Deep brain stimulation does not enhance neuroinflammation in multiple system atrophy

Highlights

• Deep brain stimulation did not enhance neuroinflammation in MSA patients.

• The number of microglia and CD3 lymphocytes was highest in the putamen.

• The proportion of neural inclusions in the putamen was higher in the operated group.

Abstract

Slowly progressive, levodopa-responsive multiple system atrophy (MSA) may be misdiagnosed as Parkinson's disease (PD). Deep brain stimulation (DBS) is mostly ineffective in these patients and may even worsen the clinical course. Here we assessed whether neuropathological differences between patients with MSA who were treated with DBS of the subthalamic nucleus because of a misleading clinical presentation and typical disease cases may explain the more benign disease course of the former, and also the rapid clinical decline after surgery. The post-mortem assessment included the subthalamic nucleus, the globus pallidus, the thalamus and the putamen in five patients with MSA who received DBS and nine typical disease cases. There was no evidence for distinct neuroinflammatory profiles between both groups that could be related to the surgical procedure or that could explain the rapid clinical progression during DBS. Patients who received deep brain stimulation displayed a higher proportion of α-synuclein bearing neuronal cytoplasmic inclusions in the putamen compared with typical cases, while the number of surviving neurons was not different between groups. Our findings suggest that DBS does not induce neuroinflammatory changes in patients with MSA, at least several years after the surgery. We further hypothesize that the peculiar pattern of α-synuclein pathology may contribute to differences in the clinical phenotype, with a greater proportion of neuronal inclusions in the putamen being associated to a milder, “PD-like” phenotype with sustained levodopa response and slower disease progression.

Introduction

Multiple system atrophy (MSA) is a fatal orphan neurodegenerative disorder that manifests in a variable combination with autonomic, parkinsonian, cerebellar, and pyramidal features (Fanciulli and Wenning, 2015; Gilman et al., 2008). The pathological hallmark is the accumulation of misfolded alpha-synuclein (α-syn) in oligodendrocytes, forming glial cytoplasmic inclusions (GCI), and to a lesser extent in neurons, typically as neuronal cytoplasmic inclusions (NCI) (Inoue et al., 1997; Kato and Nakamura, 1990; Lin et al., 2004; Nishie et al., 2004; Papp et al., 1989; Papp and Lantos, 1992). The median disease duration is 6–10 years (Gilman et al., 2008).

Current consensus diagnostic criteria include two classical phenotypes, MSA with predominant parkinsonism (MSA-P) and MSA with predominant cerebellar ataxia (MSA-C). Clinical heterogeneity has been repeatedly described in the literature, ranging from an asymmetrical, slowly progressive levodopa-responsive phenotype that may last up to two decades (Jellinger, 2012; Masui et al., 2012; Petrovic et al., 2012), to the very aggressive so-called ‘minimal change’ variant, which may lead to death within 5 years after having reached most of the clinical milestones by year 3 (Ling et al., 2015). Factors that have been associated with poor survival in MSA are older age at onset, female gender, MSA-P subtype, shorter period from onset to first clinical milestone, stridor within the first three years after symptom onset and early autonomic failure (Giannini et al., 2016; Low et al., 2015; O'Sullivan et al., 2008; Wenning et al., 2013).

In MSA, the neurodegenerative process is most pronounced in the substantia nigra (SN), putamen, globus pallidus (GP, particularly in MSA-P), vermis, cerebellar hemispheres and inferior olivary nucleus (particularly in MSA-C) (Fearnley and Lees, 1990; Wenning et al., 1997), while significant involvement of the thalamus, subthalamic nucleus (STN), cerebellar dentate nucleus and anterior horn cells is rare (Ozawa et al., 2004).

The relevance of recognizing the more benign slowly progressive, levodopa-responsive variant, lies in the fact that these patients can be misdiagnosed as having Parkinson's disease (PD) with significant levodopa-related motor complications and undergo deep brain stimulation (DBS) surgery. In this regard, we recently reported the poor clinical outcome of STN-DBS in a series of five post-mortem confirmed MSA patients. These patients were initially considered as having PD with motor fluctuations and underwent DBS surgery, followed soon after the operation by a rapid deterioration and the appearance of clinical features suggestive of MSA (Meissner et al., 2016).

The present study investigated whether neuropathological differences between these MSA-DBS cases and a cohort of typical MSA-P patients might account for the more benign disease course of the MSA-DBS cases, and also for their rapid clinical decline after DBS.