Ionic basis for membrane potential changes induced by hypoosmotic stress in guinea-pig ventricular myocytes

Objective: Causal relation between changes in action potentials and activation of several ionic currents during hypoosmotic challenge was investigated. Methods: We recorded changes in membrane potentials and currents during hypotonic stress in guinea-pig ventricular myocytes using whole-cell patch-clamp technique. Results: Exposure of ventricular myocytes to hypotonic solution (0.6 T) caused initial prolongation (|107% of control) of action potential duration at 90% repolarization (APD ) in 65% of examined myocytes. Later 90 shortening (|75% of control) of APD and depolarization of resting potential (RP) (| 4 mV) developed in all cells. Initial prolongation 90 31 of APD in hypotonic solution was mainly caused by transient activation of Gd -sensitive non-selective cation (NSC) current. Late 90 1 changes after |180 s in hypotonic solution were sustained increase in slow component of delayed rectifier K current (I ) in all cells, Ks and activation of I in 40% of cells. Prevention of APD shortening by chromanol, a selective blocker of I , was seen in about 40% Clswell 90 Ks of myocytes due to short APD in our experimental conditions. Application of 1 mM anthracene-9-carboxylic acid (9-AC) partially inhibited APD shortening in three of seven cells. Depolarization of RP was unaffected by the above-mentioned drugs, but was dependent 1 on [K ] . Conclusions: Initial prolongation followed by later shortening of APD in hypotonic solution are mostly caused by different o sequences of NSC, I and I currents activation. Depolarization of RP in hypotonic solution is probably due to dilution of Ks Clswell 1 1 1 subsarcolemmal K concentration and/or change in permeability ratio for Na and K .  2001 Elsevier Science B.V. All rights

The composition of the solutions used in this study are presented in Table 1. Standard isotonic external solution was obtained from normal Tyrode solution by reducing 2. Methods NaCl concentration to 64 mM and addition of 150 mM mannitol (osmolality was 282615 mOsm / l, measured by The investigation conforms to the Guide for the Care Vapor Pressure Osmometer 5520 (Wescor, Utah, USA). and Use of Laboratory Animals published by the US Osmolality of the pipette solution was 270611 mOsm / l. National Institutes of Health  The standard hypotonic solution was the same as the revised 1996), and was in accordance with the institutional isotonic except mannitol was reduced to 50 mM (osmolaliguidelines for animal care at Tokyo Medical and Dental ty was 177611 mOsm / l). pH was adjusted to 7.3 in the University.
bath solution with NaOH, and 7. 25  Na / K pump current. The bath solution used for isola-Isolation of single ventricular myocytes from guinea-pig tion of NSC had pH57.3 adjusted with TEA-OH and the heart was carried out by an enzymatic dissociation pro-pipette solution to pH57.25 with CsOH. The external cedure described previously [16]. Briefly, guinea-pigs, hypotonic solution for NSC recording was the same except weighing 300-400 g were anesthetized with pentobarbital mannitol concentration of 50 mM. In the bath solution sodium (20- high-K , low-Cl solution and used within 9 h from Gadolinium(III) chloride (Aldrich, Milwaukee, WI, isolation. Cells were transferred to a bath on the stage of USA) was dissolved in distilled water, freshly prepared an inverted microscope (Diaphot TMD, Nikon, Tokyo, before each experiment and added at final concentration of Japan) and were perfused with isotonic or hypotonic 20 mM to the test solution. Addition of this concentration 31 solution at 3660.58C. The average speed of perfusion was of Gd for up to 10 min to cells perfused with isotonic about 1.5 ml / min by gravity and exchange of the bath solution had no significant effect on APD (14864 ms in solution was completed within 20 s. the absence, and 14964 ms in the presence, n58) and RP (278.560.9 mV in the absence and 279.360.3 mV in the stability of action potential configurations and / or mempresence, not significant). Anthracene-9-carboxylic acid brane currents before collecting the control data. As initial (9-AC) and DIDS (Sigma Chemical, St Louis, MO, USA) prolongation of APD did not develop in all examined were dissolved in dimethylsulfoxide (DMSO) to give a myocytes, we arranged a pilot study to examine whether stock solution of 100 mM, freshly prepared, given in a cells with initial prolongation during first exposure to final concentration of 1 mM (9-AC) or 0.5 and 2 mM hypotonic solution showed similar prolongation on the (DIDS) in light-proof container. E-4031 (a gift from Eisai second exposure to hypotonic solution. In six out of six Pharmaceutical Comp., Tokyo, Japan) was dissolved in preparations with initial prolongation of APD during the distilled water, freshly prepared before each experiment first short exposure to hypotonic solution (150 s, enough and used at a concentration of 5 mM. Chromanol (a gift time for initial prolongation to appear and subside), the from Prof. H. J. Lang, Hoechst AG, Frankfurt, Germany) second exposure to hypotonic solution induced a similar was dissolved in DMSO given a stock solution of 10 mM, phenomenon. Thus, only the cells with initial prolongation while final concentration in the bath was 10 mM to 30 mM.
during first exposure to hypotonic solution were examined 31 21 Cadmium-chloride was dissolved in distilled water. Action potentials and membrane currents were recorded by a whole-cell patch-clamp technique using a patch-clamp 3. Results amplifier (Axopatch 200B, Axon Instrument, Foster City, CA, USA). Suction pipettes were made from borosilicate 3.1. Effects of hypotonic perfusion on membrane glass capillaries with inner filaments (Clark Electromedical potentials Instruments, Pangbourne, UK) pulled with a microelectrode puller (PP-83, Narishige, Tokyo, Japan). The tip Using current-clamp mode, we monitored the timeresistance of typical electrode when filled with standard course of changes in action potential duration at 90% internal solution was 1.5-2.5 MV. A 2 M KCl agar bridge repolarization (APD ) and resting potential (RP). Expo-90 between bath solution and AgCl-reference electrode im-sure to hypotonic solution produced an initial prolongation mersed in pipette solution was used to minimize liquid of APD (within 60 s of exposure) in 21 out of 33 90 junction potential. The membrane potential and current examined cells, followed by later shortening of APD in 90 signals were filtered at 2 kHz, monitored by a storage all preparations. In the remaining 12 myocytes, shortening oscilloscope DCS 7040 (Kenwood, Tokyo, Japan), digit-without initial prolongation of APD appeared. Depolar-90 ized using an analog converter (Digidata 1200, Axon ization of RP started almost simultaneously with short-Instruments, Foster City, CA, USA) and stored in a ening of APD , after about 120 s perfusion with hypo-90 personal computer (Fujitsu, Tokyo, Japan) for later analy-tonic solution in all examined preparations ( Fig. 1A and sis. Action potential was elicited by injecting supra-thres-B). APD after 60 s in hypotonic solution was prolonged 90 hold currents of 3 ms duration through the recording from 18068 ms in the control isotonic solution to 19269 electrode at a frequency of 0.1 Hz. Various ramp and ms (n521, P,0.05) and the rest of myocytes slightly voltage step protocols were used as indicated (see Results) shortened APD from 170.567 ms to 163.868 ms (n5 90 for current measurement. Membrane capacitance was 12, NS). Then, APD gradually shortened to 13265 ms 90 calculated from the integral of the current transient in (n533, P,0.01, in comparison with control isotonic response to 10-mV pulses. pCLAMP software (Version solution) at 300 s in the hypotonic solution. At the same 6.0.4., Axon Instruments) was used to generate pulse time, a small and gradual depolarization of RP developed protocols and for data acquisition. Origin software (Ver-and the average was 273.960.4 mV after 300 s, compared sion 5.0) was used for data analysis.
2.5. Experimental protocols 3.2. Effects of hypotonic solution on whole-cell membrane currents All measurements of membrane potentials and currents were performed at least 7 min after obtaining a whole-cell Using the same external and pipette solutions as for configuration. Additional 3 min were allowed to watch recording the membrane potential changes, we recorded whole-cell currents elicited by the fast descending ramp NSC) could underlie observed change in whole-cell curprotocol (46 mV/ s) in hypotonic solution (Fig. 2). During rent during the first 100 s in hypotonic solution. The early period (less then 100 s) of hypotonic stress, a small increased pump current was not likely a candidate, since 1 but significant decrease in outward currents at positive we used pipette solution without Na and the presence of 1 voltages was observed (b). At other voltage regions no Na was mandatory to activate this current [14]. Moreremarkable current changes were noted. After 150 s and over, addition of 100 mM ouabain did not affect transient later in hypotonic solution, gradual increase in outward depression of whole-cell current mentioned above (n54, current at positive voltages developed to exceed the control data not shown). Therefore, we recorded changes in level. At the same time, outward currents at 280 mV, isolated I . Fig. 3 shows changes in tail currents in Ks roughly corresponding to the RP level, were slightly but hypotonic solution in the presence of 5 mM E-4031 to significantly suppressed. Inward currents at voltages nega-block I and 1 mM nisoldipine to block I . Normalized Kr Ca,L tive to 280 mV were increased (c) with a positive shift of data obtained from five cells (Fig. 3D) clearly indicated the reversal potential (4.460.9 mV from the control, n55).
that initial (between 30 and 100 s) suppression of tail However, the slope of negative current portions was not current (from 0.7460.24 pA / pF in the control to changed (54.864.9 nS in isotonic solution vs. 51.162.5 nS 0.5860.19 pA / pF; n55, P,0.05) was followed by later in hypotonic solution, n55, NS). enhancement to 1.560.36 pA / pF after 300 s in hypotonic solution (P,0.01 vs. control). Chromanol at 30 mM 3.3. Early changes in membrane currents during completely inhibited increase in the tail current.

exposure to hypotonic solution
To rule out participation of I and I in initial Kr Ca,L prolongation of APD , we examined time-course of 90 We assumed that decrease in outward current (I and changes in isolated I and I during hypotonic chal-  out of 10 myocytes after about 60 s in hypotonic solution 31 only in the absence of 20 mM Gd (Fig. 5). The for first 60 s in hypotonic solution, then a gradual activation of this current was transient and the current depression in I was noted, from 0.4360.1 pA / pF in the subsided after about 120 s in hypotonic solution.
Kr control, to decrease significantly to 0.360.2 pA / pF (n55, 31 P,0.05) after 180 s. Pretreatment with E-4031 did not 3.4. Effects of Gd and chromanol on initial affect initial APD prolongation (Fig. 4A, Inset). Slight, but prolongation of APD not significant decrease in I was observed during Ca,L hypotonic stress for 300 s (Fig. 4B). Moreover, pretreat-The results in the previous section suggest that increased 21 ment with Cd to block I did not prevent initial APD NSC or decreased I might participate to initial prolonga-Ca,L Ks prolongation (Fig. 4B, Inset). Therefore, I and I were tion of APD in hypotonic solution. We tested effects of Kr Ca,L 90 31 Gd on action potential changes in hypotonic solution. In experimental conditions seemed to be negligible. The six out of seven examined myocytes with initial prolonga-average APD of myocytes in isotonic external solution 90 31 tion of APD , application of 20 mM Gd prevented the containing 64 mM NaCl was 17667.5 ms (n533). As the 90 prolongation in the second exposure to hypotonic solution duration of APD was critical for participation of I in Ks 31 (Fig. 6). Gd , however, did not affect depolarization of repolarization [20], we used action potential-clamp method RP (4.0360. 6 [10,21]. In the presence of Cd and ouabain, we observed hypotonic-sensitive time-dependent current in five out of five cells using voltage-step protocol shown in Inset in Fig.  7. The current increase was quickly and reversibly inhibited by 10 mM chromanol (Fig. 7C). Maximal increase in time-dependent current at 140 mV was 10.761.7 pA / pF after 300 s in hypotonic solution, from 4.3260.9 pA / pF in the control (n55, P,0.05). Addition of chromanol reduced this current to 3.8461.3 pA / pF (n55, P,0.05 vs. in the absence of chromanol).
In order to examine time-independent current changes in hypotonic solution, we used fast descending ramp protocol in Na -, K -and Ca -free solution with Cs instead of 1 K in pipette solution (see Methods for the solutions) for 300 s exposure. In nine examined myocytes, hypotonicsensitive current elicited by the ramp protocol was not affected significantly by DIDS, neither at 0.5 nor at 2.0 mM. We next used 9-AC, another blocker of I to Clswell 2 examine hypotonic-sensitive Cl current. Hypotonic-sensitive current was observed in four out of 10 cells, with a reversal potential of 119.661.6 mV, while predicted E Cl was 121 mV. It was completely inhibited by 1 mM 9-AC after 2-3 min incubation (Fig. 7D). The remaining five myocytes were not elicited hypotonic-sensitive time-independent currents. tonic stress together with the lack of effects of Cd on APD prolongation. I was depressed during hypotonic 90 Kr perfusion, which was in accordance with previous data [14,15], but the involvement of this current on early changes in APD was also excluded. We next explored possible contribution of I and NSC to this phenomenon.

Ks
Actually, time-dependent outward current at positive voltages compatible with I was initially depressed during Ks hypotonic stress, but chromanol, rather specific blocker of this current, did not prevent initial prolongation of APD . 90 This discrepancy could be explained as too short APD to 90 significantly contribute to repolarization by I in our Ks experimental condition as revealed by the action potentialclamp experiments. Therefore, decreased I as a factor for Ks the initial prolongation of APD seems negligible but a 90 small contribution by this current may not completely be excluded.
Published data indicate that the NSC channel can be activated by hypotonic stress [8,9]. We could isolate NSC currents in our control solution and the currents were transiently activated in hypotonic solution. Both currents, as well as initial prolongation of APD, were easily blocked 31 31 by 20 mM Gd . While non-selective actions of Gd on I and I [18,19] may suggest the participation of the Ca,L Kr two currents in these phenomenon, this possibility can be excluded by the reasons described above. Therefore, we assume that activation of NSC seems to be a main mechanism of initial prolongation of APD in hypotonic solution. The transient nature of this current activation, however, has not been explained in this study and its mechanism is to be further explored.  A small but significant prolongation of APD was hypotonic stress is not documented. Moreover, we did not 90 noted in two thirds of myocytes exposed to hypotonic observe time-independent hypotonic-sensitive currents solution at the early stage (within 100 s of exposure). In with reversal potential close to the reversal potential for 1 the similar conditions, the only current changes developed K when we applied fast ramp protocol (see Fig. 2   Clswell action potential changes in hypotonic solution. We failed to Taken together, our results indicate that in about 50% of inhibit time-independent currents induced by hypotonic cells shortening of APD is not due to specific current solution by 0.5 mM DIDS and obtained inhibitory effects modulation, but probably to direct effects of hypotonic by the same concentration on APD shortening in 21% of stress, as discussed below.  [27] argue against involvement of this current. Parallel