The Journal of Neurophysiology Vol. 81 No. 1 January 1999, pp. 29-38
Copyright ©1999 by the American Physiological Society

Dopamine Modulates Two Potassium Currents and Inhibits the Intrinsic Firing Properties of an Identified Motor Neuron in a Central Pattern Generator Network

Peter Kloppenburg, Robert M. Levini, and Ronald M. Harris-Warrick

Section of Neurobiology and Behavior, Seeley G. Mudd Hall, Cornell University, Ithaca, New York 14853

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Kloppenburg, Peter, Robert M. Levini, and Ronald M. Harris-Warrick. Dopamine modulates two potassium currents and inhibits the intrinsic firing properties of an identified motor neuron in a central pattern generator network. J. Neurophysiol. 81: 29-38, 1999. The two pyloric dilator (PD) neurons are components [along with the anterior burster (AB) neuron] of the pacemaker group of the pyloric network in the stomatogastric ganglion of the spiny lobster Panulirus interruptus. Dopamine (DA) modifies the motor pattern generated by the pyloric network, in part by exciting or inhibiting different neurons. DA inhibits the PD neuron by hyperpolarizing it and reducing its rate of firing action potentials, which leads to a phase delay of PD relative to the electrically coupled AB and a reduction in the pyloric cycle frequency. In synaptically isolated PD neurons, DA slows the rate of recovery to spike after hyperpolarization. The latency from a hyperpolarizing prestep to the first action potential is increased, and the action potential frequency as well as the total number of action potentials are decreased. When a brief (1 s) puff of DA is applied to a synaptically isolated, voltage-clamped PD neuron, a small voltage-dependent outward current is evoked, accompanied by an increase in membrane conductance. These responses are occluded by the combined presence of the potassium channel blockers 4-aminopyridine and tetraethylammonium. In voltage-clamped PD neurons, DA enhances the maximal conductance of a voltage-sensitive transient potassium current (I_A) and shifts its V_act to more negative potentials without affecting its V_inact. This enlarges the "window current" between the voltage activation and inactivation curves, increasing the tonically active I_A near the resting potential and causing the cell to hyperpolarize. Thus DA's effect is to enhance both the transient and resting K⁺ currents by modulating the same channels. In addition, DA enhances the amplitude of a calcium-dependent potassium current (I_O(Ca)), but has no effect on a sustained potassium current (I_K(V)). These results suggest that DA hyperpolarizes and phase delays the activity of the PD neurons at least in part by modulating their intrinsic postinhibitory recovery properties. This modulation appears to be mediated in part by an increase of I_A and I_O(Ca). I_A appears to be a common target of DA action in the pyloric network, but it can be enhanced or decreased in different ways by DA in different neurons.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

In invertebrates and vertebrates, rhythmic motor patterns are generated by central pattern generators (CPGs) (Getting 1989; Pearson 1993; Selverston and Moulins 1985). A CPG is a defined neuronal network that can generate a wide variety of variants on a basic motor pattern to adapt motor behavior to changes in the environment (Getting 1989; Harris-Warrick and Marder 1991; Pearson 1993; Selverston and Moulins 1985). These modifications in the activity pattern can be complex and vary over a wide range and are often under neuromodulatory control. Modulation can be either quantitative, for example, causing a simple change in the frequency of a given pattern, or qualitative, causing mode switches between different motor patterns. To accomplish this flexibility with a given set of neurons, the modulatory input targets two different sites within the network, 1) the synaptic connectivity and 2) the intrinsic firing properties of the component neurons, including oscillatory, plateau, and postinhibitory rebound (PIR) properties. These intrinsic cellular properties are often changed by modulating specific ionic currents in the responsive cell.

We are studying the ionic mechanisms that lead to changes in the intrinsic properties of CPG neurons in the pyloric network of the crustacean stomatogastric nervous system. This system combines the advantages of easy accessibility and a well-known circuitry (reviewed by Harris-Warrick et al. 1992b). The 14 pyloric neurons and their synaptic connections are all identified and characterized (reviewed by Selverston and Moulins 1985; Miller 1987; Mulloney 1987). The network consists of six neuron types with known electrophysiological properties (Hartline and Graubard 1992). Modulators that alter the pyloric motor pattern have differential effects on the intrinsic properties of the different cell types (Harris-Warrick et al. 1992a).

Bath application of the biogenic amine dopamine (DA) can change the pyloric motor pattern (Fig. 1) (Anderson and Barker 1981; Eisen and Marder 1984; Flamm and Harris-Warrick 1986a) by affecting both the synaptic strength (Ayali et al. 1998; Johnson and Harris-Warrick 1990; Johnson et al. 1993a,b, 1995) and the intrinsic properties of pyloric network neurons (Flamm and Harris-Warrick 1986b; Harris-Warrick and Flamm 1986; Harris-Warrick et al. 1995a, b). The lateral pyloric neuron (LP) and many of the pyloric constrictor neurons (PY) are excited and phase advanced within the pyloric cycle by DA (Harris-Warrick et al. 1995a,b). In both cell types this is, at least in part, caused by a dopaminergic increase of the intrinsic rate of recovery after inhibition. DA decreases a transient K⁺ current (I_A) in both LP and PY neurons, and in LP it additionally enhances a hyperpolarization-activated inward current (I_h).

View larger version (35K): [in this window] [in a new window]   FIG. 1. A and B: effect of dopamine (DA) on the pyloric rhythm (modified from Harris-Warrick et al. 1995b). A: simultaneous intracellular recordings of 4 cell types are shown under control conditions and during DA (104 M) application. B: phase diagrams for activity of pyloric neurons under control conditions and in the presence of DA (104 M). C: effect of DA on the resting potential of a pyloric dilator (PD) neuron that was isolated from pyloric chemical synaptic input with PTX (5 × 106 M). A 1-s DA pulse (103 M) pressure ejected across the stomatogastric ganglion (STG) hyperpolarized the cell by >5 mV and suppressed the firing of action potenials for ~1 min. AB, anterior burster neuron; PD, pyloric dilator neuron; LP, lateral pyloric neuron; PY, pyloric constrictor neuron. Vertical markers: 10 mV.

The two pyloric dilators (PDs) are members [along with the anterior burster (AB) neuron] of the pacemaker group that drives the pyloric rhythm. DA hyperpolarizes and reduces the number of action potentials per cycle in the PD neurons, leading to a phase delay of PD relative to the AB neuron and a reduction in the overall pyloric cycle frequency (Flamm and Harris-Warrick 1986a). We investigated the ionic mechanisms underlying these changes by analyzing DA's effects on three outward currents: a transient voltage-activated K⁺ current (I_A), a sustained voltage-activated K⁺ current (I_K(V)), and a voltage- and calcium-dependent K⁺ current (I_O(Ca)). Some of these results were presented in abstract form (Kloppenburg et al. 1997; Levini et al. 1996).

	METHODS

Abstract Introduction Methods Results Discussion References

Materials

California spiny lobsters, Panulirus interruptus, were obtained from Don Tomlinson and maintained 4 wk in artificial seawater at 16°C until use. All chemicals were obtained from Sigma Chemical (St. Louis, MO).

Cell identification

Animals were anesthetized by cooling in ice for 30 min before dissection. The stomatogastric ganglion (STG), along with the motor nerves and the associated commissural and oesophageal ganglia, were dissected as described by Selverston et al. (1976) and pinned in a Sylgard-coated dish. The preparations were superfused continuously at 3 ml/min with saline (16°C) of the following composition in mM: 479 NaCl, 12.8 KCl, 13.7 CaCl₂, 3.9 Na₂SO₄, 10.0 MgSO₄, 2 glucose, and 11.1 Tris base, pH 7.35 (Mulloney and Selverston 1974). Extracellular recordings were made from identified motor nerves with bipolar pin electrodes insulated by vaseline. After desheathing the STG, individual somata were impaled with glass microelectrodes (10-25 M; 2.5 M KCl) and identified with three criteria: 1) 1:1 correspondence of action potentials recorded intracellularly in the soma and extracellularly from an identified motor nerve, 2) characteristic phasing and synaptic input during the pyloric motor pattern, and 3) characteristic shape of the membrane potential oscillations and action potentials in the pyloric rhythm.

DA application

If not indicated differently, DA was bath applied. The volume of the bath was ~3 ml, and the perfusion rate was 3 ml/min. The physiological response was measured after 10 min of bath application of DA (10⁴ M), when a steady state of the DA response was well established. Effects of DA bath application on the isolated PD neuron reversed after 30 min washout with normal saline. We also applied DA by pressure ejection across the entire STG (1 s, 10³ M inside the puffer pipette) with delivery pipettes with a tip diameter of ~50 µm and a suitable ejection pressure of 2 × 10⁴ Pa. To monitor the drug application, Fast Green (1 mg ml¹) was added to the pipette solution.

Physiological recording

The response of the intact pyloric motor pattern to 10⁴ M DA was recorded as described previously, except that two or three neurons were impaled for intracellular recordings. Both intracellular and extracellular recordings were digitized and stored on videotape. The threshold for detectable inhibition of PD by DA is 10⁵ M, and a maximal effect is observed at 10⁴ M (Flamm and Harris-Warrick 1986b).

Synaptic isolation of the PD neuron

The PD neuron was isolated from all detectable synaptic input with three steps (Flamm and Harris-Warrick 1986b). First, modulatory inputs from other ganglia were eliminated with a 10⁴ M tetrodotoxin (TTX) block placed in a small vaseline well on the stomatogastric nerve, the sole input nerve to the STG (Russel 1979). Second, the AB and the other PD neuron (and in most cases also the LP and VD neurons) were photoinactivated by intracellular injection of 5,6-carboxyfluorescein and illumination with UV light (Flamm and Harris-Warrick 1986b; Miller and Selverston 1979). Third, the remaining glutamatergic synapses were blocked with 5 × 10⁶ M picrotoxin (PTX) (Bidaut 1980). The PD neuron was allowed to recover for 1 h before measurements were made. There might be additional sources of synaptic input to the PD neuron after this treatment (Nusbaum et al. 1992), but we saw no evidence for these in our experiments.

Current-clamp recordings of synaptically isolated neurons

After isolation, a PD neuron was impaled with two microelectrodes (9-11 M, 2.5 M KCl or 2.5 M K-acetate with 2 × 10² M KCl) for voltage recording and current injection with an Axoclamp-2A amplifier. The cell was held by DC current injection at a constant membrane potential of 50 mV during the entire experiment. Current protocols were generated with the aid of pCLAMP and a TL1 interface (Axon Instruments) running on a Gateway 2000 microcomputer. Typically a series of 200-ms prepulses with incrementally increasing currents was given to hyperpolarize the neuron to between 50 and 100 mV, followed immediately by a 700-ms depolarizing current pulse to 45 mV, just above threshold for action potential generation. Because DA reduces the input resistance of the PD neuron, the current pulse amplitudes were adjusted throughout the experiment to give relatively constant voltage steps, allowing us to compare the spike frequency in the presence and absence of DA at the same membrane potential.

Voltage clamp of synaptically isolated PD neurons

Synaptically isolated PD neurons were impaled with two electrodes for voltage recording and current recording/injection (9-11 M; 2.5 M KCl or 2.5 M K-acetate with 2 × 10² M KCl). The cell was voltage clamped with an Axoclamp-2A amplifier driven by pClamp. Linear leakage and capacitative currents were digitally subtracted with a P/6 protocol (see Armstrong and Bezanilla 1974).

Current isolation

Currents were isolated with a combination of pharmacological block, voltage inactivation, and digital current subtraction protocols. Sodium currents were blocked by TTX (10⁷ M). Calcium currents were blocked by CdCl₂ (2-6 × 10⁴ M). A hyperpolarization-activated inward current was blocked by CsCl (5 × 10³ M). Currents from glutamatergic synapses were blocked by 5 × 10⁶ M PTX (Bidaut 1980). Tetraethylammonium (TEA, 2 × 10² M) was used to block I_K(V) and I_O(Ca) simultaneously. I_O(Ca) was also indirectly eliminated when the Ca²⁺ currents were blocked by CdCl₂. Although 4-aminopyridine (4AP, 4 × 10³ M) has been shown to be a selective blocker of I_A in other STG neurons (Graubard and Hartline 1991; Tierney and Harris-Warrick 1992), it induced a large and reversible leak current in PD (compare currents evoked by small voltage steps in Fig. 3, A and B), and thus was not used. I_A was instead eliminated by holding the PD neuron at 40 mV, where I_A is almost completely inactivated (Baro et al. 1997).

View larger version (43K): [in this window] [in a new window]   FIG. 3. DA evoked outward current and increase in membrane conductance in a synaptically isolated PD neuron. The cell was voltage clamped at 50 mV. To monitor the conductance, brief (100 ms, 1 Hz) 10-mV hyperpolarizing voltage pulses were applied. For clarity, the peak currents evoked by each voltage pulse are marked by an asterisk. A: DA pulse (1 s, 103 M) evoked a prolonged outward current, accompanied by an increase in membrane conductance. B: DA-evoked outward current and conductance increase were occluded by the combined presence of 4AP (5 × 103 M) and tetraethylammonium (TEA; 2 × 102 M). Note that 4AP evokes a large increase in leak conductance in the PD neuron. This rapidly reversed on removal of 4AP.

Transient K⁺ current I_A

For measurement of I_A, the STG was bathed in saline containing 10⁷ M TTX, 2 × 10² M TEA, 2-6 × 10⁴ M CdCl₂, 5 × 10³ M CsCl, and 5 × 10⁶ M PTX to greatly reduce non-I_A currents. The cell was held at 50 mV. Two series of 10-mV voltage steps between 50 mV and +50 mV were delivered. The first series had no prestep, whereas the second series had a 200 ms prestep to 100 mV to maximally deinactivate I_A. The first series, which evoked the residual non-I_A currents, was digitally subtracted from the second series, which additionally possessed active I_A. The resulting outward current could be completely abolished by 4 × 10³ M 4AP. Although this digital subtraction procedure gives a relatively pure I_A, it also removes the contribution of active I_A at or below 50 mV. However, this was typically 5% of the maximal conductance.

Sustained K⁺ current (I_K(V))

To measure I_K(V), the STG was bathed in 10⁷ M TTX, 2-6 × 10⁴ M CdCl₂, 5 × 10³ M CsCl, and 5 × 10⁶ M PTX to block most of the non-I_K(V) currents, and the cell was held at 40 mV, where I_A is almost completely inactivated. Voltage steps in 10-mV increments between 40 mV and +50 mV were delivered to activate I_K(V). This current was blocked by 2 × 10² M TEA.

Calcium-dependent K⁺ current (I_O(Ca))

For measurement of I_O(Ca), the STG was bathed in 10⁷ M TTX, 5 × 10³ M CsCl, and 5 × 10⁶ M PTX to greatly reduce non-I_O(Ca) currents, and I_A was removed by setting the holding potential to 40 mV. The remaining current was then the sum of I_O(Ca) and I_K(V). After I_O(Ca) was blocked (indirectly) by Cd²⁺, the remaining component (I_K(V)) was digitally subtracted from the summed current of I_O(Ca) and I_K(V) to yield I_O(Ca).

The voltage dependence of activation of I_A and I_K(V) was determined by converting the peak current to a peak conductance, g (assuming E_K = 86 mV) (Hartline and Graubard 1992). The resulting g/V curve was fitted to a third-order (n = 3) and first-order (n = 1) Boltzmann equation of the form

(1)

where g_max is the maximal conductance and s is a slope factor. For the third order Boltzmann fit, V_act is the voltage at which half-maximal activation of the individual gating steps occurs, assuming a third-order activation relation (Hodgkin and Huxley 1952). For the first-order Boltzmann fit V_act (= V_0.5) is the voltage of half-maximal activation of the peak current. For I_O(Ca) the voltage dependence was analyzed by current/voltage plots.

Steady-state inactivation of I_A was measured from a holding potential of 50 mV. Two second voltage presteps were delivered at 10-mV increments from 120 to 50 mV, followed by a step to +20 mV, and the peak current was measured. The data, scaled as a fraction of the calculated maximal conductance, were fit to a first-order Boltzmann equation (Eq. 1 with n = 1), based on the model of Hodgkin and Huxley (1952).

Statistical analysis

For single pairwise comparisons, Student's t-tests were used to assess statistical significance. Significances were accepted at P = 0.025. Throughout this paper, all calculated ranges are reported as means ± SE.

	RESULTS

Abstract Introduction Methods Results Discussion References

DA effect on PD neurons in intact pyloric network

The firing pattern of four of the major cell types in the pyloric network is shown in Fig. 1. The AB and the two PD neurons form the pacemaker group in this circuit. They are electrically coupled and fire synchronously, inhibiting all other neurons in the network. The follower cells recover from this inhibition with different rates and fire with different phases until they are inhibited by the next burst of the pacemaker group (reviewed by Johnson and Hooper 1992; Miller 1987).

As we have previously shown, bath application of 10⁴ M DA modified the pyloric motor pattern (Ayali et al. 1998; Flamm and Harris-Warrick 1986a; Harris-Warrick et al. 1995b) (Fig. 1, A and B) by changing the strength of synaptic connections between the pyloric neurons (Johnson and Harris-Warrick 1990; Johnson et al. 1993a,b, 1995) and changing their intrinsic properties (Harris-Warrick et al. 1995a,b). The AB, LP, and PY cells were excited and increased their firing rate in response to DA. However, the PD cells were inhibited by DA, which evoked a hyperpolarization and a reduction in the firing rate; in some preparations, the PD neurons fell silent. When it was active, the PD no longer fired synchronously with the AB. It fired a few spikes in the middle of the AB burst and thus had a phase delay compared with the first AB spike of each cycle (Fig. 1B).

In Fig. 1C, a DA pulse (1 s, 10³ M) was applied to a PD neuron that was isolated from pyloric chemical synaptic input by PTX (5 × 10⁶ M). The isolated PD cell hyperpolarized by >5 mV and stopped firing for ~1 min. Similar results were obtained in four experiments.

Effects of DA on synaptically isolated PD neurons

We investigated the effects of DA on the rate of recovery after inhibition in PD cells that were isolated from all detectable synaptic input (n = 5). Figure 2 shows such an experiment. Throughout the experiment, the cell was held at 50 mV by tonic current injection. At these holding potential, the isolated PD neuron did not have spontaneous spike activity. To mimic synaptic inhibition, the cell was hyperpolarized with 200-ms current pulses of varying amplitude. This prepulse was followed by a small depolarizing step to 45 mV, which is just above threshold for the initiation of action potentials. To quantify the rate of recovery after inhibition, we measured the latency from the end of the prepulse to the first action potential. We also monitored the spike frequency by measuring the interspike interval (ISI) between the first and second action potential of the subsequent spike train (Hartline 1979; Tierney and Harris-Warrick 1992) and the total number of spikes during the depolarizing step.

View larger version (28K): [in this window] [in a new window]   FIG. 2. Effect of DA on the rate of recovery after hyperpolarizing presteps in a synaptically isolated PD neuron. A: voltage traces of a PD neuron under control conditions, during DA application, and after wash. The neuron was held at 50 mV by current injection throughout the experiment and did not have spontaneous activity at this potential. The voltage traces are offset to enhance readability. Presteps (200 ms) were applied to hyperpolarize the neuron between 50 mV and 80 mV, followed by a 700-ms depolarizing step to 45 mV. The amplitudes of the injected currents were adjusted during the experiment to maintain these voltage steps. B-D: parameters of the spike train during recovery from 200-ms hyperpolarizing presteps from the neuron shown in A under control conditions () and in 104 M DA (). B: latency to first spike. The time from the end of the hyperpolarizing prepulse to the first action potential during the subsequent 700-ms depolarizing step to 45 mV is shown as a function of the membrane potential at the end of the hyperpolarizing prestep. C: first interspike interval (ISI). The time between the first and second action potentials during the depolarizing step is shown as a function of the membrane potential at the end of the hyperpolarizing prestep. D: spikes per step. The total number of action potentials elicited during the 700-ms depolarizing step is shown as a function of the membrane potential at the end of the hyperpolarizing prestep.

At the end of the hyperpolarizing prepulse, the cell began to rebound rapidly, but this slowed down to a prolonged slow ramp depolarization delaying the onset of a spike train. Under control conditions, the latency to the first spike increased as the hyperpolarizing step increased up to approximately 80 mV because of a prolongation of the slow ramp depolarization. Below this membrane potential, the latency remained relatively constant or increased much more slowly because the prolonged slow ramp depolarization became relatively constant. During bath application of DA (10⁴ M), synaptically isolated PD cells hyperpolarized and usually became quiescent and showed reduced input resistance. To compare the intrinsic recovery properties of PD under control and DA conditions, we varied the tonic current injection to maintain the resting membrane potential of the PD at the control level and varied the amplitude of the current steps in the prestep protocol such that the voltage steps were near the control values. During bath application of DA (10⁴ M), the PD cells showed reduced intrinsic recovery properties compared with control (Fig. 2). DA increased the first spike latency at all hyperpolarizing prestep levels (Fig. 2, A and B). For example, the latency after a hyperpolarizing prestep to 80 mV was prolonged by 30.9 ± 16.3% (P < 0.015; n = 5). DA also reduced the spike frequency during the subsequent spike train (Fig. 2, A and C); after a prestep to 80 mV, the first ISI increased by 21.5 ± 10.1% (P < 0.025; n = 5). This combination of increased latency and reduced spike frequency led to a decrease in the total number of spikes during the depolarizing pulse after the hyperpolarizing presteps (Fig. 2, A and D). After the prestep to 80 mV, the number of spikes decreased by 48 ± 8.3% (P < 0.001; n = 5).

DA modulation of ionic conductances

As a first step to explore the ionic mechanisms of DA inhibition of the PD cells, a brief DA pulse (1 s, 10³ M) was applied to a synaptically isolated and voltage clamped PD neuron. DA evoked a small voltage-dependent outward current, accompanied by a 37.8 ± 10.5% (n = 5) increase in membrane conductance (Fig. 3A). As seen in Fig. 3B, both of these effects were eliminated by the combined presence of two potassium channel blockers 4AP (5 × 10³ M), which selectively blocks I_A in STG neurons (Tierney and Harris-Warrick 1992), and TEA (2 × 10² M), which blocks I_O(Ca) and I_K(V) in these neurons. TEA alone reduced the DA evoked outward current by 27.9 ± 6.8% (n = 4) and reduced the DA evoked increase in conductance by 29.4 ± 9.0% (n = 4), suggesting that 4AP blocks the majority of the DA effect. Unfortunately, 4AP activates a large leak current in addition to blocking I_A, as seen by the increase in conductance in Fig. 3B, so it was not possible to accurately ascertain the effects of 4AP alone. This 4AP sensitive leak current was only seen in PD neurons and was not seen in our earlier studies of the LP and PY cells (Harris-Warrick et al. 1995a,b).

These results suggest that modulation of voltage-activated K⁺ currents might contribute to the DA-induced changes in intrinsic rebound properties of PD. To test if K⁺ currents are indeed targets of DA modulation, we performed voltage-clamp studies on synaptically isolated PD cells and determined the effect of DA on a transient voltage-activated K⁺ current (I_A), a sustained voltage-activated K⁺ current (I_K(V)), and a calcium-dependent K⁺ current (I_O(Ca)).

Transient K⁺ current

Under control conditions, I_A activated with voltage steps above 50 mV (Fig. 4, A and B). This current was transient and decayed because of inactivation during a maintained depolarizing voltage step. The conductance/voltage relation (Fig. 4B) was determined from the peak currents evoked by each voltage step. This curve showed a typical voltage dependence for activation for I_A and was fitted to a third-order Boltzmann equation. This fit showed half-maximal activation for each of the individual gating steps at 41.9 mV, leading to half-maximal activation of the peak current (V_0.5) at 21.5 mV. Once inactivated, inactivation had to be removed by hyperpolarization. The voltage dependence of steady state inactivation (Fig. 4 B) was well fitted by a first-order Boltzmann equation, with a voltage for half-maximal inactivation of 69.1 mV. These parameters for I_A in PD were in good agreement with those described by Baro et al. (1997).

View larger version (21K): [in this window] [in a new window]   FIG. 4. Voltage-clamp analysis of the effect of DA on the transient K+ current, IA, in a PD neuron. The cell was held at 50 mV. A: current traces of IA under control conditions, during DA application, and after a wash. Each series represents current responses to increasing voltage steps between 40 and +20 mV in 10-mV increments. IA was isolated by pharmacological blockade and digital subtraction (see METHODS). B: conductance/voltage curves for activation (circles) and inactivation (squares) of IA under control (solid symbols) conditions and in 104 M DA (open symbols). Values are means ± SE (n = 7 for activation; n = 6 for inactivation), calculated as a fraction of the calculated maximal conductance under control conditions in each experiment. The activation curves were generated from peak currents after a maximally deinactivating prestep to 100 mV. The curves are fits to a third-order Boltzmann relation (Eq. 1, METHODS) with the following parameters. Control: gmax = 3.32 ± 0.2 µS; Vact = 41.9 mV; s = 15.5 mV. DA: gmax = 3.65 ± 0.2 µS; Vact = 49.5 mV; s = 15.4 mV. Steady-state inactivation is shown in curves with square symbols. The neurons were held at the indicated potential for 2 s before being stepped to +20 mV. The curves are fits to a 1st order Boltzmann relation (Eq. 1, METHODS), with the following parameters. Control: Vinact = 69.0 mV; s = 6.5. DA: Vinact = 69.8 mV; s = 6.4. C: to demonstrate the increase in tonically active IA near the resting potential, the product of the activation and inactivation curve (from B) is plotted as g/gmax = (1/1 + e(VVact)/sact)3 × (1/1 + e(VVinact)/sinact)1 for control condition and for DA application. The integral between curve and baseline is increased by 174% during DA application.

DA increased I_A by two mechanisms (Fig. 4). First, it significantly increased the maximal conductance by 10.2 ± 2.8% (P < 0.025; n = 7). Second, it shifted the voltage for half-maximal activation to more negative potentials (Fig. 4B). Under control conditions, the V_0.5 for half-maximal activation of the current (from a first-order Boltzmann fit to peak currents) was 21.5 ± 1.3 mV, which shifted significantly to 28.9 ± 1.0 mV (P < 0.02; n = 7) during DA application. However, there was no significant shift in V_0.5 for inactivation. The effects of DA on I_A were reversible after washing with normal saline (Fig. 4A).

The shift in V_act together with the increase in g_max led to an increase in the tonically active "window current" between the steady-state activation and inactivation curves. This is demonstrated in Fig. 4C by plotting the product of the steady state activation (m³) and inactivation (h) curves. These curves showed the fraction of tonically active I_A as a function of membrane potential. During control oscillations (Fig. 1A), the PD neuron hyperpolarized to a mean trough value of 61.6 ± 3.9 mV (mean ± SD; n = 5) (Harris-Warrick, unpublished data), and remained below or close to 60 mV for several hundred milliseconds, sufficient to remove a portion of inactivation at these voltages. Around the normal PD membrane potential, tonic I_A was dramatically increased by DA. For example at 50 and 60 mV, DA caused a 242 and 310% increase in resting I_A, respectively.

Calcium-activated outward current

I_O(Ca) consisted of an inactivating and a noninactivating component (Fig. 5A). Under control conditions, I_O(Ca) activated with voltage steps above 30 mV (Fig. 5, A and B). Measurements of the reversal potential determined from tail current measurements with different external K⁺ concentrations indicated that this current is primarily, but not exclusively, carried by K⁺ ions (Graubard and Hartline 1991; Hartline et al. 1985; Kloppenburg and Harris-Warrick, unpublished data). I_O(Ca) could be eliminated by removal of extracellular Ca²⁺ or addition of >2 × 10⁴ M CdCl₂ or TEA (2 × 10² M). I_O(Ca) showed rundown during prolonged voltage clamp recordings, with a decrease in amplitude of >30% during the first 20 min of recording. This rundown appeared to be caused by depolarizing voltage steps and seemed to be cumulative with regard to the number, length and especially the amplitude of the pulses. To minimize rundown during our DA study, we avoided making measurements within the first few minutes of the recording when rundown was maximal, and limited long-lasting, repetitive high-amplitude depolarizing pulses.

View larger version (14K): [in this window] [in a new window]   FIG. 5. Voltage-clamp analysis of the effect of DA on the calcium dependent outward current, IO(Ca), in a PD neuron. The cell was held at 40 mV. A: current traces of IO(Ca) under control conditions, during DA application, and after a wash. Each series represents current responses to increasing voltage steps between 40 mV and +30 mV in 10-mV increments. IO(Ca) was isolated as described in METHODS. B: current/voltage curves for activation of IO(Ca) under control conditions () and in 104 M DA (). Values are means ± SE from 5 experiments, calculated as a fraction of the maximal current measured under control conditions at +30 mV in each experiment.

In our first series of experiments, we gave a single 200-ms depolarizing pulse to +20 mV every minute before, during, and after DA application. This protocol did not cause marked rundown. DA caused a reversible increase in the magnitude of I_O(Ca). This effect was then confirmed in seven experiments where a series of depolarizing pulses from 40 mV to +30 mV in 10-mV increments was applied before, during, and after DA application (Fig. 5). From these experiments, it is clear that DA increased the magnitude of I_O(Ca) in PD at depolarized voltages; for example, the amplitude of I_O(Ca) at +30 mV was reversibly increased by 24 ± 0.5% (P < 0.01; n = 5). I_O(Ca) probably depends on both intracellular [Ca²⁺] and voltage, and the voltage activation curve combines these two independent factors. As a consequence we did not attempt to dissect further the targets of modulation of this current by DA (see DISCUSSION).

Sustained K⁺ current

When I_A and I_O(Ca) were eliminated, a small noninactivating potassium current, I_K(V), was unmasked. Under control conditions, I_K(V) activated with voltage steps above 30 mV (Fig. 6). The current was sustained and did not decay during a maintained depolarizing voltage step. The voltage activation relation (Fig. 6), constructed from the maximal currents evoked by each voltage step, showed a typical voltage dependence for activation of I_K(V). When fitted to a third-order Boltzmann equation, the current showed half-maximal activation for each of the individual gating steps at 23.7 mV, leading to half-maximal activation of the peak current at +1.9 mV. I_K(V) showed little or no inactivation even with depolarizations lasting 1 s, and there was no detectable voltage dependence of steady-state inactivation (data not shown).

View larger version (21K): [in this window] [in a new window]   FIG. 6. Voltage-clamp analysis of the effect of DA on the sustained K+ current, IK(V), in a PD neuron. The cell was held at 40 mV. A: currents traces of IK(V) under control conditions, during DA application, and after a wash. Each series represents current responses to increasing voltage steps between 40 mV and +40 mV in 10-mV increments. IK(V) was isolated as described in METHODS. B: conductance/voltage curves for activation of IK(V) under control conditions () and in 104 M DA (). Conductance was calculated assuming EK = 86 mV (Hartline and Graubard 1992). Values are means ± SE from 7 experiments, calculated as a fraction of the calculated maximal conductance under control conditions in each experiment. The curves are fits to a 3rd-order Boltzmann relation (Eq. 1, as described in METHODS), with the following parameters. Control: gmax = 0.48 ± 0.04 µS; Vact = 23.7 mV; s = 20.2 mV. DA: gmax = 0.48 ± 0.04 µS; Vact = 22.2 mV; s = 20.4 mV.

Bath application of DA (10⁴ M) had no detectable effect on I_K(V) (Fig. 6). It did not change the maximal conductance, and the small shift in V_0.5 from + 1.9 ± 3.7 mV to +3.3 ± 3.8 mV was not significant (P > 0.05; n = 7).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The starting point of this investigation was the finding that exogenously applied DA alters the rhythmic activity of the pyloric network (Anderson and Barker 1981; Eisen and Marder 1984; Flamm and Harris-Warrick 1986a) (see Fig. 1, A and B). In the two PD neurons, DA evokes a hyperpolarization and a reduction in the number of action potentials per cycle. In part, this inhibition is caused by DA alteration of synaptic inputs to the PD neurons; DA enhances LP inhibition of PD (Johnson et al. 1995). In addition, this inhibition results from direct effects of DA on the PD neuron's own intrinsic properties (Flamm and Harris-Warrick 1986b). DA's inhibition of the PD neurons has three obvious consequences for the pyloric motor pattern (Fig. 1). First, it causes a decrease in cycle frequency (see Abbott et al. 1991). This occurs despite DA's direct excitation of the AB neuron and the acceleration of its cycle frequency (Flamm and Harris-Warrick 1986b) because, when AB is electrically coupled to the hyperpolarized, "leaky" PD neuron, the net effect on the pacemaker group is to reduce cycle frequency (see Abbott et al. 1991). Second, the PD is phase delayed in its onset of spiking relative to the onset of the AB burst (0 in control conditions to around 0.1 in DA). In some preparations, the PD cells fall silent altogether and thus stop signaling to their muscles (Flamm and Harris-Warrick 1986a). Third, in the presence of DA, the PD neurons' graded inhibition of the follower neurons, including LP and PY, is markedly reduced or completely abolished (Johnson and Harris-Warrick 1990). This appears to be due to a reduction of acetylcholine (ACh) release from the PD terminals because the follower cells remain sensitive to ACh. As a partial consequence of this loss of PD inhibition, the LP and PY cells show a phase advance in their burst onset during DA (Eisen and Marder 1984; Flamm and Harris-Warrick 1986a).

In voltage clamp, brief pulses of DA evoke an outward current accompanied by an increase in membrane conductance. These effects are largely blocked by the combination of 4AP and TEA, suggesting K⁺ channel modulation by DA. Direct analyses of three K⁺ currents revealed that the 4AP-sensitive I_A and and the TEA-sensitive I_O(Ca) are indeed enhanced by DA. A detailed analysis of I_A revealed that DA increases its maximum conductance and shifts its voltage activation curve to more negative potentials. This change in conductance goes in the right direction to explain the hyperpolarization and decreased PIR evoked by DA. In particular, the window region of overlap between steady-state activation and inactivation curves is markedly enlarged during DA (Fig. 4C), and tonic I_A is increased two- to threefold in the normal voltage range of the PD neurons. It is unusual to think of the "transient K⁺ current" (I_A) as contributing to the tonic currents that set the resting potential. However, blockade of I_A with 4AP causes synaptically isolated pyloric neurons to depolarize by 10-20 mV (Tierney and Harris-Warrick 1992), suggesting that tonic I_A plays a greater role in setting the resting potential than previously thought. Thus DA enhancement of tonic I_A may explain, at least in part, DA's tonic hyperpolarization of PD. DA also delays the onset of PD firing relative to the AB burst. This phase delay is due in part to the tonic hyperpolarization of PD described previously. However, DA also delays the rate of PD recovery after inhibition (Figs. 1 and 2), and this could result not only from increase in tonic I_A but also an increase in phasic I_A activated in the critical subthreshold voltage range for rebound. Thus DA, acting on the same channels, enhances both tonic potassium currents contributing to the PD resting potential and transient potassium currents determining the rate of PIR and phasing of PD activity in the pyloric motor pattern.

In addition to modulating I_A, DA increases the magnitude of I_O(Ca). Because I_O(Ca) is not only voltage but also Ca²⁺ dependent, it has an important function during prolonged depolarizations that underlie bursts of action potentials (see Hille 1992). Because of Ca²⁺ accumulation during a burst of action potentials, I_O(Ca) will be increasingly activated during the burst. This will influence the interspike intervals and thus spike frequency, the number of spikes, and the overall duration of the burst. DA increases I_O(Ca), which should prolong the interspike interval, decrease the number of spikes per burst, and terminate the burst prematurely. This conclusion fits well with our findings from the rebound experiments in synaptically isolated neurons (Fig. 2) and DA's reduction of PD spike frequency and burst duration in the intact pyloric network (Fig. 1). The I/V curve in Fig. 5B suggests that I_O(Ca) is activated only above 30 mV and thus should not contribute to the resting potential. However, bathing the neuron with low extracellular Ca²⁺ or with Ca²⁺ channel blockers (Cd²⁺ or Co²⁺) causes the PD and other pyloric neurons to depolarize by 10-20 mV (Harris-Warrick, unpublished data). This cannot be a direct effect of reducing I_Ca, which would hyperpolarize the neurons. Instead, these experiments suggest that I_O(Ca) is partially activated at, and contributes to, the resting potential in PD neurons. This hypothesis is further supported by the finding that the DA-induced hyperpolarization from the resting potential of a synaptically isolated neuron can be partially blocked by the I_O(Ca) blocker TEA.

We did not perform a detailed analysis of which parameters of I_O(Ca) are modulated by DA. The amplitude and time course of I_O(Ca) are functions of several parameters, including membrane voltage, Ca²⁺ current kinetics, intracellular Ca²⁺ concentration and sequestration, and intrinsic channel properties. Conclusions about which specific parameters of I_O(Ca) are modulated by DA cannot be made from whole cell currents induced by simple depolarizing steps. Single channel experiments will be needed to determine if DA acts directly on the channels generating I_O(Ca), those generating I_Ca, or even on the kinetics of intracellular calcium metabolism. However, our results show clearly that DA increases the magnitude of I_A and of I_O(Ca), suggesting a synergistic effect of both currents on the rebound properties.

Modeling studies (Golowasch et al. 1992; Harris-Warrick et al. 1995a) demonstrate that modulation of I_K(V) could be an effective mechanism to change neuronal resting potential and firing frequency in STG neurons. However, DA's effects on the intrinsic properties of PD somata appear not to be mediated by this mechanism because I_K(V) was unaltered in the presence of DA.

Although our results demonstrate that DA's enhancement of I_A and I_O(Ca) contribute to its inhibition of the PD neuron, they do not prove that these are the only ionic currents modulated by DA in this neuron. Our voltage-clamp studies measure primarily currents from the cell body, whereas many of the integrative actions of the neuron arise in the neuropil, beyond the space-clamped region of the cell. We would have liked to use TEA and 4AP to try to occlude DA's effects on the firing properties of the unclamped PD neuron to address this question. Unfortunately, 4AP activates a large leak current that significantly reduces the input resistance, and it cannot be used as a simple I_A blocker.

From this and other studies two conclusions can be drawn. First, I_A is a common target of DA modulation in the pyloric network (Harris-Warrick et al. 1995a,b). Although I_A is present in all pyloric neurons, its magnitude varies significantly between neuron types (Baro et al. 1997). Thus, DA could act on I_A in different cells but have quantitatively different effects on their firing properties depending on differences in I_A density in the cells. In addition, DA can have opposite effects on I_A in different neurons. DA inhibits the PD neurons in part by enhancing I_A (this paper) and excites the PY and LP neurons in large part by reducing I_A (Harris-Warrick et al. 1995a, b). Second, the actions of DA are not limited to the modulation of a single current. DA excites the LP neuron not only by modulation of I_A but also by enhancing I_h (Harris-Warrick et al. 1995b). In the PD neurons, both I_A and I_O(Ca) are enhanced by DA, and these currents contribute to DA's actions to inhibit the PD neuron, reduce its rebound properties, and reduce its spike frequency. In addition, DA greatly reduces or abolishes synaptic transmission from PD synapses (Johnson and Harris-Warrick 1990). Although the modulation of K⁺ currents (described in this paper) may contribute to reducing release, other ionic currents (such as Ca²⁺ currents) could be selectively modulated at nerve terminals in a way that is not detectable by voltage clamp of the soma. Experiments are in progress to study these additional effects of DA in distal regions of the neuron.

	ACKNOWLEDGEMENTS

  We thank A. Ayali, D. J. Baro, L. B. French, B. R. Johnson, and J. H. Peck for comments on this manuscript and A. R. Willms for helpful discussions. Special thanks go to L. Davenport.

  This work was supported by National Institute of Neurological Disorders and Stroke Grant NS-17323 and by Office of Naval Research Grant N00003-95-0292.

	FOOTNOTES

  Address for reprint requests: P. Kloppenburg, Cornell University, Section of Neurobiology and Behavior, Seeley G. Mudd Hall, Ithaca, NY 14853.

  Received 15 December 1997; accepted in final form 18 September 1998.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ABBOTT, L. F., MARDER, E., HOOPER, S. L. Oscillating networks: control of burst duration by electrically coupled neurons. Neural Comput. 3: 487-497, 1991.
• ANDERSON, W. W., BARKER, D. L. Synaptic mechanisms that generate network oscillations in the absence of discrete synaptic potentials. J. Exp. Zool. 216: 187-191, 1981.[Medline]
• ARMSTRONG, C. M., BEZANILLA, F. Charge movement associated with the opening and closing of the activation gates of the Na channels. J. Gen. Physiol. 63: 533-552, 1974.[Abstract/Free Full Text]
• AYALI, A., JOHNSON, B. R., HARRIS-WARRICK, R. M. Dopamine modulates graded and spike evoked synaptic inhibition independently at single synapses in the pyloric network of the lobster. J. Neurophysiol. 79: 2063-2069, 1998.[Abstract/Free Full Text]
• BARO, D. J., LEVINI, R. M., KIM, M. T., WILLMS, A. R., LANNING, C. C., RODRIQUEZ, H. E., HARRIS-WARRICK, R. M. Quantitative single-cell-reverse transcription-PCR demonstrates that A-current magnitude varies as a linear function of shal gene expression in identified stomatogastric neurons. J. Neurosci. 17: 6597-6610, 1997.[Abstract/Free Full Text]
• BIDAUT, M. Pharmacological dissection of pyloric network of the lobster stomatogastric ganglion using picrotoxin. J. Neurophysiol. 44: 1089-1101, 1980.[Abstract/Free Full Text]
• EISEN, J. S., MARDER, E. A mechanism for the production of phase shifts in a pattern generator. J. Neurophysiol. 51: 1375-1393, 1984.[Abstract/Free Full Text]
• FLAMM, R. E., HARRIS-WARRICK, R. M. Aminergic modulation in the lobster stomatogastric ganglion. I. Effects on the motor pattern and activity of neurons within the pyloric circuit. J. Neurophysiol. 55: 847-865, 1986a.[Abstract/Free Full Text]
• FLAMM, R. E., HARRIS-WARRICK, R. M. Aminergic modulation in the lobster stomatogastric ganglion. II. Target neurons of dopamine, octopamine and serotonin within the pyloric circuit. J. Neurophysiol. 55: 866-881, 1986b.[Abstract/Free Full Text]
• GETTING, P. A. Emerging principles governing the operation of neural networks. Annu. Rev. Neurosci. 12: 185-204, 1989.[Medline]
• GOLOWASCH, J., BUCHHOLTZ, F., EPSTEIN, I. R., MARDER, E. The contribution of individual ionic currents to activity of a model stomatogastric ganglion neuron. J. Neurophysiol. 67: 341-349, 1992.[Abstract/Free Full Text]
• GRAUBARD, K., HARTLINE, D. K. Voltage clamp analysis of intact stomatogastric neurons. Brain Res. 557: 241-254, 1991.[Medline]
• HARRIS-WARRICK, R. M., CONIGLIO, L. M., BARAZANGI, N., GUCKENHEIMER, J., GUERON, S. Dopamine modulation of transient potassium current evokes phase shifts in a central pattern generator network. J. Neurosci. 15: 342-358, 1995a.[Abstract]
• HARRIS-WARRICK, R. M., CONIGLIO, L. M., LEVINI, R. M., GUERON, S., GUCKENHEIMER, J. Dopamine modulation of two subthreshold currents produces phase shifts in activity of an identified motor neuron. J. Neurophysiol. 74: 1404-1420, 1995b.[Abstract/Free Full Text]
• HARRIS-WARRICK, R. M., FLAMM, R. E. Chemical modulation of a small central pattern generator circuit. Trends Neurosci. 9: 432-437, 1986.
• HARRIS-WARRICK, R. M., MARDER, E. Modulation of neural networks for behavior. Annu. Rev. Neurosci. 14: 39-57, 1991.[Medline]
• HARRIS-WARRICK, R. M., MARDER, E., SELVERSTON, A. I., MOULINS, M., IN In: Dynamic Biological Networks: The Stomatogastric Nervous System. Cambridge, MA: MIT Press, 1992b.
• HARRIS-WARRICK, R. M., NAGY, F., NUSBAUM, M. P. Neuromodulation of stomatogastric networks by identified neurones and transmitters. In: Dynamic Biological Networks: The Stomatogastric Nervous System, edited by R. M. Harris-Warrick, F. Nagy, and M. P. Nusbaum Cambridge, MA: MIT Press, 1992a, p. 87-137.
• HARTLINE, D. K. Pattern generation in the lobster (Panulirus) stomatogastric ganglion. II. Pyloric network simulation. Biol. Cybern. 33: 223-236, 1979.[Medline]
• HARTLINE, D. K., GRAUBARD, K. Cellular and synaptic properties in the crustacean stomatogastric nervous system. In: Dynamic Biological Networks: The Stomatogastric Nervous System, edited by R. M. Harris-Warrick, E. Marder, A. I. Selverston, and M. Moulins Cambridge, MA: MIT Press, 1992, p. 31-85.
• HARTLINE, D. K., PALACIO, J. Y., GASSIE, D. V. Voltage clamp studies of isolated somata from identified pattern generator neurons in the lobster stomatogastric ganglion. Soc. Neurosci. Abstr. 11: 1023, 1985.
• HILLE, B. In: Ionic Channels of Excitable Membranes. Sunderland, MA: Sinauer Associates, 1992.
• HODGKIN, A. L., HUXLEY, A. F. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. (Lond.) 117: 500-544, 1952.
• JOHNSON, B. R., HARRIS-WARRICK, R. M. Aminergic modulation of graded synaptic transmission in the lobster stomatogastric ganglion. J. Neurosci. 10: 2066-2076, 1990.[Abstract]
• JOHNSON, B. R., HOOPER, S. L. Overview of the stomatogastric system. In: Dynamic Biological Networks: The Stomatogastric Nervous System, edited by R. M. Harris-Warrick, E. Marder, A. I. Selverston, and M. Moulins Cambridge, MA: MIT Press, 1992, p. 1-30.
• JOHNSON, B. R., PECK, J. H., HARRIS-WARRICK, R. M. Amine modulation of electrical coupling in the pyloric network of the stomatogastric ganglion. J. Comp. Physiol. [A] 172: 715-732, 1993a.[Medline]
• JOHNSON, B. R., PECK, J. H., HARRIS-WARRICK, R. M. Dopamine induces sign reversal at mixed chemical-electrical synapses. Brain Res. 625: 159-164, 1993b.[Medline]
• JOHNSON, B. R., PECK, J. H., HARRIS-WARRICK, R. M. Distributed amine modulation of graded chemical transmission in the pyloric network of the lobster stomatogastric ganglion. J. Neurophysiol. 74: 437-452, 1995.[Abstract/Free Full Text]
• KLOPPENBURG, P., LEVINI, R. M., HARRIS-WARRICK, R. M. Dopamine changes firing properties of PD neurons in the pyloric network by modulation of K⁺ currents. Soc. Neurosci. Abstr. 23: 478, 1997.
• LEVINI, R. M., PECK, J. H., HARRIS-WARRICK, R. M. The ionic basis of dopamine inhibition of PD neurons in the pyloric network of the lobster stomatogastric ganglion. Soc. Neurosci. Abstr. 22: 2046, 1996.
• MILLER, J. P. Pyloric mechanisms. In: The Crustacean Stomatogastric System, edited by A. I. Selverston, and M. Moulins Berlin: Springer-Verlag, 1987, p. 109-136.
• MILLER, J. P., SELVERSTON, A. I. Rapid killing of single neurons by irradiation of intracellularly injected dye. Science 206: 702-704, 1979.[Abstract/Free Full Text]
• MULLONEY, B. Neural circuits. In: The Crustacean Stomatogastric System, edited by A. I. Selverston, and M. Moulins Berlin: Springer-Verlag, 1987, p. 57-75.
• MULLONEY, B., SELVERSTON, A. I. Organization of the stomatogastric ganglion in the spiny lobster. III. Coordination of the two subsets of the gastric system. J. Comp. Physiol. 91: 53-78, 1974.
• NUSBAUM, M. P., WEIMANN, J. M., GOLOWASCH, J., MARDER, E. Presynaptic control of modulatory fibers by their neural network targets. J. Neurosci. 12: 2706-2714, 1992.[Abstract]
• PEARSON, K. G. Common principles of motor control in vertebrates and invertebrates. Annu. Rev. Neurosci. 16: 265-297, 1993.[Medline]
• RUSSEL, D. F. CNS control of pattern generation in the lobster stomatogastric ganglion. Brain Res. 177: 598-602, 1979.[Medline]
• SELVERSTON, A. I., MOULINS, M. Oscillatory neural networks. Annu. Rev. Physiol. 47: 29-48, 1985.[Medline]
• SELVERSTON, A. I., RUSSELL, D. F., MILLER, J. P., KING, D. G. The stomatogastric nervous system: structure and function of a small neural network. Prog. Neurobiol. 7: 215-290, 1976.[Medline]
• TIERNEY, A. J., HARRIS-WARRICK, R. M. Physiological role of the transient potassium current in the pyloric circuit of the lobster stomatogastric ganglion. J. Neurophysiol. 67: 599-609, 1992.[Abstract/Free Full Text]

0022-3077/99 $5.00 Copyright ©1999 The American Physiological Society