Gap26, a connexin mimetic peptide, inhibits currents carried by connexin43 hemichannels and gap junction channels.

Connexin mimetic peptides corresponding to short conserved extracellular loop sequences of connexins have been used widely as reversible inhibitors of gap junctional intercellular communication. These peptides also block movement of ATP and Ca(2+) across connexin hemichannels, i.e. hexameric channels yet to dock with partners in aligned cells and to generate the gap junction cell-cell conduit. By means of electrophysiology, we compared the effects of Gap26, a mimetic peptide corresponding to a short linear sequence in the first extracellular loop of connexin43, on connexin channel function in HeLa cells expressing connexin43. We demonstrate that Gap26 inhibited electrical coupling in cell pairs mediated by gap junctions after exposure for 30min. In contrast, Gap26 applied to single cells, inhibited hemichannel currents evoked in low Ca(2+) solution with a response time of less than 5min. The results further support the view that the likely primary and direct inhibitory effect of Gap26 is on connexin hemichannels, with gap junctions becoming inhibited later. The mechanism of action of Gap26 in blocking hemichannels and gap junction channels is discussed in the context of their different functions and locations.

Incompatibility of connexin 40 and 43 Hemichannels in gap junctions between mammalian cells is determined by intracellular domains.

Murine connexin 40 (Cx40) and connexin 43 (Cx43) do not form functional heterotypic gap junction channels. This property may contribute to the preferential propagation of action potentials in murine conductive myocardium (expressing Cx40) which is surrounded by working myocardium, expressing Cx43. When mouse Cx40 and Cx43 were individually expressed in cocultured human HeLa cells, no punctate immunofluorescent signals were detected on apposed plasma membranes between different transfectants, using antibodies specific for each connexin, suggesting that Cx40 and Cx43 hemichannels do not dock to each other. We wanted to identify domains in these connexin proteins which are responsible for the incompatibility. Thus, we expressed in HeLa cells several chimeric gene constructs in which different extracellular and intracellular domains of Cx43 had been spliced into the corresponding regions of Cx40. We found that exchange of both extracellular loops (E1 and E2) in this system (Cx40*43E1,2) was required for formation of homotypic and heterotypic conductive channels, although the electrical properties differed from those of Cx40 or Cx43 channels. Thus, the extracellular domains of Cx43 can be directed to form functional homo- and heterotypic channels. Another chimeric construct in which both extracellular domains and the central cytoplasmic loop (E1, E2, and C2) of Cx43 were spliced into Cx40 (Cx40*43E1,2,C2) led to heterotypic coupling only with Cx43 and not with Cx40 transfectants. Thus, the central cytoplasmic loop of Cx43 contributed to selectivity. A third construct, in which only the C-terminal domain (C3) of Cx43 was spliced into Cx40, i.e., Cx40*43C3, showed neither homotypic nor heterotypic coupling with Cx40 and Cx43 transfectants, suggesting that the C-terminal region of Cx43 determined incompatibility.

Formation of heterotypic gap junction channels by connexins 40 and 43.

Gap junctions formed between transfected cells expressing connexin (Cx) 40 and Cx43 (Cx43-RIN, Cx40-HeLa, and Cx43-HeLa) revealed a relationship, g(j)=f(V(j)), at steady state, that is typified by a nonsymmetrical behavior similar to that previously reported for other heterotypic channels (gap junction conductance [g(j)]; transjunctional voltage [V(j)]). The unitary conductance of the channels was sensitive to the polarity of V(j). A main state conductance of 61 pS was found when the Cx43 cell was stepped positively or the Cx40 cell negatively (V(j)=70 mV); the reverse polarities yielded a conductance of 100 pS. These heterotypic channels were permeable to carboxyfluorescein. In addition, two other heterotypic forms are illustrated to demonstrate that endogenous Cx45 expression cannot explain the results. The demonstration of heterotypic Cx40-Cx43 channels may have implications for the propagation of the electrical impulse in heart. For example, they may contribute to the slowing of the impulse propagation through the junctions between Purkinje fibers and ventricular muscle.

Electrical properties of the nexal membrane studied in rat ventricular cell pairs.

Cell pairs isolated from adult rat ventricles were used to characterize the electrical properties of the nexal membrane located between the cells. Each cell of a cell pair was connected to a suction pipette so as to enable whole-cell recordings. A double voltage-clamp method was employed which allowed the voltage gradient across the nexal membrane to be controlled. The current-voltage relationship of the nexal membrane was found to be linear over a broad range of transnexal voltages ( +/- 50 mV). The measurements revealed a mean value for the apparent nexal membrane resistance, rn(app), of 3.4 M omega. Taking into account the contribution of an uncompensated series resistance (access resistance), the effective nexal resistance, rn(eff), amounts to 1.7 M omega, approximately. The nexal membrane resistance was found to be insensitive to the sarcolemmal membrane potential, Vm (voltage range tested: -90 mV to +30 mV). The nexal membrane showed no rectifying property, i.e. it allows impulse transmission in both directions equally well. The connexons of the nexal membrane exhibited no time-dependent gating behaviour (time range investigated: 0.1-10 s).

The actions of ouabain on intercellular coupling and conduction velocity in mammalian ventricular muscle.

1. The effects of ouabain on the electrical coupling between cells and the conduction velocity, theta, were studied in ventricular muscle preparations from calf and cow hearts using a silicon-oil-chamber. 2. After 90 min of exposure to 2 X 10(-6) M ouabain, an increase of the inside longitudinal resistance, Ri, from 420 omega cm to 1032 omega CM was observed. Assuming a constant myoplasmic resistivity this presumably reflects a reduced electrical coupling between myocardial cells. 3. Concomitantly, theta was decreased from 50-3 to 29-4 cm/sec. This change could be explained by the observed alterations in the maximal rate of rise of the action potential, (dV/dt)max, the amplitude of the action potential Vp, the membrane capacity Cf, and the sum, respectively, of the inside and outside longitudinal resistance per unit distance (ri + ro). Quantitatively, about 60% of the decrease of theta could be accounted for by the experimentally determined increase of Ri. 4. Time course studies revealed a biphasic action of ouabain on Ri. An early dose-dependent drop in Ri, equivalent to an improvement of the intercellular coupling, was followed by a delayed massive increase in Ri, whose onset and magnitude were also concentration-dependent. 5. The delayed increase in Ri was associated with an increase of the diastolic tension. Toxic ouabain doses (2 X 10(-6) M) produced irreversible changes on both parameters, whereas thereapeutic doses (less than 5 X 10(-7) M) affected neither of them. Reversible effects on both parameters were observed at an intermediate drug concentration (10(-6) M). 6. The strong correlation between decoupling and contracture is consistent with the idea that the intracellular Ca concentration, [Ca]i, is involved in the control of the nexal conductance. This is supported by the finding that increasing the extracellular Ca concentration, [Ca]o, accelerated the ouabain-induced decoupling, whereas reducing [Ca]o retarded it. 7. If anything, the contracture slightly preceded the increase in Ri. From this it is concluded that the threshold [Ca]i for the electrical decoupling between cells must be somewhat larger than the threshold level for the tension activation. 8. The delayed increase in Ri is compatible with an inhibition of the Na pump which according to the Na-lag hypothesis predicts an increase of [Ca]i secondary to a Na-accumulation. The early drop in Ri can either be explained by a stimulation of the Na pump, or by a non-monotonic relationship between Ri and [Ca]i.

The permeability to tetraethylammonium ions of the surface membrane and the intercalated disks of sheep and calf myocardium.

1. Efflux experiments and longitudinal diffusion experiments were performed to determine the permeability of the surface membrane and of the nexal membrane of sheep and calf ventricular muscle to radioactive [1-(14)C]tetraethylammonium cations ([(14)C]TEA) at 25 degrees C.2. Efflux curves were analysed as the sum of three exponential processes with time constants for TEA loss of 2.7, 13.3, and 832 min.3. The first component was attributed to the exchange of extracellular free TEA, the second to extracellular, probably mucopolysaccharide bound TEA, and the third to intracellular free TEA.4. From the decay of the first phase the tortuosity factor for extracellular radial movement of TEA ions was calculated as 1.59.5. The permeability of the surface membrane to [(14)C]TEA was estimated as 6.06 x 10(-8) cm/sec.6. A tenfold change of [K](o) from 5.4 to 54 mM resulted in a 2.75-fold decrease of the permeability of the surface membrane; a fifteenfold change (5.4 to 81 mM) resulted in a 4.24-fold decrease.7. From longitudinal non-steady state diffusion experiments the permeability of the nexal membrane to [(14)C]TEA turned out to be 1.27 x 10(-3) cm/sec.8. The permeability of the nexal membrane to TEA is 21,000 times larger than that of the surface membrane. This finding is consistent with the concept of low resistance pathways between myocardial cells.9. From a comparison of the dimensions of radiopotassium, [(14)C]TEA and Procion Yellow, and the study of their diffusional properties, the pores within the nexal membrane of the intercalated disks are deduced to have a diameter of somewhat more than 10 A.

Action potential transfer in cell pairs isolated from adult rat and guinea pig ventricles.

An enzymatic procedure was used to obtain ventricular cells from adult rat and guinea pig hearts. Isolated pairs of cells were selected to study the action potential transfer from cell to cell and determine the resistance of the nexal membrane, rn. For this purpose, each cell of a cell pair was connected to a patch pipette so as to enable whole-cell, tight-seal recording. Normal impulse transmission was observed when rn ranged from 5-265 M omega. In these cases, the action potential in both cells occurred virtually simultaneously. An occasional failure in action potential transfer was seen in cell pairs whose rn had increased to 155-375 M omega. In these cases, the impulse transfer across the nexal membrane occurred with considerable delay. Impulse transfer was completely blocked once rn was larger than 780 M omega. Assuming a single connexon conductance of 100 pS, this would mean that more than 13 connexons are necessary to allow impulse transfer from cell to cell. Two single myocytes, gently pushed together, neither showed electrotonic interaction nor impulse transfer, thus rendering unlikely the possibility of an ephaptic signal transmission.

Electric current flow in cell pairs isolated from adult rat hearts.

Cell pairs were isolated from ventricles of adult rat hearts so as to study cell-to-cell coupling. Both cells of each pair were impaled with micro-electrodes connected to balanced bridge circuits. Rectangular current pulses were passed and the resulting voltage deflexions monitored. The data were analysed in terms of a delta configuration of three resistive elements, the resistances of the non-junctional membrane of cell 1 and cell 2 (rm, 1 and rm, 2), and the resistance of the nexal membrane (rn). The nexal membrane resistance was found to be insensitive to voltage gradients across the non-junctional membrane (range examined: -70 to -10 mV) and direction of current flow. The mean value of rn was 2.12 M omega ([K+]o = 12 mM). Taking into account morphological parameters, this corresponds to a specific nexal membrane resistance (Rn) of 0.1 omega cm2. Spontaneous uncoupling in which one cell remained polarized while the other one depolarized was never observed. The current-voltage relationship of the non-junctional membrane was found to be bell-shaped. The specific resistance (Rm) at the resting membrane potential (approximately -50 mV) was 3.2 k omega cm2 ([K+]o = 12 mM). Comparative studies performed on single cells revealed a similar relationship Rm versus Vm. Rm at the resting membrane potential (Vm approximately -50 mV) was 2.5 k omega cm2 ([K+]o = 12 mM). The specific capacitance of the non-junctional membrane (Cm) was determined from experiments on single cells. Cm was found to be independent of Vm (voltage range: -80 to 0 mV). The mean value of Cm was 1.66 microF/cm2 ([K+]o = 12 mM). For comparison, experiments on cell pairs and single cells were also carried out with [K+]o = 4 mM. The values obtained for Rn, Rm and Cm did not deviate significantly from those found with [K+]o = 12 mM.

Cell pairs isolated from adult guinea pig and rat hearts: effects of [Ca2+]i on nexal membrane resistance.

Cell pairs isolated from adult rat and guinea pig ventricles were used to study the resistance of the nexal membrane, rn. Each cell of a cell pair was connected to a voltage-clamp circuit to obtain simultaneous whole-cell, tight-seal recordings. With this technique, rn was determined under experimental conditions aimed at primarily modifying [Ca2+]i. Moderate changes in [Ca2+]i (produced by trains of depolarizing voltage-clamp pulses activating the slow inward current, or alterations in [Ca2+]o from 0.5 to 10 mM), resulted in no change in rn for normally coupled cell pairs (rn = 5 M omega), but small and reversible changes in slightly uncoupled preparations (rn greater than or equal to 50 M omega). Large increases in rn developed with substantial elevations in [Ca2+]i (secondary to [Na+]o-withdrawal, exposure to strophanthidin in conjunction with isi, or Ca2+-dialysis). Increases in rn brought about via elevation in [Ca2+]i always were accompanied by cell shortening consistent with a sustained contracture. The current-voltage relationship of the nexal membrane was ohmic regardless of whether rn was low (control) or elevated (after increasing [Ca2+]i).

Conductive and kinetic properties of connexin45 hemichannels expressed in transfected HeLa cells.

Human HeLa cells transfected with mouse connexin Cx45 were used to examine the conductive and kinetic properties of Cx45 hemichannels. The experiments were carried out on single cells using a voltage-clamp method. Lowering the [Ca2+]o revealed an extra current. Its sensitivity to extracellular Ca2+ and gap junction channel blockers (18alpha-glycyrrhetinic acid, palmitoleic acid, heptanol), and its absence in non-transfected HeLa cells suggested that it is carried by Cx45 hemichannels. The conductive and kinetic properties of this current, Ihc, were determined adopting a biphasic pulse protocol. Ihc activated at positive Vm and deactivated partially at negative Vm. The analysis of the instantaneous Ihc yielded a linear function ghc,inst = f(Vm) with a hint of a negative slope (ghc,inst: instantaneous conductance). The analysis of the steady-state Ihc revealed a sigmoidal function ghc,ss=f(Vm) best described with the Boltzmann equation: Vm,0= -1.08 mV, ghc,min=0.08 (ghc,ss: steady-state conductance; Vm,0: Vm at which ghc,ss is half-maximally activated; ghc,min: minimal conductance; major charge carriers: K+ and Cl-). The ghc was minimal at negative Vm and maximal at positive Vm. This suggests that Cx45 connexons integrated in gap junction channels are gating with negative voltage. Ihc deactivated exponentially with time, giving rise to single time constants, taud. The function taud = f(Vm) was exponential and increased with positive Vm (taud=7.6 s at Vm=0 mV). The activation of Ihc followed the sum of two exponentials giving rise to the time constants, taua1 and taua2. The function taua1=f(Vm) and taua2 = f(Vm) were bell-shaped and yielded a maximum of congruent with 0.6 s at Vm congruent with -20 mV and congruent with 4.9 s at Vm congruent with 15 mV, respectively. Neither taua1 =f(Vm) nor taua2 = f(Vm) coincided with taud=f(Vm). These findings conflict with the notion that activation and deactivation follow a simple reversible reaction scheme governed by first-order voltage-dependent processes.

Cell-to-cell coupling studied in isolated ventricular cell pairs.

Cell pairs isolated from adult rat and guinea pig ventricles were used to study the electrical properties of the nexal membrane. Each cell of a pair was connected to a voltage-clamp system so as to enable whole-cell, tight-seal recording. The current-voltage relationship of the nexal membrane was found to be linear, revealing a resistance rn of 2-4 M omega. rn was insensitive to the sarcolemmal membrane potential (range: -90 to +30 mV), and exerted no time-dependent gating behavior (range: 0.1 to 10 s). Lowering pHi yielded a small increase in rn. Vigorous elevations in [Ca2+]i gave rise to an increase in rn which was associated with a cell shortening. Uncoupling caused by aliphatic alcohols or halothane did not produce cell shortening. Cell pairs were also used to study action potential transfer.

Electrical properties of gap junction channels in guinea-pig ventricular cell pairs revealed by exposure to heptanol.

Cell pairs were isolated from adult guinea pig ventricles to study the electrical properties of gap junction channels. The experiments involved a double voltage-clamp approach and whole-cell, tight-seal recording. Heptanol decreased the intracellular current, In, in a dose-dependent fashion. Before complete uncoupling, In showed fluctuations suggesting the operation of gated channels. In the presence of 3 mM heptanol, In showed quantal steps arising from spontaneous opening and closing of single channels. The IV-relationship of the channels was linear (range: +/- 95 mV). Analysis of current records revealed the following single-channel conductances, gamma n: Mean value = 37 pS; median value = 33 pS. gamma n was insensitive to the non-junctional membrane potential (range: -90 to +10 mV). 3 mM ATP4- in the pipette solution had no effects on gamma n, 6 mM ATP4- produced a small decrease, and 6 mM ATP + 0.1 mM cAMP- an increase in gamma n. Channel transitions from closed to open state were variable (range of apparent time constants: 2.5-32 ms; mean: 11 ms).

Co-operativity between mouse connexin30 gap junction channels.

HeLa cells stably transfected with mouse cDNA coding for connexin30 (Cx30) were used to study the electrical properties of gap junction channels. The experiments involved the measurement of intercellular currents (Ij) from cell pairs using dual whole-cell recording with the patch-clamp method. The aim was to compare Ij from cell pairs whose gap junctions consisted of a single channel and cell pairs whose gap junctions consisted of many channels. We found that both the ensemble average currents gained from single-channel records and the currents obtained from multichannel records inactivated exponentially with time. However, the former inactivated significantly slower than the latter. At ajunctional voltage (Vj) of 50 mV, the time constants of inactivation (tau(i)) were 8.1 s and 1.6 s, respectively. Moreover, the ratio tau(i)(single-channel)/tau(i)(multichannel) turned out to be voltage sensitive, i.e. it decreased with increasing V(j) These observations suggest that the operation of Cx30 gap junction channels in the multichannel configuration involves co-operative interactions.

Voltage gating of Cx43 gap junction channels involves fast and slow current transitions.

RIN cells transfected with mouse cDNA coding for connexin43 (Cx43) were used to further examine the electrical properties of single gap junction channels. The experiments involved measuring intercellular currents from cell pairs using dual whole-cell recording with the patch-clamp method. We found that the single-channel currents exhibit two types of transitions and several conductance states. Besides fast transitions between the main open state and the residual state, the channels underwent slow transitions between an open state (i.e. main open state or residual state) and a closed state. The fast transitions lasted less than 2 ms, the slow ones ranged from 3.5 to 145 ms. The incidence of slow transitions increased with increasing transjunctional voltage. These observations are consistent with the notion that Cx43 gap junction channels possess more than one mechanism of voltage gating.

Electrical properties of gap junction hemichannels identified in transfected HeLa cells.

Human HeLa cells transfected with mouse connexin Cx30, Cx46 or Cx50 were used to study the electrical properties of gap junction hemichannels. With no extracellular Ca2+, whole-cell recording revealed currents arising from hemichannels. Multichannel currents showed a time-dependent inactivation sensitive to voltage, Vm. Plots of the instantaneous conductance, ghc,inst, versus Vm were constant; plots of the steady-state conductance, ghc,ss, versus Vm were bell-shaped. Single-channel currents showed two conductances, gammahc,main and gammahc,residual, the latter approximately or approximately equals=1/6 of the former. Single-channel currents exhibited fast transitions (1-2 ms) between the main state and residual state. Late during wash-in and early during wash-out of 2 mM heptanol, single-hemichannel currents showed slow transitions between an open state and closed state. The open channel probability, Po, was Vm-dependent. It declined from approximately =1 at Vm= 0 mV to 0 at large Vm of either polarity. Hemichannel currents showed a voltage-dependent gammahc,main, i.e., it increased/decreased with hyperpolarization/depolarization. Extrapolation to Vm=0 mV led to a gammahc,main of 283, 250 and 352 pS for Cx30, Cx46 and Cx50, respectively. The hemichannels possess two gating mechanisms. Gating with positive voltage reflects Vj-gating of gap junction channels, gating with negative voltage reflects a property inherent to hemichannels, i.e., Vm or "loop" gating. We conclude that Cx30, Cx46 and Cx50 form voltage-sensitive hemichannels in single cells which are closed under physiological conditions.

Connexin43 gap junctions exhibit asymmetrical gating properties.

A communication-deficient cell line (RIN cells, derived from a rat islet tumour), stably transfected with cDNA coding for rat connexin43 (Cx43), was chosen to further assess the mechanism of voltage gating of Cx43 gap junction channels. The experiments were carried out on preformed cell pairs using a dual whole-cell, voltage-clamp method. The junctional current, Ij, revealed a time- and voltage-dependent inactivation at transjunctional voltages Vj>+/-40 mV. When an asymmetrical pulse protocol was used (in cell 1 the holding potential was maintained, in cell 2 it was altered to establish a variable Vj), the channels exhibited an asymmetrical gating behaviour: Vj,0=-73.7 mV and 65.1 mV for negative and positive Vj, respectively (Vj at which Ij is half-maximally inactivated); gj(min)=0.34 and 0.29 (normalized minimal conductance); tau = 350 ms and 80 ms at Vj=100 mV (time constant of Ij inactivation). Hence, these parameters were more sensitive to positive Vj values. When a symmetrical pulse protocol was used (the holding potentials in cell 1 and cell 2 were altered simultaneously in steps of equal amplitude but of opposite polarity), the Vj -dependent asymmetries were absent: Vj,0=-60.5 and 59.5; gj (min)=0.27 and 0.29; tau =64 ms and 47 ms at 100mV. Putative explanations for these observations are discussed. A possibility is that the number of channels alters with the polarity of Vj.

Modification of gap junction conductance by divalent cations and protons in neonatal rat heart cells.

Myocytes were isolated from neonatal rat hearts and grown in culture dishes. Pairs of cells were selected to study the effect of divalent cations and protons on the conductance of gap junctions, gj. The experimental approach involved the dual voltage-clamp method and cell dialysis via patch pipette, i.e. gj was monitored while the cytosolic level of Ca2+, Mg2+, Sr2+, Ba2+ or H+ was modified in one of the cells. A dose-dependent decrease in gj developed when pCa of the pipette solution was lowered (range: pCa = 7.7-2.42, equivalent to a [Ca2+] of 20 nM-3.8 mM). The gj/pCa-relationship revealed a Hill coefficient n of 0.87 and a half-maximal concentration pKCa of 3.5. Pretreatment with 3 mM NiCl2 and 1 micron ryanodine to minimize the removal of cytosolic Ca2+ did not significantly affect the response to gj. Similarly, gj was decreased in a dose-dependent fashion when pHi in the pipette solution was lowered (range: pH = 7.2-5.0, corresponding to a [H+] of 63 nM-10 microns). The gj/pH-relationship yielded an n of 0.92 and a pKH of 5.85. Pretreatment with 1 mM amiloride to minimize the extrusion of protons enhanced the effects of pH on gj. Simultaneous alterations in pCa and pH demonstrated an additive type of action of Ca2+ and H+ on gj. This is consistent with the existence of two types of sensors which contribute separately to the functional state of gj. No significant decrease in gj was detectable when the pipette solution contained Mg2+ or Ba2+ (up to 5 mM). Partial uncoupling was observed with pipette solution containing 5 mM Sr2+. We conclude that gj of neonatal and adult cardiomyocytes exhibit different ionic sensitivities. This discrepancy may reflect differences in connexin expression and/or molecular intermediates involved in regulating gj.