Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Heart failure wikipedia , lookup
Remote ischemic conditioning wikipedia , lookup
Cardiac contractility modulation wikipedia , lookup
Coronary artery disease wikipedia , lookup
Mitral insufficiency wikipedia , lookup
Management of acute coronary syndrome wikipedia , lookup
Electrocardiography wikipedia , lookup
Hypertrophic cardiomyopathy wikipedia , lookup
Quantium Medical Cardiac Output wikipedia , lookup
Heart arrhythmia wikipedia , lookup
Ventricular fibrillation wikipedia , lookup
Arrhythmogenic right ventricular dysplasia wikipedia , lookup
864 Reduced Content of Connexin43 Gap Junctions in Ventricular Myocardium From Hypertrophied and Ischemic Human Hearts N.S. Peters, MRCP; C.R. Green, PhD; P.A. Poole-Wilson, MD, FRCP; N.J. Severs, PhD Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 Background.Gap junctions are a determinant of myocardial conduction. Disturbances of gap-junctional content may account for abnormalities of impulse propagation, contributing to the arrhythmic tendency and mechanical inefficiency of ischemic and hypertrophied myocardium. The aim of this study was to characterize gap junction organization in normal human ventricular myocardium and to establish whether abnormalities exist in myocardium of chronically ischemic and hypertrophied hearts. Methods and Results. Cardiac gap-junctional connexin43 antibodies and confocal microscopy were used in a quantitative immunohistochemical study of surgical myocardial samples to explore the structural basis of electromechanical ventricular dysfunction in chronic ischemic and hypertrophic heart diseases. Normal adult human left ventricular myocardium had a gap-junctional surface area of 0.0051 pm2/pm3 myocyte volume; gap junctions were confined to intercalated disks, of which there was a mean of 11.6 per cell. The right ventricle showed similar gap junction surface area. Left ventricular myocardium from ischemic hearts (distant from any fibrotic scarring), despite normal numbers of intercalated disks per cell, had a reduced gap junction surface area (0.0027 Ftm2/'lm3; P=.02), as did hypertrophied myocardium (0.0031 tm2/pm3; P=.05). The cardiac myocytes in the pathological tissues were larger than normal, and estimated gap-junctional content per cell was reduced in ischemic ventricle (P=.02) compared with normal. Conclusions. Gap junctions in normal adult human working ventricular myocardium occupy an area of 0.0051 am2/,'lm3 myocyte volume. This surface area is reduced in ventricular myocardium from hearts subject to chronic hypertrophy and ischemia, despite a normal number of intercellular abutments, and this alteration may contribute to abnormal impulse propagation in these hearts. (Circulation. 1993;88:864-875.) KEY WORDS * gap junctions * structure * G ap-junctional organization is an important determinant of intercellular conductance and the conduction properties of myocardium.1-4 The normal pattern of anisotropic conduction in ventricular myocardium, by which conduction parallel to the myocyte long axis is up to four times more rapid than that transverse to it,5,6 is dependent in part on the low resistivity of the gap-junctional membranes, their distribution, and their abundance. The underlying mechanism of many clinical cardiac arrhythmias is a myocardial reentrant electrical circuit,7 perpetuation of which requires areas of slowed conduction within the circuit, so that the advancing wave front encounters excitable tissue78 and initiation of which is enhanced by a local dispersion of recovery of excitability of the constituent myocytes.9-11 These factors are thought to provide the substrate for clinical reentrant arrhythmias.12 Efforts in basic cardiac electrophysiology have focused on how alterations of active membrane ionic properties can lead to slowing of electrical propReceived April 23, 1992; revision accepted May 12, 1993. From the National Heart and Lung Institute (N.S.P., P.A.P.-W., N.J.S.) and University College (C.R.G.), London, England. Correspondence to Dr N.S. Peters, Department of Cardiac Medicine, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY, England. hypertrophy * ischemia * microscopy agation and affect recovery of excitability, but this approach has not provided a full explanation of arrhythmogenesis. Indeed, there are reports of normal membrane potential characteristics in myocytes from myocardium with markedly abnormal electrophysiological properties and manifest arrhythmias.13-15 There has been increasing recognition that changes in the passive resistivity encountered by a propagating impulse can reduce conduction velocity and increase heterogeneity of conduction, irrespective of changes in active membrane properties.16-19 Despite the recognition of this key influence on the behavior of the propagating impulse, little information is available on the quantity and distribution or the function of gap junctions in human myocardium in health or disease. In this study, immunohistochemical detection of the principal cardiac gap-junctional protein, connexin43, by confocal laser scanning microscopy and electron microscopic techniques were used to examine the structural organization of gap junctions in normal human ventricular myocardium and to establish whether changes in gap junction content and organization occur in diseased ventricular tissue in which there are recognized alterations of myocyte form and electromechanical interaction. The results provide new insights into a possible Peters et al Gap Junction Quantification in Human Myocardium anatomic factor in the pathogenesis of electromechanical dysfunction and arrhythmias. Methods Myocardial Samples Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 Ventricular specimens were obtained from patients fulfilling the study criteria (defined below), selected at random from those undergoing cardiac surgery having given prior written consent, or from transplant donor hearts. The biopsies were taken from either the left ventricular apex or the mid right ventricular free wall as appropriate in each case, either by excision of a 3-mm ellipse of epimyocardium or by use of a transmural needle bioptome (diameter, 1.5 mm) from which the epimyocardial end was used for study. All biopsies obtained at routine surgery were taken immediately after the heart was arrested on cardiopulmonary bypass. Normal ventricular myocardium. Completely normal human myocardium for immediate fixation is not available for experimentation. For the purposes of this study, "normal" myocardium was obtained from the following sources: 1. Cardiac transplant donor hearts from which a single right ventricular biopsy was obtained on arrival at the implanting hospital after the heart was transported in cardioplegic solution on ice (for up to 4 hours). Light and electron microscopy have established that no detectable disruption of the intercalated disk structure occurs under these conditions of cardioplegia, a finding consistent with separate studies on myocardial preservation and cardioplegia.20 Samples of both ventricles of normal donor hearts that for technical reasons were not used for transplantation as intended were also studied. 2. The left ventricular apical region from patients with the Wolff-Parkinson-White syndrome (WPW). This group of patients experienced occasional but symptomatic tachyarrhythmic episodes but had no other detected cardiac dysfunction on routine preoperative assessment and were undergoing surgical division of the accessory pathway. Ventricular myocardium from hearts such as these, with infrequent arrhythmic episodes, has not been shown to have any structural abnormality and is generally considered normal.21 By these criteria, five "normal" left ventricular samples (two transplant donor hearts and three WPW hearts) and five right ventricular samples (transplant donors) were studied. Myocardium from ischemic ventricle. Left ventricular myocardium was obtained from five patients with threevessel coronary artery disease undergoing coronary artery bypass graft surgery for symptomatic relief from angina. All patients had had documented anterior myocardial infarction (with or without Q waves on ECG) at least 3 months previously and were normotensive. Preoperative `1Tl perfusion scintigraphy demonstrated symptomatic exercise-induced reversible ischemia in the remainder of the anteroapical territory, from which the biopsies of viable epimyocardium were obtained in each case. Myocardium immediately abutting scar from past infarction, shown in an earlier study22 to have a highly disrupted gap junction distribution (to a mean distance of 123 1um from the scar tissue border), was not included for analysis in the present study. For the same reason, subendocardial myocardium, which may contain wide- 865 spread small scars in patients with chronic coronary disease,23 both disrupting gap-junctional distribution and distorting tissue architecture, was also avoided. This strategy allowed sampling of viable myocardium associated with a region of recurrent reversible ischemia but distant from detectable histological abnormalities to establish whether alterations of gap junction organization, in the absence of more gross histological disruption, exist in these regions. A detectable reversible thallium perfusion defect with associated angina pectoris is suggestive of ischemia affecting the full myocardial thickness,24-28 but direct demonstration that the epimyocardial samples taken had themselves been subjected to ischemia was not feasible. Hypertrophied left ventricle. Left ventricular apical biopsies were taken during replacement of stenosed aortic valves in five patients with ECG and echocardiographic evidence of left ventricular hypertrophy. Processing of Specimens All biopsies required immediate fixation on removal, and the specimen was therefore divided as required in the operating theater, keeping note of tissue orientation to allow matching of areas analyzed by the different microscopic techniques described below. Immunohistochemistry. Tissue was put in Zamboni's fixative29 (2% paraformaldehyde, 0.2% picric acid, 0.1 mol/L phosphate-buffered saline [PBS], pH 7.4) for 2 to 6 hours. After fixation, all samples were washed in tap water, dehydrated in alcohol, placed in chloroform, and embedded in wax according to standard histological procedures. The primary antiserum used to localize cardiac gapjunctional protein was raised in rabbits against a synthetic peptide matching residues 131 through 142 of the cytoplasmically exposed segment of the cardiac gapjunctional protein connexin43.30,31 The peptide, supplied by Dr N.B. Gilula (Research Institute of the Scripps Clinic, La Jolla, Calif) was prepared by the simultaneous multiple peptide synthesis procedure of Houghten32 and was coupled by use of glutaraldehyde to the carrier protein, keyhole limpet hemocyanin, for subcutaneous injection into Sandy half-lop rabbits. Full details of the production and characterization of the polyclonal antiserum are published elsewhere.30,31,33'34 Sections (10 ,um) of wax-embedded tissue were dewaxed, rehydrated, and incubated in a trypsin solution (containing 0.1% trypsin [Sigma T-8128], 0.1% CaCl2, 20 mmol/L Trizma base, pH 7.4) for 10 minutes at room temperature to reexpose antigenic sites masked by fixation.35 The sections were washed and treated with 0.1 mol/L L-lysine (as blocking agent) in PBS containing 0.1% Triton X-100. Incubation with the primary antiserum (dilution, 1:10 in PBS) was carried out for 1 hour at 37°C. After washing, secondary antibody treatment with swine anti-rabbit fluorescein isothiocyanate (Dako; 1:20 dilution) was given for 1 hour in the dark at room temperature. After final washing in PBS, the slides were mounted with Citifluor mounting medium (Agar Scientific Ltd, Stansted, England). Under the conditions described, the immunolabeling procedure produces even and consistent staining through the depth of the sections. Immunostained sections were examined by conventional epifluorescence and confocal laser scanning mi- 866 Circulation Vol 88, No 3 September 1993 croscopy. Phase contrast microscopy was used to obtain details of tissue structure, and adjacent sections stained with hematoxylin and eosin were routinely examined by standard light microscopy. Control specimens in which the antiserum was substituted by preimmune rabbit serum or buffered saline or the second antibody by buffered saline were routinely run in parallel. The specificity of the antiserum for the peptide, for isolated cardiac gap-junctional protein, and for ultrastructurally defined gap junctions was demonstrated as part of the characterization of the antiserum.31,33 For confocal microscopy, immunolabeled sections were examined with a Bio-Rad Lasersharp MRC-500 running on standard Bio-Rad software. Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 Quantitative Analysis For quantitative assessment of myocardial gap junctions in the normal and diseased groups, all aspects of tissue processing, labeling, and image acquisition and analysis were strictly standardized as described below. Connexin43 gap junction sizes. Measurements of the longest dimension of the fluorescent spots representing gap junctions were made by use of the "length" composite command from the digital images of en face intercalated disks, which are viewed as characteristic clusters of gap junctions in transversely sectioned tissue. To confirm that the measurement of the fluorescent spots by this technique provided a reliable guide to the true size of gap junctions, despite undulating topology and different planes of orientation at the ultrastructural level,36,37 we have compared fluorescent spot sizes with gap junctions measured on freeze-fracture electron micrographs of the same tissue and demonstrated a close correspondence.33'34 Myocardial gap junction surface area. To determine the total quantity of connexin43 gap junction present in a volume of myocardium, the tissue section was cut transverse to the long axis of the constituent myocytes. The confocal microscope was configured with a small aperture (aperture control withdrawn 1.2 mm) so that the confocality of 10 optical slices of data taken at 1-pum intervals through the depth of the tissue permitted minimal overlap between consecutive images, ensuring acquisition of all fluorescence present (Fig 1). These series of images were collected using the x60 objective lens and the zoom 1 computer setting (field, 180x 120 1=m2) with a neutral density filter to eliminate significant fading of fluorescence. The "black level" was constant (at 4.0) such that the outlines of the individual crosssectioned cells were visible, and the "gain" control was adjusted so that the spectrum of label intensities spanned the full 255-level gray scale. The image data from the volume of tissue of 180x 120x 10 ,um3 were then analyzed with the standard Bio-Rad software and the more versatile PC IMAGE (Foster-Findlay Assoc, Newcastle-upon-Tyne, England) image analysis software. First, the number of crosssectioned myocytes in the fifth optical slice was counted, and the total area of the field occupied by myocytes was determined by subtracting from the total sample field all regions that were either completely devoid of signal (interstitial fibrosis) or recognizable as blood vessels. A composite image (or projection) was constructed for each 10-image series by superimposition (Fig 1), and a binary overlay was created automatically, in which each Lam Sure 1pm Inarvas Optical Section Series 1Sqim FIG 1. Diagrammatic representation of the operation of the laser scanning confocal microscope. The image of the stimulated fluorescence is acquired through a very small aperture, resulting in a narrow depth of field, in this case 1 ,m. By moving the specimen in the z axis (up and down in the diagram), one can acquire a series of optical sections at set intervals through the depth of the specimen. Superimposition of the digitized optical section series produces an in-focus single image projection of the information in the entire sampled volume (in this case, 180x120xl0 ,um3). An example of an optical section series and projection is shown in Fig 4. pixel was either on or off, as determined by a set threshold of 60 on the 255 -point gray scale, to eliminate the background cell outlines. The binary image was then edited by hand to reject all extraneous lipofuscin and blood vessel autofluorescence, which were easily identifiable, and an automatic count was made of the total number of pixels in the on state. This area represented the total area of labeled gap junction in the sampled volume of myocytes. Five randomly selected fields of each tissue sample were analyzed. To minimize errors introduced by variations between immunohistochemical labeling runs, samples of each of the disease groups and a control were included in every run. The resultant mean binary image area of three sampling sites from this positive control tissue was used to ensure consistency of the technique between labeling runs. The analysis of the optical section series by this technique allowed determination of the amount of gap junction in a defined volume of myocardium expressed in the standard manner for stereological assessment of this sort, square micrometers per cubic micrometer.38 In an effort to relate this conventional measure of surface area of gap junctions per unit volume to the size of the myocytes composing the tissue from which they were obtained, a relative index of mean cell volume of Peters et al Gap Junction Quantification in Human Myocardium Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 each of the sampled fields was derived. The volume index was obtained from the ratio of the number of cross-sectioned cells to the number of complete disks present (as an index of cell length) multiplied by the average cross-sectional area of the myocytes in the sample volume. If myocytes were cylindrical in shape with an intercalated disk at each end, the product of cell length derived by this method and the cross-sectional area would provide an absolute measure of cell volume. In reality, myocytes are not simple cylinders; they possess multiple intercalated disks, all of which lie in the plane approximately transverse to the long axis of the cell. Provided that this basic structural arrangement is present in all groups, however, with equivalent numbers of intercalated disks per cell, this index of cell volume is suitable for comparative purposes. The number and mean cross-sectional area of transected myocytes and the number of complete or nearly complete (>75% circumference) collections of gap junctions representing intercalated disks were determined from the projection image. The validity of this derivation of an index of cell volume has been confirmed in an experimental animal system by comparison with precise values for cell volume determined in isolated myocytes obtained from the same myocardium (unpublished observations). Intercalated disks per myocyte: myocardial cell arrangement. To determine the number of intercellular abutments in normal ventricular myocardial architecture and to detect alterations in disease, counts were made of the number of cell-surface clusters of gap junction label, each representing an intercalated disk, in 100 randomly selected isolated left ventricular myocytes from the normal and ischemic hearts. Myocytes were isolated by the modified enzymatic tissue dissociation method,39 as described by Harding et al.40 The cells produced were immediately fixed and processed for connexin43 labeling by use of a procedure similar to that for whole tissue, and confocal microscopy was used to make disk counts. All focal clusters of label were considered to represent intercalated disks. Electron microscopy: gap junction ultrastructure and connexon packing density. Glutaraldehyde-fixed specimens from the normal and the ischemic hearts were prepared for thin-section electron microscopy and freeze-fracture electron microscopy according to standard procedures.41 Glycerinated samples for freezefracture were frozen in Freon, and replicas were preTABLE 1. Details of Normal Patients Heart 1 Type WPW Age Ventricle (years) sampled 37 L 2 Donor 46 L+R L 3 WPW 28 4 44 L+R Donor 5 62 L WPW Donor 26 R 6 7 37 R Donor 24 R Donor 8 WPW, patient with the Wolff-Parkinson-White syndrome; Donor, transplant donor; L, left ventricular sample; R, right ventricular sample. 867 TABLE 2. Details of Patients With Ischemic Heart Disease Age Heart (years) ECG 9 54 NoQ 10 47 Q 11 46 Q 12 48 Q 13 53 NoQ Q and No Q, Q waves present/absent in anterior leads on preoperative ECG. All patients had documented anterior myocardial infarction at least 3 months previously and exercise-induced ischemia in the anteroapical territory on thallium perfusion scan. pared in a BaIzers BAF400T apparatus. Replicas and sections were examined in a Philips EM 301 microscope. Freeze-fracture electron micrographs (50 pooled from ischemic and 22 from normal hearts) were used to measure the connexon surface density within individual gap junction plaques, since the precise packing arrangement of connexons is reported to alter under hypoxic conditions (reviewed in References 33 and 37). It was important for this study that any variation in connexon density (regardless of the precise arrangement of the packing) was accounted for in the interpretation of the measured spot sizes by the immunohistochemical technique. An electron micrograph with a final magnification of x 70 000 of each gap junction plaque was used to determine the junctional area, using VIDS in software (Analytical Measuring Systems, Cambridge, England). The connexon packing density was determined by counting the number of connexons within a hole in a mask placed randomly on the micrograph. Statistical analysis. All results are expressed as mean±SD. Group data are compared by two-sample t tests of mean values for each heart. Statistical significance is defined as a value of P<.05. Results Details of the 18 patients studied are summarized in Tables 1 through 3. The ages of the patients from whom normal left ventricular tissue was obtained were 43.4±12.6 years and those from whom right ventricular tissue was obtained, 35.4+10.1 years. The five patients with ischemic heart disease were 49.6±3.7 years old. The patients with aortic stenosis were 59.6±4.6 years old. General Ultrastructure The ventricular myocardium from the transplant donor and WPW subjects had normal ultrastructural apTABLE 3. Details of Patients With Aortic Valve Stenosis Valve gradient Age Heart (mm Hg) (years) 14 60 77 15 56 50 54 16 110 17 63 80 18 65 60 All patients had ECG and echocardiographic evidence of left ventricular hypertrophy. 868 Cfrcnlation Vol 88, No 3 September 1993 Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 p of left vrmyocanumfrom a patient with Wotff-Parw*lnonWhite syndrome FI 2. Thin-se n electronmc showing an electron-dense intexalaed disk (large arrow) traversing showing normalfeatures. a, Low-power electron mi between abutting myocyes in the characteristic step-ie manner. The striations of the sarcomeres are aligned in ths healthy tiue (small arrow). Scale bar=5 ,um. b, Higher-powereectron micrograph showinggap junctions ofnormal appearance (arrows). Note the characteristic long stretch of gap-junctional embrane at the ends of the disk, consistent with the peripheral ring of large junctions on immunolabeling (see Fig 3, c). Scale bar=1 ,um pearances, illustrated in Fig 2. The myocardium sampled from the ischemic hearts had an orderly arrangement of fibers, with no tissue distortion and no apparent fibrosis. Various degrees of hypertrophy were present, but thin-section electron microscopic appearances were otherwise normal. The myocardium from patients with aortic stenosis had the classic features of hypertrophy, with myocyte enlargement and hyperchromatic rectangular nuclei, the extent of which varied from cell to cell. Immunolabeled Connexin43 Gap Junction Distribution at Confocal Microscopy Immunolocalization of connein43 gave consistent patterns of gap junction staining, common to left and right ventricle of normal hearts and the left ventricular myocardium from hearts with aortic stenosis and ischemia. In all these groups, gap junctions were visualized with the normal adult human patterns previously described42: bright fluorescent domains marking the positions of intercalated disks between myocytes, appearing as transverse lines of individually resolved spots in longitudinally sectioned myofibers (Fig 3, a) and discoid arrays delineated peripherally by large gap junctions in transversely sectioned myocardium (Figs 3, c and 2, b). Controls in which antiserum treatment was omitted or substituted with preimmune serum showed no specific fluorescent labeling (Fig 4, g). Grapelike collec- tions of lipofuscin (Fig 4, g) with characteristic pink autofluorescence were present in normal and to a greater extent in diseased myocardium. Quantitative Characterization of Gap Junctions Analsis of connexon density in freeze-fractured gap junctions. Analysis of data from freeze-fractured gap junctions .(ig 5) shows that there is no correlation between gap junction size and the density of its component connexons (r= -.22, P=.13). Fig 6 shows the plot of these variables for gap junction plaques. of <1 jum2 surface area from the ischemic hearts. There was no difference in density distribution between myocardium from the normal and ischemic hearts. Measurement of total label present in the immunohistochemical images will therefore be proportional to the number of connexons and the quantity of connexin regardless of any variation in spot size distribution between different samples. Size of immunolabeled gap junctions. Measurement of the longest axis of the labeled spots of 25 intercalated disks selected at random from the five normal left ventricular specimens revealed a mean length of 0.52±0.30 pm, with a distribution as shown in Fig 7, A. The larger junctions at the periphery of the disk had a mean length of 0.67+0.32 -m (maximum, 2.8 gum); the junctions in the central area of the disk, 0.36±0.17 gim (minimum, 0.1 gum). The distribution of the labeled Peters et al Gap Junction Quantification in Human Myocardium 869 Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 FIG 3. Confocal immunolocalization of connexin43 gap junctions in left ventricular myocardium from a patient with Wolff-Parkinson-White syndrome, showing normal features. a, A single optical section of longitudinally sectioned myocardium revealing each individual gap junction domain as a separate spot (arrow) grouped within intercalated disks. Scale bar=50 gtm. b, A projected optical series from transversely sectioned myocardium showing the characteristic features of a disk viewed en face (arrow), with a peripheral ring of large junctions and smaller central domains. Scale bar=50 ,gm. c, High-power view of en face disks, showing the entire gap junction population of each disk. Scale bar=-0 gm. 870 Circulation Vol 88, No 3 September 1993 Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 FIG 4. A confocal optical section series of l-um intervals through the depth of transversely sectioned normal human myocardium labeled for connexin43 (compare with Fig 1). Scale bars=50 gm. a through e, The first, third, fifth, seventh, and ninth optical sections in an optical series acquired for the quantification procedure. There are progressive changes in the image content throughout the series. f Theprojection of the 10 images of the optical section series (a through e), showing some completely defined intercalated disks with their entire gap junction population displayed in focus. g, A wax section adjacent to that in a through f in which preimmune serum was substituted for the HJ antiserum in the processing procedure. This control section contains no fluorescent label, demonstrating the specificity of the procedure, and shows the lipofuscin autofluorescence, which can readily be distinguished from labeling (when present) for the purposes of editing the image. 8 '71 Peters et al Gap Junction Quantification in Human Myocardium A - from normal left ventricle 25 bC 20- 10 z5 0 Measured Spot Size (pm) Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 B - from ischemic left ventricle 25 b 20- -a 0 15 'a FIG 5. Electron micrograph offreeze-fracture replica showing the densely packed connexon particles (arrow) of an entire junctional plaque (gj) from a patient with ischemic heart disease. Scale bar=0.5 gum. junction long-axis measurements pooled from the ischemic left ventricles (mean, 0.54±0.42 gm) showed no difference from normal (Fig 7, B). Gap-junctional surface area per unit volume of tissue. An example of images from an optical section series of C-M 150E 0 125 - 100 - a- = L o . .-. o 75- X 0'50 .9 0.00 0.25 .50 .75 1.00 gap junction plaque area (pm2) FIG 6. Plot of connexon surface density vs gap junction plaque area in plaques up to 1 gm2 pooled from ventricular myocardium from the ischemic hearts. There is no correlation (P=.18), and the slope is not significant. 00 5 IN. 0 0 0 000 0 w v_- Pm " W _ . - M - m m N W W Measured Spot Size (pm) FIG 7. Frequency distributions of long-axis measurements of labeled spots in confocal images (A) from the normal left ventricles (mean, 0.52 gum) and (B) from the ischemic left ventricles (mean, 0.54 gum). There is no difference between these distributions. connexin43-labeled transversely sectioned ventricular myocardium prepared for quantitative analysis is shown in Fig 4. The results of the morphometric analysis of such images to determine myocardial gap junction content in the healthy and diseased hearts are shown in Table 4. Normal adult left ventricular myocardium had a gap junction surface area of 0.0051±0.0015 gm2/,um3 myocyte volume, and normal right ventricle showed no significant difference (0.0044±0.0006 ,um2Im3). Left ventricular myocardium from recurrently ischemic hearts had a significantly reduced gap junction surface area of 0.0027±0.0009 gm2/,um3 (P=.024). Hypertrophied left ventricle from patients with aortic stenosis also had a reduced gap junction surface area of 0.0031±0.0005 11m2/gm3 (P=.05) and was not significantly different from myocardium from the ischemic hearts. 872 Circulation Vol 88, No 3 September 1993 TABLE 4. Results of Quantitative Morphometric Analysis Gap junction surface area/volume myocardium Heart 1 2 3 4 S 2 4 6 7 8 Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 9 10 11 12 13 (gm2/Im3) Cell volume index 0.00476±0.00076 0.00751±0.00158 0.00419±0.00072 0.00360±0.00117 0.00540±0.00113 29 436±5377 15 347±859 23 516±5534 24 668±6998 20 925±1170 Mean LV 0.00508±0.00151 22 778±5174 RV RV RV RV RV 0.00501±0.00106 0.00349±0.00034 0.00464±0.00123 0.00402±0.00038 0.00473±0.00050 22 799±3512 22 582±3136 26 901±8047 22 577±3146 22 639±3424 Mean RV 0.00438±0.00061 23 499±1904 ILV ILV ILV ILV ILV 0.00230±0.00098 0.00199±0.00038 0.00193±0.00046 0.00351±0.00162 0.00388±0.00099 39 067± 1049 36 756±9360 29 346±6100 25 723±6528 24 443±3072 0.00272±0.00091* 31 067±6553 0.00352±0.00029 0.00265±0.00033 0.00333±0.00021 0.00253±0.00080 0.00359±0.00118 29 070±3853 32 804±4312 34 204±4156 35 104±1179 26 052±6612 0.00313±0.00050* 31 447±3794* Tissue type LV LV LV LV LV Mean ILV 14 15 16 17 18 HLV HLV HLV HLV HLV Mean HLV Mean, mean ± SD of means of five sample fields for each heart. Values are mean±SD. LV, normal left ventricle; RV, normal right ventricle; ILV, ischemic left ventricle; HLV, hypertrophied left ventricle. *Significantly different from value for normal left ventricle (P<.05). Comparative index of myocyte volume. The cell volume index derived from the hearts with aortic stenosis demonstrated the increased ventricular myocyte volume that would be expected in hypertrophied myocardium (31 477± 3794 units) compared with normal (22 778+ 5174 units; P=.019). The mean value for myocardium from the ischemic ventricles (31 067±6553 units), although increased, did not differ significantly from normal (P=.062), owing to the presence of a wider range of cell sizes than the aortic stenosis group. Since the cell volume indexes were derived from the same tissue fields as the gap-junctional surface density values, these data allow approximation of an index of gap junction area per myocyte for each myocardial specimen, derived from the mean of the gap junction areas and the cell volume indexes of each of the five sample volumes. This value is reduced in myocardium from the ischemic hearts (10 758±2133 units) compared with normal (14 996±2268 units; P=.019). Hypertrophied myocardium, by comparison, has a value of 12 848±1533 units (P=.11). These results and their interpretation must be treated with caution because the ratio used in their derivation may artificially exaggerate differences. Intercalated disk counts. A confocal image of a labeled isolated ventricular myocyte is shown in Fig 8. When such images were used to count the number of surface clusters of connexin43 label (each representing an intercalated disk) per myocyte, there was no difference between cells from myocardium of normal (11.55±2.20) and ischemic hearts (11.87+2.49; P=.33). Discussion Quantitative studies of gap junctions in mammalian myocardium, most of which used thin-section electron microscopy, have been reviewed by Page.4 These studies are confined to species other than humans, in whom characterization of myocardial gap junctions and alterations that may accompany human myocardial disease have therefore not been determined. We have published findings at the borders of healed human infarcts22 Peters et al Gap Junction Quantification in Human Myocardium Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 FIG 8. Confocal microscopic projection image of an optical section series through a single isolated ventricular myocyte labeled for connexin43, showing the positions of the intercalated disks as clusters of transversely oriented labeled gap junctions (examples arrowed). Scale bars=20 ,um. but have not previously attempted quantification of gap junction content. The group of ischemic patients tended to be older than the group with normal left ventricles (49.6±3.7 vs 43.4±12.6 years), a difference that is difficult to overcome given the reasons for which these groups undergo surgery. However, this difference was not statistically significant (P<.35), and there were no trends to suggest age-related changes in any parameters within any of the study groups. The gap-junctional surface area in normal human left ventricular myocardium (0.0051 ,um2/pum3 myocyte volume), as determined in the present study, falls within the range of values reported from electron microscopic ultrastructural analyses of myocardium of other mammalian species. These include rabbit ventricle (0.017 I£m2/j&m3),43 rat ventricle (0.0047 and 0.0022 to 0.0154 gm2/1,m3),44,4 and canine ventricle (0.0085 11m2/4m3),46 from which Luke et a138 have also estimated a value of 0.0052 Mm2/1m3 by an immunohistochemical technique using light microscopy. In contrast with all these reports, the technique used in the present study enables the quantity of gap junction to be related directly to entire and completely visualized intercalated disks sampled through relatively large volumes of tissue. The quantitative spatial distribution data obtained facilitate meaningful comparisons between disease groups and could, in principle, allow direct correlation with myocardial function and electrophysiological data acquired before myocardial fixation. The presence of fluorescent labeling must be interpreted with some caution, since the antibody is directed to a small segment of the connexin43 molecule and labeling may not necessarily represent intact or functional gap junctions. However, the agreement of our data with the values obtained by standard electron microscopic morphometry,43-46 the correlation of the labeling patterns with gap junction distribution deter- 873 mined by ultrastructural techniques,22 and the confirmation by immunogold labeling at the electron microscopic level of specific binding of the antibodies to gap junctions,33 all serve to support the validity of the approach. Our finding of a mean of 11.55 +2.2 intercalated disks per myocyte in normal human epimyocardium is consistent with data reported by Luke and Saffitz,47 who used a quantitative electron microscopic technique to determine a mean of 11.2 intercalated disk contacts per myocyte in normal canine ventricular myocardium. These authors showed a reduced number (6.5) of such intercellular contacts in canine myocardium associated with infarction but had specifically selected regions associated with fibrotic scarring. In the present study, however, we obtained normal intercalated disk counts in ischemic hearts, despite changes in cell size distribution, suggesting that the basic architecture of intercellular abutments was not significantly altered in regions free of scar. Substrate for Abnormal Impulse Propagation It has long been suggested that fibrosis plays a part in the anatomy of arrhythmogenesis,48 but the results of the present study suggest a mechanism by which intercellular coupling may be impaired in the ischemic heart without invoking fibrosis as the explanation. Both ischemic heart disease and left ventricular hypertrophy are associated with ventricular arrhythmias and sudden cardiac death.49,50 The development of treatment strategies for cardiac arrhythmias that occur in these and other clinical settings requires a detailed and complete understanding of the mechanisms by which these arrhythmias are generated. One of the many adverse effects of cardiac ischemia is a decrease in conduction velocity,12'51 a factor increasing the likelihood of arrhythmias. Although gap junctions are established as the organelle determining intercellular and whole-tissue conductance and conduction velocity4J12'51 and an absence of gap junctions results in electrical isolation,52 the relation between the relative abundance of gap junctions, intercellular conductance, and conduction velocity is poorly understood. Evidence suggests that regions of the heart with slow conduction and relatively low intercellular conductance (such as the atrioventricular node) contain fewer gap junctions than regions of more rapid conduction.5' The possibility exists, therefore, that a reduction in gap junction surface area per unit volume of about 40% in ventricular myocytes from ischemic and hypertrophied ventricles, as demonstrated in the present study, may significantly alter patterns and rates of impulse propagation. The distribution of labeled gap junction long-axis lengths and the number of intercalated disks per cell remained unchanged in the myocardium from the ischemic hearts compared with normal, suggesting that the reduction in gap-junctional content occurs as a result of reduction in the numbers of all sizes of junction. This is of interest because electron microscopic examination of the canine infarct border region has suggested a preferential reduction of larger junctions, but fibrotic scarring of the canine myocardium may be relevant to this difference.47 The greater myocyte volume that characterizes hypertrophied myocardium associated with chronic aortic stenosis was evident in the present study. Compensatory 874 Circulation Vol 88, No 3 September 1993 hypertrophy that might be expected in the myocytes of the myocardium around a healed infarct23,53 did not occur to a statistically significant degree, despite a greater mean cell volume, owing to the large standard deviation. Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 Functional Significance: Correlating Morphology With Electrophysiology Although electrophysiological studies of arrhythmogenic cardiac disease have documented significant abnormalities of active membrane properties,8 many have failed to do So.13'14 Boyden et al14 concluded that alteration of unspecified morphologic features of the myocardium was responsible for the electrophysiological disturbances in chronically hypertrophied canine atria. Kieval et allS demonstrated normal action potential characteristics but abnormal gap-junctional intercellular conductance in postischemic myocyte pairs, and reduced junctional conductance in cultured myocytes infected with Trypanosoma cruzi is associated with a qualitative reduction in connexin43 expression.54 A reduction in gap-junctional coupling between myocytes may, therefore, be an important morphological feature that could interact with altered active membrane properties in diseased myocardium of a variety of pathogeneses. It is likely that arrhythmogenic states result from a combination of changes of both membrane properties and coupling characteristics of the myocardium, and it has been suggested that abnormal cellular coupling may be the predominant electrophysiological derangement at the border of a healed myocardial infarct.55 It has not been possible to study the interaction between the action potential and myocardial resistivity during programmed stimulation,55-58 but a number of computer models simulating this interaction have been developed. Joyner59 showed that as coupling resistance increases, the "intrinsic" cellular differences of action potential duration become increasingly manifest, and Lesh et al55 showed a coincident increase in axial resistivity, leading to slowed conduction. The combination of dispersion of action potential duration and slow conduction promotes reentrant tachycardia initiation and perpetuation,12 and Quan and Rudy's60 model suggests that cellular uncoupling is arrhythmogenic by just these mechanisms. Implications of Results A reduction in gap junction coupling may, therefore, change conduction velocity and unmask intrinsic differences in action potential characteristics, manifest at the clinical level by nonspecific changes in the QRS complex on the ECG, changes in the coordination of contraction and the mechanical efficiency of the heart, and a possible lowering of arrhythmic threshold. These phenomena are well recognized in ischemic and hypertrophied hearts. The process of uncoupling may, itself, induce changes in action potential characteristics55'61 and recovery of excitability.55 All these alterations increase the potential for spontaneous arrhythmias,62 for which there is growing evidence of the role of abnormal intercellular coupling.1"2'55'6' The results of the present study suggest that changes in resistivity, as evidenced by altered gap junction quantity in the diseased human ventricle, may indeed occur. The resistivity of myocardium, a potentially imporarrhythmias, has been ignored in many studies confined to relating active membrane properties to the behavior of the tissue and the heart as tant factor in clinical a whole. In the present study, normal human ventricular myocardium has been characterized with respect to gap junctions. Abnormalities of gap junction content, as demonstrated in ischemic and hypertrophied hearts, would be consistent with the clinically apparent disturbances of ventricular electromechanical function in these conditions, the explanation and treatment of which remain poorly understood. Little is known of the action of conventional antiarrhythmic agents on cellular coupling and anisotropy,63 and with a number of clinical trials questioning the conventional wisdom of pharmacological arrhythmia treatment,64 a broader approach to understanding electromechanical dysfunction may require greater attention to gap junction intercellular coupling. Acknowledgments This study was supported by the British Heart Foundation (grant F268). We are grateful to John Pepper, Graeme Bennett, and Sir Magdi Yacoub for providing the surgical specimens. The photographic expertise and assistance of Stephen Rothery is gratefully acknowledged. We thank Federica Del Monte for the myocyte isolation. References 1. Spach MS, Miller WT III, Dolber PC, Kootsey JM, Sommer JR, Mosher CE Jr. The functional role of structural complexities in the propagation through an electrical syncytium. Circ Res. 1982;50: 175-191. 2. Joyner RW. Effects of the discrete pattern of electrical coupling on propagation through an electrical syncytium. Circ Res. 1982;50: 192-200. 3. Gardner PI, Ursell PC, Fenoglio JJ Jr, Wit AL. Electrophysiologic and anatomic basis for fractionated electrograms recorded from healed myocardial infarcts. Circulation. 1985;72:596-611. 4. Page E. Cardiac gap junctions. In: Fozzard HA, Jennings RB, Haber E, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System: Scientific Foundations. New York, NY: Raven Press, Publishers; 1991:1003-1048. 5. 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. 6. Delmar M, Michaels DC, Johnson T, Jalife J. Effects of increasing intercellular resistance on transverse and longitudinal propagation in sheep epicardial muscle. Circ Res. 1987;60:780-785. 7. Mines GR. On circulating excitations in heart muscles and their possible relation to tachycardia and fibrillation. Trans R Soc Cardiol. 1914;46:349-383. 8. Cranfield PF, Dodge FA. Slow conduction in the heart. In: Zipes DP, Bailey JC, Elharrad I, eds. The Slow Inward Current and Cardiac Arrhythmias. Boston, Mass: Martinus Nijhoff; 1984: 149-171. 9. Allesi R, Nusinowitz M, Abildskov JA, Moe GK. Nonuniform distribution of vagal effects on the atrial refractory period. Am J Physiol. 1958;194:406-410. 10. Kuo CS, Munakata K, Reddy CP, Surawitz B. Characteristics and possible mechanisms of ventricular arrhythmias dependent on the dispersion of action potential duration. Circulation. 1983;67: 1356-1367. 11. Han J. The concepts of reentrant activity responsible for ectopic rhythms. Am J Cardiol. 1971;28:253-262. 12. Fozzard HA, Arnsdorf MF. Cardiac electrophysiology. In: Fozzard HA, Jennings RB, Haber E, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System: Scientific Foundations. New York, NY: Raven Press, Publishers; 1991:63-98. 13. Spach MS, Dolber PC. Relating extracellular potentials and their derivatives to anisotropic propagation at a microscopic level in human cardiac muscle. Circ Res. 1986;58:356-371. 14. Boyden PA, Tilley LP, Pham TD, Liu S-K, Fenoglio JJ, Wit AL. Effects of left atrial enlargement on atrial transmembrane Peters et al Gap Junction Quantification in Human Myocardium 15. 16. 17. 18. 19. 20. 21. 22. Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. potentials and structure in dogs with mitral valve fibrosis. Am J Cardiol 1982;49:1896-1907. Kieval RS, Spear JF, Moore EN. Gap junction conductance in ventricular myocyte pairs isolated from postischemic rabbit myocardium. Circ Res. 1992;71:127-136. Fozzard HA. Cardiac muscle: excitability and passive electrical properties. Prog Cardiovasc Dis. 1977;19:343-359. Arnsdorf MF. Cable properties and conduction of the action potential. In: Sperelakis N, ed. Physiology and Pathophysiology of the Heart. Boston, Mass: Martinus Nijhoff; 1984:109-140. Binah 0, Rosen MR. Mechanisms of ventricular arrhythmias. Circulation. 1992;85(suppl I):I-25-I-31. Saffitz JE, Hoyt RH, Luke RA, Kanter HL, Beyer EC. Cardiac myocyte interconnections at gap junctions. Trends Cardiovasc Med. 1992;2:56-60. Mankad PS, Severs NJ, Lachno DR, Rothery S, Yacoub MH. Superior qualities of University of Wisconsin solution for ex-vivo preservation of the pig heart. J Thorac Cardiovasc Surg. 1992;104: 229-240. Wit AL, Janse MJ. Basic mechanisms of arrhythmias. In: The Ventricular Arrhythmias of Ischemia and Infarction. Mt Kisco, NY: Futura Publishing Co; 1992:1-160. Smith JH, Green CR, Peters NS, Rothery S, Severs NJ. Altered patterns of gap junction distribution in ischemic heart disease. Am JPathol. 1991;139:801-821. Cotran RS, Kumar V, Robbins SL. The heart. In: Robbins Pathologic Basis ofDisease. 4th ed. Philadelphia, Pa: WB Saunders; 1989: 597-656. Wackers FJ, Sokole EB, Samson G, van de Schoot JB, Lie KI, Liem KL, Wellens HJJ. Value and limitations of thallium-201 scintigraphy in the acute phase of myocardial infarction. N Engl J Med. 1976;295:1-5. Mueller TM, Marcus ML, Ehrhardt JC, Chaudhuri T, Abboud FM. Limitations of thallium-201 myocardial perfusion scintigrams. Circulation. 1976;54:640-646. Heller GV, Ahmed I, Tilkemeier PL, Barbour MM, Garber CE. Comparison of chest pain, electrocardiographic changes and thallium-201 scintigraphy during varying exercise intensities in men with stable angina pectoris. Am J Cardiol. 1991;68:569-574. Sutton JM, Topol EJ. Significance of a negative exercise thallium test in the presence of a critical residual stenosis after thrombolysis for acute myocardial infarction. Circulation. 1991;83:1278-1286. Nishimura S, Mahmarian JJ, Boyce TM, Verani MS. Quantitative thallium-201 single-photon emission computed tomography during maximal pharmacologic coronary vasodilation with adenosine for assessing coronary artery disease. J Am Coil Cardiol. 1991;18: 736-745. Toshimori H, Toshimori K, Oura C, Matsuo H. Immunohistochemistry and immunocytochemistry of natriuretic peptide in porcine heart. Histochemistry. 1987;86:595-601. Gourdie RG, Harfst E, Severs NJ, Green CR. Cardiac gap junctions in rat ventricle: localization using site directed antibodies and laser scanning confocal microscopy. Cardioscience. 1990;1: 75-82. Harfst E, Severs NJ, Green CR. Cardiac myocyte gap junctions: evidence for a major connexon protein with an apparent relative molecular mass of 70 000. J Cell Sci. 1990; 96:591-604. Houghten RA. General method for the rapid solid-phase synthesis of large numbers of peptides: specificity of antigen-antibody interactions at the level of individual amino acids. Proc Natl Acad Sci USA. 1985;82:5131-5135. Severs NJ, Gourdie RG, Harfst E, Peters NS, Green CR. Intercellular junctions and the application of microscopical techniques: the cardiac gap junction as a case model. J Microsc. 1993;196: 299-328. Green CR, Peters NS, Gourdie RG, Rothery S, Severs NJ. Validation of immunohistochemical quantification in confocal scanning laser microscopy: a comparative assessment of gap junction size using confocal and ultrastructural techniques. J Histochem Cytochem. In press. Harlow E, Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1988. Forbes MS, Sperelakis N. Intercalated disks of mammalian heart: a review of structure and function. Tissue Cell. 1985;17:605-648. Severs NJ. The cardiac gap junction and intercalated disk. Int J Cardiol. 1990;26:137-173. Luke AL, Beyer EC, Hoyt RH, Saffitz JE. Quantitative analysis of intercellular connections by immunohistochemistry of the cardiac gap junction protein connexin43. Circ Res. 1989;65:1450-1457. 875 39. Farmer BB, Mancica M, Williams ES, Watanabe AM. Isolation of calcium tolerant myocytes from adult rat hearts: review of the literature and description of a method. Life Sci. 1983;33:1-18. 40. Harding SE, O'Gara P, Jones SM, Brown LA, Vescovo G, PooleWilson PA. Species dependence of contraction velocity in single isolated cardiac myocytes. Cardioscience. 1990;1:49-53. 41. Severs NJ, Slade AM, Powell T, Twist VW, Jones GE. Morphometric analysis of the isolated calcium-tolerant cardiac myocyte: organelle volumes, sarcomere length, plasma membrane surface folds and intramembrane particle density and distribution. Cell Tissue Res. 1985;240:159-168. 42. Gourdie RG, Green CR, Severs NJ. Gap junction distribution in adult mammalian myocardium revealed by an antipeptide antibody and laser scanning confocal microscopy. J Cell Sci. 1991;99:41-55. 43. Shibata Y, Nakata K, Page E. Ultrastructural changes during development of gap junctions in rabbit left ventricular myocardial cells. J Ultrastruct Mol Struct Res. 1980;71:258-271. 44. Page E. Quantitative ultrastructural analysis in cardiac membrane physiology. Am J Physiol. 1978;235:C147-C158. 45. Stewart JM, Page E. Improved stereological techniques for studying myocardial cell growth. J Ultrastruct Mol Struct Res. 1978; 65:119-134. 46. Hoyt RH, Cohen ML, Saffitz JE. The distribution and threedimensional structure of intercellular junctions in canine myocardium. Circ Res. 1989;64:563-574. 47. Luke RA, Saffitz JE. Remodeling of ventricular conduction pathways in healed canine infarct border zones. J Clin Invest. 1991;87:1594-1602. 48. Mehra R, Zeiler RH, Gough WB, El-Sharif N. Reentrant ventricular arrhythmias in the late myocardial infarction period: 9. Electrophysiologic-anatomic correlation of reentrant circuits. Circulation. 1983;67:11-24. 49. Fujimoto T, Peter R, Hamamoto H, Mandel WJ. Electrophysiological observations during the spontaneous initiation of ischemia induced ventricular fibrillation. Am Heart J. 1983;105:189-197. 50. Kannel B. Prevalence and natural history of electrocardiographic left ventricular hypertrophy. Am J Med. 1983;75(suppl):4-11. 51. Jongsma HJ, Gros D. The cardiac connection. News Physiol Sci. 1991;6:34-40. 52. Spray DC, Burt JM. Structure-activity relations of the cardiac gap junction channel. Am J Physiol. 1990;258:C195-C205. 53. Anversa P, Beghi C, Kikkawa Y, Olivetti G. Myocardial response to infarction in the rat: morphometric measurement of infarct size and myocyte cellular hypertrophy. Am J Pathol. 1985;118:484-492. 54. Campos de Carvalho AC, Tanowitz HB, Wittner M, Dermietzel R, Roy C, Hertzberg EL, Spray D. Gap junction distribution is altered between cardiac myocytes infected with Trypanosoma cruzi. Circ Res. 1992;70:733-742. 55. Lesh MD, Pring M, Spear JF. Cellular uncoupling can unmask dispersion of action potential duration in ventricular myocardium. Circ Res. 1989;65:1426-1440. 56. Tsuboi N, Kodama I, Toyama J, Yamada K. Anisotropic conduction properties of canine ventricular muscles: influence of high extracellular K' concentration and stimulation frequency. Jpn Circ J. 1985;49:487-498. 57. Delmar M, Delgado C, Chivalo D, Michaels DC, Jalife J. On the problem of anisotropic propagation in ventricular muscle. In: Rosen MR, Palti Y, eds. Lethal Arrhythmias Resulting From Myocardial Ischemia and Infarction. Boston, Mass: Kluwer Academic Publishers; 1989:181-197. 58. Spach MS, Dolber PC, Heidlage JF. Interactive roles of inhomogeneities of repolarization and anisotropic propagation in atrial reentry. Circulation. 1988;78(suppl II):II-413. Abstract. 59. Joyner RW. Modulation of repolarization by electronic interactions. Jpn Heart J. 1986;27:167-183. 60. Quan W, Rudy Y. Unidirectional block and reentry of cardiac excitation: a model study. Circ Res. 1990;66:367-382. 61. Jalife J, Sicouri S, Delmar M, Michaels DC. Electrical uncoupling and impulse propagation in isolated sheep Purkinje fibers. Am J Physiol. 1989;257:H179-H189. 62. Hoffman BF, Rosen MR. Cellular mechanisms for cardiac arrhythmias. Circ Res. 1982;49:69-83. 63. Kadish AH, Spear JF, Levine JH, Moore EN. The effects of procainamide on conduction in anisotropic canine ventricular myocardium. Circulation. 1986;74:616-625. 64. Akiyama T, Pawitan Y, Greenberg H, Chien-Suu K, ReynoldsHaertle RA, the CAST Investigators. Increased risk of cardiac death and cardiac arrest from encainide and flecainide in patients after non-Q-wave myocardial infarction in the Cardiac Arrhythmia Suppression Trial. Am J Cardiol. 1991;68:1551-1555. Reduced content of connexin43 gap junctions in ventricular myocardium from hypertrophied and ischemic human hearts. N S Peters, C R Green, P A Poole-Wilson and N J Severs Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017 Circulation. 1993;88:864-875 doi: 10.1161/01.CIR.88.3.864 Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1993 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/88/3/864 Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/