1095 ATP Directly Affects Junctional Conductance Between Paired Ventricular Myocytes Isolated From Guinea Pig Heart Hiroki Sugiura, Junji Toyama, Naoya Tsuboi, Kaichiro Kamiya, and Itsuo Kodama Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 Effects of ATP on junctional conductance (gj) were investigated in paired ventricular myocytes isolated from guinea pig hearts. One cell of the pair was voltage-clamped with a single-patch pipette, and gj was measured after the perforation of the nonjunctional membrane of the partner cell. The current-voltage relation of gj was linear between -30 and +30 mV. The control gj at 5.0 mM ATP in 88 pairs of cells ranged from 100 to 1,055 nS (average, 268 nS). ATP within the range from 0.1 to 5.0 mM increased gj in a dose-dependent manner. The Hill coefficient was 2.6, and the half-maximum effective concentration of ATP was 0.68 mM. Adenylylimidodiphosphate (2 mM) caused a transient increase in gj in the presence of 0.5 mM ATP, but forskolin (30 ,uM), cyclic AMP (50 ELM), catalytic subunit of cyclic AMP-dependent protein kinase (1 ,uM), and ADP (10 mM) had no significant effect on gj. The temperature coefficient of gj in the presence of 5.0 mM ATP was 1.29. These findings suggest that gj in paired ventricular myocytes is directly regulated by ATP probably through a specific ligand-receptor interaction between ATP and gap junctional channel protein. (Circulation Research 1990;66:1095-1102) P rimary effects of ischemia on cardiac muscle include decreases of ATP and creatine phosphate within affected fibers. These decreases result in a depression of active transport systems, which elevates intracellular concentrations of Na+, Ca2+, and protons.1 Previous studies2-4 have shown that electrical coupling was suppressed when energy metabolism was inhibited by various experimental conditions, such as hypoxia or ischemia, or in the presence of metabolic inhibitors. In some cases, uncoupling effects have been shown to be reversed by intracellular injection of ATP2 or EDTA3 or by addition of glucose to the extracellular perfusate.4 Abundant studies indicate that increases in intracellular concentrations of Na+, Ca2 , or protons5-7 cause cell decoupling, and it has been assumed that uncoupling effects of ischemia are due to changes in these ions. It has also been reported that in various tissues junctional conductance (gj) is either increased8-10 or decreased1l-13 as a consequence of From the Department of Circulation and Respiration, The Research Institute of Environmental Medicine, Nagoya University, Nagoya, Japan. Supported by Grants-in-Aid for Scientific Research (01624004 and 63480226) from the Ministry of Education, Science and Culture of Japan. Address for reprints: Hiroki Sugiura, MD, Department of Circulation and Respiration, The Research Institute of Environmental Medicine, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-01, Japan. Received August 16, 1989; accepted November 20, 1989. cyclic AMP (cAMP)-mediated phosphorylation of the channel protein. Recent advances in cardiac electrophysiology have proved that some ionic channels in sarcolemmal membrane can be regulated by ligand-receptor interaction between ATP and channel proteins14 or by phosphorylation of channel proteins.15 In the present study, we investigated the direct effect of ATP on cardiac gj by using the whole-cell clamp of paired ventricular myocytes, one of which was perforated to allow intracellular diffusion with extracellular solutions.7 We report here that gj is strongly affected by ATP from 0.1 to 5.0 mM and that this effect is direct and, thus, analogous to the effect reported on other sarcolemmal channels. Materials and Methods Preparations Enzymatic procedure for the isolation of paired ventricular cells was modified from the method as described by Mitra and Morad.16 Briefly, guinea pigs weighing 150-250 g were killed by an overdose of pentobarbital sodium. The heart was quickly removed, and the aorta was immediately cannulated for Langendorff perfusion. Subsequently, the coronary artery was perfused backward with Ca2`-free Tyrode's solution for 3-4 minutes. Thereafter, the heart was perfused by 50 ml Ca2+-free Tyrode's solution containing 50 mg collagenase (type I, Sigma Chemical, St. Louis, Missouri) and 10 mg protease 1096 Circulation Research Vol 66, No 4, April 1990 TABLE 1. Composition of Internal Solutions Na2ATP MgCl2 Aspartic acid CsOH EGTA HEPES TEA Na2CRP 5.0 3.0 110 110 4 5 20 Pipette (mM) 5 Test solution(mM) 5.0 1 (control) 3.0 110 110 4 5 20 0 1.5 1.5 110 110 4 2 5 20 0 110 110 4 5 1.0 20 3 1.0 0 0.5 0.5 110 4 110 4 5 20 0 0.1 5 0.1 110 110 4 5 20 0 TEA, tetraethylammonium chloride; CRP, creatine phosphate. pH of internal solutions was adjusted to 7.2 with CsOH. ATP concentrations in internal test solutions 2-5 are varied to 1.5, 1.0, 0.5, and 0.1 mM, respectively. In each of the test solutions, the concentration of MgCl2 was varied to keep free-Mg2' concentration less than 0.3 mM. Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 (type XIV, Sigma Chemical) for 3-10 minutes. Then, the collagenase solution was washed out by perfusing with high-K' and low-Cl- solution (KB medium).'7 At the end of the above procedure, the ventricle was cut into pieces and was stored in the KB medium at 40 C before use. Cells were dispersed in the recording chamber (0.1-0.2 ml) on the stage of an inverted phase-contrast microscope (Diaphoto-TMD, Nikon, Japan) and were continuously perfused with normal Tyrode's solution (1.8 mM Ca2) at a rate of 2-3 ml/min. About 50% of the cells were rod-shaped in normal Tyrode's solution. Paired cells were distinguished from single cells by the presence of contact regions between apposed cells. Solutions Normal Tyrode's solution consisted of (mM) NaCl 136.9, NaH2P04 0.33, KCI 5.4, CaCl2 1.8, MgC12 0.5, glucose 5, and HEPES 5; pH was adjusted to 7.4 with NaOH. Internal solutions are listed in Table 1. Ca2free Tyrode's solution was prepared by omitting CaCi2 from normal Tyrode's solution. Pipette solution was essentially the same as the internal test solution (test solution 1) except that creatine phosphate was added as a reservoir of high-energy phosphates. Internal test solutions 1-5 were used to perfuse the perforated cell interior at various ATP concentrations. To keep free-Mg2+ concentration constant in the internal test solution, the amount of MgCl2 was varied between 0.1 and 3.0 mM in proportion to ATP concentrations. Free-Ca2+ and -Mg2+ concentrations in each of internal test solutions, which had been estimated with the stability constants for ATP, Ca2+, Mg2+, and EGTA at a given pH,1819 were limited to a level less than 1 nM and 0.3 mM, respectively. Noma and Tsuboi7 showed that these Ca2+ and Mg2+ concentrations had no effects on gj. Forskolin, cAMP, and adenylylimidodiphosphate were directly dissolved in the internal test solutions. ADP was dissolved in the internal test solution containing 10 ,uM Pl,P5-di(adenosine-5') pentaphosphate (AP5A), a potent inhibitor of adenosine kinase. Catalytic subunit of cAMP-dependent protein kinase (c-subunit) was dissolved in the internal test solution containing 32.4 mM DL-dithiothreitol to prevent the inactivation of c-subunit. In preliminary experiments, AP5A and DL-dithiothreitol had no effect on gj (results not shown). All experiments were carried out at 34°-35° C unless otherwise stated. Electrophysiological Experiments The method for measuring gj and the intracellular perfusion technique were essentially similar to those as described by Noma and Tsuboi7 except that the whole-cell clamp was carried out by using a singlepatch pipette instead of the double-pipette method. In brief, one cell of the pair was voltage-clamped with a giga-sealed patch pipette filled with pipette solution (Table 1). After clamping the membrane potential of a paired cell at approximately 0 mV, bath solution was switched from normal Tyrode's solution (1.8 mM Ca'+) to a Cs+-rich and low-Ca'+ internal test solution (test solution 1). Subsequently, the nonjunctional membrane of the partner cell was perforated by crushing the tip of the other pipette against the cell. This procedure allowed access of bath solution to the junctional channel of the perforated cell. Under this condition, current from the clamped cell flows through two pathways. One is through the nonjunctional membrane of the clamped cell to the ground. The other is through the gap junction and the interior of the perforated cell to the ground. The equivalent circuit after perforation of the nonjunctional membrane of the partner cell is illustrated in Figure 1. The value of input conductance (gi) through the electrode after cell membrane perforation was given by gi = gml + gj x (gm2 + gs)/(gj +gm2 + gs) where g, is the conductance through the hole made in the partner cell and g., and g.2 are the nonjunctional membrane conductances of the clamped cell and the perforated cell, respectively. Since g, is much greater than gj and g,2 the approximate value of gj was obtained by gj=gi-gmi The value of gm, was approximated as half of the total nonjunctional membrane conductance of a paired cell (gml+gm2), which had been measured before perforation of the nonjunctional membrane of the partner cell. Series resistance of the electrode Sugiura et al Effects of ATP on Gap Junction were considered to be negligible, because the negative pressure of pipette and the input resistance under control condition at 5.0 mM ATP were kept constant. Paired cells having gjs of larger than 100 nS were used in our experiments. When membrane leak conductance increased noticeably within 3 minutes after cell membrane perforation, the cells were discarded. Ramp command potential between -30 and +30 mV with a duration of 2.5 seconds was applied to the clamped cell every 6 seconds. gi was calculated from the slope conductance of the current output. Values were presented in mean+SEM unless otherwise stated. Statistical analysis was performed by Student's paired t test, and values of p<0.05 were considered to indicate a significant difference. V rplp gml Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 Results Control g, Figure 2 illustrates changes of gi after the perforation of the nonjunctional membrane of the partner cell. Current output in response to the command potential of +30 mV was increased suddenly after the perforation (Figure 2A). The experiment was carried out at 5.0 mM ATP in the bath solution. Therefore, the average value of gi was increased from 47±3 to 292±15 nS (n=88) (Figure 2B). The current-voltage relations before and after the perforation were linear over the range from -30 to +30 mV (Figure 2C), indicating no voltage dependency of junctional as well as nonjunctional membrane conductances under such Cs+-rich and low-Ca21 conditions. The control g. was calculated to be from 100 to 1,055 nS (average, 268±15 nS). Effects of ATP Reduction on gj We measured gi with internal test solutions containing various ATP concentrations (0.1-5.0 mM). gap junction FIGURE 1. Diagrams showing voltage-clamp circuit and the equivalent circuit of the open-cell system. One cell of the pair was voltage-clamped with a patch pipette. I, current; V, voltage; s.r.c., series resistance compensation; rpip, resistance of the pipette; gml and gm2, the conductances of the clamped cell (1) and the perforated cell (2), respectively; gj, the conductance of the junctional membrane; gs, the conductance of the hole made in the partner cell. Diagram of the paired cell is shown on the right. (1-5 Mfl) was always 1% or less of input resistance (the parallel sum of the seal and nonjunctional membrane resistances) and was compensated by the series resistance compensation circuit in the amplifier. Access resistance of the electrode produced by cellular materials plugging the pipette was not compensated. However, changes of access resistance A llllll 11 llllll C +30mV l l I ll l llll I (nA) 20 nA a -~ -50 ro- 0 U) c --20 A 1 min b -30 1 min B 1097 FIGURE 2. Tracings and graphs showing change of input conductance (gi) after cell membrane perforation. Panel A: Original ramp voltage-clamp pulse between -30 and +30 mV and current output after the perforation. Closed triangle (A) indicates the time of cell membrane perforation. Panel B: Change of gi after cell membrane perforation. gi was sudV(mV) denly increased from 14 to 501 nS by the perforation and '° became stable in 1 minute. Panel C: Current (I)-voltage (V) relation before and after cell membrane perforation. Tracings obtained at the times indicated by 'a' and 'b' in panel A are illustrated. I-V relations obtained before and after the perforation were linear between -30 and +30 mV Circulation Research Vol 66, No 4, April 1990 1098 ATP (mM)[- 0.5 5 1 5 5 0.1 (3) 100 (22) 300- 000 isvU) 200 - . ',-0%-~.0 7. 00 (D 0 a 5 ZW (8) 100- 0- 0 r 5 10 15 ATP Concentration (-log(M)) Time (min) Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 FIGURE 3. Graph showing effects of ATP on junctional conductance (gj). ATP concentration in the bath solution was changed from 5.0 to 0.5, 1.0, and 0.1 mM as indicated at the top. In this experiment, gj in the presence of 5.0 mMATP was 203 nS. When ATP concentration was decreased from 5.0 mM to 1.0, 0.5, and 0.1 mM, gj decreased to 123, 46, and 34 nS, respectively. The ATP-dependent change in gj was reversible. Ordinate shows gj; abscissa shows the time after the initiation of the whole-cell clamp. Our preliminary experiments have shown that with potassium channels blocked by cesium and tetraethylammonium chloride, g,, values of ventricular myocytes were unaffected by similar changes in ATP concentrations of the bath solution (results not shown). Therefore, gj was calculated by assuming that the initial g,, value was kept constant throughout the experiment. Representative results are shown in Figure 3. When ATP concentration was reduced from control (5.0 mM) to 1.0, 0.5, or 0.1 mM, gj decreased in a concentration-dependent manner. Linear currentvoltage relation of gi indicating no voltage dependency of gj was maintained even at the lower ATP concentrations. At the resumption of control ATP (5.0 mM), gj returned toward the control level. In 18 of 40 experiments (45%), half or more gj recovery was not attained at the resumption of 5.0 mM ATP. Such irreversible data were discarded, and the remaining 22 experiments were used for data analysis. The irreversible decoupling is most likely explained by mechanical breakdown of gap junctional structure due to rigor induced by ATP depletion. Figure 4 summarizes the results obtained from 22 pairs of myocytes. Values of gj measured at each ATP concentration were normalized with reference to the average of those at 5.0 mM ATP before and after the reduction of ATP. The smooth curve is a least-squares fit of the Michaelis-Menten equation to the plotted dose-response data. The Hill coefficient was 2.6, and the half-maximum effective concentration of ATP was 0.68 mM. The calculated Michaelis-Menten curve suggests that gj would reach 0 nS at ATP concentrations less than 0.1 mM. However, we were not able to determine the exact ATP level causing complete cell FIGURE 4. Graph showing relation between ATP concentration and normalized junctional conductance (Gj). Ordinate shows the percentage of Gj with reference to the value measured at 5.0 mMA TP in each experiment. Abscissa shows ATP concentration in the bath solution in logarithmic scale. Data are presented as mean ±SEM. Closed circle shows the control value in the presence of 5.0 mM ATP. Figures in parentheses are numbers of experiments. The smooth curve is a least-squares fit of the Michaelis-Menten equation to the plotted data. The Hill coefficient was 2.6, and the halfmaximum effective concentration ofA TP was calculated to be 0.68 mM decoupling because most experiments under such conditions showed irreversible gj reduction. The above-mentioned results have revealed that gj is directly regulated by ATP under the condition in which free-Ca21, -Mg2+, and -proton concentrations are kept constant. The following experiments were designed to clarify possible underlying mechanisms for the regulation. Effects of Forskolin, cAMP, and c-Subunit on g, The ATP-dependent change in gj could be explained by phosphorylation of gap junctional channel proteins through activation of cAMP-dependent protein kinase. If such a mechanism is responsible for the opening of gap junctional channels, gj would be increased by stimulation of adenylate cyclase in the cell membrane or by an elevation of cAMP or c-subunit in the cytoplasm. This prediction was tested by applying forskolin (30 ,uM), cAMP (50 ,uM), or c-subunit (1 ,uM) to internal test solutions containing 5.0 mM or 0.5 mM ATP. Application of these three substances for 5 minutes in the medium containing 5.0 mM ATP had no significant effect on gj (Table 2). Figure 5 shows effects of forskolin (30 ,uM) and cAMP (50 ,uM) applied for 3 minutes in the low ATP (0.5 mM) medium. These substances did not reverse gj values, which had been decreased by the ATP reduction (Table 3). These findings may indicate that opening of gap junction by ATP is not mediated by cAMPdependent phosphorylation of channel proteins. Sugiura et al Effects of ATP on Gap Junction TABLE 2. Change in Junctional Conductance in the Presence of 5.0 mM ATP TABLE 3. Change in Junctional Conductance in the Presence of 0.5 mM ATP gj (nS) gj (nS) After Gj (%) Before 94±5 247±38 233±28 Forskolin (n=5) 97±2 236±27 225±23 Cyclic AMP (n= 13) 89±6 334±62 284±44 c-Subunit (n=7) 99±5 Adenylylimidodiphosphate (n=5) 195±21 192±18 168±14 169±17 102±8 ADP (n=5) Values are mean±SEM. gj, junctional conductance; Gj, normalized gj; c-subunit, catalytic subunit of cyclic AMP-dependent protein kinase. Values of Gj were normalized by the value of gj measured just before the addition of each drug in the presence of 5.0 mM ATP. Effects of Adenylylimidodiphosphate and ADP on gj Some sarcolemmal channels have been reported to be regulated by the ligand-receptor interaction Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 A ATP(mM) | 5 0.5 l S 1~~~~~~~~~~~~ 130pm forakow 5001 0.-.0- 0 C c a . .......'. . 0] O ATP(mbo)l 5 1 5 10 Q.5 5 I5#.M CAMP 300 1 ..~~ U) C c 0 15~0 10 5 Before After 36+16 27+9 Gj (%) 102+29 54±33 Forskolin (n=3) 63+11 42+11 Cyclic AMP (n=9) Adenylylimidodiphosphate (n=7) 70+ 11 112±22 156+19* 25±6 82+27 42±21 ADP (n=4) Values are mean±SEM. gj, junctional conductance; Gj, normalized gj. Values of Gj were normalized by the value of gj measured just before the addition of each drug in the presence of 0.5 mM ATP. *Significantly different from the value of gj measured before the addition of drugs atp<0.05. between ATP and channel proteins.14 The ATPdependent change in gj might be explained by a similar mechanism. We tested this alternative possibility by using nonhydrolyzable ATP analogue (adenylylimidodiphosphate) and ADP. Application of adenylylimidodiphosphate (2 mM) or ADP (10 mM) to the internal bath solution containing 5.0 mM ATP did not affect gj (Table 3). Figure 6 shows experiments in the low (0.5 mM) ATP medium. Application of adenylylimidodiphosphate (2 mM) caused a partial and transient reversal of gj that had been decreased by the ATP reduction from 5.0 to 0.5 mM. Peak gj values during adenylylimidodiphosphate application were 56% larger than the pretreatment level at 0.5 mM ATP (Table 3). However, the increment of gj was not maintained and dissipated spontaneously within several minutes (Figure 6A). Application of ADP (10 mM) at 0.5 mM ATP did not cause such a partial reversal of gj (Figure 6B, Table 3). Effects of Temperature on g, Effects of temperature on gj were examined in the control (5.0 mM) ATP medium. As the bath temperature was decreased from 350 to 28° and 240 C, gi decreased in steps, and the change was reversible (Figure 7). The temperature coefficient (Q1o) in 10 experiments was 1.29. I ~ ... ~~ ,~~~~~~~~~~~ , r 1099 B 0J Time (min) FIGURE 5. Panel A: Graph showing effects of forskolin on junctional conductance (gj). A reduction ofATP from 5.0 to 0.5 mM resulted in a decrease of gj from 452 to 42 nS. Application of forskolin (30 MM) for 3 minutes caused no significant change in gj. Panel B: Graph showing effects of cyclic AMP (cAMP) on gj. A reduction ofATP from 5.0 to 0.5 mM resulted in a decrease ofgj from 262 to 62 nS. Application of cAMP (50 MM) for 3 minutes caused no significant change in gj. For both panels, ordinate shows gj, and abscissa shows the time after the initiation of the whole-cell clamp. Discussion In the present experiments, we measured gj of ventricular cell pairs isolated from adult guinea pig hearts through the use of a single voltage-clamp approach. The current-voltage relation of junctional membrane was linear between -30 mV and +30 mV, indicating that gj has no voltage dependency under such Cs+-rich and low-Ca2' conditions. This finding is in agreement with previous studies on adult ventricular cell pairs of guinea pig or rat.7,20,21 Recently, Rook et a122 have shown that gj of neonatal rat heart pairs exhibits marked voltage and time dependency when cell coupling is inhibited by a decrease of the number of channels involved. However, the voltage dependency is not observed until the transjunctional potential exceeds ±50 mV. Therefore, our data are 1100 . Circulation Research Vol 66, No 4, April 1990 0.5 5 ATP(mM) 500 l A 5 | 2mM AMP-PNP Temp(TC) |35 1\ 400 r"**,"% 00 28 1 33.2 24 1 1 0111 0 0 0-~~~~~~~~~ o~~~~~~~~~ \Ito. A 0 5 10 0-J 1T Time (min) ATP(mM) 1 0.5 5 B _- A 5 150-1 100- 1.0- ~rlpr.'O.. 0.5- Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 Q1o=1.29 50 0 * 0- .1 6 B 0 0 itw^wo-4c i0 1'5 Time (min) FIGURE 6. Panel A: Graph showing effects of adenylylimidodiphosphate (AMP-PNP) on junctional conductance (gj). A reduction ofA TPfrom 5.0 to 0.5 mM resulted in a decrease of gj from 432 to 123 nS. Application of 2 mM AMP-PNP caused a partial and transient reversal of gj from 123 to 192 nS. This reversal was not maintained and dissipated spontaneously during application of AMP-PNP. Panel B: Graph showing effects ofADP on gj. A reduction ofATP from 5.0 to 0.5 mM caused a decrease of gj from 105 to 9 nS. Application of 10 mMADP for 3 minutes caused no significant change in gj. For both panels, ordinate shows gj, and abscissa shows the time after the initiation of the whole-cell clamp. also in agreement with Rook et a122 regarding the voltage range examined. We estimated gj values by assuming that nonjunctional membrane conductance of the voltageclamped cell (g.,) is one half of the input conductance before perforation of the other cell of the pair (gm,+gm2). There is no evidence indicating that gm, is similar to g.2 and that gm, is kept constant throughout the experiments even after perforation. Accordingly, such an assumption might introduce variable errors in estimating the precise amount of gj values. The potential error is considered to be minimal under the control condition in which gm, is less than 10% of gi, but it may increase as gi declines. Although this would limit the quantitative accuracy, we nevertheless think that it does not invalidate our conclusion. Our control gj values in ventricular cell pairs (average, 268 nS) are more or less similar to those reported by Kameyama (710 nS in guinea pig)23 and those by Weingart and Maurer (250 nS in guinea pig 0.1 - 25 30 Temperature (C) 35 FIGURE 7. Panel A: Graph showing temperature-dependent change ofjunctional conductance (gj) in the presence of 5.0 mM ATP. Temperature of the bath solution was changed within the range from 240 to 350 C, as indicated at the top. Ordinate shows gj; abscissa shows the time after the initiation of the whole-cell clamp. In this experiment, gj at 350 C was 294 nS. When temperature of the bath solution was varied to 28', 240, and 33.20 C, gj decreased to 248, 174, and 279 nS, respectively. Panel B: Graph showing relation of normalized junctional conductance (Gj) and temperature. Gj, which was normalized with the value of gj measured at 35° C at 5.0 mM ATP, was plotted in logarithmic scale against the temperature. Temperature coefficient (Q10) was 1.29. and 590 nS in rat).21,24 Noma and Tsuboi7 and Wittenberg et a125 showed somewhat higher control values (1,000 nS in guinea pig, 1,250-2,530 nS in rat). Such discrepancy might be attributed to different methods for cell isolation and for gj measurement or to different species employed. In the present experiments, gj decreased in a concentration-dependent manner when intracellular ATP concentration was reduced to less than 1.5 mM under the condition where free-Ca21, -Mg2+, and -proton concentrations were kept constant. This ATP-dependent change in gj was reversible. The half-maximum effective concentration of ATP was 0.68 mM. Intracellular ATP concentration of intact cardiac myocytes under physiological conditions is kept at a level ranging from 3.0 to 7.5 mM.26-28 Under some pathological conditions in which energy metabolism is inhibited, a depletion of creatine phosphate is followed by a significant decrease in ATP. Recent studies have shown that intracellular ATP concentration was decreased by 10-90% from control when Sugiura et al Effects of ATP on Gap Junction Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 cardiac muscles were treated by ischemia or hypoxia for 10-25 minutes.26-28 ATP-dependent active ionic transport through sarcolemma and through membranes of intracellular organelles is inhibited under such conditions leading to an accumulation of Ca21 and H' in the cytosol.2-4 Cell decoupling induced by ischemia or hypoxia has been mainly attributed to an elevation of these intracellular cations.2-4 The present data suggest that the ATP depletion may also contribute to cell decoupling. As to the effect of cAMP on cell coupling, there are some discrepancies among previous reports. In horizontal cells of the turtle retina"112 or fish retina,'3 the permeability of gap junction was shown to be inhibited by an increase of intracellular cAMP. In contrast, De Mello9 demonstrated that intracellular injection of cAMP into cardiac myocytes enhanced cell-to-cell coupling. Burt and Spray'0 have recently shown the similar improvement of cardiac cell coupling byr an elevation of intracellular cAMP under low-Ca + conditions. In addition, Wojtczak29 has shown that cAMP had no effects on cell coupling under physiological conditions but inhibited cell coupling during calcium-overload conditions, such as hypoxia. cAMP is known to affect many aspects of cell functions including energy metabolism, contraction, and intracellular ionic concentrations. An elevation of cAMP in these previous studies may have affected gj through various secondary changes in such cell functions. The present results seem to indicate that cAMP has no direct action on cardiac gj, provided intracellular ATP, Mg2+, Ca2+, and H+ are kept constant. Recently, Pressler and Hathaway30 showed that the isolated heart gap junctional protein is phosphorylated by cAMP-dependent protein kinase like liver or lens gap junctional protein.8,3' However, it is not clear whether phosphorylation of gap junctional protein increases or decreases cell coupling. In the present study, gj was not affected by cAMP. This may indicate that the open state of the gap junctional channel is not maintained by phosphorylation of the channel protein. Adenylylimidodiphosphate caused a partial and transient reversal of gj that had been depressed by the ATP reduction. Such a transient reversal of gj was not induced by ADP. These findings indicate that nonhydrolyzable ATP analogue (adenylylimidodiphosphate), which has a similar molecular structure to ATP, could partially substitute for ATP. Accordingly, we suggest that gj is maintained by ATP through the ligand-receptor interaction between ATP and gap junctional channel protein. However, the partial reversal of gj by adenylylimidodiphosphate was not maintained and dissipated spontaneously within several minutes. Further experimental studies will be required to clarify the underlying mechanism for this phenomenon. Our data showed that Qlo of the gap junction was 1.29, in agreement with experiments on the junction 1101 of earthworm septate median giant axon.32 This Qlo value is comparable with that of aqueous diffusion of ions (Q1o= 1.3) and is much lower than that of enzyme reactions (Q,0=3), such as phosphorylation regulated by high-energy phosphates.33 Such a low Qlo value seems to support our proposal that the regulation of gap junctional channel is mediated by ligand-receptor interaction between ATP and channel proteins, because it would require minimal energy consumption. In summary, ATP regulates cell coupling between paired ventricular myocytes independently of intracellular free Ca2+, Mg2+, and protons. We suggest that ATP-dependent change in gj is mediated by the ligand-receptor interaction between ATP and gap junctional channel protein. Acknowledgments We express our appreciation to Dr. David C. Spray for his useful suggestions and comments in preparing this manuscript. References 1. Baumgarten CM, Cohen CJ, McDonald TF: Heterogeneity of intracellular potassium activity and membrane potential in hypoxic guinea pig ventricle. Circ Res 1981;49:1181-1189 2. Politoff AL, Socolar SJ, Loewenstein WR: Permeability of a cell membrane junction: Dependence on energy metabolism. J Gen Physiol 1969;53:498-515 3. De Mello WC: Effect of 2-4-dinitrophenol on intercellular communication in mammalian cardiac fibres. Pflugers Arch 1979;380:267-276 4. Wojtczak J: Contractures and increase in internal longitudinal resistance of cow ventricular muscle induced by hypoxia. Circ Res 1979;44:88-95 5. De Mello WC: Influence of the sodium pump on intercellular communication in heart fibres: Effect of intracellular injection of sodium ion on electrical coupling. J Physiol (Lond) 1976; 263:171-197 6. De Mello WC: Effect of intracellular injection of calcium and strontium on cell communication in heart. J Physiol (Lond) 1975;250:231-245 7. Noma A, Tsuboi N: Dependence of junctional conductance on proton, calcium and magnesium ions in cardiac paired cells of guinea-pig. J Physiol (Lond) 1987;382:193-211 8. Saez JC, Spray DC, Nairn AC, Hertzberg E, Greengard P, Bennett MVL: cAMP increases junctional conductance and stimulates phosphorylation of the 27-kDa principal gap junction polypeptide. Proc Natl Acad Sci USA 1986;83:2473-2477 9. De Mello WC: Effect of intracellular injection of cAMP on the electrical coupling of mammalian cardiac cells. Biochem Biophys Res Commun 1984;119:1001-1007 10. Burt JM, Spray DC: Inotropic agents modulate gap junctional conductance between cardiac myocytes. Am J Physiol 1988; 254:H1206-H1210 11. Piccolino M, Neyton J, Gerschenfeld HM: Decrease of gap junction permeability induced by dopamine and cyclic adenosine 3': 5'-monophosphate in horizontal cells of turtle retina. J Neurosci 1984;4:2477-2488 12. Lasater EM, Dowling JE: Dopamine decreases conductance of the electrical junctions between cultured retinal horizontal cells. Proc Natl Acad Sci USA 1985;82:3025-3029 13. Teranishi T, Negishi K, Kato S: Dopamine modulates Spotential amplitude and dye-coupling between external horizontal cells in carp retina. Nature 1983;301:243-246 14. Taniguchi J, Noma A, Irisawa H: Modification of the cardiac action potential by intracellular injection of adenosine triphosphate and related substances in guinea pig single ventricular cells. Circ Res 1983;53:131-139 1102 Circulation Research Vol 66, No 4, April 1990 Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 15. Kameyama M, Hofmann F, Trautwein W: On the mechanism of 8-adrenergic regulation of the Ca channel in the guinea-pig heart. Pflugers Arch 1985;405:285-293 16. Mitra R, Morad M: A uniform enzymatic method for dissociation of myocytes from hearts and stomachs of vertebrates. Am J Physiol 1985;249:H1056-H1060 17. Isenberg G, Klockner U: Calcium tolerant ventricular myocytes prepared by preincubation in a "KB Medium." Pflugers Arch 1982;395:6-18 18. Fabiato A, Fabiato F: Calculator programs for computing the composition of the solutions containing multiple metals and ligands used for experiments in skinned muscle cells. J Physiol (Paris) 1979;75:463-505 19. Blinks JR, Wier WG, Hess P, Prendergast FG: Measurement of Ca21 concentrations in living cells. Prog Biophys Mol Biol 1982;40:1-114 20. 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 21. Weingart R: Electrical properties of the nexal membrane studied in rat ventricular cell pairs. J Physiol (Lond) 1986; 370:267-284 22. Rook MB, Jongsma HJ, van Ginneken ACG: Properties of single gap junctional channels between isolated neonatal rat heart cells. Am J Physiol 1988;255:H770-H782 23. Kameyama M: Electrical coupling between ventricular paired cells isolated from guinea-pig heart. J Physiol (Lond) 1983; 336:345-357 24. Maurer P, Weingart R: Cell pairs isolated from adult guinea pig and rat hearts: Effects of [Ca2+], on nexal membrane resistance. Pflugers Arch 1987;409:394-402 25. Wittenberg BA, White RL, Ginzberg RD, Spray DC: Effect of calcium on the dissociation of the mature rat heart into 26. 27. 28. 29. 30. 31. 32. 33. individual and paired myocytes: Electrical properties of cell pairs. Circ Res 1986;59:143-150 Allen DG, Morris PG, Orchard CH, Pirolo JS: A nuclear magnetic resonance study of metabolism in the ferret heart during hypoxia and inhibition of glycolysis. J Physiol (Lond) 1985;361:185-204 Grinward PM, Hearse DJ, Segal MB: A possible mechanism of glycolytic impairment after adenosine triphosphate depletion in the perfused rat heart. J Physiol (Lond) 1980; 301:337-347 Kupriyanov VV, Lakomkin VL, Steinschneider AY, Severina MY, Kapelko VI, Ruuge EK, Saks VA: Relationships between pre-ischemic ATP and glycogen content and post-ischemic recovery of rat heart. J Mol Cell Cardiol 1988;20:1151-1162 Wojtczak J: Influence of cyclic nucleotides on the internal longitudinal resistance and contractures in the normal and hypoxic mammalian cardiac muscle. J Mol Cell Cardiol 1982; 14:259-265 Pressler ML, Hathaway DR: Phosphorylation of purified dog heart gap junctions by c-AMP dependent protein kinase (abstract). Circulation 1987;76(suppl IV):IV-18 Johnson KR, Lampe PD, Hur KC, Louis CF, Johnson RG: A lens intercellular junction protein, MP26, is a phosphoprotein. J Cell Biol 1986;102:1334-1343 Verselis V, Brink PP: The gap junction channel: Its aqueous nature as indicated by deuterium oxide effects. Biophys J 1986;50:1003-1007 Hille B: Counting channels, in Hille B (ed): Ionic Channels of Excitable Membranes. Sunderland, Mass, Sinauer Associates, Inc, 1984, pp 205-225 KEY WORDS * paired ventricular myocytes * junctional conductance * ATP * ligand-receptor interaction ATP directly affects junctional conductance between paired ventricular myocytes isolated from guinea pig heart. H Sugiura, J Toyama, N Tsuboi, K Kamiya and I Kodama Downloaded from http://circres.ahajournals.org/ by guest on October 2, 2016 Circ Res. 1990;66:1095-1102 doi: 10.1161/01.RES.66.4.1095 Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1990 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7330. Online ISSN: 1524-4571 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circres.ahajournals.org/content/66/4/1095 Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation Research is online at: http://circres.ahajournals.org//subscriptions/