The technique of deep brain stimulation (DBS) has become a preferred surgical choice for the treatment of advanced Parkinson's disease. The subthalamic nucleus (STN) is presently the most promising target for such DBS. In this study, whole-cell patch-clamp recordings were made from 46 STN neurons in rat brain slices to examine the effect of high-frequency stimulation (HFS) of the STN on glutamatergic synaptic transmission in STN neurons. HFS, consisting of trains of stimuli at a frequency of 100 Hz for 1 min, produced three types of synaptic plasticity in 17 STN neurons. First, HFS of the STN induced short-term potentiation (STP) of evoked postsynaptic current (EPSC) amplitude in four neurons. STP was associated with a reduction in the EPSC paired-pulse ratio, suggesting a presynaptic site of action. Second, HFS of the STN generated long-term potentiation (LTP) of EPSC amplitude in eight neurons. Although the EPSC paired-pulse ratio was reduced transiently in the first 2 min following HFS, ratios measured 6-20 min after HFS were unchanged from control. This suggests that LTP is maintained by a postsynaptic mechanism. Third, HFS produced long-term depression (LTD) of EPSC amplitude in five STN neurons. LTD was associated with a significant increase in EPSC paired-pulse ratios, indicating a presynaptic site of action. These results suggest that HFS can produce long-term changes in the efficacy of synaptic transmission in the STN. HFS-induced synaptic plasticity might be one mechanism underlying the effectiveness of DBS in the STN as a treatment of advanced Parkinson's disease.

We used whole-cell patch recordings in current clamp to investigate the ionic dependence of burst firing induced by N-methyl-d-aspartate (NMDA) in neurons of the subthalamic nucleus (STN) in slices of rat brain. NMDA (20 microm) converted single-spike firing to burst firing in 87% of STN neurons tested. NMDA-induced bursting was blocked by AP5 (50 microm), and was not mimicked by the non-NMDA receptor agonist AMPA (0.6 microm). Tetrodotoxin (1 microm) converted bursts to oscillations of membrane potential, which were most robust when oscillations ranged between -50 and -70 mV. The NMDA bursts were blocked by an elevated extracellular concentration of Mg(2+), but superfusate containing no added Mg(2+) either reduced or increased burst firing, depending upon the amount of intracellular current injection. Block of K(+) conductances by apamin and tetraethylammonium prolonged burst duration, but iberiotoxin had no effect. NMDA-induced burst firing and membrane oscillations were completely blocked by superfusate containing no added Ca(2+), and they were significantly reduced when patch pipettes contained BAPTA. Selective antagonists for T-type (mibefradil, 10 microm), L-type (nifedipine, 3 microm), and N-type (omega-conotoxin GVIA, 1 micro m) Ca(2+) channels had no effect on NMDA burst firing. Superfusate containing a low concentration of Na(+) (20 mm) completely abolished NMDA-induced burst firing. Flufenamic acid (10 microm), which blocks current mediated by Ca(2+)-activated nonselective cation channels (I(CAN)), reversibly abolished NMDA-depended bursting. These results are consistent with the hypothesis that NMDA-induced burst firing in STN neurons requires activation of either an I(CAN) or a Na(+)-Ca(2+) exchanger.

No abstract

Information processing in the brain requires adequate background neuronal activity. As Parkinson's disease progresses, patients typically become akinetic; the death of dopaminergic neurons leads to a dopamine-depleted state, which disrupts information processing related to movement in a brain area called the basal ganglia. Using agonists of dopamine receptors in the D1 and D2 families on rat brain slices, we show that dopamine receptors in these two families govern the firing pattern of neurons in the subthalamic nucleus, a crucial part of the basal ganglia. We propose a conceptual frame, based on specific properties of dopamine receptors, to account for the dominance of different background firing patterns in normal and dopamine-depleted states.

The subthalamic nucleus (STN) receives direct dopaminergic innervation from the substantia nigra pars compacta, but the importance of this input in the pathophysiology of parkinsonism remains to be determined. We used whole-cell patch-clamp recordings in brain slices to study presynaptic dopaminergic modulation of synaptic inputs to the STN in unilateral 6-hydroxydopamine (6-OHDA)-lesioned rats. Here, we report that dopamine was more potent for inhibiting GABA IPSCs and glutamate EPSCs in the STN ipsilateral to the lesion, and was less potent for suppressing IPSCs and EPSCs in the STN contralateral to the lesion, compared with the effects of dopamine in control STN. Dopamine reduced IPSCs with an IC50 value of 20.9 +/- 3.6 microM in control STN, whereas IC50 values were 0.83 +/- 0.15 and 55.1 +/- 11.1 microM in STN ipsilateral and contralateral to 6-OHDA lesions, respectively. Dopamine also inhibited EPSCs with an IC50 value of 12.8 +/- 2.8 microM in control STN, whereas IC50 values were 4.5 +/- 0.9 and 41.6 +/- 9.8 microM in STN ipsilateral and contralateral to 6-OHDA lesions, respectively. Results with paired stimuli to evoke EPSCs and IPSCs suggest that endogenous dopamine acts presynaptically to inhibit transmitter release in the STN. These results show that chronic dopamine denervation significantly alters the regulation of synaptic input to the STN. Our results also suggest that the STN may be an important target for levodopa therapy in Parkinson's disease.

No abstract

Tyramine, an endogenous ligand for mammalian trace amine-associated receptors, may act as a neuromodulator that regulates neuronal activity in basal ganglia. Using whole-cell patch recordings of subthalamic nucleus (STN) neurons in rat brain slices, we found that bath application of tyramine evoked an inward current in voltage-clamp in over 60% of all STN neurons. The inward current induced by tyramine was mimicked by the D 2 -like dopamine receptor agonist quinpirole, but was only partially blocked by the D 2 -like receptor antagonist sulpiride. In contrast, the D 1 -like receptor agonist SKF38393 evoked no current in STN neurons. Inward current evoked by tyramine was significantly reduced by the catecholamine uptake inhibitor nomifensine, and by exhausting catecholamines in the brain via pre-treatment with reserpine. Tyramine also reduced the amplitude of GABA A receptor-mediated IPSCs that were evoked by focal electrical stimulation of the slice. Inhibition of IPSCs by tyramine was mimicked by quinpirole and was blocked by sulpiride but not by SCH23390, a D 1 receptor antagonist. Moreover, tyramine-induced inhibition of IPSCs was reduced in slices pre-treated with reserpine, and this inhibition could be restored by briefly superfusing the slice with dopamine. These results suggest that tyramine acts as an indirect dopamine agonist in the STN. Although inhibition of IPSCs are mediated by D 2 -like receptors, the dopaminedependent inward currents evoked by tyramine do not fit a typical dopamine receptor pharmacological profile.