Sulfonylurea-sensitive potassium current evoked by sodium-loading in rat midbrain dopamine neurons

~

Pergamon

0306-4522(95)00489-0

Neuroscience Vol. 71, No. 3, pp. 709-719, 1996 Elsevier ScienceLtd Copyright © 1996 IBRO Printed in Great Britain. All rights reserved 0306-4522/96 $15.00 + 0.00

SULFONYLUREA-SENSITIVE POTASSIUM CURRENT E V O K E D BY S O D I U M - L O A D I N G IN R A T M I D B R A I N DOPAMINE NEURONS V. S E U T I N , * K.-Z. S H E N , R. A. N O R T H t and S. W. J O H N S O N ~ Vollum Institute and Departments of Physiology & Pharmacology, and the Department of Neurology, Oregon Health Sciences University, Portland, OR 97201, U.S.A. A~traet--In Parkinson's disease, there is evidence of impaired mitochondrial function which reduces the capacity to synthesize ATP in dopamine neurons. This would be expected to reduce the activity of the sodium pump (Na +/K ÷ ATPase), causing increased intracellular levels of Na ÷ . Patch pipettes were used to introduce Na ÷ (40 mM in pipette solutions) into dopamine neurons in the rat midbrain slice in order to study the electrophysiological effects of increased intracellular Na ÷ . We found that intracellular Na + loading evoked 100-300 pA of outward current (at - 6 0 mV) and increased whole-cell conductance; these effects developed gradually during the first 10 min after rupture of the membrane patch. Extracellular Ba 2÷ reduced most of the outward current evoked by Na ÷ loading; this Ba2÷-sensitive current reversed direction at the expected reversal potential for K ÷ (E~), and was also blocked by extracellular tetraethylammonium (30 mM) and intracellular Cs ÷ (which replaced K ÷ in pipette solutions). The sulfonylurea drugs glipizide 0c50 = 4.9 nM), tolbutamide 0c50 = 23/aM) and glibenclamide (1 #M) were as effective as 300/aM Ba 2÷ in reducing the K ÷ current evoked by Na + loading. When recording with "control" pipettes containing 15 mM Na ÷ , diazoxide (300/a M) increased chord conductance and evoked outward current at - 60 mV, which also reversed direction near E K. Effects of diazoxide were blocked by glibenclamide (1 pM) or glipizide (300 nM). Diazoxide (300/aM) and baclofen (3/aM), which also evoked K÷-mediated outward currents recorded with control pipettes, caused little additional increases in outward currents during Na ÷ loading. Raising ATP concentrations to 10 mM in pipette solutions failed to significantly reduce currents evoked by diazoxide or Na ÷ loading, suggesting that these currents may not be mediated by ATP-sensitive K ÷ channels. Finally, Na + loading using pipettes containing Cs ÷ in place of K ÷ evoked a relatively small outward current (50-150 pA at - 6 0 mV), which developed gradually over the first 10 min after rupturing the membrane patch. This current was reduced by dihydro-ouabain (3/aM) and a low extracellular concentration of K ÷ (0.5 mM instead of 2.5 mM), but was not affected by Ba2+. We conclude that intracellular Na + loading evokes a current generated by Na+/K ÷ ATPase in addition to sulfonylurea-sensitive K + current. This Na+-dependent K ÷ current is unusual in its sensitivity to sulfonylureas, and could protect dopamine neurons against toxic effects of intracellular Na ÷ accumulation. Key words: substantia nigra, ventral tegrnental area, diazoxide, sodium pump, glibenclamide, tolbutamide

Degeneration o f d o p a m i n e - c o n t a i n i n g n e u r o n s in the substantia nigra causes muscular rigidity, tremor, bradykinesia and postural instability o f P a r k i n s o n ' s disease. Results o f several studies suggest that a defect in complex I in m i t o c h o n d r i a reduces the capacity to synthesize ATP, which lowers the threshold for inducing death o f d o p a m i n e - c o n t a i n i n g neurons in P a r k i n s o n ' s disease. 3s,4°,'~ One i m p o r t a n t consequence o f energy depletion is increased intracellular N a ÷ c o n c e n t r a t i o n due to the diminished activity o f N a + / K ÷ ATPase, which is needed to

maintain intracellular N a ÷ homeostasis, t3 In order to study the electrophysiological effects o f intracellular N a + loading, we used patch pipettes to introduce N a ÷ into single d o p a m i n e neurons in the rat midbrain slice. A preliminary report o f these studies has been published in abstract form. 3~ EXPERIMENTAL PROCEDURES

Tissue preparation

*Present address: Department of Pharmacology, Tour de Pathologie (B23), University of Lirge, Belgium. tPresent address: Glaxo Institute for Molecular Biology, 1228 Plan-les-Ouates, Geneva, Switzerland. ~To whom correspondence should be addressed. Abbreviations: EGTA, ethylene glycol bis (aminoethyl ether) tetra-acetate; HEPES, N-2-hydroxyethylpiperazine-N'2-ethanesulfonic acid; TEA, tetraethylammonium. 709

Male Sprague-Dawley rats (Bantin & Kingman, WA, U.S.A.) were anesthetized with halothane and killed by severing major thoracic vessels in accordance with institutional guidelines. Midbrain slices were prepared from rats as described previously.27,29 Briefly, the scalp and skull were removed and the brain was excised. A block of tissue containing the midbrain was placed ventral surface up in a Vibratome (Series 1000, Technical Products International, Inc., St Louis, MO) and serial horizontal slices (thickness 300 #m) were cut starting at the ventral surface. Slices were

710

V. Seutin et al.

discarded until the floor of the interpeduncular fossa was reached; the next slice was placed on a nylon mesh in a recording chamber (volume 500/~1). The tissue was held in position with two electron microscopy grids, which were weighed down by short (2-3 m m ) pieces of 0.5 m m diameter platinum wire. The slice was immersed in a continuously flowing (2 ml/min), heated (36°C) solution containing (in mM): NaC1, 126; KC1, 2.5; NaH2PO4, 1.2; MgC12, 1.2; CaCI 2, 2.4; glucose, 11; N a H C O 3 , 18; saturated with 95% 02 and 5% CO 2 (pH 7.4).

Whole-cell recordings Recordings were made in the whole-cell configuration from neurons in the ventral tegmental area or substantia nigra zona c o m p a c t a ? 6 The "control" pipette solution contained (mM): potassium gluconate, 125; NaC1, 15; CaC12, 1; MgCI,, 2; HEPES, 10; E G T A , 11; K2ATP , 1.5; N % G T P , 0.3. In some experiments, potassium gluconate was replaced by cesium gluconate. The pH of pipette solutions were adjusted to 7.25 with either K O H or CsOH. Osmolalities of pipette solutions were measured with a Wescor 500 vapor pressure o s m o m e t e r (Logan, UT) and adjusted to 290mOsmol/1 by adding either a solution of 15 m M NaCI or one containing 25 m M sodium gluconate and 15 m M NaC1. The osmolality of the external solution was 295-300 mOsmol/l. Patch pipettes were made from 1.5 m m o.d. borosilicate glass capillary tubing (WPI, New Haven, CT) pulled on a Flaming Brown Micropipette Puller (Model P-80/PC, Sutter Instrument Co., San Rafael, CA). Pipettes had resistances of 2 - 4 Mr2 and outer tip diameters o f 1/~m. After forming a gigaseal and breaking into the cell, voltage-clamped membrane currents were amplified by an Axopatch IB amplifier (Axon Instruments, Foster City, CA). Stated voltages are corrected for liquid junction potential. Series resistance (R~) was monitored repeatedly by measuring the time course of currents during 1 0 m V hyperpolarizing c o m m a n d s lasting 20 ms. Series resistance was calculated by the equation R S= r/Cm, where r is the time constant of the current change in response to the voltage step and Cm is whole-cell capacitance obtained by integrating the capacity current transient over time) Time constants and capacity currents were measured using the A x o G r a p h curve-fitting program (Axon Instruments) on a Macintosh computer. Na + loading failed to evoke an outward current when R~ > 20 Mf~; this was presumably because poor access inhibited the movement of Na + into the cell. Therefore, data from such recordings were excluded.

Sodium loading The term " N a + loading" is used t h r o u g h o u t the text to refer to the process of recording from neurons in the whole-cell configuration using patch pipettes containing 4 0 m M Na ÷ instead of the usual 15raM Na +. In these experiments, the concentration o f Na ÷ in pipette solutions was raised to 4 0 m M by replacing 2 5 m M potassium (or cesium) gluconate with sodium gluconate, while the concentration of NaCl (15 m M ) was not altered. M o s t data from Na + loading experiments were acquired at least 10 min after rupturing the m e m b r a n e patch; this allowed changes in chord conductance to approach a steady state. The term "control" pipettes is used t h r o u g h o u t the text to refer to patch pipettes containing 15 m M Na ÷ .

Identification of cell type Most recordings were m a d e from "principal" neurons; they characteristically fire action currents spontaneously at 1-5 Hz, generate K+-mediated outward currents (at - 6 0 m V ) in response to dopamine (30/IM) but not to [MetS]enkephalin (10 # M), and have the hyperpolarizationactivated time-dependent inward current k n o w n as I h. Principal neurons contain tyrosine hydroxylase and are presumably dopaminergic. 23'z935"67 Recordings were also

made from "secondary" cells, which are characterized by a lack of spontaneous firing of action currents, the generation of K+-mediated outward current in response to [MetS]enkephalin, but not to dopamine, and little or no I h. Neurons with these electrophysiological characteristics do not contain tyrosine hydroxylase 233s,67 and are presumed to be GABA-containing interneurons? 8 In the present study, a cell was classified as a "principal" neuron if a hyperpolarizing voltage step (from --60 to -- 120 mV, 250 ms duration) evoked > 200 p A of Ih, whereas little or no I h (0-50 pA) indicated a " s e c o n d a r y " neuron. Responses to dopamine (30 # M ) and [MetS]enkephalin (10/~M) were used to classify cell type when amplitudes o f l h were between 50 and 200 pA.

Current-voltage studies Current-voltage plots were obtained by measuring currents during hyperpolarizing voltage steps (10 60mV, duration 250ms) from a holding potential of - 6 0 m V . Currents were measured immediately after capacitative transients to minimize the influence of I h on current-voltage plots. In some figures, current-voltage plots show "subtracted" currents in which currents recorded during the experimental treatment have been subtracted from the average of those currents recorded before (control) and after (recovery) the experimental treatment. Therefore, subtracted currents represent " n e t " currents produced (or blocked) by an experimental condition or treatment.

Materials Apamin, baclofen, dihydro-ouabain, dopamine hydrochloride, [MetS]enkephalin, tetraethylammonium chloride (TEA), tetrodotoxin and tolbutamide were obtained from Sigma Chemical Co. (St Louis, MO); diazoxide, glibenclamide and glipizide were obtained from Research Biochemicals Inc. (Natick, MA). Stock solutions of dopamine were m a d e daily and kept on ice to retard oxidation. Stock solutions of glipizide, glibenclamide and tolbutamide were m a d e in dimethyl sulfoxide then diluted at least 1:1000 before use. Control solutions of 1:1000 dimethyl sulfoxide had no effect on m e m b r a n e current.

Statistics Numerical data in the text and error bars in figures are expressed as the mean _+ S.E.M. In current-voltage plots, the chord conductance (slope of the current-voltage curve) and reversal potential (Erev) were determined by linear regression for each cell, and mean chord conductance and Ere` were calculated by averaging the results from all cells. Two-way analysis of variance (ANOVA) with repeated measures was used to test for significant differences between current-voltage curves (SigmaStat, Jandel Scientific, San Rafael, CA); significant differences between data points were evaluated using the pairwise multiple comparison method of S t u d e n t - N e w m a n - K e u l s . Two-tailed t-tests were used to test for significant differences between chord conductances, holding currents and series resistances of pipettes. In evaluating concentration-dependent drug effects, an I%0 was calculated for each cell using the KaleidaGraph curve-fitting program (Synergy Software, Reading, PA) on a Macintosh computer; concentration-response curves were fitted to the equation y = ax/(x + b), where y is the magnitude of drug effect, a is the m a x i m u m drug effect, x is concentration of drug and b is the IC50. The mean lcs0 and S.E.M. were calculated by averaging the 1c50 values determined for all cells. RESULTS

General effects o f N a + loading in dopamine neurons When the patch pipette contained a higher than n o r m a l c o n c e n t r a t i o n o f N a ÷ (40 m M in place o f

Intracellular Na ÷ evokes sulfonylurea-sensitive K ÷ currents 15 m M ) , a net o u t w a r d current (at - 6 0 m V ) slowly developed after r u p t u r i n g the m e m b r a n e p a t c h (Fig. 1A). Ten m i n u t e s after b e g i n n i n g recording in the whole-cell configuration, a significantly larger holding current ( 1 8 9 + 1 8 p A , n = 4 6 ) was required to clamp m e m b r a n e potential at - 60 m V in N a + - l o a d e d n e u r o n s c o m p a r e d with the c u r r e n t ( - 2 3 + 20 pA, n = 14) required in n e u r o n s recorded with control pipettes (i.e. pipettes c o n t a i n i n g 15 m M N a ÷) ( P < 0.0001). As illustrated in the curr e n t - v o l t a g e plots in Fig. I B, the m e a n c h o r d c o n d u c t a n c e following N a ÷ loading (8.18 + 0.43 nS, n = 46) was significantly greater t h a n t h a t m e a s u r e d using control pipettes (5.68 + 0 . 6 9 n S , n = 14) ( P = 0.0019). It should be n o t e d t h a t currents evoked by N a + loading were in the o u t w a r d direction c o m p a r e d with those recorded using control pipettes, a l t h o u g h currents evoked by N a + loading were smaller at more h y p e r p o l a r i z e d test potentials (see Fig. IB). Series resistance in pipettes c o n t a i n i n g 40 m M N a ÷ (13.1 + 0.5 MD, n = 46) was not significantly different from t h a t in pipettes

A •

*

mie

o mm (i 5 ~ N=*)

8az* (300 i~ L

5 m~

10 m~

B

300-

200 i I00

U

.

.

I0

.

.

.

1 O0

.

1000

Barium( p M ) Ba 2+ ( 1 0 0 i.tM)

I 100

pA

5 min

B 4 0 rnM N a *

~

200

.

1 5 m M Na +

mV , -130

asaz, (3OO.ML

lOO ms

0

A

711

;z ,

,

,

, / ,

~ ~ ~ "

Fig. 2. Barium blocks current evoked by Na + loading. (A) Current traces from two dopamine neurons recorded with pipettes containing 40 or 15 mM Na ÷. In all records, the holding potential was - 6 0 m V except during hyperpolarizing voltage steps (10-60 mV, 250ms duration; see inset). Membrane conductance increases over time in the cell recorded with 40 mM Na ÷ (upper traces), but not when the pipette contained 15 mM Na ÷ (lower traces). Barium significantly reduced whole-cell conductance in the neuron loaded with Na ÷. The broken line indicates zero current. (B) Barium concentration-response curve for inhibition of outward current (at - 6 0 m V ) evoked by Na ÷ loading (40 mM in pipettes). Each data point is the mean + S.E.M. of six cells.

~//'==

, ~ . ~

~

0

pA

containing ( P = 0.42).

15mM

N a ÷ (12.1 + 1.1Mf~,

n = 14)

Ba 2+- sensitive current evoked by N a + loading ~~ 0 O

-400 Fig. 1. lntracellular Na ÷ loading evokes outward currents in dopamine neurons. (A) Membrane current (at - 6 0 mV) recorded with a pipette containing 40 mM Na + . An outward current develops during the first 10 min after rupturing the membrane patch (arrow); this outward current is largely reduced by Ba 2÷ (solid bar). The broken line represents zero current. Vertical lines in the figure were caused by passing current for determining access resistance or current-voltage plots. (B) Whole-cell current-voltage plots in neurons recorded with pipettes containing 40 mM (n = 46) or 15 mM (n = 14) Na +. Na ÷ loading evokes an outward current at all test potentials, but current produced by Na ÷ loading is greatest at less hyperpolarized potentials. All measurements were made 10min after rupturing the membrane patch. Asterisks signify data points which were significantly different from control•values (P < 0.05).

Ba 2+ (300 # M ) blocked most of the o u t w a r d currents p r o d u c e d by N a + loading (Figs 1A, 2A), but h a d little effect when n e u r o n s were recorded with control pipettes (Fig. 2A). Effects of Ba 2+ were c o n c e n t r a t i o n dependent, with an Ic~0 of 84 _ 24 p M (n = 6); these data, summarized in Fig. 2B, were o b t a i n e d at least 10 m i n after r u p t u r i n g m e m b r a n e patches in N a + - l o a d e d d o p a m i n e neurons. Effects o f Ba 2+ were usually completely reversible 5 - 1 0 m i n after washout, a l t h o u g h recovery was sometimes incomplete after applying the highest c o n c e n t r a t i o n of Ba 2+ ( 1 0 0 0 # M ) . As s h o w n in the subtracted c u r r e n t - v o l t a g e plots in Fig. 3A, Ba 2+ ( 1 0 0 p M ) caused a significantly greater reduction in c h o r d c o n d u c t a n c e (4.21 _+ 0.53 nS, n = 12) following N a ÷ loading c o m p a r e d with the reduction in c o n d u c t a n c e recorded with control pipettes (0.99 _+ 0.18 nS, n = 5)

V. Seutin et al.

712

(P < 0.0001). In neurons loaded with N a + , the Ba 2+sensitive current reversed direction ( - 103 +_ 5 mV; n = 12; Fig. 3A) near the expected reversal potential (Erev) for K + as predicted by the N e r n s t equation (-98mV, with 100raM K + in the pipette and 2.5 m M in the superfusate).

A 0 rain

5 rain

10 rain

Baz+ 300 pM

500pA 100ms

Effects o f other K + channel blockers Subtracted c u r r e n t - v o l t a g e plots in Fig. 3B show that T E A (30raM) blocked significantly more outward current during N a + loading c o m p a r e d with that recorded with control pipettes (P <0.05). Measured 10rain after r u p t u r i n g the m e m b r a n e patch, T E A blocked significantly more o u t w a r d current at - 6 0 m V in N a + loaded neurons (180_+63pA, n = 5) c o m p a r e d with that recorded with control pipettes (18 _+ 4 p A , n = 4) (P = 0.037). T E A also p r o d u c e d a greater decrease in c h o r d c o n d u c t a n c e following N a + loading (4.49 _+ 1.22 nS, n = 5) c o m p a r e d with the reduction in c o n d u c t a n c e recorded with control pipettes (1.50 + 0.50 nS, n = 4), but this difference was n o t statistically significant ( P = 0.06). The Erev o f the TEA-sensitive current

A

~r

200

40 mM Na+

I O0

mV

-130 ~

~

-60

0 pA

B

200 40 mM Na +

100

mV ' -130 ~~"

~"

~

-

w

~

-gO

0 pA -100

Fig. 4. Na + loading fails to evoke Ba2+-sensitive outward currents in the presence of intracellular Cs ÷. (A) Current traces recorded during a series of hyperpolarizing voltage steps using a pipette containing 40raM Na + and cesium gluconate (100 mM). The broken line indicates zero current. Note that the small outward current which develops over 10min is not blocked by Ba 2+. Also note that the timedependent inward current (lh) is not blocked by Cs + in the pipette. (B) Subtracted Ba2+-sensitive currents using pipettes containing 40raM Na + and cesium gluconate. These currents were calculated by subtracting currents recorded in Ba 2+ (300#M) from the average of those currents recorded before and after Ba2+ at each test potential in five cells. The current-voltage curve has zero slope because intracellular Cs + completely blocks K + currents evoked by Na + loading.

-100 g

W 40 mM Na+ 15 mM Na+

200

T 2 .

~

~_

100

mV, , -130 ~

,

0

0

pA

-100 Fig. 3. Barium and TEA block Na+-evoked currents which reverse direction near the expected reversal potential for K +. (A) Subtracted currents blocked by 100 #M Ba 2÷ recorded at various test potentials with pipettes containing either 40mM (n = 12)or 15mM (n = 5 ) N a ÷. (B) Subtracted currents blocked by 30mM TEA recorded with pipettes containing either 40mM (n = 5) or 15 mM (n = 4) Na ÷. Currents in A and B were calculated by subtracting currents recorded during the experimental condition (Ba 2+ or TEA) from the average of those recorded before and after the experimental condition at each test potential for each cell. Asterisks signify those test potentials at which data points were significantly different (P < 0.05).

evoked by N a + loading ( - 9 3 _+ 10 mV, n = 4) was near that expected for K +. Because high concentrations o f intracellular N a + are k n o w n to reduce the activity o f the N a + / C a 2+ exchanger and thereby raise the level o f intracellular Ca 2+,3'8'59 we next considered the possibility that the outward current evoked by Na + loading was a Ca2+-dependent K + current. However, currents evoked by N a + loading were not reduced by either Cd 2+ (100 p M ; n = 4), a nonselective Ca -,+ channel blocker, or apamin (300 nM; n = 3), which blocks CaZ+-dependent K + currents in d o p a m i n e neurons. 5s'56 These data suggest that Ca 2+activated K + channels do not underlie outward currents evoked by N a + loading. W h e n Cs + replaced K + in pipettes, Na + loading p r o d u c e d a small o u t w a r d current which was not sensitive to Ba 2+ ( 3 0 0 # M ) (Fig. 4A). The lack of Ba2+-sensitive currents is further illustrated in the subtracted c u r r e n t - v o l t a g e plot in Fig. 4B; Ba 2+ ( 3 0 0 p M ) reduced the chord conductance by only 0.19_ 0.13 nS (n = 5), which is not significantly different from zero. Intracellular Cs + blocks most K + channels, 5° which m o s t likely explains its ability to prevent the a p p e a r a n c e o f Ba2+-sensitive currents

lntracellular Na + evokes sulfonylurea-sensitive K + currents following N a + loading. It should also be noted that neither intracellular Cs + (Fig. 4A) nor extracellular T E A or Ba 2+ (Fig. 2A) reduced the magnitude of lh; these findings argue against the possibility that outward currents evoked by Na ÷ loading are caused by reducing lh inward currents.

Effects o f sulfonylureas in dopamine neurons As shown in Fig. 5A, tolbutamide reversibly blocked the outward current at - 6 0 m V and dramatically reduced membrane conductance, which had been increased following Na + loading. The sulfonylurea agents glipizide and glibenclamide were also effective, but recovery from these drugs was incomplete 2 h after washout. Due to considerable cellto-cell variation in the amount of outward currents evoked by Na + loading, sulfonylurea-sensitive currents shown in Fig. 5B are expressed as percentages of currents blocked by a standard concentration (300/aM) of Ba 2+. [This concentration of Ba 2+ was chosen because it inhibited most (about 85% of the current evoked by Na + loading, and it washed out more quickly and completely than did higher concen-

A

I --

--

--

B

--

~

.

.

.

.

100

'~

8o 60

~ 40 ~

2o 0

i -9

-7

Concentration

-5

t

i -3

(log M )

Fig. 5. Sulfonylurea drugs block currents evoked by Na + loading. (A) The conductance increase evoked by Na ÷ loading (40 mM in pipette) is blocked by Ba 2+ and tolbutamide. Currents were measured during a series of hyperpolarizing voltage steps; all current traces were recorded from the same dopamine neuron. Broken lines indicate zero current. (B) Concentration-response curves for glipizideand tolbutamide-induced inhibition of outward current (at - 6 0 mV) evoked by Na + loading. Effects of sulfonylureas were normalized to the effect produced by Ba 2÷ (300/aM) in each cell. Concentrations of drugs were increased cumulatively. Each data point represents the mean + S.E.M. of three to six dopamine neurons. NSC 7 1 / 3 - - E

trations.] Expressing the data in this way, glipizide (1/aM) blocked 100 +_ 3% (n = 3) of the Ba2+-sensi tive current (at - 6 0 m V ) , glibenclamide ( l ~ l M ) blocked 96_+ 6% (n = 4 ) and tolbutamide (1 m M ) blocked 92 _+ 2% (n = 4). Figure 5B shows concentration-response curves for glipizide and tolbutamide on inhibition of outward current (at - 6 0 m V ) evoked by Na + loading; these data have also been normalized for effects produced by Ba 2+ (300/aM). The ICs0 for glipizide was 4.9 _+ 1.4 nM (n = 5) and that for tolbutamide was 22.8 + 7.6 # M (n = 4); these values are similar to the Ka values for glipizide (2 nM) and tolbutamide (7/aM) obtained from binding studies in brain microsomes. 6 Reversal potentials for the subtracted currents were close to that expected for K + (glibenclamide - 9 8 _ 5 mV, n = 4); glipizide -107+2mV, n=7; tolbutamide - 9 5 + _ 6 m V , n = 6). Sulfonylureas had no effect on m e m b r a n e conductance when cells were recorded with control pipettes, as reported by others. 25 Because sulfonylureas block ATP-sensitive K + channels in many neuronal and non-neuronal tissues, 57 we tested the possibility that these channels might produce the K + currents evoked by Na + loading. However, ATP, which blocks sulfonylureasensitive K + channels in other cells, 57 apparently did not block current generated by Na + loading because our patch pipettes always contained t.5 m M A T P . Moreover, a significant amount of Ba 2+ (300/aM)sensitive outward current (226-t-53 pA, n = 4) was recorded (at - 6 0 mV) during Na + loading when the concentration of A T P in pipette solutions was raised to 1 0 m M . It is possible that A T P failed to diffuse into cells, but this seems unlikely because the series resistance of these pipettes was relatively low (12.8 _+ 2.7 Mr1, n = 4).

Potassium channel openers

Glipizide - - )-- Tolbutamide

..E: .E

713

Diazoxide (300/aM), which opens sulfonylureasensitive K + channels in pancreatic /~-cells and smooth muscle, 2 significantly increased the outward current at - 6 0 mV (194 + 42 pA, n = 8) and wholecell conductance (3.86 + 0.78 nS, n = 8) in dopamine neurons recorded with control pipettes, as seen in the subtracted current-voltage plot in Fig. 6A. Currents evoked by diazoxide reversed direction at -102___ 5 m V (n = 8 ) , and diazoxide failed to increase chord conductance when Cs ÷ replaced K ÷ in the pipette (n = 3). These data suggest that currents evoked by diazoxide are mediated by K ÷. Glibenclamide (1 # M ) completely blocked outward currents produced by diazoxide (300/aM), as shown in Fig. 6B. The effect of glibenclamide reached a m a x i m u m in 3-7 min, but recovery was incomplete 1 h after washout. The whole-cell current-voltage plots in Fig. 6C were all recorded from the same three dopamine neurons; they show that diazoxide (300/aM) significantly increased chord conductance from 3.05 4-0.34 nS to 6.95-t-1.17 nS (P < 0.0001), and the addition of glibenclamide (1/aM) to

714

V. Seutin et aL 200

A

I00

mV

0 pA

-I 3'0 i-lO0

B

So ms

C

~

Control Diazoxide

- ~-

Glibenclamide & diazoxide

. ~

200 1 O0

mL . . . .

0 pA -I O0 -200

1 0 m M , the diazoxide ( 3 0 0 / i M ) - i n d u c e d outward current (126 _+ 24 pA at --60 mV, n = 5) and increase in c h o r d c o n d u c t a n c e (2.30 ___0.22 nS, n = 5) were not significantly different from those produced by diazoxide using pipettes containing 1.5 m M ATP (P > 0 . 1 ) . On the other hand, diazoxide (300/tM) p r o d u c e d no significant o u t w a r d current (10 _+ 14 pA, n = 5) or increase in chord conductance (0.27 _+ 0.37 nS, n = 5) in d o p a m i n e neurons following N a + loading. Therefore, it appears that effects of diazoxide are prevented or masked by Na + loading. Baclofen (3/~M), a GABAB receptor agonist, evoked a significant outward current (190 + 37 pA at - 6 0 mV, n = 6) and an increase in chord conductance ( 5 . 0 0 _ 1.07 nS, n = 6) in d o p a m i n e neurons recorded with control pipettes. As seen in the subtracted c u r r e n t - v o l t a g e plots in Fig. 7A, the baclofen-induced currents reversed direction ( - 102 + 3 mV, n = 6) near the expected reversal potential for K +, in agreement with results o f others. 34 Figure 7A also shows that glibenclamide (1 ~ M ) had no significant effect on baclofen ( 3 / t M ) - i n d u c e d currents (n = 4; P > 0.7). However, baclofen evoked significantly less outward current (38_+ 1 6 p A at - 60 mV, n = 6) and produced a smaller increase in chord c o n d u c t a n c e (0.86 _+ 0.31 nS, n = 6) (P < 0.01) in the presence o f outward currents evoked by N a + loading. However, as seen in Fig. 7A and B, the

-300

Fig. 6. Diazoxide evokes sulfonylurea-sensitive K ÷ currents in dopamine neurons. (A) Currents produced by diazoxide (300 MM) reverse direction near the expected reversal potential for K + (n = 8). This current-voltage relationship was calculated by subtracting the average of currents recorded before and after diazoxide from those recorded in the presence of diazoxide at each test potential for each celt. Diazoxide evoked a significant increase in conductance (P = 0.005) which returned to control level 20-30 min after washout. (B) Currents evoked by hyperpolarizing voltage steps show that the diazoxide-induced increase in membrane conductance is blocked by adding glibenclamide to the superfusate. All records are from the same neuron; the broken line indicates zero current. (C) These current-voltage plots illustrate the ability of glibenclamide (1 pM) to block currents evoked by diazoxide (300pM). These non-subtracted whole-cell currents were obtained from the same three dopamine neurons. Asterisks indicate test potentials at which effects of diazoxide are significantly different from both control data and data obtained in glibenclamide plus diazoxide (P < 0.05). Neurons in A-C were recorded with control pipettes containing 15 mM Na +.

diazoxide in the superfusate returned the c h o r d conductance to the initial level (2.84 ___0. ! 6 nS). Glipizide (300 nM) also rapidly and completely blocked the conductance increase evoked by diazoxide (n = 2; data not shown). Diazoxide, which has been reported to open ATPsensitive K ÷ channels in o t h e r cells, 5L52~63 increased chord conductance in d o p a m i n e neurons despite the presence o f 1.5 m M A T P in control pipettes. W h e n the A T P c o n c e n t r a t i o n in pipettes was raised to

A

300

Control --C)-- With 91ibenclam~

200 I00

mV

,

-I 30

'

,

~

,

~-'~""¢~

'

,

,

0 pA

-60 -I00 -200

B

Baclolen(3 ~4) & r

m

*

,

am

so ms Fig. 7. Baclofen (3#M) evokes glibenclamide-insensitive currents. (A) Glibenclamide (1/~M) does not significantly change currents evoked by baclofen in dopamine neurons recorded with control pipettes. These currents were calculated by subtracting the average of currents recorded before and after baclofen from those recorded in the presence of baclofen. Filled circles represent data recorded with control pipettes (with 15 mM Na+; n = 4) and open circles represent data obtained following Na ÷ loading during superfusion with glibenclamide (n = 6). (B) Baclofen evokes an outward current (--60mV) and increases membrane conductance after the effects of Na + loading are blocked by glibenclamide. Each trace is from the same dopamine neuron; the broken line indicates zero current.

Intracellular Na + evokes sulfonylurea-sensitive K + currents efficacy of baclofen (3 p M ) could be restored once the K ÷ currents evoked by N a t loading were blocked by glibenclamide (1 # M ) . Finally, raising the conc e n t r a t i o n of A T P to 10 m M in pipettes (n = 3) h a d no significant effect on baclofen-induced o u t w a r d currents or c h o r d c o n d u c t a n c e c o m p a r e d with those recorded with control pipettes ( P = 0.29).

A

715

-I 30

-6O

i

-200

-400

Sodium pump A l t h o u g h the substitution of Cs ÷ for K ÷ in pipette solutions prevented the a p p e a r a n c e o f BaZ+-sensitive currents following N a ÷ loading, 5 0 - 1 5 0 p A o f o u t w a r d current (at - 6 0 mV) still developed slowly d u r i n g the first 10 rain after r u p t u r i n g the m e m b r a n e p a t c h (Fig. 4A). Using C s ÷ - c o n t a i n i n g pipettes, the h o l d i n g current for d o p a m i n e n e u r o n s r e c o r d e d following N a ÷ loading was significantly m o r e o u t w a r d ( - 40 _ 22 p A at - 60 mV, n = 7) c o m p a r e d with t h a t for neurons recorded with control pipettes ( - 148 ± 8 pA, n = 5) ( P < 0.001). Figure 8A shows t h a t the whole-cell c u r r e n t - v o l t a g e relationship for N a + - l o a d e d neurons was significantly shifted in a n o u t w a r d direction with respect to t h a t o b t a i n e d in n e u r o n s n o t loaded with N a ÷ (P < 0.0001). H o w e v e r , the c h o r d c o n d u c t a n c e of N a + - l o a d e d n e u r o n s (5.19 + 0.80 nS, n = 7) was n o t significantly different f r o m t h a t recorded from n e u r o n s with control pipettes (5.38 + 0.69, n = 5) ( P = 0.86). These d a t a suggest t h a t N a ÷ loading, in the presence o f Cs +, produces a n o u t w a r d current with n o significant voltage dependence; this is seen as a parallel o u t w a r d shift in the two c u r r e n t - v o l t a g e curves in Fig. 8A. D i h y d r o - o u a b a i n ( 3 # M ) , which blocks N a + / K ÷ ATPase, p r o d u c e d a significantly greater i n w a r d current in N a + - l o a d e d n e u r o n s ( - 1 5 4 + 1 4 p A , n = 3) c o m p a r e d with the inward c u r r e n t in n e u r o n s recorded with control (Cs+-containing) pipettes (60 + 2 2 p A , n = 3 ) ( P = 0 . 0 2 3 ) . As seen in the " s u b t r a c t e d " c u r r e n t - v o l t a g e plots in Fig. 8B, N a ÷ loading significantly increased the a m p l i t u d e o f dihydro-ouabain-sensitive currents c o m p a r e d with those recorded in n e u r o n s with control pipettes ( P = 0.025). Dihydro-ouabain-sensitive currents d u r i n g N a t loading h a d a positive slope c o n d u c t a n c e (1.36 ± 0.22 nS, n = 3), whereas d i h y d r o - o u a b a i n sensitive currents recorded with c o n t r o l pipettes did n o t (see Fig. 8B). D i h y d r o - o u a b a i n was perfused for 10 min, a n d its effect was partially reversible 30 rain after washout. R e d u c i n g the extracellular K ÷ c o n c e n t r a t i o n f r o m 2.5 to 0.5 m M p r o d u c e d significantly m o r e i n w a r d current (at - 6 0 m V ) in N a ÷ - l o a d e d n e u r o n s (-78±17pA at - 6 0 m V , n = 6 ) c o m p a r e d with inward currents recorded with control (Cs+-filled) pipettes ( - - 2 1 + 8 p A , n = 5) ( P = 0 . 0 1 4 ) . As seen in the " s u b t r a c t e d " c u r r e n t - v o l t a g e plots in Fig. 8C, 0 . 5 r a M K ÷ blocked significantly m o r e c u r r e n t following N a ÷ loading c o m p a r e d with t h a t recorded from n e u r o n s with control pipettes ( P = 0 . 0 2 1 5 ) . C u r r e n t s blocked by low K ÷ following N a ÷ loading

0 pA

rnV

15 mM Na+

600

B

200 W

W

40 rnM Na*

1 O0

mV

-1

pA .

C

.

.

.

.

.

- 6'0

100

40 mM Na+

mY@ ' -130

~Y

÷

-"

±

I I-60

0 pA

Fig. 8. Na t loading evokes a ouabain-sensitive pump current. All recordings were made with pipettes containing Cs ÷ in place of K ÷. (A) Whole-cell current-voltage plots recorded with pipettes containing 40 mM (n = 7) or 15 mM (n = 5 ) Na t . These whole-cell currents were recorded 10min after rupturing membrane patches. Na + loading evoked significantly more outward current than did those cells recorded with control pipettes (P < 0.0001). (B) Subtracted current-voltage plots showing that dihydro-ouabain (3 #M) blocks significantly more outward current in Na ÷ loaded neurons (n = 3) compared with neurons not loaded with Na ÷ (n = 3) (P = 0.0252). Asterisks indicate test potentials at which effects of dihydro-ouabain recorded during Na ÷ loading were significantly different from those recorded with control pipettes (P <0.05). (C) Subtracted current-voltage plots showing that a reduced extraeellular concentration of K ÷ (0.5 raM) blocks significantly more outward current in Na+-loaded neurons (n = 6) compared with neurons not loaded with Na ÷ (n = 5) (P = 0.0215). Tetrodotoxin (0.5#M) was added to the superfusate to block spikes which appeared in 0.5 mM K +.

h a d a positive slope c o n d u c t a n c e ( 0 . 8 8 _ 0.30nS, n = 6), in c o n t r a s t to those recorded in n e u r o n s not loaded with N a ÷ . Low K ÷ was perfused for 10 min, a n d its effect was completely reversible 5 m i n after

716

V. Seutin et al.

returning to 2 . 5 m M K +. Because both dihydroouabain and low K + inhibit N a + / K + ATPase, 42~ these data suggest that Na ÷ loading increases the current generated by the electrogenic sodium pump. Glibenclamide (1/~M) had no effect on outward currents evoked by Na + loading with Cs+-filled pipettes (n = 2), suggesting that sulfonylureas do not affect the sodium pump.

Effects of Na + loading in secondary neurons Na + loading in secondary neurons evoked 100-300 pA of outward current (at - 60 mV) during the first 10 min after rupturing the m e m b r a n e patch. Ba 2+ ( 3 0 0 ~ M ) reduced the whole-cell conductance by 2.62 ___0.64 nS (n = 10) in neurons following Na + loading; this is significantly greater than the reduction in conductance (0.92 ___0.05 nS, n = 3) produced by Ba 2+ when recording with control pipettes (P =0.02). The subtracted Ba2+-sensitive current evoked by Na + loading reversed direction at - 1 0 9 _ + 4 m V ( n - - 1 0 ) , which is consistent with a current mediated by K +. Tolbutamide (100 +~tM) reduced 87_+5% (n = 3) of the outward current evoked by Na + loading (normalized to the reduction in outward current produced by 3 0 0 ~ M Ba2+). Therefore, it appears that a raised intracellular concentration of Na + also evokes a sulfonylureasensitive K ÷ current in secondary neurons. DISCUSSION

Our results clearly show that intracellular loading of Na + evokes a sulfonylurea-sensitive K + current in dopamine neurons. Although there is a high density of binding sites for sulfonylureas in midbrain regions containing cell bodies and dendrites of dopamine neurons, 4~'62 the majority of binding sites are located on nerve terminals of the striatonigral p a t h w a y ? 6 However, our results cannot be explained by a presynaptic site of action because sulfonylurea drugs had no effect on membrane conductance unless N a + was introduced postsynaptically into dopamine neurons. We suggest that the block of an Na+-depen dent K + conductance in dopamine neurons may be an important effect of sulfonylurea drugs in the brain. ATP-sensitive K + channels are blocked by sulfonylureas in many neuronal and non-neuronal cells, 57 which raises the possibility that currents generated by Na + loading are mediated by these channels in dopamine neurons. Moreover, conditions which reduce levels ofintracellular A T P are known to evoke sulfonylurea-sensitive K + current in dopamine neurons, 2449'58 and A T P ( 2 m M ) closes these K + channels. 24 Because Na + increases A T P utilization by increasing the activity of the sodium pump, one might expect that Na + loading opens ATP-sensitive K + channels by depleting intracellular supplies of ATP. However, in our experiments it is unlikely that the Na +-evoked outward current results from a progressive fall in intracellular A T P levels because pipettes

routinely contained 1.5mM ATP. Furthermore, raising the A T P concentration to 1 0 m M failed to prevent the outward current produced by Na + loading. Another possibility is that the increased sodium p u m p activity raises the intracellular ratio of A D P to A T P , a condition which favors the opening of ATP-sensitive K + channels in other cells. 32'5~'64ATPsensitive K ÷ channels can also be opened by intracellular acidosis, which reduces the affinity of A T P for its binding site. 12 Although Na + loading could produce acidosis by increasing the activity of the N a + / H + exchanger, 2° any such change in pH would have to be rather localized so as not to be buffered by the 10raM H E P E S contained in our pipette solutions. On the whole, our data do not support the hypothesis that ATP-sensitive K + channels mediate outward currents evoked by Na + loading. It could be argued that outward currents evoked by N a ÷ loading result from a reduction in the transmembrane concentration gradient of Na ÷. For example, a reduction in the Na ÷ gradient might be expected to slow the exchange ofintracellular Ca 2+ for extracellular N a + by the N a + / C a 2+ antiporter, 3 and the resulting increase in intracellular Ca 2÷ could subsequently activate a Ca2+-dependent K ÷ current. 7 However, this hypothesis is not supported by our finding that Cd 2+, which blocks CaZ+-dependent K + currents, ~°54 failed to block outward currents evoked by Na ÷ loading. Alternatively, one could argue that N a ÷ loading produces a net outward current by reducing the inward current 1h. However, I h in dopamine neurons is not activated at the holding potential of - 6 0 mV used in our experiments. Also, outward currents evoked by Na + loading were largely blocked by K + channel blockers, which should have no effect on the mixed cation current I h in dopamine neurons. 36'39 Finally, Ba 2+, T E A and intracellular Cs ÷ completely blocked the outward current evoked by N a + loading without affecting I h. Our results do not support a role for I h in outward currents produced by Na ÷ loading. Results of the present studies are most consistent with the hypothesis that outward currents evoked by Na ÷ loading are mediated by Na+-dependent K + channels. The intracellular concentration of Na + ( 4 0 m M ) which evoked K + current in our experiments is in the same range as that which activates K ÷ channels in other cells. 33"37'45"65 Furthermore, large conductance Na+-dependent K ÷ channels are widely distributed throughout the brain, including olfactory bulb, 18 brainstem, 17 neocortex, 53 cerebellum, spinal cord and hippocampusJ 9 There are, however, some features of Na+-dependent K ÷ currents in dopamine neurons which differ from those described in other cells. F o r example, inward rectification is a prominent feature of Na+-dependent K ÷ currents in many cells, 16,33'37'65 whereas K + currents evoked by N a + loading in dopamine neurons showed relatively linear current-voltage relationships. A lack of rectification

Intracellular Na + evokes sulfonylurea-sensitive K + currents of currents in our studies could be due in part to the narrow range of voltages tested. A n o t h e r unusual finding in our studies was the time delay between rupture of the membrane patch and the appearance of outward current evoked by N a ÷ loading. The delayed onset of the outward current could be due to the need for Na + to diffuse into distal dendrites, although activation of a second messenger is another possibility. ~6'18 The ability of sulfonylureas to block Na÷-dependent K ÷ channels has not been reported previously and may suggest the existence of a novel type of K ÷ channel. However, our finding that a sulfonylurea-sensitive K + current could also be evoked in non-dopaminergic " s e c o n d a r y " cells suggests that this K ÷ channel is not unique to dopamine neurons. It is unclear if the K ÷ channels opened by diazoxide are the same as those opened by intracellular N a ÷ loading in dopamine neurons. Currents evoked by diazoxide and Na ÷ loading were similar in that they were both blocked by sulfonylurea drugs, and neither was blocked by raised internal concentrations of ATP. Although our finding that the effect of diazoxide was occluded by Na ÷ loading suggests that both treatments open the same set of K + channels, this "occlusion" of currents was non-selective because outward currents produced by baclofen were also greatly reduced by Na ÷ loading. Our finding that sulfonylureas do not block baclofen-induced currents agrees with the work of others 66 and supports the conclusion that these agents act on separate receptors. Although these results conflict with those of Roeper et al., 48 these differences could be due to the use of different species (rat vs guinea-pig) or tissue preparations (brain slice vs acute dissociation of dopamine neurons). The finding that Na ÷ loading evokes current generated by a ouabain-sensitive sodium p u m p was not unexpected 22'6~ and is consistent with previous results from our laboratory showing that N a ÷ influx through N-methyl-D-aspartate-activated channels evokes sodium pump current in dopamine neurons in vitro. 3° Pump currents were relatively small, perhaps because patch pipettes introduced N a ÷ into the soma, which may contain a lower density of N a + / K + ATPase compared with dendrites and axons. 9'47 Our finding that pump currents remained

717

outward even at relatively hyperpolarized test potentials is consistent with data from other studies j4"21'42 showing that sodium pump currents have little voltage dependence at the test potentials ( - 6 0 to - 120 mV) used in our studies. The presence of pump current, which shows little voltage dependence, most likely explains the failure of currents evoked by Na ÷ loading to reverse direction at the K + equilibrium potential (see Fig. 1B). Increased intracellular N a + reduces the activity of the N a + / C a + exchanger 3 and thereby facilitates the intracellular build-up of Ca 2+, which can trigger events leading to cell d e a t h ] 1'43"6°D o p a m i n e neurons in patients with Parkinson's disease may be especially vulnerable to Na ÷ accumulation, because these neurons reportedly have an impaired ability to synthesize ATP, 5"38"4°'~ which is needed by N a + / K + ATPase to maintain intracellular Na ÷ homeostasis. However, the K ÷ current evoked by Na ÷ loading could have a neuroprotective role in dopamine neurons. 16'33For example, this K + current will tend to cause membrane hyperpolarization, which would improve the electrical gradient for effective N a +/Ca 2+ exchange and thereby slow the rate of intracellular accumulation of Ca 2+ 15.33 Furthermore, membrane hyperpolarization would also reduce Ca 2÷ influx through voltage-gated channels. If the sulfonylureasensitive K ÷ current evoked by N a ÷ loading is indeed neuroprotective, then one might predict that sulfonylurea drugs would hasten the death of dopamine neurons in patients with Parkinson's disease. It is possible that sulfonylurea drugs used clinically do not reach sufficient concentrations in the brain to make this a realistic concern. CONCLUSIONS

We conclude that intracellular Na ÷ loading increases a sulfonylurea-sensitive K ÷ current and a current generated by the sodium pump in dopamine neurons. The Na+-activated K ÷ current may play a neuroprotective role during times of metabolic stress. Acknowledgements--This work was supported by grants

from the North Atlantic Treaty Organization, the Fondation L~on Fredericq of the University of Lifige and PHS grant NS31889. We also wish to thank Dr Dawn Shepherd for her critical review of the manuscript.

REFERENCES

1. Armstrong C. M. and Gilly W, F. (1992) Access resistance and space clamp problems associated with whole-cell patch clamping. In Methods in Enzyrnology. Ion Channels. (eds Rudy B. and Iverson L. E.), Vol. 207, pp. 100-122, Academic Press, New York. 2. Ashcroft F. M. (1988) Adenosine 5'-triphosphate-sensitive potassium channels. A. Rev. Neurosci. 11, 97-118. 3. Baker P. F., Blaustein M. P., Hodgkin A. L. and Steinhardt R. A. (1969) The influence of calcium on sodium efflux in squid axons. J. Physiol., Lond. 200, 431-458. 4. Baker P. F., Blaustein M. P., Keynes R. D., Manil J., Shaw T. 1. and Steinhardt R. A. (1969) The ouabain-sensitive fluxes of sodium and potassium in squid giant axons. J. Physiol., Lond. 200, 459-596.

718

V. Seutin et al.

5. Benecke R., Strumper P. and Weiss H. (1993) Electron transfer complexes I and IV of platelets are abnormal in Parkinson's disease but normal in Parkinson-plus syndromes. Brain 116, 1451-1463. 6. Bernardi H., Fosset M. and Lazdunski M. (1988) Characterization, purification, and affinity labeling of the brain [3H]glibenclamide-binding protein, a putative neuronal ATP-regulated K + channel. Proc. nam. Acad. Sci. U.S.A. 85, 9816-9820. 7. Blatz A. L. and Magleby K. L. (1987) Calcium-activated potassium channels. Trends Neurosci. 10, 463-467. 8. Blaustein M. P. (1977) Effects of internal and external cations and of ATP on sodium-calcium and calcium-calcium exchange in squid axons. Biophys. J. 20, 79-I 11. 9. Brines M. L. and Robbins R. J. (1993) Cell-type specific expression of Na +, K+-ATPase catalytic subunits in cultured neurons and glia: evidence for polarized distribution in neurons. Brain Res. 631, 1-1 I. 10. Brown D. A. and Higashida H. (1988) Voltage- and calcium-activated potassium currents in mouse neuroblastoma x rat glioma hybrid cells. J. Physiol., Lond. 397, 149-165. 11. Choi D. W. (1987) Ionic dependence of glutamate neurotoxicity. J. Neurosci. 7, 369-379. 12. Davies N. W., Standen N. B. and Stanfield P. R. (1992) The effect of intracellular pH on ATP-dependent potassium channels of frog skeletal muscle. J. Physiol., Lond. 445, 549-568. 13. Deri Z. and Adam-Vizi V. (1993) Detection of intracellular free Na + concentration of synaptosomes by a fluorescent indicator, Na+-binding benzofuran isophthalate: the effect of veratridine, ouabain, and alpha-latrotoxin. J. Neurochem. 61, 818-825. 14. De Weer P., Gadsby D. C. and Rakowski R. F. (1988) Voltage dependence of the N a - K pump. A. Rev. Physiol. 50, 225-241. 15. Dryer S. E. (1991) Na+-activated K + channels and voltage-evoked ionic currents in brain stem and parasympathetic neurones of the chick. J. Physiol., Lond. 435, 513-532. 16. Dryer S. E. (1994) Na+-activated K + channels: a new family of large-conductance ion channels. Trends Neurosci. 17, 155-160. 17. Dryer S. E., Fujii J. T. and Martin A. R. (1989) A Na+-activated K + current in cultured brain stem neurones from chicks. J. Physiol., Lond. 410, 283-296. 18. Egan T. M., Dagan D., Kupper J. and Levitan I. B. (1992) Properties and rundown of sodium-activated potassium channels in rat olfactory bulb neurons. J. Neurosci. 12, 1964-1976. 19. Egan T. M., Dagan, D., Kupper J. and Levitan I. B. (1992) Na+-activated K + channels are widely distributed in rat CNS and in Xenopus oocytes. Brain Res. 584, 319-321. 20. Frelin C., Vigne P. and Lazdunski M. (1984) The role of the Na+/H + exchange system in cardiac cells in relation to the control of the internal Na + concentration. J. biol. Chem. 259, 8880-8885. 21. Gadsby D. C., Bahinski A. and Nakao M. (1989) Voltage dependence of N a / K pump current. Curr. Topics Membranes Transport 34, 269-288. 22. Gadsby D. C. and Nakao M. (1989) Steady-state current-voltage relationship o f the N a / K pump in guinea pig ventricular myocytes. J. gen. Physiol. 94, 511-537. 23. Grace A. A. and Onn S.-P. (1989) Morphology and electrophysiological properties o f immunocytochemically identified rat dopamine neurons recorded in vitro. J. Neurosci. 9, 3463-3481. 24. Hausser M. A., de Weille J. R. and Lazdunski M. (1991) Activation by cromakalim of pre- and post-synaptic ATP-sensitive K ÷ channels in substantia nigra. Biochern. biophys. Res. Commun. 174, 909-914. 25. Hicks G. A. and Henderson G. (1992) Lack of evidence for coupling of the dopamine D 2 receptor to an adenosine triphosphate-sensitive potassium (ATP-K ÷) channel in dopaminergic neurones o f the rat substantia nigra. Neurosci. Lett. 141, 213-217. 26. Hicks G. A., Hudson A. L. and Henderson G. (1994) Localization of high affinity [3H]glibenclamide binding sites within the substantia nigra zona reticulata of the rat brain. Neuroscience 61, 285-292. 27. Jiang Z.-G., Pessia M. and North R. A. (1993) Dopamine and baclofen inhibit the hyperpolarization-activated cation current in rat ventral tegmental neurones. J. Physiol., Lond. 462, 753-764. 28. Johnson S. W. and North R. A. (1992) Opioids excite dopamine neurons by hyperpolarization of local interneurons. J. Neurosci. 12, 483-488. 29. Johnson S. W. and North R. A. (1992) Two types of neurone in the rat ventral tegmental area and their synaptic inputs. J. Physiol., Lond. 450, 455-468. 30. Johnson S. W., Seutin V. and North R. A. (1992) Burst firing in dopamine neurons induced by N-methyl-D-aspartate: role of electrogenic sodium pump. Science 258, 665 667. 31. Johnson S. W., Shen K.-Z., North R. A. and Seutin V. (1994) Intracellular sodium evokes a sulfonylurea-sensitive potassium current in dopamine neurons. Soc. Neurosci. Abstr. 20, 1522. 32. Kakei M., Kelly R. P., Ashcroft S. J. H. and Ashcroft F. M. (1986) The ATP-sensitivity o f K ÷ channels in rat pancreatic /~-cells is modulated by ADP. Fedn. Eur. biochem. Socs. Lett. 208, 63-66. 33. Kameyama M., Kakei M., Sato R., Shibasaki T., Matsuda H. and Irisawa H. (1984) Intracellular Na ÷ activates a K ÷ channel in mammalian cardiac cells. Nature 309, 354-356. 34. Lacey M. G., Mercuri N. B. and North R. A. (1988) On the potassium conductance increase activated by GABA-B and dopamine D-2 receptors in rat substantia nigra neurons. J. Physiol., Lond. 401, 437-453. 35. Lacey M. G., Mercuri N. B. and North R. A. (1989) Two cell types in rat substantia nigra zona compacta distinguished by membrane properties and the actions of dopamine and opioids. J. Neurosci. 9, 1233-1241. 36. Lacey M. G. and North R. A. (1988) An inward current activated by hyperpolarization (I-h) in rat substantia nigra zona compacta neurones in vitro. J. Physiol., Lond. 406, 18P. 37. Luk H.-N. and Carmeliet E. (1990) Na÷-activated K + current in cardiac cells: rectification, open probability, block and role in digitalis toxicity. Pfliigers Arch. 416, 766-768. 38. Mann V. M., Cooper J. M., Krige D., Daniel S. E., Schapira A. H. V. and Marsden C. D. (1992) Brain, skeletal muscle and platelet homogenate mitochondrial function in Parkinson's disease. Brain 115, 333-342. 39. Mercuri N. B., Bonci A., Calabresi P., Stefani A. and Bernardi G. (1995) Properties of the hyperpolarization-activated cation curent I h in rat midbrain dopaminergic neurons. Eur. J. Neurosci. 7, 462-469. 40. Mizuno Y., Ohta S., Tanaka M. et al. (1989) Deficiencies in complex I subunits of the respiratory chain in Parkinson's disease. Biochem. biophys. Res. Commun. 163, 1450--1455.

Intracellular Na + evokes sulfonylurea-sensitive K + currents

719

41. Mourre C., Widmann C. and Lazdunski M. (1990) Sulfonylurea binding sites associated with ATP-regulated K + channels in the central nervous system: autoradiographic analysis of their distribution and ontogenesis, and of their localization in mutant mice cerebellum. Brain Res. 519, 29-43. 42. Nakao M. and Gadsby D. C. (1989) [Na] and [K] dependence of the Na/K pump current-voltage relationship in guinea pig ventricular myocytes. J. gen. Physiol. 94, 539-565. 43. Orrenius S., Burkitt M. J., Kass G. E. N., Dypbukt J. M. and Nicotera P. (1992) Calcium ions and oxidative cell injury. Ann. Neurol. 32, $33-$42. 44. Parker W. D. Jr, Boyson S. J. and Parks J. K. (1989) Abnormalities of the electron transport chain in idiopathic Parkinson's disease. Ann. Neurol. 26, 719-723. 45. Partridge L. D. and Thomas R. C. (1976) The effects of lithium and sodium on the potassium conductance of snail neurones. J. Physiol., Lond. 254, 551-563. 46. Paxinos G. and Watson C. (1986) The Rat Brain in Stereotaxic Coordinates. Academic Press, San Diego. 47. Pietrini G., Matteoli M., Banker G. and Caplan M. J. (1992) Isoforms of the Na,K-ATPase are present in both axons and dendrites of hippocampal neurons in culture. Proc. ham. Acad. Sci. U.S.A. 89, 8414-8418. 48. Roeper J., Hainsworth A. H. and Ashcroft F. M. (1990) Tolbutamide reverses membrane hyperpolarisation induced by activation of D-2 receptors and GABA-B receptors in isolated substantia nigra neurones. Pfliigers Arch. 416, 473-475. 49. Rrper J. and Ashcroft F. M. (1993) Reduced internal ATP concentrations and mitochondrial inhibition activate K+-ATP channels in isolated dopaminergic substantia nigra neurones from the rat. J. Physiol., Lond. 473, 32P. 50. Rudy B. (1988) Diversity and ubiquity of K channels. Neuroscience 25, 729-749. 51. Schwanstecher C., Dickel C., Ebers I., Lins S., Zunkler B. J. and Panten U. (1992) Diazoxide-sensitivity of the adenosine 5'-triphosphate-dependent K ÷ channel in mouse pancreatic beta-cells. Br. J. Pharmac. 107, 87-94. 52. Schwanstecher C. and Panten U. (1994) Identification of an ATP-sensitive K ÷ channel in spiny neurons of rat caudate nucleus. Pflfigers Arch. 427, 187-189. 53. Schwindt P. C., Spain W. J. and Crill W. E. (1989) Long-lasting reduction of excitability by a sodium-dependent potassium current in cat neocortical neurons. J. Neurophysiol. 61, 233-244. 54. Segal M. and Barker J. L. (1986) Rat hippocampal neurons in culture: Ca 2÷ and Ca2+-dependent K ÷ conductances. J. Neurophysiol. 55, 751-766. 55. Seutin V., Johnson S. W. and North R. A. (1993) Apamin increases NMDA-induced burst-firing of rat mesencephalic dopamine neurons. Brain Res. 630, 341-344. 56. Shepard P. D. and Bunney B. S. (1991) Repetitive firing properties of putative dopamine-containing neurons in vitro: regulation by an apamin-sensitive Ca++-activated K + conductance. Expl Brain Res. 86, 141-150. 57. Stanfield P. R. (1987) Nucleotides such as ATP may control the activity of ion channels. Trends Neurosci 10, 335-339. 58. Stanford I. M. and Lacey M. G. (1995) Regulation of a potassium conductance in rat midbrain dopamine neurons by intracellular adenosine triphosphate (ATP) and the sulfonylureas tolbutamide and glibenclamide. J. Neurosci. 15, 4651-4657. 59. Stys P. K., Waxman S. G. and Ransom B. R. (1991) Reverse operation of the Na÷-Ca 2÷ exchanger mediates Ca 2+ influx during anoxia in mammalian CNS white matter. Ann. N.Y. Acad. Sci. 639, 328-332. 60. Stys P. K., Waxman S. G. and Random B. R. (1992) Ionic mechanisms of anoxic injury in mammalian CNS white matter: role of Na ÷ channels and Na+-Ca ~+ exchanger. J. Neurosci. 12, 430-439. 61. Thomas R. C. (1969) Membrane current and intracellular sodium changes in a snail neurone during extrusion of injected sodium. J. Physiol., Lond. 201, 495-514. 62. Treherne J. M. and Ashford M. L. J. (1991) The regional distribution of sulphonylurea binding sites in rat brain. Neuroscience 40, 523-531. 63. Trube G., Rorsman P. and Ohno-Shosaku T. (1986) Opposite effects of tolbutamide and diazoxide on the ATP-dependent K ÷ channel in mouse pancreatic beta-cells. Pflfigers Arch. 407, 493-499. 64. Tsuchiya K., Wang W., Giebisch G. and Welling P. A. (1992) ATP is a coupling-modulator of parallel Na/K ATPase-K channel activity in the renal proximal tubule. Proc. natn. Acad. Sci. U.S.A. 89, 6418-6422. 65. Wang Z., Kimitsuki T. and Noma A. (1991) Conductance properties of the Na+-activated K ÷ channel in guinea-pig ventricular cells. J. Physiol., Lond. 433, 241-257. 66. Watts A. E., Hicks G. A. and Henderson G. (1994) Diazoxide induces a sulphonylurea-sensitive outward current in rat substantia nigra zona compacta neurons in vitro. J. Physiol., Lond. 475, 146P. 67. Yung W. H., Hausser M. A. and Jack J. J. B. (1991) Electrophysiology of dopaminergic and non-dopaminergic neurones of the guinea-pig substantia nigra pars compacta in vitro. J. Physiol., Lond. 436, 643-667. (Accepted 17 October 1995)