Published online before print February 9, 2006, doi: 10.1161/01.RES.0000209770.72203.01
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 American Heart Association, Inc.
UltraRapid Communication |
Structure-Related Initiation of Reentry by Rapid Pacing in Monolayers of Cardiac Cells
From the Department of Biomedical Engineering, The Johns Hopkins University, Baltimore, Md. Current address for W.B.: Department of Biomedical Engineering, Duke University, Durham, NC.
Correspondence to Dr Leslie Tung, Department of Biomedical Engineering, The Johns Hopkins University, 720 Rutland Ave, Baltimore, MD 21205. E-mail ltung{at}bme.jhu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
This study examines how a zigzag pattern of conduction, a form of structural heterogeneity frequently found in old or diseased hearts, affects the vulnerability to reentry during rapid pacing. A central rectangular island (8x4 mm) containing a predefined zigzag pattern was created in cultured isotropic monolayers of neonatal rat ventricular myocytes. Impulse propagation was optically mapped from 253 sites using voltage-sensitive dye and was anisotropic within the zigzag island. With increasing interval between neighboring transverse connections (a), relative to the distance between longitudinal strands (b), transverse conduction velocity (CV) decreased to 66±6%, 20±2%, and 15±2% of CV in the surrounding isotropic region, whereas longitudinal CV increased to 102±8%, 113±12%, and 131±23% for a:b ratios of 1:1, 1:5, and 1:9, respectively. During rapid pacing, propagation distal to the island was steered from the side of the island with more transverse connections ("dominant" side) toward the side with fewer connections ("weak" side). Increased asymmetry in the pattern accentuated this effect, and resulted in increased rate-dependent differences in CV on the 2 sides. Consequently, a functional obstacle formed on the weak side, followed by development of single loop reentry. The reentrant wave revolved around a line of block defined by the border of the island. Reentry chirality was determined by the weak side location, and the pacing rate needed to initiate reentry decreased with increased asymmetry in the pattern. In conclusion, reentry is readily induced by rapid pacing in confluent cardiac cell monolayers containing a central and asymmetric island of zigzag conduction.
Key Words: reentry • arrhythmia • cardiac electrophysiology • optical mapping • cell culture
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Reentry is the circus movement of electrical waves in cardiac tissue and underlies various arrhythmias, including atrial flutter and fibrillation,1,2 and ventricular arrhythmias that follow myocardial ischemia or infarction.3 To induce reentry in experimental preparations or computer models, rapid pacing is typically applied.4 Previous studies have shown that dynamic instabilities may arise during rapid pacing as a result of restitution kinetics (for action potential duration [APD] and conduction velocity [CV]) or calcium cycling and can produce temporal and spatial alternans, local conduction block, wave break, and reentry-based tachycardias or fibrillation.5–10 On the other hand, spatial heterogeneity of active ionic currents in the cell membrane has been shown to cause wave break and reentry in the absence of unstable restitution dynamics.11 However, the possibility that heterogeneity in tissue structure, in and of itself, may also serve as a locus for the initiation of reentrant activity during rapid pacing has not been systematically investigated. Such was the goal of this study. Our particular interest lies in a frequent and clinically relevant form of structural heterogeneity—the zigzag pattern that results from infiltration of fibrotic tissue into the interstitial spaces between myocardial fibers.
Zigzag conduction was first reported by Spach et al12 in aged human atria and has also been found in diseased hearts with dilated cardiomyopathy13 or myocardial infarction.14 It was first visualized by Koura et al15 in old canine myocardium using high-resolution optical mapping. The presence of zigzag trajectories leads to immensely slowed transverse conduction perpendicular to the fiber orientation and promotes the vulnerability to reentrant arrhythmias,12,15,16 but the detailed mechanism of initiation of reentry remains unclear.
The cell culture approach eliminates much of the complexity found in native tissue preparations (eg, presence of microvasculature) and the tissue injury from excision for an ex vivo experimental preparation. Methods of patterned growth have been used to create linear strands of cultured cardiac cells for the study of impulse propagation at a microscopic scale17 and have clarified the functional effects of tissue expansion and branching geometry.18–20 Using advanced cell patterning techniques, we designed and created simplified 2D tissue models consisting of isotropic monolayers of cardiac cells that contain a central region with a zigzag structure. Optical mapping was used to monitor the macroscopic electrical activity in the cell monolayers, enabling us to study how this type of structural heterogeneity contributes to pacing-induced initiation of reentry.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Cell Culture
All experiments involving animals conformed to the protocols in the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, Revised 1996), and the animal protocol was approved by the Johns Hopkins Animal Care and Use Committee. Cardiac cells were dissociated from ventricles of 2-day-old Sprague-Dawley rats (Harlan, Indianapolis, Ind) using trypsin (Amersham Life Sciences, Arlington, Ill) and collagenase (Worthington, Lakewood, NJ) and resuspended in M-199 culture medium (Life Technologies, Rockville, Md) supplemented with 10% FBS (Life Technologies) as previously described.21 Approximately 1x106 cells were plated on prepared 22-mm diameter circular cover slips. At day 2 following cell plating, serum was reduced to 2% to reduce proliferation of noncardiac cells. Such a reduction has been shown to have no adverse effects on cell size, cellularity index, capture rate, connexin-43 expression, or conduction velocity.22
Microcontact Printing
Directed cell growth following a predefined zigzag pattern was achieved using a modified method of microcontact printing of adhesive proteins.23 Briefly, desired patterns were produced on 3-inch silicon wafers (Ultrasil, Hayward, Calif) using SU-8 photoresist (MicroChem, Newton, Mass) by standard photolithography. Degassed polydimethyl siloxane (PDMS) prepolymer mixture (10 part of base and 1 part of curing agent by weight) (Sylgard 182, Dow Corning, Midland, Mich) was poured on the silicon master. After overnight baking, the solidified PDMS polymer was peeled off the master, cut into individual stamps and coated with 50 µg/mL human fibronectin (Sigma, St Louis, Mo) for 1 hour. Twenty-two millimeter diameter glass cover slips (Fisher Scientific, Pittsburgh, Pa) were spin coated with PDMS prepolymer mixture (Sylgard 184; Dow Corning), baked in an oven at approximately 70°C for 1 hour and treated with UV-generated ozone (UVO) for 8 minutes (PSD-UV system, Novascan Technologies Inc, Ames, Iowa) for sterilization and increased hydrophilicity. Within 20 minutes after UVO treatment, the fibronectin-coated stamps were washed with deionized and distilled water and applied to the PDMS surface for 30 to 45 minutes. The cover slips were then dipped in 0.2% (wt/vol) Pluronic F-127 (Molecular Probes, Eugene, Ore) for 20 minutes to block the regions uncovered by fibronectin. After rinsing with PBS, the cover slips were ready for cell plating.
Experimental Setup
Cover slips with cell monolayers were placed in a custom-designed chamber filled with warm (T=36±0.5°C) oxygenated Tyrode’s solution (in mmol/L: 135 NaCl, 5.4 KCl, 1.8 CaCl2, 1 MgCl2, 0.33 NaH2PO4, 5 HEPES, 5 glucose), stained with voltage-sensitive dye di-4-ANEPPS (10 µmol/L) (Molecular Probes) for 5 minutes, and continually superfused with Tyrode’s solution afterward. A plexiglass cover was placed over the chamber to minimize motion artifact resulting from solution flow, and a point stimulus was delivered by a unipolar platinum electrode through one of several holes drilled in the cover. A field stimulus was delivered by 2 parallel platinum line electrodes spaced 2.7 cm apart. Transmembrane potentials of cardiac cells were visualized and monitored by an in-house contact fluorescence imaging (CFI) system that has been previously described24,25 and has now been upgraded as follows. An array of 253 optic fibers arranged in a 17-mm-diameter hexagon was placed directly under the experimental chamber to collect the optical fluorescent signals. A red filter made of a No. 1 glass coverslip spin coated with 3 layers of red photosensitive resin (PSCred; Brewer Science, Rolla, Mo) was placed at the bottom of the chamber to block the excitation light and pass the red emission signal. An LED light source26 consisting of 26 high power green LEDs (Kingbright, Taipei, Taiwan) with an interference filter (530±25 nm) delivered excitation light from above the chamber. Optical signals were low-pass filtered at 320 Hz, amplified, and sampled at 1 kHz by four 64-channel, 16-bit A/D acquisition boards (Sheldon Instruments, San Diego, Calif). Data were displayed and analyzed using software written in Visual C++ (Microsoft, Redmond, Wash), LabVIEW (Texas Instrument, Dallas), and MATLAB (MathWorks, Natick, Mass). A high-resolution charge-coupled device camera (Hitachi, Tokyo, Japan) with a 50 mm, f/1.4 lens with no. 1 and no. 2 close-up lenses (Nikon, Tokyo, Japan) was placed above the chamber to capture images of the cell monolayer and the optical bundle. The images were then superimposed to verify the position of the optical recording sites relative to the pattern in the monolayer.
Experimental Protocol
Before recording, the cover slips were carefully inspected using light microscopy to exclude any monolayers with broken cell strands, transverse connections lying out of place in the central patterned area or underconfluence of cells in the surrounding region. After the cover slip was placed in the experimental chamber, the central zigzag pattern was moved to the center of the optical recording area and aligned horizontally with the rows of recording sites, as shown in Figure 1A. For measurement of the transverse and longitudinal conduction velocity (TCV and LCV) in the zigzag region, a 3-Hz point stimulus was delivered at the top or bottom side (viewed from above), and left or right side of the monolayer, respectively. To induce reentry by rapid pacing, the point electrode was placed on the top or bottom side of the monolayer, so that the excitation wave propagated transversely across the cell strands in the zigzag pattern. The excitation threshold was first determined by delivering several test pulses, and then the stimulus intensity was increased to 1.5x threshold for the remainder of the experiment. A train of 30 pulses was delivered through the point electrode starting at 3 Hz and incremented by 1 Hz up to 5 Hz and 0.5 Hz above 5 Hz until reentry was induced or 1:1 capture (1 stimulus: 1 response) was lost (0.5 Hz above the maximum capture rate [MCR]). A 2-sec recording was made at the end of the pulse train that included the last few pulses and the period immediately following termination of pacing. If reentry was induced, 2-sec recordings were made at 1 minute, 5 minutes, and 10 minutes after initiation to check the stability of the reentry. Single or multiple field stimuli were then delivered to terminate the ongoing reentry. If the reentry was successfully terminated, the pacing train was reapplied at a rate 0.5 Hz higher than the previous reentry induction rate (RIR) and then incremented by 0.5 Hz for the determination of MCR. This cycle was repeated until either loss of capture and identification of MCR was obtained, or reentry could not be terminated by field stimuli. If MCR was less than 7 Hz, pinacidil (50 µmol/L), a KATP channel opener, was used to accelerate repolarization, shorten the effective refractory period and raise MCR (in 8 of 31 monolayers), so that too low an MCR would not become a confounding factor that could influence the reentry initiation outcomes (by not permitting high pacing rates to be applied). Dynamic CV restitution relations were obtained by temporally averaging the last 4 beats of the 30 pulse stimulus trains applied at the stepwise increasing rates.
|
Data Analysis
The raw optical signals were detrended by subtracting a fitted second-order polynomial curve and then low-pass filtered with a fourth order elliptical filter. Isopotential and isochrone maps were generated from the processed signals. The activation time was defined as the instant of maximum positive slope during the depolarization phase of the action potential. CV was measured along 4 to 5 manually selected paths and averaged spatially and temporally over 3 to 4 beats. LCV and TCV in the zigzag pattern were normalized to the CV in the surrounding isotropic tissue (ICV). RIR for each monolayer was normalized to MCR for that monolayer.
Statistical Analysis
Data were expressed as mean±SD and analyzed using Student t test or 1-way ANOVA followed by Tukey post test. Differences were considered to be significant when P<0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Anisotropic Conduction in the Region With Zigzag Course of Activation
Rectangular islands (8x4 mm) containing a predefined zigzag pattern were created in the center of 22-mm diameter cultured monolayers of neonatal rat ventricular myocytes (NRVMs). The position of the zigzag pattern relative to the 253 optical recording sites is depicted in Figure 1A and was verified by superimposing video images of the pattern and optical recording bundle. Outside the island, cells were confluent and randomly oriented. The action potentials outside (traces a through c) and inside (traces d through f) the island were similar, akin to the observations that action potentials in healed infarcts are like those in normal tissue, despite deranged cellular connections.14,27 APD at 80% repolarization (APD80) showed no significant difference for the 2 groups of sites inside and outside the zigzag island in 27 of 34 monolayers (averaged P=0.53±0.34). The zigzag structure is shown in detail in Figure 1B, and the ratio of the separation between neighboring cell strands (a) to the distance between neighboring transverse connections (b) was set to 1:1 (a=300 µm, b=300 µm), 1:5 (a=300 µm, b=1500 µm), or 1:9 (a=300 µm, b=2700 µm), with the width of the cell strand=100 µm in all cases. These structures have physical dimensions and patterns intended to mimic the zigzag structures found in infarcted or aged myocardium, albeit detailed morphological information is limited.14,15 Isochrone maps derived from the optical signals are shown in Figure 2A and 2B for point stimuli at the top and right sides of the monolayer. In Figure 2A, it can be seen that the paced wave front slows maximally in the transverse direction of the zigzag pattern when it encounters the proximal edge of the pattern and then accelerates afterward as the result of the invasion of action currents from the adjacent normal regions on the left and right sides. In Figure 2B, a small increase in CV is evident along the longitudinal midline of the zigzag region, because of less electrotonic loading imposed by the adjacent normal regions at the upper and lower sides. Figure 2C and 2D summarize LCV, TCV, and the anisotropic ratio of conduction (AR=LCV/TCV) in the zigzag pattern for the different a:b aspect ratios. TCV and LCV were normalized to the velocity measured in the surrounding isotropic tissue in the same monolayer (average ICV=17±4 cm/sec, n=38). TCV decreased significantly with aspect ratio (0.66±0.06, 0.20±0.02, and 0.15±0.02, for a:b=1:1, 1:5, and 1:9, respectively, P<0.0001), with significance for all pair-wise combinations. LCV increased slightly but significantly with aspect ratio (1.02±0.08, 1.13±0.12, and 1.31±0.23 for a:b=1:1, 1:5, and 1:9, respectively; P=0.0004) with a significant difference for 1:1 versus 1:9 ratios. Consequently AR increased significantly (1.5±0.2, 5.8±0.8, and 8.7±1.3, for a:b=1:1, 1:5, and 1:9, respectively, P<0.0001), with significant differences for all pair-wise comparisons, as shown in Figure 2D.
|
Structure-Related Asymmetric Impulse Propagation Enhanced by Rapid Pacing
With point stimulation, wave-front propagation around the central zigzag island is shown in the isopotential maps of Figure 3 for a zigzag pattern with aspect ratio of 1:9 (see movie Bian_1.mpg in the online data supplement, available at http://circres.ahajournals.org, for full animation of Figure 3). At a 3-Hz pacing rate (Figure 3A), propagation was symmetric around the vertical axis bisecting the island (Map 3). At a 7-Hz pacing rate (Figure 3B), propagation was nearly symmetric, but the wave front on the right side became slightly more advanced compared with that on the left (Map 3). At a pacing rate of 9 Hz (Figure 3C), the difference in propagation times down the 2 sides was even more evident (Map 3). This difference in propagation was observed only with patterns with aspect ratios of 1:9, and not with ratios of 1:1 or 1:5.
|
Closer examination of our patterns revealed a subtle but significant difference between the left and right sides of the pattern (Figure 1B). It can be seen that the transverse connections were not perfectly symmetric with respect to the central vertical axis of the pattern, and that 1 side (in this case, the right) had a greater number of connections compared with the other side (in this case, the left). We term the 2 sides to be the "dominant" and "weak" sides, respectively. The existence of dominant and weak sides has little influence at low or high pacing rates for aspect ratios of 1:1 or 1:5. Even with a ratio of 1:9 (Figure 3), propagation was symmetric at a pacing rate of 3 Hz but became somewhat asymmetric at 7 Hz and then clearly asymmetric at 9 Hz, a rate at which CV was greatly slowed and most sensitive to small differences in source-load conditions. Moreover, the wave front was always more advanced on the dominant side of the pattern (in this case, the right side). Also, a comparison of the time indices for corresponding maps in Figure 3A through 3C shows an overall slowing of the wave fronts as pacing frequency increased. Similar results were observed in 15 other monolayers having the same pattern. Thus, our results show that asymmetry in microstructure can result in asymmetry in macroscopic propagation that is accentuated at high pacing rates and in patterns with sparse transverse connections.
Structural Asymmetry and Reentry Induction by Rapid Pacing
Rapid pacing was applied to monolayers containing central islands with various zigzag aspect ratios (n=10 for a:b=1:1, 9 for a:b=1:5, and 16 for a:b=1:9). No reentry was initiated when the aspect ratio was 1:1 or 1:5, but single loop reentry was frequently (13 of 16) formed at a sufficiently high pacing rate (9.1±1.8 Hz, n=13) for an aspect ratio of 1:9. Figure 4 depicts 1 such example of reentry induction (see supplemental movie Bian_2.mpg). Sequential isopotential maps in Figure 4A show that the asymmetric structure not only caused the wave on the dominant (right) side to propagate around the zigzag pattern faster than the wave on the weak (left) side, as seen previously in Figure 3, but also elicited a conduction block in the weak side of the zigzag region, as indicated by the pair of white solid lines in Maps 2 and 7. At the same time, the 2 waves on either side of the zigzag region collided on the weak side instead of in the center. The collision point of the 2 waves (denoted by the pairs of opposing white arrows in Maps 3 and 8) shifted more and more to the weak side with successive beats. The size and location of the region of conduction block varied somewhat from monolayer to monolayer but always occurred on the weak side, usually appearing inside the zigzag region and extending with following beats outside the region. Occasionally, additional islands of conduction block appeared on the weak side distal to the zigzag region. Consequently, the wave front on the weak side became blocked or diverted to the edge of the monolayer, whereas the wave front from the dominant side was able to advance through the region of block (which had time to recover because of its asymmetric location on the weak side) and collide with the weak-side wave front arising from the subsequent pacing pulse (Map 9). This resulted in the annihilation of both waves and left the next wave from the dominant side free of collision (Map 10) and able to evolve into a reentrant wave (Maps 11 and 12).
|
Figure 4B shows the events leading up to the conduction block in more detail for 6 sites along the conduction pathway passing through the block site. Responses to the last 17 pulses of a train of 30 stimuli are shown, as well as reentrant activity after stimulation was turned off. Conduction proceeded rapidly from site 1 through 6 following each stimulus and reversed direction during reentry. Successive beats led to a progressive slowing of conduction from sites 4 to 5, until conduction block occurred (fourth arrow from left terminating in double line). In the beats leading up to the conduction block, the amplitudes of the action potentials in sites 4 and 5 decreased with each beat and became increasingly delayed (decremental conduction). Excitation continued to occur distal to the block in site 6 because of lateral propagation around the area of block and arrival of the wave from the dominant side. After 3 more beats, the wave from site 6 (originating from the 29th beat) penetrated the block at site 5 and collided with the wave from site 4 (originating from the 30th beat, 5th arrow from left). The next wave from site 6 (originating from the 30th beat) then commenced reentry. After pacing was turned off, reentrant activity persisted, with conduction now proceeding from site 6 to site 1. The relatively longer delay from site 6 to site 5 (rightmost arrow) was because of pivoting of the wave front around the lower left corner of the zigzag region.
To further investigate the impact of asymmetry in the zigzag pattern on the initiation of reentry, we designed additional patterns with various degrees of asymmetry. Pattern Z1 (Figure 5A), referred to as slightly asymmetric, is the pattern already discussed and contained a zigzag structure (a:b=1:9, a=300 µm) over the entire rectangular island with few transverse connections. Pattern Z2 (Figure 5B), referred to as moderately asymmetric, had a zigzag pattern (a:b=1:3, a=300 µm) with a moderate number of transverse connections in one half of the island. Pattern Z3 (Figure 5C), referred to as strongly asymmetric, was similar to Z2 except that the number of transverse connections was very high in half of the island (a:b=1:1, a=300 µm) (see supplemental movie Bian_3.mpg for example of reentry development with this pattern). Pattern S0 (Figure 5D) was structurally symmetric (a=300 µm) and served as the control (lacking any transverse connections). The success rate of initiation of reentry is summarized in Figure 5E. Reentry induction was 100% successful with patterns Z2 and Z3, but had 3 failures out of 16 monolayers with pattern Z1 (presumably because of the weak asymmetry) and totally failed with symmetrical pattern S0. RIRs, normalized in each monolayer to MCR, were 0.98±0.03 (n=7), 0.86±0.07 (n=6), and 0.78±0.07 (n=6) for patterns Z1, Z2, and Z3, respectively (Figure 5F) and decreased significantly with an increase in the degree of structural asymmetry (P<0.0001), with significant differences for Z2 versus Z1 and Z3 versus Z1.
|
As described earlier (Figure 4), the rate-dependent difference in CV between the weak and dominant sides of the zigzag region was an important factor that enabled reentry to be initiated. Average CVs were calculated from 4 to 5 manually selected paths oriented along the directions of propagation just outside the weak and dominant sides of the zigzag region. The dynamic restitution relation was consistently steeper on the weak side at short cycle lengths compared with the dominant side, in a manner that became more pronounced with increasing asymmetry from pattern Z1 to Z3 (Figure 6).
|
The chirality of the reentrant wave was also directly related to the asymmetry in geometry, as summarized in Figure 7. Grouping all reentry episodes induced in patterns Z1, Z2, and Z3 together, when the dominant side of the pattern was on the right relative to the position of the pacing source, all of the reentries (19 of 19) were clockwise. Conversely, when the dominant side was on the left, all of the reentries (12 of 12) were counterclockwise. These results also correlated in 100% of the cases with the location of the collision points and conduction block being on the weak side (left for dominant right side and right for dominant left side) during reentry induction.
|
Sustained Reentry With a Line of Block
All of the induced reentries were stable and lasted for at least 10 minutes (the time of observation). The reentrant wave revolved around a line of block that was always located at either the distal or proximal edge of the rectangular island containing the zigzag pattern as shown in Figure 8 (also see supplemental movie Bian_2.mpg). It can be seen that the length of the line of block was defined by the length of the edge of the island. The pivot points of the reentrant wave were exactly at the corners of the island (Maps 2, 3, 6, and 7), where the wave front could spread rapidly in the transverse direction. Also note that when the wave front traversed the inner edge of the island (Maps 1, 2, 7, and 8), the segment inside the island moved more quickly than that outside the island, giving a protrusion-like appearance to the wave front. This was not the case when the wave front traversed the outer edge of the island (Maps 3, 4, and 5). Similar findings were obtained in all of our induced reentries (31 monolayers).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Rapid activity can occur in vivo from focal activity,28 microreentry,29 or waves emanating from a "mother rotor."30 It has been shown that rapid pacing can induce wave breaks and reentrant activity, primarily from the viewpoint of dynamic instabilities arising from rate-dependent kinetics or calcium cycling5–9 or spatial gradients in repolarization properties.11,31,32 On the other hand, the gradients of conduction and conduction block necessary for the development of reentry can also arise from other forms of nonuniformity present in a tissue substrate, including heterogeneity in cellular coupling and gap junctional proteins29,33,34 or variations in tissue structure at a cellular level.35 Our experimental results show that (1) structural heterogeneity at the microanatomical level can play an integral role in the development of reentry; (2) for zigzag structures, slow conduction and asymmetry in the zigzag pathways are key contributing factors; and 3) the resulting reentry will anchor to the site of structural discontinuities. Regions of precisely defined microstructure in the form of a zigzag pattern create a substrate that allows the reproducible induction of reentry by rapid pacing. Zigzag activation supports very slow transverse conduction (down to 2.6 cm/sec for aspect ratio of 1:9, which is comparable to that found in periinfarcted tissue,36 old myocardium,15 and infarcted muscle16), which is considered to be important for the initiation of reentrant arrhythmias.14,36 However, our experiments also suggest that a certain degree of structural asymmetry in the zigzag pattern is necessary for reentry to be readily initiated. The asymmetry contributes after many beats to the development of a functional block on the weak side, typically inside and with subsequent beats, outside the zigzag region.
Anisotropic Conduction Arising From Zigzag Activation
Our results support the notion that a zigzag course of activation underlies the abnormally slow transverse conduction in fibrotic tissue caused by aging, cardiomyopathy, or myocardial infarction.13,14,33 At the same time, longitudinal conduction in the zigzag region can be near normal, with velocity comparable to or higher than that of randomly oriented cells (isotropic tissue). This results in a very high degree of anisotropic conduction. The highest aspect ratio (1:9) in the zigzag pattern resulted in a transverse CV and anisotropy ratio that respectively were much slower and higher than that reported by Fast and Kleber37 and Bursac et al38 in anisotropic monolayer cultures of cardiomyocytes. Bursac et al found that with increasing AR, LCV increased and TCV decreased in cultures with small ARs (
1.3 to 3.7), and both LCV and TCV decreased in cultures with large ARs (3.5 to 5.6). In contrast, we found an increase in LCV and decrease in TCV over the full and higher range of ARs (1.6 to 8.8) in the zigzag structure. The increased LCV in the zigzag pattern compared with CV in the surrounding isotropic tissue was primarily attributable to cell coalignment and elongation in strands that comprised the zigzag pattern.38 The trend of a decrease in LCV when the ratio of a:b decreased can be explained by the conduction slowing effect of a branching tissue geometry as demonstrated by Kucera et al.20 In their experiments, conduction was significantly slower in cell strands with multiple branch points compared with those with just a single branch point, and in strands with interbranch distance of 150 µm compared with those with distance of 300 µm. Similarly, longitudinal propagation along a zigzag pattern with a shorter distance between neighboring transverse bridges encounters more branch points per unit length and therefore results in a slower CV.
Structural Asymmetry, CV Restitution, and Reentry Induction
In our experiments, the slowed transverse conduction of the zigzag region separated the advancing wave front into two halves. In control experiments (pattern S0), conduction slowing was symmetric in both halves, so reentry initiation could not be achieved. However, the introduction of asymmetry into the zigzag structure led to different slowing of the 2 halves of the advancing wave front, an effect that was enhanced with an increase in pacing rate and with increased number of pulses. As demonstrated by Lammers et al.39 rapid pacing induces a higher degree of inhomogeneity in CV compared with single or multiple premature beats. Our results demonstrate that the property of CV restitution (rate dependence of CV) as measured in tissue arises not only from cellular membrane properties but also from the tissue structure. At low pacing rates, CVs on the dominant and weak sides of the zigzag pattern are about the same, but at high pacing rates, CV on the weak side is lower than that on the dominant side because of fewer transverse connections (Figure 3). Therefore, the weak side has a CV restitution curve that is steeper than that of the dominant side, with a difference that increases with increasing asymmetry in the zigzag pattern (Figure 6). The formation of conduction block on the weak side of the zigzag pattern (Figure 4B) sets the stage for the development of reentry having a preset chirality, and appears to be the result of the steepened slope of CV restitution that occurs at high pacing rates.
The possibility exists that the conduction block is also the result of discordant alternans arising from intrinsic repolarization gradients that are enhanced in the presence of rectangular structural barriers, as shown by Pastore and Rosenbaum.40 Although we cannot entirely rule out gradients in repolarization, our isotropic monolayers of NRVMs possess macroscopically homogeneous ionic membrane properties and stable cellular dynamics, as judged by smooth and circular isochrones during point stimulation,38 moderate slopes in APD restitution,41 the general absence of wave breaks during rapid pacing of monolayers lacking central zigzag islands (data not shown), and electrotonic coupling that will reduce intrinsic gradients among neighboring cells.42 Furthermore, discordant alternans was not apparent in our experiments, and the structural barrier was not an anatomical obstacle but rather a region containing conductive pathways that would be expected to less effective in augmenting regional repolarization gradients.
Once formed, the region of conduction block persists and becomes a functional obstacle. The obstacle tends to grow in size with time and blocks the wave from the weak side of the monolayer. This allows the wave from the dominant side (from the nth pacing pulse) to evade the nth wave from the weak side and is the key event that permits the dominant side wave to survive. Advancement through the obstacle is possible because of the longer delay time for the dominant side wave to reach the obstacle. The unidirectional block is similar to that which occurs in an anatomical circuit by a premature stimulus that is applied asymmetrically on 1 side of the circuit,43 except that here the block evolves over many beats in the absence of extrastimuli. Subsequently, the nth wave on the dominant side will either collide with the n+1th wave on the weak side (and repeat with subsequent beats) or, if pacing has been terminated, evolve into a reentrant wave. As expected from this scenario, the chirality of the reentrant wave depends on the asymmetry (right versus left) of the zigzag pattern (Figure 7). An increase in degree of structural asymmetry results in a decrease of the pacing rate required for reentry induction (Figure 5F) and facilitates the onset of reentry.
Line of Block in Sustained Reentry
The sustained single-loop reentry rotates around a line of block located along the longitudinal edge of the island containing the zigzag course of activation. Because this line of block is fixed at a particular location, it is related to the anatomical microarchitecture. However, it is different from conventional inexcitable anatomical obstacles44,45 because paced wave fronts can pass across the site of block (Figure 3). It also differs from the functional line of block that occurs in uniform anisotropic tissue4 or homogeneous46–48 isotropic media because the zigzag line of block remains fixed in location with a size determined by the width of the zigzag region. Thus, we see a different type of behavior brought about by the microanatomical tissue structure. Similarly, lines of "apparent block" during sustained reentry in the infarct border zone were described many years ago by Dillon et al and hypothesized to be the result of nonuniform anisotropy related to the microanatomical structure.36 Peters and Wit have speculated that an area of myocardium with enhanced anisotropy caused by impaired transverse coupling might form the common central pathway of a reentrant circuit and define lines of functional block at its longitudinal interface with the surrounding normal tissue.49 The line of block that we observed at the border of the zigzag region is evidence that supports their hypothesis.
Limitations of the Study
The major limitation of the present study is the simplification of the 3D tissue architecture in real hearts into a 2D bounded substrate. Although the central position and minimal size of the zigzag region lessened the influence of boundaries on reentry initiation, the initiation mechanism of reentry in 3D myocardium may not be as straightforward as that shown in our cultured monolayers of cardiac cells. Moreover, the portion of the monolayer surrounding the zigzag region contained randomly oriented cardiomyocytes that yield homogeneously isotropic properties rather than uniformly anisotropic properties as in real cardiac tissue. This difference may have enhanced the contrast in functional properties between the zigzag patterned region and the surrounding area and facilitated the initiation of reentry even further. Additional experiments, in which the surrounding region of the monolayer is made to be anisotropic, will be necessary to resolve this issue.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
We have demonstrated the reproducible initiation of reentry in cultured monolayers of cardiac cells containing a central region of zigzag structure. We have shown how tissue microstructure can exert subtle effects on macroscopic behavior, via spatial variation in CV restitution, creation of a functional obstacle, and formation of a line of block, none of which is evident during pacing at basal rates. Asymmetry in the zigzag region, such that 1 side has more transverse connections than the other side, increases the vulnerability to reentry, so that rapid pacing can initiate reentrant waves.
|
Acknowledgments |
|---|
This project was supported by NIH grant HL66239. We thank Christopher Chen for his advice on microcontact printing and method to coat cover slips with PDMS; Felipe Aguel, Zhan Yang Lim, Yelena Nabutovsky, Barun Maskara, Marvin Chang, Alexander Jow, and Daniel Loeser for their assistance with development of the mapping system; Roland Emokpae for supplying the cardiac cells, and Yibing Zhang for assistance with the movies in the on-line supplement.
|
Footnotes |
|---|
This manuscript was sent to Michael R. Rosen, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
Original received November 4, 2005; revision received January 22, 2006; accepted January 25, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
1. Allessie MA, Bonke FI, Schopman FJ. Circus movement in rabbit atrial muscle as a mechanism of tachycardia. Circ Res. 1973; 33: 54–62.
2. Waldo AL. Mechanisms of atrial fibrillation. J Cardiovasc Electrophysiol. 2003; 14: S267–S274.[CrossRef][Medline] [Order article via Infotrieve]
3. Antzelevitch C. Basic mechanisms of reentrant arrhythmias. Curr Opin Cardiol. 2001; 16: 1–7.[CrossRef][Medline] [Order article via Infotrieve]
4. Schalij MJ, Boersma L, Huijberts M, Allessie MA. Anisotropic reentry in a perfused 2-dimensional layer of rabbit ventricular myocardium. Circulation. 2000; 102: 2650–2658.
5. Pruvot EJ, Katra RP, Rosenbaum DS, Laurita KR. Role of calcium cycling versus restitution in the mechanism of repolarization alternans. Circ Res. 2004; 94: 1083–1090.
6. Goldhaber JI, Xie LH, Duong T, Motter C, Khuu K, Weiss JN. Action potential duration restitution and alternans in rabbit ventricular myocytes: the key role of intracellular calcium cycling. Circ Res. 2005; 96: 459–466.
7. Shiferaw Y, Sato D, Karma A. Coupled dynamics of voltage and calcium in paced cardiac cells. Phys Rev E Stat Nonlin Soft Matter Phys. 2005; 71: 021903.[Medline] [Order article via Infotrieve]
8. Cao JM, Qu Z, Kim YH, Wu TJ, Garfinkel A, Weiss JN, Karagueuzian HS, Chen PS. Spatiotemporal heterogeneity in the induction of ventricular fibrillation by rapid pacing: importance of cardiac restitution properties. Circ Res. 1999; 84: 1318–1331.
9. Fox JJ, Riccio ML, Hua F, Bodenschatz E, Gilmour RF Jr. Spatiotemporal transition to conduction block in canine ventricle. Circ Res. 2002; 90: 289–296.
10. Pak HN, Hong SJ, Hwang GS, Lee HS, Park SW, Ahn JC, Moo Ro Y, Kim YH. Spatial dispersion of action potential duration restitution kinetics is associated with induction of ventricular tachycardia/fibrillation in humans. J Cardiovasc Electrophysiol. 2004; 15: 1357–1363.[CrossRef][Medline] [Order article via Infotrieve]
11. Qu Z, Garfinkel A, Chen PS, Weiss JN. Mechanisms of discordant alternans and induction of reentry in simulated cardiac tissue. Circulation. 2000; 102: 1664–1670.
12. Spach MS, Dolber PC. Relating extracellular potentials and their derivatives to anisotropic propagation at a microscopic level in human cardiac muscle. Evidence for electrical uncoupling of side-to-side fiber connections with increasing age. Circ Res. 1986; 58: 356–371.
13. Wu TJ, Ong JJ, Hwang C, Lee JJ, Fishbein MC, Czer L, Trento A, Blanche C, Kass RM, Mandel WJ, Karagueuzian HS, Chen PS. Characteristics of wave fronts during ventricular fibrillation in human hearts with dilated cardiomyopathy: role of increased fibrosis in the generation of reentry. J Am Coll Cardiol. 1998; 32: 187–196.
14. de Bakker JM, van Capelle FJ, Janse MJ, Tasseron S, Vermeulen JT, de Jonge N, Lahpor JR. Slow conduction in the infarcted human heart. "Zigzag" course of activation. Circulation. 1993; 88: 915–926.
15. Koura T, Hara M, Takeuchi S, Ota K, Okada Y, Miyoshi S, Watanabe A, Shiraiwa K, Mitamura H, Kodama I, Ogawa S. Anisotropic conduction properties in canine atria analyzed by high-resolution optical mapping: preferential direction of conduction block changes from longitudinal to transverse with increasing age. Circulation. 2002; 105: 2092–2098.
16. de Bakker JM, Coronel R, Tasseron S, Wilde AA, Opthof T, Janse MJ, van Capelle FJ, Becker AE, Jambroes G. Ventricular tachycardia in the infarcted, Langendorff-perfused human heart: role of the arrangement of surviving cardiac fibers. J Am Coll Cardiol. 1990; 15: 1594–1607.[Abstract]
17. Rohr S, Kleber AG, Kucera JP. Optical recording of impulse propagation in designer cultures. Cardiac tissue architectures inducing ultra-slow conduction. Trends Cardiovasc Med. 1999; 9: 173–179.[CrossRef][Medline] [Order article via Infotrieve]
18. Fast VG, Kleber AG. Cardiac tissue geometry as a determinant of unidirectional conduction block: assessment of microscopic excitation spread by optical mapping in patterned cell cultures and in a computer model. Cardiovasc Res. 1995; 29: 697–707.[CrossRef][Medline] [Order article via Infotrieve]
19. Rohr S, Salzberg BM. Characterization of impulse propagation at the microscopic level across geometrically defined expansions of excitable tissue: multiple site optical recording of transmembrane voltage (MSORTV) in patterned growth heart cell cultures. J Gen Physiol. 1994; 104: 287–309.
20. Kucera JP, Kleber AG, Rohr S. Slow conduction in cardiac tissue, II: effects of branching tissue geometry. Circ Res. 1998; 83: 795–805.
21. Bursac N, Papadaki M, Cohen RJ, Schoen FJ, Eisenberg SR, Carrier R, Vunjak-Novakovic G, Freed LE. Cardiac muscle tissue engineering: toward an in vitro model for electrophysiological studies. Am J Physiol. 1999; 277: H433–H444.[Medline] [Order article via Infotrieve]
22. Papadaki M, Bursac N, Langer R, Merok J, Vunjak-Novakovic G, Freed LE. Tissue engineering of functional cardiac muscle: molecular, structural, and electrophysiological studies. Am J Physiol Heart Circ Physiol. 2001; 280: H168–H178.
23. Tan JL, Liu W, Nelson CM, Raghavan S, Chen CS. Simple approach to micropattern cells on common culture substrates by tuning substrate wettability. Tissue Eng. 2004; 10: 865–872.[CrossRef][Medline] [Order article via Infotrieve]
24. Iravanian S, Nabutovsky Y, Kong CR, Saha S, Bursac N, Tung L. Functional reentry in cultured monolayers of neonatal rat cardiac cells. Am J Physiol Heart Circ Physiol. 2003; 285: H449–H456.
25. Entcheva E, Lu SN, Troppman RH, Sharma V, Tung L. Contact fluorescence imaging of reentry in monolayers of cultured neonatal rat ventricular myocytes. J Cardiovasc Electrophysiol. 2000; 11: 665–676.[Medline] [Order article via Infotrieve]
26. Entcheva E, Kostov Y, Tchernev E, Tung L. Fluorescence imaging of electrical activity in cardiac cells using an all-solid-state system. IEEE Trans Biomed Eng. 2004; 51: 333–341.[CrossRef][Medline] [Order article via Infotrieve]
27. Ursell PC, Gardner PI, Albala A, Fenoglio JJ Jr, Wit AL. Structural and electrophysiological changes in the epicardial border zone of canine myocardial infarcts during infarct healing. Circ Res. 1985; 56: 436–451.
28. Tsai CF, Tai CT, Hsieh MH, Lin WS, Yu WC, Ueng KC, Ding YA, Chang MS, Chen SA. Initiation of atrial fibrillation by ectopic beats originating from the superior vena cava: electrophysiological characteristics and results of radiofrequency ablation. Circulation. 2000; 102: 67–74.
29. Spach MS, Josephson ME. Initiating reentry: the role of nonuniform anisotropy in small circuits. J Cardiovasc Electrophysiol. 1994; 5: 182–209.[Medline] [Order article via Infotrieve]
30. Samie FH, Berenfeld O, Anumonwo J, Mironov SF, Udassi S, Beaumont J, Taffet S, Pertsov AM, Jalife J. Rectification of the background potassium current: a determinant of rotor dynamics in ventricular fibrillation. Circ Res. 2001; 89: 1216–1223.
31. Pastore JM, Girouard SD, Laurita KR, Akar FG, Rosenbaum DS. Mechanism linking T-wave alternans to the genesis of cardiac fibrillation. Circulation. 1999; 99: 1385–1394.
32. Banville I, Gray RA. Effect of action potential duration and conduction velocity restitution and their spatial dispersion on alternans and the stability of arrhythmias. J Cardiovasc Electrophysiol. 2002; 13: 1141–1149.[CrossRef][Medline] [Order article via Infotrieve]
33. 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.
34. Guerrero PA, Schuessler RB, Davis LM, Beyer EC, Johnson CM, Yamada KA, Saffitz JE. Slow ventricular conduction in mice heterozygous for a connexin43 null mutation. J Clin Invest. 1997; 99: 1991–1998.[Medline] [Order article via Infotrieve]
35. Spach MS, Heidlage JF, Barr RC, Dolber PC. Cell size and communication: role in structural and electrical development and remodeling of the heart. Heart Rhythm. 2004; 1: 500–515.[CrossRef][Medline] [Order article via Infotrieve]
36. Dillon SM, 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.
37. Fast VG, Kleber AG. Anisotropic conduction in monolayers of neonatal rat heart cells cultured on collagen substrate. Circ Res. 1994; 75: 591–595.
38. Bursac N, Parker KK, Iravanian S, Tung L. Cardiomyocyte cultures with controlled macroscopic anisotropy: a model for functional electrophysiological studies of cardiac muscle. Circ Res. 2002; 91: e45–e54.[CrossRef][Medline] [Order article via Infotrieve]
39. Lammers WJ, Schalij MJ, Kirchhof CJ, Allessie MA. Quantification of spatial inhomogeneity in conduction and initiation of reentrant atrial arrhythmias. Am J Physiol. 1990; 259: H1254–H1263.[Medline] [Order article via Infotrieve]
40. Pastore JM, Rosenbaum DS. Role of structural barriers in the mechanism of alternans-induced reentry. Circ Res. 2000; 87: 1157–1163.
41. Bursac N, Tung L. Acceleration of functional reentry by rapid pacing in anisotropic cardiac monolayers: formation of multi-wave functional reentries. Cardiovasc Res. 2006; 69: 381–390.
42. Lesh MD, Pring M, Spear JF. Cellular uncoupling can unmask dispersion of action potential duration in ventricular myocardium. A computer modeling study. Circ Res. 1989; 65: 1426–1440.
43. Rudy Y. Reentry: insights from theoretical simulations in a fixed pathway. J Cardiovasc Electrophysiol. 1995; 6: 294–312.[Medline] [Order article via Infotrieve]
44. Boersma L, Brugada J, Kirchhof C, Allessie M. Mapping of reset of anatomic and functional reentry in anisotropic rabbit ventricular myocardium. Circulation. 1994; 89: 852–862.
45. Girouard SD, Pastore JM, Laurita KR, Gregory KW, Rosenbaum DS. Optical mapping in a new guinea pig model of ventricular tachycardia reveals mechanisms for multiple wavelengths in a single reentrant circuit. Circulation. 1996; 93: 603–613.
46. Krinsky VI, Efimov IR, Jalife J. Vortices with linear cores in excitable media. Proc R Soc Lond A. 1992; 437: 645–655.
47. Biktashev VN, Holden AV. Re-entrant activity and its control in a model of mammalian ventricular tissue. Proc R Soc Lond B. 1996; 263: 1373–1382.[CrossRef][Medline] [Order article via Infotrieve]
48. Efimov IR, Krinsky VI, Jalife J. Dynamics of rotating vortices in the Beeler-Reuter model of cardiac tissue. Chaos Solitons Fractals. 1995; 5: 513–526.[CrossRef]
49. Peters NS, Wit AL. Myocardial architecture and ventricular arrhythmogenesis. Circulation. 1998; 97: 1746–1754.
This article has been cited by other articles:
 |
|
 |
|
  T. Nakagami, H. Tanaka, P. Dai, S.-F. Lin, T. Tanabe, H. Mani, K. Fujiwara, H. Matsubara, and T. Takamatsu Generation of reentrant arrhythmias by dominant-negative inhibition of connexin43 in rat cultured myocyte monolayers Cardiovasc Res, July 1, 2008; 79(1): 70 - 79. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||