This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Verheule, S. Articles by Jongsma, H. J. Search for Related Content PubMed PubMed Citation Articles by Verheule, S. Articles by Jongsma, H. J.

(Circulation Research. 1997;80:673-681.)
© 1997 American Heart Association, Inc.

Articles

Characterization of Gap Junction Channels in Adult Rabbit Atrial and Ventricular Myocardium

Sander Verheule, Marjan J. A. van Kempen, Pascal H. J. A. te Welscher, Brenda R. Kwak, , Habo J. Jongsma

From the Department of Medical Physiology and Sports Medicine, Utrecht University, the Netherlands.

Correspondence to Sander Verheule, Department of Medical Physiology and Sports Medicine, Utrecht University, PO 80043, 3508 TA Utrecht, the Netherlands. E-mail verheule{at}med.ruu.nl

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract For effective cardiac output, it is essential that electrical excitation spread rapidly throughout the atria and ventricles. This is effected by electrical coupling through gap junction channels at contact sites between myocytes. These channels form a low-resistance pathway between adjacent myocytes and consist of connexin proteins. The connexin family is a large multigene family, and the channels formed by different members of this family have distinct electrical and regulatory properties. We have studied gap junction channels between adult rabbit atrial and ventricular myocytes using immunocytochemical and electrophysiological methods. Gap junctions of ventricular myocytes were immunoreactive to antibodies directed against connexin43 (Cx43) and Cx45, but not to antibodies against Cx37 or Cx40. Gap junctions between atrial myocytes showed immunostaining with anti-Cx40, -Cx43, and -Cx45 antibodies, but not with anti-Cx37 antibody. Endocardial and endothelial tissue were labeled with both Cx37 and Cx40 antibodies. The conductance of rabbit myocardial gap junctions was measured using the double whole-cell voltage-clamp method. The average macroscopic junctional conductance, corrected for series resistance, of atrial and ventricular cell pairs did not differ significantly (169±146 and 175±147 nS, respectively), and both were at most only slightly sensitive to the applied transjunctional potential difference. The difference in connexin expression between atrial and ventricular myocytes was reflected in the distribution of single gap junction channel conductances. A single population of unitary channel conductances with an average of 100 pS was observed between ventricular myocyte pairs. In addition to this population, a population with an average conductance of 185 pS was present between atrial myocyte pairs. The observed difference in connexin expression between atrial and ventricular myocardium may enable differential regulation of conduction in these tissues.

Key Words: gap junction • atrium • electrophysiology • ventricle • immunocytochemistry •

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
It has long been recognized that the working myocardium behaves as an electrical syncytium.¹ Taken together, the findings that the myocardium consists of distinct cells² and that the electrotonic length constant of cardiac muscle greatly exceeds the dimensions of a single cell³ suggest that there must exist a low-resistance pathway between neighboring myocytes. More recent research has shown that this pathway consists of clusters of gap junction channels that are present at cell-cell contacts (reviewed in Reference 4⁴ ). Because these channels allow passage of ions and small molecules up to 1 kD,⁵ myocytes are both electrically and metabolically coupled.

A single gap junction channel is formed by two hemichannels, with each of the apposing cells contributing one hemichannel. The hemichannel (or connexon) is a hexagonal aggregate of six connexin protein molecules.⁶ Connexins belong to a large multigene family, of which at present 13 members have been identified in mammals.⁷

Different cell types express different connexin proteins. Gap junction channels consisting of a single type of connexin protein (homomeric channels) have been shown to differ in biophysical properties, such as single-channel conductance, ionic selectivity (see eg, Reference 8⁸ ), and sensitivity to the transjunctional voltage (eg, see Reference 9⁹ ) Different homomeric channels also vary in their sensitivity to, for example, intracellular second messengers¹⁰ and pH.¹¹ In addition, differential connexin expression may electrically isolate different cell types, since some combinations of connexons consisting of a single type of connexin protein (homomeric connexons) do not readily form functional gap junction channels (for an overview, see Reference 12¹² ).

In the adult mammalian heart, several connexins are expressed. Connexin37 (Cx37), Cx40, Cx43, Cx45, and Cx46 mRNA have all been detected in homogenates of adult hearts.^{13 14 15 16} The distribution pattern may vary somewhat among species, but the similarities are striking. Cx43 is the most ubiquitous in myocardia of species investigated: it is abundantly present in atria and ventricles of rats, guinea pigs, pigs, cows, dogs, and humans.^{17 18} Cx40 is present in the atrial myocardium of guinea pigs, pigs, cows, dogs, and humans; the rat seems to be an exception, in that Cx40 is immunohistochemically not detectable in atrial myocardium.^{18 19} The presence of Cx45 in adult canine and human ventricular myocardium has also been reported.^{13 20} Cx37 and Cx40 are expressed in endothelium,^{14 21} whereas the localization of Cx46 within the heart is unknown.¹⁶ Connexin expression in the rabbit heart has hitherto not been described.

Whereas detailed accounts have been published of single gap junction channel recordings in cultured neonatal rat ventricular myocytes (see "Discussion"), only a few reports describe adult myocardial gap junctions electrophysiologically at the single-channel level. Rüdisüli and Weingart²² have studied gap junction channels between guinea pig ventricular cell pairs and have reported a population of single-channel conductances of 37 pS with cesium aspartate as a charge carrier (78 pS in CsCl). Lal and Ansdorf²³ have studied rat atrial gap junctions and found a mean single-channel conductance of 36 pS (in CsCl). The latter authors also described a pronounced dependence on transjunctional voltage of the macroscopic gap junctional conductance. In contrast, other researchers found ventricular gap junctions to be voltage insensitive.^{22 24 25 26}

We have performed a combined electrophysiological and immunocytochemical characterization of atrial and ventricular myocardial gap junctions in the adult rabbit. We report that atrial myocytes express Cx40, Cx43, and Cx45, whereas ventricular myocytes express only Cx43 and Cx45. The differential expression of Cx40 might enable differential modulation of gap junctional coupling in these tissues. In recent years, a number of reports have shown that several types of cardiac arrhythmias that occur in the human heart can also be induced in the rabbit heart.^{27 28} The similarity in connexin distribution in human and rabbit cardiac gap junctions demonstrated in the present study makes the rabbit heart a suitable and convenient model system for studying the role of gap junctions in cardiac arrhythmias.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Myocyte Isolation
Adult male New Zealand White rabbits weighing 1500 to 1700 g were anesthetized by injection of 1 to 1.5 mL Nembutal (60 mg/mL pentobarbital, Sanofi Sante b.v.) in the marginal ear vein. After opening the chest cavity, 0.3 mL heparin (5000 IU/mL, Leo) was injected into the left ventricle, and the heart was rapidly excised. Using a modified Langendorff perfusion setup,²⁹ the heart was retrogradely perfused via the aorta with the following solutions, at 37°C: (1) normal modified Tyrode's solution, 5 minutes, (2) zero-Ca²⁺ modified Tyrode's solution, 6 minutes, (3) zero-Ca²⁺ modified Tyrode's solution with 0.2 mg/mL collagenase B (0.75 U/mg, Boehringer) and 0.2 mg/mL collagenase P (2.2 U/mg, Boehringer), 8 minutes, and (4) solution 3, but with 0.1 mg/mL protease (Sigma), 6 minutes. Subsequently, the atria and ventricles were shaken separately in low-Ca²⁺ modified Tyrode's solution to release dissociated cells. After gradually raising the Ca²⁺ concentration of the solution to 1.8 mmol/L, the isolated myocytes were transferred to 35-mm culture dishes (Falcon 3801 Primaria, Becton Dickinson). Cells were allowed to attach for at least 30 minutes and were used within 8 hours after isolation.

Composition of salines was as follows: normal modified Tyrode's solution (mmol/L): NaCl 140, KCl 5.4, MgCl₂ 1, CaCl₂ 1.8, glucose 5, and HEPES 5, pH 7.4 (NaOH); zero-Ca²⁺ modified Tyrode's solution (mmol/L): NaCl 140, KCl 5.4, MgCl₂ 0.5, KH₂PO₄ 1.2, glucose 5, and HEPES 5, pH 7.2 (NaOH); and low-Ca²⁺ modified Tyrode's solution (mmol/L): NaCl 140, KCl 5.4, MgCl₂ 0.6, CaCl₂ 0.18, KH₂PO₄ 1.1, glucose 5, and HEPES 5, pH 7.2 (NaOH).

Immunochemistry
For immunolabeling of isolated myocytes, cells were plated on glass coverslips. After 1 hour, the cells were fixed by a 5-minute incubation in 100% methanol at -20°C. To detect Cx37, Cx40, and Cx43, rabbit polyclonal antibodies were used, kindly provided by Dr D. Gros, Department of Biology, University of Aix-Marseille II, France. Rabbit polyclonal anti-Cx45 was kindly provided by Dr E.C. Beyer, Department of Pediatrics, Washington University School of Medicine, St. Louis, Mo. These antibodies were raised against synthetic peptides corresponding to the following (intracellular) carboxy-terminal regions: AA₃₁₅ to AA₃₃₁ of mouse Cx37, AA₃₃₅ to AA₃₅₆ of rat Cx40,¹⁸ AA₃₁₄ to AA₃₂₂ of rat Cx43,³⁰ and AA₂₈₅ to AA₂₉₈ of dog Cx45.¹³ The specificity of the Cx37 antibody has been tested on communication-deficient cells transfected with mouse Cx37, and a full characterization will be published elsewhere (D. Gros, unpublished data, 1996). In some experiments, mouse monoclonal antibody directed at AA₂₅₂ to AA₂₇₀ of rat Cx43 (Zymed) or mouse monoclonal anti-desmin antibody (DAKO) was used. Cells were permeabilized with 0.2% Triton X-100 in PBS (1 hour) and incubated with 2% BSA in PBS for 30 minutes and with one of the primary antibodies overnight. Subsequently, cells were incubated with 2% BSA for 30 minutes and then for 2 hours with a secondary antibody (either goat anti-rabbit or donkey anti-mouse, depending on the primary antibody) labeled with a FITC or Texas red fluorophore. Between incubation steps, cells were washed with PBS. To test for the specificity of the antisera, cells were either incubated with a mixture of primary antibody and a 500-fold excess of peptide homologous to the epitope against which the antiserum was directed, or the first antibody was omitted altogether. For immunohistochemistry of intact tissue, excised hearts were perfused with normal modified Tyrode's solution for 10 minutes, after which pieces of atrium and ventricle were frozen rapidly in liquid nitrogen. Ten-micrometer cryosections were cut from the ventricles and the left atrium. The procedure for antibody labeling of these sections was the same as for isolated cells. All incubations were performed at room temperature.

Electrophysiology
A symmetrical setup with two independent DAGAN 8900 patch-clamp amplifiers was used. Data were stored on a Biologic DTR-1800 digital tape recorder (Biologic). Afterward, the data were fed into an Apple Macintosh Quadra 650 computer for further analysis. Borosilicate electrodes were pulled on a Narishige PB-7 electrode puller and fire-polished. Electrode resistances were 1 to 5 M. Junctional currents were recorded using the double whole-cell voltage-clamp technique.^{26 31} In short, both cells of a cell pair were voltage-clamped at 0 mV (the reversal potential of the membrane for the pipette solution used; see below). When a step in potential was applied to one cell, the current response in that cell was the sum of membrane currents, leak current, and junctional current. In the other cell, however, which was held at a constant potential of 0 mV, only the junctional current was recorded. The junctional conductance could then be calculated by dividing the observed junctional current by the applied transjunctional potential difference. The voltage drop across the series resistance (5 to 10 M) will cause an error in an applied transjunctional potential difference.³² The actual series resistance could be measured by canceling the fast initial voltage step in response to a small current step in the current-clamp mode. The validity of this method was ascertained using a model circuit with a variable series resistance. In this model circuit, the estimate for the series resistance was accurate within ±20%. To reduce the series resistance error, the series resistance was partly compensated in voltage clamp (between 50% and 80%), and the remaining series resistance (the resistance determined in current-clamp mode minus the resistance compensated in voltage-clamp mode) was corrected for afterward by deducting the uncompensated series resistance from the total resistance.

Halothane was used to reduce the gap junctional conductance to such an extent that currents through single gap junctional channels could be recorded. Data were filtered at 500 Hz and sampled at 2 kHz. The amplitude of single-channel transitions was measured by hand, using the MacDaq analysis software developed by Dr A.C.G. van Ginneken, Department of Physiology, University of Amsterdam, the Netherlands. Only transitions from the closed state to a fully open state were analyzed. Single-channel conductance histograms were fitted in KaleidaGraph (Adelbeck Software) with one, two, or three gaussian distributions using the least-squares method.

Action potentials were recorded at 35°C with the following pipette solution (mmol/L): potassium gluconate 130, KCl 5, MgCl₂ 2, CaCl₂ 0.6, Na₂ATP 4, EGTA 5, and HEPES 5, pH 7.2 (KOH). Bath medium consisted of normal modified Tyrode's solution. In recordings of junctional conductance, noise from channels in the nonjunctional membrane was reduced by adding 6 mmol/L NiCl₂, 2 mmol/L CsCl, and 1 mmol/L BaCl₂ to the bath, and the pipette solution consisted of (mmol/L) CsCl 120, tetraethylammonium chloride 10, MgCl₂ 3, CaCl₂ 1, Na₂ATP 2, EGTA 10, and HEPES 5, pH 7.2 (CsOH). All junctional conductance measurements were performed at room temperature.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Immunohistochemistry of Rabbit Cardiac Tissue
To investigate connexin expression in the adult rabbit heart, cryosections from atrium and ventricle were immunolabeled with Cx37, Cx40, Cx43, and Cx45 antibodies. No labeling with Cx37 antibody could be observed in the ventricular and atrial myocardium (Fig 1A and 1B, respectively). Interestingly, endothelial cells lining the blood vessels (Fig 1A) and endocardial cells lining the interior surface of the myocardium (Fig 1B) displayed specific Cx37 labeling at cell borders. In Fig 1C, it can be observed that Cx40 protein can not be detected in ventricular myocytes. However, this gap junction protein is clearly present in atrial myocardium (Fig 1D). Like Cx37, Cx40 was present in endothelial cells (Fig 1C and 1D).

View larger version (88K): [in this window] [in a new window]   Figure 1. Immunohistochemistry of intact cardiac tissue. A, Labeling with anti-Cx37 antibody: three blood vessels (positive) surrounded by ventricular myocardium (negative). B, Cx37 labeling: negative in atrial myocardium and positive in endocardium. C, Cx40 labeling: ventricular myocardium (negative) with blood vessel (positive). D, Cx40 labeling: positive in atrial myocardium and endothelium (upper right corner). E, Cx43 labeling (mouse monoclonal antibody) in ventricular myocardium (positive). F, Cx43 labeling (mouse monoclonal antibody) in atrial myocardium (positive). G, Cx45 labeling: positive in ventricular myocardium. H, Cx45 labeling: positive in atrial myocardium. Bar=100 µm.

Cx43 exhibits yet another distribution pattern: whereas it is absent in endothelial and endocardial cells (not shown), it is abundantly present in both atrial and ventricular myocytes (Fig 1E and 1F, respectively). Moreover, Cx43 protein was expressed in atrial epicardial tissue, which did not express either Cx37 or Cx40 (not shown).

The distribution pattern of Cx45 was similar to that of Cx43: it was present in both ventricular and atrial myocardium (Fig 1G and 1H) and absent in endothelial and endocardial cells.

Thus, all five connexin proteins investigated here have different and distinct distribution patterns in the adult rabbit heart. Connexin expression in endothelial and endocardial cells did not differ between atrium and ventricle. Myocardial tissue, on the other hand, showed a pronounced difference in connexin expression between atrium and ventricle. For convenience, these results have been summarized in the Table.

View this table: [in this window] [in a new window]   Table 1. Immunohistochemical Distribution of Cx37, Cx40, Cx43, and Cx45 in Adult Rabbit Cardiac Tissue

Immunocytochemistry of Isolated Cells
Enzymatic dissociation of adult rabbit hearts yielded predominantly single cells. However, numerous cell pairs remained intact (Fig 2A and 2B). These were mostly connected laterally, but because of the branched nature of cardiac myocytes, longitudinal junctions (perpendicular to the length axes of the cells) were sometimes present. Labeling with an anti-desmin antibody showed the typical cross striation of cardiomyocytes, illustrating that the cytoskeletal organization was not disrupted by the isolation and fixation procedures. Desmin labeling was especially intense in the intercalated disk areas and at lateral contact sites between myocytes (Fig 2C and 2D). Other cell types were also present in the isolate. Most notable of these was a round cell type that tended to occur in large flat clusters, typical of endothelial cells. However, the large size of these clusters and the fact that the clusters occurred relatively more frequently in the atrial than in the ventricular preparation make it plausible that these cells originated from the endocardium rather than from endothelium.

View larger version (84K): [in this window] [in a new window]   Figure 2. Morphology of isolated pairs. A and B, Ventricular myocyte pair (A) and atrial myocyte pair (B) (viewed with Hoffmann modulation contrast optics). C and D, Labeling with anti-desmin antibody in ventricular myocytes (C) and atrial myocytes (D). Bar=50 µm.

To determine whether the isolation procedure affected the connexin localization pattern in single myocytes, we labeled isolated cells with antibodies directed against Cx37, Cx40, Cx43, and Cx45. As expected from the results obtained from intact tissue, neither ventricular nor atrial myocytes were labeled specifically with anti-Cx37 antibody (Fig 3A and 3B, respectively). In some places, sparse intracellular labeling of ventricular cells with Cx37 antibody could be observed. This labeling did not disappear in the presence of an excess of Cx37 peptide and can thus be considered as nonspecific. However, islets of cells of putative endocardial origin were labeled strongly and specifically with this antibody (Fig 3B, center).

View larger version (67K): [in this window] [in a new window]   Figure 3. Connexin expression in isolated cells. Isolated ventricular (A, C, E, and G) and atrial (B, D, F and H) myocytes, labeled with anti-Cx37 (A and B), anti-Cx40(C and D), anti-Cx43 (E and F), and anti-Cx45 antibody (G and H). In panel B, a cluster of putative endocardial cells, labeled with anti-Cx37 antibody at cell borders, is also shown. Bar=50 µm.

Ventricular myocytes did not show specific binding of Cx40 antibody (Fig 3C). In contrast, atrial myocytes displayed specific labeling with Cx40 antibody in distinct spots along cell borders (Fig 3D), similar to the pattern observed in intact tissue. Putative endocardial cells were also labeled with Cx40 antibody (not shown). Whereas these latter cells were not immunoreactive to Cx43 antibody (not shown), both ventricular (Fig 3E) and atrial (Fig 3F) myocytes displayed abundant anti-Cx43 staining. Labeling was present in the intercalated disk areas and along the lateral cell borders. Similarly, anti-Cx45 staining was present between ventricular (Fig 3G) and atrial (Fig 3H) myocytes. To establish whether Cx40 and Cx43 in atrial myocytes are distributed in the same pattern, we performed double-labeling experiments with monoclonal mouse anti-Cx43 antibody and rabbit polyclonal anti-Cx40 antibody. As can be seen by comparing Fig 4A with 4B, Cx40 and Cx43 were strongly colocalized at the light-microscopic level. Most labeled areas contained both connexins.

View larger version (33K): [in this window] [in a new window]   Figure 4. Double labeling of Cx40 and Cx43: atrial myocyte pair labeled with rabbit polyclonal anti-Cx40 antibody and mouse monoclonal anti-Cx43 antibody as primary antibodies and goat anti-rabbit FITC and donkey anti-mouse Texas red as secondary antibodies. A, FITC fluorescence (Cx40). B, Texas red fluorescence (Cx43). The fluorescence spectra of the secondary antibodies did not overlap, as determined in experiments with a single primary antibody. Bar=50 µm.

In conclusion, connexin distribution in the adult rabbit myocardium is similar to the distribution reported by other researchers in guinea pigs, pigs, cows, and humans in that Cx43 and Cx45 are present in atrial and ventricular myocardium, but Cx40 is specifically expressed in atrial myocardium. Furthermore, enzymatic dissociation of myocardial tissue does not affect the distribution pattern of connexins in myocytes.

Macroscopic Junctional Conductance
When both cells of a cell pair were recorded in current clamp, synchronous action potentials could be elicited by stimulating one of the two cells. Action potentials had the shape that would be expected from ventricular and atrial myocytes (Fig 5A and 5B, respectively), illustrating that both cell types remained viable after the isolation procedure.

View larger version (6K): [in this window] [in a new window]   Figure 5. Evoked action potentials in myocyte pairs. A, Ventricular cell pair. Stimulation of the upper cell by a short current injection elicits synchronous action potentials in both cells. B, Synchronous action potential in an atrial cell pair, elicited by current injection in the upper cell. Recordings were at 35°C.

To determine the junctional conductance, myocyte pairs were recorded in the double whole-cell voltage-clamp mode. CsCl was used as the main charge carrier. After correction for series resistance, macroscopic junctional conductances ranged from 26 to 647 nS in ventricular pairs and from 30 to 635 nS in atrial pairs. Frequency histograms illustrating the wide range in total junctional conductance encountered between ventricular and atrial myocyte pairs are depicted in Fig 6A and 6B, respectively. The mean junctional conductance in atrial pairs (175±147 nS [mean±SD]; n=41; median, 116 nS) did not differ significantly from that in ventricular pairs (169±146 nS [mean±SD]; n=33; median, 111 nS).

View larger version (17K): [in this window] [in a new window]   Figure 6. Macroscopic junctional conductance (Gj) in myocyte pairs. Both cells of a pair were voltage-clamped at a common holding potential of 0 mV. Small voltage steps, 5 mV in amplitude, were applied to one cell of a pair, and the resulting junctional current in the other cell was measured. The total resistance was calculated by dividing the applied potential difference by the junctional current. The uncompensated series resistances were then subtracted from the total resistance to yield the true junctional resistance (Rj). Frequency histograms of total gap junctional conductance (Gj=1/Rj) in pairs of ventricular (A) and atrial (B) myocytes are shown. In both cases, CsCl was used as the main charge carrier.

To determine whether macroscopic junctional conductance is sensitive to the transjunctional potential difference, we applied large voltage steps in myocyte pairs. The true transjunctional potential difference was calculated by subtracting the voltage drops across the uncompensated part of the series resistances from the total applied potential afterward. A modest decay of the junctional current with time could be observed only in relatively weakly coupled pairs. In Fig 7A and 7B are shown representative traces from weakly coupled ventricular and atrial cell pairs, respectively.

View larger version (7K): [in this window] [in a new window]   Figure 7. Voltage sensitivity in relatively poorly coupled cell pairs. A, Current decay during a 4.5-second step to a transjunctional potential of 97 mV in a ventricular myocyte pair with a macroscopic junctional conductance of 32 nS. B, Current decay during a 4.5-second step to a transjunctional potential of 84 mV in an atrial myocyte pair with a junctional conductance of 30 nS. The transjunctional potential differences and junctional conductances have been corrected for the voltage drops across the series resistances.

Single-Channel Recordings
The macroscopic junctional current is the result of stochastic opening and closing of a large number of single gap junction channels. Single-channel transitions can be revealed under circumstances of almost complete uncoupling. Substances like higher alkanols and halothane are thought to uncouple gap junctions by dramatically lowering the open probability, without affecting the open-channel conductance.^{33 34} Single-channel conductances measured using these agents have been found to be characteristic for a given connexin and may vary widely among different connexins (see "Discussion"). Therefore, the immunochemically observed differences between connexin distribution in atrial and ventricular myocardium would be expected to correlate with differences in single gap junction channel conductances. To investigate this hypothesis, we recorded single-channel transitions in atrial and ventricular cell pairs under halothane-induced uncoupling. Halothane led to a fast and reversible reduction in gap junctional coupling (Fig 8A). Just before total uncoupling and in the early phase of recoupling, single-channel transitions could be resolved. In these periods, a continuous transjunctional potential difference was applied. Transitions resulting from gap junction channel gating could be recognized easily by the fact that they occur in both cells simultaneously, but in opposite directions. Fig 8B shows a short stretch of single-channel activity recorded from a ventricular cell pair. Single-channel transitions recorded in atrial cell pairs showed a wider variation in open-channel current, as exemplified in Fig 8C. Here, channels with an open-channel amplitude similar to the channels observed in ventricular cell pairs and channels with a notably higher amplitude could be observed.

View larger version (15K): [in this window] [in a new window]   Figure 8. Single-channel recording. A, Time course of halothane-induced uncoupling. Halothane was added to the bath medium at t=30 seconds and washed out at t=150 seconds. Black bars indicate the periods in which currents through single gap junction channels could be recorded. Between these two periods, the pair was completely uncoupled. The recovery from halothane-induced uncoupling is not shown completely. B, Single-channel openings and closings recorded from a ventricular myocyte pair at a driving force of 50 mV, showing an open amplitude of 4.5 pA. C, Single-channel openings and closings from an atrial myocyte pair at a driving force of 50 mV, containing both 5- and 8-pA single-channel amplitudes.

To investigate the influence of the applied transjunctional driving force on the single gap junction channel conductance, we have recorded single-channel activity at transjunctional potentials of 25, 50, and 75 mV. Data from several experiments were pooled into the single-channel conductance histograms depicted in Fig 9. These single-channel recordings were limited by high noise levels, probably originating from unblocked nonjunctional membrane channels and unresolved gap junction channel openings. The estimated average detection limits in these experiments are indicated by dotted lines.

View larger version (38K): [in this window] [in a new window]   Figure 9. Distribution of single gap junction channel conductance (gj) at several driving forces. N indicates the number of experiments; n, the number of events. Dotted lines represent average detection limits. A, Conductance frequency histogram of single gap channels in ventricular myocyte pairs at 25-mV transjunctional potential difference under halothane-induced uncoupling, fitted with one gaussian (mean±SD, 97.7±18 pS; average noise bandwidth, 1.2 pA). B, Distribution of single atrial gap junction channel conductances at transjunctional potential of 25 mV, fitted with one gaussian (mean±SD, 185.9±59 pS; average noise bandwidth, 2.3 pA). C, Ventricle, transjunctional potential of 50 mV, fitted with one gaussian (mean±SD, 97.7±29 pS; average noise bandwidth, 1.8 pA). D, Atrium, transjunctional potential of 50 mV, fitted with a sum of two gaussians (mean±SD, 99.3±20 and 184.2±53 pS; average noise bandwidth, 2.9 pA). E. Ventricle, transjunctional potential of 75 mV, fitted with one gaussian (mean±SD, 92.9±23; average noise bandwidth, 2.3 pA). F, Atrium, transjunctional potential of 75 mV, fitted with a sum of two gaussians (mean±SD, 108.5±39 and 175.9±39 pS; average noise bandwidth, 4.8 pA).

Ventricular gap junction channels showed a single predominant population. Indeed, the pattern in Fig 9C (50-mV driving force) could best be fit with a single gaussian distribution with an average conductance of 97.7 pS. The single-channel conductance was not affected by the applied driving force (Fig 9A and 9E).

Strikingly, at a transjunctional potential difference of 50 mV, atrial gap junction channels showed a much wider distribution (Fig 9D). This distribution could best be fit with a sum of two gaussian distributions. In this case, the smaller population had a mean conductance of 99.3 pS, similar to the population found between ventricular cells. The larger population had a mean conductance of 184.2 pS. At a transjunctional potential difference of 25 mV (Fig 9B), a broad distribution between 100 and 300 pS was observed. This distribution had an average of 185.9 pS. Compared with recordings from ventricular pairs, noise in recordings from atrial pairs was notably higher (average noise bandwidths at 25 mV were 1.2 and 2.3 pA, respectively). Therefore, the presence of channels in the 100-pS range (with a single-channel current amplitude of 2.5 pA) was impossible to ascertain at 25-mV driving force in atrial cell pairs. The distribution at 75-mV driving force (Fig 9F) was best fitted by the sum of two gaussian distributions, with mean conductances of 108.5 and 175.9 pS. Thus, the conductance of the smaller population was shifted slightly. This is probably related to higher noise from nonjunctional membrane channels, leading to a poorer signal-to-noise ratio compared with recordings at 50 mV and a correspondingly higher detection limit. In comparison, the number of events above 200 pS is smaller at 75 mV than at 25 mV. This might reflect a genuine dependence on transjunctional potential difference of the high-conductance channels present between atrial myocytes.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
The principal findings of the present study may be summarized as follows: Cx43, classically known as the "cardiac connexin," and Cx45 are abundantly present in the atrial and ventricular myocardium of the adult rabbit. This can be concluded from immunostaining of cardiac tissue and isolated myocytes. In addition, atrial myocytes also express Cx40. This distribution is similar to that in humans, pigs, cows, and guinea pigs.^{18 19} Cx37 and Cx40, which have been shown to be expressed by endothelial cells,^{14 21} are present in both endothelial and endocardial cells in the rabbit heart. It may be somewhat surprising that connexin distribution in the rabbit can be assessed with antibodies raised in the rabbit. Apparently, Langendorff perfusion before fixation caused a sufficient washout of immunoglobulins to which an anti-rabbit secondary antibody would have bound. Nevertheless, background fluorescence in experiments with rabbit primary antibodies was higher than in experiments with mouse monoclonal anti-Cx43 antibody but low enough to allow the detection of specific binding. In fact, the higher background fluorescence facilitated orientation in intact tissue.

Cx43 and Cx45 protein in ventricular myocytes and Cx40, Cx43, and Cx45 protein in atrial myocytes was mainly localized at lateral contacts to other myocytes and in intercalated disk areas (as assessed by double labeling with anti-desmin antibody, not shown). In electrophysiological experiments, we recorded mainly from laterally connected cell pairs. Our values for the average macroscopic junctional conductance are on the lower side of the range obtained from ventricular myocyte pairs by other researchers, which range from 100 to 1000 nS.^{22 25 26 35} In contrast, we never found values for macroscopic coupling as low as the values frequently encountered in rat atrial myocyte pairs by Lal and Ansdorf²³ (0.3 to 2 nS). An important methodological difference with their study is that we have recorded from cell pairs within a few hours after isolation, whereas Lal and Ansdorf have recorded after culturing myocytes for 18 hours, during which period connexin expression may have changed or previously unconnected cells may have formed new small gap junctions. A related difference with the study of Lal and Ansdorf on rat atrial gap junctions is that in the rabbit the gap junctional conductance of atrial myocytes is, at most, only slightly sensitive to the applied transjunctional potential difference. This is probably related to the observation that fully developed cardiac gap junctions consist of large clusters of gap junction channels (reviewed in Reference 4⁴ ). Wilders and Jongsma³² have shown that even after taking the voltage drops across the series resistances into account, only part of the true transjunctional potential difference is actually sensed by the individual channels. Their theoretical considerations imply that the electrical field line density around a gap junction channel decreases with increasing gap junction size, thus reducing the fraction of the total potential difference that can be detected by voltage-sensitive structures in the junctional channels. In this way, the voltage sensitivity of the individual channel is blunted by the presence of neighboring channels. Therefore, we expect that in the intact working myocardium under nonpathological conditions, voltage sensitivity of gap junctions will not play an important role in the conduction of the action potential.

As stated in the introduction, Cx37, Cx40, Cx43, Cx45, and Cx46 mRNA are all present in heart homogenates.^{13 14 15 16} The biophysical properties of gap junction channels formed by these connexins vary widely within a given species, but the properties of homologous connexins in different mammalian species bear a remarkable degree of similarity. This also pertains to single-channel conductances (for an overview see References 8 and 36^{8 36} ); therefore, single-channel conductance may be said to be characteristic of a connexin.

With the limitation of a relatively high noise level, inherent in large cardiomyocytes, we have attempted to correlate the expression of connexin proteins and the distribution of single-channel conductances. Using CsCl as the main charge carrier, single-channel conductance histograms obtained from ventricular cell pairs showed a single predominant peak at 100 pS. This is in fair agreement with the single-channel conductance in guinea pig ventricular myocytes of 80 pS in CsCl, as reported by Rüdisüli and Weingart,²² which was similarly unaffected by the trans-junctional potential difference. In neonatal rat ventricular myocytes, which express mainly Cx43, conductance states of 20, 40 to 45, and 70 pS in potassium gluconate have been observed.³⁷ Other researchers, who have recorded from Cx43 channels in expression systems, have found single-channel conductance states of 30, 60, and 90 pS for rat Cx43 in CsCl¹⁰ and states at 60 to 70 and 90 to 100 pS for human Cx43 in CsCl.³⁸ On the basis of the observed expression of Cx43 in rabbit ventricular myocytes, we believe that the 100-pS channel population corresponds with channels formed from Cx43 protein. This value is similar to the highest conductance state of Cx43 in other studies. Possibly, we were not able to resolve the lower conductance states of Cx43 because of an insufficiently high signal-to-noise ratio. It is likely that lower conductance states are present, because it has been reported that Cx43 occurs as a phosphoprotein in the ventricular myocardium,³⁹ and the phosphorylated forms are thought to correspond to the lower conductance states.^{38 40}

We found a much wider distribution of single-channel conductances in atrial myocyte pairs, with peaks at 100 and 185 pS. These data diverge from the results from rat atrial myocytes reported by Lal and Ansdorf,²³ who found a single peak at 40 pS in CsCl. However, as mentioned before, the rat differs from other species examined, in that Cx40 is not detectable in the atrial myocardium.^{18 19} Assuming the 100-pS peak to be caused by gating of Cx43 channels, we believe that the higher conductances are derived from Cx40 channels. Mouse Cx40 has single-channel conductance states of 153 and 121 pS (KCl as a charge carrier).⁴¹ For rat Cx40, conductances of 158 pS in potassium glutamate and 180 pS in KCl have been reported.⁴² Therefore, it seems reasonable to ascribe the observed 185-pS single-channel conductance to gating of Cx40 channels. If all openings in the 140- to 250-pS range resulted from gating of Cx40, then it is possible that Cx40 has more than one conductance state, which is unresolved in our conductance histograms. The shift away from higher conductances, observed with increasing transjunctional potential difference, might reflect a greater voltage sensitivity of the open probability of higher conductance states (>200 pS). Indeed, Bukauskas et al⁴³ have reported that the open probability of mouse Cx40 is strongly voltage sensitive, but these researchers only found a single main conductance state.

The absence of Cx37 from myocardial gap junctions is substantiated by the absence of specific anti-Cx37 antibody binding. Moreover, the reported high single-channel conductance of Cx37, 356 pS in KCl,⁴⁴ is not present in our single-channel conductance histograms. Unfortunately, Cx45 forms channels with a conductance of only 30 pS with CsCl as a charge carrier,^{10 45 46} which is too low to be detected, given the detection limit in our recordings. Therefore, our recording method does not allow us to study a possible modulation of the single-channel conductance of Cx45 channels.

From double-labeling experiments, it appears that in atrial myocytes, Cx40 is colocalized with Cx43 at the light-microscopic level. It is unlikely that Cx40 and Cx43 are colocalized at the level of the single atrial gap junction channel, because Cx40 and Cx43 have been shown to be incompatible and do not form heterotypic channels in expression systems.^{21 47} Cx45 is able to form heterotypic channels with both Cx40 and Cx43 in expression systems,⁴⁸ but this does not necessarily have to occur in myocytes. In principle, heterotypic and heteromeric Cx37/Cx40 channels could exist between endothelial and endocardial cells.²¹ However, channels between endothelial cells and myocytes are not likely to be formed, since these cell types are separated by connective tissue.

The observed differential expression of Cx40 in atrial and ventricular myocytes may have important consequences for the regulation of conduction in these tissues. For example, angiotensin II⁴⁹ and carbachol^{35 50} reduce junctional coupling between ventricular myocytes, whereas isoproterenol has been reported to increase coupling.⁴⁷ To date, no examples have been reported of modulation of atrial gap junction channels by these or other compounds. However, if Cx40 channels are affected differently, then this could provide a mechanism for differential modulation of atrial and ventricular conduction by the autonomous nervous system and humoral factors.

Cx43 channels close as a result of a number of factors commonly associated with myocardial ischemia (reviewed in Reference 51⁵¹ ). This may lead to "sealing off" of the ischemic area from healthy myocardium. On the other hand, the resulting slow conduction may give rise to reentrant circuits and lethal arrhythmias. We propose that the rabbit myocardium is a suitable system for studying the role of gap junctional communication in cardiac arrhythmias, because the connexin expression pattern in rabbits seems comparable to that observed in humans and because several types of cardiac arrhythmias can actually be evoked in rabbit hearts.^{27 28}

	Acknowledgments

 
Drs B.R. Kwak and H.J. Jongsma were supported by a combined grant from the Netherlands Organization for Scientific Research and the Netherlands Heart Foundation (No. 900-516-093). Dr van Kempen was supported by grant 91.059 from the Netherlands Heart Foundation. The authors thank Anton van der Wardt for his expert technical assistance. We thank Dr Daniel Gros for kindly providing the polyclonal Cx37, Cx40, and Cx43 antibodies and Dr Eric Beyer for kindly providing the Cx45 antibody.

Received December 9, 1996; accepted February 6, 1997.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Engelmann TW. Über die Leitung der Erregung im Herzmuskel. Pflugers Arch. 1875;11:465-480.
2. Sjöstrand FS, Anderson E. Electron microscopy of the intercalated discs of cardiac muscle tissue. Experientia. 1954;10:369-370.[Medline] [Order article via Infotrieve]
3. Weidmann S. Elektrophysiologie der Herzmuskelfaser. Bern, Switzerland: Hans Huber; 1956.
4. Severs NJ. The cardiac gap junctions and intercalated disc. Int J Cardiol. 1990;26:137-173.[Medline] [Order article via Infotrieve]
5. Simpson I, Rose B, Loewenstein WR. Size limit of molecules permeating the junctional membrane channels. Science. 1977;197:294-296.
6. Yeager M, Gilula NB. Membrane topology and quaternary structure of cardiac gap junction ion channels. J Mol Biol. 1992;223:929-948.[Medline] [Order article via Infotrieve]
7. Goodenough DA, Goliger JA, Paul DL. Connexins, connexons and intercellular communication. Annu Rev Biochem. 1996;65:475-502.[Medline] [Order article via Infotrieve]
8. Veenstra RD, Wang H-Z, Beblo DA, Chilton MG, Harris AL, Beyer EC, Brink PR. Selectivity of connexin-specific gap junctions does not correlate with channel conductance. Circ Res. 1995;77:1156-1165.[Abstract/Free Full Text]
9. Veenstra RD, Wang H-Z, Westphale EM, Beyer EC. Multiple connexins confer distinct regulatory and conductance properties of gap junctions in developing heart. Circ Res. 1992;71:1277-1283.[Abstract/Free Full Text]
10. Kwak BR, Hermans MMP, de Jonge HR, Lohmann SM, Jongsma HJ, Chanson M. Differential regulation of gap junction channels by similar phosphorylating conditions. Mol Biol Cell. 1995;6:1707-1719.[Abstract]
11. Hermans MMP, Kortekaas P, Jongsma HJ, Rook MB. pH sensitivity of the cardiac gap junction proteins, connexin 45 and 43. Pflugers Arch. 1995;431:138-140.[Medline] [Order article via Infotrieve]
12. Bruzzone R, White TW, Paul DL. Connections with connexins: the molecular basis of direct intercellular signaling. Eur J Biochem. 1996;238:1-27.[Medline] [Order article via Infotrieve]
13. Kanter HL, Saffitz JE, Beyer EC. Cardiac myocytes express multiple gap junction proteins. Circ Res. 1992;70:438-444.[Abstract/Free Full Text]
14. Reed KE, Westphale EM, Larson DM, Wang H-Z, Veenstra RD, Beyer EC. Molecular cloning and functional expression of human connexin37, an endothelial cell gap junction protein. J Clin Invest. 1993;91:997-1004.
15. Haefliger J-A, Bruzzone R, Jenkins NA, Gilbert DJ, Copeland NG, Paul DL. Four novel members of the connexin family of gap junction proteins. J Biol Chem. 1992;267:2057-2064.[Abstract/Free Full Text]
16. Paul DL, Ebihara L, Takemoto LJ, Swenson KI, Goodenough DA. Connexin 46, a novel lens gap junction protein, induces voltage-gated currents in nonjunctional plasma membrane of Xenopus oocytes. J Cell Biol. 1991;115:1077-1089.[Abstract/Free Full Text]
17. Gourdie RG, Green CR, Severs NJ. Gap junction distribution in adult mammalian myocardium revealed by an anti-peptide antibody and laser scanning confocal microscopy. J Cell Sci. 1991;99:41-55.[Abstract/Free Full Text]
18. van Kempen MJA, ten Velde I, Wessels A, Oosthoek PW, Gros D, Jongsma HJ, Moorman AFM, Lamers WH. Differential connexin distribution accommodates cardiac function in different species. Microsc Res Tech. 1995;31:420-436.[Medline] [Order article via Infotrieve]
19. Gros D, Jarry-Guichard T, Ten Velde I, de Maziere A, van Kempen MJA, Davoust J, Briand JP, Moorman AFM, Jongsma HJ. Restricted distribution of connexin40, a gap junctional protein, in mammalian heart. Circ Res. 1994;74:839-851.[Abstract/Free Full Text]
20. Davis LM, Rodefeld ME, Green K, Beyer EC, Saffitz JE. Gap junction protein phenotypes of the human heart and conduction system. J Cardiovasc Electrophysiol. 1995;6:813-822.[Medline] [Order article via Infotrieve]
21. Bruzzone R, Haefliger J-A, Gimlich RL, Paul DL. Connexin40, a component of gap junctions in vascular endothelium, is restricted in its ability to interact with other connexins. Mol Biol Cell. 1993;4:7-20.[Abstract]
22. Rüdisüli A, Weingart R. Electrical properties of gap junction channels in guinea-pig ventricular cell pairs revealed by exposure to heptanol. Pflugers Arch. 1989;415:12-21.[Medline] [Order article via Infotrieve]
23. Lal R, Ansdorf MF. Voltage-dependent gating and single-channel conductance of adult mammalian atrial gap junctions. Circ Res. 1992;71:737-743.[Abstract/Free Full Text]
24. Kameyama M. Electrical coupling between ventricular cells isolated from guinea-pig heart. J Physiol (Lond). 1983;336:345-357.[Abstract/Free Full Text]
25. Weingart R. Electrical properties of the nexal membrane studied in rat ventricular cell pairs. J Physiol (Lond). 1986;370:267-284.[Abstract/Free Full Text]
26. White RL, Spray DC, Campos de Carvalho AC, Wittenberg BA, Bennett MVL. Some electrical and pharmacological properties of gap junctions between adult ventricular myocytes. Am J Physiol. 1985;249:C447-C455.[Abstract/Free Full Text]
27. Gray RA, Jalife J, Panfilov A, Baxter WT, Cabo C, Davidenko JM, Pertsov AM. Nonstationary vortexlike reentrant activity as a mechanism of polymorphic ventricular tachycardia in the isolated rabbit heart. Circulation. 1995,91:2454-2469.
28. Vermeulen JT, McGuire MA, Opthof T, Coronel R, de Bakker JMT, Klopping C, Janse MJ. Triggered activity and automaticity in ventricular trabeculae of failing human and rabbit hearts. Cardiovasc Res. 1994;28:1547-1554.[Medline] [Order article via Infotrieve]
29. Langendorff, O. Untersuchungen am überlebenden Säugethierherzen. Pflugers Arch. 1895;61:291-332.
30. El Aoumari A, Fromaget C, Dupont E, Reggio H, Durbec P, Briand JP, Beller K, Kreitman B, Gros D. Conservation of a cytoplasmic carboxy-terminal domain of Cx43, a gap junction protein, in mammal heart and brain. J Membr Biol. 1990;115:229-240.[Medline] [Order article via Infotrieve]
31. Neyton J, Trautmann A. Single channel currents of an intercellular junction. Nature. 1985;317:331-335.[Medline] [Order article via Infotrieve]
32. Wilders R, Jongsma HJ. Limitations of the dual voltage clamp method in assaying conductance and kinetics of gap junction channels. Biophys J. 1992;63:942-953.[Abstract/Free Full Text]
33. Burt JM, Spray DC. Single-channel events and gating behaviour of the cardiac gap junction channel. Proc Natl Acad Sci U S A. 1988;85:3431-3434.[Abstract/Free Full Text]
34. Takens-Kwak BR, Jongsma HJ, Rook MB, van Ginneken ACG. Mechanism of heptanol induced uncoupling of cardiac gap junctions: a perforated patch-clamp study. Am J Physiol. 1992;262:C1531-C1538.[Abstract/Free Full Text]
35. Wittenberg BA, White RL, Ginzberg RD, Spray DC. Effect of calcium on the dissociation of the mature rat heart into individual and paired myocytes: electrical properties of cell pairs. Circ Res. 1986;59:143-150.[Abstract/Free Full Text]
36. Beyer EC, Davis LM, Saffitz JE, Veenstra RD. Cardiac intercellular communication: consequences of connexin distribution and diversity. Braz J Med Biol Res. 1995;28:415-425.[Medline] [Order article via Infotrieve]
37. Takens-Kwak BR, Jongsma HJ. Cardiac gap junctions: three distinct single channel conductances and their modulation by phosphorylating treatments. Pflugers Arch. 1992;422:198-200.[Medline] [Order article via Infotrieve]
38. Moreno AP, Sáez JC, Fishman GI, Spray DC. Human connexin43 gap junction channels: regulation of unitary conductances by phosphorylation. Circ Res. 1994;74:1050-1057.[Abstract/Free Full Text]
39. Kadle R, Zhang JT, Nicholson BJ. Tissue-specific distribution of differentially phosphorylated forms of Cx43. Mol Cell Biol. 1991;11:363-369.[Abstract/Free Full Text]
40. Kwak BR, Sáez JC, Wilders R, Chanson M, Fishman GI, Hertzberg EL, Spray DC, Jongsma HJ. Effects of cGMP-dependent phosphorylation on rat and human connexin43 gap junction channels. Pflugers Arch. 1995;430:770-778.[Medline] [Order article via Infotrieve]
41. Traub O, Eckert R, Lichtenberg-Fraté, Elfgang C, Bastide B, Scheidtmann KH, Hülser DF, Willecke K. Immunochemical and electrophysiological characterization of murine connexin40 and -43 in mouse tissues and transfected human cells. Eur J Cell Biol. 1994;64:101-112.[Medline] [Order article via Infotrieve]
42. Beblo DA, Wang H-Z, Beyer EC, Westphale EM, Veenstra RD. Unique conductance, gating, and selective permeability properties of gap junction channels formed by connexin40. Circ Res. 1995;77:813-822.[Abstract/Free Full Text]
43. Bukauskas FF, Elfgang C, Willecke K, Weingart R. Biophysical properties of gap junction channels formed by mouse connexin40 in induced pairs of transfected human HeLa cells. Biophys J. 1995;68:2289-2298.[Abstract/Free Full Text]
44. Veenstra RD, Wang H-Z, Beyer EC, Ramanan SV, Brink PR. Connexin37 forms high conductance gap junction channels with subconductance state activity and selective dye and ionic permeabilities. Biophys J. 1994:66:1915-1928.
45. Moreno AP, Laing JG, Beyer EC, Spray DC. Properties of gap junction channels formed of connexin 45 endogenously expressed in human hepatoma (SKHep1) cells. Am J Physiol. 1995;268:C356-C365.[Abstract/Free Full Text]
46. Veenstra RD, Wang H-Z, Beyer EC, Brink PR. Selective dye and ionic permeability of gap junction channels formed by connexin45. Circ Res. 1994;75:483-490.[Abstract/Free Full Text]
47. Elfgang C, Eckert R, Lichtenberg-Fraté H, Butterweck A, Traub O, Klein RA, Hülser DF, Willecke K. Specific permeability of gap junction channels in connexin-transfected HeLa cells. J Cell Biol. 1995;129:805-817.[Abstract/Free Full Text]
48. Elfgang C, Eckert R, Lichtenberg-Frate H, Butterweck A, Traub O, Klein RA, Hulser DF, Willecke K. Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J Cell Biol. 1995;129:805-817.
49. De Mello WC. Is an intracellular renin-angiotensin system involved in control of cell communication in heart? J Cardiovasc Pharmacol. 1994;23:640-646.[Medline] [Order article via Infotrieve]
50. Burt JM, Spray DC. Inotropic agents modulate gap junctional conductance between cardiac myocytes. Am J Physiol. 1988;23:H1206-H1210.
51. Jongsma HJ, Rook MB. Morphology and electrophysiology of cardiac gap junction channels. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. 2nd ed. Philadelphia, Pa: WB Saunders; 1995.

This article has been cited by other articles:

				N. Gassanov, D. Devost, B. Danalache, N. Noiseux, M. Jankowski, H. H. Zingg, and J. Gutkowska Functional Activity of the Carboxyl-Terminally Extended Oxytocin Precursor Peptide During Cardiac Differentiation of Embryonic Stem Cells Stem Cells, January 1, 2008; 26(1): 45 - 54. [Abstract] [Full Text] [PDF]

				D. Volonte, C. F. McTiernan, M. Drab, M. Kasper, and F. Galbiati Caveolin-1 and caveolin-3 form heterooligomeric complexes in atrial cardiac myocytes that are required for doxorubicin-induced apoptosis Am J Physiol Heart Circ Physiol, January 1, 2008; 294(1): H392 - H401. [Abstract] [Full Text] [PDF]

				I. Banerjee, J. W. Fuseler, R. L. Price, T. K. Borg, and T. A. Baudino Determination of cell types and numbers during cardiac development in the neonatal and adult rat and mouse Am J Physiol Heart Circ Physiol, September 1, 2007; 293(3): H1883 - H1891. [Abstract] [Full Text] [PDF]

				J. O. Tellez, H. Dobrzynski, I. D. Greener, G. M. Graham, E. Laing, H. Honjo, S. J. Hubbard, M. R. Boyett, and R. Billeter Differential Expression of Ion Channel Transcripts in Atrial Muscle and Sinoatrial Node in Rabbit Circ. Res., December 8, 2006; 99(12): 1384 - 1393. [Abstract] [Full Text] [PDF]

				C. Cabo, J. Yao, P. A. Boyden, S. Chen, W. Hussain, H. S. Duffy, E. J. Ciaccio, N. S. Peters, and A. L. Wit Heterogeneous gap junction remodeling in reentrant circuits in the epicardial border zone of the healing canine infarct Cardiovasc Res, November 1, 2006; 72(2): 241 - 249. [Abstract] [Full Text] [PDF]

				A. E. Pollard, W. M. Smith, and R. C. Barr Feasibility of cardiac microimpedance measurement using multisite interstitial stimulation Am J Physiol Heart Circ Physiol, December 1, 2004; 287(6): H2402 - H2411. [Abstract] [Full Text] [PDF]

				S. Poelzing and D. S. Rosenbaum Altered connexin43 expression produces arrhythmia substrate in heart failure Am J Physiol Heart Circ Physiol, October 1, 2004; 287(4): H1762 - H1770. [Abstract] [Full Text] [PDF]

Y.-S. Ko, H.-I Yeh, Y.-L. Ko, Y.-C. Hsu, C.-F. Chen, S. Wu, Y.-S. Lee, and N. J. Severs Three-Dimensional Reconstruction of