Motor Thalamic Deep Brain Stimulation Alters Cortical Activity and Shows Therapeutic Utility for Treatment of Parkinson’s Disease Symptoms in a Rat Model

Highlights

• VA-VL DBS increased spike frequencies of primary and secondary motor cortical pyramidal cells in 6-OHDA lesioned rats.

• VA-VL DBS increased spike frequencies of secondary motor cortical interneurons in 6-OHDA lesioned rats.

• VA-VL and STN DBS had comparable therapeutic windows and induced dyskinesia at similar rates in 6-OHDA lesioned rats.

• VA-VL and STN DBS adversely affected memory in cognitively normal and mildly impaired sham rats.

• VA-VL DBS improved memory in a subset of sham rats exhibiting more severe cognitive deficits.

Abstract

Deep brain stimulation (DBS) in Parkinson’s disease (PD) alters neuronal function and network communication to improve motor symptoms. The subthalamic nucleus (STN) is the most common DBS target for PD, but some patients experience adverse effects on memory and cognition. Previously, we reported that DBS of the ventral anterior (VA) and ventrolateral (VL) nuclei of the thalamus and at the interface between the two (VA|VL), collectively VA-VL, relieved forelimb akinesia in the hemiparkinsonian 6-hydroxydopamine (6-OHDA) rat model. To determine the mechanism(s) underlying VA-VL DBS efficacy, we examined how motor cortical neurons respond to VA-VL DBS using single-unit recording electrodes in anesthetized 6-OHDA lesioned rats. VA-VL DBS increased spike frequencies of primary (M1) and secondary (M2) motor cortical pyramidal cells and M2, but not M1, interneurons. To explore the translational merits of VA-VL DBS, we compared the therapeutic window, rate of stimulation-induced dyskinesia onset, and effects on memory between VA-VL and STN DBS. VA-VL and STN DBS had comparable therapeutic windows, induced dyskinesia at similar rates in hemiparkinsonian rats, and adversely affected performance in the novel object recognition (NOR) test in cognitively normal and mildly impaired sham animals. Interestingly, a subset of sham rats with VA-VL implants showed severe cognitive deficits with DBS off. VA-VL DBS improved NOR test performance in these animals. We conclude that VA-VL DBS may exert its therapeutic effects by increasing pyramidal cell activity in the motor cortex and interneuron activity in the M2, with plausible potential to improve memory in PD.

Introduction

Parkinson’s Disease (PD) affects 1% of people age 60 or older (Tysnes and Storstein, 2017). Males are diagnosed with PD at rates 1.37 to 3.7 times that of females, making male sex the greatest risk factor after aging (Gillies et al., 2014). The two key pathophysiological correlates of PD are expression of protein aggregates known as Lewy bodies containing alpha synuclein and loss of 50–70% of substantia nigra pars compacta (SNc) dopamine (DA) neurons at diagnosis (Chen et al., 2013).

The motor symptoms of PD include bradykinesia, muscle rigidity, resting tremor, postural instability, gait dysfunction and reduced facial expression (Postuma et al., 2015). The non-motor symptoms of PD include mood disorders such as anxiety and depression, sleep disorders, orthostatic hypotension, constipation, hyposmia, fatigue, pain, urinary problems such as urgency and incontinence, sexual dysfunction, hallucinations, and delusions (Stacy, 2011). Related to our study, one fifth of PD patients experience cognitive impairment without dementia (CIND) at diagnosis and an increased likelihood of memory problems (Watson and Leverenz, 2010). These symptoms can progress to dementia and psychosis as the disease affects the cortex (Stacy, 2011). Impaired memory and cognitive decline in PD patients is associated with hippocampal atrophy (reviewed in Krajcovicova et al., 2019) and is more severe in patients with akinetic-rigid than tremor dominant PD (Wojtala et al., 2019). While deep brain stimulation (DBS) has been effective for patients with CIND, DBS is contraindicated for patients with a diagnosis of dementia (Hartmann et al., 2019).

Levodopa (L-DOPA) is a pharmacological replacement for the loss of SNc DA efferents to the basal ganglia (BG) and is the first-line treatment for PD motor symptoms. Levodopa-induced dyskinesia (LID) is experienced by 90% of PD patients after 10 years of treatment with the drug (Ahlskog and Muenter, 2001). These sometimes disabling abnormal movements can make it more difficult to manage drug titration and motor symptoms of PD, and diminish quality of life (Fabbrini et al., 2007). Interestingly, while subthalamic nucleus (STN) DBS allows for a greater reduction in levodopa equivalent dose (LED), globus pallidus pars interna (GPi) DBS produces a more pronounced improvement in dyskinesia than STN DBS (Ryu et al., 2017, Liu et al., 2019). Cognitive and memory decline is observed more frequently in PD patients receiving STN DBS than those receiving GPi DBS or those that did not undergo surgery (Hariz et al., 2008, Williams et al., 2011, Cernera et al., 2019), possibly owing to the associative and limbic connectivity of the STN (Parent and Hazrati, 1995, Temel et al., 2005, Arnold Anteraper et al., 2018). A meta-analysis found the adverse effects of STN DBS on human cognition included mildly significant declines in executive function and in verbal learning and memory, with moderately significant declines in verbal fluency (Parsons et al., 2006). Still, bilateral STN DBS remains the prevailing approach for treatment of PD motor symptoms (Williams et al., 2014).

In our previous study, we explored the efficacy of motor thalamic DBS for the treatment of akinesia, partially based on a report that optogenetic stimulation of the ventral anterior nucleus of the thalamus (VA) improved haloperidol-induced PD akinesia in rats [5]. We found that electrical stimulation at mode settings of 10 Hz, 100 µs. and 100 µA at the VA, ventrolateral (VL) and at the interface between the two (VA|VL) similarly improved forelimb akinesia in a 6-hydroxydopamine (6-OHDA) rat model of PD (Tucker et al., 2019). These three regions were collectively referred to as the VA-VL, due to their similar therapeutic effectiveness (Tucker et al., 2019). In vitro, VA|VL stimulation increased induced action potential frequencies in both VA and VL neurons and caused VL neurons to transition from bursting to single spike action potentials (Tucker et al., 2019). In vivo, VA|VL DBS reduced beta and gamma oscillations in the primary motor cortex (M1) (Tucker et al., 2019). Notably, the impact of VA-VL DBS on spike frequencies in the motor cortices was not explored, which could reveal mechanisms underlying its efficacy. Additionally, the translational viability of VA-VL DBS would be revealed by comparing its therapeutic window and cognitive effects to those of standard STN DBS.

We addressed these unanswered questions here by examining how VA-VL DBS affects average spike frequencies in the primary and secondary (M2) motor cortices of anesthetized hemiparkinsonian rats. We also compared therapeutic windows and onset rate of stimulation-induced dyskinesia in hemiparkinsonian rats, and cognitive effects in sham rats receiving VA-VL DBS to those receiving STN DBS.