Altered function of glutamatergic cortico-striatal synapses causes output pathway abnormalities in a chronic model of parkinsonism.

OBJECTIVE: In Parkinson's disease, chronic striatal dopamine depletion results in over-activity and under-activity of the indirect and direct striatal output pathways respectively. In this study, we investigated changes in the function of glutamatergic cortico-striatal synapses that contribute to abnormalities in striatal efferents. METHODS: Whole-cell recordings were performed in striatal slices prepared from adult bacterial artificial chromosome mice, chronically lesioned with 6-hydroxydopamine (6-OHDA). Paired pulse facilitation, spontaneous synaptic activity, the ratio of AMPAR to NMDAR-mediated components of excitatory postsynaptic currents, AMPAR and NMDAR kinetics, current-voltage relationship and intrinsic membrane properties were assessed in indirect and direct pathway medium spiny neurons (MSNs), which were identified on the basis of expression of GFP, driven by the promoters of A2A or D1 receptor expression. The trajectory of striatal efferents, with respect to selective targeting of the globus pallidus and substantia nigra was also compared in sham-operated versus 6-OHDA-lesioned mice. RESULTS: Dopamine depletion did not affect the number of pathway specific output neurons or the trajectory of striatal outputs. In sham-operated animals, cortico-striatal synapses of both striatal efferent populations exhibited paired pulse facilitation and similar ratios of AMPAR to NMDAR-mediated components of excitatory postsynaptic currents. Following striatal dopamine depletion, indirect pathway neurons exhibited decreased levels of paired pulse facilitation, enhanced sensitivity to presynaptic stimulation and an increase in the relative contribution of NMDAR to the EPSC but no change in spontaneous synaptic activity. In sham-operated mice, neurons of the direct pathway exhibited lower firing frequency compared to the indirect pathway following current injection. However, in 6-OHDA-lesioned mice, in the direct pathway, firing threshold was reduced, spike frequency adaptation developed and the frequency of spontaneous activity was also reduced. In addition, changes in the properties of NMDAR kinetics suggest that these receptors were desensitised. DISCUSSION: Increased synchronicity between pre and postsynaptic neurons, as indicated by decreased paired pulse facilitation, and increased sensitivity to extracellular stimulation, combined with an increase in the contribution of NMDARs to the EPSC at cortico-striatal synapses, may contribute to the over-activity of indirect pathway neurons in the parkinsonian striatum. In contrast, a decrease in spontaneous activity, postsynaptic desensitisation to excitatory stimuli and spike frequency adaptation of cortico-striatal synapses may underlie under-activity of the direct pathway.

Generation of a model of L-DOPA-induced dyskinesia in two different mouse strains.

Prolonged use of the dopamine precursor L-DOPA for the treatment of Parkinson's disease commonly results in abnormal involuntary movements, which are termed L-DOPA-induced dyskinesia (LID). Over-activity at corticostriatal synapses onto neurons of the direct and indirect striatal output pathways has been implicated in the development of dyskinesia, but it has proved difficult to investigate the pathways separately due to their morphological similarities. The recent development of bacterial artificial chromosome mice that express green fluorescent protein in either the direct or indirect pathway allows visual identification of the output neurons in each pathway. Here we describe the use of two different strains of these transgenic mice (pure FVB and FVB crossed with C57BL6) in the development of mouse models of L-DOPA-induced dyskinesia. This model will allow the direct and indirect pathways to be studied individually to delineate the cellular and molecular mechanisms underlying dyskinesias. These studies demonstrate that mouse strain impacts on behavioural responses and L-DOPA sensitivity. Therefore, when generating mouse models of LID, strain must be taken into consideration when choosing the L-DOPA dosing regimen.