Habo J. Jongsma

Limitations of the dual voltage clamp method in assaying conductance and kinetics of gap junction channels

The electrical properties of gap junctions in cell pairs are usually studied by means of the dual voltage clamp method. The voltage across the junctional channels, however, cannot be controlled adequately due to an artificial resistance and a natural resistance, both connected in series with the gap junction. The access resistances to the cell interior of the recording pipettes make up the artificial resistance. The natural resistance consists of the cytoplasmic access resistances to the tightly packed gap junction channels in both cells. A mathematical model was constructed to calculate the actual voltage across each gap junction channel. The stochastic open-close kinetics of the individual channels were incorporated into this model. It is concluded that even in the ideal case of complete compensation of pipette series resistance, the number of channels comprised in the gap junction may be largely underestimated. Furthermore, normalized steady-state junctional conductance may be largely overestimated, so that transjunctional voltage dependence is easily masked. The model is used to discuss conclusions drawn from dual voltage clamp experiments and offers alternative explanations for various experimental observations.

Pacemaker activity of the rabbit sinoatrial node. A comparison of mathematical models

In the past decade, three mathematical models describing the pacemaker activity of the rabbit sinoatrial node have been developed: the Bristow-Clark model, the Irisawa-Noma model, and the Noble-Noble model. In a comparative study it is demonstrated that these models, as well as subsequent modifications, all have several drawbacks. A more accurate model, describing the pacemaker activity of a single pacemaker cell isolated from the rabbit sinoatrial node, was constructed. Model equations, including equations for the T-type calcium current, are based on experimental data from voltage clamp experiments on single cells that were published during the last few years. In contrast to the other models, only a small amount of background current contributes to the overall electrical charge flow. The action potential parameters of the model cell, its responses to voltage clamp steps and its current-voltage relationships have been computed. The model is used to discuss the relative contribution of membrane current components to the slow diastolic depolarization phase of the action potential.

Heptanol-induced decrease in cardiac gap junctional conductance is mediated by a decrease in the fluidity of membranous cholesterol-rich domains.

To assess whether alterations in membrane fluidity of neonatal rat heart cells modulate gap junctional conductance (gj), we compared the effects of 2mm 1-heptanol and 20 μm 2-(methoxyethoxy)ethyl 8-(cis-2-n-octylcyclopropyl)-octanoate (A2C) in a combined fluorescence anisotropy and electrophysiological study. Both substances decreased fluorescence steady-state anisotropy (rss), as assessed with the fluorescent probe 1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene (TMA-DPH) by 9.6±1.1% (mean ±sem,n=5) and 9.8±0.6% (n=5), respectively, i.e., both substances increased bulk membrane fluidity. Double whole-cell voltage-clamp experiments showed that 2mm heptanol uncoupled cell pairs completely (n=6), whereas 20 μm A2C, which increased bulk membrane fluidity to the same extent, did not affect coupling at all (n=5).Since gap junction channels are embedded in relatively cholesterol-rich domains of the membrane, we specifically assessed the fluidity of the cholesterol-rich domains with dehydroergosterol (DHE). Using DHE, heptanol increased rss by 14.9±3.0% (n=5), i.e., decreased cholesterol domain fluidity, whereas A2C had no effect on rss (−0.4±6.7%,n=5).Following an increase of cellular “cholesterol” content (by loading the cells with DHE), 2mm heptanol did not uncouple cell pairs completely:gj decreased by 80±20% (range 41–95%,n=5). The decrease ingj was most probably due to a decrease in the open probability of the gap junction channels, because the unitary conductances of the channels were not changed nor was the number of channels comprising the gap junction. The sensitivity of non-junctional membrane channels to heptanol was unaltered in cholesterol-enriched myocytes.These results indicate that the fluidity of cholesterol-rich domains is of importance to gap junctional coupling, and that heptanol decreasesgj by decreasing the fluidity of cholesterol-rich domains, rather than by increasing the bulk membrane fluidity.

Effects of cGMP-dependent phosphorylation on rat and human connexin43 gap junction channels

The effects of 8-bromoguanosine 3′:5′-cyclic monophosphate (8Br-cGMP), a membrane-permeant activator of protein kinase G (PKG), were studied on rat and human connexin43 (Cx43), the most abundant gap junction protein in mammalian heart, which were exogenously expressed in SKHep1 cells. Under dual whole-cell voltage-clamp conditions, 8Br-cGMP decreased gap junctional conductance (gj) in rat Cx43-transfected cells by 24.0±3.7% (mean±SEM, n=5), whereas gj was not affected in human Cx43-transfected cells by the same treatment. The relaxation of gj in response to steps in transjunctional voltage observed in rat Cx43 transfectants was best fitted with three exponentials. Time constants and amplitudes of the decay phases changed in the presence of 8Br-cGMP. Single rat and human Cx43 gap junction channels were resolved in the presence of halothane. Under control conditions, three single-channel conductance states (γj) of about 20, 40–45 and 70 pS were detected, the events of the intermediate size being most frequently observed. In the presence of 8Br-cGMP, the γj distribution shifted to the lower size in rat Cx43 but not in human Cx43 transfectants. Immunoblot analyses of Cx43 in subconfluent cultures of rat Cx43 or human Cx43 transfectants showed that 8Br-cGMP did not induce changes in the electrophoretic mobility of Cx43 in either species. However, the basal incorporation of [32P] into rat Cx43 was significantly altered by 8Br-cGMP, whereas this incorporation of [32P] into human Cx43 was not affected. We conclude that 8Br-cGMP modulates phosphorylation of rat Cx43 in SKHep1 cells, but not of human Cx43. This cGMP-dependent phosphorylation of rat Cx43 is associated with a decreased gj, which results from both an increase in the relative frequency of the lowest conductance state and a change in the kinetics of these channels.

Cardiac gap junctions: three distinct single channel conductances and their modulation by phosphorylating treatments

SummaryThe effects of an increase in intracellular cyclic GMP (cGMP)-concentration on gap junctional current (Ij) were studied in cultured neonatal rat heart cells using both the whole-cell and perforated patch voltage-clamp method. In whole-cell measurements, exposure to 8-bromo-cGMP or carbachol reduced Ij. With the perforated patch technique, on the other hand, Ij was not affected by either 8-bromo-cGMP or carbachol. Addition of alkaline phosphatase prevented the carbachol-induced decrease in Ij in whole-cell measurements. Reduction of Ij in well-coupled cell pairs by application of heptanol allowed us to study the effects of these substances on the single gap junction channel level. We found that cGMP-treatment shifts the single channel conductance (γj) from 43 to 21 pS in whole-cell measurements and that intracellular addition of phosphatase prevents this shift. In contrast, intracellular phosphatase-treatment itself shifts γj to 70 pS. Our results indicate that Cx43-gap junction channels may exhibit three conductance levels (21 pS, 40–45 pS and 70 pS), depending on the phosphorylation state of the protein.

Single channel currents of homo- and heterologous gap junctions between cardiac fibroblasts and myocytes

Recently, the use of the double whole-cell patchclamp technique enable conductance measurements of single gap junctional channels. Different values have been measured in pairs of rat lacrimal cells (6), murine acinar cells and chinese hamster ovary cells (9), embryonic chick heart- (10) and neonatal rat heart myocytes (7). We here present evidence that the conductance of gap junction channels between two different cell types originating from the same tissue, neonatal rat heart, is different. In mixed cultures of cardiac fibroblasts and myocytes, gap junction channels between fibroblasts have a single channel conductance of only 22 pS, while those between myocytes have a conductance of 43 pS. Fibroblasts can be electrically coupled to myocytes through channels having an intermediate conductance of 29 pS, a value which matches very well with te theoretically expected conductance of a gap junction channel composed of a fibroblast- and a myoblast connexon (hemichannel). These data provide direct evidence on the single channel level that in heterologous gap junction channels the composing connexons retain their cell-specific properties.

Spatial and functional relationship between myocytes and fibroblasts in the rabbit sinoatrial node

In an attempt to understand better the directional differences in conduction velocity in the rabbit sinoatrial node, a possible conductive role of the abundant connective tissue surrounding the myocytes has been investigated. In particular, starting from the finding of communicating junctions between heart muscle cells and fibroblasts in tissue culture, heterologous gap junctions were searched for in thin sections of the rabbit sinoatrial node. Within and at the edge of nodal cell clusters, fibroblasts often show thin sheet-like extensions parallel to the surface of myocytes. In contrast to the intimately contacting myocytes, fibroblast extensions are kept separated from the myocytes by the basement membrane of the latter. Besides some rare undefined membrane appositions a single tiny gap junction-like structure was found between a fibroblast and a myocyte in a tissue area in which the calculated number of gap junctions between myocytes amounts from 1.10(4) to 3.10(4). Yet, fibroblasts are linked together regularly by small gap junctions containing a wider gap than the junctions between the myocytes (1.4 +/- 0.4 nm vs. 1.0 +/- 0.4 nm, resp., P less than 0.05). As an alternative to direct electrical coupling, the possibility of interaction between fibroblasts and nodal cells by capacitive coupling has been considered. Model calculations based on the reconstruction of some fibroblast extensions parallel to nodal cells show that the current which can be transmitted from discharging nodal cells to fibroblasts is negligible. It is concluded that fibroblasts do not participate in the impulse conduction within the sinoatrial node. The origin of the directional differences in conduction velocity in the sinoatrial node must be found in the spatial arrangement of the myocytes and the distribution of the gap junctions between these cells only.

Beating irregularity of single pacemaker cells isolated from the rabbit sinoatrial node

Single pacemaker heart cells discharge irregularly. Data on fluctuations in interbeat interval of single pacemaker cells isolated from the rabbit sinoatrial node are presented. The coefficient of variation of the interbeat interval is quite small, approximately 2%, even though the coefficient of variation of diastolic depolarization rate is approximately 15%. It has been hypothesized that random fluctuations in interbeat interval arise from the stochastic behavior of the membrane ionic channels. To test this hypothesis, we constructed a single channel model of a single pacemaker cell isolated from the rabbit sinoatrial node, i.e., a model into which the stochastic open-close kinetics of the individual membrane ionic channels are incorporated. Single channel conductances as well as single channel open and closed lifetimes are based on experimental data from whole cell and single channel experiments that have been published in the past decade. Fluctuations in action potential parameters of the model cell are compared with those observed experimentally. It is concluded that fluctuations in interbeat interval of single sinoatrial node pacemaker cells indeed are due to the stochastic open-close kinetics of the membrane ionic channels.

Spatial Distribution of Connexin43, the Major Cardiac Gap Junction Protein, Visualizes the Cellular Network for Impulse Propagation From Sinoatrial Node to Atrium

Myocytes are electrically coupled by gap junctions, which are composed of low-resistance intercellular channels. The major cardiac gap junction protein is connexin43 (Cx43). The distribution of Cx43 has been studied by immunofluorescence to visualize the electrical coupling between atrial tissue and sinoatrial node. From modeling studies, this coupling was inferred to be gradual in order to shield the sinoatrial node from the atrial hyperpolarizing influence. The actual Cx43 labeling pattern did not show the expected gradient but instead a rather black and white staining in a striking pattern of strands of cells. We used an immunohistochemical marker (anti-alpha-smooth muscle actin [alpha SMA]) that specifically cross-reacts with guinea pig sinoatrial node cells together with Cx43 antibody to stain previously electrophysiologically mapped sinoatrial nodes. We found that in the guinea pig sinoatrial node the impulse originates in an alpha SMA-positive, virtually Cx43-negative, region (primary pacemaker region). The impulse then travels obliquely upward to the crista terminalis through a region where layers of alpha SMA-positive cells alternate with layers of Cx43-positive SMA-negative cells. The layers of Cx43-positive cells appear to become broader and thicker in the direction of the crista terminalis, whereas the layers of alpha SMA-positive cells become thinner and narrower. Lateral contacts between Cx43- and alpha SMA-positive cells were very sparse and only detected where the Cx43-positive strands ended (the region where alpha SMA-positive cells fill the whole space between endocardium and epicardium, ie, the putative primary pacemaker region). From these results, we conclude that the primary pacemaker is shielded from the hyperpolarizing influence of the atrium by a gradient in coupling brought about by tissue geometric factors rather than by a gradient of gap junction density.

Channelopathies: Kir2.1 mutations jeopardize many cell functions.

Andersens syndrome is caused by mutations in the potassium channel Kir2.1, a major determinant of resting membrane potential. The clinical features of this disease illustrate the importance of a stable resting membrane potential for many cell functions.