Stimulus Response of Firing Properties and Synaptic Plasticity in Subthalamic and Nigral Neurons of Parkinson's Disease Patients

Milosevic, Luka 1 ; Kalia, Suneil 2 ; Hodaie, Mojgan 2 ; Lozano, Andres 2 ; Hutchison, William 3 ; Popovic, Milos 1  

1. Institute of Biomaterials and Biomedical Engineering, University of Toronto; 2. Department of Neurosurgery, Toronto Western Hospital; 3. Department of Physiology, University of Toronto

Background: Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is an effective procedure for the symptomatic treatment of Parkinson’s disease (PD). The therapeutic benefits of DBS are frequency-dependent, but the underlying physiological mechanisms remain unclear. We previously reported short-term plasticity changes in substantia nigra pars reticulata (SNr) with short trains of high frequency stimulation (HFS), but long-trains have not been investigated.

Objectives:  (i) Compare the frequency-dependent effects on cell firing in STN and SNr neurons, (ii) quantify the frequency-dependent dynamics of short-term plasticity in SNr, and (iii) deliver continuous long-train HFS and compare the effects on short-term potentiation to our previous study.

Methods: In patients undergoing stereotactic DBS surgery for PD, two closely spaced (600um) microelectrodes were advanced into the STN and SNr. One microelectrode recorded individual action potential firing and evoked field potentials (fEPs) during stimulation trains of different frequencies (1Hz for 10s up to 100Hz for 0.5s) from the adjacent microelectrode.

Results: STN neuronal firing showed significant attenuation with 20Hz (p<0.01) stimulation and greater (silenced at 100Hz), while SNr decreased with 3Hz (p<0.05) and greater (silenced at 50Hz). Both structures showed increasing transient silent periods following stimulation. The average peak amplitude of the fEP in SNr neurons was attenuated above 30Hz (p<0.05). However, the first-fEP within the train was potentiated above 30Hz (p<0.01). This is suggestive of synaptic facilitation, followed by rapid synaptic depression. Furthermore, the average fEP amplitude during 1Hz pulses showed significant inhibitory synaptic potentiation after continuous HFS (100Hz, 10s). The amplitude of the fEP increased by a factor of 1.72 (p<0.001), while the time delay between the stimulation pulse and first spike increased by a factor of 1.88 (p<0.01).

Conclusions: STN neurons exhibited a higher frequency threshold to electrical stimulation either due to a differing ratio of GABA:glutamate terminals on the soma compared to SNr, and/or the nature of their GABAergic inputs (pallidal vs. striatal). Nevertheless, this supports the hypothesis that HFS produces predominantly GABA release from afferent synaptic terminals resulting in a reduction of neuronal firing through excitation of pre-synaptic axon fibers. We also showed enhancement of inhibitory synaptic plasticity in SNr by continuous HFS, and the frequency-dependent dynamics of short-term synaptic plasticity (which are believed to be modulated by neurotransmitter release properties) and consider these to be additional putative mechanisms of DBS.