INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A widespread phenomenon in the nervous system, in both neurons and neuroendocrine cells, is the modulation of cellular communication by transmitters and intracellular messengers. Modulators that act via metabotropic receptors can enhance or inhibit the output of the receiving cell through various mechanisms. Presynaptic inhibition of neurotransmitter release can result from a reduction of Ca²⁺ influx by inhibition of voltage-gated Ca²⁺ channels, which has the greatest impact on neurotransmission (Wu and Saggau 1997). In addition, it can result from modulation of the release machinery downstream of Ca²⁺ entry that affects the strength of the stimulus-secretion coupling or from an increase in potassium conductance shortening the duration of action potentials and increasing the threshold of firing.

In neuroendocrine cells, the hormonal output can be modulated by similar mechanisms and through multiple intracellular messengers. For example, adrenaline stimulates glucagon secretion by both affecting Ca²⁺ channels and the docking of large dense-cored vesicles (LDCVs) in pancreatic A cells (Gromada et al. 1997). In pancreatic  cells, cAMP potentiates insulin release by affecting Ca²⁺ channels as well as the release machinery (Ammala et al. 1993a). In chromaffin cells, activating protein kinase A via cAMP enhances release by activating a facilitation Ca²⁺ channel (Artalejo et al. 1990, 1994). Protein kinase C augments secretion by increasing the size of the readily releasable pool of secretory vesicles in these cells (Gillis et al. 1996; Vitale et al. 1995). Thus modulation of release by simultaneously altering Ca²⁺ channel functioning and modifying the exocytotic machinery downstream of Ca²⁺ entry appears to be a common mechanism both in neurons and in neuroendocrine cells (Kits and Mansvelder 2000).

Melanotropes of the intermediate pituitary produce hormones such as -melanocyte stimulating hormone (-MSH) and -endorphin, which are packaged in LDCVs that are subsequently released into the bloodstream. As in other neuroendocrine cells and neurons, the influx of Ca²⁺ ions through voltage-gated Ca²⁺ channels is the most important stimulus to trigger the release of the contents of LDCVs (Lee 1996; Thomas et al. 1990; Tomiko et al. 1981). Melanotropes receive synaptic inhibition from GABA- and dopamine-containing synapses (MacVicar and Pittman 1986). GABA directly inhibits the melanotrope by hyperpolarization of the cell membrane through activation of GABA_A receptors. Dopamine inhibits the hormonal output of the melanotrope by activating D₂ receptors that subsequently trigger different G-protein-dependent mechanisms. The cell membrane is hyperpolarized by activation of a potassium conductance and inhibition of voltage-gated Ca²⁺ channels, presumably through a direct G-protein-channel interaction (Keja et al. 1992; Nussinovitch and Kleinhaus 1992; Stack and Surprenant 1991; Williams et al. 1989, 1990). However, it is unknown whether dopamine directly interacts with exocytosis of LDCVs independent of Ca²⁺ entry.

Melanotropes express five different Ca²⁺ channel types (Mansvelder et al. 1996). Our laboratory previously reported that both high-voltage activated Ca²⁺ channels and low-voltage-activated T-type Ca²⁺ channels are inhibited by dopamine (Keja et al. 1992). However, it is unknown how the different types of Ca²⁺ channels are affected by dopamine individually. Moreover it has not been studied how dopamine's modulation of Ca²⁺ channels affects the stimulus-secretion coupling in these cells. In this study we address these questions. We find that dopamine differentially affects high-voltage-activated Ca²⁺ channels and that it sensitizes exocytosis downstream of Ca²⁺ entry. This outlines a previously unidentified mechanism by which dopamine reduces Ca²⁺ entry but leaves the stimulus-secretion coupling intact.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell culture

Pituitary melanotropic cells of male Wistar rats (200-300 g, Harlan CPB, Netherlands) were isolated as described previously (Keja et al. 1991). The cells were cultured on poly-L-lysine coated coverslips (7 × 7 mm) at a density of 0.1 intermediate lobe per coverslip. The culture medium consisted of Biorich I (Flow), 26.2 mM NaHCO₃, 5% Ultroser G (Gibco), 200 U/ml penicillin G (Sigma), 50 µg/ml streptomycin (Sigma), and 1 µM cytosine arabinosine (Sigma). Cells were maintained in a 37°C incubator with a humidified atmosphere comprising 5% CO₂ in air. Recordings were made up to 4 days after isolation.

Pharmacology of the whole cell Ba²⁺ current

Coverslips bearing melanotropic cells were transferred to the recording chamber, containing 300 µl external solution flowing through the chamber at a rate of approximately 1.5 ml/min. The external solution contained (in mM): 130 TEACl, 10 glucose, 13 BaCl₂, 10 HEPES, and 1 4-aminopyridine, pH 7.4 adjusted with TEAOH. Apart from the control experiments, all experiments were performed with 1 µM dopamine in the external solution. Because of the oxidation of dopamine, new external solution containing dopamine was prepared after 3 h. The pipette solution contained (in mM): 135 CsCl₂, 2 MgCl₂, 1 CaCl₂, 10 HEPES, 11 ethylene glycol-bis(beta-aminoethyl ether) N,N,N',N'-tetraacetic acid (EGTA, Sigma), 2 MgATP, 0.1 TrisGTP, pH 7.4 adjusted with CsOH.

Electrodes were pulled on a Flaming/Brown P-87 (Sutter Instruments, Novato, CA) horizontal microelectrode puller from thick-walled Clark GC-150 borosilicate glass (Clark Electromedical Instruments, Pangbourne Reading, UK). To reduce the pipette capacitance, the tip of the electrodes were covered with silicone elastomer (Sylgard). The impedance of the pipettes after fire polishing was 2-4 M. Whole cell Ba²⁺ currents were filtered at 1 kHz, using the 4-pole low-pass Bessel filter of the Axopatch 200A amplifier (Axon Instruments, Foster City, CA), before digitizing with a Digidata 1200 (Axon Instruments) at a sampling frequency of 10 kHz. Access resistance after establishment of the whole-cell configuration was 7.9 ± 1.28 M (n = 80) and was not compensated to reduce noise. Experiments started at least ten minutes after transference of the cells to the recording chamber, to allow dopamine to reach a steady state of block. Recordings were made at a holding potential of 80 mV, and effects of blockers were tested at a potential of +10 mV. The effects of the blockers were measured when a steady-state level of block was reached. Experiments were performed and analyzed using pClamp 6 software (Axon Instruments).

Capacitance measurements

MEASURING EXOCYTOSIS INDUCED BY STEP DEPOLARIZATIONS. The external solution consisted of (in mM): 142 TEACl, 10 glucose; 5 CaCl₂, 10 HEPES, 1 4-aminopyridine; pH adjusted to 7.4 with TEAOH. The internal solution for these experiments contained (in mM): 145 CsCl, 2 MgCl₂, 0.1 EGTA, 10 HEPES, and 2 MgATP; pH adjusted to 7.4 with CsOH. The whole cell membrane current was monitored and digitized with an EPC 9 amplifier (HEKA, Lambrecht Germany). Capacitance measurements were made using the PULSE software running on an IBM compatible Intel Pentium based computer. The membrane capacitance, access conductance and membrane conductance were calculated according to the Lindau-Neher technique (Gillis 1995), implemented as the "sine + DC" feature of the PULSE lock-in module. A sinewave of 1,000 Hz, 40 mV peak-to-peak, was added to a holding potential of 80 mV. The reversal potential of the lock-in module was set to 0 mV. The membrane current was sampled at 10 kHz before, during, and after either the step depolarization or the action potential template, after it was filtered at 2 kHz by the Bessel filter of the EPC 9. The membrane capacitance, access conductance, and membrane conductance were calculated at 1 kHz. The experiments concerning exocytosis induced by step depolarizations were performed at 33°C.

1

1

MEASURING EXOCYTOSIS INDUCED BY 3 µM FREE INTERNAL CA²⁺. Continuous monitoring membrane capacitance for several minutes of whole cell recording was done using the "piecewise-linear" technique, with a software-based phase-sensitive detector (Joshi and Fernandez 1988; Neher and Marty 1982). These experiments were performed at room temperature (20-22°C). The external solution consisted of (mM): 130 TEACl, 10 glucose, 13 BaCl₂, 10 HEPES, and 1 4-AP. The pH was adjusted to 7.4 with TEAOH. The internal solution containing 3 µM free internal Ca²⁺ was (mM): 135 CsCl, 10.81 CaCl₂, 11 EGTA, 2 MgATP, 2 MgCl₂, 0.1 trisGTP, and 10 HEPES. The pH was adjusted to 7.4 with CsOH. The whole cell membrane current was monitored with an Axopatch 200B amplifier (Axon Instruments) and was digitized with a Digidata 1200 interface (Axon Instruments). The source codes of the acquisition and analysis software, which run in an Axobasic environment (Axon Instruments), were acquired from Axon Instruments and were modified in our laboratory. A 40-mV peak-to-peak, 1.2-kHz sine wave was added to a holding potential of 80 mV, and the resulting membrane current was filtered at 2 kHz (4-pole low-pass Bessel filter on the Axopatch 200A) and sampled at 20 kHz. Before analysis at two orthogonal phase angles, 10 cycles of the sine wave were averaged. The correct phase angle was determined every 19 s by repetitive computer-controlled switching of a 500 k resistor in series with ground until changes in the capacitance trace were minimal. An independent measure of the membrane conductance was obtained by applying a hyperpolarizing pulse of 20 mV and 6-ms duration to the membrane in between two groups of 10 sine waves. The resulting temporal resolution of each capacitance, total conductance, and membrane conductance point was 19 ms. Changes in capacitance were calibrated by a temporary 100 fF change in the compensation circuitry of the amplifier, and the conductance trace was calibrated by the 500 k dither resistor.

Data analysis

The amount of exocytosis was calculated as the difference between the average of 100 membrane capacitance samples before and the first 10 samples following a particular depolarizing pulse or action potential template. Quantifications of exocytosis were corrected for the transient capacitance change (C_t) (Horrigan and Bookman 1994; Mansvelder and Kits 1998), which was 2.0 ± 0.29 fF. This values was determined by depolarizing the cell in the presence of Ni²⁺ (40 µM) and Cd²⁺ (100 µM), which blocked all calcium currents (n = 4 cells, not shown). Statistical significance of differences of means were determined with Student's t-test, with the Systat software (Evanson, IL). Fitting of data were done using the NLREG v3.4 software of Ph. H. Sherrod (Nashville, TN). Means mentioned in the text are given with standard errors of the mean (SE) unless mentioned otherwise. Error bars signify SE.

Drugs and application

Calcium channel blockers were applied with pressure ejection from a glass pipette. Nimodipine was purchased from RBI (Natick, MA), -conotoxin GVIA, -agatoxin TK, and -conotoxin MVIIC were obtained from Alamone Labs (Jerusalem, Israel). The toxins were stored in stock solutions at 20°C in small aliquots for single use, and these were diluted in external medium to reach their final concentrations immediately before the experiments started. Nimodipine was stored in a stock solution of 10² M in methanol at 20°C. This was diluted in external solution to the final concentration daily, immediately before the experiments. The percentage methanol was below 0.1%, which did not affect Ba²⁺ currents. Dopamine was obtained from Sigma (St. Louis, MO).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Kinetics of dopamine inhibition of Ba²⁺ currents

On basis of pharmacological sensitivity, we previously identified five different types of Ca²⁺ channels in these cells: L, N, P, Q, and T-type (Keja et al. 1992; Mansvelder et al. 1996). It is well established that dopamine affects voltage-dependent Ca²⁺ channels in rat melanotropic cells (Douglas and Taraskevich 1982; Keja et al. 1992; Stack and Surprenant 1991; Williams et al. 1990). However, none of the studies examined the effect of dopamine on the various high-voltage-activated (HVA) components of the Ca²⁺ current in isolation. Therefore we first determined the time course of the effect of dopamine on the whole cell Ca²⁺ current and subsequently studied the modulation of the individual HVA currents. With Ba²⁺ as charge carrier, the average peak whole cell current under control condition was 184 ± 79.4 (SD) pA. Selective D₂ receptor agonists like quinpirole and LY171555 had a maximal effect on Ba²⁺ currents at 1 µM (Keja et al. 1992; Stack and Surprenant 1991). Dopamine at this concentration reduced the whole cell Ba²⁺ current by 22 ± 4.2% after 30 min of application (Fig. 1). The time constant of inhibition of the Ba²⁺ current was 7.7 min. During the entire application, there was no sign of desensitization.

View larger version (15K): [in this window] [in a new window]   Fig. 1. The effect of dopamine (1 µM) on the whole cell Ba2+ current. Dopamine was applied after 3.3 min whole cell recording. Ba2+ currents were evoked every 20 s. , the average of the peak current of 6 experiments that were normalized to their maximal control current. The averages were corrected for washout of the Ba2+ current, which inevitably occurs during long recordings, even with GTP and ATP present in the pipette solution. After 37 min of recording, washout of 21.5% of the Ba2+ current was observed (n = 4). A single exponential decay function was fitted to these averages. The time constant of this function was 7.7 min.

Sustained Ba²⁺ currents are blocked by dopamine

Our next aim was to determine the effect of dopamine on the different components of the Ba²⁺ current. In a previous study (Mansvelder et al. 1996), we found that the L-type current blocked by nimodipine (1 nM) as well as the P-type current blocked by low concentrations -AgTx IVA (-AgTx; 10 nM) show no inactivation during a 200-ms step depolarization to +10 mV (Fig. 2, A and B, right insets). The contribution of the L-type current to the total Ba²⁺ current under control condition was 34 ± 4.5% (Mansvelder et al. 1996). With 1 µM dopamine present in the bath, nimodipine (1 nM) had only a very minor effect on the Ba²⁺ current (Fig. 2, A and left inset). It reduced the Ba²⁺ current by 9 ± 3.2% (n = 7). This suggests that dopamine largely blocked the L-type current, leaving almost no current to be blocked by 1 nM nimodipine.

View larger version (23K): [in this window] [in a new window]   Fig. 2. Dopamine blocks sustained L- and P-type Ca2+ currents. A: example of a representative experiment, wherein 1 nM nimodipine had only a minor effect on the whole cell Ba2+ current. , the peak Ba2+ current during a 200-ms step depolarization. Left inset: example traces in the absence and presence of 1 µM dopamine of the Ba2+ current evoked by such step depolarizations. Scale bars: horizontal, 50 ms; vertical: 50 pA (scale bars also apply to the left inset of B). Right inset: the normalized average of the sustained current that is blocked by 1 nM nimodipine in the absence of dopamine (n = 10). B: example of a representative experiment showing that 10 nM -AgTx has no effect on the whole cell Ba2+ current in the presence of 1 µM dopamine. Left inset: example traces in the absence and presence of dopamine. Right inset: the normalized average of the current that is blocked by 10 nM -AgTx in the absence of dopamine. Right insets are taken from Mansvelder et al. (1996).

The current sensitive to low concentrations of -AgTx IVA contributed 19 ± 3.4% to the Ba²⁺ current under control conditions. The right inset of Fig. 2B shows that in the absence of dopamine, -AgTx blocked a sustained component of the Ba²⁺ current. In the presence of dopamine, 10 nM -AgTx had no effect on the Ba²⁺ current (n = 7; Fig. 2, B and left inset). These data suggest that, in addition to reducing the L-type current, dopamine completely blocked the P-type current, leaving 10 nM -AgTx ineffective in blocking part of the Ba²⁺ current.

Dopamine alters kinetics of inactivating Ba²⁺ currents

Both the snail toxin -conotoxin GVIA (-CgTx) and higher concentrations of -AgTx (0.1-1 µM) block current components distinct from the sustained L- and P-type currents in the absence of dopamine (Mansvelder et al. 1996). These components have been identified as N and Q type. The N-type current contributed 26 ± 3.6% to the whole cell Ba²⁺ current. With 1 µM dopamine, -CgTx still blocked a substantial part of the whole-cell Ba²⁺ current (38 ± 4.6%; Fig. 3A). The time course of block was similar to the time course observed under control conditions. This shows that despite the presence of dopamine, the N-type channel is activated by a step depolarization. The Q-type current contributed 12 ± 2.6% to the whole cell Ba²⁺ current under control conditions. In the presence of dopamine, 1 µM -AgTx blocked 21 ± 4.7% of the Ba²⁺ current (Fig. 3B). Thus the Q-type current is activated during the step depolarization despite the presence of dopamine.

View larger version (20K): [in this window] [in a new window]   Fig. 3. Dopamine changes the kinetics of the inactivating Ba2+ currents. A: the application of 1 µM -CgTx to the whole cell Ba2+ current in the presence of 1 µM dopamine results in a reduction of the peak amplitude. Each dot represents the peak current during a 200-ms step depolarization. Inset: example traces in the absence and in the presence of -CgTx. B: typical example showing that 1 µM -AgTx blocks a part of the whole cell Ba2+ current in the presence of 1 µM dopamine. Inset: the effect on example traces. Note that the difference between the traces increases during the depolarization. C: difference plots showing the current blocked by 1 µM -CgTx and 1 µM -AgTx in the presence of dopamine (top traces). Bottom traces: the currents blocked by 1 µM -CgTx and 1 µM -AgTx in the absence of dopamine. Both currents were inactivating under control conditions. Dopamine changes the inactivation properties and introduces a slow activation time constant in the -AgTx sensitive current. The control traces were taken from Mansvelder et al. (1996). Scale bars A and C: horizontal 50 pA, vertical 50 ms. The scale bars of the inset in A also apply to the inset in B.

Without dopamine, the currents blocked by -CgTx and -AgTx inactivate with time constants of inactivation of approximately 100 and approximately 180 ms, respectively (controls in Fig. 3C). In the presence of 1 µM dopamine, the blocked currents no longer inactivated (traces marked dopamine in Fig. 3C). The current blocked by -AgTx had a phase of slow activation. This is reminiscent of the slow voltage-dependent disinhibition of Ca²⁺ channels, presumably by unbinding of the subunit of the G protein from the Ca²⁺ channel ₁ subunit (Dolphin 1995, 1998; Hille 1994), which was reported for the melanotropic cell (Keja and Kits 1994; Keja et al. 1992). The slow time constant of activation of the Q-type current in the presence of dopamine was approximately 40 ms, which is in the same order as the previously reported short-term inhibition of the total Ba²⁺ current by selective D₂-receptor activation (Keja and Kits 1994; Keja et al. 1992). Thus although the N- and Q-type channels are still opening during a step depolarization in the sustained presence of dopamine, their activation and inactivation properties are affected by dopamine receptor activation.

Figure 4 summarizes the relative contribution of different current types to the total whole cell Ba²⁺ current in the absence and presence of dopamine. Both the contributions of the L- and P-type current are reduced by dopamine. As a result, the contribution of the N- and Q-type current to the whole cell current is enhanced by the presence of dopamine (Fig. 4, A and B). These results together show that dopamine differentially affects high-voltage activated Ca²⁺ currents in melanotropic cells, either by reducing the amplitude or the changing the kinetics of the current.

View larger version (23K): [in this window] [in a new window]   Fig. 4. Statistical analysis of the contribution of different Ca2+ channel types to the whole cell Ba2+ current in the absence (A) and presence of dopamine (B). The number of experiments is indicated in parentheses above each bar. Differences of mean contributions between control and in the presence of dopamine, for the different channel types, were all statistically significant at P < 0.05.

Stimulus-secretion coupling is not affected by dopamine

Our next aim was to examine the effect of dopamine on exocytotic output of the melanotropic cell. We used high-resolution membrane capacitance measurements to monitor membrane surface area, as an assay for exocytosis (Neher and Marty 1982). Exocytosis has a linear relation to the amount of Ca²⁺ that enters the cell during a step depolarization in melanotropic cells (Mansvelder and Kits 1998, 2000a,b). Because dopamine reduces the whole cell Ca²⁺ current, our most straightforward expectation was that the amount of stimulated exocytosis would be reduced proportionally to the reduction in Ca²⁺ entry. In these experiments, Ca²⁺ was used as charge carrier to stimulate exocytosis, and the experiments were performed at 33°C. Under control conditions, a step depolarization of 40 ms to +10 mV generated a Ca²⁺ current with a peak amplitude of 173 ± 13.0 pA (n = 18; Fig. 5). In the presence of 1 µM dopamine, the peak amplitude of the Ca²⁺ current was reduced to 122 ± 14.2 pA (P < 0.05; n = 13, Figs. 5 and 6, top right). To our surprise, the amount of exocytosis elicited by these Ca²⁺ currents in the presence of dopamine did not differ significantly from control. In the absence of dopamine, the membrane capacitance (C_m) changed by 17 ± 2.3 fF as a result of the step depolarization and the subsequent Ca²⁺ influx (Fig. 5). The amount of C_m change in the presence of dopamine was 15 ± 1.7 fF (P = 0.5; Figs. 5 and 6, top left). This shows that, although the peak Ca²⁺ current is reduced by dopamine, this does not affect the amount of exocytosis that is elicited by a step depolarization.

View larger version (13K): [in this window] [in a new window]   Fig. 5. Dopamine reduces the amplitude of the Ca2+ current in melantropes, but the amount of exocytosis is hardly affected. Top: examples of membrane capacitance, Cm, trace in the absence and presence of 1 µM dopamine. Each cell was subjected to 5 step depolarizations of which 1 Cm response is shown. During the depolarizations, the Cm measurement was interrupted. Bottom: examples of Ca2+ current evoked by the step depolarizations. Note the difference in time scale with the upper traces.

View larger version (27K): [in this window] [in a new window]   Fig. 6. Dopamine increases the efficacy of Ca2+ ions to stimulate exocytosis. Summary of the experiments on the effect of dopamine on the Ca2+ current and exocytosis (control n = 18, dopamine n = 13). Top left: exocytosis is not significantly affected. Top right: the peak Ca2+ current is reduced. Bottom left: the number of Ca2+ ions the entered the cell during the depolarization is reduced. Bottom right: the efficacy of Ca2+ ions to stimulate exocytosis is enhanced. *, a significant difference with the control group at P < 0.05. For each experiment, a different cell was used. Per cell the mean of the response to 5 depolarizations was calculated. The average of these means is shown.

As a result of the reduction of the overall amplitude of the Ca²⁺ current, the number of Ca²⁺ ions that enter the cell during the step depolarization was also reduced. Under control conditions, approximately 11 million Ca²⁺ ions entered the cell during the depolarization. With dopamine, this number was reduced to approximately 7 million (P < 0.05; Fig. 6, bottom left). In Fig. 6, bottom right, the amount of exocytosis is expressed as a fraction of the number Ca²⁺ ions that entered the cell, thus expressing the efficacy of Ca²⁺ ions to stimulate exocytosis. Entry of Ca²⁺ more efficiently stimulated exocytosis in the presence of dopamine than in the absence of dopamine (Fig. 6, bottom right). Without dopamine, influx of 1 million Ca²⁺ evoked a Cm increase of 1.8 ± 0.25 fF, whereas with dopamine, entry of the same amount of Ca²⁺ ions induced a Cm increase of 2.6 ± 0.31 fF (P < 0.05).

Exocytosis is sensitized downstream of Ca²⁺ entry

There are different possible explanations for the lack of effect of dopamine on the amount of exocytosis and the increased efficacy of Ca²⁺ ions to stimulate exocytosis. It could be that the amount of Ca²⁺ entering the cell during a 40-ms step depolarization in the presence of dopamine already saturates exocytosis. As a result, the amount of exocytosis in the presence of dopamine would not differ from control conditions. However, in two previous studies, we addressed the issue of saturation both with step depolarizations and action potential waveforms (Mansvelder and Kits 1998, 2000a). In both cases, exocytosis increased linearly with the amount of calcium entry, without showing any sign of saturation. With increasing step depolarization duration, from 2 to 40 ms, exocytosis increased linearly with the amount of calcium entry (Fig. 7), thereby excluding saturation as an explanation for the absence of effect of dopamine on exocytosis. Alternatively, it could be that different types of Ca²⁺ channels couple differentially to exocytosis. If dopamine shuts down those channels that are involved to a lesser extend in stimulating exocytosis, an increased efficacy of Ca²⁺ ions to stimulate exocytosis would result. However, recently we showed that all different Ca²⁺ channel types present in the melanotropic cell couple equally efficient to exocytosis (Mansvelder and Kits 2000a), thus ruling out the possibility that dopamine preferentially blocks Ca²⁺ channels that contribute little to the stimulation of exocytosis. Therefore we considered a third possible explanation. We hypothesized that the observed increase in efficiency results from a direct effect of dopamine on exocytosis independent from its effect on Ca²⁺ channel functioning and Ca²⁺ entry. To test this, we stimulated exocytosis by dialyzing cells with buffered Ca²⁺ solutions containing 3 µM free Ca²⁺, thereby stimulating exocytosis without activating voltage-gated Ca²⁺ channels.

View larger version (12K): [in this window] [in a new window]   Fig. 7. Exocytosis increases linearly with the amount of Ca2+ entry, without showing saturation. Increasing step depolarization duration from 2 to 40 ms proportionally increases Ca2+ entry and exocytosis. All pulse durations were tested within the same recording from one cell. Plotted are the average data from 8 cells. The amount of Ca2+ entry and exocytosis at the end of a train of 15 pulses at 1 Hz are shown. The numbers above the data points indicate the pulse duration of each step during the pulse train. Data taken from Figs. 10 and 11 in Mansvelder and Kits (1998) and presented in modified form.

In the absence of dopamine, 3 µM free intracellular Ca²⁺ evoked a gradual increase in Cm (Fig. 8A). This is in accordance with the findings of other groups (Lee 1996; Okano et al. 1993; Sikdar et al. 1998). Three minutes after establishing the whole cell configuration C_m was increased by 330 ± 38.0 fF (Fig. 8C). In the presence of 1 µM dopamine, C_m increased much faster (Fig. 8B). After 1 min of recording, C_m was increased by 217 ± 44.6 fF, which was significantly higher than under control conditions (P < 0.05, Fig. 8C). After 90 and 180 s, C_m was still significantly higher with dopamine. However, at 180 s the difference between control and dopamine no longer increased (Fig. 8C), suggesting that after 180 s the release with dopamine is slowed down to the level of control conditions. This is confirmed by examination of the rate of release at different times after patch rupture (Fig. 8C). The rate of release is enhanced by dopamine primarily early in the recording. After 90 s of recording the difference in rate was most prominent, 1.8 ± 0.30 fF s¹ in control versus 5.5 ± 1.3 fF s¹ with 1 µM dopamine (P < 0.05). After 180 s, the rate of exocytosis dropped to 1.5 fF s¹ both in control and in the presence of dopamine. Although the average rate of exocytosis in the presence of dopamine slows down between 90 and 180 s after break in, we hardly ever encountered the response reaching a plateau during this time, neither in control nor in dopamine conditions. Of the 20 controls, three cells showed an average rate of exocytosis below 0.1 fF/s at 180 s of recording. Of the 21 cells in the presence of dopamine, the number of cells with an average rate of exocytosis below 0.1 fF/s after 180 s of recording was 4. 

View larger version (21K): [in this window] [in a new window]   Fig. 8. Dopamine affects the early phase of exocytosis independent of its effect on Ca2+ channel functioning. A, top: example Cm traces in the absence and presence of dopamine. Exocytosis was stimulated by dialyzing the cell with a buffered Ca2+ solution containing 3 µM free Ca2+, in the absence and presence of 1 µM dopamine. Bottom: membrane conductance (Gm) traces recorded from the same cells as in the top graph. Note the absence of parallel changes with the Cm traces. In our recordings, a change of 500 pS in membrane conductance resulted in a cross talk of 10 fF on the Cm trace (not shown). We have determined this value by specifically changing Gm using GABA applications while performing membrane capacitance measurements. B: average of 20 control experiments and 21 experiments in the presence of dopamine. The change in Cm was determined after 30, 60, 90, and 180 s. C: average rate of Cm increase plotted as function of time for the same data as in B. Due to the presence of occasional fluctuations in the Cm traces, the average rate was determined by taking the average increase over a period of 30 s, i.e., 1,578 data points. This is plotted as the average rate during period of 30 s. The rates during the 1st 90 s were significantly higher in the presence of dopamine than controls (P < 0.05). , control. , in the presence of 1 µM dopamine. Experiments were performed at room temperature.

Together, these results suggest that dopamine affects the probability of release of the vesicles that fuse in the early exocytotic phase. Vesicles that fuse later may not be affected by dopamine, or to a lesser extent. These results confirm the hypothesis that dopamine increases exocytosis independent of its effect on Ca²⁺ channel functioning. The increased probability of release of the early vesicles may cause the increased efficacy of Ca²⁺ ions entering through voltage-gated Ca²⁺ channels observed in the experiments described in Figs. 5 and 6.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

It is well established that dopamine reduces the Ca²⁺ influx in melanotropic cells (Douglas and Taraskevich 1982; Keja et al. 1992; Lee 1996; Nussinovitch and Kleinhaus 1992; Stack and Surprenant 1991; Williams et al. 1989). However, it was unknown how dopamine affects the different Ca²⁺ channel types individually. It was also not known whether dopamine interacts with exocytosis directly, independent of Ca²⁺ entry. In this study, we examined the effect of dopamine on the different Ca²⁺ channel types and on exocytosis. We conclude that the stimulus-secretion coupling is unaffected by dopamine, despite its modulation of Ca²⁺ channels, due to a direct effect on the exocytotic machinery.

Modulation of the HVA Ca²⁺ channels

It is reported that D₂ receptor activation reduces the peak amplitude and changes the kinetics of the macroscopic Ca²⁺ current in melanotropes (Keja and Kits 1994; Keja et al. 1992; Stack and Surprenant 1991). We dissected out what the specific effects of dopamine on each of the different Ca²⁺ channel types are that underlie these changes of the macroscopic Ca²⁺ current. We found that dopamine reduced the amplitude of the total current by reducing the sustained L- and P-type current. Moreover, dopamine changed the kinetics of the inactivating N- and Q-type currents.

Activation of dopamine D₂ receptors results in a decrease of intracellular cAMP levels through a reduction of adenylate cyclase activity (Frey et al. 1982). This occurs via activation of a pertussis toxin-sensitive G protein (Cote et al. 1982; Taraskevich and Douglas 1990). Generally speaking, Ca²⁺ channels can be modulated by phosphorylation and dephosphorylation or directly by G proteins (Dolphin 1995, 1998; Herlitze et al. 1996; Ikeda 1996). Phosphorylation by the cAMP-dependent protein kinase A often results in an enhancement of the current amplitude, whereas dephosphorylation leads to a reduction of the Ca²⁺ current (Dolphin 1995). Since D₂ receptor activation leads to a reduction of cAMP levels in melanotropes, the reduction of the L-and P-type currents may result from reduced phosphorylation conditions by dopamine.

Direct interaction of Ca²⁺ channels with the G protein leads to a slow down of activation of N- and P/Q-type channels (Dolphin 1995, 1998; Hille 1994). It has been shown that it is the subunit of the G protein that interacts with the ₁ subunit of the Ca²⁺ channel (Dolphin 1998; Herlitze et al. 1996; Ikeda 1996). The slow time constant of activation reflects the voltage-dependent recovery from the inhibition by G. In the melanotropic cell, application of selective dopamine D₂ receptor agonists slows the time constant of activation of the Ba²⁺ current (Keja and Kits 1994; Keja et al. 1992). This effect is G protein mediated and voltage-dependent (Keja and Kits 1994). We found that the Q-type current lost its inactivation and acquired a slow time constant of activation in the presence of dopamine. This can explain the slow time constant of the macroscopic Ca²⁺ current (Keja and Kits 1994; Keja et al. 1992). Most likely, it occurs through a direct G interaction with the ₁ subunit of the Q-type channel. The change in the kinetics of the N-type channel might also result from a direct G protein interaction, although no kinetic slowing of activation was observed. Alternatively, it might be modulated by dephosphorylation.

Serotonin has similar effects as dopamine on the different HVA channel types in melanotropes by activating 5-HT_1A and 5-HT_2B receptors (Ciranna et al. 1996). Like dopamine, it scales down the L-type channels and it slows down the activation kinetics of the Q-type current, voltage-dependently and G protein mediated. The P-type channels seem however unaffected by serotonin. Thus dopamine and serotonin may activate a similar, but not identical signaling cascade in the melanotrope. Whether dopamine acts as an agonist on serotonin receptors in these cells is unknown.

Modulation of the release machinery

Although the total Ca²⁺ current amplitude was reduced, the amount of exocytosis per depolarization was not affected by dopamine. As a result, the amount of exocytosis per million Ca²⁺ ions that entered the cell during the depolarization was increased. This does not result from a preestablished differential coupling of Ca²⁺ channels to exocytosis. We recently showed that each of the different Ca²⁺ channel types contributes equally to exocytosis in the absence of dopamine (Mansvelder and Kits 2000a). By dialyzing cells with solutions containing 3 µM free Ca²⁺ we found that the increased efficacy of Ca²⁺ ions to stimulate exocytosis results from a modulatory effect of dopamine on the release machinery downstream of the Ca²⁺ channels. This strengthens the conclusion that the increased efficacy does not result from dopamine favoring a Ca²⁺ channel that is better coupled to exocytosis. However, in pancreatic A-cells, adrenaline causes vesicles to dock preferentially near L-type Ca²⁺ channels, thereby changing the strength of the coupling between the L-type channel and exocytosis (Gromada et al. 1997). It is not known whether dopamine induces a redistribution of LDCVs in melanotropes.

There are several intracellular messengers that affect exocytosis at a level downstream of the Ca²⁺ channels in melanotropes (Lee 1996; Okano et al. 1993; Sikdar et al. 1998). Dialyzing cells with high concentrations cAMP (200 µM) increases secretion in response to elevated internal Ca²⁺ (Lee 1996; Sikdar et al. 1998). The effect is Ca²⁺ concentration-dependent and is most pronounced at 3 µM free Ca²⁺ (Lee 1996). The mechanism underlying the enhancement by cAMP is still unknown. One report suggested that high levels of cAMP stimulate the fusion of larger vesicles (Sikdar et al. 1998). In mouse pancreatic -cells cAMP potentiates secretion by direct modulation of the release machinery (Ammala et al. 1993b, 1994). Dopamine most likely caused a reduction in the intracellular cAMP concentration in our experiments. From the effects of cAMP on exocytosis one might expect a reduction in the release efficiency. However, this is not what we observed. Instead we observed that 1 µM dopamine enhances the early phase of exocytosis. Therefore it is unlikely that cAMP mediates the enhanced efficacy of Ca²⁺ to stimulate exocytosis.

The direct effect of dopamine on the release machinery might be mediated by G proteins. Stimulation of the dopamine D₂ receptor activates heterotrimeric G proteins. Dialysis of melanotropes with GTPgS (100 µM), thereby irreversibly stimulating GTP-binding proteins, stimulates large amounts of exocytosis (Okano et al. 1993). Recently, it was shown that specific activation of trimeric G proteins increases Ca²⁺-induced exocytosis in melanotropes (Kreft et al. 1999). In a lamprey neuromuscular synapse, the G subunit affects neurotransmitter release downstream of Ca²⁺ entry (Blackmer et al. 2001). Although the G subunit inhibited neurotransmitter release in this system, unlike the Ca²⁺-independent modulation of exocytosis by dopamine in melanotropes described here, it shows that G protein subunits interact with the release machinery in other systems as well.

Alternatively, the dopamine D₂ receptor might couple to a diacylglycerol-IP₃ second messenger pathway. Release of Ca²⁺ from IP₃-sensitive stores stimulates exocytosis in melanotropic cells (Lee 1996), just as in pancreatic -cells and pituitary gonadotropes (Ammala et al. 1993a; Tse et al. 1993). In gonadotropes, IP₃-sensitive Ca²⁺ stores are located close to the membrane and local release of Ca²⁺ from these stores stimulates release (Tse et al. 1997). Both the activation by dopamine of G proteins and IP₃-mediated local release of Ca²⁺ are good candidates for directly altering release properties of LDCVs in melanotropes.

Dopamine and the melanotropic cell output

Dopamine inhibits the melanotropic cell at different levels and on different time scales. At longer terms, dopamine depresses gene expression and affects posttranslational processing of its release products (Millington et al. 1987; Oyarce et al. 1996). The expression of the L-type Ca²⁺ channel is depressed by D₂ receptor stimulation in vitro (Chronwall et al. 1995), and in vivo the HVA Ca²⁺ channel expression decreases during development, as soon as the intermediate lobe gets innervated by dopaminergic fibers (Gomora et al. 1996). It will be interesting to learn how the stimulus-secretion coupling is affected by long-term dopamine exposure.

At short term, dopamine inhibits Ca²⁺ channels and opens a potassium conductance, thereby hyperpolarizing the membrane potential (Stack and Surprenant 1991; Williams et al. 1989). Lee (1996) suggested that the hyperpolarization is the most important mechanism to reduce the Ca²⁺ influx by shutting down voltage-dependent Ca²⁺ channels. Inhibition of the melanotrope by dopamine will most likely decrease the number of action potentials that are fired. However, at the same time dopamine enhances the efficacy of Ca²⁺ ions to stimulate exocytosis and the stimulus-secretion coupling for a given depolarization remains intact. Therefore the hormonal output of the melanotrope per action potential might be unaffected. Thus although the melanotrope fires less action potentials in the presence of dopamine, the amount of hormone release per action potential can still be the same.