Thus, although there were electrophysiological differences between IZc and IZo, both cell types showed abnormal cell surface staining for the Nav1.5, -subunit of the cardiac Na+ channel. This is consistent with the observed reduction in Na+ current density.
Ca2+ Currents
Peak L-type Ca2+ currents in IZc and IZo were reduced, similar to each other, but differed from NZs (Figure 4). Reduced Ca2+ currents in IZo and IZc occurred at most voltages (Figure 4B). However, the decays of the small peak IZc currents were faster than those in IZo and NZs (Figure 4C). This is interesting, considering the well-known current dependence of Ca2+ channel inactivation. Activation curves were similar in all 3 cell types. However, there were differences in steady-state availability curves between IZc and IZo, with there being a significant negative shift in IZc versus NZs (Table 4). Although there is a significant slowing in recovery in both IZc and IZo versus NZs, there is no difference between the 2 IZ cell groups (Table 5).
In a subset of cells from each group, we tested whether we could augment peak Ca2+ currents with -adrenergic stimulation as described by Aggarwal and Boyden.20 Figure II (Data Supplement) shows that compared with NZs, in which isoproterenol 1 μmol/L increases peak current 2.72±0.15-fold (n=10), it had a similar effect in IZc (2.13±0.2-fold, n=10) and a diminished effect in IZo (1.79±0.13-fold) (n=14) (P<0.05 versus NZ).
K+ Currents
As reported previously,7 most IZ cells have no transient outward current that can be elicited with depolarizing clamp steps. In both IZc and IZo, this is also true. Only 5 of 11 IZc and 2 of 11 IZo cells had a measurable Ito in this study (Figure III, Data Supplement). However, the densities of these transient currents in both cell groups did not differ from each other and were small compared with NZs (&20 pA/pF at +40 mV).21 Furthermore, there were no differences in Isus or Cs-sensitive currents (IK1) in cells of the 2 IZ groups.
In sum, IZc and IZo have reduced ICaL. However, there was no difference in the peak ICaL in IZc and IZo. IZc cells showed changes in Ca2+ current kinetics: an acceleration of current decay, a shift in availability, and a slowing in recovery from inactivation. The small ICaL of IZc responded normally to -adrenergic stimulation, whereas IZo ICaL did not. There are similar changes in K currents in both IZc and IZo.
Computer Simulations
Reentry in a Uniform Model of the EBZ
To simulate reentry in a substrate with identically remodeled cells (same APs), we used our generic IZ cell model14 for both the center and outer regions. Reentry was stable for the duration of the simulation (5 seconds), with a CL=210 ms (Movie I, Data Supplement). Although it is not unexpected to obtain stable reentry in such a medium, experimental results reported above show that there are significant differences in INa and ICaL between IZc and IZo. Therefore, to understand the stability of reentrant circuits in the EBZ, we need to understand the conditions under which EBZ circuits are stable in this remodeled substrate.
Effect of INa Remodeling on Reentry
Therefore, we adapted our generic IZ cell AP model14 to generate a cell model of an IZc and an IZo by incorporating the INa data (for changes in rate constants/equations used, see Data Supplement). Figure 5 shows the calculated INa density-voltage curves, activation and inactivation curves, and simulated APs of the IZc (solid line) and IZo (dotted line). Except for a small reduction in AP plateau, there were no differences in APD90 (Figure 5D). Notably, there were differences in the maximum depolarization rate and refractory period as measured in a 1D cable using a S1-S2 protocol.14 As a result, conduction velocity was decreased by &10% in the IZc cable. Both cell types exhibited postrepolarization refractoriness, with IZc exhibiting a longer refractory period (215 ms) compared with that of IZo (198 ms).
To understand whether differences only between INa in the IZc and IZo cells promote or prevent stability of the reentrant circuit, we performed the computer simulations illustrated in Figure 6A. Note that the reentrant wave was not stationary terminating after 6 rotations (&1400 ms) (Movie II, Data Supplement). In sum, although the tip trajectory and the lines of functional block during the 6 beats stayed within 5 mm of the boundary of the central and outer regions (Figure 6A, right), differences in refractoriness between cells of the central and outer regions produced by the changes in INa in these cells caused the spiral to drift into the IZc region. Eventually, it is extinguished by colliding with the top boundary of the preparation and refractory tissue of the IZo cells.
Effect of ICaL Remodeling on Reentry
To understand how the kinetic differences in ICaL seen in IZc and IZo would affect the AP, we incorporated the averaged data into our cell model. While these simulations showed differences in APs (the ERP was 188 ms in the IZc and 224 ms in the IZo) between IZc and IZo, only a nonsustained tachycardia could be induced in the 2D array (Movie III, Data Supplement). The large difference in ERP (36 ms) caused a drift in the reentrant circuit along the boundary.8,9 Thus, differences only in ICaL in the different regions of the reentrant circuit did not promote a stable line of functional block.
Effects of Remodeling of Both INa and ICaL on Reentry
We next produced cell models of IZc and IZo that incorporated both INa and ICaL data. Results from our simulations are shown in Figure 5, E and F. The combination of the regional differences of both inward currents produced an IZo AP (and ERP) that was longer than IZc AP, but the ERP difference (13 ms) was smaller than when INa (17 ms) or ICaL (36 ms) differences were considered alone. This is because remodeling in INa and ICaL affects the ERP in the central and outer pathway in opposite directions: remodeling of INa makes the ERP longer in IZc than in the IZo; remodeling of ICaL makes the ERP longer in IZo than in IZc. Overall, when the remodeling of both currents is taken into account, differences in ERP decrease. Notably, although ERP was longer in IZo, velocity was increased over IZc (Figure 5F).
To understand whether together, INa and ICaL differences in IZc and IZo promote or prevent stability of the reentrant circuit, we performed the simulations illustrated in Figure 6, C and D. The circulating wave was stationary, forming a line of functional block between the 2 regions (Figure 6C) for 9 rotations, after which it terminated (Movie IV, Data Supplement). Figure 6C shows isochronal maps of the last 5 beats of the reentrant wave. During beats 5, 6, and 7, the reentrant circuit is stable at the boundary between IZc and IZo (Figure 6C). During beats 5 and 6, sites g–a (Figure 6C) activate transversely without delay. However, the wave of beat 7 cannot propagate from f to e because of postrepolarization refractoriness of site e (Figure 6D, asterisk, small hump in e). Site e is eventually activated after the wave turns around a longer line of functional block (Figure 6C). The premature activation of IZc cells during beat 7 causes a decrease in APD. Figure 7 illustrates the dynamics of INa (Figure 7B) and ICaL (Figure 7C) during reentry at 2 sites, one in the central pathway (solid line) and another in the outer pathway (dashed line) and shows that the decrease in APD in the central pathway is the result of a reduced ICaL (Figure 7C) that results from the acceleration in current decay and its delay in recovery (Table 2). Because IZc APD is shortened, the reentering wave in beat 8 propagates readily across the boundary between the central and outer pathway, causing a further reduction in IZc APD (note the decrease in APD during beats 6 to 7–8 at sites b, c, and d in Figure 6B), thus allowing the reentering wave to rotate inside the center region.
Discussion
Location of the central isthmus of a reentrant circuit or the area of slowed conduction of VTs has been used as an approach for ablation of these tachycardias.22–24 Although this has been useful, sometimes the lines of block and therefore the isthmus are difficult to locate, and the reentry is nonsustained or not inducible. Thus, we have approached this problem by determining the ionic determinants of the lines of functional block that form during a reentrant tachycardia. For these studies, we used the well-studied VTs that occur in the canine epicardium 5 days after coronary occlusion.25 In our studies, we mapped reentrant circuits with CLs similar to those previously described in this model.26,27 Our maps showed that lines of block formed between the center and outer regions. Using tissue slices obtained at each side of the line of block, we describe here the differences in ionic current function in the cells from these 2 regions.
Similar to our previous findings16 using somewhat larger tissue slices from the EBZ (and for which tachycardias had not been mapped), we found that IZc and IZo cells have a reduced INa density. However, we found that there are significant differences between the kinetic parameters of INa in IZc versus IZo. IZc INa tended to enter inactivation from a closed state more rapidly, displaying shifts in both availability and activation curves. These differences are not a result of differences in experimental techniques, because we sampled both IZc and IZo, controlling for time after membrane rupture (see figure legends).
The mechanism of these differences is unknown at this time. With the immunocytochemistry experiments presented here, we show that there is a loss of the protein of the -subunit Nav1.5 in cells from the EBZ (Figure 3). Thus, there is marked structural remodeling of this protein in IZs. However, quite different from the structural remodeling of other proteins in cells of the EBZ,18,19 we find that there is a loss of Nav1.5 protein at the cell surface in both IZc and IZo, consistent with reduced peak INa. This staining differs from the staining pattern we observed in NZs, which showed significant SL, non-T-tubular staining of Nav1.5 protein with some gap junctional staining and from the Nav1.5 pattern seen in mouse ventricle by our laboratory (data not shown) and by others.17 Our findings in normal canine ventricle cells are consistent with the findings of several patch-clamp reports in which investigators have studied single Na channels in cardiac cells (see, eg, Berman et al28). In none of these reports was it stated that single sodium channel records could be obtained only if the investigator clamped regions of the gap junction. In both IZc and IZo, we found that surface staining for Nav1.5 was nonuniform, non-T-tubular, and gap junctional. In fact, in some IZs, gap junctional staining was the only staining observed. The change in the location of Nav1.5 proteins may have little effect on the cell electrophysiology but may affect propagation.29
Ionic Current Heterogeneity and Stability of Reentrant Circuits in the EBZ: Pharmacological Implications
One of the main results of our study is that despite differences in INa and ICaL channel properties in the IZc and IZo, the resulting change in refractory period tends to stabilize reentry in the EBZ. An important consequence of this is that the stability of the reentrant circuit is not a result of the remodeling of a single current (ie, INa or ICaL) but rather is determined by how any combination of remodeled currents affects the refractory period. This insight has important pharmacological implications. We suggest that one therapeutic strategy for converting a sustained VT to a nonsustained VT would be to make the refractory periods of the central and outer pathway as different as possible. Such a difference in ERPs would cause the reentrant wave to drift and eventually stop. For example, if we chose to use an ICaL agonist to decrease the differences in ICaL between IZc and IZo, we might predict that the circuit would become unstable and terminate, because regional differences in INa alone show unstable circuits (Figure 8). In fact, this has been reported for this model.27 Conversely, if we chose to normalize INa differences between IZc and IZo (eg, by using a drug that would relieve closed-state inactivation of IZc cells), we might expect the circuit to be unstable and tachycardias to terminate. This idea is not in conflict with the idea that pharmacological manipulation of anisotropy could lead to destabilization of reentry in this model (but this was not tested here). What we show here is that manipulation of anisotropy is not the only way to destabilize reentry. From our computer simulations, we suggest that the possibility of destabilizing anisotropic reentry can be done by creating gradients of refractory period in the highly remodeled center/outer regions. All that this requires is sufficient quantitative knowledge of the ion channels that are critical (as opposed to bystanders) in the remodeling process.
Time Course of Remodeling and Inducibility of VT
Our results here may also explain why in this experimental model, in which there is a time course in the inducibility of the VT, a sustained VT is induced in some animals but not in others.2 The remodeling of ion channels is a dynamic process, as we know from both ion channel studies on cells from acute and more healed infarcts3,21 and from this study, in which the dynamics of the ion channel changes in different regions of a circuit affect the stability of the circuit. Each ionic current contributing to the AP waveform has its own time course of remodeling and reversal. Therefore, if the time course of remodeling in cells in each region for each current were known, we should be able to predict the occurrence of sustained VT in the substrate in question. Sustained VTs would occur only when the sum of all ionic current remodeling results in center/outer regions with similar ERPs. In fact, if the time course of remodeling of one inward current were altered, we predict that we would never have a substrate that would sustain a reentrant wave.
Ion Channel Remodeling, Gap Junction Remodeling, and Stability of Reentrant Circuits
In this model of reentry, called anisotropic reentry, lines of block form in regions of slow transverse propagation associated with the nonuniform anisotropic nature of this substrate. Peters et al30 suggested a relationship between the location of the stable lines of block and the microanatomy of the EBZ as well as the distribution of abnormal connexin43 staining in EBZ cells. A limitation of our in silico representation of the EBZ is that it is one cell deep and has no structural components (macro or micro). It is clear from several of our previous studies, as well as those of others, that the electrical remodeling occurring in the 5-day EBZ is not limited to SL ionic channels. Gap junction distribution in the central pathway of reentrant circuits and gap junction conductance between side-to-side coupled cell pairs (direction transverse to the fiber orientation)31 are also remodeled. Although the functional effects of gap junction remodeling on anisotropy are still unknown (and currently under investigation), it is possible that changes in cell coupling and anisotropy contribute to further stabilization of reentry in the EBZ.32 This is supported by the fact that even though the incorporation of the remodeling of INa and ICaL in the models of IZo and IZc tends to stabilize reentry, the simulated reentry is not as sustained as it is in the experimental model. Again, we do not believe that the findings we report here invalidate the concept that anisotropy is important for stabilizing reentry. Rather, we contend that any mechanism of reentry in the EBZ should take into account all new findings, such as those presented here, where cells of the center and outer pathways show differences in ionic current function.
Limitations
We have shown that when we incorporate in a computer model ionic currents measured in IZc and IZo, reentrant circuits tend to stabilize. However, in the in situ infarcted hearts from which the IZc and IZo cells were harvested, reentrant tachycardias were stable for more that 30 seconds. Thus, it is possible that the remodeling of SL currents alone is insufficient for stabilization of reentrant circuits for longer periods. Other factors like gap junction remodeling, anchoring to anatomic obstacles, or the presence of areas of discontinuities may contribute to the stability of reentrant circuits.
Acknowledgments
This study was supported by National Institutes of Health grants HL-66140 and HL-30557.
References
Dillon S, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988; 63: 182–206.
Wit AL, Janse MJ. The Ventricular Arrhythmias of Ischemia and Infarction: Electrophysiological Mechanisms. Mount Kisco, NY: Futura Publishing Co, Inc; 1993.
Pinto JMB, Boyden PA. Electrophysiologic remodeling in ischemia and infarction. Cardiovasc Res. 1999; 42: 284–297.
Pu J, Boyden PA. Alterations of Na+ currents in myocytes from epicardial border zone of the infarcted heart: a possible ionic mechanism for reduced excitability and postrepolarization refractoriness. Circ Res. 1997; 81: 110–119.
Aggarwal R, Boyden PA. Diminished calcium and barium currents in myocytes surviving in the epicardial border zone of the 5 day infarcted canine heart. Circ Res. 1995; 77: 1180–1191.
Jiang M, Cabo C, Yao J-A, Boyden PA, Tseng G-N. Delayed rectifier K currents have reduced amplitudes and altered kinetics in myocytes from infarcted canine ventricle. Cardiovasc Res. 2000; 48: 34–43.
Lue W-M, Boyden PA. Abnormal electrical properties of myocytes from chronically infarcted canine heart: alterations in Vmax and the transient outward current. Circulation. 1992; 85: 1175–1188.
Fast VG, Pertsov AM. Shift and termination of functional reentry in isolated ventricular preparations with quinidine-induced in homogeneity on refractory period. J Cardiovasc Electrophysiol. 1992; 3: 255–365.
ten Tusscher KH, Panfilov A. Reentry in heterogeneous cardiac tissue described by the Luo-Rudy ventricular action potential model. Am J Physiol. 2003; 284: H542–H548.
Harris AS. Delayed development of ventricular ectopic rhythms following experimental coronary occlusion. Circulation. 1950; 1: 1318–1328.
Coromilas J, Costeas C, Deruyter B, Dillon S, Peters NS, Wit AL. Effects of pinacidil on electrophysiologic properties of epicardial border zone of healing canine infarcts: possible effects of KATP channel activation. Circulation. 2002; 105: 2309–2317.
Pu J, Balser J, Boyden PA. Lidocaine action on sodium currents of ventricular myocytes from the epicardial border zone of the infarcted heart. Circ Res. 1998; 83: 431–440.
Baba S, Dun W, Boyden PA. Can PKA activators rescue Na channel function in epicardial border zone cells that survive in the infarcted canine heart Cardiovasc Res. 2004; 64: 260–267.
Cabo C, Boyden PA. Electrical remodeling of the epicardial border zone in the canine infarcted heart: a computational analysis. Am J Physiol. 2003; 284: H372–H384.
Yao J-A, Hussain W, Patel P, Peters NS, Boyden PA, Wit AL. Remodeling of gap junctional channel function in epicardial border zone of healing canine infarcts. Circ Res. 2003; 92: 437–443.
Pu J, Boyden PA. Alterations of Na+ currents in myocytes from epicardial border zone of the infarcted heart: a possible ionic mechanism for reduced excitability and postrepolarization refractoriness. Circ Res. 1997; 81: 110–119.
Maier SKG, Westenbroek RE, Schenkman KA, Feigl EO, Scheuer T, Catterall WA. An unexpected effect for brain type sodium channels in coupling of cell surface depolarization to contraction in the heart. Proc Natl Acad Sci U S A. 2002; 99: 4073–4078.
Mays DJ, Boyden PA, Tamkun MM. Redistribution of the Kv1.5 K+ channel protein on the surface of myocytes from the epicardial border zone of the infarcted canine heart. Cardiovasc Pathobiol. 1997; 2: 79–87.
Peters NS, Wit AL. Myocardial architecture and ventricular arrhythmogenesis. Circulation. 1998; 97: 1746–1754.
Aggarwal R, Boyden PA. Altered pharmacologic responsiveness of reduced L-type calcium currents in myocytes surviving in the infarcted heart. J Cardiovasc Electrophysiol. 1996; 7: 20–35.
Dun W, Baba S, Yagi T, Boyden PA. Dynamic remodeling of K+ and Ca2+ currents in cells that survived in the epicardial border zone of the canine infarcted heart. Am J Physiol. 2004; 287: H1046–H1054.
Stevenson WG, Friedman PL, Sager PT, Saxon LA, Kocovic DZ, Harada T, Wiener I, Khan H. Exploring postinfarction reentrant tachycardia with entrainment mapping. J Am Coll Cardiol. 1997; 29: 1180–1186.
Harada T, Stevenson WG, Kocovic DZ, Friedman PL. Catheter ablation of ventricular tachycardia after myocardial infarction: relation of endocardial sinus rhythm late potentials to reentry circuit. J Am Coll Cardiol. 1997; 30: 1015–1023.
de Bakker JM, van Capelle FJL, Janse MJ, Wilde AA, Coronel R, Becker AE, Dingemans KP, van Hemel NM, Hauer RN. Reentry as a cause of ventricular tachycardia in patients with chronic ischemic heart disease: electrophysiologic and anatomic correlation. Circulation. 1988; 77: 589–606.
Janse MJ, Wit AL. Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction. Physiol Rev. 1989; 69: 1049–1169.
Dillon S, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988; 63: 182–206.
Cabo C, Schmitt H, Wit AL. New mechanism of antiarrhythmic drug action: increasing inward L type calcium current prevents reentrant tachycardia in the infarcted canine heart. Circulation. 2000; 102: 2417–2425.
Berman MF, Camardo JS, Robinson RB, Siegelbaum SA. Single sodium channels from canine ventricular myocytes: voltage dependence and relative rates of activation and inactivation. J Physiol. 1989; 415: 503–531.
Kucera JP, Rohr S, Rudy Y. Localization of sodium channels in intercalated disks modulates cardiac conduction. Circ Res. 2002; 91: 1176–1182.
Peters NS, Coromilas J, Severs NJ, Wit AL. Disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation. 1997; 95: 988–996.
Yao J-A, Chen S, Hussain W, Peters N, Boyden PA, Wit AL. Gap junction coupling of myocytes in central common pathway of reentrant circuits in infarct epicardial border zone: comparison with outer pathway type myocytes. Heart Rhythm. 2004; 1: S4.
Cabo C, Boyden PA. Gap junction remodeling stabilizes reentry causing ventricular tachycardia in infarcted canine hearts: a computational study. Heart Rhythm. 2004; 1: S84.(Shigeo Baba, MD; Wen Dun,)
Abstract
Background— Anisotropic reentrant excitation occurs in the remodeled substrate of the epicardial border zone (EBZ) of the 5-day infarcted canine heart. Reentry is stabilized because of the formation of functional lines of block. We hypothesized that regional differences of ionic currents in cells of the EBZ form these lines of block. Therefore, we first mapped reentrant circuits of sustained tachycardias, then dispersed cells (infarct zone cells, IZs) from the central common pathway of the circuit (IZc) as well as from the other side of the line of block (outer pathway, IZo) for study.
Methods and Results— We mapped reentrant circuits in the EBZ of infarcted hearts during sustained ventricular tachycardias (>30 seconds, n=17 episodes, cycle lengths=218±7.9 ms). INa density was reduced in both IZc and IZo, and the kinetic properties of IZc INa were markedly altered versus IZo. Structural remodeling of the sodium channel protein Nav1.5 occurred in IZs, with cell surface localization differing from normal cells. Both IZc and IZo have similar but reduced ICaL, whereas IZc showed changes in Ca2+ current kinetics with an acceleration of current decay. Computer simulations of the 2D EBZ showed that incorporating only differences between INa in IZc and IZo prevented stability of the reentrant circuit. Incorporating only differences between ICaL in the IZc and IZo cells also prevented stability of the circuit. However, incorporating both INa and ICaL current differences stabilized the simulated reentrant circuit, and lines of block formed between the 2 distinct regions.
Conclusions— Despite differences in INa and ICaL properties in cells of the center and outer pathways of a reentrant circuit, the resulting changes in effective refractory periods tend to stabilize reentry in this remodeled substrate.
Key Words: myocardial infarction ion channels reentry arrhythmia electrophysiology
Introduction
Structural and electrical nonuniform anisotropy underlies reentrant excitation and ventricular tachycardias (VTs) in the epicardial border zone (EBZ) of 5-day infarcted canine heart.1,2 This substrate is highly remodeled compared with normal noninfarcted epicardium.3 For example, Na+, Ca2+, and K+ currents in cells isolated from the EBZ of 5-day infarcted heart (IZs) have altered current amplitudes as well as critical changes in kinetics.4–7 Furthermore, studies have shown a different pharmacological response of the center and outer pathway tissues, which again suggests differences in currents between cells of the center and outer pathways. Despite results from experiments in normal tissue and computer simulations8,9 that indicate that a nonuniform substrate causes drift and self-termination of reentrant waves, VTs induced in the EBZ are sustained (>30 seconds).
To understand this paradox, we measured regional differences of ionic currents (Na+, Ca2+ and K+) in cells of the EBZ and studied how they contribute to the perpetuation of reentrant tachycardias. Accordingly, we first mapped reen-
trant circuits in the EBZ of 5-day infarcted hearts with sustained tachycardias. Then, we dispersed cells from regions of the EBZ in the central common pathway of the circuit (IZc), as well as cells from regions directly adjacent to the central path but on the other side of the line of functional block (outer pathway, IZo). Finally, we formulated computer models of the IZc and IZo cells to study the contribution of the remodeled ion channel activity to stability of reentry.
Clinical Perspective p 2396
Methods
Healthy mongrel male dogs (12 to 15 kg, 2 to 3 years old) were used in these studies. Under isoflurane anesthesia (30 mg/kg) and sterile conditions, myocardial infarction was produced by a 2-step total occlusion of the left coronary artery using the Harris procedure.10 Dogs were treated with lidocaine (2 mg/kg IV) if multiple ventricular beats occurred at the time of the surgical procedure. Five days after surgery, mapping experiments were performed with animals under sodium pentobarbital (30 mg/kg IV) anesthesia.
We mapped reentrant circuits in the EBZ of 5-day infarcted hearts during sustained VTs (>30 seconds (cycle lengths [CLs]=218±7.9 ms, n=17 episodes in 17 animals) induced by programmed stimulation.11 After at least 2 sequential beats of the VT had been mapped and the location of the reentrant pathway determined (Figure 1), the location of the line of block that formed during the tachycardia was carefully marked on a transparent electrode grid. The size and location of the central and outer pathways were noted, and the heart was removed. Thin sections of EBZ were then made from the center and outer paths (Figure 1A). Each was subjected to our cell dispersion protocol.7 Single ventricular myocytes were enzymatically dispersed from both the central common path (IZc) and the outer path (IZo), as well as from normal noninfarcted epicardium (NZ).
Electrophysiology
For study of sodium currents, an aliquot of cells was transferred onto a polylysine-coated glass coverslip placed at the bottom of a chamber that had been mounted on the stage of a Nikon inverted microscope (Nikon Diaphot). Myocytes were continuously superfused (2 to 3 mL/min) with Tyrode’s solution containing (in mmol/L) NaCl 137, NaHCO3 24, NaH2PO4 1.8, MgCl2 0.5, CaCl2 2.0, KCl 4.0, and dextrose 5.5 (pH 7.4). The solution was bubbled with 5% CO2/95% O2. Whole-cell INa was recorded using the patch-clamp technique and recording conditions (5 mmol/L Nao, etc) as previously described4,12,13 (see Data Supplement at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.534784/DC1).
For Ca2+ currents, pipette resistances were 1 to 2 M when filled with an internal solution of (in mmol/L) CsOH 125, aspartic acid 125, tetraethylammonium chloride 20, HEPES 10, Mg-ATP 5, EGTA 10, and phosphocreatine 3.6 (pH adjusted to 7.3 with CsOH). After rupture, 10 minutes was allowed for intracellular dialysis to begin before switching to the sodium-free recording solution containing (in mmol/L) CaCl2 5.0, MgCl2 0.5, tetraethylammonium chloride 140, HEPES 12, dextrose 10, and 4-aminopyridine 2 (pH adjusted to 7.3 with CsOH) at a temperature of 35°C as in Reference 5 (Data Supplement).
For K+ currents, patch pipette resistances equaled 1 to 2 M when filled with the internal solution (in mmol/L): KCl 140, MgCl2 1, EGTA 10, MgATP 5, creatine phosphate 5, GTP 0.2, and HEPES 10, pH adjusted to 7.2 with KOH. Cells were superfused with a Na+-free solution containing (in mmol/L) N-methyl-D-glucamine (NMDG)-Cl 144, KCl 5.4, MgCl2 1, CaCl2 2.5, CdCl2 0.5, and HEPES 10, pH 7.4, 30°C to 31°C for study (Data Supplement).
Computer Simulations
Data from experiments were used to formulate ionic models of the IZc and IZo action potential (AP) (Data Supplement).
Effective refractory periods (ERPs) and action potential durations (APDs) were calculated in 1D simulations as described by Cabo and Boyden.14 The dynamics of reentry were studied in 2D arrays. A 2D sheet (5x2 cm2) was divided into 2 vertical regions of equal area to represent the center and outer pathways. To initiate a clockwise wave, we electrically isolated the central and outer regions and initiated a propagating planar wave by external stimulation along the top boundary of the outer region. Once the wave was created, the electrical connection between the central and outer regions was established, and a spiral rotating reentering wave was initiated. The spatial resolution was 100 μm in the direction of the fiber orientation and 20 μm in the direction transverse to the fibers. The intracellular resistivity in the longitudinal direction was 0.5 k · cm and in the transverse direction was 10 k · cm to simulate the reduced side-to-side coupling between IZ cell pairs as reported by Yao et al.15
We completed 4 sets of simulations. In the first, to simulate reentry in a substrate with identical cell APs, we used our generic model14 of an IZ at nodes in both the center and outer regions. In the second, all ionic currents except INa as well as changes in ionic concentrations and intracellular calcium handling were formulated as reported in the model of an IZ.14 In the third set, all ionic currents except ICaL were formulated as reported for IZ.14 In the fourth set, we combined regional changes in INa and ICaL and K+ currents to formulate complete cell models of IZc and IZo.
Immunocytochemistry
Some cells dispersed for the electrical studies described above were plated on laminin-coated glass chamber slides and then fixed with 4% paraformaldehyde for 30 minutes, rinsed with PBS (Sigma), blocked in 2% avidin-PBS and rinsed in PBS, then blocked in 2% biotin-PBS and rinsed in PBS. Cells were then stained for Nav1.5 (Data Supplement).
Statistics
All values are represented as mean±SEM. n is the number of cells, and N is the number of animals. A value of P<0.05 was considered significant. For multiple comparisons (eg, see Table 1), ANOVA was used to determine whether sample mean values between groups were significantly different. If so, a modified t test with Bonferroni correction was used (Sigmastat, Jandel Scientific). Fisher’s exact test was used to determine the change in number of cells with specific location of Na+ channel antibody (see Figure 3B below).
Results
Sodium Currents
Compared with NZs, average peak INa densities of IZc and IZo were significantly smaller, but IZo peak INa did not differ from that of IZc (Vh=–100 mV) (Figure 2, A and B). Average current density-voltage relations of IZc and IZo also differed from that of NZs (Figure 2C). Current tracings of INa in a NZ, IZc, and IZo are shown in Figure 2B. Note that in the IZo, the time constants of current decay and times to peak of INa were greater than those of IZc (P<0.05) (Table 1). I/Imax curves of IZc showed a significant negative shift (&8 mV) compared with that of NZs (Table 2). IZo V0.5 values did not differ from the NZ value. The average activation curve of IZc also showed a significant negative shift (&5 mV) compared with that of NZs (P<0.05). This differed significantly from IZo (Table 2). Thus, a negative shift in both activation and I/Imax curves was observed in IZc. INa recovery from inactivation in IZc and IZo was slowed compared with that of NZs; however, values in IZc and IZo did not differ (Figure I, Data Supplement). Finally, clamp protocols designed to test closed-state inactivation were completed as before16 and showed that the time course of closed-state inactivation was markedly accelerated in IZc (Table 3). In sum, although INa density was reduced in both IZc and IZo, the kinetic properties of IZc INa were markedly altered.
Immunocytochemistry: Nav1.5 in Canine Epicardial Cells
Previous reports in mouse ventricular cells from normal hearts have suggested that the cardiac Na+ channel isoform, Nav1.5, localizes primarily to gap junction regions.17 Because other proteins also localize to this area in normal canine epicardial cells,18,19 we hypothesized that like Cx43 and Kv1.5 proteins, the cardiac Na+ channel protein may redistribute around the cell membrane (lateralize) in IZs. Thus, we completed a series of immunostaining experiments in IZc and IZo and compared their Nav1.5 staining pattern with that of NZs. We found that Nav1.5 staining in NZs was uniform, with staining all along the sarcolemma (SL) in all optical sections as well as in the gap junction region. There was no staining of T-tubular membranes. SL staining of an NZ showed a specific pattern, with staining being separated every 1.9 μm by a small region of no staining (because of T-tubular invaginations) (Figure 3A, inset). Most NZs studied showed this staining pattern (Figure 3B), whereas 20% showed a gap staining pattern accompanied by nonuniform staining of the SL. In contrast, most cells from the EBZ (62% and 68% IZc and IZo, respectively) showed Nav1.5 gap staining with variable, nonuniform SL staining (Figure 3, A and B) (P<0.05 each IZ group versus NZ). However, there was no difference in the percentage of cells with this abnormal staining pattern in IZc versus IZo.
Thus, although there were electrophysiological differences between IZc and IZo, both cell types showed abnormal cell surface staining for the Nav1.5, -subunit of the cardiac Na+ channel. This is consistent with the observed reduction in Na+ current density.
Ca2+ Currents
Peak L-type Ca2+ currents in IZc and IZo were reduced, similar to each other, but differed from NZs (Figure 4). Reduced Ca2+ currents in IZo and IZc occurred at most voltages (Figure 4B). However, the decays of the small peak IZc currents were faster than those in IZo and NZs (Figure 4C). This is interesting, considering the well-known current dependence of Ca2+ channel inactivation. Activation curves were similar in all 3 cell types. However, there were differences in steady-state availability curves between IZc and IZo, with there being a significant negative shift in IZc versus NZs (Table 4). Although there is a significant slowing in recovery in both IZc and IZo versus NZs, there is no difference between the 2 IZ cell groups (Table 5).
In a subset of cells from each group, we tested whether we could augment peak Ca2+ currents with -adrenergic stimulation as described by Aggarwal and Boyden.20 Figure II (Data Supplement) shows that compared with NZs, in which isoproterenol 1 μmol/L increases peak current 2.72±0.15-fold (n=10), it had a similar effect in IZc (2.13±0.2-fold, n=10) and a diminished effect in IZo (1.79±0.13-fold) (n=14) (P<0.05 versus NZ).
K+ Currents
As reported previously,7 most IZ cells have no transient outward current that can be elicited with depolarizing clamp steps. In both IZc and IZo, this is also true. Only 5 of 11 IZc and 2 of 11 IZo cells had a measurable Ito in this study (Figure III, Data Supplement). However, the densities of these transient currents in both cell groups did not differ from each other and were small compared with NZs (&20 pA/pF at +40 mV).21 Furthermore, there were no differences in Isus or Cs-sensitive currents (IK1) in cells of the 2 IZ groups.
In sum, IZc and IZo have reduced ICaL. However, there was no difference in the peak ICaL in IZc and IZo. IZc cells showed changes in Ca2+ current kinetics: an acceleration of current decay, a shift in availability, and a slowing in recovery from inactivation. The small ICaL of IZc responded normally to -adrenergic stimulation, whereas IZo ICaL did not. There are similar changes in K currents in both IZc and IZo.
Computer Simulations
Reentry in a Uniform Model of the EBZ
To simulate reentry in a substrate with identically remodeled cells (same APs), we used our generic IZ cell model14 for both the center and outer regions. Reentry was stable for the duration of the simulation (5 seconds), with a CL=210 ms (Movie I, Data Supplement). Although it is not unexpected to obtain stable reentry in such a medium, experimental results reported above show that there are significant differences in INa and ICaL between IZc and IZo. Therefore, to understand the stability of reentrant circuits in the EBZ, we need to understand the conditions under which EBZ circuits are stable in this remodeled substrate.
Effect of INa Remodeling on Reentry
Therefore, we adapted our generic IZ cell AP model14 to generate a cell model of an IZc and an IZo by incorporating the INa data (for changes in rate constants/equations used, see Data Supplement). Figure 5 shows the calculated INa density-voltage curves, activation and inactivation curves, and simulated APs of the IZc (solid line) and IZo (dotted line). Except for a small reduction in AP plateau, there were no differences in APD90 (Figure 5D). Notably, there were differences in the maximum depolarization rate and refractory period as measured in a 1D cable using a S1-S2 protocol.14 As a result, conduction velocity was decreased by &10% in the IZc cable. Both cell types exhibited postrepolarization refractoriness, with IZc exhibiting a longer refractory period (215 ms) compared with that of IZo (198 ms).
To understand whether differences only between INa in the IZc and IZo cells promote or prevent stability of the reentrant circuit, we performed the computer simulations illustrated in Figure 6A. Note that the reentrant wave was not stationary terminating after 6 rotations (&1400 ms) (Movie II, Data Supplement). In sum, although the tip trajectory and the lines of functional block during the 6 beats stayed within 5 mm of the boundary of the central and outer regions (Figure 6A, right), differences in refractoriness between cells of the central and outer regions produced by the changes in INa in these cells caused the spiral to drift into the IZc region. Eventually, it is extinguished by colliding with the top boundary of the preparation and refractory tissue of the IZo cells.
Effect of ICaL Remodeling on Reentry
To understand how the kinetic differences in ICaL seen in IZc and IZo would affect the AP, we incorporated the averaged data into our cell model. While these simulations showed differences in APs (the ERP was 188 ms in the IZc and 224 ms in the IZo) between IZc and IZo, only a nonsustained tachycardia could be induced in the 2D array (Movie III, Data Supplement). The large difference in ERP (36 ms) caused a drift in the reentrant circuit along the boundary.8,9 Thus, differences only in ICaL in the different regions of the reentrant circuit did not promote a stable line of functional block.
Effects of Remodeling of Both INa and ICaL on Reentry
We next produced cell models of IZc and IZo that incorporated both INa and ICaL data. Results from our simulations are shown in Figure 5, E and F. The combination of the regional differences of both inward currents produced an IZo AP (and ERP) that was longer than IZc AP, but the ERP difference (13 ms) was smaller than when INa (17 ms) or ICaL (36 ms) differences were considered alone. This is because remodeling in INa and ICaL affects the ERP in the central and outer pathway in opposite directions: remodeling of INa makes the ERP longer in IZc than in the IZo; remodeling of ICaL makes the ERP longer in IZo than in IZc. Overall, when the remodeling of both currents is taken into account, differences in ERP decrease. Notably, although ERP was longer in IZo, velocity was increased over IZc (Figure 5F).
To understand whether together, INa and ICaL differences in IZc and IZo promote or prevent stability of the reentrant circuit, we performed the simulations illustrated in Figure 6, C and D. The circulating wave was stationary, forming a line of functional block between the 2 regions (Figure 6C) for 9 rotations, after which it terminated (Movie IV, Data Supplement). Figure 6C shows isochronal maps of the last 5 beats of the reentrant wave. During beats 5, 6, and 7, the reentrant circuit is stable at the boundary between IZc and IZo (Figure 6C). During beats 5 and 6, sites g–a (Figure 6C) activate transversely without delay. However, the wave of beat 7 cannot propagate from f to e because of postrepolarization refractoriness of site e (Figure 6D, asterisk, small hump in e). Site e is eventually activated after the wave turns around a longer line of functional block (Figure 6C). The premature activation of IZc cells during beat 7 causes a decrease in APD. Figure 7 illustrates the dynamics of INa (Figure 7B) and ICaL (Figure 7C) during reentry at 2 sites, one in the central pathway (solid line) and another in the outer pathway (dashed line) and shows that the decrease in APD in the central pathway is the result of a reduced ICaL (Figure 7C) that results from the acceleration in current decay and its delay in recovery (Table 2). Because IZc APD is shortened, the reentering wave in beat 8 propagates readily across the boundary between the central and outer pathway, causing a further reduction in IZc APD (note the decrease in APD during beats 6 to 7–8 at sites b, c, and d in Figure 6B), thus allowing the reentering wave to rotate inside the center region.
Discussion
Location of the central isthmus of a reentrant circuit or the area of slowed conduction of VTs has been used as an approach for ablation of these tachycardias.22–24 Although this has been useful, sometimes the lines of block and therefore the isthmus are difficult to locate, and the reentry is nonsustained or not inducible. Thus, we have approached this problem by determining the ionic determinants of the lines of functional block that form during a reentrant tachycardia. For these studies, we used the well-studied VTs that occur in the canine epicardium 5 days after coronary occlusion.25 In our studies, we mapped reentrant circuits with CLs similar to those previously described in this model.26,27 Our maps showed that lines of block formed between the center and outer regions. Using tissue slices obtained at each side of the line of block, we describe here the differences in ionic current function in the cells from these 2 regions.
Similar to our previous findings16 using somewhat larger tissue slices from the EBZ (and for which tachycardias had not been mapped), we found that IZc and IZo cells have a reduced INa density. However, we found that there are significant differences between the kinetic parameters of INa in IZc versus IZo. IZc INa tended to enter inactivation from a closed state more rapidly, displaying shifts in both availability and activation curves. These differences are not a result of differences in experimental techniques, because we sampled both IZc and IZo, controlling for time after membrane rupture (see figure legends).
The mechanism of these differences is unknown at this time. With the immunocytochemistry experiments presented here, we show that there is a loss of the protein of the -subunit Nav1.5 in cells from the EBZ (Figure 3). Thus, there is marked structural remodeling of this protein in IZs. However, quite different from the structural remodeling of other proteins in cells of the EBZ,18,19 we find that there is a loss of Nav1.5 protein at the cell surface in both IZc and IZo, consistent with reduced peak INa. This staining differs from the staining pattern we observed in NZs, which showed significant SL, non-T-tubular staining of Nav1.5 protein with some gap junctional staining and from the Nav1.5 pattern seen in mouse ventricle by our laboratory (data not shown) and by others.17 Our findings in normal canine ventricle cells are consistent with the findings of several patch-clamp reports in which investigators have studied single Na channels in cardiac cells (see, eg, Berman et al28). In none of these reports was it stated that single sodium channel records could be obtained only if the investigator clamped regions of the gap junction. In both IZc and IZo, we found that surface staining for Nav1.5 was nonuniform, non-T-tubular, and gap junctional. In fact, in some IZs, gap junctional staining was the only staining observed. The change in the location of Nav1.5 proteins may have little effect on the cell electrophysiology but may affect propagation.29
Ionic Current Heterogeneity and Stability of Reentrant Circuits in the EBZ: Pharmacological Implications
One of the main results of our study is that despite differences in INa and ICaL channel properties in the IZc and IZo, the resulting change in refractory period tends to stabilize reentry in the EBZ. An important consequence of this is that the stability of the reentrant circuit is not a result of the remodeling of a single current (ie, INa or ICaL) but rather is determined by how any combination of remodeled currents affects the refractory period. This insight has important pharmacological implications. We suggest that one therapeutic strategy for converting a sustained VT to a nonsustained VT would be to make the refractory periods of the central and outer pathway as different as possible. Such a difference in ERPs would cause the reentrant wave to drift and eventually stop. For example, if we chose to use an ICaL agonist to decrease the differences in ICaL between IZc and IZo, we might predict that the circuit would become unstable and terminate, because regional differences in INa alone show unstable circuits (Figure 8). In fact, this has been reported for this model.27 Conversely, if we chose to normalize INa differences between IZc and IZo (eg, by using a drug that would relieve closed-state inactivation of IZc cells), we might expect the circuit to be unstable and tachycardias to terminate. This idea is not in conflict with the idea that pharmacological manipulation of anisotropy could lead to destabilization of reentry in this model (but this was not tested here). What we show here is that manipulation of anisotropy is not the only way to destabilize reentry. From our computer simulations, we suggest that the possibility of destabilizing anisotropic reentry can be done by creating gradients of refractory period in the highly remodeled center/outer regions. All that this requires is sufficient quantitative knowledge of the ion channels that are critical (as opposed to bystanders) in the remodeling process.
Time Course of Remodeling and Inducibility of VT
Our results here may also explain why in this experimental model, in which there is a time course in the inducibility of the VT, a sustained VT is induced in some animals but not in others.2 The remodeling of ion channels is a dynamic process, as we know from both ion channel studies on cells from acute and more healed infarcts3,21 and from this study, in which the dynamics of the ion channel changes in different regions of a circuit affect the stability of the circuit. Each ionic current contributing to the AP waveform has its own time course of remodeling and reversal. Therefore, if the time course of remodeling in cells in each region for each current were known, we should be able to predict the occurrence of sustained VT in the substrate in question. Sustained VTs would occur only when the sum of all ionic current remodeling results in center/outer regions with similar ERPs. In fact, if the time course of remodeling of one inward current were altered, we predict that we would never have a substrate that would sustain a reentrant wave.
Ion Channel Remodeling, Gap Junction Remodeling, and Stability of Reentrant Circuits
In this model of reentry, called anisotropic reentry, lines of block form in regions of slow transverse propagation associated with the nonuniform anisotropic nature of this substrate. Peters et al30 suggested a relationship between the location of the stable lines of block and the microanatomy of the EBZ as well as the distribution of abnormal connexin43 staining in EBZ cells. A limitation of our in silico representation of the EBZ is that it is one cell deep and has no structural components (macro or micro). It is clear from several of our previous studies, as well as those of others, that the electrical remodeling occurring in the 5-day EBZ is not limited to SL ionic channels. Gap junction distribution in the central pathway of reentrant circuits and gap junction conductance between side-to-side coupled cell pairs (direction transverse to the fiber orientation)31 are also remodeled. Although the functional effects of gap junction remodeling on anisotropy are still unknown (and currently under investigation), it is possible that changes in cell coupling and anisotropy contribute to further stabilization of reentry in the EBZ.32 This is supported by the fact that even though the incorporation of the remodeling of INa and ICaL in the models of IZo and IZc tends to stabilize reentry, the simulated reentry is not as sustained as it is in the experimental model. Again, we do not believe that the findings we report here invalidate the concept that anisotropy is important for stabilizing reentry. Rather, we contend that any mechanism of reentry in the EBZ should take into account all new findings, such as those presented here, where cells of the center and outer pathways show differences in ionic current function.
Limitations
We have shown that when we incorporate in a computer model ionic currents measured in IZc and IZo, reentrant circuits tend to stabilize. However, in the in situ infarcted hearts from which the IZc and IZo cells were harvested, reentrant tachycardias were stable for more that 30 seconds. Thus, it is possible that the remodeling of SL currents alone is insufficient for stabilization of reentrant circuits for longer periods. Other factors like gap junction remodeling, anchoring to anatomic obstacles, or the presence of areas of discontinuities may contribute to the stability of reentrant circuits.
Acknowledgments
This study was supported by National Institutes of Health grants HL-66140 and HL-30557.
References
Dillon S, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988; 63: 182–206.
Wit AL, Janse MJ. The Ventricular Arrhythmias of Ischemia and Infarction: Electrophysiological Mechanisms. Mount Kisco, NY: Futura Publishing Co, Inc; 1993.
Pinto JMB, Boyden PA. Electrophysiologic remodeling in ischemia and infarction. Cardiovasc Res. 1999; 42: 284–297.
Pu J, Boyden PA. Alterations of Na+ currents in myocytes from epicardial border zone of the infarcted heart: a possible ionic mechanism for reduced excitability and postrepolarization refractoriness. Circ Res. 1997; 81: 110–119.
Aggarwal R, Boyden PA. Diminished calcium and barium currents in myocytes surviving in the epicardial border zone of the 5 day infarcted canine heart. Circ Res. 1995; 77: 1180–1191.
Jiang M, Cabo C, Yao J-A, Boyden PA, Tseng G-N. Delayed rectifier K currents have reduced amplitudes and altered kinetics in myocytes from infarcted canine ventricle. Cardiovasc Res. 2000; 48: 34–43.
Lue W-M, Boyden PA. Abnormal electrical properties of myocytes from chronically infarcted canine heart: alterations in Vmax and the transient outward current. Circulation. 1992; 85: 1175–1188.
Fast VG, Pertsov AM. Shift and termination of functional reentry in isolated ventricular preparations with quinidine-induced in homogeneity on refractory period. J Cardiovasc Electrophysiol. 1992; 3: 255–365.
ten Tusscher KH, Panfilov A. Reentry in heterogeneous cardiac tissue described by the Luo-Rudy ventricular action potential model. Am J Physiol. 2003; 284: H542–H548.
Harris AS. Delayed development of ventricular ectopic rhythms following experimental coronary occlusion. Circulation. 1950; 1: 1318–1328.
Coromilas J, Costeas C, Deruyter B, Dillon S, Peters NS, Wit AL. Effects of pinacidil on electrophysiologic properties of epicardial border zone of healing canine infarcts: possible effects of KATP channel activation. Circulation. 2002; 105: 2309–2317.
Pu J, Balser J, Boyden PA. Lidocaine action on sodium currents of ventricular myocytes from the epicardial border zone of the infarcted heart. Circ Res. 1998; 83: 431–440.
Baba S, Dun W, Boyden PA. Can PKA activators rescue Na channel function in epicardial border zone cells that survive in the infarcted canine heart Cardiovasc Res. 2004; 64: 260–267.
Cabo C, Boyden PA. Electrical remodeling of the epicardial border zone in the canine infarcted heart: a computational analysis. Am J Physiol. 2003; 284: H372–H384.
Yao J-A, Hussain W, Patel P, Peters NS, Boyden PA, Wit AL. Remodeling of gap junctional channel function in epicardial border zone of healing canine infarcts. Circ Res. 2003; 92: 437–443.
Pu J, Boyden PA. Alterations of Na+ currents in myocytes from epicardial border zone of the infarcted heart: a possible ionic mechanism for reduced excitability and postrepolarization refractoriness. Circ Res. 1997; 81: 110–119.
Maier SKG, Westenbroek RE, Schenkman KA, Feigl EO, Scheuer T, Catterall WA. An unexpected effect for brain type sodium channels in coupling of cell surface depolarization to contraction in the heart. Proc Natl Acad Sci U S A. 2002; 99: 4073–4078.
Mays DJ, Boyden PA, Tamkun MM. Redistribution of the Kv1.5 K+ channel protein on the surface of myocytes from the epicardial border zone of the infarcted canine heart. Cardiovasc Pathobiol. 1997; 2: 79–87.
Peters NS, Wit AL. Myocardial architecture and ventricular arrhythmogenesis. Circulation. 1998; 97: 1746–1754.
Aggarwal R, Boyden PA. Altered pharmacologic responsiveness of reduced L-type calcium currents in myocytes surviving in the infarcted heart. J Cardiovasc Electrophysiol. 1996; 7: 20–35.
Dun W, Baba S, Yagi T, Boyden PA. Dynamic remodeling of K+ and Ca2+ currents in cells that survived in the epicardial border zone of the canine infarcted heart. Am J Physiol. 2004; 287: H1046–H1054.
Stevenson WG, Friedman PL, Sager PT, Saxon LA, Kocovic DZ, Harada T, Wiener I, Khan H. Exploring postinfarction reentrant tachycardia with entrainment mapping. J Am Coll Cardiol. 1997; 29: 1180–1186.
Harada T, Stevenson WG, Kocovic DZ, Friedman PL. Catheter ablation of ventricular tachycardia after myocardial infarction: relation of endocardial sinus rhythm late potentials to reentry circuit. J Am Coll Cardiol. 1997; 30: 1015–1023.
de Bakker JM, van Capelle FJL, Janse MJ, Wilde AA, Coronel R, Becker AE, Dingemans KP, van Hemel NM, Hauer RN. Reentry as a cause of ventricular tachycardia in patients with chronic ischemic heart disease: electrophysiologic and anatomic correlation. Circulation. 1988; 77: 589–606.
Janse MJ, Wit AL. Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction. Physiol Rev. 1989; 69: 1049–1169.
Dillon S, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988; 63: 182–206.
Cabo C, Schmitt H, Wit AL. New mechanism of antiarrhythmic drug action: increasing inward L type calcium current prevents reentrant tachycardia in the infarcted canine heart. Circulation. 2000; 102: 2417–2425.
Berman MF, Camardo JS, Robinson RB, Siegelbaum SA. Single sodium channels from canine ventricular myocytes: voltage dependence and relative rates of activation and inactivation. J Physiol. 1989; 415: 503–531.
Kucera JP, Rohr S, Rudy Y. Localization of sodium channels in intercalated disks modulates cardiac conduction. Circ Res. 2002; 91: 1176–1182.
Peters NS, Coromilas J, Severs NJ, Wit AL. Disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation. 1997; 95: 988–996.
Yao J-A, Chen S, Hussain W, Peters N, Boyden PA, Wit AL. Gap junction coupling of myocytes in central common pathway of reentrant circuits in infarct epicardial border zone: comparison with outer pathway type myocytes. Heart Rhythm. 2004; 1: S4.
Cabo C, Boyden PA. Gap junction remodeling stabilizes reentry causing ventricular tachycardia in infarcted canine hearts: a computational study. Heart Rhythm. 2004; 1: S84.(Shigeo Baba, MD; Wen Dun,)