Based on precise neuron-to-neuron signaling, synaptic transmission is proposed to be the basic tenet of the neurone doctrine. Over the last decade, however, another mode for interneuronal communication in the central nervous system (CNS) has been advanced and has gained experimental support (Otellin & Arushanian, 1989; Sacharov, 1990; Agnati et al., 1995; Bach-y-Rita, 1993; Grace, 1991; Zigmond et al., 1990). New concept is based on diffusion of neurotransmitters and other biologically active compounds through the brain extracellular fluid to distant receptors. Agnati et al. suggested the term volume transmission to define this complementary means of intercellular communication (Agnati et al., 1995). By volume transmission, neurotransmitters may spread for distances beyond the point of release through the extracellular space and exert their activity at multiple receptor sites within a brain area; this may permit the area to operate as a unified whole.
Several investigators have undertaken the historical analysis of this ideas (Agnati et al., 1995; Bach-y-Rita,1993). This concept can be traced to Golgi's reticular tenet postulating that the CNS operates as a global neuronal continuity in which all elements are connected to others (see Agnati et al., 1995). In Russian physiological school, D.A.Sacharov has proposed a critical revision of synaptic theory in his concept of "heteron" (Sacharov,1990). Using morphological studies, V.A. Otellin has provided evidence in favour of non-synaptic nature of interactions between different neurotransmitter systems in the CNS (Otellin & Arushanian, 1989).
The first indications that neurotransmission by diffusion may exist in the CNS were provided by morphological studies showing receptors for glutamate, GABA, monoamines, neuropeptides outside classical synapses in many brain areas (Otellin & Arushanian, 1989; Agnati et al., 1995; Bjorklund & Lindvall, 1986; Groves et al., 1994; La Gamma et al., 1994; Levey et al., 1993; Martin et al., 1993; Petralia et al., 1996; Yung et al., 1995). This hypothesis gained further support from studies using in vivo microdialysis, which gave a direct opportunity to describe processes occurring in the brain extracellular space. These studies have revealed that among the classical neurotransmitters found in the extracellular fluid at concentrations within the range required for activation of at list metabotropic non-synaptic receptors are monoamines, glutamate, GABA (Abercrombie et al., 1990; Conor et al., 1991; Gonon et al., 1994; Imperato et al., 1992; Parsons et al., 1991; Westerink et al., 1987). The efflux of neurotransmitter dopamine from the synaptic cleft (Garris et al., 1994) and its diffusion through the extracellular fluid over long distance (Parsons et al.,1991; Stamford et al., 1988) has been demonstrated in direct experiments. Taken together, these results strongly imply that release and spread of classical neurotransmitters within the extracellular fluid is the mode for informational handling in the CNS rather than non-functional component of the synaptic release of neurotransmitters that have already exerted their physiological action. This hypothesis has raised a number of important questions which currently remain unanswered. For example, what information is conveyed by volume transmission? How important is this information for the expression of adaptive behaviour? What mechanisms underlie this mode of interneuronal communication? Most of these questions represent a major challenge to future research in the field.
Here we discuss current ideas concerning some problems mentioned above. This paper focuses on three important classical neurotransmitters of the striatum (dopamine, glutamate and GABA) which may act via volume transmission mode of action in this brain area. We review the evidence for physiological importance of volume transmission of these neurotransmitters in controlling striatal functions and in expression of behaviour regulated by this brain area.
1. THE EXISTENCE OF VOLUME TRANSMISSION IN THE STRIATUM.
The striatum is a large forebrain structure, which is implicated in the normal control of motor functions as well as emotional and motivational processes (Otellin & Arushanian, 1989; Shapovalova et al., 1992). The main function of the striatum is proposed to gather information from different cortical areas and then to convey integrated signals to the brain pre- motor area, on the one hand, and back to the cortex, on the other hand (Carlsson & Carlsson, 1990; Goldman-Rakis & Selemon, 1990; Shapovalova et al., 1992).
Structural units of the striatum are considered to be principal GABA-ergic medium-sized spiny neurones that represent target cells for the major afferent systems to the striatum and simultaneously projection neurones of this brain area (Smith & Bolam, 1990). In addition to cortical glutamatergic input, all regions of the striatum receive topographical dopaminergic input from the ventral tegmental area and the substantia nigra (Shapovalova et al., 1992; Bjorklund & Lindvall, 1986). Morphological studies have revealed a convergence of cortical (glutamatergic) and dopaminergic inputs on the same dendritic spine of principal GABA-ergic striatal neurones (Smith & Bolam, 1990) that underlies interactions between dopaminergic, glutamatergic and GABA-ergic systems in this brain region.
During the past two decades, several models of striatal function have been advanced (Shapovalova et al., 1992; Carlsson & Carlsson, 1990; Goldman Rakic & Selemon,1990). All of them are solely based on synaptic transmission as the means of interneuronal communication in this brain area. Nonetheless, convincing arguments have been currently made that volume transmission as well may underlie the cross-talk between axon terminals and neurones in the striatum.
In this respect, most investigators have concentrated on studies of diffuse action of dopamine in the striatum. Since the discovery of dopaminergic systems in the CNS, the idea that dopamine might be considered as a classical neurotransmitter, has been called into question several times. Some studies revealed non-classical morphological features of dopamine synapses: forming multiple presynaptic varicosites en passage and lack of postsynaptic densities in some of these synaptic contacts (Bjorklund & Lindvall,1986; Smith & Bolam, 1990). Nevertheless, subsequent analysis using electron microscopic preparation, has revealed evidence of classic pre- and postsynaptic densities at least in some dopamine synapses (Smith & Bolam, 1990). A very striking analysis of processes underlying dopamine efflux from the synaptic cleft in the ventral striatum has been undertaken by Garris et al., (Garris & Wightman, 1994; Garris et al., 1994).
Dopamine synapses in the striatum have been described as two parallel, thickened membranes, 300 nm in length, with a synaptic cleft of 15 nm (Garris et al., 1994; Groves et al., 1994). Synaptic vesicles are densely packed not only in the presynaptic region but also in the adjacent axon segments. As stressed by Garris et al., (Garris & Wightman, 1994; Garris et al., 1994) originally, high-affinity dopamine uptake (Near et al., 1985) was suggested to terminate the dopamine action in the synaptic cleft. Moreover, the extracellular space was considered as a separating zone between synapses (Gonon et al., 1987). Nevertheless Garris et al., (Garris & Wightman, 1994; Garris et al., 1994) have shown, that striatal membranes express dopamine uptake sites underlying dopamine reuptake at a concentration of 5,9 pmol/mg protein. Taking into account the density of dopaminergic synapses (one synapses per 4mm), the number of uptake sites per dopamine synapse is calculated to be 1750 uptake sites/synapse (Garris et al., 1994). However, in spite of such high density of dopamine uptake sites and approximately the same calculated density of dopamine receptors (1655 D1 receptor sites and 433 D2 receptor sites per synapse), recent investigations using fast-scan cyclic voltammetry have revealed that dopamine released in response to single stimulus pulse (approximately 1000 molecules) escape from a synaptic cleft and penetrate into the extracellular space (Garris et al., 1994). These calculations imply that the majority of dopamine reuptake sites and dopamine receptors are located outside dopamine synapses (Garris et al., 1994). Garris et al., postulated that dopamine synapse is designed for the effective efflux of dopamine from the synaptic cleft to the extracellular space. Reuptake system is proposed to regulate extrasynaptic dopamine levels and the distance that dopamine can diffuse from the synapse (Garris & Wightman, 1994; Garris et al., 1994). The hypothesis of non-synaptic action of dopamine in the striatum has been further substantiated by studies showing that in the rat striatum, there is no spatial correspondence between sites of dopamine release and sites of dopamine receptor concentration. Indeed, morphological evidence has been obtained that dopamine terminals in the striatum form synaptic contact on the neck of dendritic spines of principal striatal neurones (Groves et al., 1994; Smith & Bolam., 1990) whereas the majority of striatal D1 and D2 receptors are located on spine heads, i.e. far from sites of dopamine release (Levey et al., 1993). Interestingly, glutamate terminals that originate from cortical areas, also form synaptic contact on spine heads of principal striatal neurones (Smith & Bolam., 1990)(fig 1).
- Fig. 1. Tentative receptor mechanisms involved in volume transmission of dopamine, glutamate and GABA in the striatum (Otellin & Arushanian,1989; Agnati et al., 1995; Calabresi et al., 1990; Grace, 1991;Groves et al., 1994; Levey et al., 1993; Martin et al.,1993; Petralia et al., 1996; Shi & Rayport, 1994, Smith & Bolam., 1990; Yung et al., 1995). GABA- principal GABA-ergic medium-sized spiny neurones of the striatum that provide integration of the input information to form output signals; D1, D2 - dopamine receptors, DA - dopamine; NMDA - NMDA receptor; AMPA - AMPA/kainate receptors. mGLU - metabotropic glutamate receptor. GLU - glutamate. GABAA, GABAB- GABA receptors.
The presynaptic interactions in the striatum might be still another important evidence for the existence of volume transmission in this brain region that involves not only dopamine but also glutamate, GABA and acetylcholine Abercrombie & Keefe, 1991; Connor et al., 1991; Zigmond et al., 1990). As mantioned in these and other studiese, these neurotransmitters exert a direct tetrodotoxin-insensitive presynaptic action on release of each other despite an apparent absence of axo-axonic synaptic contacts (7 axo-axonic synapses per 100 000 synaptic junctions)(Otellin & Arushanian, 1989; Kornhuber & Korhuber., 1986; Zigmond et al., 1990). These data have led some investigators to conclusion that presynaptic interactions in the striatum are based primarily on diffuse action of neurotransmitters (Otellin & Arushanian, 1989; Grace., 1991; Zigmond et al., 1990). Diffusion of dopamine throughout the extracellular space in the dorsal and ventral striatum has been demonstrated in direct experiments (Parsons et al., 1991; Stamford et al., 1988). The length of the diffusion path for a dopamine molecule in the striatum is about 100 mm (Parsons et al., 1991). However, in the dopamine-denervated striatum, this length normally increases up to 1 mm (Stamford et al., 1988), that corresponds to the size of striatal cell clusters - a small group of medium- sized spiny principal neurones lying around a large aspyny cholinergic interneurone (Goldman-Rakic & Selemon, 1990). Recent studies using electron microscopy combined with immunostaining with monoclonal antibody against choline acetiltransferase have revealed that only 8% of cholinergic axon terminals in the rat striatum form classical synaptic junctions (Contant et al., 1996), whereas striatal neurones express high amount of acetylcholine receptors (Shapovalova et al., 1992; Smith & Bolam., 1990). This finding makes it possible to suggest that a cholinergic interneurone located to the centre of a cell cluster of the striatum, interacts with other neurones in the cluster presumably via volume transmission. Irrespective of the role that cell clusters play in the striatum, it is likely that the diffusion of extracellular dopamine and acetylcholine within the cluster serves as an important mechanism of integration between cells in the cluster that influences functional state of the cluster as a whole.
Molecular diffusion of glutamate and GABA in the striatal extracellular space have not been investigated. However, studies using microdialysis, have shown that glutamate and GABA are permanently present in the striatal extracellular fluid at concentrations of 10-6 M and 10-7 M correspondingly (Saulskaya & Marsden, 1995-a; 1996; Connor et al., 1991; Saulskaya & Marsden,1995-c).
A current belief is that in the striatum, GABAB receptors serve as presynaptic ones. Indeed, as demonstrated in morphological and electrophysiological studies, striatal neurones express both GABAA and GABAB receptors that appear to be nonuniformly distributed (Calabresi et al., 1990; Shi & Rayport, 1994). In particular, GABAA receptors appear to show preferential localisation to postsynaptic sites and they are exclusively responsible for synaptic action of GABA, whereas GABAB receptors show non-synaptic and presynaptic localisation (Shi & Rayport, 1994). Taken together these data have led to suggest that GABAB receptors in the striatum account for most of GABA receptors for volume transmission in this brain area.
Non-synaptic glutamate receptor have been revealed in many brain regions (Agnati et al., 1995). Striatal neurones express a high density of NMDA, AMPA/kainate and metabotropic glutamate receptor subtypes. However, postsynaptic action of glutamate in the striatum appear to be mediated solely via AMPA/kainate receptors (Herrling ., 1985) that are normally located to postsynaptic sites (Martin et al.,1993). This receptor subtype have not been observed at presynaptic sites in the striatum (Martin et al.,1993). These morphological findings suggest, but of cause they do not prove, that in the striatum, extracellular glutamate exerts its action presumably via presynaptic and extrasynaptic NMDA and metabotropic glutamate receptors. Future research needs to address more specifically the question of the glutamate receptor subtypes involved in volume transmission in this brain area.
Membrane receptors of astroglia has been proposed to be another important target for non-synaptic action of extracellular neurotransmitters (Agnati et al., 1995; Bach-y-Rita,1993; La Gamma et al., 1994; Martin et al.,1993). At the present time, studies using electron microscopy combined with immunostaining have revealed that striatal astrocytes express dopamine, metabotropic glutamate, but not AMPA/kainate receptors (La Gamma et al., 1994; Petralia et al., 1996; Martin et al.,1993).
2. MECHANISMS UNDERLYING VOLUME TRANSMISSION IN THE STRIATUM
Dopamine level in the striatal extracellular fluid is proposed to be independently regulated by several mechanisms. The first one is a synaptic vesicular release due to exocytosis. This process is impulse- and Ca++ - dependent (Abercrombie & Keefe, 1991). The second process is high affinity uptake of released dopamine, that is the primary mechanism by which dopamine is inactivated (Abercrombie & Keefe, 1991). The third mechanism is the non-vesicular carrier-mediated release of dopamine due to reversal of dopamine re-uptake mechanism. This process is impulse- and Ca++ -independent (Levi & Raiteri, 1993). The balance between these processes is suggested as the major determinant of extracellular dopamine level in the striatum (Abercrombie & Keefe, 1991). As has been recently shown, enzymatic degradation of released dopamine does not play a role in the determination of basal dopamine level in this brain area (Justice et al., 1994).
The origin of extracellular glutamate and GABA in the striatum appears to be vesicular and nonvesicular release limited by re- uptake (Smolders et al., 1994). A significant proportion of glutamate and GABA (60-80%) in the striatal extracellular space arises via vesicular and non-vesicular release from neurones (Smolders et al., 1994). In addition, extracellular glutamate and GABA may arise from glia cells and also leakage from metabolic pools.
Electrophysiological recording from identified dopaminergic neurones have shown that under physiological conditions, these cells typically exhibit two patterns of discharge activity (Fig. 2): either single spikes at frequencies averaging 3-4 Hz (pacemaker-like firing) or bursts of action potentials (2 to 6 action potentials at a frequency of 15 Hz)(Grace., 1991).
- Fig. 2. Tentative neuronal mechanisms underlying synaptic (A) and volume (B) transmission of dopamine in the striatum. Two patters of activity of dopaminergic neurones, single spikes and burst firing, result in synaptic and volume transmission of dopamine in the striatum correspondingly (Garris et al., 1994;Gonon, 1994;Gonon et al., 1987;Grace, 1991) Bust firing of dopaminergic neurones is induced by glutamatergic cortical signals through NMDA receptor activation (Gariano & Groves, 1988; Sanghera et al., 1984; Saud-Chagny et el., 1991; Svensson & Tung,1989). D1, D2 - dopamine receptors, DA- dopamine; NMDA - NMDA receptor; GLU - glutamate.
Electrical stimulation of nigro-striatal dopaminergic pathway mimicking the spontaneous bursting pattern is several times as portent as regularly spaced ones having the same average frequency in affecting dopamine release (Gonon., 1994). These data have led to conclusion that physiological significance of burst-like firing of dopaminergic neurones in the striatum is to initiate volume transmission. In contrast, pacemaker-like firing of dopaminergic neurones results in synaptic transmission, and under these conditions, dopamine does not escape from the synaptic cleft and can exert its action on synaptic receptors within close proximity to sites of release.
Since bust firing is never observed in experiments in vitro (Sanghera et al., 1984), this pattern appear to depend on activity of afferent fibbers (Kalivas, 1993). A detailed analysis of this phenomenon undertaken by Kalivas, reveals that glutamatergic cortical inputs to dopaminergic cells are, at least partly, responsible for converting pacemaker-like firing in dopaminergic cells of the ventral tegmental area into burst-firing patterns (Kalivas, 1993). Electrical stimulation of the prefrontal cortex converts dopamine neuronal activity into bursting patterns (Gariano & Groves., 1988; Kalivas, 1993) and cooling the prefrontal cortex converts spontaneous burst firing dopamine cells back to pacemaker-like firing (Svensson & Tung,1989; Kalivas, 1993). Moreover, the local administration of NMDA to the ventral tegmental area induces burst firing accompanied by dopamine release in the ventral striatum (Saud-Chagny et el., 1991).
These data allow to draw a conclusion that in the striatum, dopaminergic volume transmission is typically initiated by bust firing of dopaminergic neurones induced by glutamatergic cortical signals through NMDA receptor activation.
In addition, prefrontal cortex, as well as other glutamatergic afferent sites that project to the striatum, may influence dopaminergic volume transmission in this brain area via presynaptic mechanisms as administration of excitatory amino acids or their analogues into the striatum elicits increase in striatal dopamine extracellular level in a tetrodotoxin-insensitive manner (Shapovalova et al., 1992; Abercrombie & Keefe, 1991). Interestingly, experiments have shown that the local application of glutamate antagonists into the striatum via the microdialysis probe (Shapovalova et al., 1992; Abercrombie & Keefe, 1991) and, as revealed in our studies, damage to the cortical area projecting to the striatum (Saulskaya & Gorbachevskaya, 1997; Saulskaya et al., 1996), does not influence basal dopamine release into the striatal extracellular space. Therefore, this data suggest that glutamatergic inputs to the striatum only exert transient influence on dopamine extracellular outflow in this brain area and this influence does not occur under the rest conditions. In contrast, impulse activity in nigrostriatal dopaminergic neurones appear to be the principle determinant of extracellular dopamine concentrations under basal conditions, as evidenced by the dramatic decreases in basal dopamine extracellular level in the striatum after injections of tetrodotoxin into the medial forebrain bundle (Abercrombie & Keefe, 1991). However, in our recent study, we have demonstrated that learning causes long-lasting changes in the mechanisms involved in the presynaptic glutamatergic control of basal dopamine release into the extracellular space in the striatum (Saulskaya & Marsden, 1995-b). Using microdialysis, two hours after an acquisition of conditioned emotional response, we revealed an appearance of NMDA-dependent component of basal dopamine release in the ventral striatum of learned rats (but not of untrained animals) although apparent "basal" dopamine release has returned to normal.
Therefore, volume dopaminergic transmission in the striatum appear to be under the double control of cortical glutamatergic areas. Dopamine release into the striatal extracellular space may be induced by either burst firing of dopaminergic neurones initiated by cortico-nigral glutamatergic signals or presynaptic cortico-striatal influence.
3. FUNCTIONAL ROLE OF VOLUME TRANSMISSION IN THE STRIATUM
Although a number of investigations using in vivo microdialysis, have provided evidence that striatal extracellular level of dopamine, glutamate and GABA changes in response to behavioural challenge (Saulskaya, 1993; Saulskaya & Marsden, 1994; 1995-a; 1995-b; 1996; Imperato et al., 1992; McCulloughet al., 1993; Phillips et al., 1991; Shi & Rayport, 1994;), until recently it has not been established if these changes are essential for the expression of behavioural activity.
The first direct evidence for the functional importance of dopamine diffusion within the striatal extracellular fluid has been obtained in studies that revealed compensations after lesions of central dopaminergic neurones. These studies using neurotoxin, 6-hydroxidopamine, suggest that the subtotal loss of nigral dopaminergic neurones is compensated, to some extent, by increased release of dopamine from residual dopaminergic neurones (Zigmond et al., 1990). This maintains the constancy of striatal dopamine extracellular level correlated with the absence of functional impairments (Abercrombie et al 1990; Parsons et al., 1991; Zigmond et al., 1990). Significant decrease in striatal extracellular dopamine level accompanied by akinesia, only develops if degeneration of the dopaminergic neurones is almost complete (not less then 90%) (Abercrombie et al 1990; Zigmond et al., 1990).
Moreover, we have observed a comparable phenomenon in studies of glutamatergic systems of the striatum. In particular, using microdialysis and measurements of glutamate release, we have revealed that partial excitotoxic lesions of the hippocampal glutamatergic inputs to the ventral striatum made by infusion of ibotenic acid, influence neither basal glutamate extracellular level in this area nor open-field activity. Only complete loss of neurones in the ventral subiculum and CAI projecting to the ventral striatum, causes both decrease in striatal extracellular glutamate level and hyperlocomotion.
Therefore, extensive impairment of impulse-flow dependent synaptic transmission in the striatum due to damage to dopaminergic and glutamatergic inputs does not influence locomotion, whereas impairment of volume transmission influences this type of behavioural activity, that is known to be under the control of the striatum.
Taken together, these data suggest firstly that compensatory events occur to maintain extracellular level of dopamine and glutamate after subtotal degeneration of dopaminergic and glutamatergic inputs to the striatum. Secondly, they indicate the functional significance of volume transmission of dopamine and glutamate for the expression of behavioural activity regulated by the striatum.
Regulation of locomotion by dopaminergic and glutamatergic systems in the striatum involving volume transmission, is thought to be an integral part of more global functions operating via this brain area. As postulated by Stricker & Zigmond, "...motivated behaviours can be described in terms of two components: one that directs activities toward distinctive goal and one that energizes activities regardless of their goal" (Stricher & Zigmond,1989). Important functions of meso-striatal dopaminergic systems are considered to provide non-specific activational component of motivated behaviours (Stricker & Zigmond,1989). The way they operate is a modulation of the striatal responsiveness to cortical glutamatergic inputs and, via stria-pallido-thalamo-cortical loop, the responsiveness of cortical areas to sensory inputs that influence the degree of arousal (Carlsson & Carlsson, 1990, Stricker & Zigmond,1989). Various lines of evidence have indicated that this is volume transmission that underlies these functions of the dopaminergic systems (Phillips et al., 1991; Joseph et al., 1991). Data from literature and our studies have shown that polymodal environmental stimuli (e.g. mild stress, food and water intake, social and sexual contacts, novelty) are associated with prolonged dopamine release into the extracellular space of the striatum (Saulskaya, 1993; Saulskaya & Marsden, 1994; 1995-b; Imperato et al., 1992; Joseph et al., 1991; McCulloughet al., 1993; Phillips et al., 1991). Studies utilising brain microdialyses have observed increased extracellular level of dopamine in the striatum during all behavioural situation studied i.e.: those involving movements (McCulloughet al., 1993) as well as inhibition of movements (Saulskaya, 1993; Saulskaya & Marsden, 1994;1995-b]; accompanied by either reward (Phillips et al., 1991), or punishment (Saulskaya, 1993; Saulskaya & Marsden, 1994; 1995-b). Both stress and relief from stress causes dopamine release into the ventral striatum (Imperato et al., 1992). Therefore, initiation of volume transmission reflected in dopamine release into the extracellular space of the striatum, is hypothesized to be a non-specific response occurring during situations that require enhanced behavioural responsiveness (Joseph et al., 1991). This response depends neither on motivation underlying ongoing behaviour, nor a pattern of movement activity.
The standard view holds that dopamine action in the striatum is mediated by two families of G-protein-coupled receptors whose stimulation exerts both immediate and delayed action on striatal neurones. One of the most important consequence of dopamine action in the region of the striatum is the control of signal-to-noise ratio. Enhanced dopamine tone appears to be associated with inhibition of spontaneous activity of principal striatal neurones and with an increase in their response elicited by glutamatergic inputs (Rolls et al., 1984; Stamford et al., 1988). Furthermore, extracellular dopamine via gap junctions decreases the degree of electrotonic coupling between striatal neurones (Donnel & Grace, 1995). These temporal anatomical changes make the activity of individual striatal neurones more segregated and, therefor, more selective for input signals. These findings suggest an important contribution of dopamine extracellular action to behaviour.
In addition, dopamine release into the striatal extracellular space occur slowly with delayed peak and extended duration; the entire process occurring over periods of tens of minutes and reaching a maximum after the behavioural session (Imperato et al., 1992; Joseph et al.,1991; Phillips et al., 1991; Saulskaya,1993; Saulskaya & Marsden,1994; 1995-b). The question arises, what for and through what mechanisms such prolonged activation of volume transmission occur in response to behaviour.
We believe that the necessity of prolonged dopamine release is accounted for the importance of extended dopaminergic extracellular stimulation, in particular, for the expression of immediate-early genes encoding transcriptional factors that regulate the expression of late specific target genes providing coupling short-term stimulus-response cascades to long term changes in neurones (Berretta et al., 1992).
Besides an influence of volume transmission on processes underlying plasticity of striatal neurones, regulation mechanisms of volume transmission as such also involve memory processes. In particular, findings from our studies have shown that dopamine and glutamate release into the striatal extracellular space can be caused by conditioned stimuli previously paired with reinforcement (Saulskaya & Marsden, 1994; 1995-a; 1995-b; 1995-c).
In studies of mechanisms underlying prolonged dopamine release into the striatal extracellular space in response to behaviour, we have revealed that at least under certain conditions presynaptic glutamatergic influence through activation of NMDA presynaptic receptor may increase the duration of dopamine release after the behavioural session (Saulskaya & Marsden, 1994; 1995-b). In particular, NMDA antagonist MK-801, but not AMPA/kainate antagonist CNQX administered into the extracellular space of the ventral striatum, had no significant effect on the immediate increase in dopamine release induced by conditioned emotional response but completely prevented the later phase lasted for an hour after the behavioural session (Saulskaya & Marsden, 1994; 1995-b). This supposition was later substantiated by our studies of glutamate release in the striatal extracellular space, which have further demonstrated the existence of the delayed increase in extracellular glutamate that might be responsible for the maintained delayed phase of the increase in dopamine release following conditioned emotional response (Saulskaya & Marsden, 1995-a; 1995-c). We suggest that during conditioned emotional response, glutamate, released by terminals located on a head of a dendritic spine of principal striatal neurones, escapes the synaptic cleft and diffuses through the extracellular space to the dopaminergic terminals located on a neck of the dendritic spine. Thus, cortical glutamatergic inputs may be capable of modulating the extracellular concentration of glutamate in the vicinity of the dopaminergic terminals so that glutamate may exert an additional action on NMDA presynaptic receptors located there, and therefore it causes additional dopamine release registered as NMDA dependent delayed component.
Therefore, glutamatergic volume transmission may be also capable of setting level of responsivity of principal striatal neurones by increasing the duration of extracellular dopamine release in response to behaviour.
There is also evidence that extracellular GABA would be expected to regulate the responsivity of striatal neurones. Our data have indicated that all behavioural situations studied (exploratory activity, mild stress, response to, as well as relief from danger) are associated with increased extracellular GABA release in the ventral striatum (Saulskaya & Marsden, 1995-c; 1996). Consistent with these observations, recent investigations of cultured striatal neurones have provided evidence for implications of the intrinsic strialal GABA-ergic system which is known to set the level of extracellular GABA in the striatum (Smolders et al., 1994), into the regulation of signal-to-noise ratio in the striatal information processing (Shi & Rayport, 1994).
CONCLUSIONS
In this paper, we review evidence that, in addition to synaptic transmission, another neurotransmitter mechanism, named volume transmission, exists in the striatum. This complimentary neurotransmitter mechanism operating independently and differently from synaptic transmission, is based on diffusion of dopamine, glutamate, GABA and other compounds through the striatal extracellular fluid. Neurones, axon terminals, astroglia might be candidate targets for volume transmission signals. D1, D2 dopamine receptors, NMDA and, probably, metabotropic glutamate receptors as well as GABA-B receptors are expected to be involved as receptors for volume transmission in the striatum (Fig. 1).
One of the primary functions of volume transmission in the striatum is to maintain the levels of striatal and cortical responsivity that influences the degree of arousal depending on ongoing behavioural situation. At present, it can safely be said that under the rest conditions, extracellular dopamine sets the background level of neuronal excitability in the striatum via D2 nonsynaptic, tonically active receptors (Connor et al., 1991). D1 receptors whose activation requires higher concentrations of dopamine than those exist in the striatal extracellular space under basal conditions, as well as NMDA receptors because of Mg++ block, are not tonically active (Abercrombie & Keefe, 1991; Connor et al., 1991; Saulskaya & Marsden,1994;1995-b). There is no available evidence in favour of tonic activation of GABA nonsynaptic receptors and metabotropic glutamate receptors in the striatum.
One of the most important properties of volume transmission in the striatum appears to be an exchange of receptor subtypes involved during situations in which a resting animal starts behavioural activity. Indeed, relevant environmental stimuli have been shown to cause additional dopamine release in the striatal extracellular space (Imperato et al., 1992; McCulloughet al., 1993; Phillips et al., 1991; Saulskaya & Marsden, 1994; 1995-b). Studies using voltammetry have shown that under these conditions, local increases in extacellular dopamine level in the striatum can be higher than those obtained by microdialysis and high enough to activate D1 dopamine receptors. It should be mentioned that an increase in extracellular dopamine in the striatum causes qualitative rather than quantitative changes. Another receptor subtype, D1, exhibiting different electrophysiological and pharmacological profile and located at output striatal neurones of another type (Gerfen et al., 1990), gets involved in the response. This view has gained further support from our studies showing that the behavioural activity causes tonic activation of NMDA presynaptic receptors that results in the appearance of NMDA dependent component of basal dopamine release in trained animals (Saulskaya & Marsden, 1994;1995-b). While the precise mechanisms underlying the activation of NMDA receptors without AMPA/kainate receptor activation still remains to be investigated, it is clear that NMDA receptors become available for glutamate released in response to behaviour. Therefore, one could suggest the existence of two receptor systems involved in volume transmission: the system for the rest and the system for the functional activity. We believe that a switch from one receptor system to another is an important component in the organisation of behaviour.
It should be stressed that volume transmission is not a unique mechanism of mammalian CNS (Agnati et al., 1995; Arshavsky et al., 1988; Bach-y-Rita,1993). Arshavsky et al (Arshavsky et al., 1988) provided direct evidence for the functional significance of volume transmission in ganglia of invertebrates. After complete disruption of synaptic contacts, a single neurone of isolated pedal ganglia was still able to express its functional activity (Arshavsky et al., 1988). V.A.Otellin (Otellin & Arushanian, 1989) and Agnati et al.(Agnati et al., 1995) have suggested that in phylogeny, volume transmission is more ancient mechanism of neuronal communication in the CNS than synaptic transmission. Nevertheless, volume transmission still plays a very important role throughout phelogeny. In fact, neurones in the youngest brain areas, such as cortex, express a maximum of nonsynaptic receptors for acetylcholine, excitatory amino acids, monoamines (see for ref. Agnati et al., 1995; Contant et al., 1996; Martin et al.,1993).
It would be wrong to oppose volume and synaptic transmission, considering them as alternative mechanisms of neuronal communication. As suggested by D.A.Sacharov (Sacharov,1990), synaptic transmission may be defined as a particular case of volume transmission occurring if a diffusion area is restricted within a synaptic cleft. We believe that in the CNS, both neurotransmitter mechanisms operate as complimentary processes. Volume transmission plays a role of "tuning" whereas synaptic transmission is responsible for quick "executive" processes.
In general, it must be emphasised that in spite of the accumulating data in this field, our ideas concerning functional role of volume transmission in the CNS are mostly speculative. Future research needs to address more specifically some questions mentioned here. In particular, our knowledge about physiological relevance of extracellular GABA in the striatum is far from being completed. We currently know very little of glutamate and GABA receptor mechanisms for volume transmission in the striatum. It still remains to be seen how glia cells are involved in volume transmission in the striatum. These and perhaps other questions are going to be high on the agenda in the nearest future.
Acknowledgement: The research is supported by Russian Fond for Fundamental Research (grant N 95-04-11524a)
REFERENCES
Otellin V.A., Arushanian E.B. (1989) The Nigro-Strio-Nigral System (In Russian). Medicina, Moskow.
Saulskaya N.B. Synaptic release of dopamine during an acquisition of passive avoidanse task. (1993) (In Russian) Zhurn. Vyssh. Nerv. Dejat. 43. 1025-1029.
Saulskaya N.B., Marsden C.A. (1994) The role of glutamatergic inputs to the n. accumbens in control of synaptic dopamine release during association learning. (In Russian) Physiol. Journ. 80. 45-54.
Saulskaya N.B., Marsden C.A. (1995-a) Extracellular glutamate levels in the n. accumbens during conditioned emotional response. Physiol. Journ. (In Russian) 81. 161-165.
Saulskaya N.B., Marsden C.A. (1996) The GABA extracellular level in the n. accumbens in the course of a conditioned nociceptive response in rats. (In Russian ) Physiol. Journ. 82. 37-43.
Saulskaya N.B., Gorbachevskaya A.I. (1997)Conditioned dopamine release in the nucleus accumbens in rats with lesions of the hippocampal formation. (In Russian) Physiol Journ. 83. 76-82.
Sakharov D.A. (1990) A diversity of neutotransmitters: functional meaning. (In Russian ) Zhurn. Evol. Biochim. Fiziol. 26. 733-741.
Shapovalova K.B. Gorbachevskaya A.I., Saulskaya N.B. (1992) Structural organization and neurochemical mechanisms of participation of the nucleus accumbens in the interaction of limbic and motor systems and in the regulation of motor behaviour. (In Russian) Zhurn. Vyssh. Nerv. Dejat. 42. 226-276.
Abercrombie E.D., Bonatz A.E., Zigmond M.J. (1990) Effects of L-DOPA on extracellular dopamine in striatum of normal and 6-hydroxydopaminetreated rats. Brain Res . 525. 36-44.
Abercrombie E.D., Keefe K.A. Determinants of extracellular dopamine concentration as measured by microdialysis. (1991) In: Monitoring Molecules in Neuroscience. Eds. H. Rollema, B. Westerink, J.D. Drijfhout. University Centre for Pharmacy, Groningen. 308-311.
Agnati L.F., Zoli M., Stromberg I., Fuxe K. (1995) Intercellular communication in the brain: Wiring versus volume transmission. Neurosci. 69. 711-726.
Arnt J. (1981)Hyperactivity following injection of a glutamate agonist and 6.7.ADIN into the rat nucleus accumbens and its inhibition by THIP. Life Sci. . 28. 1597-1603.
Arshavsky Yu.I., Deliagina T.G., Gerlfand I.M., Orlovsky G.N.,Panchin Yu.V.,Pavlova G.A., Popova G.A. (1988) Non- synaptic interaction between neurons in molluscs. Comparative Biochem. Physiol. 91C. 199-203.
Bach-y-Rita P. (1993) Neurotransmission in the brain by diffusion through the extracellular fluid: A review. NeuroReport. 4. 343-350.
Berretta S., Robertson H.A., Graybiel A.M. (1992) Dopamine and glutamate agonists stimulate neuron-specific expression of fos-like protein in the striatum. J. Neurophysiol. 68. 767-777.
Bjorklund A. Lindvall O. (1986) Dopamine-containing systems in the CNS. In: Handbook of Chemical Neuroanatomy. Classical Transmitters and Transmitter Receptors in the CNS. Part II, 3. Eds. A.Bjorklund, T.Hokfelt, M.J.Kuhar, Elsevier, Amsterdam, New York, Oxford. 55-111.
Boldry R.C., Willins D.L., Wallace L.J., Uretsky N..J. (1991) The role of endogenous dopamine in the hypermotility response to intra-accumbens AMPA. Brain Res. 559. 100-108
Calabresi P., Mercuri N.B., DeMurtas M., Bernardi G. (1990) Endogenous GABA mediates presynaptic inhibition of spontaneous and evoked excitatory synaptic potentials in the rat neostriatum. Neurosci. Lett. 118 . 99-102.
Carlsson M., Carlsson A. (1990) Interactions between glutamatergic and monoaminergic systems within the basal ganglia - implications for schizophrenia and Parkinson's disease. TINS. 13. 272-276.
Connor W.T.O., Herrera-Marschits M., Lindefors N., Osborne P.G.,Drew K.L., Reid M., Ungerstedt U. (1991) Dopamine- GABA interactions in the striatum. In: Monitoring Molecules in Neuroscience. Eds. H. Rollema, B. Westerink, J.D. Drijfhout, University Centre for Pharmacy, Groningen. 93-95.
Contant C., Umbriaco D., Garcia S., Watkins K.S., Descarries L. (1996) Ultrastracrtural characterization of the acetilcholine innervation in adult rat neostriatum. Neurosci. 7. 937-947.
Donnel P.O, Grace A. (1995) Different effects of subchronic clozapine and haloperidol on dye-coupling between neurons in the rat striatal complex. Neurosci. 66. 763-767.
Gariano R.F., Groves P.M. (1988) Burst firing induced in midbrain dopamine neurons by stimulation of the medial prefrontal and anterior cingulate cortex. Brain Res. 462. 194-198.
Garris P.A., Ciolkowski E.L., Pastore P., Wightman R.M. (1994) Efflux of dopamine from the synaptic clift in the nucleus accumbens of the rat brain. J. Neurosci. 14. 6084-6093.
Garris P.A. & Wightman R.M. (1994) Efflux of dopamine from the synaptic cleft in the nucleus accumbens of the rat brain:evidence for an extrasynaptic mode of dopamine transmission. In: Monitoring Molecules in Neurosciense. Eds. A. Louilot, T. Durkin, U.Spampinato, M.Cador. Bordeaux University. Bordeaux. 62-63.
Gerfen C.R., Engber T.M., Mahan L.C., Susel Z.,Chase T.N. (1990) D1 and D2 dopamine receptor-regulated gene expression of striatonigral and srtiatopallidal neurons. Science. 250. 1429-1432.
Goldman-Rakic P.S., Selemon L. (1990) New frontiers in basal ganlia research. TINS. 13. 241-244.
Gonon F. (1994) Kinetics of catecholamine release and elimination in the peripheral and central nervous system. In: Monitoring Molecules in Neuroscience. Eds. A.Louilot, T. Durkin, U. Spampinato, M.Cador. Bordeaux University. Bordeaux. 60-61.
Gonon F.G., Mermet C.C., Marcenac F.H. (1987) In vivo voltammetry with carbon fiber electrodes: a tool for studying dopaminergic neurons of the basal ganglia. In: Neurotransmitter Interactions in the Basal Ganglia. Eds. M. Sandler, C. Feuerstein, B. Scatton. Raven Press. New York. 101-110.
Grace A.A. (1991) Phasic versus tonic dopamine release and the modulation of dopamine system responsivity: A hypothesis for the etiology of schizophrenia. Neurosci. 41. 1-24.
Groves P.M., Linder J.C., Young S.J. (1994) 5-hydroxydopamine-labeled dopaminergic axons: Three dimensional reconstructions of axons, synapses and postsynaptic targets in the neostriatum. Neurosci. 58. 593-604.
Hannigan J.H., Springer J.E., Isaacson R.L. (1984) Differentiation of basal ganglia dopaminergic invovement in the behaviour after hippocampectomy. 291. 83-91.
Herrling P.L. (1985) Pharmacology of corticocaudate exitatory postsynaptic potential in the cat: Evidence for its mediation by quisqualate or kainate-receptors. Neurosci. 14. 417-426.
Imperato A., Angelucci L., Casolini P., Zocchi A., Puglisi-Allegra S. (1992) Repeated stressfull experience differentially affect limbic dopamine release during and following stress. Brain Res . 57. 194-199. J
oseph M.H., Young A.M.J., Gray J.A. (1991) Reinforcement, conditioning and dopamine function in the nucleus accumbens. In: Monitoring Molecules in Neurosciense. Eds. H. Rollema, B. Westerink, J.D. Drijfhout. University Centre for Pharmacy. Groningen. 200-203.
Justice J.B., Smith A.D., Sam P.M. (1994) Characterization of dopamine microdialysis and application to regulation of the dopamine transport. In: Monitoring Molecules in Neurosciense. Eds. A. Louilot, T. Durkin, U.Spampinato, M.Cador. Bordeaux University. Bordeaux. 73-74.
Kalivas P.T. (1993) Neurotransmitter regulation of dopamine neurons in the ventral tegmental area. Brain Research Rev. 18. 75-113. Kornhuber J., Kornhuber M.E. (1986) Presynaptic dopaminergic modulation of cortical input to the striatum. Life Sci. 39. 669-674.
La Gamma E.F., Strecker E., Lenn N.J., DeCristofaro J.D., Weisinger G. (1994) Dopamine regulation of transfected preproenkephaline promoter in primary rat astrocytes in vitro and in vivo. Exptl. Neurol. 130. 304-310.
Levey A.I., Hersch H., Rue D.B., Sunahara R.K, Niznik H.B., Kitt C.A., Price D.L., Maggio R., Brann M.R., Ciliax B.J. (1993) Localization of D1 and D2 dopamine receptors with subtype-specific antibodies. Proc.Natl. Acad. Sci. USA. 90. 8861- 8865.
Levi G., Raiteri M. (1993) Carrier-mediated release of neurotransmitters. TINS. 16. 415-419. Martin L.J., Blackstone C.D., Levey A.I., Huganir R.L., Price D.L. (1993) AMPA glutamate receptor subunits are defferentially distributed in rat brain. Neurosci. 53. 327-358.
McCullough L.D., Sokolowski J.D., Salamone J.D. (1993) A neurochemical and behavioral investigation of the involvement of nucleus accumbens dopamine in instrumental avoidance. Neurosci . 52. 919-925.
Near J.A., Bigelow J.B. Wightman R.M. (1985) Comparison of uptake of dopamine in rat striatal chopped tissue and synaptosom. J. Pharmacol. Exp. Ther. 245. 921-927. Parsons L.H., Smith A.D., Justice J.B. (1991) The in vivo microdialysis recovery of dopamine is altered independently of basal level by 6-hydroxydopamine lesions to the n. accumbens. J. Neurosci. Methods. 40. 139-147.
Phillips A.G., Atkinson L.J., Blackburn J.R., Blaha C.D. (1991) In vivo analyses of dopamine neurotransmission during anticipatory and consummatory phases of feading behavior. In: Monitoring Molecules in Neuroscience. Eds. H. Rollema, B. Westerink, J.D. Drijfhout. University Centre for Pharmacy. Groningen. 204-297.
Petralia, R.S., Wang,Y.X., Niedzielski A.S., Wenthold R.J. (1996) The metabotropic glutamate receptors, MGLUR2 and MGLUR3, show unique postsynaptic, presynaptic and glial localizations. Neuroscience. 71. 949-956.
Rolls E.T., Thorpe S.J., Boytim M., Szabo I., Perrett D.I. (1984) Responses of striatal neurons in the behaving monkey. 3. Effects of iontophoretically applied dopamine on normal responsiveness. Neurosci. 12. 1201-1212.
Sanghera M.K., Trulson M.E., German D.C. (1984) Electrophysiological properties of mouse dopamine neurons: In vivo and in vitro studies. Neurosci. 12. 793-801.
Saud-Chagny M.F., Ponec J., Gonon F. (1991) Presynaptic autoinhibition of the electrically evoked dopamine release studied in the rat olfactory tubercule by in vivo electrochemistry. Neurosci. 45. 641-652.
Saulskaya N., Marsden C.A. (1995-b) Conditioned dopamine release: Dependence upon N-methyl-D-aspartate receptors. Neurosci. 67. 57-63.
Saulskaya N., Marsden C.A. (1995-c) Extracellular glutamate in the nucleus accumbens during a conditioned emotional response in the rat. Brain Res. 698. 114-120.
Saulskaya N., Marsden C.A., Gorbachevskaya A.I. (1996) Conditioned emotional response and glutamate - dopamine interaction. In: Monitoring Molecules in Neuroscience. Eds. J.L.Gonzales-Mora, R.Borges, M.Mas. University of La Laguna. Santa Cruz de Tenerife. 329-330.
Shi W.-X., Rayport S. (1994) GABA synapses formed in vitro by local axon collaterals of nucleus accumbens neurons. Neurosci. 14. 4548-4560.
Smith A.D., Bolam J.P. (1990) The neural network of the basal ganglia as revealed by the study of synaptic connections of identified neurones. TINS. 13. 259-265.
Smolders I., Herregodts P., Michotte Y., Ebinger G. (1994) Modulation of the extracellular levels of GABA and glutamate in rat striatum by meanes of local adminisrtation of uptake inhibitors and after decortication. In: Monitoring Molecules in Neuroscience. Eds. A.Louilot, T.Durkin, U.Spampinato, M.Cador, 1994, Bordeaux University, Bordeaux, p. 291.
Stamford J.A., Zurmunt L., Kruh P. (1988) Diffusion and uptake of dopamine in rat caudate and nucleus accumbens compared using fast cyclic voltammetry. Brain. Res. 448. 381-386.
Stricher E.M., Zigmond M.J. (1986) Brain monoamines, homeostasis and adaptive behavior. In: Handbook of Physiology. The Nervous System IV. Eds. V.B. Mountcastle., F.E.Bloom., S.R.GeigerAmer. Physiol. Socicty. Bethesda. Mariland. 677- 700.
Svensson T.H., Tung C.-S. (1989) Local cooling of prefrontal cortex induced pacemaker-like firing of dopaminergic neurons in rat ventral tegmental area in vivo. Acta Physiol. Scand. 136. 135-136.
Tamminga C.A., Thaker G.K., Bchanan R., Kikpatric B., Alphs L.D., Chase T.N., Carpenter W.T. (1992) Limbic systems abnormalities identified in schizophrenia using position emission tomography with fluorodeoxyglucose and neocortical alteration with defici syndrom. Arch. Gen. Psychiatry. 49. 522-530.
Westerink B.H.C., Tuntler J., Damsma G., Rollema H., De Vries J.B. (1987) The use of tetrodotoxin for the characterization of drug-enhanced dopamine release in conscious rats studied by brain dialysis. Naunyn-Schmiedeberg's Arch. Pharmac. 336. 502-507.
Wu M., Brudzynski M. (1995) Mesolimbic dopamine terminals and locomotor activity induced from the subiculum. Neuroreport.. 6. 1601-1604.
Yung K.K.L., Bolam J.P., Smith A.D., Hersch S.M., Ciliax J., Levey A. (1995) Immunocytochemical localization of D1 and D2 dopamine receptors in the basal ganglia of the rat: light and electron microscopy. Neurosci. 65. 709-730.
Zigmond M., Abercrombie E.D., Berger T.W., Grace A.A., Stricker E.M. (1990) Compensations after lesions of central dopaminergic neurons: some clinical and basic implications. TINS. 13. 290-295.
Contact address: Saulskaya N.B. Pavlov Institute of Physiology RAS,