Ethics statement

The human studies were approved by the Ethics Committee of Zhujiang Hospital of Southern Medical University, and all patients provided written informed consent prior to recruitment into the study and was conducted in accordance with the principles of the Helsinki Declaration revised in 1975 and the National Institutes of Health Policy and Guidelines for Human Subjects issued in 1999. All mouse experiments were approved by the Experimental Animal Ethics Committee of Zhujiang Hospital of Southern Medical University.

Clinical study

From Sep 2021 to Jan 2022, 60 PD patients were prospectively recruited in a randomized, double-blinded study (the main researchers, raters, experienced neurologists and patients involved in the study were not aware of the rTMS treatment protocol, except the therapists who did rTMS treatment for PD) from Zhujiang Hospital (Trial Registration Chinese ClinicalTrials. gov Identifier: ChiCTR2100051140). All participants provided written consent to participate in the investigation and allowed investigators to examine their blood samples. Informed written consent was obtained from patients and family members. The enrolled PD patients were interviewed by two experienced neurologists. Characteristics of PD patients, clinical ratings and neuropsychological tests were performed in Table 1. The exclusion Criteria include: 1.Patients with secondary Parkinson's syndrome or Parkinson's overlap syndrome caused by vascular factors, toxins, medications, etc.; 2. PD patients with persistent head tremors; 3. History of Deep Brain Stimulation (DBS) or ablative surgery in the past;4. Other contraindications for rTMS, such as history of epileptic seizures, pregnant women, individuals with cardiac pacemakers, those with metal implants in the body, intracranial hypertension, or severe bleeding tendencies;5. Individuals who have received rTMS treatment within the last 6 months;6. As assessed by the researchers, other factors that make the patient unsuitable for participation in the study or those who are unable to commit to completing follow-up visits. All participants with PD received an EEG before inclusion and fulfilled the Movement Disorder Society Clinical Diagnostic Criteria for Idiopathic PD [31] and underwent extensive clinical examinations. 54 PD patients were ultimately recruited (6 PD patients were ultimately excluded due to four individuals elected not to partake in the study and two subjects were identified as having secondary parkinsonism) and received rTMS treatment or sham rTMS treatment according to random numbers. Random numbers were performed as follows. Computer SAS software was used to randomly group according to a single center, generate random numbers, and the treatment scheme is hidden using opaque envelopes. The envelopes are kept by the treatment provider and opened and sealed by the treatment provider in the order of the patient's visit. Unless the conditions for unblinding are met, the treatment plan cannot be disclosed under any circumstances. The follow-up examinations were done by the same rater and under the same time and all the patients were ON-condition. Sample size was based on a calculation of the results of pre-experiments. The rTMS were delivered by MagPro R30 (Tonica Elektronik A/S, Denmark). The specific rTMS treatment parameters are as follows: initially, high-frequency rTMS (100% resting motor threshold, 10 Hz, 1000 pulses) is administered to the primary motor cortex of the left hemisphere. Subsequently, after a 5-min treatment interval, high-frequency rTMS (100% resting motor threshold, 10 Hz, 1000 pulses) is applied to the primary motor cortex of the right hemisphere. This bilateral hemispheric treatment is conducted daily. The treatment protocol spans over two weeks, with a consecutive 10 working days of treatment for each hemisphere. During the 2-week rTMS treatment period, the dosage of the original treatment drug remained unchanged. The sham rTMS treatment protocol was the same as the rTMS treatment protocol, except that a Cool-B65 P CO coil was used for the sham treatment, while a Cool-B65 A CO coil was used for the rTMS treatment. The alternative sham stimulation was also proceeded for two-week. At the start of each treatment, the resting motor threshold (RMT) of the patient's cortex was measured according to relative frequency method [32]. Specifically, the patients were seated in an armchair with a silversilver chloride surface electrode placed over the abductor pollicis brevis muscle contralateral to each hemisphere. The hot spot was determined using the MagPro R30 Stimulator TMS System (Tonica Elektronik A/S, Denmark) and octagonal coil. The octagonal coil was placed over the scalp and repositioned until the maximal motor evoked potential (MEP) was elicited. After determining the hot spot, the RMT was obtained by delivering single pulse transcranial magnetic stimulation to the hot spot. The RMT was defined as the lowest TMS intensity capable of eliciting a MEP greater than a 50 μV peak-to-peak amplitude in five of the ten subsequent trials. CD4 + CD25 + CD127- cells, CD4 + CD25(low)CD45RA + [naive Treg (nTreg) cells] and CD4 + CD25(high)CD45RA- [activated Treg (aTreg) cells] subsets in the peripheral blood were assessed by flow cytometry and IFN-γ, TNF-α, IL-17α, IL4, IL10, TGF-β1 in the peripheral blood were assessed by Enzyme-linked immunosorbent assay (ELISA) prior to treatment and day 13. Hoehn and Yahr and MDS-UPDRS III were performed prior to treatment and on Days 13, 19, and 40 following the culmination of the two-week rTMS therapy.

Table 1 Baseline characteristics of PD patients in sham and rTMS group

The peripheral blood was obtained from PD patients [33] and lysed in ACK lysis buffer (Biosharp, China).CD4 T cells were then purified using magnetic bead separation (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). The cells were stained with anti-CD25 and anti-CD127 (all from BD Biosciences) for analysis by flow cytometry. Following the acquisition of sample data on a FACSCalibur flow cytometer (BD Biosciences), the results were generated in graphical and tabular formats using FlowJo V10 software (TreeStar Inc., Ashland, USA).

Animal study

Based on our clinical findings of an association between rTMS and changes in circulating Tregs of PD patients, we next conducted studies in a PD mouse model to validate the results. We determined whether the Tregs in the spleen of the PD mouse model changed after rTMS treatment and investigated the potential mechanisms that led to these changes. We applied a similar rTMS treatment protocol in MPTP mice. Male C57BL/6 mice (8 weeks old, 22–24 g) received intraperitoneal (i.p.) injection of 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) (Sigma–Aldrich, USA) (30 mg/kg) for 5 days as PD animal model, while control mice received the same dose of normal saline. Another group of MPTP mice received i. p. injection of anti-CD25 monoclonal antibody (1 mg/kg; Thermo, USA) for three consecutive days to block Tregs in the spleen, to investigate whether the therapeutic effect of rTMS on MPTP mice is related to changes in Treg level in the spleen. In addition, we also evaluated the behavior of mice by pole climbing test and detected the expression of dopamine cells, neurotrophic factors BDNF, GDNF, microglia and inflammatory factors in the SN of mice. We next screened the hub node proteins related to rTMS on MPTP mice by quantitative proteomics analysis with TMT labeled of whole proteins. Finally, to better understand the role of hub proteins in MPTP mice, RNA interference (RNAi) mediated Adeno-associated virus (AAV) vector was injected into midbrain via a stereotactic midbrain approach to block hub proteins level in SN. Immunofluorescent staining and western blots (WB) were used to detect the transfection efficiency. The changes of hub protein levels in SN were detected by immunofluorescence staining and the inflammatory cytokines in the ventral midbrain were detected by ELISA after the hub proteins were interfered by the virus.

For the sample size in animal experiments, a power analysis was performed. A sample size of at least n = 8 per group was determined to reach a statistical significance of 0.05, in detecting an effect size of at least 1.06 with 80% power. Mice were assigned randomly to the experimental groups. The investigators were blinded to the identities of treatment groups.

Animals and MPTP-PD model

Male C57BL/6 mice (8 weeks old, 22–24 g) were obtained from Liaoning Changsheng Biotechnology Co., Ltd. (Liaoning, China). Before the experiment, all mice were housed at 20–22 °C with a 12 h: 12 h light/dark cycle and food and water provided ad libitum for 7 days. The CD25 receptor is a key site for activating Treg, and as the hypothesis is that rTMS may exert its modulatory effects on Treg in PD, we used anti CD25 injections to specifically block Treg to determine if rTMS still has an impact on PD. A total of 189 male mice were randomly divided into 9 groups as follows: (1) the saline normal control group (NC group, n = 21 mice) was only administered saline by intraperitoneal (i. p.) injection for 5 days; (2) the NC + sham rTMS group (NC + sham group, n = 21 mice), which were treated with i. p. injections of saline (30 mg/kg) for 5 days followed by sham rTMS treatment; (3) the NC + rTMS (10 Hz) group (n = 21 mice), which were administered i. p. doses of saline (30 mg/kg) for 5 days before receiving rTMS (10 Hz) therapy; (4) the MPTP group (n = 21 mice), which received i. p. injections of 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) (Sigma–Aldrich, USA) (30 mg/kg) for 5 days; (5) the MPTP + sham rTMS group (n = 21 mice), which were treated with i. p. injections of MPTP (30 mg/kg) for 5 days followed by sham rTMS treatment; (6) MPTP + rTMS (10 Hz) group (n = 21 mice), which were administered i. p. doses of MPTP (30 mg/kg) for 5 days before receiving rTMS (10 Hz) therapy; (7) the MPTP + block group (n = 21 mice), which received 3 days of continuous i. p. injections of anti-CD25 monoclonal antibodies (1 mg/kg) followed by 5 days of i. p. injections of MPTP (30 mg/kg); (8) the MPTP + block + sham (10 Hz) group (n = 21 mice), which was administered i. p. injections of anti-CD25 monoclonal antibodies (1 mg/kg, Thermo, USA) for 3 days before receiving i. p. injections of MPTP (30 mg/kg) for 5 days before receiving shamtherapy; and (9) the MPTP + block + rTMS (10 Hz) group (n = 21 mice), which was administered i. p. injections of anti-CD25 monoclonal antibodies (1 mg/kg, Thermo, USA) for 3 days before receiving i. p. injections of MPTP (30 mg/kg) for 5 days before receiving sham therapy.

rTMS treatment

A transcranial magnetic stimulator (MagVenture A/S, Denmark) connected to a circular coil (Cool-40 Rat Coil) with a diameter of 40 mm was used to deliver rTMS. the rTMS treatment regimen for mice is consistent with the treatment regimen for PD patients. In the sham rTMS group, the stimulus was delivered by fixing the coil 10 cm above the heads of the mice to ensure that the mice felt the clicking sound or vibrations generated by the rTMS coil but did not receive brain stimulation. To minimize the discomfort of the mice, their movements were carried out in a soft plastic funnel, therefore, on the first day, they were restrained without stimulation for 10 s to familiarize them with the experimental procedure. At the start of each treatment, the resting motor threshold was measured as previously described [34]. Briefly, a standard electromyographic (EMG) machine (Counterpoint, Dantec Medical Inc., Denmark) was used for recording of the motor evoked potentials (MEPs). Sampling frequency was 51.4 kHz, high and low pass filter settings were 10 and 3000 Hz. A 0.5 mm bipolar EMG needle (DANTEC) placed in the right hindlimb biceps femoris muscle with a ground electrode 10 mm proximal to the recording electrode was used for recording the muscle activity. The motor threshold was defined as a reproducible motor evoked potential in five consecutive stimuli with an interstimulus interval > 3 s and an amplitude > 50 mV. The real and sham rTMS treatments did not induce seizures or other behavioral changes during the treatment period.

Pole climbing test

We assessed motor dysfunction by performing pole tests [35] Briefly, the mice were placed head-up on the top of the pole (height 50 cm, diameter 1 cm), and their return to the bottom was timed. Timing began when the mouse was released and stopped when one hindlimb reached the bottom. The test was repeated 3 times for each mouse after they were trained once at 30 min intervals, and the average time was used for data analysis. The raters were blinded for the treatment group.

Flow cytometric analysis of CD4 + CD25 + Foxp3 + Tregs in the spleen

The spleens of male C57BL/6 J mice [19] were removed and homogenized by a tissue grinder and a wire mesh screen. Red blood cells were then lysed in ACK lysis buffer. Then CD4 + T cells were then purified using magnetic bead separation (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). The cells were stained with anti-CD25 and anti-Foxp3 (BD Biosciences, USA, 567,456) for flow cytometric analysis. Following the acquisition of sample data on a FACSCalibur flow cytometer (BD Biosciences), the results were generated in graphical and tabular formats using FlowJo V10 software (TreeStar Inc., Ashland, USA).

Immunohistochemical staining

The mice were anesthetized with sodium pentobarbital, sacrificed, and transcardially perfused with normal saline containing 0.5% sodium nitrate and heparin (10 U/ml) followed by 4% paraformaldehyde (PFA) dissolved in 0.1 M phosphate buffer. The brain was dissected and postfixed in 4% PFA at 4 °C for 18–24 h and then immersed in 20% sucrose at 4 °C until the brain dropped to the bottom, followed by immersion in 30% sucrose at 4 °C for 24 h. The brains were embedded in optimal cutting temperature compound (O. C. T. Compound, Tissue-Tek, USA.) We prepared coronal sections of the frozen brains with a cryostat microtome (CM1950, Leica, Germany). Frozen brains were sectioned into 40-µm-thick coronal sections. All sections were collected and processed for immunohistochemical staining. For antigen retrieval, the tissue slices were submerged in 0.01 M sodium citrate buffer (pH 6.0) and rinsed twice in PBS. The tissue slices were then treated for 1 h at 37 °C in PBS containing 0.2% v/v Triton X-100, 0.02% w/v sodium azide, and 5% v/v goat serum. Primary antibodies against tyrosine hydroxylase TH (1: 800, Proteintech, USA, 25,859–1-AP), BDNF (1: 500, Abcam, England, ab108319), GDNF (1: 500, Abcam, England, ab18956) and Iba-1 (1: 500, Abcam, England, ab178846) SYT6 (1: 250, SANTA CRUZ, sc390321), TLR4 (1: 250, Invitrogen, 710,185) were added to the slices and incubated overnight at 4 °C. Then, the sections were treated with the appropriate biotinylated secondary antibody, followed by Vectastain ABC reagent (Vector Laboratories, Burlingame, CA, USA) and 3, 3’-diaminobenzidine (Sigma–Aldrich, Vienna, Austria). The stained slices were dehydrated and coverslipped with Entellan before being mounted on slides (Merck, Darmstadt, Germany). IgG conjugated with Alexa 594 (1: 500, Abcam, England, ab150080) was used for immunofluorescent staining. Mounting medium was used to cover the sections (Dianova, Hamburg Germany). The numbers of TH + , SYT6, TLR4 in the substantia nigra pars compacta (SNc) were estimated using unbiased stereological methods (Stereo Investigator, MicroBrightField, VT). The numbers of activated Iba-1-positivemicroglia in the SNc were estimated using the optical fractionator probe with the stereologist. Microglia were classified as activated if the cell body was visibly increased in diameter and the cell had shortened and thickened processes. The three stereologists were blinded to the treatment received. Additionally, we determined the average optical density value of BDNF and GDNF immunoreactivity through densitometry, employing the freely available ImageJ software (National Institutes of Health, Bethesda, MD, USA). All histochemical data statistics was conducted in a single-blind manner by four experimenters. FX renumbers each group of fluorescent films in an experiment and randomly divides them into three groups (random number method) and assigns them to three investigators (YQL, ZCX and Q.W). Each film takes 5 fields of view for statistical averaging, and finally the data is summarized with FX.

Quantitative proteomics technology with TMT labeling of whole protein

Twelve brain tissue from 4 groups (saline normal control, MPTP, MPTP + rTMS, MPTP + block + rTMS) were collected for proteome Tandem Mass Tag (TMT) analysis. The collected samples were processed for TMT analysis in JingJie Bio-tech. Output was visualized using R and CummeRbund. Heatmaps were generated in R using pHeatmap and R ColorBrewer. Result was analyzed according to Maxquant (v1.6.15.0) database.

Molecular docking method

The protein data bank (PDB) database (https://www.rcsb.org/) was used to obtain the PDB file which has the active structure of the target protein, The smaller Resolution value and complete protein binding pockets were used as the screening conditions to select the crystal structure for subsequent analysis. Refer to the previous literature [36,37,38,39,40], according to the amino acid sequence provided by Unirpot, SWISS-MODEL was used (https://swissmodel.expasy.org/interactive) to perform online homology modeling of proteins without ready-made protein crystal structures in the PDB database. Select the result with homology ≥ 30%, and then according to the scoring screen to evaluate the model quality with GMQE and QMEAN values when the homology reaches the standard, the template with larger GMQE and QMEAN close to 0 was chose. Then, online protein–protein docking is performed through Z-dock (https://zdock.umassmed.edu), the version is selected as V3.0.2 [41, 42], and the docking result image uses PyMol 2. 3.5 for image rendering.

Path enrichment analysis and PPI network construction

GO database (http://geneontology.org/) was used for pathway enrichment analysis [43], obtain immune-related pathway genes, and String database (https://www.string-db.org/) was used to construct protein online Protein interaction network (PPI network) [44], the species is selected as Mus musculus, the minimum required interaction score is set to 0.4, and the other conditions are the default settings.

Western blotting

Tissues were lysed in a detergent-based lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS and 1 mM PMSF). Samples were then centrifuged for 10 min at 13, 000 g at 4 °C, and the supernatant was collected and quantified with the BCA Protein Assay Kit (Thermo Scientific, USA). Equal amounts of protein sample were separated by 10% SDS-PAGE. Antibodies used for western blotting were: Syt6 (1: 1000, abcam, USA), TLR4 (1: 1000, abcam, USA), TH (1: 1000, proteintech, USA), Slc6a3 (1: 1000, abcam, USA) and β-actin (1: 10,000, Thermo, USA). Antibody binding was detected using the secondary antibodies (Peroxidase-Conjugated AffiniPure Goat Anti-Rabbit/Mouse IgG (H + L), 1: 5000, thermofisher, USA). All samples were normalized to β-actin. All quantification was calculated with Alpha Ease FC software (Alpha Innotech Corporation, USA).

ELISA assay

Cytokine levels of IFN-γ, TNF-α, IL-17α, IL4, IL10 and TGF-β1 in the serum of PD patients were measured using commercial ELISA kits (IFN-γ ELISA kit (Abcam, England, ab46025), TNF-α ELISA kit (Abcam, England, ab181421), IL-17α ELISA kit (Abcam, England, ab216167), IL4 ELISA Kit (Abcam, England, ab215089), IL-10 ELISA kit (Abcam, England, ab100549), TGF-β1 ELISA kit (Sigma–Aldrich, USA,RAB0460) in accordance with the manufacturer’s protocols. Brain tissue was collected from the different groups of mice, and then the tissue from the ventral midbrain was separated and homogenized with lysis buffer on ice. The supernatant was collected at 12, 000 rpm at 4 °C and centrifuged for 10 min for subsequent experiments. The proinflammatory cytokines IL-6, IFN-γ, TNF-α, IL-1β, IL-10 and TGF-β1 were measured using commercial ELISA kits (IL-6 ELISA kit (Abcam, England, ab100713), IFN-γ ELISA kit (Abcam, England, ab100690), TNF-α ELISA kit (Abcam, England, ab229393), IL-1β ELISA Kit (Sigma–Aldrich, USA, RAB0275), IL-10 ELISA kit (Abcam, England, ab100697), TGF-β1 ELISA kit (Abcam, England, ab119557) in accordance with the manufacturer’s protocols.

Virus constructs and stereotaxic surgery

The scramble short hairpin RNA (shRNA) (5’-CCTAAGGTTAAGTCGCCCTCG-3’) and shRNA coding sequences targeting mous syt6 (5’-ATGAAAGCGAGACGCTGATTG-3’), (5’- ATGTCTCCAGCGTGGACTATG-3’) and (5’- GCGGAAGTTCTGACCCTTA-3’) were cloned into the PAAV-U6-shRNA(Syt6)-CMV-EGFP-WPRE vector (Obio Technology, Shanghai, China). Adeno-associated viruses (AAVs) described above were packaged by Obio Technology into serotype 2/9.After mice were anesthetized and placed in the stereotaxic frame, 2 μl of purified and concentrated AAV(1012 IU/mL) was unilateral injected into the right SNc (AP =  − 3.2 mm; ML =  ± 1.2 mm, DV =  − 4.6 mm) at a rate of 0.2 μl/min according to previously described protocols [45]. The 5-μl Hamilton syringe was kept in place for 10 min before being slowly retracted within 5 min.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8.0 software. Data with a normal distribution are presented as the mean ± SD. Intra-group comparisons of squared differences were conducted using paired sample t-tests, while inter-group comparisons were made using two independent samples t-tests. Data that did not conform to a normal distribution are presented with comparisons made using non-parametric tests. Counting data are presented as frequencies, with comparisons made using the χ2 test. A two-sided test was chosen, and a P-value less than 0.05 was considered statistically significant. Correlation between the change of Treg ratio in peripheral blood and the change of MDS-UPDRSIII before and after treatment with rTMS were calculated by the Pearson correlation. After flowcytometry, CD4 subsets were Determined as a percentage of all lymphocytes gated using forward and side scatter, and as a percentage of all leukocytes bydual-platform analysis using full blood cell counts from the same blood collection. Mouse flow cytometry data, total times of pole climbing test, TH, BDNF, GDNF, Syt6, TLR4 immunohistochemistry, IL-6, IFN-γ, TNF-α, IL-1β, IL-10 and TGF-β1 enzyme-linked immunosorbent assay and Syt6, TLR4, Slc6a3 and TH western blotting were analyzed using One-way ANOVA followed by the least significant difference (LSD) for post hoc comparisons. The data are representative of at least three independent experiments.