|Home | About | Journals | Submit | Contact Us | Français|
Synaptic transmission mediated by G-protein coupled receptors (GPCR) is not generally thought to be point-to-point. To determine the extent over which dopamine signals in the midbrain, the present study examined the concentration and time course of dopamine that underlies a D2-receptor inhibitory post-synaptic current (D2-IPSC) in the ventral tegmental area (VTA). Extracellular dopamine was measured electrochemically while simultaneously recording D2-IPSCs. The presence of dopamine was brief relative to the IPSC, suggesting that G-protein dependent potassium channel activation determined the IPSC time course. The activation kinetics of D2 receptor-dependent potassium current was studied using outside-out patch recordings with rapid application of dopamine. Dopamine applied at a minimum concentration of 10 µM for a maximum of 100 ms mimicked the IPSC. Higher concentrations applied for as little as 5 ms did not change the kinetics of the current. The results indicate that both the intrinsic kinetics of G-protein coupled receptor signaling and a rapidly rising high concentration of dopamine determine the time course of the IPSC. Thus dopamine transmission in the midbrain is more localized then previously proposed.
The time course of synaptic transmission is determined by the kinetics of receptor activation, the concentration of neurotransmitter and the clearance of transmitter from the synapse (Katz and Miledi, 1973; Lester et al., 1990; Barbour et al., 1994; Diamond and Jahr, 1995, 1997; Balakrishnan et al., 2009). Two distinct classes of receptors mediate transmission: ligand-gated ion channels and metabotropic G-protein coupled receptors (GPCRs). Fast synaptic currents mediated by ligand-gated ion channels result from a rapid rise and fall (~1 ms) of a high concentration (mM) of transmitter (Clements et al., 1992; Barbour et al., 1994; Diamond and Jahr, 1997; Overstreet et al., 2000; Mozrzymas et al., 2003; Beato, 2008). Less clear however is the concentration and time course of transmitter that mediates synaptic currents through the activation of GPCRs.
In numerous regions a variety of GPCRs mediate IPSCs through G-protein coupled potassium channels (GIRK/Kir3) channels (Luscher et al., 1997; Sodickson and Bean, 1998). The activation of these channels occurs with a time constant of several hundred milliseconds following a lag to onset of ~50 ms (Logothetis et al., 1987; Wickman et al., 1994; Huang et al., 1995; Kofuji et al., 1995; Sodickson and Bean, 1996; Ingram et al., 1997; Riven et al., 2006). In spite of the slow intrinsic signaling between GPCRs and the potassium conductance, the kinetics of IPSCs mediated by these receptors are highly sensitive to the block of metabolism or reuptake of transmitter (Isaacson et al., 1993; Beckstead et al., 2004). Thus, although GPCR-mediated IPSCs have been well described (Otis and Mody, 1992; Isaacson et al., 1993; Kulik et al., 2002), the mechanisms that determine the time course of transmission through this pathway is not completely understood.
Dendritic release of dopamine from neurons in the substantia nigra (SNc) and ventral tegmental area (VTA) has been studied over many years (Bjorklund and Lindvall, 1975; Geffen et al., 1976; Wilson et al., 1977; Cheramy et al., 1981; Nirenberg et al., 1996; Rice et al., 1997; Jaffe et al., 1998). Recent work indicates that dendritic dopamine release activates D2 receptors to produce an inhibitory post-synaptic current (IPSC) via activation of GIRK channels (Beckstead et al., 2004). The time course of the dopamine-dependent IPSC is, however, short relative to reports on the presence of extracellular dopamine (Chen and Rice, 2001; Chen et al., 2006). This raises the possibility that the prolonged, low concentrations of extracellular dopamine may not represent the time course and concentration of dopamine underlying transmission.
Using the combination of electrochemistry and electrophysiology, the present study compared the time course of the rise and fall of dopamine and the dopamine IPSC in the VTA. Recordings of the current induced by dopamine in outside-out patches from dopamine cells were used to define the intrinsic kinetics of the D2-receptor dependent activation of GIRK channels. The results indicate that the concentration of dopamine required to activate a D2-receptor dependent IPSC is surprisingly high and the rise and fall of dopamine in the extracellular space is faster than previously reported.
Horizontal slices (220 µm) of midbrain from male and female DBA/2J mice (4–8 week old, Jackson Laboratories, Bar Harbor, ME) were cut in ice-cold physiological saline solution containing (in mM) 126 NaCl, 2.5 KCl, 1.2 MgCl2, 2.4 CaCl2, 1.4 NaH2PO4, 25 NaHCO3, 11 d-glucose, and 0.005 MK-801 using a vibratome (Leica, Nussloch, Germany). Slices were incubated in warm (35°C) 95%O2/5%CO2 oxygenated saline containing MK-801 (10 µM) for at least 30 min and transferred to a chamber that was constantly perfused (1.5 ml/min) with oxygenated saline (35°C). Slices were visualized with an Olympus BX51WI (Olympus America, Inc., San Diego, CA) microscope. Dopamine cells were classified as within the VTA, were medial to the medial lemniscus (Ford et al., 2006).
An Axopatch 200A amplifier (Molecular Devices, Foster City, CA) was used to make whole cell recordings using 1.5–2.5 MΩ pipettes. Pipette internal solution contained (in mM) 115 K-methylsulphate, 20 NaCl, 1.5 MgCl2, 5 HEPES, 10 BAPTA, 2 ATP, 0.3 GTP; pH 7.3, 270 mOsm. Cells were voltage clamped at −60 mV. Dopamine neurons were identified by established criteria for the identification of mesolimbic dopamine neurons in the mouse (the presence of a D2-sensitive dopamine conductance, pacemaker firing (1–5 Hz) with spikes exhibiting AP widths of ≥ 1.2 ms (measured in cell-attached voltage clamp mode from the initial inward current to the peak of the outward current), a whole-cell input resistance < 400 MΩ, and a whole-cell input capacitance >25 pF (Ford et al., 2006; Lammel et al., 2008). Electrophysiological data was acquired using Axograph X (AxographX, Axograph Scientific). Series resistance was not compensated. Cells were discarded if the access resistance exceeded 10 MΩ.
Dopamine release from the VTA was elicited with an extracellular mono-polar saline-filled glass electrode (~5 MΩ) placed 100–200 µm from both the neuron being recorded and the tip of the exposed carbon electrode. As stated, either single pulses (0.6 ms) or trains (0.5 ms at 40 Hz) of stimuli (~30 µA) were used to drive dopamine release within the VTA. For the pharmacological isolation of the dopamine D2-receptor synaptic current (D2-IPSC), the external solution contained picrotoxin (100 µM), DNQX (10 µM), CGP 55845 (200 nM) and MK-801 (10 µM). BAPTA (10 mM) included in the internal solution was used to block mGluR signaling.
Glass encased carbon fiber electrodes with an exposed final length of 30–50 µm were prepared from 7 µm diameter carbon fibers (34–700, Goodfellow, PA). Prior to experimentation, the cut electrode tip was placed in isopropanol purified with activated carbon for 10 min. The tip of the carbon fiber electrodes was placed in the slice ≤ 75 µm below the neuron being recorded. Triangular waveforms (holding at −0.4V) at 10Hz (−0.4 to 1.0 V versus Ag/AgCl at 300 V/S) were used. Background subtracted cyclic voltammogram currents were obtained by subtracting 10 cyclic voltammograms (oxidation-reduction profiles) obtained before stimulation from voltammograms obtained after stimulation. After subtraction, two-dimensional voltammetric color plots were used to examine the data. To determine the time course of dopamine, the current at the peak oxidation (~600 mV) was plotted against time. After the experiment the electrode was calibrated using dopamine solutions of known concentration.
To confirm the chemical identity of the fast-scan cyclic voltammetry (FSCV) signal, voltammograms from exogenously applied dopamine were compared to voltammograms obtained from evoked release. Dopamine was applied by the iontophoretic application of dopamine hydrochloride or serotonin hydrochloride (data not shown) as a single pulse (25–100 nA, 5–25 ms, [DA] 1M) from thin walled iontophoretic electrodes (70≥100 MΩ). Iontophoretic electrodes were placed within 50 µm from the tip of the carbon fiber and a retention current of 1–5 nA was applied to prevent passive leakage.
Amperometry was also used to monitor the evoked dopamine overflow as the kinetics of the presence of extracellular dopamine are faster than when observed with FSCV. Electrodes were fabricated and placed in the slice as for FSCV. A constant potential of 0.3 V was applied. The current was plotted at 80 Hz, each data point was the average of 1000 points taken for 5ms. Ascorbate (600 µM) was added to the external Krebs buffer to allow for the catalytic regeneration of dopamine. Ascorbic acid (600 µM) prevents antioxidant depletion and stabilizes the catalysis of dopamine. Under these conditions, amperometry then detects individual analyte molecules multiple times. Thus, the resultant current is not an index of the absolute number of molecules, but its concentration in bulk solution and is a rate that is dependent upon the radius of the diffusion layer maintained by ascorbic acid and the concentration of the analyte in the extracellular fluid (outside the diffusion layer) (Venton et al., 2002).
To estimate the extracellular time course of dopamine resulting from a single stimulation quantitative analysis of individual records were not possible due to the low signal to noise ratio. However the group average (n=17) approximated the time course of a 250 ms application of dopamine (normalized integral single stimulus 293 mA S−1; normalized integral 250 ms ‘open tip’ potential 247 mA S−1).
Large nucleated macro patches were pulled from VTA dopamine cells using standard intracellular solution with standard patch pipettes (~1.3 MΩ). To obtain nucleated patches, gentle suction was applied over ~1 min until the nucleus was positioned directly at the tip of the patch pipette. While continually applying gentle suction, the patch pipette was slowly (~1 min) pulled away from the cell. Care was taken to ensure that the nucleus remained positioned at the tip of the patch pipette. Separation of the nucleated patch from the cell was indicated by a rapid drop in the input capacitance (from > 25 pF to ~5 pF). Often obtaining nucleated patches was unsuccessful, such that small inside-out patches were pulled. To ensure that nucleated patches were pulled, access resistance was continuously monitored such that an access resistance <5 MΩ was maintained. Nucleated patches were pulled out from the slice and placed in front of a theta-tube flow pipe that was situated in the bath ~500 µm above the slice. Once patches were pulled, solution was allowed to flow from the flow pipes. The bath volume was adjusted so that final ~1cm of the tip of the flow pipe was in contact with bath solution (35°c). The flow rate was set to ~50 µL per minute so that the temperature of the flow pipe solution reached equilibrium with that of the warmed bath solution. The flow pipe perfused two separate solutions: control containing in mM: 126 NaCl, 2.5 KCl, 1.2 MgCl2, 2.4 CaCl2, 1.4 NaH2PO4 0.6 ascorbate, and a similar solution with a known concentration of dopamine. The flow pipe was attached to a piezoelectric bimorph. Passing 4.4 V across the bimorph could rapidly change the location of the tip of the flow pipe such that the solution passing over the patch could be rapidly switched from control to dopamine with rapid kinetics. To determine the kinetics, after each experiment, the patch was blown from the end of the pipette and the open tip current recorded between control solutions and test solution containing an additional 20 mM NaCl was measured, with the average 10–90% rise time being 0.9 ± 0.2 ms.
Values listed are means ± SEM. Statistical significance was assessed using either paired, or unpaired t-tests where appropriate. Normality of distribution was tested before running parametric t-tests. Paired comparisons were done with a Wilcoxon matched paired test. In cases where the distribution was not normal, a non-parametric Mann-Whitney U test was used. A difference of p < 0.05 was considered significant. Statistical tests were performed with InStat (GraphPad Software, San Diego, CA).
Quinpirole and CGP 55845 were obtained from Tocris (Ellisville, MO). Cocaine hydrochloride was obtained from NIDA-NIH (Bethesda, MD). All other chemicals were obtained from Sigma-Aldrich.
To relate the time course of the D2-IPSC to the presence of dopamine, simultaneous whole cell voltage clamp and fast-scan cyclic voltammetry (FSCV) recordings were made in mouse brain slices containing the VTA. In the presence of antagonists to block GABAA, GABAB, AMPA, NMDA and mGluR receptors, a single stimulation via an extracellular stimulating electrode evoked a D2-IPSC (53.7 ± 9 pA, n =17). The IPSC was blocked by the D2-receptor antagonist sulpiride (data not shown), confirming that the current was mediated by D2 receptor activation (Beckstead et al., 2004; Ford et al 2007). The time to peak of the D2-IPSC (260 ± 10 ms) was slower than the concurrent rise in extracellular dopamine (61 ± 11 nM, 144 ± 15 ms, n = 17, p = 0.002, Fig 1a). The difference between the peak of the dopamine transient and the IPSC most likely represents the time required for metabotropic signaling (receptor/G-protein/GIRK channel activation).
In vivo, dopamine neurons fire action potentials in a pacemaker or burst pattern (Grace and Onn, 1989). A train of 5 stimuli evoked larger IPSCs (164 ± 18 pA, n = 17) and increased the concentration of extracellular dopamine (284 ± 52 nM, n =17; Fig. 1b). The cyclic voltammograms (oxidation-reduction profile) confirmed that dopamine was detected following stimulation (reduction peak of evoked release: −379 ± 7 mV, n=9; versus the reduction peak of exogenous dopamine: −358 ± 9 mV, n =3, Fig. 1d). The voltammogram did not match that produced by the exogenous application of serotonin (data not shown). Thus serotonin was most likely not released in the VTA with our stimulation protocol and does not contribute to the generation of the D2-IPSC. Like the single stimulation, the train of five stimuli evoked a rise in the concentration of extracellular dopamine ([DA]o) that peaked before the D2-IPSC (Fig. 1b). The time to peak for the [DA]o was 244 ± 26 ms (n=17), which was faster than that of the IPSC (374 ± 15 ms; n=17; p = 0.0001; measured from the first stimulus of the train, Fig. 1b).
Following stimulation the D2-IPSC decayed to baseline with a time course that preceded the extracellular FSCV dopamine transient. This was most likely due to an underestimate of the rate of clearance of extracellular dopamine by voltammetry (Bath et al., 2000). This temporal distortion does not occur with amperometry, where a constant positive potential is applied to the carbon fiber so that it does not promote adsorption of cationic species. Due to the decreased sensitivity of amperometry, a train of stimuli was used. The decay in [DA]o, measured with amperometry, was faster than the IPSC (n =12, Fig 1b), indicating that the duration of the [DA]o transient was less than the IPSC (duration to within 80% of baseline amperometry: 542 ± 39 ms, n = 12; vs. IPSC: 858 ± 48 ms, n = 17, p = 0.0001; Fig 1c).
Based on the size of the carbon fiber, dopamine was measured in the extracellular space from numerous release sites. The distance from each release site to the probe will affect the time required for dopamine to reach the probe. Despite the spatial and temporal limitations of measuring the bulk dopamine transient in the extracellular space, the rise and fall in dopamine occurred more rapidly than the IPSC. Thus a transient exposure of dopamine at postsynaptic D2 receptors mediates the IPSC.
The dopamine transient measured in the extracellular space with the carbon fiber most likely represents an underestimate of the concentration of dopamine at post-synaptic D2 receptors that mediate the IPSC. To determine the concentration and duration of dopamine at the receptor it was necessary to apply a known concentration of dopamine for a controlled period. Due to diffusion barriers this was not possible using brain slices. Therefore the current induced by dopamine was examined using outside-out nucleated macro-patches from VTA cells. Patches were pulled from the soma of dopamine neurons and positioned out from the slice directly in front of a flow pipe constructed from theta tubes. One barrel contained control saline, while the other contained known concentrations of dopamine. The flow pipe was attached to a piezoelectric bimorph for rapid exchange of the two solutions (0.9 ± 0.2 ms, 10–90% rise time, Fig 2a, inset).
The amplitude and the kinetics of the current activated by dopamine were dependent on concentration. Applications of 250 ms were examined initially as this was similar to the time course of extracellular dopamine measured with amperometry when evoked with a single stimulation (Fig 2a, inset). A low concentration of dopamine (100 nM) evoked a small current in three out of ten patches (6 ± 1 pA; n = 3 of 10, Fig 2a,c). The rate of activation and amplitude of these currents were slower than the IPSC (time to 10% of the peak patch: 282 ± 10 ms; n = 3; time to 10% of the peak IPSC: 87 ± 4 ms, n = 28; p < 0.0001; Fig 2 b,d). Increasing the concentration of dopamine ten-fold (1 µM) evoked larger currents that exhibited a faster rate of activation (amplitude 1 µM, 250 ms: 32 ± 2 pA, n = 7 out of 12 patches). However, the kinetics and amplitude of the current induced by DA (1 µM) were also slow relative to the IPSC (time to 10% peak 1 µM, 250 ms patch: 156 ± 15 ms; p = 0.002; Fig 2 a, b). Increasing the concentration of dopamine to 10 µM increased the amplitude and decreased the delay until the activation of the current (amplitude 52 ± 7 pA, time to 10% peak 101 ± 6 ms, n = 12, p < 0.05). Increasing the concentration to 100 µM evoked rapidly activating currents that were similar to the IPSC (amplitude 65 ± 11 pA, time to 10% peak 92 ± 6 ms, n =13, p > 0.4, Fig 2a,c).
The major effect of increasing the concentration of dopamine to 100 µM was that currents could be evoked with short applications. The high concentration of dopamine (100 µM) evoked a current that activated with a similar rate of activation regardless of the duration of application (Fig 2d). Decreasing the concentration of dopamine (10 µM) also evoked kinetically similar currents (Fig 2d). However at this concentration, a longer duration (25 ms) of dopamine application was required and currents could only be observed in half of the patches (Fig 2c,d). Regardless of the duration of dopamine applied, low concentrations of dopamine (100 nM - 1µM) failed to evoke currents that kinetically matched the rate of activation of the D2-IPSC. These results suggest that the minimum concentration present at post-synaptic D2 receptors during the peak of the IPSC is ≥10 µM. While lower concentrations of dopamine evoked outward currents, the rate of activation was slower than the IPSC. Thus although D2 receptors exhibit a high affinity for dopamine (Richfield et al., 1989; Stormann et al., 1990), a much higher concentration was required to produce the rapid activation of receptors to kinetically match that seen during transmission.
To confirm that the activation kinetics of the potassium conductance was similar at different sites in the neuron, dopamine/D2 receptor signaling in the soma and dendritic arbor was next examined. Dopamine was applied by iontophoresis at somatic and dendritic (>50 µm from the soma) locations. Dopamine neurons were filled with Alexa Fluor 594 (1 µM) and visualized using a 2-photon microscope (Fig 3a). The outward current evoked from dendrites was smaller in amplitude than the current evoked in the cell body (amplitude soma 142 ± 25 pA; n =5; amplitude dendrites 91 ± 20 pA; n = 5; p < 0.05; Fig 3a), but the kinetics of the current recorded in both sites were similar (time to peak soma: 575 ± 37 ms; n = 5; time to peak dendrites: 691 ± 67 ms p > 0.1; Fig 3a). Both the D2-IPSC and the current induced by iontophoretic application of dopamine are mediated by D2 receptors (Beckstead et al., 2004). The current produced by dopamine applied by iontophoresis to the soma was also blocked by sulpiride (200 nM; n = 6; 6 ± 4% of control; Fig 3b,c) and not blocked by alpha-2-adrenoceptor antagonist, idazoxan (1 µM; n = 6; 102 ± 5% of control; Fig 3b,c). Finally the dopamine-evoked current measured in nucleated patches was blocked by sulpiride (200 nM; n = 5; 7 ± 2% of control; Fig 3b,c) and not blocked by alpha-2-adrenoceptor antagonist, idazoxan (1 µM; n = 3; 100 ± 16% of control; Fig 3b,c). This confirmed that the dopamine current recorded in nucleated patches resulted from the activation of D2 –receptors that are kinetically similar to those receptors mediating the IPSC.
To further examine the time course of D2 receptor activation, dopamine (100 µM) was applied to nucleated macro-patches for various durations to determine the minimum time course of D2-receptor/GIRK signaling. Increasing the duration of application from 5 ms to 100 ms evoked larger amplitude currents (amplitude: 5 ms = 12 ± 2 pA; n = 9; 100 ms = 37 ± 5 pA, n = 14; p = 0.001) that peaked with a similar latencies (time to peak 5 ms = 215 ± 17 ms, n = 9; 100 ms = 239 ± 9 ms; n = 14; p > 0.8; Fig 4 a,b,c). Thus applications of dopamine (100 µM) for 5, 25, or 100 ms produced currents that, when scaled to the peak amplitude, were super imposable (Fig 4a). In order to mimic the time to peak of the D2-IPSC evoked from a single stimulation, the maximum time that dopamine (100 µM) could be applied was 100 ms (time to peak 100 ms = 239 ± 9 ms, n = 14, time to peak IPSC = 255 ± 38 ms n = 28, p > 0.5, Fig 4c). An application of dopamine for 250 ms evoked a current that took 373 ± 16 ms to peak (n = 13), significantly longer than the IPSC and the current induced by a 100 ms application (p = 0.0005). This places an upper limit on the duration of the peak concentration of dopamine that mediates the IPSC. As the duration and time to peak of the current did not vary between 5 ms and 100 ms, the result suggests that brief exposure to dopamine results in a fixed period of D2-receptor/G-protein/GIRK channel activation. It is this intrinsic period of receptor signaling, not the presence of dopamine that determines the time course of the peak of the IPSC.
If the IPSC was mediated by a dopamine transient that was less than 100 ms, a small increase in the presence of dopamine could increase the amplitude without altering the kinetics of the IPSC. The presence of dopamine was increased by blocking dopamine uptake with cocaine. In the presence of a low concentration of cocaine (100 nM, n= 9), the amplitude of the D2-IPSC was increased 163 ± 7% (n = 9, p = 0.01) yet the kinetics remained unchanged (p > 0.1) (Fig 4d). A higher concentration of cocaine (500 nM, n = 8) further increased the amplitude (283 ± 25 %, n = 8, p = 0.005), the time to peak (138 ± 4 %, n = 8, p = 0.001) and the decay of the D2-IPSC (Fig 4d). Taken together with the patch recordings, the results suggest that dopamine mediating the D2-IPSC was present at D2 receptors for no longer than 100 ms.
The results indicate that the time course of the IPSC was limited by the kinetics of receptor activation. To distinguish receptor activation from the activation of the GIRK conductance, the outward current evoked by the rapid application of the synthetic agonist quinpirole was examined. Quinpirole has a higher affinity for the D2 receptor than dopamine (Levant et al., 1992), and was therefore predicted to result in currents that were prolonged relative to the current induced by dopamine. A concentration of quinpirole (30 µM) was chosen that evoked a maximal outward current (100 ms: 41 ± 4 pA, n = 7). The application of quinpirole (30 µM, 100 ms) evoked a current that had a longer latency to activate (time to 10% of the peak: 119 ± 4 ms, n = 7) and peaked later (437 ± 8 ms, n = 7, p < 0.0001) than the current induced by dopamine (100 µM, 82.5 ± 5 ms, n = 15, p < 0.0001; Fig 5a, b). In addition, the time constant of the decay (τdecay) of the current induced by quinpirole (30 µM, 100 ms: 397 ± 40 ms, n = 7) was greater than dopamine (100 µM, 100 ms: 138 + 11 ms, n = 15, p < 0.0001) (Fig 5a, c). This suggests that both the on- (Kon) and off-rate (Koff) were slower than dopamine. Thus the time course of the IPSC is determined by the kinetics of dopamine-D2 receptor interaction.
To further examine the signaling processes underlying the dopamine transmission, the release of dopamine, the D2-IPSC and the current measured in excised patches at 35°C and 23°C were examined. Decreasing the temperature was expected to have relatively minor effects on the kinetics of free diffusion of dopamine compared to intracellular signaling. At 23°C the amplitude of the IPSC and current evoked in nucleated patches was decreased (D2-IPSC decreased to 34 ± 6%; n = 6, p < 0.01; nucleated patches decreased to 33 ± 9 %; n =9; p < 0.01). However, the amount of extracellular dopamine measured electrochemically increased by 170 ± 13% (n = 6, p < 0.05; Fig 6). Despite the differences on the effects of amplitude, the kinetics of all three measurements were slowed at 23°C (half maximal duration [DA]o: 201 ± 23%, n = 6; p < 0.05; D2-IPSCs: 226 ± 14%, n = 6, p<0.05; nucleated patch: 143 ± 16 %, n = 9; Fig 6). As the amplitude and kinetics of the IPSC and the direct application of dopamine were both reduced at room temperature the results indicate that the intracellular signaling cascade, not the extracellular diffusion of dopamine limited the time course of the IPSC.
The D2-IPSC had a longer duration than the current measured in outside-out patch recordings (Fig 7a). Simply increasing the duration of dopamine application to patches did not change the kinetics of the current in a way that matched the IPSC (Fig 4c). The presence of a late component of the IPSC suggests that receptor activation may be prolonged. Two (not mutually exclusive) possibilities exist that could explain the late component of the IPSC. Dopamine could diffuse some distance and recruit additional receptors, or the concentration of dopamine could decrease slowly over time resulting in a prolonged activation of local D2 receptors (Fig 7b). While these two possibilities could not be distinguished, it was possible to qualitatively replicate the D2-IPSC by adding the current produced by a brief high concentration of dopamine (100 ms, 100 µM; red trace) to the current produced by a prolonged low concentration (500 ms, 100 nM; gray trace, Fig 7c,d). Thus, although a low concentration of dopamine may contribute to the late phase of the IPSC and a high concentration is required to induce the rising phase of the IPSC.
This study compares the kinetics of D2-receptor mediated activation of potassium conductance induced by dopamine applied to outside-out patches with the D2-receptor dependent IPSC measured in brain slices. When dopamine (10–100 µM) was applied for 100 ms the current rose to a peak in ~250 ms and declined within ~500 ms to baseline. The duration of the current did not decrease further even when dopamine was applied for 5 ms. Thus, the current in the patch recordings induced by dopamine (10 −100 µM) applied for 100 ms or less, mimicked the early component of the dopamine-IPSC. The most conservative interpretation is that the IPSC is activated by a concentration of dopamine that is ≥10 µM that is present for ≤100 ms. The results challenge the hypothesis that dendritic release of dopamine activates receptors through a mechanism involving diffusion over a great distance.
Synaptic transmission mediated by GPCRs has been described in multiple sites in the nervous system. The majority of studies have focused on GABAB receptors. These receptors are most often located at extra-synaptic sites (Kulik et al., 2002; Lopez-Bendito et al., 2002). The high affinity of GABAB receptors allows for their activation at the lower concentrations of agonist found at these extra-synaptic sites (Otis and Mody, 1992; Isaacson et al., 1993; Sodickson and Bean, 1996; Scanziani, 2000). Pooling of GABA from multiple release sites in the extracellular space is then though to be important for the activation of GABAB IPSCs (Scanziani, 2000). This may account for the necessity to use multiple stimuli to evoke GABAB mediated synaptic potentials (Otis and Mody, 1992; Isaacson et al., 1993; Scanziani, 2000).
The ability to measure dopamine in the extracellular space with high temporal fidelity by electrochemical means has allowed estimation of the diffusion of dopamine in the extracellular space (Garris et al., 1994; Cragg and Rice, 2004; Staal et al., 2004; Rice and Cragg, 2008a). At both dendritic (SNc) and terminal (striatum) release sites pooling of dopamine in the extracellular space is thought to allow dopamine to signal over a large area (Garris et al., 1994; Cragg and Rice, 2004; Staal et al., 2004; Rice and Cragg, 2008b). Anatomical studies have found that D2 receptors are primarily located at extrasynaptic sites, lending support to the hypothesis that dopamine may signal at a distance from multiple sources (Sesack et al., 1994; Pickel et al., 2002). However, unlike reports in terminal regions, the density of release sites and distance from the release site to D2 receptors has not been anatomically quantified in the VTA.
Amperometric studies using midbrain dopamine neurons in slices and culture have measured events that had rapid kinetics resulting from the fusion of single vesicles (Jaffe et al., 1998; Staal et al., 2004). Other work has shown that the concentration of. The activation of receptors by dopamine is therefore dependent on the distance that dopamine diffuses and the affinity of post-synaptic receptors (Rice and Cragg, 2008b). The affinity of dopamine for the D2 receptor when measured at equilibrium is ~10 nM (Richfield et al., 1989; Stormann et al., 1990). However the transient rise in the concentration of transmitter during synaptic transmission does not reach steady-state so that equilibrium binding can not be used to predict receptor occupation (Clements et al., 1992; Diamond and Jahr, 1997; Beato, 2008). For example, EPSCs mediated by glutamate depend on a concentration ~1000 fold greater than the EC50 value(Patneau and Mayer, 1990; Clements et al., 1992).
Diffusion of dopamine away from the release site in the substantia nigra has been simulated with radial diffusion models. Those models predict that over a distance of 2 to 8 µm from the release site, the peak concentration of dopamine falls from 1 µM to 10 nM (Cragg and Rice, 2004; Rice and Cragg, 2008b). The present study indicates that at least 10 µM dopamine was required for the phasic activation of D2 receptors required to evoke an IPSC. When this concentration was used in combination with the diffusion model, the outcome suggests that synaptically released dopamine acted on receptors within ~ 1 µm of the release site.
The GPCR-dependent activation of potassium conductance has been examined at several brain regions and in each case the kinetics of the underlying currents are similar (Pan et al., 1989; Otis et al., 1993). Likewise, the activation of a variety of GPCRs using rapid application of saturating concentrations of agonist report a lag prior to the onset of current that peaks within several hundred milliseconds (Pan et al., 1989; Sodickson and Bean, 1996; Ingram et al., 1997). This most likely represents the time for G-protein activation of the potassium channel (Riven et al., 2006). These values match well with the kinetics of D2 receptor mediated GIRK currents. In the present study, lower concentrations of dopamine resulted in smaller currents that activated with slower kinetics. Similar results have been obtained in a study of the GABA-B receptor activation of potassium current where an EC50 concentration of baclofen activated currents at a rate 3 times slower than with a saturating concentration (Sodickson and Bean, 1996).
To distinguish agonist receptor binding from second messenger signaling, two experiments were performed. First, the temperature of the bath was changed. At low temperature more dopamine remained in the extracellular space for longer, whereas the D2-IPSC and the current induced by application of dopamine to outside-out patches were smaller and slower. The effect of decreased temperature was expected to have a greater effect on the kinetics of intracellular signaling than the free diffusion of dopamine. Thus the steps subsequent to agonist binding determined the rate of current activation. The second experiment examined the current induced by application of quinpirole to outside-out patches. Quinpirole evoked currents that had a slower rate of rise and decay than dopamine. This suggests that the higher affinity of quinpirole for the D2 receptor lead to prolonged receptor activation (Levant et al., 1992). The fact that the current induced by the two agonists was different suggests that the on and off rates of the two agonists regulate the macroscopic current. Thus the specific interaction between agonist and the D2 receptor as well as second messenger signaling are key determinants regulating the time course of the D2-IPSC.
We thank CE Jahr for advice and support with the rapid application of dopamine and AC Riegel, ST Hentges and CE Jahr for critical advice on the manuscript. Supported by NIH grants K99DA026417 (CPF), DA04523 (JTW) DA17155 (PEMP), DA24140 (PEMP). CPF was the Wyeth fellow of the Life Sciences Research Foundation and a fellow of the Alberta Heritage Foundation for Medical Research. JTW was a NARSAD Ritter Foundation Investigator