Poster
# 95
Poster |
|
6th Internet World Congress for Biomedical Sciences |
Nicola B. Mercuri(1), Pernilla Grillner(2)
(1)IRCCS Santa Lucia, University of Rome Tor Vergata - Rome. Italy
(2)Dept Physiology and Pharmacology. Karolinska Institutet - Stockholm. Sweden
|
 |
|
|
[Cell Biology & Cytology] |
[Neuroscience] |
The glutamatergic synaptic input to midbrain dopamine neurones previously described is predominantly mediated through AMPA/kainate receptors and only to a smaller extent through NMDA receptors. However, the dopamine neurones are sensitive to NMDA application (Mercuri et al. 1992, Overton and Clark 1992, Seutin et al. 1994, Christofferson and Meltzer 1995) and NMDA receptor activation is considered to be responsible for the burst firing activity seen in vivo (Grenhoff et al. 1988, Svensson and Tung 1989, Charléty et al. 1991, Chergui et al. 1993).
Changes in the pattern of electrical stimulation from a single pulse that evokes the fast AMPA/kainate-mediated EPSP to repetitive pulses evoked a synaptic potential mediated through NMDA receptor activation. In fact, local stimulation with a short train of repetitive stimulating pulses (40 ms, 100-400 Hz) elicited a slow EPSP that was not affected by the application of the AMPA/kainate receptor antagonist CNQX, and enhanced by the application of the GABAA antagonist, picrotoxin (PTX), and the GABAB antagonist saclofen. This slow EPSP was dose-dependently inhibited by the application of the NMDA receptor antagonists AP5 (10-100 µM) and ketamine (30 µM), and was enhanced by Mg2+ removal from the extracellular solution, demonstrating that the slow potential is indeed mediated by activation of NMDA receptors. The application of the Na+ channel blocker tetrodotoxin (TTX) completely abolished the potential, demonstrating that it is synaptically mediated and not due to an unspecific direct stimulation of the dopamine neurones. (Mercuri et al., 1996)
The characteristics of the NMDA-mediated synaptic transmission were studied using the whole-cell patch clamp technique. We investigated the role of voltage-dependent Ca2+ channels of the L-type in mediating the slow EPSC. The L-type Ca2+ channel blockers nifedipine (1-100 µM), nimodipide (1-100 µM) and isradipine (0.03-100 µM) were able to depress the amplitude of the slow EPSC by 30-40 %, when evoked by a stimulating train with a duration of 80 ms or more. Moreover, the L-type channel agonist Bay-K 88644 enhanced the amplitude of the slow EPSC by 20 %. Nifedipine did not affect neither the fast EPSC nor the fast and slow IPSC. The change of the postsynaptic membrane holding potential from –100 to +30 mV did not affect the degree of depression of the slow EPSC by nifedipine. Furthermore, the postsynaptic response to glutamate was not affected by the presence of nifedipine, indicating that the effect of the DHPs is presynaptic (Bonci et al., 1998).
These two papers demonstrate that NMDA receptors contribute substantially to excitatory synaptic transmission to the midbrain dopamine neurones during high frequency stimulation of the glutamatergic afferents, that possibly correspond to bursting activity of the input neurones in vivo. This finding is in accordance with studies showing an important role of NMDA receptors in the bursting activity of the dopamine neurones in vivo (Grenhoff et al. 1988, Svensson and Tung 1989, Charléty et al. 1991, Chergui et al. 1993). The involvement of L-type Ca2+ channels in the slow EPSP/C is consistent with high frequency stimulation or a bursting activity, since the biophysical properties of these channels require high threshold activation and conduct large currents with a long activation time. This can provide a larger Ca2+ influx leading to a larger release of glutamate from the nerve terminal. This higher concentration of glutamate in the synaptic cleft might be required for the activation of NMDA receptors while the AMPA receptors that desensitise rapidly (Trussel et al. 1993) do not contribute to the slow EPSP. L-type channels have previously not been considered to play a major role in synaptic transmission, but this study shows that under certain circumstances the L-type channels contribute to the release of glutamate. Interestingly, they are not involved in the slow GABAB-mediated IPSP that is evoked by the same stimulating parameters. This suggests that L-type channels might be present exclusively on the excitatory glutamatergic input to the dopamine neurones.
In conclusion, we describe in paper I an NMDA-mediated slow EPSP/C that is evoked by a repetitive stimulating pulse, and in paper II, that it involves activation of presynaptic L-type Ca2+ channels, in contrast to the fast EPSC and the fast and slow IPSCs recorded in midbrain dopamine neurones which mainly involve N and P/Q-type Ca2+ channels.
Activation of metabotropic glutamate receptors (mGluRs) has previously been shown to induce an inward current in midbrain dopamine neurones (Mercuri et al. 1993). To test if mGluRs also mediate presynaptic modulation of the synaptic input to dopamine neurones the effects of specific agonists and antagonists to the mGluRs of group I, II and III were tested on the fast EPSC as well as on the fast and slow IPSC.
The group III agonists L-AP4 and L-SOP depressed all three types of synaptic potentials recorded. The specific group III antagonist MAP4 blocked this depression. The fast EPSC was depressed to a larger degree than the IPSCs. The group II agonist L-CCGI also depressed all the three types of synaptic events to a similar degree as the group III agonists. The group I agonist 3,5-DHPG, on the other hand, reduced the GABAA- and GABAB-mediated synaptic currents to a larger extent than the excitatory synaptic currents. In addition, L-CCGI and 3,5-DHPG both induced an inward current. The effect of L-CCGI and 3,5-DHPG on the synaptic events were partially blocked by the broad spectrum antagonist MCPG, but not by MAP4. To determine if the effects were presynaptic, a paired-pulse protocol was used (DelCastillo and Katz 1954, Rausche et al. 1988, Zucker 1989, Mennerick and Zorumski 1995). Two stimuli of identical intensity were applied with 50 ms interval. The EPSC showed
paired-pulse depression while the IPSC displayed paired-pulse facilitation. During the application of agonists of the different mGluRs the ratio between the second and first pulse (p2/p1) was increased for both the fast EPSC and IPSC, indicating that the depression of these synaptic events is mediated by presynaptic mechanisms (Bonci et al., 1997).
The present work demonstrates, that in addition to the direct postsynaptic excitatory effect on the dopamine neurones (Mercuri et al. 1993, Shen and Johnson 1997), mGluRs also mediate presynaptic inhibition of excitatory and inhibitory inputs to these cells. mGluR-mediated presynaptic inhibition has previously been demonstrated in several regions of the CNS including the cortex, hippocampus, striatum, and thalamus (Baskys and Malenka 1991, Calabresi et al. 1993, Burke and Hablitz 1994, Stefani et al. 1994, Lovinger and McCool 1995, Salt and Eaton 1995). The group II and III mGluRs are most efficient in depressing synaptic events and glutamatergic synaptic events seems more sensitive than the GABA-mediated events. The presynaptic locus of action is indicated by the paired-pulse experiments. The functional role of the mGluRs in VTA and SNPC could be to act as autoreceptors on the glutamatergic input to the dopamine neurones to prevent excessive release of glutamate, but still maintaining some excitability of the dopamine cells by directly stimulating the postsynaptic receptors and, at the same time, presynaptically inhibiting the release of GABA.
Muscarine has previously been shown to excite midbrain dopamine neurones in vitro through activation of M1-like receptors (Lacey et al. 1990). Since muscarinic receptors are also known to mediate presynaptic modulation of transmitter release in several regions of the mammalian nervous system (Sugita et al. 1991, Hasselmo and Bower 1992, Hsu et al. 1995, Sim and Griffith 1996) we were interested in studying the possible existence of such a modulation of synaptic input to the midbrain dopamine neurones.
We investigated the effects of muscarine and carbachol on the evoked fast EPSP recorded in midbrain dopamine neurones. The EPSP was depressed by bath application of carbachol (0.1–30 µM) in a dose-dependent manner and by muscarine (30 µM). To elucidate the receptor subtype involved four different antagonists were used. Only the M3/M1 antagonist 4-DAMP (1 µM) was able to completely block the effect of carbachol and muscarine on the EPSP. The M1 specific antagonist pirenzepine (1 µM) could not significantly prevent the depression of the EPSP. To demonstrate that the suppressing effect on the EPSP was not due to a change in sensitivity of the glutamate receptors at the postsynaptic membrane, the depolarising response to a short application of glutamate (1 mM; in the presence of 0.5 µM TTX) was compared during control conditions and in the presence of muscarine (30 µM). The depolarisation was of the same amplitude both in control and in the presence of muscarine (30 µM). A paired-pulse protocol (described above) was also used to further confirm the presynaptic site of action of muscarine. A paired-pulse stimulus caused a depression of the second EPSP and application of muscarine depressed the first EPSP to a larger extent than the second, thereby increasing significantly the ratio (p2/p1) (Grillner et al., 1999a). These results thus indicate that M3 receptor activation mediates depression of glutamatergic transmission through a presynaptic mechanism of action. Moreover, the input resistance of the dopamine neurones was not changed by the application of muscarine excluding the possibility that electrical shunting could account for the depression of the EPSP. By the use of physostigmine, an inhibitor of the ACh degrading enzyme ACh-esterase, we could show that ACh, endogenously released from the slice, has the same effect as muscarine. The depression of the EPSP by physostigmine was also blocked by 4-DAMP. To investigate the second messenger pathways involved we used N-ethylmaleimide (NEM) a blocker of the pertussis-toxin-sensitive Gi/o-protein. The preincubation with NEM did not affect the muscarine-induced depression of the EPSP. The presynaptic inhibition was not mediated through Ca2+ channels of N- or P/Q-type since specific antagonists to these channels did not affect the presynaptic inhibition by muscarine of the excitatory potential. Also the involvement of K+ channels were studied by blocking these channels with 4-AP or barium. The blockade of K+ channels did not change the response to muscarine.
We also investigated the muscarinic modulation of GABAA-mediated synaptic transmission to the dopamine neurones (Grillner et al., 1999b). Firstly, we studied the effect of bath applied muscarine on the evoked fast GABAA-mediated IPSP recorded in midbrain dopamine neurones. Muscarine at a concentration of 30 µM depressed the IPSP amplitude. To characterise the receptor subtype involved, the M3/M1 antagonist 4-DAMP (100 nM) and the M1 antagonist pirenzepine (1 µM) were used. The antagonists were applied 15 min prior to the application of muscarine to enable the tissue to be well incubated. Only 4-DAMP could completely block the effect of muscarine, while pirenzepine failed to significantly affect the muscarinic depression of the IPSP. To exclude the possibility that the depressant effect of muscarine was due to a change in sensitivity of the postsynaptic GABAA receptors the hyperpolarisation induced by bath applied GABA (200 µM; in the presence of TTX (0.5 µM), CNQX (10 µM), AP5 (100 µM) and saclofen (200 µM)) was studied under control conditions and in the presence of muscarine (30 µM). The amplitude of the hyperpolarisation was the same in control and in the presence of muscarine. The effect of muscarine on mIPSCs was analysed in the presence of pirenzepine to block the postsynaptic receptors (Lacey et al. 1990). Under these conditions muscarine (30 µM) significantly and reversibly decreased the frequency of mIPSCs without affecting their amplitude.
These results demonstrate that M3 muscarinic receptors depresses the evoked fast glutamate-mediated EPSP and the fast GABAA-mediated IPSP to the midbrain dopaminergic neurones. The M3 receptor involvement is suggested by the counteraction of muscarinic receptor-mediated depression of synaptic transmission by 4-DAMP but not by pirenzepine. This is consistent with autoradiography studies showing a relative enrichment of M3 receptors in relation to M1 and M2 receptors in the substantia nigra (Zubieta and Frey 1993). Furthermore, we demonstrate that the endogenously released transmitter can mediate this presynaptic inhibition. The cerebral cortex and the subthalamic nucleus that both send excitatory afferents to SNPC and VTA have been reported to contain mRNA for m3 receptors (Weiner et al. 1990). However, the presence of m3 receptor expression has not been demonstrated in any of the areas sending GABAergic input to the VTA and SNPC (Weiner et al. 1990). The previously reported postsynaptic action of muscarine is confirmed in these studies since muscarine application gives rise to a depolarisation or an inward current that is prevented by pirenzepine. The role of muscarinic receptors in the SNPC and VTA is complex since activation of muscarinic receptors leads to a direct postsynaptic excitatory action mediated through M1 receptors as well as a presynaptic inhibition of both the GABAergic and the glutamatergic inputs to the dopamine neurones. The functional significance of this dual action on the dopamine neurones might be to enhance excitation of the dopamine neurones when glutamatergic afferents are inactivated, but to prevent excessive excitation of the dopamine neurones when both cholinergic and glutamatergic afferents are active simultaneously.
Nicotine is known to excite dopamine neurones in vitro by a direct action on nicotinic receptors on the cell membrane (Calabresi et al. 1989, Pidiplichko et al. 1997, Picciotto et al. 1998). However, in several brain regions nicotine has been shown to facilitate synaptic release of glutamate and GABA through activation of presynaptic nicotinic receptors (Léna et al. 1993, Vidal and Changeux, 1993, McGehee et al. 1995, Gray et al. 1996). In this study, we investigated a possible additional presynaptic action of nicotinic receptors on glutamatergic terminals besides their direct effect on the midbrain dopamine neurones.
Thus, we characterised the effect of nicotine on the dopamine neurones (Grillner and Svensson 1999). Nicotine (10 µM) depolarised the membrane potential by 5 mV and reduced the postspike afterhyperpolarisation (AHP) leading to an increase in firing frequency. The effects of nicotine on the membrane potential, the AHP and the firing frequency were blocked by the nicotinic antagonists dihydro-b-erythroidine (DHBE; 50 µM) and mecamylamine (10 µM), but not by methyllycaconitine (MLA; 10 nM). To investigate whether part of the excitatory role of nicotine is mediated through ionotropic glutamate receptor activation, antagonists to NMDA and AMPA/kainate receptors were used to find out whether the presence of these antagonists would affect the nicotine-induced excitation. Both the NMDA receptor antagonist AP5 (90 µM) and the AMPA/kainate receptor antagonist CNQX (10 µM) reduced the effect of nicotine on the firing frequency. These results indicate that the nicotine-induced excitation of the dopamine neurones involves activation of ionotropic glutamate receptors.
To further analyse if this nicotine-mediated excitation involves activation of presynaptic receptors, spontaneous synaptic currents were studied using whole-cell patch clamp recordings (Grillner et al., 1999)). sEPSC were not frequent in these neurones; however, their occurrence increased by application of 4-AP (20-100 µM). They were recorded in the presence of PTX (100 µM), in order to abolish GABAA-mediated responses. Bath or local pressure application of nicotine did not affect neither the frequency nor the amplitude of the sEPSCs. The effect of nicotine was also studied on GABAergic mIPSCs. No effect of nicotine was, however, seen on the frequency or on the amplitude of the mIPSCs neither by local pressure application nor by bath application of nicotine.
Our results thus demonstrate that nicotine excites dopamine neurones through a specific activation of nicotinic receptors. The effect partly involves ionotropic glutamate receptor activation. We found, however, no evidence for activation of presynaptic nicotinic receptors on glutamate or GABA terminals onto the midbrain dopamine neurones. Our results thus confirm earlier studies (Calabresi et al. 1989, Pidiplichko et al. 1997, Picciotto et al. 1998) which suggest that nicotine acts primarily through nicotinic receptors located at the somato-dendritic membrane of the dopamine. The receptor subtype that mediates the nicotinic effect might be a4b2 or a7 since nicotine-induced currents mediated through both subtypes have been demonstrated in midbrain dopamine neurones (Pidiplichko et al. 1997, Picciotto et al. 1998). However, our results show that nicotine induced excitation of the dopamine neurones in vitro does not require activation of a7 receptors, since the relatively specific antagonist MLA did not prevent the increase in firing frequency. The reduced excitatory effect of nicotine in the presence of glutamate antagonists may be explained by a general reduction of the excitability of the dopamine neurones or by the possible release of glutamate from the dopamine neurones themselves (Sulzer et al. 1998).
|
|
|
|
|
|
[Cell Biology & Cytology] |
[Neuroscience] |
