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

Pergamon

PII:

Neuroscience Vol. 86, No. 1, pp. 233–240, 1998 Copyright ? 1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(98)00021-9

CHARACTERIZATION OF THE ELECTROGENIC Na+–K+ PUMP IN HORIZONTAL CELLS ISOLATED FROM THE CARP RETINA M. SHIMURA,*† M. TAMAI,† I. ZUSHI* and N. AKAIKE*‡ *Department of Physiology, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan †Department of Ophthalmology, School of Medicine, Tohoku University, Sendai 980-77, Japan Abstract––The electrogenic Na+–K+ pump current in horizontal cells acutely dissociated from the carp retina was investigated using a nystatin-perforated patch recording configuration under voltage-clamp conditions. In the presence of suitable blockers for known voltage-dependent Na+, K+ and Ca2+ conductances, the pump current was activated in a concentration-dependent manner by adding K+ ions to external solution. The EC50 value and Hill coefficient for the external K+ concentration were 0.66 mM and 1.39, respectively. The pump current did not show any significant voltage dependency at the physiological potential range between "90 and 20 mV either with or without external Na+ ions. In the presence of 120 mM external Na+ concentration, the addition of 3 mM K+ to the external solution induced a steady outward pump current even when the patch-pipette (internal) solution did not contain Na+. A large outward shift of the holding current was observed by removing external Na+. The result thus suggests that continuous Na+ influxes exist across the plasma membrane in the presence of external Na+. When Na+ was removed from both external and internal solutions, a transient outward pump current was observed by adding K+ to the external solution, thus indicating that the transient pump current was activated by the residual intracellular Na+ ions. The pump current was suppressed by ouabain in a concentration-dependent manner, and the ouabain-sensitive inhibition curve was fitted by two components. The IC50 values of high- and low-sensitive pump currents for ouabain were 20 nM and 10.4 ìM, respectively, indicating the existence of at least two isoforms of the pump in the horizontal cells. ? 1998 IBRO. Published by Elsevier Science Ltd. Key words: carp retina, horizontal cell, sodium–potassium pump.

The Na+–K+ pump located in an integral membrane is responsible for establishing the electrochemical gradient of Na+ and K+ ions 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,26 for osmotic balance and cell volume regulation,13 and for the maintenance of the resting membrane potential in excitable cells.9,24 The pump expels three intracellular Na+ ions and uptakes two extracellular K+ ions in a hydrolysed ATP molecule, thus resulting in the generation of a net outward current. Therefore, the pump directly affects the resting membrane potential and thus exerts a profound effect on the activities of excitable cells.1,5,11 In the retina, the maintenance of ionic environment in the intra- and extracellular ionic concentration is very important. At the outer plexiform layer in the retina, Na+ ions enter the outer segment ‡To whom correspondence should be addressed. Abbreviations: HEPES, N-2-hydroxyethylpiperazine-N*-2ethanesulphonic acid; Ip, pump current; [K+]o, external K+ concentration; [K+]i, internal K+ concentration; [Na+]o, external Na+ concentration; [Na+]i, internal Na+ concentration; NMG, N-methyl-glucamine; TTX, tetrodotoxin; VH, holding potential.

of the photoreceptor cell in the dark, while this Na+ influxes stop in the light. To maintain the normal intracellular Na+ concentration ([Na+]i) in the photoreceptor cell, Na+–K+ pump in the inner segment is activated.20 Otherwise at the inner plexiform layer, during the electrical activities, retinal neurons as well as other CNS neurons release K+ ions to the extracellular space. This may lead to considerable increases in the extracellular K+ concentration ([K+]o) both during and after illumination.22 Therefore, to limit the alterations of the ionic environment, the magnitude and duration of increased [K+]o are reduced by various mechanisms of K+ clearance.19 It is well known that in the retina, the glial cells like retinal pigment epithelial cells and Muller cells contribute to ion transport mechanisms by activating the Na+–K+ pump to uptake extracellular K+ ions.17,18 Although there have been several studies on the ion transporters responsible for the Na+ and K+ fluxes in retinal cells, the basic properties of the Na+–K+ pump in the horizontal cells, one of the retinal neurons, have not yet been investigated. In the present study, the electrogenic Na+–K+ pump current in the horizontal cells acutely dissociated from the carp retina was investigated under voltage-clamp conditions by using a nystatin

233

234

M. Shimura et al.

perforated-patch recording mode, which almost maintains the intact intracellular environment.2 EXPERIMENTAL PROCEDURES

Cell isolation Carp (14–20 cm in length) were dark-adapted for a minimum of 60 min. After the carp were anaesthetized on ice, the eyes of the fish were then enucleated and hemisected. The retinas were removed from the back of the eye under dim red light and then were incubated in 25 ml of enzyme solution containing 3 mg/ml papain (Wako Chemicals, Osaka, Japan) for 20 min. -cysteine (Wako Chemicals, Osaka, Japan) of 0.3 mg/ml was 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, U.S.A.), and then further four times with only rinse solution. Then the retinas were triturated 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, New Jersey, U.S.A.) filled with standard solution. All experiments were performed for at least 60 min after placing the fragments into the dishes. Electrophysiological study The electrical measurements were performed with the nystatin perforated-patch recording configuration under voltage-clamp conditions. The patch pipettes were pulled from the glass capillaries (o.d., 1.5 mm) using a vertical 2-step puller (Narishige, PP-83, Tokyo, Japan). The resistance of the patch pipette was 3–7 MÙ. The ionic currents were measured with a patch-clamp amplifier (EPC-7, ListElectronic), low-pass filtered at 1 KHz (NF Electronic Instruments, Tokyo, Japan), and monitored on both a storage oscilloscope (Iwatsu, HS-5100A, Tokyo, Japan) and a pen-recorder (Nippondenki San-ei, RECTIHORITZ-8K21, Tokyo, Japan). All data were simultaneously recorded on video tape after digitalization with a PCM processor (Nihon Kohden, type PCM 501ESN, Tokyo, Japan). All experiments were carried out at room temperature (23–25)C). In most experiments, the horizontal cells were maintained at "40 mV, at this holding potential (VH) the membrane patch seemed to be most stable in the present experimental conditions. Solutions 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, Na-pyruvate 1, HEPES 4 and glucose 16, and the rinse solution was (in mM): NaCl 119.9, KCl 2.6, MgCl2 1, CaCl2 2.5, NaHCO3 1, NaH2PO4 0.5, Napyruvate 1, HEPES 4 and glucose 16. The standard solution has a composition (in mM): NaCl 115.0, KCl 7.5, MgCl2 1, CaCl2 2.5, NaHCO3 1, NaH2PO4 0.5, Na-pyruvate 1, HEPES 4 and 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 and 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 with equimolar KCl. In the condition of Na+-free external solutions, NaCl was replaced with equimolar N-methyl-glucamine chloride (NMG-Cl). For the ramp-clamp measurement, 10"6 M tetrodotoxin (TTX) was added to the test solution. In most experiments, the patch pipette (internal) solution for the nystatin perforated-patch recording had the following composition (in mM): CsCl 80, NaCl 40 and HEPES 10.

The Na+-free patch pipette solution included (in mM): CsCl 120 and HEPES 10. The pH was adjusted to 7.2 with Tris-base. Nystatin (Sigma, U.S.A.) was first dissolved in methanol at a concentration of 10 mg/ml, and then was further diluted in the internal solution which thus resulted in a final concentration of 200 ìg/ml just before use. Drugs The drugs used in this study were TTX (Sankyo, Tokyo, Japan), ouabain (Sigma, U.S.A.), NMG (Tokyokasei, Tokyo, Japan). TTX and ouabain were directly dissolved in the test solution. All test solutions were applied to horizontal cells using the ‘‘Y-tube’’ technique which enabled a rapid change of the external solution within 20 ms.16 Statistics The data were calculated as the mean&S.E.M. The continuous theoretical lines and curves were constructed by the given equations (see text) using a least fitting routine. Some data were also calculated using the pCLAMP system (Axon Instruments, Foster City, CA, U.S.A.). RESULTS

The current induced by the Na+–K+ pump activity The Na+–K+ pump current was measured as the K+-activated current in the external and internal solutions containing appropriate blockers for known ionic conductances. At first, the horizontal cell was perfused with standard external solution and internal solution containing 40 mM Na+ at a holding potential (VH) of "40 mV. When the holding current level reached a steady state, the standard external solution was changed to a K+-free test external solution. After a few minutes perfusion with the solution, the test external solution containing 3 mM K+ was applied to activate the pump for 10 s. Such a change in the extracellular K+ concentration ([K+]o) from 0 to 3 mM caused a steady outward current, and the current was completely suppressed by adding 3#10"4 M ouabain, a Na+–K+ ATPase blocker (Fig. 1A), indicating that the K+-activated outward current is the electrogenic Na+–K+ pump current (Ip) which resulted from the pump activity. To obtain the current–voltage (I–V) relationship for Ip, a ramp command was applied before (Fig. 1Aa) and during (Fig. 1Ab) an application of 3 mM K+. To block the voltage-dependent Na+ channels, 10"6 M TTX was added to these test external solutions. The ramp command was changed from a VH of "40 to +20 mV, then to "90 mV and back to "40 mV for 1 s. Figure 1B shows the actual I–V relationships for each ramp command obtained from the descending limbs of the ramp. The inset of Fig. 1B shows a subtracted I–V relationship between the currents obtained in test external solutions with 3 mM K+ and 0 mM K+. The Ip did not indicate any significant voltage dependency. Activation of pump current by external K+ When the K+-free test external solution was exchanged by the external solutions containing

Na+–K+ pump in carp horizontal cells

Fig. 1. The electrogenic Na+–K+ pump current (Ip) in horizontal cells acutely dissociated from the carp retina. The cells were perfused with internal solution containing 40 mM Na+. (A) At a holding potential (VH) of "40 mV, the application of 3 mM K+ to K+-free test external solution activated an 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 actual current– voltage (I–V) relationships obtained from the descending limbs of the respective ramp method. (Inset) Subtraction between the currents obtained in the test external solutions containing 3 mM K+ and 0 mM K+. [K+]o is the external K+ concentration. Representative data were also obtained from four other cells. Extracellular Na+ concentration ([Na+]o)= 120 mM, intracellular Na+ concentration ([Na+]i)=40 mM.

+

various K concentrations, the Ip was activated in a concentration-dependent manner at a VH of "40 mV. The activated Ips were a steady current. But at a high concentration of external 30 mM K+, Ip gradually decreased with time (Fig. 2A). A continuous theoretical curve for the peak Ip was drawn as a function of [K+]o by a modified Michaelis–Menten equation using a least fitting routine after normalizing for the current amplitude induced by the external solution with 3 mM K+: Cn I=Imax n C +EC50n

(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. Figure 2B shows a relationship between Ip and [K+]o. The EC50 and Hill coefficient were 0.66 mM and 1.39, respectively.

235

Fig. 2. The activation of Ip by adding K+ to K+-free test external solution. The cells were perfused with internal solution containing 40 mM Na+. (A) The application of K+ caused an outward current in a concentration-dependent manner. Ip gradually decreased in the presence of 30 mM K+. (B) The Ip increased in a sigmoidal fashion as [K+]o increased from 0.01 to 30 mM. Each point corresponds to the mean value obtained from five to six cells, and each bar indicates &S.E.M. [Na+]o =120 mM, [Na+]i =40 mM.

This figure also shows that Ip is fully activated by [K+]o of over 3 mM. Effect of external Na+ on pump current As shown in Figs 1A, 2A and 3A, the application of 3 mM K+ to the K+-free test external solution induced Ip. When [Na+]o was changed from 120 to 0 mM in the K+-free test medium, an outward shift of the holding current was observed (Fig. 3A). The successive application of 3 mM K+ could also fully activate the Ip. The duration of 0 mM [Na+]o perfusion varied and was either 5 s, 1 min or 10 min, but no differences were observed among each amplitude of the Ip. To see the 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. Figure 3B 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 relationship (c) in the Na+-free test external solution without K+ was thus subtracted from that (d) in the Na+-free test external solution with 3 mM K+. Similarly, the I–V relationship for Ip in the

236

M. Shimura et al.

Fig. 3. The I–V relationships for Ip with or without [Na+]o using the ramp-clamp method. Cells were perfused with internal solution containing 40 mM Na+. (A) The ramp commands ranging "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) Ips obtained from the subtraction of the I–V curves b"a (in the presence of [Na+]o) and d"c (in the absence of 120 mM [Na+]o). Representative data were also obtained from three other cells. [Na+]o =120 mM or 0 mM, [Na+]i =40 mM.

120 mM Na+ external solution was obtained by subtracting a from b. These subtracted Ips (d"c and b"a) did not show any significant voltagedependency either in the presence or absence of 120 mM [Na+]o (Fig. 3B inset). Pump current activated by intracellular residual Na+ Even when the cells were perfused with Na+-free internal solution, the external application of 3 mM K+ could induce Ip in the presence of 120 mM [Na+]o, though there was a slight decrease in the peak amplitude of Ip (13.6&3.7 pA, n=7) during the first 1 min (Fig. 4A, arrow a). When [Na+]o was completely removed from 120 to 0 mM, the holding current showed a large outward shift (Fig. 4B, arrow b, 47.3&2.6 pA, n=7). In this condition, the subsequent application of 3 mM K+ to the test external solution could still induce Ip, but the Ip appeared transiently and then completely disappeared within 40–50 s (Fig. 4B, arrow c). This transient Ip was similarly observed at 0 mM [Na+]o perfusion times of 5 s, 1 min or 10 min. After the complete reduction of

Fig. 4. The effect of [K+]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 decreased a little to a steady-state current level in the test external solution with 120 mM Na+ (arrow a). The Ip was completely blocked by 10"4 M ouabain. (B) When the [Na+]o 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 K+-free test external solution did not induce any current (arrow d). [Na+]o = 120 mM or 0 mM, [Na+]i =0 mM.

Ip, the second application of 3 mM K+ to the external solution never induced any current (Fig. 4B, arrow d). When [Na+]o returned to the original concentration of 120 mM, however, the Ip was fully activated again by adding 3 mM K+ to K+-free test external solution (data not shown). Inhibition of pump current by ouabain Since ouabain selectively inhibits the Na+-K+ ATPase,29 the effect of ouabain on Ip was quantitatively investigated. Ouabain blocked the Ip in a concentration-dependent manner (Fig. 5), though the inhibition curve was biphasic (Fig. 6). The measured inhibition curve was constructed using the following theoretical equation, I=f

IC50 An 1

IC50nB2 +(1"f ) C n1+IC50nA1 C n2+IC50nB2

(2)

where I is the Ip amplitude and C is the concentration of ouabain. IC50A and IC50B represent the

Na+–K+ pump in carp horizontal cells

Fig. 5. Effects of ouabain at various concentrations on Ip. Ouabain suppressed concentration-dependently the Ip. [Na+]o =120 mM, [Na+]i =40 mM.

concentrations of ouabain that inhibits 50% of Na+–K+ pump activities for the endogenous isoforms. The n1 and n2 denote the Hill coefficient for each isoform, respectively. The f represents the fraction of Na+–K+ pump activity contributed by the endogenous isoforms. The IC50 values calculated from the biphasic curve were 20 nM (IC50A) and 10.4 ìM (IC50B) for high- and low-ouabain sensitive components, respectively. Both the Hill coefficients were 1.15 while the f value was 0.326. DISCUSSION

We have investigated for the first time the functional properties of the electrogenic Na+–K+ pump in horizontal cells acutely dissociated from the carp retina. Using suitable blockers for known conductances and nystatin perforated patch recording mode, the electrogenic Na+–K+ pump current (Ip) was successfully isolated under voltage-clamp conditions. Activation of pump current In the carp horizontal cells, the EC50 and Hill coefficient for K+ were 0.66 mM and 1.39, respectively. As to the other cell types, each value was 1.23 mM and 1.36 in carp bipolar cells and 0.73 mM and 1.2 in rat neostriatal neurons. In the retinal pigment epithelial cells, EC50 and Hill coefficient were 1.06 mM and 2.25, respectively (unpublished

237

Fig. 6. Relationship between the Ip amplitude and ouabain concentration. (A) The inhibition curve was biphasic. Each point was obtained from four to six cells (12 cells in only 10"7 M ouabain), and each bar indicates &S.E.M. The data-points were fitted by equation (2). (B) Each component calculated from the biphasic curve was separated, and the characteristics of ouabain sensitivity for each subunit were thus depicted. Curve a indicates the ouabain highsensitive component subunit (32.6%) while curve b indicates the low-sensitive one (67.4%). [Na+]o =120 mM, [Na+]i = 40 mM.

observations). These values were almost the same regardless of the difference of cell types. Current–voltage relationship for pump current It is well known that the Ip in cardiac cells monotonically increased with the membrane depolarization at the potential range between "160 and "20 mV in the presence of extracellular Na+.7 In squid giant axons, the Na+–K+ pump current is strongly voltage dependent in 400 mM [Na+]o seawater but only weakly in Na+-free seawater.4 In addition, in our recent study, the Ip in the rat neostriatal neurons also had a slight voltage dependency only in the presence of [Na+]o (unpublished observations). In the present study, however, even in the presence of [Na+]o, the Na+–K+ pump in the carp horizontal cell had no voltage dependency at the holding potential range between "90 and 20 mV (Fig. 3). The physiological meanings of voltage dependency of the Na+–K+ pump in neurons are still fully unknown. In the retina, [Na+]o always changes

238

M. Shimura et al.

around the photoreceptor cell associated with phototransduction (see Introduction). Therefore, such voltage independency of the Na+–K+ pump in the carp horizontal cell may be related to the speciality of the ionic environment of extracellular Na+ ions in the retina.

[Na+]o. It is also worth mentioning that the amplitude of a steady-state current of Ip in the presence of external Na+ corresponded to one third of the Na+dependent inward current. These findings thus agreed with the supposition that the Na+ flux into the cell, and the accumulated Na+, was expelled by the pump at a ratio of 3:2.

Contribution of Na+ influx on pump current In the horizontal cells perfused with the internal solution containing 40 mM Na+, the complete removal of [Na+]o from 120 to 0 mM induced the outward shift of the holding current, thus suggesting a decrease in the Na+ influxes. Therefore, if there is no Na+ influx into the cell from the extracellular solution and if the cells are perfused with Na+-free internal solution, the pump might not be activated, because the presence of intracellular Na+ is necessary to activate Ip. The Na+ influx was not influenced by either 1 mM furosemide or 50 ìM bumetanide, a blocker of the Na+, K+–2Cl" cotransporter or 1 mM metylisobutyl amirolide, a blocker of the Na+–H+ exchanger (data not shown). Thus the Na+ influx was not mediated by the Na+, K+–2Cl" cotransporter, Na+, 2HCO3" cotransporter or Na+–H+ exchanger, and therefore some other mechanism of Na+ influx was thus suggested to possibly exist. When the cells were perfused with Na+-free external and internal solutions, the addition of 3 mM K+ to the external solution could still activate the Ip, though the Ip only appeared transiently (Fig. 4B). The result indicates that the transient Ip was activated by the intracellular residual Na+ ions which had been carried across the plasma membrane when the cells had been perfused with K+-free test external solution with 120 mM Na+. After the intracellular residual Na+ was exhausted by adding 3 mM K+, [Na+]o was increased from 0 to 120 mM. In this condition, the Ip was fully activated again and the Ip was the same as the control Ip recorded in 120 mM [Na+]o. These results thus suggest that the Na+ influxes are large enough to cause a continuous activation of the Na+–K+ pump activity in the carp horizontal cells. Since such Na+ influxes are persistently observed, the Na+–K+ pump is activated at all times. Therefore, to maintain a normal [Na+]i in the cells, the excess Na+ influxes are compensated by the Na+ extrusion to the extracellular space by the pump activities, resulting in the appearance of Ip. In the present preparation perfused with Na+-free internal solution, the peak amplitude of a transient Ip activated by adding 3 mM K+ was 3.8&1.1 pA (Fig. 4Bc). On the other hand, a steady Ip activated by adding K+ in the presence of 120 mM [Na+]o was 13.6&3.7 pA (Fig. 4Aa). These results also suggest that the persistent Na+ influxes contribute greatly to the pump activation. Moreover, the large outward shift of the holding current was 47.3&2.6 pA (Fig. 4Ab), and this value was about three times larger than the value of Ip in the presence of 120 mM

Inhibition of pump current by ouabain In the concentration-dependent inhibition of ouabain on the Ip evoked by 3 mM K+, at least two isoforms were observed. As shown in Fig. 6B, one had a high sensitivity to ouabain (IC50A =20 nM), while the other isoform had low sensitivity to the drug (IC50B =10.4 ìM). If two isozymes exist independently in the different cells, the Ip of the ouabain low-sensitive cell group might show a small inhibition by 10"7 M ouabain while a high-sensitive cell group might be completely inhibited. In this experiment, 10"7 M ouabain, at which concentration a notch in the concentration-inhibition curve appeared, did not inhibit more than 50% of the total Ip in all cells (n=12). These results thus indicate that two isoforms exist in the same cell. Na+–K+–ATPase is composed of á and â subunits.9 The á subunit is a catalytically active subunit, whereas the function of the â subunit may be to assure the proper localization and assembly in the plasma membrane.14 Isoforms of the á subunit are encoded by separate genes;21 i.e. á1 subunit is more resistant than the á2 and á3 subunits for ouabaininduced inhibition.23,27,28 Therefore, in the present study, the isoform which was resistant to the inhibition of ouabain (IC50B =10.4 ìM) might be á1 isoform, and the other (IC50A =20 nM) á2 and/or á3 isoforms. The expression of the á3-subunit mRNA is characteristic in the neurons, whereas the á2-subunit transcripts are predominant in the glia.30 In the rat retina and optic nerve, a recent study demonstrated the presence of three isozymes (á1, á2 and á3) thus showing a complex expression pattern among the retinal cells, and also showing that the dissociated horizontal cells stain for the á1 and á3 subunits, while not staining for á2.15 In the present study in the carp horizontal cells, the Ip fractions of the ouabain low- and high- sensitivity subunits were 0.674 and 0.326, respectively (Fig. 6B). That means that the low-sensitive subunit exists almost twice as frequently as the high-sensitive ones in horizontal cells, but the functional difference of each isozyme was not investigated. The Na+–K+ pump not only maintains the intracellular ionic conditions but also has a direct action in the neuronal activities because of its electrogenic nature. Several studies have reported that the electrogenic Na+–K+ pump is accelerated after a longlasting spike discharge8 and glutamate-induced depolarization.25 This acceleration of the pump results in the membrane hyperpolarization and thus functions as a negative feedback system against

Na+–K+ pump in carp horizontal cells

over excitation. The pump also generates the Na+ gradient which provides the major driving force for high-affinity neurotransmitter uptake systems.3 The horizontal cells are second-order neurons that receive the excitatory synaptic inputs from photoreceptors.12 The resting membrane potential of the horizontal cells was about "20 to "30 mV in the dark.6,10,31 The cells respond to levels as low as "70 mV for a bright stimulus.10,32 Horizontal cells also possess a number of intrinsic time- and voltage-dependent channels, and some of these channels form the membrane potential of the horizontal cell. The Na+–K+ pump is electrogenic, so in neurons the turnover rate of the pump that transports the net charge is affected

239

by the membrane potential.4 But the possible role of the Na+–K+ pump in making the membrane potential of the cell has yet to be fully determined. The present study clearly shows the characterization of the electrogenic Na+–K+ pump in horizontal cells of the carp retina. Acknowledgements—The authors would like to thank Drs M. Munakata, Y. Hayashida and T. Yagi for their helpful comments and Dr B. Quinn for reading the manuscript. This study was supported by Grant-in-Aid for Scientific Research (No. 07407002) and Grant-in-Aid for Scientific Research on Priority Areas (No. 07276101) to N. Akaike from the Ministry of Education, Science and Culture, Japan.

REFERENCES

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

Akaike N. (1975) Contribution of an electrogenic sodium pump to membrane potential in mammalian skeletal muscle fibres. J. Physiol. 245, 499–520. Akaike N. and Harata N. (1994) Nystatin perforated patch recording and its applications to analyses of intracellular mechanisms. Jap. J. Physiol. 44, 433–473. Arrindell E. L., McKay B. S., Jaffe G. J. and Burke J. M. (1992) Modulation of potassium transport in cultured retinal pigment epithelium and retinal glial cells by serum and epidermal growth factor. Expl Cell Res. 203, 192–197. De Weer P., Gadsby D. C. and Rakowski R. F. (1988) Voltage dependence of the Na–K pump. A. 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. and Ripps H. (1973) Effect of magnesium on horizontal cell activity in the skate retina. Nature 242, 101–103. Gadsby D. C., Kimura J. and Noma A. (1985) Voltage dependence of Na/K pump current in isolated heart cells. Nature 315, 63–65. Gustafsson B. and Wigstro¨m H. (1983) Hyperpolarization following long lasting tetanic activation of hippocampal pyramidal cells. Brain Res. 275, 159–163. Jewell E. A. and 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, 16,925–16,930. Kaneko A. (1970) Physiological and morphological identification of horizontal, bipolar and amacrine cells in goldfish retina. J. Physiol. 207, 623–633. Kerkut G. A. and York B. (1969) The oxygen sensitivity of the electrogenic sodium pump in snail neurons. Comp. Biochem. Physiol. 28, 1125–1134. Lasater E. M. (1991) Membrane properties of distal retinal neurons. Prog. Retinal Res. 11, 215–246. MacKnight A. D. C. and Leaf A. (1977) Regulation of cellular volume. Physiol. Rev. 57, 510–573. McDonough A. A., Geering K. and Farley R. A. (1990) The sodium pump needs its beta subunit. Fedn Am. Socs exp. Biol. J. 4, 1598–1605. McGrail K. M. and Sweadner K. J. (1989) Complex expression patterns for Na+, K+–ATPase isoforms in retina and optic nerve. Eur. J. Neurosci. 2, 170–176. Murase K., Randic M., Shirasaki T., Nakagawa T. and Akaike N. (1990) Serotonin suppresses N-methyl--aspartate responses in acutely isolated spinal dorsal horn neurons of the rat. Brain Res. 525, 84–91. Newman E. A. (1987) Regulation of potassium levels by Muller cells in the vertebrate retina. Can. J. Physiol. Pharmac. 65, 1028–1034. Reichelt W., Dettmer D., Bruckner G., Brust P., Eberhardt W. and Reichenbach A. (1989) Potassium as a signal for both proliferation and differentiation of rabbit retinal (Muller) glia growing in cell culture. Cell Signal. 1, 187–194. Reichenbach A., Henke A., Eberhardt W. and Reichelt W. (1992) K+ ion regulation in retina. Can. J. Physiol. Pharmac. 70, S239–S247. Saari J. C. (1992) The biochemistry of sensory transduction in vertebrate photoreceptors In Adler’s Physiology of the Eye (ed. Hart W. M. Jr), Vol. 14, pp. 460–484. Mosby Year Book, St Louis, Missouri. Shull M. M. and Lingrel J. B. (1987) Multiple genes encode the human Na+, K+–ATPase catalytic subunit. Proc. natn. Acad. Sci. U.S.A. 84, 4039–4043. Steinberg R. H., Oakley B. and Niemeyer G. (1980) Light-evoked changes in [K+]o in retina of intact cat eye. J. Neurophysiol. 44, 897–921. Sweadner K. J. (1989) Isozymes of the Na/K–ATPase. Biochim. biophys. Acta 988, 185–220. Thomas R. C. (1972) Electrogenic sodium pump in nerve and muscle cells. Physiol. Rev. 52, 563–594. Thompson S. M. and Prince D. A. (1986) Activation of electrogenic sodium pump in hippocampal CA1 neurons following glutamate-induced hyperpolarization. J. Neurophysiol. 56, 507–522. Ullrich K. J. A. (1979) Sugar, amino acid, Na+ cotransport in the proximal tubule. A. Rev. Physiol. 41, 181–195. Urayama O., Shutt H. and 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. Urayama O. and Sweadner K. J. (1988) Ouabain sensitivity of the alpha 3 isozymes of rat Na, K–ATPase. Biochem. biophys. Res. Commun. 156, 796–800. Vasilets L. A. and Schwarz L. A. (1993) Structure-function relationships of cation binding in the Na+/K+–ATPase. Biochim. biophys. Acta 1154, 201–222.

240

M. Shimura et al.

Watts A. G., Watts G. S., Emanuel J. R. and Levenson R. (1991) Cell-specific expression of mRNAs encoding Na+, K+–ATPase a- and b-subunit isoforms within the rat central nervous system. Proc. natn. Acad. Sci. U.S.A. 88, 7425–7429. 31. Werblin F. S. and Dowling J. E. (1969) Organization of the retina of the mudpuppy, Necturus maculosus. II. Intracellular recording. J. Neurophysiol. 32, 339–355. 32. Yang G., Morgan I. G. and Dvorak D. (1989) Ganglion cell responses in ECMA-lesioned chicken retina. Neurosci. Lett. 30, Suppl. S142. 30.

(Accepted 5 January 1998)