Characterization of the electrogenic Na+–K+ pump in bipolar cells isolated from carp retina

Neuropharmacology 37 (1998) 1053 – 1061

Characterization of the electrogenic Na + –K + pump in bipolar cells isolated from carp retina Ikuko Zushi a, Masahiko Shimuraa b, Makoto Tamai b, Yasuhiro Kakazu a, Norio Akaike a,* a b

Department of Physiology, Faculty of Medicine, Kyushu Uni6ersity, Fukuoka 812 -82, Japan Department of Ophthalmology, School of Medicine, Tohoku Uni6ersity, Sendai 980 -77, Japan Accepted 28 April 1998

Abstract The electrogenic Na + –K + pump current (Ip) in carp bipolar cells was investigated under voltage-clamp conditions. The Ip was activated in a concentration-dependent manner by adding external K + (Ko+ ), and was completely suppressed with 10 − 4 M ouabain (EC50 =1.23 mM; Hill coefficient =1.36). The Ip was suppressed in a concentration-dependent manner by ouabain (IC50 =1.90 mM; Hill coefficient =0.93). The Ip did not show a distinct voltage dependency either with or without Nao+ . A large outward shift of the holding current was observed by completely removing Nao+ . In the presence of Nao+ , a steady Ip was observed even in the absence of internal Na + (Nai+ ). These results suggest that continuous Na + influxes exist across the membrane. When external and internal Na + was removed, a transient Ip was observed (half decay time (t1/2) was 5.0 90.6 s), thus indicating that the transient Ip was activated by the residual Nai+ . In the absence of Nao+ , the transient Ip was also observed with lower than 8 mM Nai+ . The t1/2 depended on Nai+ . However, a steady Ip was observed with 10 mM Nai+ or more. The functional properties of the Ip are discussed. © 1998 Elsevier Science Ltd. All rights reserved. Keywords: Bipolar cell; Carp; Patch clamp; Retina; Sodium–potassium pump

1. Introduction The Na + –K + pump, located in an integral membrane, is responsible for establishing the electrochemical gradient of Na + and K + across the plasma membrane of living cells. The ion gradients formed by this pump are necessary for the active transport of essential nutrients into cells (Ullrich, 1979), for osmotic balance and cell volume regulation (MacKnight and Leaf, 1977), as well as for the maintenance of the resting membrane potential in excitable cells (Thomas, 1972, Jewell and Lingrel, 1991). The pump expels three intracellular sodium ions (Nai+ ) and takes up two extracellular potassium ions (Ko+ ) in one hydrolyzed ATP molecule, resulting in the generation of a net outward charge transfer across the cell membrane. The pump thus directly affects the resting membrane potential and exerts a profound effect on the activities of * Corresponding author. Tel.: +81 92 6411151; fax: + 81 92 6336748; e-mail: [email protected]. 0028-3908/98/$19.00 © 1998 Elsevier Science Ltd. All rights reserved. PII: S0028-3908(98)00084-7

excitable cells (Desmedt, 1953, Kerkut and York, 1969, Akaike, 1995). The bipolar cells of the retina are the pathway in the retina that passes information from the distal retina to the proximal retina. These cells receive input from photoreceptors, horizontal cells and interplexiform cells and then provide output to the amacrine cells, interplexiform cells and ganglion cells (Dowling, 1987). Like other neuronal cells, bipolar cells also have various ion channels to transduce and signal information. These channels play an integral role in generating and shaping the retinal neuronal response to light (Lasater, 1991). With regard to the membrane properties of the bipolar cell, the membrane conductances have already been examined in the goldfish (Kaneko and Tachibana, 1985), axolotl (Tessier-Lavigne et al., 1988), tiger salamander (Lasater, 1988), mouse (Kaneko et al., 1989) and rat (Karschin and Wa¨ssle, 1990). The resting membrane potential of the bipolar cell is − 35 to −45 mV in the dark. Their light-evoked response to small spot stimuli in their receptive field center can be either

1054

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

depolarizing or hyperpolarizing. This in turn is determined by the postsynaptic receptor which is activated by the photoreceptor neurotransmitter. Curiously, bipolar cells lack the fast, transient Na + current found in horizontal cells. Moreover, the characteristics of K + and Ca2 + currents have also been investigated in many species (Lasater, 1991), but the basic properties of the Na + –K + pump in bipolar cells have not yet been investigated. The analyses in this study were obtained using a nystatin perforated patch recording mode (Akaike and Harata, 1994). The pores formed by nystatin are permeable to monovalent ions (mainly Na + , K + and to a lesser extent Cl − ) but are not permeable to larger organic substances that are likely to be intracellular mediators of agonist-induced responses (Horn and Marty, 1988, Amedee et al., 1990, Shimura et al., 1996). Therefore, in this mode, [Na + ]i appears to be controlled by the ionic composition of the patch pipette solution with little influence on the intracellular environment. In this study, we investigated for the first time the properties of the Na + – K + pump in bipolar cells acutely dissociated from the carp retina under voltageclamp conditions.

vertical two-step puller (PP-83, Narishige, Tokyo). The resistance of the patch pipette was 3–7 MV. The ionic currents were measured with a patch-clamp amplifier (EPC-7, List-Electronic), low-pass filtered at 1 kHz (NF Electronic Instruments, Tokyo) and monitored on both a storage oscilloscope (HS-5100A, Iwatsu, Tokyo) and a pen-recorder (RECTIHORITZ-8K21, NippondenkiSan-ei, Tokyo). All data were simultaneously recorded on video tape after digitalization with a PCM processor (type PCM 501ESN, Nihon Kohden, Tokyo). All experiments were carried out at room temperature (23–25°C). In most experiments, the bipolar cells were maintained at − 40 mV, at which holding potential (VH) the membrane patch seemed to be the most stable in the present experimental conditions.

2.3. Solutions

Carp (14–20 cm long) were dark-adapted for a minimum of 60 min. After the carp had been anesthetized on ice, the eyes of the fish were enucleated and hemisected. The retinas were removed from the back of the eye under dim red light and incubated for 15 min in 25 ml of enzyme solution containing 4 mg/ml papain (Wako); then 0.3 mg/ml of L-cysteine crystals (Wako) were added to the solution to activate the papain. Following the incubation period, the retinas were initially washed twice with rinse solution containing 1 mg/ml of bovine serum albumin (Sigma), and then a further four times with only rinse solution. The retinas were then tritulated by turning the tissue specimens up and down several times in a 10-ml sterile pipette. Following three to four passages, the tissue fragments were removed and placed into 35-mm dishes (Falcon) in standard solution. All experiments were performed for at least 60 min after placing the fragments into the dishes.

The ionic composition of the enzyme solution was (in mM): NaCl 119.9, KCl 2.6, MgCl2 1, CaCl2 0.5, NaHCO3 1, NaH2PO4 0.5, sodium pyruvate 1, N-2-hydroxyethylpiperazine-N%-2-ethanesulfonic acid (HEPES) 4, glucose 16. The composition of the rinse solution was (in mM): NaCl 119.9, KCl 2.6, MgCl2 1, CaCl2 2.5, NaHCO3 1, NaH2PO4 0.5, sodium pyruvate 1, HEPES 4, glucose 16. The standard solution had a composition of (in mM): NaCl 115.0, KCl 7.5, MgCl2 1, CaCl2 2.5, NaHCO3 1, NaH2PO4 0.5, sodium pyruvate 1, HEPES 4, glucose 16. The pH of both solutions was adjusted to 7.4 with NaOH. The external test solution for measuring the pump current was (in mM): NaCl 120, MgCl2 3, BaCl2 3, CdCl2 0.3, HEPES 10, glucose 16. The pH was adjusted to 7.4 with tris(hydroxymethyl)aminoethane (Tris base). In the external test solutions containing K + for activating the Na + –K + pump, NaCl in the external solution was replaced by equimolar KCl; for the Na + free external solutions, NaCl was replaced by equimolar N-methylglucamine chloride. The patch pipette (internal) solution for the nystatin perforated patch recording had the following composition (in mM): CsCl 120, HEPES 10. In the internal solution containing Na + for activating the Na + –K + pump, CsCl in the internal solution was replaced by equimolar NaCl. The pH was adjusted to 7.2 with Tris base. Nystatin (Sigma) was first dissolved in methanol at a concentration of 10 mg/ml, and then was further diluted in the internal solution just before use, giving a final concentration of 200 mg/ml.

2.2. Electrophysiological study

2.4. Drugs

The electrical measurements were performed with the nystatin perforated patch recording configuration under voltage-clamp conditions. The patch pipettes were pulled from glass capillaries (1.5 mm o.d.) using a

The drugs used in this study were ouabain (Sigma) and N-methylglucamine (NMG; Tokyokasei, Tokyo). Ouabain was directly dissolved in the test solution. All test solutions were applied to bipolar cells with the use

2. Methods

2.1. Cell isolation

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

of the Y-tube technique which enabled the rapid exchange of the external solution within 20 ms (Murase et al., 1990).

2.5. Statistics The data were calculated as mean9standard error of the mean (S.E.M.). The continuous theoretical lines and curves were constructed by Eqs. (1) and (2) (see Section 3) using a least fitting routine. Some data were also calculated using the pCLAMP system (Axon Instruments, Foster City, CA).

3. Results

3.1. The current induced by Na + – K + pump acti6ity The Na + –K + pump current (Ip) was measured as the K + -activated current when the external and internal solutions contained K + , Ca2 + channel blockers such as Ba2 + , Cs + and Cd2 + , and which was blocked with ouabain. At first, the bipolar cell was superfused with standard external solution and dialyzed with an internal solution containing 40 mM Na + at a VH of − 40 mV. After VH had reached a steady state, the standard external solution was changed to a K + -free test external solution. After a few minutes of perfusion with the solution, the test external solution containing 3 mM K + was applied to activate the pump for 15 s. Such a change in the extracellular K + concentration ([K + ]o) from 0 to 3 mM caused a steady outward current, which was then completely suppressed by adding 10 − 4 M ouabain, a Na + – K + ATPase blocker (Fig. 1A), indicating that the K + -activated current was the electrogenic Na + – K + pump current (Ip) that resulted from the pump activity. To obtain the current – voltage (I – V) relationship for Ip, a ramp command was applied before (Fig. 1A, a) and during (Fig. 1A, b) the application of 3 mM K + . The ramp command changed from a VH of −40 to +20 mV, then to −90 mV and back to −40 mV for 1 s. Fig. 1B shows the actual I – V relationships for each ramp command obtained from the descending limbs of the ramp. The inset in Fig. 1B shows the subtracted I – V relationship between the currents obtained in the test external solutions with 3 mM K + and 0 mM K + (b − a). The Ip did not indicate a significant voltage dependency in the potential range − 90 mV to + 20 mV.

3.2. Acti6ation of Ip by external K + When the K + -free test external solution was exchanged with external solutions containing various [K + ]o, the Ip was activated in a concentration-dependent

1055

manner at a VH of −40 mV. The activated Ips demonstrated a steady current and the currents were completely blocked by 10 − 4 M ouabain (Fig. 2A). A continuous theoretical curve for Ip as a function of [K + ]o was drawn by a modified Michaelis–Menten equation using a least fitting routine after normalizing for the current amplitude induced by external solution with 3 mM K + : I= Imax

Cn C + (EC50)n n

(1)

where I is the relative Ip amplitude and C is [K + ]o. EC50 and n denote the half-activation concentration and the Hill coefficient, respectively. Fig. 2B shows the relationship between Ip and [K + ]o. The EC50 and Hill coefficient were 1.23 mM and 1.36, respectively.

Fig. 1. Electrogenic Na + – K + pump currents (Ip) in bipolar cells acutely dissociated from the carp retina. The cells were perfused with internal solution containing 40 mM Na + . External solution contained 120 mM Na + . (A) At a holding potential (VH) of −40 mV, the application of 3 mM K + to K + -free test external solution activated a ouabain-sensitive steady outward current. The dashed line shows the original holding current level (control). The ramp commands were applied before (a) and during (b) 3 mM K + application. (B) The actul current – voltage (I – V) relationships obtained from the descending limbs of the respective ramp method. Inset: subtraction of the currents obtained in the test external solutions containing 3 mM K + and 0 mM K + . [K + ]o is the external K + concentration.

1056

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

3.4. Effect of external Na + on Ip As shown in Fig. 1AFig. 2A, the application of 3 mM K + to the K + -free test external solution induced a steady Ip. When the extracellular Na + concentration ([Na + ]o) was changed from 120 to 0 mM in the K + free test external solution, an outward shift of the holding current was observed (Fig. 4, arrow a). The successive application of 3 mM K + was also able to fully activate the outward current (Fig. 4, arrow b) which was completely suppressed by 10 − 4 M ouabain (Fig. 4, arrow c). The duration of 0 mM [Na + ]o perfusion varied and was either 5 s, 1 min or 10 min, but no differences were observed in the amplitude of the Ip. To see I– V relationships for each Ip before and during the application of 3 mM K + with and without [Na + ]o, a ramp-voltage command was applied (Fig. 5A). Fig. 5B shows the actual I–V relationships for each ramp command obtained from the descending limbs of the ramp. To obtain the I–V curve for Ip (d− c) in the Na + -free test external solution, the I–V

Fig. 2. The activation of Ip by adding K + to a K + -free test external solution with 120 mM Na + . The cells were perfused with internal solution containing 40 mM Na + . (A) The application of 0.3– 10 mM K + caused a steady outward current in a concentration-dependent manner. Each Ip was completely suppressed by 10 − 4 M ouabain. (B) The Ip increased in a sigmoidal fashion as [K + ]o increased from 0.01 to 10 mM. Each point corresponds to the mean value obtained from five to six cells, and each bar indicates 9 S.E.M.

3.3. Inhibition of Ip by ouabain Ouabain selectively inhibits Na + – K + ATPase, resulting in the suppression of Ip activity (Vasilets and Schwarz, 1993). When the effect of ouabain on the Ip was investigated, ouabain blocked the Ip in a concentration-dependent manner (Fig. 3A). The measured inhibition curve was constructed based on the following equation: I=

(IC50)n C n + (IC50)n

(2)

where I is the Ip amplitude and C is the concentration of ouabain. IC50 represents the concentration of ouabain that inhibits 50% of Na + – K + pump activity; n denotes the Hill coefficient. The IC50 value calculated from the theoretical curve was 1.90 mM, and the Hill coefficient was 0.93.

Fig. 3. The relationship between Ip amplitude and ouabain concentration. (A) Ouabain suppressed the Ip in a concentration-dependent manner. (B) The concentration – inhibition relationship for ouabain on the Ip. Each point was obtained from four to six cells and each bar indicates 9S.E.M. All data points were fitted by Eq. (2).

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

1057

Fig. 4. The activation of the Ip with or without 120 mM Na + . The cells were perfused with an internal solution containing 40 mM Na + . In the presence of 120 mM [Na + ]o, Ip was activated by the external application of 3 mM K + . When [Na + ]o was changed to 0 mM, the holding current level was outwardly shifted (arrow a). The subsequent application of 3 mM K + induced an outward current (arrow b) which was suppressed by adding 10 − 4 M ouabain (arrow c). The lower dashed line shows the holding current level in 120 mM [Na + ]o; the upper dashed line shows the holding current level in 0 mM [Na + ]o.

relationship in the Na + -free test external solution with 3 mM K + (d) was thus subtracted from that in the Na + -free test external solution without K + (c). Similarly, the I –V relationship for Ip in the 120 mM Na + external solution was obtained by subtracting b from a. The subtracted Ips (d − c and b− a) did not show significant voltage dependency either in the presence or the absence of 120 mM [Na + ]o (Fig. 5B, inset). The equilibrium potential of a pump (ENa/K) transporting 3Na + and 2K + per molecule of ATP hydrolyzed is defined as the voltage at which the free energy of ATP hydrolysis and the work required for the reversible stoichiometric transfer of Na + and K + are equal and opposite. According to De Weer et al. (1988), ENa/K is defined by the equation: ENa/K =

(EATP +3ENa −2EK) (3−2)

(3)

The values for ENa and EK in our conditions were calculated as follows: ENa =RT[ln([Na + ]o/[Na + ]i)] =0.0821(273+25) ln(120/40) = + 26.88 mV EK =RT[ln([K + ]o/[K + ]i)] =0.0821(273 + 25)ln(3/120) = − 90.25 mV where R and T have their usual meanings. The approximate value for EATP in respiring animal cells is −600 mV (Veech et al., 1979, Tanford, 1981), and in our conditions it was −340 mV for ENa/K. This value was more hyperpolarized than the previously reported value of −200 to −240 mV (Hurlburt, 1970,

Caldwell, 1973, Chapman and Johnson, 1978). We tried to isolate the Ip in the potential range under − 340 mV but failed because of membrane stability. Thus, in these conditions, it did not show distinct voltage dependency, at least in the physiological ranges (− 90 mV to +20 mV) of holding potential.

3.5. Ip acti6ated by intracellular residual Na + To investigate the activation of Ip with 0 mM [Na + ]i, Na + in the patch pipette solution was replaced by Cs + . Even when the cells were dialyzed with Na + -free internal solution, the external application of 3 mM K + was able to induce the Ip in the presence of 120 mM [Na + ]o. The peak amplitude of the Ip was 14.794.1 pA, though there was only a slight decrease in the peak amplitude of the Ip during the first minute (Fig. 6A, arrow a). When [Na + ]o was completely removed, from 120 to 0 mM, the holding current showed a large outward shift (Fig. 6A, arrow b, 48.69 3.3 pA). In these conditions, the subsequent application of 3 mM K + to the test external solution could still induce the Ip, but the Ip only appeared transiently and thereafter completely disappeared within 30–40 s. This transient Ip was similarly observed at 0 mM [Na + ]o with perfusion times of 5 s, 1 min or 10 min. The peak amplitude of the transient Ip was 7.691.2 pA (Fig. 6A, arrow c), and the half decay time (t1/2) of the Ip was 5.090.6 s. After the complete reduction of the Ip, the second application of 3 mM K + to the external solution never induced any current (Fig. 6A, arrow d). When [Na + ]o was returned to the original concentration of 120 mM

1058

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

for 5 min, however, the Ip was fully activated again by adding 3 mM K + to the K + -free test external solution (Fig. 6A, arrow e). The Ip in the presence of 120 mM [Na + ]o was completely sensitive to 10 − 4 M ouabain (Fig. 6A, arrow f). In the Na + -free test external solution, the transient Ip induced with 3 mM K + test external solution was also blocked by 10 − 4 M ouabain (Fig. 6B). When the intracellular Na + concentration ([Na + ]i) was changed to various concentrations lower than 10 mM, the application of 3 mM K + to the test external solution induced the Ip in the presence of 120 mM [Na + ]o. When [Na + ]o was completely removed, the holding current showed a large outward shift as shown in Figs. 4 and 5A. In these conditions, the subsequent application of 3 mM K + to the test external solution could still induce the Ip. At lower concentrations than 8 mM [Na + ]i, the Ip appeared transiently and then disappeared with time (Fig. 7A, arrows), while the second application of 3 mM K + after a complete

Fig. 6. The effects of [K + ]o and [Na + ]o on Ip activation in the cells perfused with Na + -free internal solution. (A) The external application of 3 mM K + caused an outward current which slightly decreased to a steady state current level in the test external solution with 120 mM Na + (arrow a). When the [Na + ]o was decreased from 120 to 0 mM, the holding current showed an outward shift (arrow b). The subsequent application of 3 mM K + only induced a transient Ip (arrow c). The second application of 3 mM K + to a K + -free test external solution did not induce any current (arrow d). [Na + ]o was returned to the original concentration of 120 mM for 5 min; the Ip was then activated again by adding 3 mM K + to the K + -free test external solution (arrow e). The Ip was completely suppressed by 10 − 4 M ouabain (arrow f). (B) In Na + -free external and internal solutions, the transient Ip was also completely blocked by 10 − 4 M ouabain.

Fig. 5. The I – V relationship for the Ip with or without [Na + ]o using the ramp-clamp method. The cell was perfused with an internal solution containing 40 mM Na + . (A) The ramp commands ranging from − 90 mV to + 20 mV were applied before (a, c) and during (b, d) the application of 3 mM K + with (a, b) and without (c, d) 120 mM Na + . (B) The actual I– V relationships for each ramp command. Inset: Ip obtained from subtraction of the I–V curves b− a (in the presence of [Na + ]o and d−c (in the absence of [Na + ]o).

reduction of the Ip never induced any current. Otherwise, in 10 mM [Na + ]i, the Ip showed a steady state current and the second application of 3 mM K + also induced the Ip (Fig. 7A, asterisk). For a quantitative comparison of the time course of the disappearance of the transient Ip, the half decay time (t1/2) was calculated. The t1/2 became smaller as the [Na + ]i decreased (Fig. 7B).

4. Discussion We have investigated the functional properties of the electrogenic Na + –K + pump in bipolar cells acutely

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

dissociated from carp retina. Using suitable blockers for known conductances, the electrogenic Na + –K + pump current (Ip) was successfully isolated under voltage-clamp conditions.

4.1. Inhibition of Ip by ouabain Na + ,K + -ATPase is composed of a and b subunits (Jewell and Lingrel, 1991), the a subunit being related to ouabain sensitivity (Urayama and Sweadner, 1988, Sweadner, 1989, Urayama et al., 1989). In our previous study, the Ip in the horizontal cells of the fish retina showed biphasic ouabain inhibition, indicating the existence of two components of the a subunit (unpublished data). However, in the present study, ouabain inhibited the Ip monophasically (IC50 =1.9 mM). In the retina, both bipolar cells and horizontal cells lie in the inner nuclear layer and receive input from photoreceptors, even though their functional roles are

Fig. 7. The Ips with and without [Na + ]o observed at various [Na + ]i. (A) In Na + -free external solution conditions, the arrows indicate the transient Ips observed in 0 and 1 mM [Na + ]i. The asterisk indicates the Ips induced by a second application of 3 mM K + to cells internally perfused with 10 mM [Na + ]i. (B) The relationship between [Na + ]i and the half-time decay of the transient Ip (t1/2) in Na + -free external solution. Each point was obtained from four to five cells, and each bar indicates 9 S.E.M.

1059

different (Cohen, 1992). The membrane properties of each cell are also different in many ways. For example, bipolar cells lack the fast, transient Na + current found in horizontal cells (Lasater, 1991). Regarding the characteristics of the electrogenic Na + –K + pump, it is well known that the isoforms of the Na + ,K + -ATPase in bipolar cells are different from those in horizontal cells. The concentration–inhibition relationship for ouabain on the Ip was also different, indicating a different composition of the isoforms of Na + ,K + -ATPase. The functional differences of the isoforms of Na + ,K + -ATPase have not yet been fully determined, and further studies are required to clarify these remaining questions.

4.2. Current–6oltage relationship for Ip It is well known that the Ip in cardiac cells monotonically increases with membrane depolarization in the potential range − 160 to − 20 mV in the presence of extracellular Nao+ (Gadsby et al., 1985). The physiological meaning of the voltage dependency of the Na + – K + pump in neurons is unknown. The Na + –K + pump is electrogenic. Therefore, if, in neurons, the turnover rate of the pump that transports net charge is affected by the membrane potential in the same manner as other cells (De Weer et al., 1988), acceleration of the pump would result in membrane hyperpolarization and thus function as a negative feedback mechanism against over-excitation. A decrease in the pump current at hyperpolarized potentials may play a role in stabilizing the membrane potential. In squid giant axons, the Na + –K + pump current is strongly voltage dependent in 400 mM Na + seawater but only weakly so in Na + -free seawater (De Weer et al., 1988). In our recent study, the Ip in rat neostriatal neurons also demonstrated slight voltage dependency, but only in the presence of [Na + ]o (unpublished data). The presence of Nao+ seemed to contribute to the voltage dependency of the Na + –K + pump in neurons (De Weer et al., 1988). Otherwise, in this study, even in the presence of [Na + ]o, the Na + –K + pump did not have a distinct voltage dependency at the holding potential range between − 90 and 20 mV (Figs. 1 and 5B), which was seen physiologically in carp bipolar cells (Lasater, 1991). In contrast to other neurons, the bipolar cells lack the fast, transient Na + current. Moreover, at the outer nuclear layer in which the bipolar cells are located, the ionic environment of [Na + ]o and [K + ]o is altered according to light stimulation (Reichenbach et al., 1992). These characteristics of the ionic environment in the retina may be related to the slight degree of voltage dependency of the Na + –K + pump in the physiological holding potential range in bipolar cells. Further investigation is required to clarify the apparent lack of voltage dependency of the Na + –K + pump in bipolar cells.

1060

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061

4.3. Contribution of Na + influx to the Ip In bipolar cells dialyzed with an internal solution containing 40 mM [Na + ]i, complete removal of [Na + ]o from 120 to 0 mM induced an outward shift of the holding current, suggesting a decrease in the Na + influx. Therefore, if there is no Na + influx into the cell from the extracellular solution and if the cells are dialyzed with Na + -free internal solution, the pump might not be activated because the presence of Nai+ is required to activate the Ip. When the cells were dialyzed with Na + -free external and internal solutions, the addition of 3 mM K + to the external solution could still activate the Ip, even though the Ip only appeared transiently (Fig. 6A, arrow c). One possible explanation is that the transient Ip was activated by the residual submembranous Na + which had been carried across the plasma membrane when the cells were perfused with K + -free test external solution with 120 mM [Na + ]o. After the residual submembranous Na + was exhausted by adding 3 mM [K + ]o, [Na + ]o increased from 0 to 120 mM. In these conditions, the Ip was activated in a similar way to the control Ip recorded in 120 mM [Na + ]o. These findings indicate that the Na + influxes are large enough to cause a continuous activation of the Na + –K + pump activity in carp bipolar cells. In addition, the peak amplitude of the transient Ip which was observed in 0 mM [Na + ]o (7.6 91.2 pA) was about half that of the Ip in 120 mM [Na + ]o (14.7 94.1 pA). These results also suggest that persistent Na + influxes contribute greatly to pump activation.

4.4. Time course of the decay of transient Ip When the cells were perfused with Na + -free external solution, a steady Ip was observed using an internal solution containing 40 mM [Na + ]i, while a transient Ip was observed with Na + -free internal solution. When the [Na + ]i was changed in the presence of a Na + -free external solution, a transient Ip was observed at concentrations lower than 8 mM [Na + ]i, and the time course of its decay was [Na + ]i dependent. In Na + -free external and internal solutions, the Ip was activated by the residual Nai+ by adding external K + . To activate a steady Ip, a [Na + ]i of at least 8 and 10 mM was required in the present preparation (Fig. 7). In recent studies, the [Na + ]i measured by a Na + -sensitive fluorescent dye was 9.29 0.4 mM in rat ventricular myocytes (Katoh et al., 1995), 14.2 9 0.6 mM in rat proximal tubules (Reilly et al., 1995), 17.092.0 mM in sheep parotid secretory endpieces (Poronni et al., 1995) and 12.8 9 5.2 mM in guinea pig nasal gland acinar cells (Ikeda et al., 1995). These values were almost the same as those in our preparation.

The action of the Na + –K + pump expels Nai+ , so the Na + –K + pump activity and Nai+ are thought to be closely related each other. There were some previous reports that inhibition of the Na + –K + pump increases [Na + ]i (Agrawal and Fehlings, 1996) and intracellular Na + entry potentiates hypoxic-ischemic cell death by causing cytotoxic cell edema (Wang et al., 1993). However, the mechanism of the link between Nai+ , Na + influx and Na + –K + pump activity has yet to be determined. In the present study, Na + influx and [Na + ]i were closely related to each other and both were found to contribute to Na + –K + pump activity. The present study is considered to form a useful framework for analyzing the Na + –K + pump in the retina, since the regulation of Na + in retina has, until now, only been rarely investigated.

Acknowledgements The authors thank Dr Takao Matsui, the President of Oshima Eye Hospital, for helpful comments and Dr Brian Quinn for editing the English. This study was supported by a Grant-in-Aid for Scientific Research (No. 07407002) and a Grant-in-Aid for Scientific Research on Priority Areas (No. 07276101) to N. Akaike from the Ministry of Education, Science and Culture, Japan.

References Agrawal, S.K., Fehlings, M.G., 1996. Mechanisms of secondary injury to spinal cord axons in vitro: role of Na + , Na + –K + -ATPase, the Na + – H + exchanger, and the Na + – Ca2 + exchanger. J. Neurosci. 16, 545 – 552. Akaike, N., 1995. Contribution of an electrogenic sodium pump to membrane potential in mammalian skeletal muscle fibers. J. Physiol. 245, 499 – 520. Akaike, N., Harata, N., 1994. Nystatin perforated patch recording and its applications to analyses of intracellular mechanisms. Jpn. J. Physiol. 44, 433 – 473. Amedee, T., Large, W.A., Wang, Q., 1990. Characterization of chloride currents activated by noradrenaline in rabbit ear artery cells. J. Physiol. 428, 501 – 516. Caldwell, P.C., 1973. Possible mechanisms for the linkage of membrane potentials to metabolism by electrogenic transport processes with special reference to Ascaris muscle. Bioenergetics 4, 201–209. Chapman, J.B., Johnson, E.A., 1978. The reversal potential for an electrogenic sodium pump. A method for determining the free energy of ATP breakdown? J. Gen. Physiol. 72, 403 – 408. Cohen, A.I., 1992. In: Hart, J.W.M. (Ed.), The Retina. Adler’s Physiology of the Eye, vol. 19. Stamathis, St Louis, MO, pp. 579 – 615. De Weer, P., Gadsby, D.C., Rakowski, R.F., 1988. Voltage dependence of the Na – K pump. Annu. Rev. Physiol. 50, 225–241. Desmedt, J.E., 1953. Electrical activity and intracellular sodium concentration in frog muscle. J. Physiol. 121, 191 – 295. Dowling, J.E., 1987. An approachable part of the brain. In: The Retina, The Belknap Press of Harvard University Press, Cambridge, MA.

I. Zushi et al. / Neuropharmacology 37 (1998) 1053–1061 Gadsby, D.C., Kimura, J., Noma, A., 1985. Voltage dependence of Na/K pump current in isolated heart cells. Nature 315, 63 – 65. Horn, R., Marty, A., 1988. Muscarinic activation of ionic currents measured by a new whole-cell recording method. J. Gen. Physiol. 92, 145 – 159. Hurlburt, W.P., 1970. Ion movements in nerve. In: Membranes and Ion Transport. Wiley, New York, pp. 95–143. Ikeda, K., Ishigaki, M., Wu, D., Sunose, H., Takasaka, T., 1995. Na + transport processes in isolated guinea pig nasal gland acinar cells. J. Cell Physiol. 163 (1), 204–209. Jewell, E.A., Lingrel, J.B., 1991. Comparison of the substrate dependence properties of the rat Na,K-ATPase a1, a2, and a3 isoforms expressed in HeLa cells. J. Biol. Chem. 266 (25), 16925–16930. Kaneko, A., Tachibana, M., 1985. A voltage-clamp analysis of membrane currents in solitary bipolar cells dissociated from Carassius auratus. J. Physiol 358, 131–152. Kaneko, A., Pinto, L.H., Tachibana, M., 1989. Transient calcium current of retinal bipolar cells of the mouse. J. Physiol. 410, 613 – 629. Karschin, A., Wa¨ssle, H., 1990. Voltage- and transmitter-gated currents in isolated rod bipolar cells of rat retina. J. Neurophysiol. 63, 860 – 876. Katoh, H., Noda, H., Hayashi, H., Satoh, H., Terada, H., Ohno, R., Yamazaki, N., 1995. Intracellular sodium concentration in diabetic rat ventricular myocytes. Jpn. Heart J. 36 (5), 647–656. Kerkut, G.A., York, B., 1969. The oxygen sensitivity of the electrogenic sodium pump in snail neurons. Comp. Biochem. Physiol. 28, 1125 – 1134. Lasater, E.M., 1988. Membrane currents of retinal bipolar cells in culture. J. Neurophysiol. 60, 1460–1480. Lasater, E.M., 1991. Membrane properties of distal retinal neurons. Prog. Retinal Res. 11, 215–246. MacKnight, A.D.C., Leaf, A., 1977. Regulation of cellular volume. Physiol. Rev. 57, 510–573. Murase, K., Randic, M., Shirasaki, T., Nakagawa, T., Akaike, N., 1990. Serotonin suppresses N-methyl-D-aspartate responses in acutely isolated spinal dorsal horn neurons of the rat. Brain Res. 525, 84 – 91.

.

1061

Poronni, P., Schumann, S.Y., Cook, D.I., 1995. HCO3− -dependent Ach activated Na + influx in sheep parotid secretory endpieces. Pflu¨gers Arch. 429 (6), 852 – 858. Reichenbach, A., Henke, A., Eberhardt, W., Reichelt, W., 1992. K + ion regulation in retina. Can. J. Physiol. Pharmacol. 70, S239– S247. Reilly, A.M., Harris, P.J., Williams, D.A., 1995. Biphasic effect of angiotensin II on intracellular sodium concentration in rat proximal tubules. Am. J. Physiol. 269 (3pt2), F374 – 380. Shimura, M., Harata, N., Tamai, M., Akaike, N., 1996. Allosteric modulation of GABAA receptor in acutely dissociated neurons of the suprachiasmatic nucleus. Am. J. Physiol. 270, C1726–1734. Sweadner, K.J., 1989. Isozymes of the Na/K-ATPase. Biochim. Biophys. Acta 988, 185 – 220. Tanford, C., 1981. Equilibrium state of ATP driven ion pumps in relation to physiological ion concentration gradients. J. Gen. Physiol. 77, 223 – 229. Tessier-Lavigne, M., Attwell, D., Mobbs, P., Wilson, M., 1988. Membrane currents in retinal bipolar cells of the axolotr. J. Gen. Physiol. 91, 49 – 72. Thomas, R.C., 1972. Electrogenic sodium pump in nerve and muscle cells. Physiol. Rev. 52, 563 – 594. Ullrich, K.J.A., 1979. Sugar, amino acid, Na + cotransport in the proximal tubule. Annu. Rev. Physiol. 41, 181 – 195. Urayama, O., Sweadner, K.J., 1988. Ouabain sensitivity of the alpha 3 isozymes of rat Na,K-ATPase. Biochem. Biophys. Res. Commun. 156, 796 – 800. Urayama, O., Shutt, H., Sweadner, K.J., 1989. Identification of three isozyme proteins of the catalytic subunit of the Na,K-ATPase in rat brain. J. Biol. Chem. 264, 8271 – 8280. Vasilets, L.A., Schwarz, L.A., 1993. Structure – function relationships of cation binding in the Na + /K + -ATPase. Biochim. Biophys. Acta 1154, 201 – 222. Veech, R.L., Lawson, J.W.R., Cornel, N.W., Krebs, H.A., 1979. Cystilic phosphorylation potential. J. Biol. Chem. 254, 6538– 6547. Wang, G.X., Schmied, R., Ebner, F., Korth, M., 1993. Intracellular sodium activity and its regulation in guinea-pig atrial myocardium. J. Physiol. 465, 73 – 84.