SYNAPSE 17:23@-240 (1994)

Dopamine Neuron Membrane Physiology: Characterization of the Transient Outward

Current (Ia) and Demonstration of a Common Signal Transduction Pathway for IA and I,

L W N LIU, ROH-YU SHEN, GREGORY KAPATOS, AND LOUIS A. CHIODO Cellular and Clinical Neurobiology Program, Department of Psychiatry, Wayne State University School of

Medicine, Detroit, Michigan 48201

KEY WORDS Dopamine neurons, Whole-cell patch recording, Dissociated mono- layer neuronal cultures, Voltage-clamp, A-current, Potassium cur- rent, D2 receptors, DA autoreceptors

ABSTRACT Dopamine neurons derived from the mesencephalon of embryonic rats were maintained in primary culture, identified and studied with whole-cell patch recording techniques. These neurons demonstrated a rapidly activating and inactivating voltage-de- pendent outward current which required the presence of K+ ions. This current was termed I, because of its transient nature. It was elicited by step depolarizations from holding poten- tials more negative than -50 mV and exhibited steady-state inactivation at a membrane potential more positive than -40 mV and half-maximal inactivation observed at -65 mV. This current rapidly achieved peak activation in less than 8 msec and decayed with a time constant (7) of 58 * 5 msec. This current was observed in the presence of tetraethylammo- nium but was readily blocked by 4-aminopyridine (2-4 mM). This current was also observed to be modulated by stimulation of D2 dopamine receptors (DA autoreceptors) located on the dopamine neurons. Thus, both DA and the Dz receptor agonist quinpirole enhanced the peak IA observed, while the partial D, receptor agonist SKF 38393 was without effect. The en- hancement of IA was confirmed to be due to the activation of D, receptors as the effects of either DA or quinpirole were blocked by the D, receptor antagonists eticlopride and sulpir- ide, but not by the D, receptor antagonist SCH 23390. Since we have previously demon- strated that the IK present in these cells is also enhanced by D, receptor stimulation, we investigated the signal transduction pathways involved in coupling DA autoreceptors to both IA and I,. The response of both these potassium currents to DA autoreceptor stimulation was completely abolished by the preincubation of cultures with pertussis toxin, indicating the possible involvement of the G proteins Gi and Go. In an attempt to further characterize which G protein may be involved, additional experiments were performed. The ability of DA autoreceptor stimulation to augment both currents was also blocked completely when G protein activation was prevented by the intracellular application of GDPPS (100 @I). In contrast, irreversible activation of G proteins by intracellular application of the nonhydrolyz- able GTP analog GTPyS (100 FM) mimicked the effects of DA autoreceptor stimulation on both IA and I= In addition, the intracellular application of a polyclonal antibody that was selective for the a-subunit of Go completely abolished the DA autoreceptor modulation of both currents while preimmune serum was without effect. Taken together, these data dem- onstrate that the enhancement of IA and IK in response to stimulation of DA autoreceptors is dependent upon the activation of Go and appears to involve a Go, subunit. 0 1994 Wiley-Liss, Inc.

INTRODUCTION Over the last several years there have been many Received March 8,1993; accepted in revised form February 4,1994.

Address reprint requests to Dr. Louis A. Chiodo, Department of Pharmacology, School of Medicine, Texas Tech Health Sciences Center, 3601 4th Street, Lub-

advances in Our Of the membrane properties of mesencephalic (MES) DA neurons. This bock, TX 79430.

0 1994 WILEY-LISS, INC.

TRANSIENT OUTWARD CURRENT IN DA NEURONS 231

work has been essential for a better understanding of- the electrophysiological activity of this important class of monoamine neurons since these cells normally dis- play a range of firing patterns including single spike and bursting modes of action potential generation (see Grace and Bunney, 1984a,b; Chiodo, 1988). For exam- ple, it is now known that DA neurons display several inward currents mediated by both sodium and calcium ions (Chiodo, 1992; Chiodo and Kapatos, 1987, 1992; Liu et al., 1991). These currents underlie both the so- matic action potential and the putative nonsomatic cal- cium-dependent spikes observed in these cells (see Chiodo and Kapatos, 1987,1992; Grace and Onn, 1989; Kitai et al., 1984; Llinas et al., 1984). In addition, it is clear that DA neurons possess at least six distinct po- tassium currents (Chiodo and Kapatos, 1992; Roeper et al., 1990; Silva et al., 1990). These include the anoma- lous rectifier, the delayed rectifier, two pharmacologi- cally distinct calcium-dependent delayed-onset cur- rents, an ATP-sensitive current, and an early transient A-current.

Because it has been known for some time that the spontaneous electrophysiological activity of DA cells can be readily inhibited by the stimulation of somato- dendritic DA autoreceptors (Aghajanian and Bunney, 1977), it is important to examine the sensitivity of dif- ferent ionic currents to DA autoreceptor activation. We have recently shown that the delayed rectifier, IK, present in DA neurons is modulated by stimulation of somatodendritic DA autoreceptors (Chiodo and Kapa- tos, 1992; Chiodo, 1992). This modulation consists of an increase in the magnitude of the voltage-depen- dent current which is observed. Since the early tran- sient potassium current (IA) has been shown to be in- volved in regulating the pacing between action potentials in several different neurons (Conner and Stevans, 1971; Rogawski, 1985; Segal et al., 19841, the present study evaluated if this current was also modu- lated by stimulation of somatodendritic DA auto- receptors.

It has been shown that D, DA receptors are part of the family of G protein-coupled neurotransmitter receptors within the central nervous system (CNS; see Sibley and Monsma, 1992; Vallar and Meldosi, 1989). For example, studies conducted in the anterior pituitary demonstrate that the D, receptor is coupled to two different G proteins (Lledo et al., 1992). Sub- sequently, numerous signal transduction pathways have been linked to D, receptors, including phospholi- pase C, alteration in phosphotidyl inositol turnover and the adenylyl cyclase-CAMP pathway (Canonico, 1989; Kanterman et al., 1991; Vallar and Meldolesi, 1989; Vallar et al., 1990). Therefore, in the present study, we also examined the signal transduction pathways in- volved in mediating the increases in I, and IK that result from stimulation of somatodendritic DA autore- ceptors.

MATERIALS AND METHODS Tissue culture

Primary cultures of rat mesencephalon were pro- duced and then maintained in serum-free medium as previously described (Granneman and Kapatos, 1990; Kapatos, 1991). Briefly, pregnant rats (Hilltop Farms, Scottdale, PA) were anesthetized under CO, and killed by cervical dislocation. The 15 to 16-day-old embryos, timed by limb development, were removed and placed in dissociating medium composed of Dulbecco’s phos- phate-buffered saline (DPBS) containing no calcium or magnesium, and supplemented with 25 mM glucose, 15 mM sucrose and 25 mM HEPES (pH 6.7,330-340 mOs- moles). After removal of the brain, the mesencephalon was dissected under microscopic control as described by Berger et al. (1982). A single cell suspension was pro- duced by mechanical trituration through fire-polished Pasteur pipettes of decreasing bore diameter. Suspen- sions containing 7545% viable cells were placed in a modified N2 medium (Bottenstein and Sato, 1979) con- taining 5% heat-inactivated fetal calf serum and 5% horse serum and plated at a density of 4 x lo5 viable cells per well of 24-well culture dishes containing a 12 mm diameter acid-washed glass coverslip previously coated with a solution of 10 pg/ml of poly-D-lysine (62,000 mw). Modified N2 medium was composed of a 1:l mixture of Dulbecco’s modified Eagles medium and Ham’s F12 medium and contained in addition: 1.26 g/L sodium bicarbonate, 1 mM glutamine, 2.5 mg/ml bovine serum albumin, 50 pg/ml human transferrin, 2.5 pg/ml catalase, 5 pgml insulin, 100 p M putresine, 20 nM progesterone, 1 pM p-estradiol, 40 nM corticosterone, 300 pM triiodothyronine, 30 nM selenium, 540 pM li- noleic acid-albumin complex and 10 p.M 5-fluoro-2’- deoxyuridine and 100 pM uridine to inhibit glial cell proliferation. Cultures were maintained in an atmo- sphere of 10% CO, at 35°C. After 24 h this serum- containing medium was completely exchanged with se- rum-free modified N2 medium which was changed three times per week. Cultures were used for electro- physiological studies 10-14 days after plating.

Identification of DA neurons with 5,7-dihydroxytryptamine

In order to visualize the DA neurons prior to studying them electrophysiologically, they were incubated in N, (without serum) which contained 0.1% ascorbic acid (used as an antioxidant) which was buffered to pH 7.3 with 0.1 N NaOH, and contained 15-25 pM 5,7 dihy- droxytryptamine (5,7-DHT) as described previously (Dacey, 1988; Silva et al., 1988). Cultures were incu- bated for 30-60 minutes in an incubator. At the time of recording, the N, medium with 5,7-DHT was removed and the coverslip was washed twice with 1.5 ml of the fresh recording solution (see below). The coverslip was examined under UV light (excitation 365 nm, long-pass

232 L. LIU ET AL.

420 nm), and cells displaying a deep blue-violet fluores- cence were located for subsequent electrophysiological study.

Electrophysiology Visually identified DA neurons were voltage-

clamped in the whole-cell configuration using an EPC- 7List patch clamp amplifier (Adams-List Associates, Ltd., Westbury, NY). Patch electrodes were fabricated from 1.5 to 1.6 O.D. Micro-Hematocrit capillary tubes (VWR Scientific) pulled in a two-stage process on a Narishigie vertical puller (model PA-81) and polished on a Narishigie microforge (model MF-83). The holding potential and all voltage steps were controlled by PCLAMP software and hardware (Axon Instruments, Foster City, CA) using the CLAMPEX program. The normal recording solution contained the following (in mM): choline chloride 109; KCl 5; MgCl, 1; CaCl, 3; glucose 10; HEPES 5; L-tyrosine 50 pM; tetraethylam- monium (TEA) chloride 30; tetrodotoxin (TTX) 2 yM (pH 7.3; osmolarity adjusted as needed with sucrose to 320-330 mosmoles). The patch electrode solution con- tained the following (in mM): KCI-EGTA 140; MgCl,; CaC1,l; HEPES 10; adenosine triphosphate (ATP, as dipotassium salt) 2; and cyclic adenosine monophos- phate (CAMP) 0.25; (pH adjusted to 7.3 with KOH, os- molarity adjusted to 320-330 mosmoles). When the convulsant 4-aminopyridine (4-AP) was added to the external solution to block transient outward currents as previously described (Gustafsson et al., 1982; Segal et al., 1984; Thompson, 1977,1982), an equimolar con- centration of choline chloride was removed to control for changes in osmolarity. The patch electrode resis- tance ranged between 2 and 4 megohms. The seal resis- tances were typically greater than 5 gigohms and after membrane rupture the series resistance was between 4.3 and 10 megohms (mean 2 S.E.M.: 7.1 & 1.8 mego- hms; n = 100) and the level of capacitance compensa- tion ranged between 10 and 22 pF (mean 2 S.E.M.: 16.4 k 6.6 pF; n = 100). All recordings were carried out at room temperature (18-22°C). A 35 mm culture dish that held the coverslip upon which the cells were plated was visualized on the two-axis mechanical stage of a Zeiss ICM 405 inverted microscope under Hoffman modulation condenser and objective (~400). The re- cording headstage and electrode assembly were held on an articulated arm of a three-position joystick-con- trolled micromanipulator (MM-8000F, Activational Systems Inc., Warren, MI) during recording.

All drugs were dissolved in bathing medium and ap- plied to the outside of the soma with a 5-10 pm diame- ter pressure ejection pipette placed 20-30 ym from the cell soma (1-5 psi pulses delivered by a Picospritzer 11; General Valve corporation, Fairfield, NJ). A puff-jump protocol controlled by the PCLAMP software was em- ployed. Specifically, the drug was applied to the cell for 450 ms, followed 50 msec later by the voltage step com- mand. Control experiments consisted of the pressure

ejection of the vehicle (external solution) onto the cells. All solutions were filtered through a 0.45 ym mem- brane filter.

In some experiments, either guanosine 5'-0-(2-thio- diphosphate) (GDPPS) or guanosine 5'-0-(3-thio- triphosphate) (GTPyS) were included in the patch elec- trode solution at a concentration of 100 pM. To study the role of the a-subunit of Go, we employed an antisera which was generated against a synthetic fragment of this protein corresponding to amino acids 117-126. The generation and specificity of this antisera has been published previously (Granneman and Kapatos, 1990). When either the antisera or preimmune serum were employed, it was mixed with the patch electrode solu- tion to yield a final concentration of 10 pLvml.

Pertussis toxin preincubation Cultures were treated with pertussis toxin as follows:

500 pl of conditioned medium was removed from the well of a 24-well plate and replaced with fresh N2 me- dium containing 1,000 ng/ml of pertussis toxin yielding a final concentration 500 ng/ml. The cultures were re- turned to the incubator for 3-5 hours. Prior to record- ing, the coverslip was washed twice with external re- cording solution and placed in a final 1 ml of recording solution for electrophysiological examination. This method is known to completely ADP ribosylate G pro- teins present in mesencephalic cultures.

Data anlaysis Analysis of current records was performed with the

CLAMPFIT program. This included the analysis of time constants and obtaining the leakage current-sub- tracted values for the generation of current-voltage (I-V) plots. Voltage and current traces were converted to an ASCII format and imported into Sigmaplot (Jan- del Scientific, San Rafael, CA) to generate all figures. Leakage current was calculated from the conductance determined at potentials where no active currents were present (usually between -70 and -80 mV). The half- maximal activation and inactivation were calculated by fitting a third-order polynomial regression to the data points obtained from 15 cells. The time constant for inactivation was determined using CLAMPFIT which fit a single exponential function to the decay of I,.

RESULTS Characterization of IA

A total of 118 MES DA neurons were studied. All neurons were observed to possess IA. This current was examined first by holding the ceIls at -40 mV and applying a 200 msec conditioning hyperpolarizing step to -90 mV prior to stepping to test potentials between -80 and 30 mV for 250 msec (Fig. 1). The observed IA developed in elss than 8 msec with step depolarizations to potentials more positive than -40 mV. The peak amplitude of I, at a membrane potential of 30 mV ranged between 1 and 8 nA. The observed current inac- tivated in the presence of continued depolarization and

A

TRANSIENT OUTWARD CURRENT IN DA NEURONS

B

4-AP (4 mM)

1 n A L 100 ms

CONTROL

-80 -40 0 40

MEMBRANE POTENTIAL (mv)

233

Fig. 1. Typical transient outward current, I,, observed in MES DA neurons maintained in primary culture. This cell was sampled using whole-cell patch recording and the membrane was held at -40 mV and stepped to the various membrane potentials between -80 and 30 mV (after a conditioning hyperpolarization to -90 mV for 200 ms). A. The current records observed for a control DA neuron (CONTROL) and one

studied in the presence of 4 mM 4-aminopyridine (4-AP). It can be observed that 4-AP dramatically reduced the observed amplitude of IA. B: Leakage-subtracted peak I, current is plotted for the traces in A. Other voltage-dependent outward currents normally present in these cells were blocked by TEA (see Materials and Methods).

A B

I CONTROL I

30 ' 20

&- -40 d-L -40

-90 -90

c

+ z W d d 3 0 I- z v) z I- I- z W 0 U w a

II!

2

MEMBRANE POTENTIAL (mv)

Fig. 2. Demonstration of the activation-inactivation properties of MES DA neurons. Voltage-clamp records obtained from a single DA neuron studied for the voltage-dependent nature of activation (A) and inactivation (B) under control conditions (upper current traces). As may be observed in the leakage-subtracted plot in B, the current- voltage profile demonstrates that activation occurs a t -40 mV with the apparent half-maximal activation occurring at approximately 5 mV (C, solid triangles). Inactivation (C, solid circles) is evident at -80 mV with the apparent half-maximal inactivation occurring a t approx-

imately -65 mV. Activation and inactivation in this same cell was studied aRer the application of DA (50 pM, middle trace in A and B; open circles and open triangles in C). It may be observed that DA had no effects on the activation-inactivation plot even though it can be observed to increase the peak I, under both conditions [before (upper traces), after (middle traces) the application of DA (50 &MI]. The bot- tom traces in A and B are the voltage-clamp step commands applied to the cell.

234

A DA 100 FM

L. LIU ET AL.

B

1.0 - H

0.8 -

0.6 -

2 H - 1. 0.4 -

20 rnV

-20 rnv miJJj-

-90 rnV

0.0 I I I I I

0 50 100 150 200 250

CONDITIONING PULSE LENGTH (ms)

Fig. 3. The recovery of I, after complete inactivation was also stud- ied in MES DA neurons. A: Steady-state inactivation was removed by stepping from the holding potential of -20 mV to a conditioning poten- tial of -90 mV for increasing durations and then stepping to a test potential of 20 mV. B: When the relative current amplitude is plotted it may be seen t,hat inactivation is removed entirely after approxi-

mately 125 ms both before (CONTROL, open circles) and after the application of DA (100 pM, closed circles). The lack of change in the recovery from inactivation by DA application is apparent even though it may be seen in A that DA increased the peak I, observed. The bottom trace in A is the voltage-clamp step commands applied to the cell.

the decaying portion of the current trace could be readily fit with a single exponential function which yielded a mean time constant of inactivation (7) of 58 2 5 msec ( n = 12). The calculated T for inactivation was not observed to vary with membrane potential be- tween -40 and 20 mV, indicating it was not voltage- dependent. The IA was observed to be almost com- pletely blocked by bath application of 4-AP ( 2 4 mM; Fig. 11, and was absent when CsCl replaced KC1 in the patch recording pipette.

Voltage dependence of activation-inactivation was also examined (n = 15). This was accomplished by varying the 200 msec conditioning prepulse amplitude between -90 and 0 mV and stepping the membrane to a test potential of 20 mV (for inactivation) or by holding the conditioning pulse fixed at -90 mV and stepping to test potentials between -80 and 30 mV (activation). Figure 2 shows that activation occurs at 40 mV and half-maximal activation occurs at approximately 5 mV. Inactivation was observed at a conditioning membrane potential of -80 mV and half-maximal inactivation was reached at -65 mV (Fig. 2). We also examined the ki- netics of recovery from complete inactivation by vary- ing the length of the -90 mV hyperpolarizing condi- tioning pulse from 10 to 210 msec (n = 10). Half- maximal activation was observed after 30 msec and a

conditioning hyperpolarization pulse duration of 125 msec or more was necessary to remove all of the inacti- vation observed (Fig. 3).

IA sensitivity to DA autoreceptor stimulation The effects of DA autoreceptor stimulation were also

tested. The application of DA or the D, receptor agonist quinpirole was observed to significantly enhance the peak amplitude of IA in each cell studied (n = 55, Fig. 4). The degree of DA autoreceptor enhancement of IA ranged between 30 and 60% (at 30 mV) with the appli- cation of 100 p M quinpirole. The application of the par- tial D1 receptor agonist SK?? 38393 (100 pM) had no effect on IA (n = 10, data not shown). The effects of DA and quinpirole could be readily blocked by the coappli- cation of either of the D, receptor antagonists, sulpiride ( 1 0 4 0 pM, n = 7; Fig. 5) and eticlopride (10-40 pM, n = 71, but not the D1 receptor antagonist SCH 23390 (n = 5, 1040 pM). The augmentation of I, observed was dose-dependent and changes in the maximum cur- rent were never observed when pressure ejection con- trols were performed (n = 10). In addition, the effects of DA on the activation-inactivation profile (Fig. 2, n = 15) and recovery from inactivation (Fig. 3, n = 10) for IA were studied. Neither of these measures was al- tered significantly by the application of DA.

TRANSIENT OUTWARD CURRENT IN DA NEURONS 235

A

B P .-

DO ms

C

-1 00 -50 0 50

MEMBRANE POTENTIAL (mVJ

Fig. 4. Stimulation of DA autoreceptors augments the peak I, observed in MES DA neurons. A The voltage-clamp traces obtained from a DA cell before (CONTROL, open circles) and after the application of DA (50 pM, closed circles). B: The leakage-subtracted current-voltage plot obtained from the traces shown in A. It can be observed that DA application significantly increased the peak I, observed.

Coupling of I, and I, to DA autoreceptors We next examined if I, and I, (previously shown to

be modulated by DA autoreceptor stimulation; Chiodo and Kapatos, 1992) coupled to DA autoreceptors in a similar fashion. When the cultures were pretreated with pertussis toxin, the ability of DA application to enhance either I, or IK was completely abolished while the control currentlvoltage profile for both currents was unchanged (Fig. 6). To examine the role of G proteins further, we conducted additional experiments in which the patch pipette contained 100 pM GDPPS. Under these conditions, DA application again failed to en- hance either I, or I, while not altering the control cur- rents observed. (Fig. 7). These observations suggested that the activation of G proteins was directly involved in the transduction of DA autoreceptor stimulation. Therefore, we attempted to mimick the effects of au- toreceptor stimulation by completely activating the G proteins present by including GTPyS (100 pM) in the patch pipette. As is observed in Figure 7, GTPyS in- creased the magnitude of both 1, and IK to the same degree as was observed by the application of DA (100 FM). In addition, the subsequent application of DA onto these cells did not produce a further enhancement of either potassium current.

Go is the major signal-transducing protein in brain and is expressed in these cultures (Granneman and Kapatos, 1990). We examined whether or not the G

50 pM DA + 10 ,ILM SULPlRlDE B

CONTROL A

n A

C h

2

a v

l-

a (L 3 0 Y 6 w a

- 1 d

50

MEMBRANE POTANTIAL (mV)

Fig. 5. The DA autoreceptor-mediated increase in I, is antagonized by the D, receptor antagonist sulpiride. The top panel shows the cur- rent trace obtained before (A) and after the application of both DA (50 pM) and the D, receptor antagonists sulpiride (10 pM, B). C: The leakage-subtracted current-voltage plot clearly demonstrates that sulpiride was able to readily block the augmentation of I, observed after application of DA (see Fig. 4). Open circles are the contro1 data points while those obtained during drug application are represented by closed circles.

236

A

L. LIU ET AL.

’ r

100 -50 0 50 MEMBRANE POTENTIAL (mV)

B CONTROL 100 pIM DA

Fig. 6. Illustration of the effects of local application of DA on both IA and I, observed in mesencephalic DA neurons. The current traces in A show the I, current observed in a DA neurons before (CONTROL) and after the application of 100 p M DA. As may be observed in the adjacent current-voltage plot, the application of DA significantly enhanced the

protein involved in the signal transduction of DA au- toreceptor stimulation might be Go, by testing whether an antibody directed against the a-subunit of Go could block the DA stimulation of potassium currents nor- mally observed. As shown in Figure 8, inclusion of the antibody completely abolished the ability of DA to en- hance either I, or IK. In control experiments, preim- mune serum was included in the patch pipette. This manipulation had no effect on the ability of DA applica- tion to enhance either IA or IK (n = 10, data not shown).

DISCUSSION The present study examined the current-voltage be-

havior of the transient outward K+ current, I,, ob- served in cultured mesencephalic DA neurons. I, was activated by depolarization of the membrane from a hyperpolarized holding potential to a value more posi- tive than -40 mV, displayed a half-maximal activation at around 5 mV, had a T of decay of 55 msec, and re- quired approximately 125 ms of hyperpolarization to remove inactivation. These findings are in agreement with the observations made in DA neurons isolated from early postnatal animals (Silva et al., 1990) as well as numerous other classes of neurons (Belluzzi et al., 1985; Segal et al., 1984; Williams et al., 1984). A tran- sient outward conductance thought to represent IA,

3 r

d

50 I-

v, MEMBRANE POTENTIAL (mV)

peak current observed. Likewise, when I, was studied in another cell (B), it may be observed that the local application of DA also signifi- cantly enhanced the magnitude of this current. The opened and closed symbols represent the data obtained before and after the application of DA, respectively.

which appears after the repolarization resulting from the injection of hyperpolarizing current, has also been observed in DA neurons sampled in the in vitro slice preparation (Grace and Onn, 1989; Harris et al., 1989). However, this conductance was reported to be insensi- tive to 4-AP. The reason for this apparent discrepancy is not immediately apparent.

We suggested earlier (Chiodo and Kapatos, 1987; also see Silva et al., 1990) that one important way by which DA cells regulate their spike frequency is through the voltage-dependent activation and inactiva- tion of IA. It is now clear that this current is generally inactive at the resting membrane potential normally observed for these neurons (-59 mV for DA cells in culture; Chiodo and Kapatos, 1992) because the thresh- old for IA activation is around -40 mV. Therefore, when a DA neuron is depolarized from rest, via the slow depo- larization that normally precedes action potential gen- eration, I, will remain inactive. It has been suggested that the afterhyperpolarization (AHP) of the mem- brane, which is normally observed after an action po- tential in these cells, might remove inactivation such that subsequent depolarizations would activate 1, (Chiodo and Kapatos, 1987; Silva et al., 1990). Al- though the AHP of these cells corresponds to a mem- brane potential of around -65 mV, which would be

237 TRANSIENT OUTWARD CURRENT IN DA NEURONS

7- N=6

5 3 u LT LT 3 2 0

4 - 4 a

0

N=6

* T 5 - L [ N=8

TREATMENTS

a CONTROL = +DA(IOOpM)

PTX GDPpS GTPyS

Fig. 7. A bar graph depicting the effects of various manipulations of G proteins on the ability of D, autoreceptor stimulation by DA to modulate either IA (upper panel) or I, (lower panel) present in DA neurons. For both currents, it may be seen that pertussis toxin pre- treatment (PIX) and intracellular application of GDPpS (100 m) abolished the modulation normally observed by DA autoreceptor stim- ulation. The enhancement of both IA and IK by DA could be mimicked by the intracellular application of GTPyS (100 p,M) which causes the persistent activation of G proteins. This effect appeared maximal since the application of DA produced no additional increase in either cur- rent. *P < 0.01.

expected to remove some inactivation, the fact that the AHP typically lasts only 96 ms (Chiodo and Kapatos, 1992) suggests that the removal of inactivation would be incomplete. For example, the present data indicate that at a potential of -90 mV, it takes at least 125 ms to completely remove inactivation. However, because the AHP which follows a burst of action potentials is of both greater magnitude and duration than that observed after a single action potential (Chiodo and Kapatos, 1987, 1992; Grace and Bunney, 1984b), it would be expected that a complete removal of inactivation would occur, and that maximal activation of IA would result from the next depolarization. This means that while the activation of IA would play some role in pacing action potentials during the single-spike mode of firing, its major influence would be seen only after the occurrence of a burst. Thus, this current plays an important role in determining the length of the postburst inhibitory pe- riod. In addition, it might be expected that the greater

the number of action potentials in a given burst (burst length) the greater the magnitude of the AHP and thus, the more complete the removal of inactivation. In possi- ble support of this we have noted that there is a positive correlation between the number of action potentials contained in a burst and the length of postburst inhibi- tory period (unpublished observations).

We are now able to begin to understand the mem- brane events associated with the reduction in DA cell excitability which accompanies stimulation of somato- dendritic DA autoreceptors. It has been known for a while that D, DA autoreceptors increase K+ conduc- tances in DA neurons (Lacey et al., 1987). We have now shown that this change in conductance is due to the ability of DA autoreceptor stimulation to increase the magnitude of the delayed rectifier (I,, Chiodo and Kap- atos, 1987, 1992) and IA. In particular, the gradual slowing of DA neuronal firing during low-to-moderate levels of autoreceptor stimulation would appear to cor- respond to the increased voltage-dependent activation of IA and the resulting increase in overall interspike interval duration. In addition, the simultaneous in- crease in the magnitude of the I,, which persists once the influence of IA has decayed (55 msec or greater given the observed T), would also play an important role. However, it is clear that additional transmem- brane ionic current must be involved in the complete inhibition of spontaneous activity produced by autore- ceptor stimulation because both I, and I, are inactive at, or a t potentials more negative than, the resting membrane potential. Likely candidates include the in- ward rectifier (Lacey et al., 1987; Silva et al., 1990) as well as the different calcium currents (Liu et al., 1991) present in these cells. In this regard, we have recently observed that the inward rectifier is increased (Chiodo et al., unpublished observations) and two out of three different calcium currents are inhibited (Liu et al., 1991) by stimulation of DA autoreceptors.

The present study evaluated the signal transduction mechanisms involved in the augmentation of two dis- tinct potassium currents in DA neurons after stimula- tion of DA autoreceptors. It was observed that modula- tion of both IA and I, were dependent upon a pertussis toxin-sensitive G protein. Moreover, the influence of DA autoreceptors upon these currents could also be blocked by the intracellular application of GDPPS, which blocks the activation of G proteins and the disso- ciation of the dimer from the GTP-bound a-subunit. Conversely, the intracellular application of GTPyS mimicked the DA-induced augmentation of I, and I,. This enhancement appeared to be maximal as the ap- plication of DA to these cells did not further enhance either current. Taken together, these data suggest that Go may be involved in the transduction of DA autore- ceptor influences on IA and I,. In support of this hy- pothesis, we observed that an antibody directed against the Go, subunit (see Granneman and Kapatos, 1990)

238 L. LIU ET AL.

ANTI Goa

CONTROL 100 pM DA

I, A

MEMBRANE POTENTIAL (mv)

: i- 100 rns

100 pM DA

I \ c Z w K K 3 0

W l- a c v) I > n a - F

5

3 4l i 1 Ld

*:a 100 -%-Lo

L, MEMBRANE POTENTIAL (mV)

Fig. 8. The a-subunit of Go is involved in the coupling of DA autoreceptors to potassium currents. In an attempt to indicate whether or not Go might be the G protein involved in the transduction of DA autoreceptor activation, we recorded from cells with patch pipette which contained a polyclonal antibody directed against Goa. This antibody completely blocked the ability of DA to enhance either IA (A) or I, (B).

resulted in a complete blockade of DA autoreceptor- mediated enhancement of I, and I,.

Previous studies show that the a-subunit of Go mod- ulates muscarinic (Toselli and Lux, 1988; also see Ya- tani et al., 1988), opiate (Heschler et al., 19871, and neuropeptide Y (Ewald et al., 1988) receptors. Using a similar approach to ours, it was shown that Go, is in- volved in mediating the DA regulation of calcium cur- rents in the snail (Harris-Warrick et al., 1988). In a recent study, GoU was shown to couple D, receptors to voltage-dependent calcium channels in rat anterior pi- tuitary cells (Lledo et al., 1992). Although the direct regulation of I, and I, by Go, may have occurred in the present study, we cannot rule out the possibility of ad- ditional intermediary steps. Finally, there are known to be two forms of Go derived via alternative splicing (Hsu et al., 1990; Strathmann et al., 1990). Recent work em- ploying the direct injection of antisense DNA into GH3 cells demonstrates that Goal is involved in coupling muscarinic receptors to calcium currents, whereas GO,, is involved in somatostatin-mediated inhibition of these currents (Kleuss et al., 1991). In contrast, in snail neurons, GOaS, and not Goal, appears to be involved in mediating the neuropeptide modulation of calcium cur- rents (Man-Son-Hing et al., 1992). The present studies did not address which isoform of Go, might be directly

involved in coupling DA autoreceptors to 1, and IK. Further studies will be required to directly address this question.

It has been demonstrated that at least a subpopula- tion of mesocortical DA neurons which project to the prefrontal cortex in the rat are devoid of impulse-regu- lating somatodendritic DA autoreceptors (Chiodo et al., 1984; however, see Shepard and German, 1984). Al- though it is not clear at the present time, several possi- ble reasons for the insensitivity of this group of DA cells to DA autoreceptor stimulation may now be advanced. First, it may be possible that these cells do not possess somatodendritic D, receptors. However, because these cells do possess release-modulating nerve terminal au- toreceptors (Galloway et al., 19861, this seems unlikely unless this represents a different type of DA receptor protein or the coupling of a given receptor protein to different intracellular events in various regions of the same cell. Second, these cells may be devoid of I, and/or I,. Although there is no direct evidence for this to date, it should be noted that DA cells projecting to the pre- frontal cortex which are devoid of impulse-regulating autoreceptors have significantly higher spontaneous firing rates when compared to any other population of DA cells sampled (Chiodo et al., 1984). Given our un- derstanding of the role of I, and IK in the pacing of

239 TRANSIENT OUTWARD CURRENT IN DA NEURONS

action potentials in these cells, this remains a possibil- ity. Third, these cells may display these two currents but lack the necessary signal transduction mechanisms to allow for their coupling to the DA autoreceptor. Addi- tional studies in identified mesocortical cells which project to the prefrontal cortex will be necessary to ad- dress these possibilities directly.

ACKNOWLEDGMENTS We thank Dr. A.S. Freeman for his comments on this

manuscript and J. Rubin and G. Bora for their technical assistance. This work was supported by USPHS grants MH-41557 (L.A.C.) and NS-26081 (G.K.).

REFERENCES Aghajanian, G.K., and Bunney, B.S. (1977) Dopamine “autoreceptors”:

Pharmacological characterization by microiontophoretic single-cell recording studies. Naunyn-Schmiedeberg‘s Arch. Pharmacol., 297:

Belluzzi, O., Sacchi, O., and Wanke, E. (1985) A fast transient outward current in the rat sympathetic neurone studied under voltage clamp conditions. J . Physiol. (Lond.), 358:91-108.

Berger, B., Verney, C., Gaspar, P, and Febvret, A. (1982) Transient expression of tyrosine hydroxylase immunoreactivity in some neu- rons on the rat neocortex during postnatal development. Dev. Brain Res., 23:141-144.

Bottenstein, J.E., and Sato, G.H. (1979) Growth of rat neuroblastoma cell lines in serum-free supplemented medium. Proc. Natl. Acad. Sci. USA, 76514-517.

Canonico, P.L. (1989) D-2 dopamine recptor activation reduces free (3H)arachidonate release induced by hypophysiotropic peptides in anterior pituitary cells. Endocrinology, 125:118&1186.

Chiodo, L.A. (1988) Dopamine-containing neurons in the mammalian central nervous system: Electrophysiology and pharmacology. Neu- rosci. Biobehav. Rev., 1249-90.

Chiodo, L.A. (1992) Dopamine autoreceptor signal transduction in the DA Cell body: A “current view”. Neurochem. Int., 20 (suppl):8lS- 84s.

Chiodo, L.A., and Kapatos, G. (1987) Mesencephalic neurons in pri- mary culture: Immunocytochemistry and membrane physiology. In: Neurophysiology of Dopaminergic Systems-Current Status and Clinical Perspectives. L.A. Chiodo and A.S. Freeman, eds. Lake- shore Publishing Co., Grosse Pointe, pp. 67-91.

Chiodo, L.A., and Kapatos, G. (1988) Dopamine-containing neurons: Intracellular analysis and characterization. In: Pharmacology and Functional Regulation of Dopaminergic Neurons. P.M. Beart, G.N. Woodruff, and D.M. Jackson, eds. MacMillan Press, London, pp. 229-235.

Chiodo, L.A., and Kapatos, G. (1992) Membrane properties of identi- fied mesencephalic dopamine neurons in primary dissociated cell culture. Synapse, 11:294-309.

Chiodo, L.A., Bannon, M.J., Grace, A.A., Roth, R.H., and Bunney, B.S. (1984) Evidence for the absence of impulse-regulating somatoden- dritic and synthesis-modulating nerve terminal autoreceptors on subpopulations of mesocortical dopamine neurons. Neuroscience, 12:l-16.

Conner, J.A., and Stevens, C.F. (1971) Prediction of repetitive firing behaviour from voltage clamp data on an isolated soma. J . Physiol. (Lond.), 213:31-53.

Dacey, D.M. (1988) Dopamine-accumulating retinal neurons revealed by invitro fluorescence display a unique morphology. Science, 240: 119C1198.

Ewald, D.A., Sternweis, P.C., and Miller, R.J. (1988) Guanine nucle- otide-binding protein GO-induced coupling of neuropeptide Y recep- tors to calcium channels in sensory neurons. Proc. Natl. Acad. Sci.

Galloway, M.P., Wolf, M.E., and Roth, R.H. (1986) Regulation of dopamine synthesis in the medial prefrontal cortex is mediated by release-modulating autoreceptors: Studies in uiuo. J. Pharmacol. Exp. Ther., 236:689-698.

Grace, A.A., and Bunney, B.S. (1984a) The control of firing pattern in nigral neurons: Single spike firing. J. Neurosci. 4:286&2876.

Grace, A.A., and Bunney, B.S. (1984b) The control of firing pattern in nigral neurons: Burst firing. J . Neurosci. 4:2877-2890.

Grace, A.A., and Onn, S.-P. (1989) Morphology and electrophysiologi- cal properties of immunocytochemically identified rat dopamine neurons recorded in uitro. J. Neurosci., 9:346%3481.

1-7.

USA, 8813633-3637.

Granneman, J.G., and Kapatos, G. (1990) Developmental expression of Go in neuronal cultures from rat mesencephalon and hypothala- mus. J. Neurochem., 541995-2001.

Gustafsson, B., Galvan, M., Grafe, P., and Wigstrom, H. (1982) A transient outward current in a mammalian central neurone blocked by 4-aminopyridine. Nature, 229:252-254.

Harris, N.C., Webb, C., and Greenfield, S.A. (1989) A possible pace- maker mechanism in pars compacta neurons of the guinea-pig sub- stantia nigra revealed by various ion channel blocking agents. Neu- roscience, 31:355-362.

Harris-Warrick, R.M., Hammond, C., Paupardin-Tritsch, D., Hom- burger, V., Rouot, B., Brockaert, J., and Gerschenfeld, H.M. (1988) An subunit of a GTP-binding protein immunologically related to Go mediates a dopamine-induced decrease of Ca2+ current in snail neurons. Neuron, 1:27-32.

Heschler, J., Rosenthal, W., Trautwein, W., and Schuyltz, G. (1987) The GTP-binding protein, Go, regulates neuronal calcium channels. Nature, 325445447.

Hsu, W.H., Rudolph, U., Sanford, J., Bertrand, P., Glate, J., Nelson, C., Moss, L.G., Boyd, A.E., Codina, J., and Birnbaumer, L. (1990) Molecular cloning of a novel splice variant of the a subunit of the mammalian Go protein. J . Biol. Chem., 265:11220-11226.

Kapatos, G. (1991) Tetrahydrobiopterin synthesis rate and turnover time in neuronal cultures from embryonic rat mesencephalon and nypothalamus. J. Neurochem., 55:129-136.

Kanterman, R.Y., Mahan, L.C., Briley, E.M., Monsma, F.J., Sibley, D.R., Axelrod, J., and Felder, C.C. (1991) Transfected D, dopamine receptors mediate the potentiation of arachidonic acid release in Chinese hamster ovary cells. Molec. Pharmacol., 39:364-369.

E t a , T., Kita, H., and Kitai, S.T. (1986) Electrical membrane proper- ties of rat substantia nigra compacta neurons in an in uitro slice preparation. Brain Res., 372:21-30.

Kleuss, C., Hescheler, J., Ewel, C., Rosenthal, W., Schultz, G., and Wittig, B. (1991) Assignment of G-protein subtypes to specific recep- tors inducing inhibition of calcium currents. Nature, 353:43-48.

Lacey, M.G., Mercuri, N.B., and North, R.A. (1987) Dopamine acts on D2 receptors to increase potassium conductance in neurons of the rat substantia nigra zona compacta. J. Physiol. (Lond.), 392:397- 416.

Liu, L.-X., Kapatos, G., and Chiodo, L.A. (1991) Voltage-clamp analy- sis of calcium-dependent inward currents in cultured mesencephalic dopamine neurons. SOC. Neurosci. Abstr., 17:1350.

Lledo, P.M., Homburger, V., Bockaert, J., and Vincent, J.-D. (1992) Differential G rotein-mediated coupling of D, dopamine receptors to K+ and Cap+ currents in rat anterior pituitary cells. Neuron, 8:455-463.

Llinas, R., Greenfield, S.A., and Jahnsen, H. (1984) Electrophysiology of pars compacta cells in the in uitro substantia nigra-A possible mechanism for dendritic release. Brain Res., 294:127-132.

Man-Son-Hing, H.J., Codina, J., Abramowitz, J., and Haydon, P.G. (1992) Microinjection of the a-subunit of the G protein G02, but not G01, reduces a voltage-sensitive calcium current. Cellular Signal- ling, 4:429441.

Rogawski, M.A. (1985) The A-current: How ubiquitous a feature of excitable cells is it? Trends Neurosci., 8:214-219.

Roeper, J., Hainsworth, A.H., and Aschroft, F.M. (1990) Tolbutamide reverses membrane hyperpolarization induced by activation of D2 receptors and GABAB receptors in isolated substantia nigra neu- rons. Pflugers Arch., 416:473-475.

Segal, M., Rogawski, M.A., and Barker, J.L. (1984) A transient potas- sium conductance regulates the excitability of cultured hippocampal and spinal neurons. J . Neurosci., 4:604-609.

Shepard, P.D., and German, D.C. (1984)A subpopulation of mesocorti- c a l dopamine neurons possess autoreceptors. Eur. J. Pharmacol., 114:401-402.

Sibley, D.R., and Monsma, F.J., Jr . (1992) The molecular biology of dopamine receptors. Trends Pharmacol. Sci., 13:61-69.

Silva, N.L., Mariani, A.P., Harrison, N.L., and Barker, J.L. (1988) 5,7-dihydroxytryptamine identifies living dopaminergic neurons in mesencephalic cultures. Proc. Natl. Acad. Sci. USA, 85:734&7350.

Silva, N.L., Pechura, C.M., and Barker, J.L. (1990) Postnatal rat ni- grostriatal dopaminergic neurons exhibit five types of potassium conductances. J . Neurophysiol., 64:262-272.

Strathmann, M., Wilkie, T.M., and Simon, M.I. (1990) Alternative splicing produces transcripts encoding two forms of the subunit of GTP-binding protein GO. Proc. Natl. Acad. Sci. USA, 87:6477-6481.

Thompson, S.H. (1977) Three pharmacologically distinct potassium channels in molluscan neurones. J. Physiol. (Land.), 265:46!%488.

Thompson, S.H. (1982) Aminopyridine block of transient potassium current. J . Gen. Physiol., 8O:l-18.

Toselli, M., and Lux, H.D. (1988) GTP-binding protein mediates ace- tylcholine inhibition of voltage-dependent calcium channels in hip- pocampal neurons. Pflugers Arch., 412:319-321.

240 L. LIU ET AL.

Vallar, L., and Meldolesi, J. (1989) Mechanisms of signal transduc- tion at the dopamine D2 receptor. Trends Pharmacol. Sci., 10:74- 77. science, 13:137-156.

Vallar, L., Muca, C,. Magni, M., Albert, P., Bunzow, J., Meldolesi, J., and Civelli, 0. (1990) Differential coupling of dopaminergic D2 re- ceptors expressed in different cell types. J. Biol. Chem., 265:10320- 10326. 828-831.

Williams, J.T., North, R.A. Shefner, S.A., Nishi, S., and Egan, T.M. (1984) Membrane properties of rat locus coeruleus neurons. Neuro-

Yatani, A,, Hamm, H., Codina, J., Mazzoni, M.R., Birnbaumer, L., and Brown, A.M. (1988) The monoclonal antibody to the a subunit of G, blocks muscarinic activation of atrial K' channels. Science, 241: