Six questions on the subthalamic nucleus: lessons from animal models and from stimulated patients.

Since the early 90s, the subthalamic nucleus (STN) has started to be the subject of an increasing interest not only in the community of the basal ganglia scientists but also for neurosurgeons and neurologists, thanks to the development of the surgical treatment for Parkinson's disease. The involvement of the STN in cognitive and motivational processes has been demonstrated since, and psychiatrists are now considering this small structure as a possible target for the treatment of various disorders. In this review, we will address six questions to highlight (1) How increased knowledge has led us from a strictly motor model to an integrative one. (2) How knowledge acquired in animal models can be similar or (3) different from the effects observed in human patients. (4) How clinical trials are sometimes ahead of fundamental research carried out in animals, showing effects that could not be predicted on the basis of animal studies, thus questioning the relevance of some animal models, especially for psychiatric disorders. We will also address the possible future orientations (5) and how the use, or precaution not to use, certain key words in animal research dedicated to STN functions can lead to the omission of a certain amount of available data in the literature (6).

Deficits of glutamate transmission in the striatum of experimental hemiballism.

Hemiballism (HB) is a quite rare disorder, generally secondary to stroke, neoplasms or demyelinating plaques, classically considered as almost pathognomonic of a lesion in the subthalamic nucleus (STN). This alteration causes involuntary movements in the chorea-ballism spectrum. One theory is that the output nuclei of the basal ganglia are overinhibited in HB, while little is known about the physiological state of the striatum, the major input structure of the basal ganglia. In the present study, we recorded spontaneous and miniature excitatory and inhibitory postsynaptic currents (sEPSCs, mEPSCs, sIPSCs, mIPSCs) from projection neurons of the striatum of experimental HB. We found a selective reduction of striatal sEPSC and mEPSC frequency following chemical lesion of the STN of the rat, suggesting that reduced synaptic excitation of the input structure of the basal ganglia represents a physiological correlate of HB.

Subthalamic nucleus lesion reverses motor abnormalities and striatal glutamatergic overactivity in experimental parkinsonism.

Subthalamic nucleus (STN) is a target of choice for the neurosurgical treatment of Parkinson's disease (PD). The therapeutic effect of STN lesion in PD is classically ascribed to the rescue of physiological activity in the output structures of the basal ganglia, and little is known about the possible involvement of the striatum. In the present study, therefore, we electrophysiologically recorded in vitro single striatal neurons of DA-depleted rats unilaterally lesioned by 6-hydroxydopamine, treated or not with therapeutic doses of levodopa (l-DOPA), or with a consecutive ipsilateral STN lesion. We show that the beneficial motor effects produced in parkinsonian rats by STN lesion or l-DOPA therapy were paralleled by the normalization of overactive frequency and amplitude of striatal glutamate-mediated spontaneous excitatory postsynaptic currents (sEPSCs). Since neither l-DOPA treatment nor STN lesion affected sEPSCs kinetic properties, the reversal of these abnormalities in striatal excitatory synaptic transmission can be attributable to the normalization of glutamate release.

Lesions of the medial and lateral striatum in the rat produce differential deficits in attentional performance.

Excitotoxic lesions of the medial frontal cortex and anterior cingulate cortex in rats have been shown to produce dissociable impairments on a reaction time visual attention (5-choice) task. Because these cortical areas project to the medial striatal region, the authors predicted similar deficits after lesions of this striatal area compared with the lateral area. Compared with sham-operated controls, rats with quinolinic acid-induced medial striatal lesions showed all the behavioral changes associated with medial frontal cortex and anterior cingulate cortex lesions. In contrast, lateral striatal lesions produced profound disturbances in the performance of the task. Control tests showed little evidence of gross deficits in either group of rats in terms of motivation, locomotor function, or Pavlovian appetitive conditioning. These data suggest that the medial and lateral striatum have contrasting roles in the control of instrumental responding related to the primary sources of their cortical innervation.

Effects of STN lesions on simple vs choice reaction time tasks in the rat: preserved motor readiness, but impaired response selection.

The subthalamic nucleus (STN) is a key structure within the basal ganglia, inactivation of which is a current strategy for treating parkinsonism. We have previously shown that bilateral lesions of the STN or pharmacological inactivation of this structure in the rat induce multiple deficits in serial reaction time tasks. The aim of the present study was to investigate further a possible role for the STN in response preparatory processes by using simple (SRT) and choice (CRT) reaction time tasks. In contrast to the CRT procedure, the information related to the location of where the response had to be made was given in advance in the SRT procedure. Accurate performance on these tasks requires not only the selection of the correct response (i.e. which response), but also preparation in order to perform when required. A comparison between the two tasks allows assessment of whether STN lesions affect which response ("which") or when to perform it ("when"). As previously observed in these procedures, the responses were faster as a function of the variable foreperiod preceding the trigger stimulus. This well-known effect, termed "motor readiness, was maintained after STN lesions, suggesting that STN lesions did not affect the "when" phase of action preparation. However, while performance on the SRT was faster than on the CRT task preoperatively, STN lesions slowed RTs and abolished the beneficial effect of advance information, suggesting a deficit in the selection ("which") phase of response preparation. This deficit in the selection phase was further supported by deficits in accuracy of responding after STN lesions, as well as increases in mislocated premature responding in the SRT condition. Together, these results suggest that the STN plays an important role in response preparatory processes, including response selection and inhibitory control processes.

Effects of dopamine depletion of the dorsal striatum and further interaction with subthalamic nucleus lesions in an attentional task in the rat.

The present study investigated the effects of 6-hydroxydopamine lesions of the dorsal striatum on a five choice serial reaction time task which assesses visual sustained and divided attention. Striatal dopamine loss by itself produced no deficits in accuracy on the standard form of the task, but lengthened response latencies and increased omissions and perseverative behaviour. Reducing the temporal predictability of the visual event led to impaired accuracy, contrasting with previously published effects of ventral striatal dopamine depletion. To further investigate the interactions between dopaminergic and glutamatergic systems within the basal ganglia, we have tested the effects of 6-hydroxydopamine lesions in animals bearing subthalamic nucleus lesions. Previous evidence [C. Baunez and T. W. Robbins, (1997) Eur. J. Neurosci. 9, 2086-2099] has revealed multiple deficits after bilateral lesions of the subthalamic nucleus. The present study replicated these effects. In combination with subthalamic nucleus lesions, striatal dopamine loss antagonised the increase in premature responding but did not counteract any of the other impairments. These results show the involvement of the dopaminergic nigrostriatal pathway in motor attention and arousal. Furthermore, they underline the independence of subthalamic nucleus lesion-induced effects from dopaminergic systems.

Effects of transient inactivation of the subthalamic nucleus by local muscimol and APV infusions on performance on the five-choice serial reaction time task in rats.

Within the basal ganglia circuitry, recent conceptions of the subthalamic nucleus are that it fulfils integrative functions. We have previously shown that bilateral excitotoxic lesions of the subthalamic nucleus induce behavioural deficits in a five-choice serial reaction time task in the rat, consistent with attentional impairments and suggesting important roles of this basal ganglia structure in mechanisms of behavioural control. In the present study, we tested the effects of (i) blocking its excitatory inputs (originating mainly in the cerebral cortex and the parafascicular nucleus of the thalamus) via the NMDA receptors and (ii) stimulating its GABA receptors to mimick the influence of its inhibitory inputs (mainly from the globus pallidus). Bilateral microinfusions of APV (NMDA receptor antagonist) or muscimol (GABA-A receptor agonist) into the subthalamic nucleus were administered to rats trained in the same five-choice serial reaction time task. Both APV (0.125-0.5 microg) and muscimol (1-3 ng) reduced choice accuracy, slowed correct responses and increased omissions and perseverative responses. Premature responses tended to increase after APV but decrease after muscimol. Increased perseverations at the food magazine occurred only after muscimol infusions. These results reproduce many of the effects of lesions of the STN and are consistent with an integrative role for this structure in pallidal and thalamo-cortical processing.

Impaired performance in a conditioned reaction time task after thermocoagulatory lesions of the fronto-parietal cortex in rats.

The present study examined whether cortical damage in rats may disrupt the integrative processes and motor control involved in the performance of a reaction time (RT) task. To investigate the nature of the deficits in the conditioned task, rats were subjected, after learning, to a coagulation of pia brain surface of varying extent, including the frontal and parietal cortical areas. They were then tested daily for over one month. The behavioural task required the rats to hold a lever down during a variable and random delay and react quickly to the onset of a visual cue by releasing the lever within a RT limit for food reinforcement. Extensive bilateral cortical lesions had no effect on spontaneous motor activity, but severely impaired RT performance. Latencies to release the lever after the cue were dramatically increased during the first postoperative sessions and gradually returned to baseline levels within 3 weeks, whereas less dramatic but long-lasting increase in premature responding (anticipatory response before the visual cue) was observed throughout the testing sessions. More restricted lesions to the frontoparietal cortex produced a similar pattern of incorrect responding with a faster recovery of delayed responses and a strong deficit in premature responding. The major effects of lesions confined to the rostral pole of the frontal cortex were observed on premature responding, however. The present results demonstrate that the impairment in movement initiation is rapidly recovered within 2-3 weeks even after extensive thermocoagulatory lesions of the frontal and parietal areas. This recovery suggests the involvement of adaptive processes developing progressively and probably reflecting the remarkable synaptic plasticity of the extrapyramidal motor output. In contrast, the long-lasting increase in premature responding, supposed to reflect some attentional deficits, may produce anatomofunctional long-term disorganization of subcortical structures such as the basal ganglia. Interestingly enough, these results show that the rat neocortex supports functions very similar to those of primates and provide a good model for studying these higher functions in operant motor procedures that require prior associative learning and appropriate motor coordination.

Bilateral lesions of the subthalamic nucleus induce multiple deficits in an attentional task in rats.

Lesioning the subthalamic nucleus (STN) has been suggested as possible therapy for the treatment of parkinsonism. Previous experiments investigating this hypothesis in rats confirmed that excitotoxic STN lesions alleviate the motor impairment induced by striatal dopamine depletion, which reproduced the degeneration observed in parkinsonism, but elicited presumed non-motor deficits such as premature responding, suggesting that the STN could be involved in other aspects of response control. The aim of the present study was to extend this analysis to choice paradigms. We thus investigated the behavioural effects of bilateral excitotoxic lesions of the STN in rats performing a five-choice test of divided and sustained visual attention, modelled on the human continuous performance task. This task required the animals to detect a brief visual stimulus presented in one of five possible locations and respond by a nose-poke in this illuminated hole within a fixed delay, for food reinforcement. Bilateral lesions of the STN severely impaired several aspects of performance, including discriminative accuracy, but also increased premature, anticipatory responding as well as perseverative panel pushes and nose-poke responses. While increasing the stimulus duration and reducing the waiting period for the stimulus partially alleviated the accuracy deficit and the premature responding deficit respectively, other deficits, such as perseverative panel pushes and nose-poke responses, were sustained under these conditions. Systemic injection of the mixed dopaminergic D1/D2 receptor antagonist, alpha-flupenthixol (0.03-0.18 mg/kg), reduced premature responses and perseverative panel pushing without affecting the perseverative nose-poke responses, suggesting that some of the deficits were independent of striatal dopaminergic transmission. These results suggest that STN lesions have multiple, dissociable effects on attentional performance, including discriminative deficits, impulsivity and perseverative behaviour. They are consistent in part with a hypothesized role of the STN in recent models of basal ganglia function in action selection and inhibition. The results also show that other aspects of behaviour should be monitored when examining the capacity of STN lesions to reverse the parkinsonian deficit induced by striatal dopamine depletion.

Evidence for functional differences between entopeduncular nucleus and substantia nigra: effects of APV (DL-2-amino-5-phosphonovaleric acid) microinfusion on reaction time performance in the rat.

Overactivity of the excitatory amino acid outputs of the subthalamic nucleus (STN) has recently been found to be one of the cascade of subsequent disruptions caused by nigrostriatal dopaminergic degeneration in Parkinson's disease. The respective contribution of the excitatory glutamatergic output structures of the STN [i.e. the globus pallidus (GP), entopeduncular nucleus (EP) and substantia nigra pars reticulata (SNr)] to the control of movement is not known, however. To investigate further the function of glutamatergic transmission through NMDA receptor subtypes in these three structures, the effects of discrete local infusion of a competitive receptor antagonist, DL-2-amino-5-phosphonovaleric acid (APV), into the EP, GP and SNr were tested in rats performing a reaction time task. Bilateral infusion of APV into the different output structures of the STN differentially impaired the performance of rats trained to release a lever after the onset of a visual stimulus within a time limit to obtain a food reward. Infusion of APV (0.25 and 0.5 microgram/0.5 microliter) into the SNr was found to induce behavioural deficits characterized by a dramatic increase in the number of premature lever releases and decreased mean reaction time. In contrast, the infusion of APV at a dose of 0.25 microgram into the GP or EP was found to induce a motor initiation deficit characterized by an increased number of delayed responses (lever release after the time limit) and increased mean reaction time. At a dose of 0.5 microgram, a premature responding deficit was added to the previous motor impairment. Interestingly, when APV was infused simultaneously into the GP and SNr in the same animals, the behavioural effects tended to be similar to those observed after a single infusion into the SNr. Altogether, these results reveal that the different functional weight of the three main output pathways originating at the STN level is t.o. The behavioural deficits induced by NMDA receptor blockade in the SNr were similar to those observed previously after a neurotoxic lesion of the STN, suggesting that NMDA receptors in this structure play a major role as a functional output of the STN. Furthermore, regarding the differential effects produced by the same dose of APV in the SNr and the EP, these two structures, which are classically believed to be functionally linked should not be considered as the same functional entity in the organization of basal ganglia outflow.

In a rat model of parkinsonism, lesions of the subthalamic nucleus reverse increases of reaction time but induce a dramatic premature responding deficit.

Lesions of the subthalamic nucleus (STN) have been found to reduce the severe akinetic motor symptom produced in animal models of Parkinson's disease, such as in monkeys treated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) or in monoamine-depleted rats. However, little is known about the effect of STN exclusion on subtle motor deficits induced by moderate dopaminergic lesions in complex motor tasks. The present study was thus performed on rats trained in a reaction time (RT) task known to be extremely sensitive to variations of dopamine transmission in the striatum. Animals were trained to release a lever after the onset of a visual stimulus within a time limit to obtain a food reward. Discrete dopamine depletion produced by infusing the neurotoxin 6-hydroxydopamine (6-OHDA) bilaterally into the dorsal part of the striatum, produced motor initiation deficits which were revealed by an increase in the number of delayed responses (lever release after the time limit) and a lengthening of RTs. In contrast, bilateral excitotoxic lesion of the STN with ibotenic acid induced severe behavioral deficits which were opposite to those produced by the dopaminergic lesion, as shown by an increase in the number of premature responses (lever release before the onset of the visual stimulus) and a decrease of RTs. Surprisingly, the performance of the animals bearing a double lesion (striatal dopaminergic lesion followed 14 d later by STN ibotenic lesion) was still impaired 40 d after the ibotenic lesion. As expected, the 6-OHDA-induced motor initiation deficits were reversed by a subsequent STN lesion. However, the dramatic increase of premature responses contributing to major behavioral deficits induced by the STN lesion remained unchanged. Thus, the bilateral lesion of the STN was found to alleviate the motor deficits in this model of parkinsonism, but essentially produced over time, long lasting deficits that might be related to dyskinesia or cognitive impairment. The present results strongly support the recent concept of a predominant control of the STN on basal ganglia output structures.

Does the blockade of excitatory amino acid transmission in the basal ganglia simply reverse reaction time deficits induced by dopamine inactivation?

It has recently been hypothesized that excessive excitatory amino acid (EAA) activity in the corticostriatal pathway and in the subthalamic nucleus could account for the expression of the motor deficits resulting from alteration in dopamine function in the basal ganglia. The present study investigated the potential benefit of blocking excitatory amino acid transmission in the basal ganglia, subsequent to the inactivation of dopaminergic function of rats performing a reaction time (RT) task. Disruption of dopamine activity by the neurotoxin 6-hydroxydopamine (6-OHDA) injected in the striatum or by systemic administration of the D2 dopamine receptor antagonist raclopride, impaired the performance of rats trained to release a lever quickly after a visual stimulus. RTs, measured by the time elapsing from the stimulus onset to the lever release, were lengthened after both treatments. The blockade of EAA transmission at the N-methyl-D-aspartate (NMDA) receptor, by systemic injections of the NMDA receptor antagonist dizocilpine or by excitotoxic lesions of the subthalamic nucleus, in animals with dopamine lesions, significantly reversed the increase of RTs. Performance of animals with subthalamic nucleus lesions did not return to pre-operative values, however. The blockade of NMDA receptors in the striatum, by a local injection of the NMDA receptor antagonist DL-2-amino-5-phosphonovaleric acid (APV), in animals treated with raclopride, was found to decrease RTs and improve performance. Analysis of RT distributions in the three groups of animals revealed that blocking EAA activity with NMDA receptor antagonists improved performance by shifting RTs back towards baseline values, preserving a normal distribution. In contrast, lesions of the subthalamic nucleus disrupted performance, as shown by the scattered distribution of RTs. The results indicate that treatment with NMDA receptor antagonists but not subthalamotomy provides a possible beneficial treatment in the present model of Parkinsonism.

Dopamine and complex sensorimotor integration: further studies in a conditioned motor task in the rat.

Rats were trained to depress a lever and wait for the onset of a light stimulus, occurring after four equiprobable and variable intervals. At the stimulus onset, they had to release the lever within a reaction time limit for food reinforcement. This paradigm required time estimation of the various intervals and high attentional load for correct performance. Following activation of the dopaminergic transmission after systemic injection of d-amphetamine (0.6 and 0.8 mg/kg) or intrastriatal injection of dopamine (2.5 microgram/microliters), the rat's performance was impaired. Compared with control animals, the performance deficits were expressed as an increased number of premature lever releases before the conditional stimulus onset ("premature responses") and decreased reaction times. Indeed, the reaction times distribution was shifted to the left towards shortened reaction times. Although the number of premature responses was increased, the time estimation of the four different equiprobable intervals was not disturbed after stimulation of dopaminergic activity. A delay-dependent shortening of reaction times as a result of the conditional probability of the stimulus occurrence (i.e. reaction times are shorter as the duration of the delay increases) was found in control and drug sessions, indicating that the animals were still able to prepare their motor response (lever release) even after overstimulation of the dopaminergic transmission. In contrast, blocking dopamine receptors with the selective D2 antagonist raclopride was found to induce opposite effects on the reaction time performance. The number of delayed responses (i.e. occurring with a latency > 600 ms) was found to be significantly enhanced. Furthermore, the reaction times distribution showed a shift of the values to the right revealing a general tendency to lengthened reaction times. These results indicate that a "critical level" of dopamine activity (neither too low nor too high) in the striatum is necessary for a correct execution of the movement in a conditioned motor task with temporal constraint. Moreover, while delayed responses might reflect a motor impairment, anticipatory responses might reflect a "motor facilitation" revealed by a higher level of motor readiness, without disturbing time estimation nor attentional processes.

Functional interactions between glutamate and dopamine in the rat striatum.

The functional role of NMDA receptors in a spontaneous (locomotion) and a conditioned behaviour (reaction-time task) known to preferentially involve dopamine transmission in the ventral or the dorsal part of the striatum, respectively, was studied in the rat. The non-competitive NMDA receptor antagonist MK-801 systemically injected produced a dose-dependent increase in locomotor activity and impaired the performance of the animals trained to release a lever after a visual stimulus within a time limit by increasing the number of anticipatory errors (lever releases occurring before the stimulus onset). Similar behavioural changes were obtained after bilateral striatal microinjections of the competitive NMDA-antagonist APV into the ventral or dorsal striatum, respectively, suggesting that MK-801-induced behavioral effects after systemic injection might be mediated through a blockade of EAA transmission within the striatum. Dopamine injected in the same striatal locations induced effects similar to APV on locomotion and reaction-time performance, in agreement with the proposal for a functional antagonism between the glutamatergic and the dopaminergic transmission at striatal level. The conjoint administration of APV and dopamine directly into the striatum did not alter the behavioural effect induced by each compound injected alone showing that these effects are not additive. This latter observation actually suggests the occurrence of a functional interaction between the two neuronal systems probably acting on a common striatal target relaying dopaminergic and glutamatergic antagonistic influences on locomotion and conditioned motor behaviours.

N-methyl-D-aspartate receptor blockade impairs behavioural performance of rats in a reaction time task: new evidence for glutamatergic-dopaminergic interactions in the striatum.

The effects of blocking glutamate transmission at the N-methyl-D-aspartate receptor subtype were studied in rats performing a conditioned reaction time motor task. Rats were trained to release a lever after the onset of a visual stimulus within a time limit to obtain food reward. The results showed that the performances of the groups receiving the N-methyl-D-aspartate receptor antagonists dizocilpine maleate (0.1 mg/kg) injected systemically or DL-2-amino-5-phosphonovaleric acid at the highest dose tested (5.0 micrograms/microliter/side) injected locally into the striatum changed significantly as compared to controls. The effects of these antagonists, consisting of an increase in the number of lever releases occurring before the visual stimulus onset ("anticipated responses"), were similar to those induced by injecting dopamine into the same striatal location. Both dizocilpine maleate and DL-2-amino-5-phosphonovaleric acid (5.0 micrograms/microliter) reversed the motor deficits, resulting in an increase in the number of lever releases after the time limit ("delayed responses") that were induced by the D2 dopamine receptor antagonist raclopride. Although these results partly confirm the existence of a functional antagonism between the glutamatergic and the dopaminergic systems in the striatum, opposite findings were obtained with the group that received intrastriatal DL-2-amino-5-phosphonovaleric acid at the lowest dose (0.5 micrograms/microliter/side). When given alone, 0.5 micrograms/microliter DL-2-amino-5-phosphonovaleric acid had no behavioural effects, but when jointly administered with dopamine or raclopride, it was found to reverse the effects of dopamine and to potentiate the motor deficits induced by raclopride. These opposite effects on the reaction time task observed after the intrastriatal injection of DL-2-amino-5-phosphonovaleric acid, depending on the dose tested, occurred only after a combined treatment with a dopaminergic agonist or antagonist and suggest that the level of the striatal dopaminergic activity may play a critical role in regulating the glutamate transmission via the N-methyl-D-aspartate receptors during the performance of complex sensorimotor tasks of this kind.