Activated Schwann cells and increased inflammatory cytokines IL‐1β, IL‐6, and TNF‐α in patients' sural nerve are lack of tight relationship with specific sensory disturbances in Parkinson's disease

Abstract Aims Neuroinflammation is one of the most important processes in the pathogenesis of Parkinson's disease (PD). Sensory disturbances are common in patients with PD, but the underlying pathophysiological mechanisms remain unknown. This study aimed to characterize the activation of Schwann cells (SCs) and the increase of expression of inflammatory cytokines IL‐1β, IL‐6, and TNF‐α in the sural nerve of PD, and further explore whether peripheral nerve inflammation is the cause of PD sensory disturbances. Methods A total of 14 patients with PD (including 5 with sensory disturbances and 9 without sensory disturbances) and 6 controls were included. The excitation and conduction function of sural nerve was detected by sural nerve electrophysiological examination. With sural nerve biopsy samples, ultrastructural changes of sural nerve were observed by electron microscopy; Schwann cell biomarker glial fibrillary acid protein (GFAP) and inflammatory cytokines including interleukin‐1beta (IL‐1β), interleukin 6 (IL‐6), and tumor necrosis factor‐alpha (TNF‐α) were detected by immunohistochemistry, and the outcome of immunostaining slice was semiquantitatively counted; double immunofluorescence was used to identify the locus immunoreactive for inflammatory cytokines. Results Compared with healthy controls, nerve conduction velocity (NCV) slowed down and sensory nerve action potential (SNAP) amplitude decreased in PD patients, accompanied by axonal degeneration and demyelinating lesions, and expression of GFAP and inflammatory cytokines was increased. Inflammatory cytokines were significantly colocalized with GFAP and slightly colocalized with NF. These indicators did not differ significantly between PD patients with and without sensory disturbances. Conclusion Our study results suggest that peripheral sensory nerve injury exists in PD patients, accompanied by Schwann cell activation and inflammation, thus demonstrate peripheral nerve inflammation participates in the pathophysiological process of PD but it is not necessarily related to the patient's sensory disturbance.


| INTRODUC TI ON
It is estimated that 30%-85% of patients with PD suffered from sensory disturbances such as pain, numbness and paresthesia. 1,2 It has been pointed out that central lesions such as degeneration of noradrenergic neurons in the locus coeruleus and hypofunction of striatal dopaminergic system may be the pathophysiological mechanism of PD sensory disturbances. 3 Large amount of evidence has suggested that neuroinflammation and oxidative stress might play a vital role in the pathophysiological process of PD. 4,5 Specifically, triggering factors such as environmental exposure or age-related conditions will lead to the activation of glia including microglia and astrocytes who serve as immune cells in central nervous system (CNS). 6 Subsequently, inflammatory cytokines such as interleukin 1-beta (IL-1β), interleukin 6 (IL-6), and tumor necrosis factor-alpha (TNF-α) will be secreted robustly as a "second hit" to cause more damage. [7][8][9] Although extensive studies have focused on the central mechanisms of sensory disturbances and inflammatory processes in CNS, it is unclear whether the peripheral nervous system (PNS) is associated with these processes by far.
SCs are the most important glia in PNS, serving as supporting and protecting axons. Besides, they could also be activated into immune competent cells after peripheral nerve is injured and then produce inflammatory cytokines such as IL-1β, IL-6, and TNF-α. 10 In the last few decades, studies have reported structural and functional abnormalities in peripheral nerve from PD patients: loss of unmyelinated nerve fibers in sural nerve, increased thermal and tactile thresholds, reduced pain perception, and epidermal denervation. 11,12 Intriguingly, in our previous study, phosphorylated α-synuclein (pSNCA) deposition, the classic pathological biomarker of PD, was found in Schwann cells (SCs) in sural nerve from PD patients using biopsy methods, demonstrating that SCs might be critically involved in peripheral nerve abnormalities in PD. 13 Therefore, in this study, we hypothesized that Schwann cells activation and inflammatory cytokine release in peripheral sensory nerves of PD patients, which may lead to their sensory disturbances. Medical University as controls who were suspicious of myopathy and accepted muscle and sural nerve biopsy, but were ultimately confirmed without neuropathy change. Also, these patients did not have a history of diabetes or other diseases that might damage the peripheral nerve, so these nerves were deemed as healthy controls. The demographic data for all participants were summarized in Table 1.

| Electrophysiology
Nerve conduction velocity (NCV) and sensory nerve action potentials (SNAP) amplitudes of bilateral sural nerve from all subjects were completed in our laboratory by conventional surface electrodes using six channels electromyogram (Keypoint, Dantec Corporation). The history and physical examination of PD patients were performed by the Conclusion: Our study results suggest that peripheral sensory nerve injury exists in PD patients, accompanied by Schwann cell activation and inflammation, thus demonstrate peripheral nerve inflammation participates in the pathophysiological process of PD but it is not necessarily related to the patient's sensory disturbance.

K E Y W O R D S
neuroinflammation, Parkinson's disease, peripheral nerve, schwann cells, sensory disturbance formally trained neurologist prior to the electrophysiological examination (All subjects had no obvious abnormalities in muscle strength and muscle tone and pathological reflex was negative.). Results of NCV and SNAP amplitudes were classified as normal or abnormal based on our own laboratory collected age-adjusted standardized database. In our study, when the age was ≥50 years, NCV reference range of sural nerve was >45.3 m/s, and SNAP amplitude reference range was >1.4 μV.

| Immunohistochemistry and immunofluorescence
To analyze the expression of glial fibrillary acid protein (GFAP,       Table 1), a 71-y-old female PD patient without sensory disturbances (E-H) (case PD patient 10 in Table 1), and a 63-y-old male control without neuropathy (I-L) (case Control 1 in Table 1

| The demographic and clinical characteristics
The clinical data were summarized in

| Electrophysiology
Electrophysiological results were shown in Table 2

| Immunohistochemistry and immunofluorescence
Sections were viewed by two pathologists who were blind to the information. With a quick view, controls had low reactivity with GFAP and three inflammatory factors IL-1β, IL-6, and TNF-α. In contrast, 14 PD samples showed increased expression of GFAP and inflammatory cytokines (Figure 1).
To test if the expression of cytokines were associated with sensory disturbances, we set aside controls and analyzed the results of PD patients alone. We divided PD patients into two groups according to whether they had sensory symptoms or not and com-

F I G U R E 3
Double-staining immunofluorescence for cytokines with GFAP and NF in sural nerve from a PD patient (case PD patient 10 in Table 1). A, Immunostaining for GFAP with TNF-α, IL-1β, and IL-6. Partial cytokines was colocalized with Schwann cells as indicated by thin arrows in A, and the rest cytokines were surrounded by the colocalization as indicated by thick arrows. B, Immunostaining for NF with TNF-α, IL-1β, and IL-6. Thick arrows indicate cytokines colocalized with axons, and thin arrows indicate cytokines surround axons. Bar 50 μm. Abbreviations: GFAP, glial fibrillary acid protein; IL-1β, interleukin-1-beta; IL-6, interleukin-6; NF, neurofilament; PD, Parkinson's disease; TNF-α, tumor necrosis factor-alpha

| Electron microscopy
The electron photomicrographs clearly showed both myelinated and unmyelinated fibers of sural nerve from PD patients and controls.
In addition, we observed the debris and foam-like body inside the axons in the sural nerve of PD patients with or without sensory disturbances, demonstrating that the axons were degenerating (white arrow in Figure 4B and D); the swollen myelin and fragmentized subunit of Schwann cells indicated demyelination of sural nerve of PD ( Figure 4B-D).

| D ISCUSS I ON
Most of the existing studies center around the central mechanism of PD sensory disturbances and the inflammatory process of CNS.
Our study focused on peripheral nerve inflammation and sought to find peripheral mechanisms of sensory disturbances. Sural nerve is composed of the axons of primary sensory neurons conducting feelings mainly from the lower limbs and Schwann cells sheathing around axons. In this study, analyzing sural nerve from patients with PD and controls, we obtained the following findings. First, the nerve conduction velocity (NCV) slowed down and the sensory nerve action potential (SNAP) amplitude decreased in PD patients compared to healthy controls. This indicated that conduction function of sural nerve was impaired and the peripheral sensory nerve injury in PD patients was definitely present.
However, there was no significant difference in electrophysiological findings between PD patients with or without sensory impair-  Figure 4A), a PD patient with sensory disturbances ( Figure 4B) and a PD patient without sensory disturbances ( Figure 4C-E). A, Normal sural nerve from a control (case Control 2 in Table 1); (B) sural nerve from a PD patient with sensory disturbances (case PD patient 3 in Table 1). C-E, sural nerve from a PD patient without sensory disturbances with different magnifications (case PD patient 9 in Table 1). A, In normal sural nerve, myelinated and unmyelinated fibers are distributed evenly with uniform size. B and C, The abnormality in PD is fewer myelinated and unmyelinated fibers than control. B and D, The foam-like axon indicated by the white arrow may represent axonal degeneration. B-E, The myelin is swollen and partly degrades into debris, and the axon-Schwann cell contact is therefore damaged. "mf" as myelinated fibers and "umf" as unmyelinated fibers  There is no controversy that inflammation exists in PD and our study just extends this knowledge to primary somatic sensory afferent nerve. There are some limitations in our study. Firstly, the sample size is relatively small which weakens the power of the statistic. Secondly, this is a cross-sectional study in which we are not able to analyze the correlation between the cytokines and the disease progression of PD patients. Thirdly, we failed to quantify the number of sural nerve fibers and evaluating the number of Schwann cells and axons in all samples could bring more convincing results to our study. And we did not explore in depth the possible difference between SCs with myelinated nerve fibers and SCs with unmyelinated nerve fibers. We hope further work could overcome these limitations and dig more information from peripheral nerve in PD.

| CON CLUS ION
In conclusion, our study results suggest that peripheral sensory nerve injury exists in PD patients, accompanied by Schwann cell activation and inflammation, thus demonstrate peripheral nerve inflammation participates in the pathophysiological process of PD and this may be related to sensory disturbances of PD patients but not necessarily related.

ACK N OWLED G M ENTS
We are grateful to all the subjects who participated in this study, and all members of the pathology and neurology staff at The