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] |
In the present study the synaptic input to midbrain dopamine neurones and its modulation was studied in vitro. We have described and characterised an NMDA-mediated synaptic event that involves L-type Ca2+ channel activation, demonstrated that mGluRs and muscarinic receptors mediate presynaptic inhibition of glutamatergic and GABAergic synaptic input and provided additional support for the notion that nicotine activates these cells via postsynaptically located receptors. These findings thus allow for a more detailed analysis of the role of the glutamatergic and cholinergic afferent inputs in regulating the dopaminergic neuronal activity in the VTA and SNPC.
Glutamatergic afferent input to midbrain dopamine neurones seems to play a critical role for maintaining the spontaneous, burst firing activity of the dopamine neurones in vivo (Grenhoff et al. 1988, Svensson and Tung 1989, Charléty et al. 1991). In particular, NMDA receptors are involved in mediating this bursting activity (Chergui et al. 1993, 1994), which is functionally coupled to a markedly enhanced dopamine efflux in the terminal areas when compared to the efflux associated with regular firing (Gonon and Buda 1985, Gonon 1988, Bean and Roth 1991). The highly regular firing activity seen in vitro can be changed into burst activity if NMDA is bath applied together with apamine, a blocker of the KCa channels which reduces the postspike AHP (see Seutin et al. 1993). Moreover, repetitive stimulation of afferents in the slice preparation can induce burst firing of the dopamine neurones (Johnson 1998). Generally, activation of NMDA receptors seems to require a larger amount of glutamate release than AMPA/kainate receptors, since single stimulating pulses evoke synaptic potentials with only a small contribution of NMDA receptors (Mereu et al. 1991, Johnson and North 1992b), whereas high frequency stimulation is required to evoke synaptic potentials with a large contribution of NMDA receptors. The present study thus suggests that the burst activity seen in vivo results from a high level of activity of the glutamatergic afferents to the dopamine neurones. The requirement of higher amount of glutamate release for the activation of NMDA receptors is in one aspect surprising, because NMDA receptors have a higher affinity for glutamate than AMPA receptors (Hollmann and Heinemann 1994). Possible explanations might be: (1) that the NMDA receptors are located more distantly from the release site than the AMPA receptors, (2) that the voltage dependence of the NMDA receptors requires a stronger depolarisation for its opening, and/or (3) that the AMPA receptors desensitise more rapidly than NMDA receptors (Trussel et al. 1993).
The involvement of L-type Ca2+ channels in mediating the slow NMDA EPSC is indeed interesting since these channels display a larger and longer activation than other Ca2+ channels. Therefore they may provide the additional Ca2+ influx into the glutamate terminal needed for the release of a sufficient amount of glutamate. Such a specific involvement in NMDA-mediated synaptic transmission may in turn allow the opportunity to selectively affect the NMDA-related activity, i.e. burst firing, of dopamine neurones by a specific pharmacological manipulation of L-type channels.
The glutamatergic afferents acting on NMDA receptors are most likely of critical importance for the normal functioning of dopaminergic systems. Accordingly, dysfunction in pathological states thought to be related to midbrain dopamine systems may thus depend, on altered level of glutamatergic afferent activity or dysfunctional L-type channels, and not only to intrinsic dysfunction of the dopamine neurones themselves.
The role of mGluRs in VTA and SNPC seems quite complex because of the large variety of functions that have been described. Postsynaptic receptors have been demonstrated by the sensitivity of dopaminergic neurones to the mGluR agonist t-ACPD (Mercuri et al. 1993) and by the finding of slow mGluR-mediated postsynaptic events evoked by repetitive focal stimulation of the slice that can be either excitatory (Shen and Johnson 1997) or inhibitory (Fiorillo and Williams 1998). In addition, we here provide evidence that also presynaptic mGluRs are present on both glutamatergic and GABAergic terminals within the VTA and SNPC. These results were recently confirmed (Wigmore and Lacey 1998).
All postsynaptic effects of mGluR activation on midbrain dopamine neurones seem to be mediated by group I mGluRs (Shen and Johnson 1997, Fiorillo and Wiliams 1998). Interestingly, these phospholipase C-coupled receptors seem to be able to mediate two opposite actions on the dopamine neurones. The excitatory action, that was first described (Mercuri et al. 1993, Shen and Johnson 1997), is preceded by an inhibition if the agonist application or stimulating pulse is brief enough (Fiorillo and Williams 1998). Fiorillo and Williams (1998) have recently proposed that the pathway mediating this inhibition desensitises rapidly, while the excitatory effect persists. The desensitisation is most likely occurring downstream from the receptor since the excitation is not affected. A similar dual pathway activation by group I receptors has also been described at the presynaptic level (Herrero et al. 1998), where a rapidly desensitising enhancement of glutamate is followed by an inhibition of the release. The dual postsynaptic action by the group I mGluRs at the dopaminergic neurones is yet another example of the importance of the pattern of activity of the afferents for the final effect on the postsynaptic cells.
The presynaptic inhibition of both glutamatergic and GABAergic synaptic inputs demonstrated in the present study involves group II and III mGluRs but also, to a lesser degree, group I mGluRs. Presynaptic inhibition of glutamate and GABA release has previously been demonstrated in several regions of the brain such as cortex, hippocampus, striatum and thalamus (Baskys and Malenka 1991, Calabresi et al. 1993, Burke and Hablitz 1994, Salt and Eaton 1995). The presynaptic mGluRs located on the glutamatergic terminals projecting to the dopamine neurones might well function as autoreceptors as has already been demonstrated in the hippocampus (Scanziani et al. 1997). The autoreceptor function at glutamatergic afferents to dopamine neurones could be of importance in limiting glutamate-induced cytotoxicity resulting from excess glutamate release. The presynaptic inhibition of the GABA afferents to dopaminergic neurones would, in principle, have an opposite effect, i.e. enhancing excitation of the dopamine neurones. A functional role of this effect might be to augment the glutamatergic input when GABAergic input is activated at the same time. An overall function of the mGluRs as presynaptic inhibitory receptors on glutamatergic and GABAergic synaptic input may, consequently, be to allow for maintenance of homeostasis within the dopaminergic system.
Glutamatergic afferents to the midbrain dopamine neurones arise from several brain regions (prefrontal cortex, subthalamic nucleus, PPTg and LDTg), whereas the GABAergic input originates in the striatum and pallidum (Fonnum et al. 1978, Bolam and Smith 1990, Smith and Bolam 1990), and from intrinsic GABAergic interneurones of the midbrain (Di Chiara et al. 1979, Grofova et al. 1982, Stanford and Lacey 1996). Since it is not possible to determine the origin of the afferents that are activated by the stimulating electrodes in the slice preparation, it is likely that afferents from several areas are stimulated. The different afferents might contain different mGluR subtypes depending on their site of origin. Therefore it may, in principle, be possible to use specific pharmacological tools to selectively modulate some, but not other afferent inputs to the dopamine neurones.
It was previously known that both muscarinic and nicotinic receptors mediate a postsynaptic action on the dopamine neurones, but presynaptic modulation by cholinergic receptors of the synaptic input to the dopamine neurones had not previously been investigated. We show in this study that muscarinic, but not nicotinic receptors, presynaptically modulate the glutamatergic and GABAergic input. ACh acting on muscarinic or nicotinic postsynaptic receptors can thus not only exert a direct excitatory action of the dopamine neurones but may also via presynaptic muscarinic receptors modulate the synaptic input from other brain regions.
The muscarinic effect on dopamine neurones of the midbrain is, like that of mGluRs, dual with a postsynaptic excitation and a presynaptic inhibition. The postsynaptic action is sensitive to pirenzepine (Lacey et al. 1990) and is likely to be mediated through M1 receptors. However, studies using in situ hybridisation techniques have demonstrated that SNPC exclusively expresses mRNA for m5 receptors (Vilaro et al. 1990, Weiner et al. 1990). m5 receptors display an intermediate affinity for pirenzepine (Vilaro et al. 1990) and are, as the M1 receptors, coupled to the IP3 cascade. The M1-like postsynaptic action on the dopamine neurones might thus rather be mediated by activation of m5 receptors. Muscarinic agonists have previously been demonstrated to enhance NMDA-mediated synaptic transmission and NMDA-mediated postsynaptic inward currents in the auditory cortex (Aramakis et al. 1997, 1999). These authors demonstrated that the enhancement is mediated through M1-like receptors coupling to G-protein-mediated activation of the IP3 system. This mechanism might also be present in the midbrain dopamine neurones. Muscarine could, through activation of postsynaptic M1 or M5 receptors, thus not only increase the firing frequency of the dopamine neurones but also enhance ongoing NMDA receptor-mediated burst firing.
The action of muscarine on presynaptic receptors on glutamate and GABA afferents to the midbrain dopamine neurones described in the present work shows a pharmacological profile similar to M3 receptors. A functional role of the muscarinic presynaptic inhibition might be to prevent excess excitation when both glutamatergic and cholinergic afferents are active, whereas it could serve to prevent excessive inhibition of the synaptic input via inhibition of the GABAergic input.
The nicotinic effect on midbrain dopamine neurones in vitro can in all probability not be attributed to an enhanced glutamate release by presynaptic nicotine receptors, as was suggested by previous in vivo and in vitro experiments from our laboratory (Schilström et al. 1998a, Results, Paper V). Presynaptic facilitation of glutamate release was first described in the prefrontal cortex by Vidal and Changeux (1993), and was subsequently demonstrated in a number of other brain regions (McGehee et al. 1995, Gray et al. 1996). It was even suggested that presynaptic modulation of transmitter release may represent the predominant role of nicotinic receptors in the brain, especially considering the lack of evidence for fast synaptic transmission mediated through nicotinic receptors (McGehee et al. 1995). However, even if it now seems clear that nicotine can act on both pre- and postsynaptic receptors in the brain, its effect on VTA and SNPC neurones seems essentially to be postsynaptic.
Both nicotinic and muscarinic receptors in the VTA appears to contribute to rewarding brain stimulation of the lateral hypothalamus, since both nicotinic and muscarinic antagonists applied locally in the VTA reduce the rewarding effect of the brain stimulation (Yeomans and Batista 1997). However, the muscarinic receptors seem to contribute more than the nicotinic receptors. Yeomans and Batista (1997) proposed that rewarding stimulation of the lateral hypothalamus is mediated through an activation of the cholinergic neurones of the PPTg and the LDTg, that in turn activate the dopaminergic neurones of the midbrain. In contrast, nicotine self-administration does not seem to involve muscarinic receptors in the VTA or the cholinergic neurones in PPTg and LDTg (Corrigall et al. 1994).
Several small clinical trials with Ca2+ antagonists, such as verapamil, diltiazem and dihydropyridines (all blockers of the L-type channels), in the treatment of psychiatric diseases have been published (see for review Hollister and Garza-Trevino 1996). The predominant diagnoses have included bipolar disorder, unipolar depression and schizophrenia. The effect on bipolar patients is relatively well documented and a therapeutic effect has frequently been observed in the manic phase, sometimes reversing it to a depression requiring treatment with antidepressants. Interestingly, in other studies the patients stabilised in their mood fluctuations between mania and depression. Unipolar depression has sometimes been found to be aggravated by Ca2+ antagonists, but in other cases to be improved. The effect of Ca2+ antagonists in schizophrenia appears equivocal, but some reports describing an effect similar to that of conventional antipsychotics have actually been published. Overall the potential usefulness of Ca2+ antagonists on psychiatric indications remains, as yet, to be conclusively established.
The putative role of the dopaminergic systems in the pathophysiology of e.g. bipolar disorder makes, in principle, the L-type-mediated NMDA transmission to the dopaminergic neurones described in the present study a tentative site of action for the Ca2+ antagonists, in particular as regards the mood stabilising action.
The clinical usefulness of mGluR agonists for the treatment of psychosis was recently proposed (see Moghaddam and Adams 1998). These authors showed that the phencyclidine (PCP)-induced glutamate release in the prefrontal cortex as well as the PCP-induced locomotor hyperactivity and stereotyped behaviour could be prevented by pre-treatment with an agonist to group II mGluRs. However, the PCP-induced increase in dopamine output in the nucleus accumbens and the prefrontal cortex was not affected. Needless to say, none of these effects of the mGluR agonist show a high predictive value for antipsychotic activity per se. The effect of the group II mGluR agonist might be mediated via presynaptic inhibition on the glutamatergic terminals in the prefrontal cortex, although the group II mGluR-mediated presynaptic inhibition of the glutamatergic input to the dopamine neurones demonstrated in Paper III did not seem to be reflected in any alteration of the basal extracellular concentrations of dopamine in terminal areas. Nevertheless, specific agonists or antagonists to mGluRs might prove useful in modulating the phasic responsivity of the dopamine neurones, since we here show that the presynatically located mGluRs are predominantly of group II or III (Paper III), whereas the postsynaptically located mGluRs have been shown to be of group I (Mercuri et al. 1993, Shen and Johnson 1997, Fiorillo and Williams 1998).
The psychotomimetic effect of antimuscarinics (see for review Yeomans 1995) as well as clinical results obtained already in the 1970’s (Davies et al. 1978), showing a reversal of the manic phase of bipolar disorder by the AChesterase inhibitor, physostigmine, suggests a possibility for muscarinic pharmacotherapy of psychiatric disorders involving potentially hyperactive or hyperreactive dopaminergic systems. For example, Yeomans (1995) suggested that blockade of muscarinic autoreceptors of the m2 subtype on the cholinergic neurones in the PPTg and LDTg leads to an increased activity of the cholinergic input to the midbrain dopamine neurones, thereby inducing psychosis. The demonstrated M3 receptor-mediated presynaptic inhibition of glutamatergic input to dopamine neurones could also help to explain an activation of the dopamine neurones by antimuscarinic agents. Indeed, M3 agonists are currently being evaluated as potential antipsychotic drugs (Shannon et al. 1998). One site of action might thus be the presynaptic receptors on glutamatergic terminals to the midbrain dopamine neurones described in this study.
Nicotine itself is currently being used clinically as a moderately effective treatment in smoking cessation, although more specific ligands to certain nicotinic receptor subtypes might provide interesting alternatives. The nicotinic receptor subtypes mediating nicotinic excitation of the dopamine neurones in VTA seems to be a4b2 and possibly a7 (Pidiplichko et al. 1997, Sorenson et al. 1998, Pichiotto et al. 1998, Schilström et al. 1998b) and these may allow for pharmaceutical development of novel therapies in this regard as well as in other dopamine-related neuropsychiatric disorders, such as Parkinson’s disease.
Generally the midbrain dopaminergic neurones, that have been implicated in reward-related behaviour, cognitive function and motor performances, thus have several afferent inputs, which control their activity in vivo. The incoming activity patterns of the different afferent inputs, which are critical for the postsynaptic response in the dopamine neurones, can thus be balanced against each other through presynaptic modulation via several G-protein-coupled receptors. This fine regulation of the afferent activity may in turn enable the dopamine neurones to adapt their activity to incoming signals from several brain regions (Hoffman et al. 1995). Thus, the dopaminergic neurones seem to play an integrative role in optimising the shaping of behaviour and motor acts in relation to the demands of the environment. By inference, distortions of this presynaptic neuromodulatory machinery might contribute to impaired capacity of the individual to adapt to the environment e.g. in psychiatric disorders or drug-dependent states, or to dysfunctional motor control and neurodegeneration in neurological disorders such as Parkinson’s disease.
|
|
|
|
|
|
[Cell Biology & Cytology] |
[Neuroscience] |
