Parkinson’s disease (PD) is a neurodegenerative disorder characterized by the presence of Lewy bodies, which primarily consist of ubiquitin and a-synuclein (a-Syn), and the progressive loss of dopaminergic neurons in the substantia nigra [1]. Current options for the treatment of PD, including deep brain stimulation and dopaminergic therapies, can alleviate symptoms but cannot directly modify the disease [24,39]. Hence, there is a pressing need to develop neurorestorative strategies to slow the progression of PD [1,14].

Parkinson’s disease is a debilitating neurodegenerative disorder, with growing evidence linking gut microbiota alterations to its pathogenesis and progression. The gut microbiota is a complex ecological system containing up to 40 trillion bacteria [31]. This highly complex ecological system can influence both normal host physiology and disease predispositions through metabolic activities and direct host-microbe interactions [38,44]. In individuals with PD, microbial species richness has been found to be reduced, and the composition of the gut microbiota was altered compared with that of healthy controls [16]. Studies have revealed microbiome differences in patients with PD, including increases in Verrucomicrobiaceae and Akkermansia, and decreases in Prevotellaceae [2,29].

Regulatory T cells (Tregs) play an essential role in the maintenance of immune homeostasis and self-tolerance [27]. Two mechanisms of the anti-cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) action have been proposed; one involves the constitutive expression of CTLA-4 to selectively deplete Tregs, which may not be in agreement with the temporary expression on active conventional T cells. Although scientific debate persists regarding the precise mechanisms of anti-CTLA-4, it is still agreed that the maximal therapeutic efficacy of anti-CTLA-4 antibodies is achieved by preserving both the antagonist activity and the Treg depletion function [15,22,23].

Given the key role of gut microbiota in PD, the present study investigated the relationship between intestinal microbiota translocation and PD. It was found that dextran sulfate sodium (DSS)-induced intestinal microbiota translocation could exacerbate motor dysfunction in a-Syn preformed fibril (PFF)-induced PD via regulation of Tregs/CTLA-4.

Animals

Eight-week-old male C57BL/6 mice with 30-35 g obtained from Hunan Slack Jingda Experimental Animal Co., Ltd were housed in an SPF grade animal room at the Animal Center of University of South China, with good ventilation and lighting conditions maintained at a relative humidity of 40-60% and an indoor temperature of 23 ±2°C. Mice had ad libitum access to food and water, and the bedding was changed regularly to ensure a comfortable environment for their growth.

Groups and treatments

The mice were randomly allocated to five groups, with 10 rats in each group: Control group, model group (PFF group), DSS group (DSS + PFF group), antibiotic mixture group (DSS + PFF + Abx group), and CTLA-4 activator group (DSS + PFF + CTLA-4 fusion protein group). The model group was first fed with an equal amount of sterile distilled water for 7 days and then injected with 5 µl of PFF for colonic modeling. The remaining groups received 3% DSS solution for 7 days before PFF administration to induce acute intestinal barrier dysfunction. Subsequently, 5 µl of PFF was injected into the colon for colonic modeling. Starting on the first day after modeling, the antibiotic mixture group was first given 0.1 ml of antibiotic mixture (4 mg/ml kanamycin, 0.35 mg/ml gentamicin, 0.85 mg/ml polymyxin, 2.15 mg/ml metronidazole, and 0.45 mg/ml vancomycin) by gavage for 7 days. After 21 days, the antibiotic mixture was diluted 50-fold (excluding vancomycin) and added to the drinking water for consumption. The remaining groups were fed sterile distilled water for 3 weeks, with the CTLA-4 group receiving weekly injections of 9D9 (5 mg/kg) beginning on the third day after completing PFF modeling surgery, for a total of 3 weeks.

Gut microbiota analysis

Mouse fecal samples were collected and processed as soon as possible or stored under appropriate conditions (typically frozen storage). Microbial cells in the fecal samples were disrupted to release DNA: The samples were frozen quickly in liquid nitrogen and then rapidly thawed, and this was repeated several times to break the cell walls. Total DNA from stool samples was isolated using the QIAamp DNA Stool Mini Kit (Qiagen) based on the modified instructions provided by the manufacturer. The total genomic DNA was assessed as described in Appendix A. DNA was amplified using dual-unique barcoded primers targeting V1-V3, V3-V4, or V4. The size and purity of PCR products were determined using agarose gel electrophoresis to ensure specific amplification of the expected size. FASTP (version 0.18.0) was used to remove raw reads with > 10% unknown nucleotides (N) and reads where < 60% of bases had a quality value < 20. FLASH (version 1.2.11) was used to merge paired-end clean reads into raw tags, requiring a minimum overlap length of 10 bp and allowing a 2% mismatch error rate. The QIIME (version 1.9.1) pipeline was used to filter noisy raw tags. The filtering conditions included breaking tags from the first low-quality base site, where the number of contiguous low-quality bases reached a set length, and then filtering tags with a contiguous high-quality base length of < 75% of the tag length. The UCHIME algorithm was used to check for chimeras in the clean tags. After removing chimeric tags, the final tag set was obtained. To study the species diversity of the samples, the clean data were imported into Qiime2 for analysis. DADA2 was used for filtering and quality control, and to remove chimeras, merge paired-end sequences, and denoise the sequencing data to generate feature information, i.e., amplicon sequence variants (ASVs). To obtain the taxonomic classification, a classifier was trained using the V3-V4 region of the Silva138 99% clustered sequences as the reference genome. The trained classifier was used to classify the ASVs and obtain the corresponding taxonomic information, thus determining the community composition for each sample. The naive Bayes classifier and q2-feature-classifier were used for taxonomic analysis of the feature sequences and to statistically analyze the community composition at various taxonomic levels for each sample.

Behavioral tests

Pole test

A pole that was suitable for the size of the experimental animal was selected, usually a wooden or metal pole with a diameter of 1-3 cm. The height of the pole was usually 50-100 cm. Before formal testing, the animals underwent 1 week of adaptation training to improve their coordination ability. Adaptation training usually involved placing the animal on top of a pole and allowing the mouse to naturally climb down the pole from the top and gradually adapt. During the testing phase, animals were placed on top of the pole, and the time required for them to arrive at the bottom of the pole was determined. Before each test, the pole was sprayed with 75% alcohol to eliminate the influence of odors between different mice. Multiple tests were usually conducted to obtain reliable results.

Gait experiment

A self-made running track was created for gait experiments, with opaque fences around the track (requiring a track size of 10 × 10 × 50 cm). The designed track was placed on a sheet of paper used for recording gait, and a top cover was placed on two-thirds of the end of the track. Before formal testing, animals underwent a 1-week adaptation period, during which mice were trained to crawl on a self-made opaque track until they could successfully complete the task. During the testing phase, the front and rear paws of each group of mice were marked with red and blue pigments using cotton swabs to distinguish them. The mice were placed at the starting point of the track and induced to pass through the track, ensuring that they left clear footprints on the recording paper and analyzed. The step distance measurement recorded at least three consecutive steps of the mice after crawling. The two consecutive footprints of one hind paw were connected to the corresponding footprint of the opposite hind paw to form a triangle. The distance between two consecutive footprint points and the vertical distance from the opposite footprint point to the line, which are the step length and step width, respectively, were measured.

Grip experiment

Before formal testing, the animals underwent a 1-week training period to improve their coordination ability. Adaptation training usually involved grabbing the two front paws of the mouse onto a metal wire, allowing it to adapt and hang on the wire for a period of time. The two front paws of C57BL/6 mice were grasped onto a horizontal metal wire, with the mouse body in a suspended state. The grasping of the metal wire by the two hind paws was observed and rated, with a maximum evaluation time of 1 min. The scoring criteria were as follows: dropped from the wire within 30 s, 0 points; front and rear claws could not grasp the metal wire at the same time, 1 point; grasped the metal wire with the front and back paws together, but without the tail, 2 points; grasped the metal wire with the front claw, rear claw, and tail, 3 points; moved along the wire with the front claw, rear claw, and tail, 4 points; moved along the steel wire with the front and rear claws, 5 points.

Histochemical staining

Immunofluorescence staining

Immunofluorescence staining was carried out as described previously [3,5,21]. Sections were blocked with 10% normal donkey serum diluted in PBS for 1 h at room temperature, and then incubated with a mixture of the following primary antibodies at 4°C overnight: Rabbit monoclonal anti-zonula occludens-1 (ZO-1; 1 : 500, ab307799, Abcam) and mouse monoclonal anti-occludin (1 : 500, ab242202, Abcam). Following rinsing with PBS, immunoreactivity was visualized via incubation with Alexa Fluor 488 fluorescent secondary antibody (1 : 1,000; Invitrogen, Thermo Fisher Scientific, Inc.). Subsequently, the slides were mounted using ProLong Gold Anti-fade reagent and were imaged using AxioVision4.6 software (Carl Zeiss AG) under an AxioObserver A1 microscope (Carl Zeiss AG).

Immunohistochemical staining

Immunohistochemical staining was carried out as described previously [7]. Sections were incubated in a 3% H₂O₂ solution for endogenous peroxidase clearance at room temperature for 10 min. Sections were subsequently washed in PBS three times for 5 min each. After blocking in 10% normal goat serum diluted in PBS for half an hour, the slides were incubated with the following primary antibodies: Mouse monoclonal anti-dopamine transporter antibody (ab128848, Abcam) or anti-ionized calcium binding adapter protein 1 (IBA1) antibody (ab283319, Abcam) at 4°C overnight. Subsequently, the slides were incubated using an Enhanced Polymer DAB Detection kit (PV-900, Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.) and an AEC kit (ZLI-9036, Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.). Finally, the sections were mounted, dehydrated, and sealed with coverslips.

Hematoxylin-eosin (H&E) staining

H&E staining was carried out as described previously [8,17,18,43]. Firstly, the slices were immersed in xylene for 20 min, and the process was repeated three times. Subsequently, the slices were soaked in 100%, 100%, 95%, 85%, and 75% ethanol in sequence, maintaining each concentration for 5 min. Afterwards, the slices were soaked in distilled water for 5 min. Hematoxylin staining was performed for 1-10 min, and then slices were rinsed with distilled water and returned to blue with PBS. Eosin staining was performed for 1-5 min, followed by rinsing with distilled water. Gradient alcohol (95-100%) dehydration was performed, maintaining each concentration for 5 min (or directly drying the slices). After removal, the slices were soaked in xylene for 10 min, which was repeated twice, and then the slices were sealed with neutral gum and finally observed under a microscope.

Western blotting

Western blotting was carried out as described previously [6,8,20,36,41,42]. Tissue lysates were mixed with the loading buffer containing 20% glycerol, 0.125 M Tris-HCl, pH 6.8, 0.1% bromophenol blue, 10% sodium dodecyl sulfate, and 5% b-mercaptoethanol. All samples were heated at 95°C for 15 min, subjected to 10% SDS-PAGE and then electroblotted onto polyvinylidene difluoride membranes (MilliporeSigma). Incubation in 5% non-fat milk powder or fatty acid-free BSA diluted in Tris-HCl saline buffer supplemented with 0.1% Tween-20 (TBST; pH 7.4) for 1 h was used to block non-specific binding sites. Membranes were then incubated overnight at 4°C with mouse monoclonal anti-a-Syn (1 : 500, ab309157, Abcam). After washing with 0.1% TBST three times at room temperature for 5 min each, the membranes were incubated with horseradish peroxidase-conjugated goat anti-mouse secondary antibodies (1 : 2,000, BA1051, Boster) diluted in 0.1% TBST. Next, membranes were washed in 0.1% TBST three times for 5 min each at room temperature. Finally, the immunoreactive bands were detected using an enhanced chemiluminescence kit (170-5061, Bio-Rad Laboratories, Inc.). Band intensities were quantified using ImageJ software (version 7.0).

Enzyme-linked immunosorbent assay

Enzyme-linked immunosorbent assay (ELISA) was performed as previously described [9]. Protein levels of sIL-2Ra (HZ0210, Gaide Chem, China), CTLA-4 (Feien Biological Technology, Ltd., Wuhan, China), CD-4 (XG-E990036, Gaide Chem), and IL-2 (ab221834, China) were measured in the supernatant using available kits according to the manufacturer’s protocol.

Statistical analysis

All statistical analyses were performed using GraphPad Prism 6 software (GraphPad Software, Inc.). Data were expressed as the mean ±SD and analyzed with one-way ANOVA followed by a post hoc Bonferroni test. P < 0.05 was considered to indicate a statistically significant difference.

DSS can exacerbate gut microbiota translocation induced by intraperitoneal injection of PFF in PD model mice

To determine the influence of DSS on gut microbiota translocation in PD mice, 16S technology was used to detect the distribution of microbial communities in the colon, spleen, and brain.

In comparison to the PFF group, the abundances of the Gram-negative bacteria Bacteroidia and Desulfovibrionia, which are associated with peripheral endotoxemia, in the gut, spleen, and brain of mice was higher in the DSS + PFF group (Fig. 1A-E).

DSS-induced gut microbiota translocation can regulate Tregs/CTLA-4 in the PD mouse model

To determine the influence of DSS on Tregs/CTLA-4 in PD mice, ELISA was carried out to detect the expression of soluble interleukin 2 receptor a (sIL-2Ra) and CTLA-4 at the protein level in the serum of PD mice.

The sIL-2Ra level was higher in the PFF group than in the Control group. The sIL-2Ra level was upregulated in the DSS + PFF group compared with the PFF group. However, compared with the DSS + PFF group, the sIL-2Ra level was lower in the DSS + PFF + 9D9 group (Fig. 2A). In contrast, CTLA-4 levels exhibited the opposite pattern to those of sIL-2Ra (Fig. 2B).

DSS can exacerbate the disruption of intestinal barrier integrity in the PFF-induced PD mouse model via regulation of Tregs/CTLA-4

To investigate the effect of DSS on the intestinal barrier integrity in PD mice, H&E staining and immunofluorescence staining of ZO-1 and occludin were carried out in the colon of PD mice.

Compared with that in the Control group, crypt height was lower in the PFF group. Compared with that in the PFF group, crypt height was lower in the DSS + PFF group. However, compared with that in the DSS + PFF group, crypt height was higher in the DSS + PFF + 9D9 group (Fig. 3A, C). A similar pattern was observed for villus height (Fig. 3B, C).

Compared with that in the Control group, the ZO-1 intensity was lower in the PFF group. Compared with that in the PFF group, the ZO-1 intensity was lower in the DSS + PFF group. However, compared with that in the DSS + PFF group, the ZO-1 intensity was higher in the DSS + PFF + 9D9 group (Fig. 3D, E). A similar pattern to that for crypt height was observed for villus height (Fig. 3D, F).

DSS can exacerbate motor dysfunction in the PFF-induced PD mouse model via regulation of Tregs/CTLA-4

To investigate the effect of DSS on motor dysfunction in PD mice, behavioral tests, including the pole test, gait experiment, and grip experiment, were carried out in PD mice.

In the pole test, compared with that in the Control group, the step width was higher in the PFF group. Compared with that in the PFF group, the step width was higher in the DSS + PFF group. However, compared with that in the DSS + PFF group, the step width was lower in the DSS + PFF + 9D9 group (Fig. 4A). An opposite pattern was observed for step length relative to step width (Fig. 4B).

In the gait experiment, compared with that in the Control group, the time was higher in the PFF group. Compared with that in the PFF group, the time was higher in the DSS + PFF group. However, compared with that in the DSS + PFF group, the time was lower in the DSS + PFF + 9D9 group (Fig. 4C).

In the grip experiment, compared with that in the Control group, the score was lower in the PFF group. Compared with that in the PFF group, the score was lower in the DSS + PFF group. However, compared with that in the DSS + PFF group, the score was higher in the DSS + PFF + 9D9 group (Fig. 4D). A similar pattern for the time to score was also observed (Fig. 4E).

DSS can exacerbate dopaminergic neuron loss in the PFF-induced PD mouse model via regulation
of Tregs/CTLA-4

To investigate the effect of DSS on neuron death in PD mice, immunohistochemical staining of DA and western blot analysis of a-Syn were carried out in the midbrain of PD mice.

Compared with that in the Control group, the DA level was lower in the PFF group. Compared with that in the PFF group, the DA level was lower in the DSS + PFF group. However, compared with that in the DSS + PFF group, the DA level was higher in the DSS + PFF + 9D9 group (Fig. 5A, B).

Compared with that in the Control group, the a-Syn level was higher in the PFF group. Compared with that in the PFF group, the a-Syn level was higher in the DSS + PFF group. However, compared with that in the DSS + PFF group, the a-Syn level was lower in the DSS + PFF + 9D9 group (Fig. 5C, D).

DSS can exacerbate neuroinflammation in the PFF-induced PD mouse model via regulation of Tregs/CTLA-4

To investigate the effect of DSS on neuroinflammation in PD mice, ELISA analysis of CD4 and IL-2 levels in the serum, and immunohistochemical staining of IBA1 in the midbrain were carried out in PD mice.

Compared with that in the Control group, the CD4 level was higher in the PFF group. Compared with that in the PFF group, the CD4 level was higher in the DSS + PFF group. However, compared with that in the DSS + PFF group, the CD4 level was lower in the DSS + PFF + 9D9 group (Fig. 6A). A similar pattern for the IL-2 level relative to the CD4 level was also observed (Fig. 6B).

Compared with that in the Control group, the IBA1 level was higher in the PFF group. Compared with that in the PFF group, the IBA1 level was higher in the DSS + PFF group. However, compared with that in the DSS + PFF group, the IBA1 level was lower in the DSS + PFF + 9D9 group (Fig. 6C, D).

A previous study reported that healthy microbiota may be helpful for ameliorating motor deficits in patients suffering from PD [13]. The present study revealed that intestinal microbiota translocation could exacerbate the motor symptoms in PD through the regulation of Tregs/CTLA-4.

The formation of a-Syn oligomers in the misfolding of a-Syn is mainly responsible for the neurotoxicity of a-Syn, as well as the cell-to-cell transmission of a-Syn pathology [10,35]. Therefore, a-Syn oligomers may serve as an ideal biomarker candidate for the progression and diagnosis of PD [11]. Studies in animals have pinpointed the gut microbial community as key in sparking neuroinflammation, digestive issues, and movement problems in mice overexpressing a-Syn [28]. Antigens of the gut microbiota can ultimately influence the immune system, and the metabolic activity of intestinal bacteria can affect the production of neurotransmitters such as dopamine [33]. The present study revealed higher abundances of the Gram-negative bacteria Bacteroidia and Desulfovibrionia after treatment with a-Syn.

Reduced numbers of Tregs in patients with PD and impaired suppression of T-cell proliferation have been observed in patients with PD. Restoration of Treg numbers and suppressive function may attenuate disease-associated inflammation and confer neuroprotection in animal models of PD [26,30]. Additionally, restoring and enhancing the suppressive functions of Tregs may be a novel cell therapeutic strategy for patients suffering from PD [37]. The present study revealed that the levels of sIL-2Ra and CTLA-4 in the serum of PD mice were upregulated.

Patients with PD tend to have a more permeable intestinal barrier and inflammation in the colon, which may be related to certain gut member abundance [12,19]. The present study demonstrated impaired intestinal barrier integrity in PD mice.

In patients with PD, motor dysfunction commonly occurs. Neurological function assessment is commonly performed to evaluate the therapeutic effect of different approaches. The present study demonstrated an influence on motor dysfunction.

Parkinson’s disease is characterized by progressive degeneration of dopaminergic neurons [40]. The resulting loss of dopamine contributes to many of the clinical symptoms associated with progression of the disease, such as mood alterations, motor dysfunction, and cognitive impairment [25]. In the present study, dopaminergic neuron loss was observed.

Individuals with PD often exhibit signs of inflammation and oxidative stress in their gut, possibly due to harmful effects from abnormal a-Syn proteins in the gut nerves, potentially initiating and worsening the disease [4]. The disease is characterized by the buildup of a-Syn protein aggregates within neurons, leading to cell death and widespread neuroinflammation [34]. Early on, the buildup of a-Syn in the gut and certain nerves, linked to inflammation, appears to activate the body defense mechanisms in the digestive system [32]. The present study demonstrated that neuroinflammation was exacerbated.

Overall, intestinal microbiota translocation may exacerbate the motor symptoms in PD, providing a rationale for regulating the intestinal microbiota as a novel therapeutic strategy for PD.

The present study was supported by the University of South China Clinical Research 4310 Program (grant no. 20224310NHYCG06), the Hunan Provincial Natural Science Foundation of China (grant no. 2025JJ90149), Major scientific research project of national clinical key specialties (grant no. 20230277), and the general project of Hunan Health Commission (grant no. 202203074319).

All experimental procedures were approved by The Laboratory Animal Ethics Committee of The First Affiliated Hospital of the University of South China (Approval No. 2022LL1001002) and were conducted in accordance with the committee’s guidelines and relevant regulations for the ethical use and care of laboratory animals.

The authors report no conflict of interest.