The application of fecal microbiota transplantation in Parkinson’s disease

PD is a neurodegenerative disorder predominantly affecting middle-aged and elderly populations, clinically characterized by cardinal motor manifestations including resting tremor, rigidity, bradykinesia, and postural instability. Substantial evidence indicates that non-motor symptoms, particularly gastrointestinal disturbances, affective disorders, sleep abnormalities, sensory deficits, cognitive impairment, and autonomic dysfunction—frequently precede motor manifestations by several years and demonstrate progressive exacerbation with disease duration.

Among these non-motor manifestations, gastrointestinal dysfunction represents the most prevalent symptom complex in PD patients. The human gastrointestinal tract harbors a diverse microbial ecosystem that bidirectionally modulates both motility and endocrine functions. This intestinal microbiota comprises a dynamic consortium of microorganisms, including bacteria, archaea, viruses, and eukaryotic organisms (e.g., fungi and protozoa) that occupying distinct ecological niches throughout the gastrointestinal lumen, collectively termed the gut microbiome (Arumugam et al., 2011). The metabolic activities and complex interactions of gut microbiota significantly influence host physiological homeostasis and disease susceptibility. As the "second genome" of the human body, the gut microbiome plays a fundamental role in both the initiation and progression of neurodegenerative disorders (Zhu et al., 2010). The commensal gut microbiota maintains intestinal homeostasis through three principal mechanisms: biosynthesis of essential micronutrients,including vitamins, amino acids, and lipids; modulation of neural signaling pathways; and regulation of immune-endocrine crosstalk. These coordinated functions collectively sustain the MGBA and preserve intestinal microenvironmental stability (Ma et al., 2019; Cook and Mansuy-Aubert, 2022). In recent years, the gut microbiome has emerged as a focal point in PD research, with accumulating evidence implicating its involvement in multiple facets of PD pathogenesis (Scheperjans et al., 2015), clinical phenotype (Vascellari et al., 2020), therapeutic effect of levodopa (Beckers et al., 2022) and relevant effects of disease-modification (Chen and Lin, 2022; Lin et al., 2023). A landmark study recently published in Nat Microbiol (Procházková et al., 2024) has established that intra-individual variations in gut microbiome homeostasis and metabolic activity significantly correlate with alterations in fecal water content and pH levels. Non-motor symptoms of PD are predominantly characterized by gastrointestinal dysfunction and may precede motor symptoms by many years.

Braak and colleagues postulated that α-syn pathology originates in the dorsal motor nucleus of the vagus nerve (DMNV) and the anterior olfactory nucleus. Through systematic postmortem neuropathological examinations of PD patients, they advanced the "gut-origin hypothesis" of PD pathogenesis (Klingelhoefer and Reichmann, 2015; Breen et al., 2019). This hypothesis proposes that pathological α-syn initially aggregates in the ENS and subsequently undergoes retrograde axonal transport via vagal nerve fibers to the substantia nigra and cerebral cortex. The hallmark neuropathological feature of PD involves the formation of Lewy bodies—intraneuronal cytoplasmic inclusions composed of misfolded α-syn aggregates. Current evidence suggests that these aberrant α-syn assemblies contribute to neurodegeneration in both PD patients and animal models through mitochondrial dysfunction-mediated pathways (Sato et al., 2013). The aggregated α-syn form fibrillar structures that propagate throughout distinct brain regions in a prion-like manner, exhibiting self-templating properties and cell-to-cell transmission characteristics (Peng et al., 2020). Trancikova et al. (2012) have proposed that mitochondrial dysfunction and oxidative stress associated with Parkinson's disease (PD) may contribute to pathological α-syn aggregation. Notably, the misfolding of α-syn has been identified as a primary causative factor underlying the degeneration of dopaminergic neurons in the substantia nigra of PD patients (Odaka et al., 2003). Accumulating evidence suggests that prodromal features of Parkinson's disease (PD), including constipation and α-syn pathology, can be detected several years prior to clinical diagnosis, potentially serving as early biomarkers for PD risk stratification (Pfeiffer, 2016). However, the precise timing of α-syn entry into the ENS relative to the onset of motor symptoms remains experimentally undetermined. Based on existing evidence, including the "Braak hypothesis" (Braak et al., 2006; Hawkes et al., 2010), we reasonably estimate that pathological α-syn likely infiltrates the ENS 10 to 20 years before motor manifestations of PD. Mounting experimental and neuropathological evidence demonstrates that α-syn pathology originates preferentially in the ENS, providing compelling support for the gut-to-brain propagation hypothesis via vagal nerve pathways in PD pathogenesis (Wang J. et al., 2022). A prospective study (Abbott et al., 2001), the Honolulu Heart Program (HHP), investigated bowel habits in 6,790 elderly men without PD. Participants were monitored for 24 years, during which 96 individuals developed PD after an average latency of 10 years (range: 5 months to 19 years). Subsequent autopsies of deceased PD patients and controls revealed phosphorylated α-syn deposits in both the central and enteric nervous systems during early disease stages, with distribution patterns closely aligning with Braak staging. Intriguingly, injection of pathological α-syn from PD patients into the gut of rodents demonstrated its transmission to the brain via the vagus nerve, ultimately leading to substantia nigra neuronal loss and motor deficits. We posit that presence does not equate to pathogenicity; while α-syn pathology in the enteric nervous system may represent an initial event, progression to clinically diagnosable disease requires prolonged accumulation and propagation. This emerging frontier necessitates further rigorous clinical investigation for validation.

Tang et al. (2013) provided mechanistic evidence that α-syn undergoes bidirectional transport across the BBB. Their findings further suggest that lipopolysaccharide (LPS) challenge may enhance α-syn brain uptake by compromising BBB integrity. In murine models, LPS exposure was shown to induce mitochondrial dysfunction through microglial NADPH oxidase activation, thereby initiating neurotoxic cascades (Hu and Wang, 2016). Consequently, LPS has been widely adopted in murine models to recapitulate Parkinsonian pathology. Emerging evidence also highlights an intriguing association between gut-derived proteins and cognitive function. Notably, bacterial production of amyloid proteins has been experimentally demonstrated to exacerbate α-syn pathology in aged rodent models.

Rotenone, a naturally occurring pesticide with high lipophilicity, readily crosses both the BBB and cellular membranes. These properties have established its widespread utility in generating experimental models of PD (Ibarra-Gutiérrez et al., 2023). Rotenone serves as a potent and selective high-affinity inhibitor of mitochondrial complex I (NADH: ubiquinone oxidoreductase) in the electron transport chain (Hineno et al., 2011), Inhibition of mitochondrial complex I activity induces a cascade of pathological events: Elevated reactive oxygen species (ROS) production; Impaired adenosine triphosphate (ATP) synthesis; Mitochondrial membrane depolarization. These alterations disrupt electron transfer from NADH to coenzyme Q, ultimately triggering apoptotic neuronal death (Ratan et al., 2024). Rotenone-induced PD models faithfully recapitulate both the behavioral phenotypes and neuropathological hallmarks observed in clinical PD. Notably, emerging evidence from chronic rotenone exposure models demonstrates that gastrointestinal dysfunction precedes motor impairments, further supporting the gut-origin hypothesis of PD pathogenesis (Bhattarai et al., 2021). Researches have demonstrated that chronic rotenone administration induces tyrosine hydroxylase-positive neuronal loss independent of gut microbiota status, while the subsequent development of motor deficits requires microbial colonization. Furthermore, their work revealed that prolonged rotenone exposure facilitates retrograde α-syn propagation to the CNS. Kim et al. (2019) provided experimental evidence that vagotomy effectively blocks the retrograde transport of α-syn fibrils to the brain, thereby preventing associated neurodegeneration and behavioral deficits in their PD model system. A research of professor Jintai Yu (Wu et al., 2025) teamidentified FAM171A2, a neuronal membrane protein, as a critical mediator of pathological α-syn propagation. Building on this discovery, they developed a novel FAM171A2-targeting therapeutic agent that effectively inhibits α-syn transmission throughout all stages of PD progression. Notably, within the context of gut microbiota-PD interactions, murine models of PD exhibit significant alterations in intestinal function that manifest prior to the onset of motor symptoms, further supporting the prodromal gut involvement in disease pathogenesis (Gries et al., 2021). Collectively, these findings provide compelling evidence for a robust association between specific gut microbiota compositions and the pathophysiology of PD.

The bidirectional communication between the human gut microbiota and the CNS, termed the MGBA, involves complex interactions across multiple pathways. The MGBA modulates host brain activity, behavior, and neurodevelopment through neural networks (e.g., enteric and central neural circuits), the neuroendocrine HPA axis, and immune signaling (Figure 2). This crosstalk is facilitated by vagal, sympathetic, and enteric neural fibers, with regulatory input from the HPA axis.

Given the correlation between gut microbiota and PD, replacing disrupted microbial communities with healthy microbiota consortia may represent a potential therapeutic strategy (de Groot et al., 2017). Recent literature suggests that microbial intervention holds substantial potential for modulating microbiota dysbiosis-driven neurodegenerative disorders. Among these approaches, fecal microbiota transplantation (FMT) represents an emerging and safe therapeutic strategy (Figure 4).

FMT represents the most effective approach for reconstructing a patient's overall gut microbiota. This technique involves transplanting functional microbial communities from healthy donor feces into the gastrointestinal tract of PD patients, thereby establishing a new gut microbiota ecosystem. The resulting enhancement of microbial diversity and functionality may contribute to effective PD treatment (Zhang et al., 2018). FMT is primarily administered via endoscopy, enema, or oral delivery of freeze-dried preparations. This therapeutic approach has now gained widespread recognition for its potential in treating neurological disorders. FMT may modulate the gut microbiota through neural, immune, and endocrine pathways, thereby influencing neurological symptoms. Currently, this technique is clinically applied in combination with antibiotics for treating Clostridioides difficile infection (Zhernakova et al., 2016; Boicean et al., 2023). FMT restores bacterial equilibrium in the gut, thereby reducing opportunities for pathogenic bacterial overgrowth. Previous studies have demonstrated that FMT significantly ameliorates gut microbial dysbiosis in PD mouse models, attenuates intestinal inflammation and barrier disruption, improves blood–brain barrier integrity, and reduces microglial and astrocytic activation in the nigrostriatal pathway. Furthermore, FMT suppresses neuroinflammation by downregulating TLR4/TNF-α signaling components in both gut and brain tissues, ultimately protecting dopaminergic neurons while increasing striatal dopamine (DA) and 5-HT levels (Sun et al., 2018; Zhang et al., 2021).

Currently, no single probiotic strain has been definitively proven to directly remodel dopaminergic neurons or ameliorate clinical manifestations of PD. However, emerging research has identified several key bacterial taxa that may exert neuroprotective effects through indirect mechanisms, representing promising candidates for microbiota-based PD intervention. We focus on four primary categories: SCFA (notably Butyrate), Bacteroides fragilis, Bifidobacterium, and Lactobacillus. Butyrate, as a pleiotropic metabolite, demonstrates potent anti-neuroinflammatory properties, enhances intestinal barrier integrity, and provides neuroprotection, thereby fostering a conducive microenvironment for dopaminergic neuron survival (Abreu et al., 2025). SCFAs derived from gut microbiota sustain host-microbial symbiosis by suppressing excessive immune responses and protecting commensal bacteria from elimination. Simultaneously, they enhance intestinal barrier stability and reduce permeability to prevent invasion by pathogenic microorganisms. Evidence indicates that SCFAs restrict neutrophil chemotaxis and monocyte–macrophage activity while promoting regulatory T cell (Treg) differentiation. Furthermore, SCFAs mitigate intestinal epithelial damage and modulate tight junction proteins to maintain barrier integrity (Wang et al., 2023). A 2024 randomized, double-blind, controlled clinical trial (Zeng et al., 2024) demonstrated that the combination of Bacteroides fragilis BF839 and earthworm protein effectively ameliorates both motor and non-motor symptoms in PD patients without significant adverse effects, offering a novel therapeutic approach. The underlying mechanism involves this formulation significantly increasing fecal Enterococcus abundance in PD patients, wherein Enterococcus has been shown to elevate striatal dopamine levels and improve behavioral phenotypes in PD mouse models. Bifidobacterium and Lactobacillus often function synergistically to produce neurotransmitters such as GABA and acetylcholine, while also facilitating the generation of L-tyrosine—a precursor to dopamine—thereby indirectly modulating neurological function (Yunes et al., 2016). Furthermore, Bifidobacterium alone exerts neuroprotective effects by preventing the reduction of spine density in Parkinson's disease models. PD models treated with Bifidobacterium also demonstrate improved motor symptoms and reduced TNF-α levels in the striatum (Li et al., 2022).

Recent studies demonstrate that FMT from healthy donors exhibits significant therapeutic effects in PD. Experimental investigations revealed that oral administration of 200 μL fecal bacterial suspension from healthy donors for 7 consecutive days induced the following physiological alterations in murine models: suppressed activity of the TLR4/TBK1/NF-κB/TNF-α signaling pathway in both gut and brain tissues; restored gut microbial homeostasis accompanied by decreased SCFAs levels; and reduced dopamine and 5-HT concentrations, ultimately leading to markedly attenuated glial cell activation. These findings provide novel insights for PD treatment strategies. Zhou et al. (2019) investigated an innovative dietary intervention strategy. Their study demonstrated that a fasting-mimicking diet (FMD) effectively inhibited the degeneration of dopaminergic neurons, thereby attenuating neurodegenerative progression. In the experimental paradigm, researchers transplanted gut microbiota from FMD-treated healthy mice into antibiotic-pretreated PD model mice, observing significantly elevated dopamine levels in the substantia nigra of recipient animals. This finding suggests that FMD may exert its neuroprotective effects through modulation of gut microbial communities. In 2021, Zhao et al. (2021) proposed the application of fecal microbiota transplantation (FMT) for therapeutic intervention in PD patients. Their experimental protocol involved daily FMT administration to 15 mice in the treatment group, with motor function assessments conducted at weeks 4 and 6 using standardized behavioral tests: rotarod test for motor coordination, adhesive removal test for sensorimotor integration, grip strength test for limb muscle strength, and pole test for movement balance. The results demonstrated significant improvement in motor symptoms in FMT-treated mice compared to rotenone-intoxicated controls. Furthermore, gastrointestinal function evaluations revealed that FMT treatment markedly increased fecal output in rotenone-induced PD mice. A recent Chinese study (Yang H. et al., 2024) administered fecal microbiota from PD patients to mice via oral gavage and systematically evaluated motor/intestinal functions, along with inflammatory and pathological changes in both colon and brain tissues. Fecal microbiota from PD patients exacerbated inflammation and neurodegeneration in A53T mice. This was evidenced by aggravated gut inflammation, impaired intestinal barrier function, enhanced microglial and astrocytic activation, aberrant α-synuclein deposition, and loss of dopaminergic neurons in the brain.

Since 2017, levodopa-carbidopa intestinal gel (LCIG) delivered via percutaneous endoscopic gastrostomy (PEG) tubes has been clinically implemented for PD treatment (Bernier et al., 2017). This therapeutic approach involves continuous intestinal levodopa infusion through an enteral feeding system with intragastric placement. The standardized protocol commences with enteral nutrition initiation using a 15-Fr (5-mm) LCIG PEG-J tube via the gastric port, followed by nocturnal nutrition delivery through a feeding pump while temporarily suspending LCIG infusion. Over subsequent weeks, five daily bolus administrations are gradually introduced alongside continuous LCIG delivery. While particularly beneficial for advanced PD patients presenting with dysphagia, this treatment modality may require discontinuation due to complications including severe gastroparesis or new-onset aspiration pneumonia, ultimately necessitating surgical jejunostomy to establish alternative jejunal feeding access. In 2021, Freire-Alvarez et al. (2021) similarly demonstrated the efficacy of LCIG in managing motor complications in advanced PD patients. In 2018, another study (Fernandez et al., 2018) employed continuous levodopa administration via PEG-J delivery of LCIG, achieving stable plasma levodopa concentrations and reduced motor fluctuations in advanced PD cases. While this approach significantly decreased 'off' time in PD patients, 94% of participants nevertheless experienced device-related complications and adverse events. A 2023 report (Chernova et al., 2023) described a technical modification involving LCIG administration via PEG-J tubes for modulating gut microbiota in PD patients. LCIG infusion represents an effective symptomatic treatment for motor fluctuations refractory to conventional oral therapy. However, despite LCIG treatment, the patient exhibited progressive symptom deterioration characterized by persistent dyskinesias and off periods, accompanied by apathy, depression, and constipation. FMT has demonstrated significant efficacy in alleviating PD-associated constipation, characterized by an increased abundance of Firmicutes and decreased proportions of Proteobacteria and Bacteroidetes in treated patients. These microbial compositional changes correlate with effective amelioration of both constipation and tremor symptoms (Huang et al., 2019). Multiple studies comparing fecal samples between Parkinson's disease patients and healthy controls have consistently revealed a characteristic pattern: PD patients typically exhibit a reduced Firmicutes/Bacteroidetes ratio. In 2015, Keshavarzian et al. (2015) conducted differential analysis of mucosal and fecal microbial communities in PD patients, definitively demonstrating a significant decrease in the Firmicutes/Bacteroidetes ratio. This dysbiotic signature was significantly correlated with α-syn misfolding and increased intestinal permeability, providing direct microbiological and pathological evidence supporting the "gut-origin" hypothesis of PD. A 2022 meta-analysis by Nishiwaki et al. (2022) systematically synthesized multiple studies and conclusively demonstrated that PD patients exhibit a consistent reduction in butyrate-producing Firmicutes alongside an increase in mucin-degrading Bacteroidetes within the gut microbiota. These findings directly support the mechanistic link between a decreased Firmicutes/Bacteroidetes ratio and both intestinal barrier disruption and exacerbated inflammation. Furthermore, this dysbiotic pattern was correlated with higher UPDRS-III scores, indicating more severe motor symptoms. In 2021, a study demonstrated that FMT significantly reduces Bacteroides abundance while increasing Prevotella and Blautia populations in PD patients with constipation. These microbial alterations were associated with marked improvements in both PAC-QOL (Patient Assessment of Constipation Quality of Life) and Wexner constipation scores, indicating substantial relief of constipation symptoms (Kuai et al., 2021). In 2021, Segal et al. (2021) conducted colonoscopic infusion of donor FMT in six PD patients, demonstrating significant improvements in motor symptoms, non-motor symptoms, and constipation over a six-month observation period. Two years later, DuPont et al. (2023) conducted a 12-week randomized, controlled, double-blind study involving 12 patients with mild-to-moderate PD. The study demonstrated that oral administration of freeze-dried FMT products effectively alleviated constipation and improved intestinal motility in PD patients by enhancing gut microbiome diversity. Several months later, Cheng et al. (2023) developed a novel fecal microbiota powder capsule formulation, which was subsequently validated in clinical trials to improve autonomic nervous system and gastrointestinal symptoms in PD patients. In 2024, Bruggeman et al. (2024) conducted a nasoduodenal FMT trial involving 43 early-stage PD patients. Post-transplantation assessments revealed significant improvements in both motor symptoms and constipation compared to baseline. Notably, the study also demonstrated FMT-mediated enhancement of cognitive functions, including amelioration of anxiety, depressive symptoms, and sleep disturbances. However, this investigation had several limitations. Specifically, since early-stage PD patients typically receive levodopa-based pharmacotherapy, it was methodologically challenging to completely eliminate confounding effects attributable to levodopa treatment in this cohort. Some researchers propose that the partial therapeutic effects of FMT on PD symptoms may be attributed to reduced abundance of tyrosine decarboxylase-producing bacteria in feces, consequently enhancing levodopa bioavailability (Beckers et al., 2024). A recent study (Yang X. et al., 2024) employed fucoidan-assisted FMT in a PD mouse model, demonstrating successful mitigation of both peripheral and central inflammation alongside amelioration of dopaminergic neuronal damage (Table 2 comprehensively summarizes animal studies and clinical trials investigating FMT for PD treatment).

Current evaluations of FMT efficacy for PD symptom improvement across various administration routes primarily utilize the Hoehn-Yahr (H-Y) staging, Unified Parkinson's Disease Rating Scale (UPDRS), and Non-Motor Symptoms Questionnaire (NMSS) to assess both motor and non-motor symptom improvements. However, existing clinical studies provide limited evidence to determine the optimal administration route. Experimental models have not yet established whether microbial-host interactions in the colon versus small intestine are more relevant to PD pathology, though different FMT routes may preferentially affect specific gastrointestinal regions. Given that most PD patients present with both motor and non-motor symptoms, effective management of these manifestations remains a clinical priority. Based on comparative analysis of FMT delivery methods, colorectal FMT infusion appears to be the most efficacious and durable approach for PD treatment. This method demonstrates comprehensive improvements in motor/non-motor symptoms and constipation relief in most patients. However, potential levodopa bioavailability alterations may exacerbate dyskinesias, necessitating further validation in larger cohorts (Segal et al., 2021). Additionally, while colorectal FMT infusion is generally safe, its invasive nature limits patient acceptability. For practicality, oral FMT capsules represent an emerging and more acceptable alternative, particularly for gastrointestinal symptom relief, though their efficacy for motor symptom improvement currently remains inferior to colorectal administration.

In summary, FMT demonstrates therapeutic potential in ameliorating intestinal dysfunction, modulating neuroinflammation, and regulating neurotransmitter activity, thereby conferring neuroprotective effects. Compared to conventional PD medications, FMT exhibits superior safety profiles and broader mechanisms of action. However, various factors may influence recipient outcomes. Notably, a recent study (D'Amato et al., 2020) revealed that FMT from aged donor mice impaired spatial learning and memory in young recipients, while transplantation from aged mouse models to germ-free (GF) mice reduced fecal short-chain fatty acid (SCFA) production, exacerbating depression-like behaviors and short-term memory deficits. Current research still lacks standardized donor screening protocols, and inadequate preparation may potentially yield countertherapeutic effects. The existing FMT technique involves transplanting fecal microbiota into recipient intestines, which reduces lipopolysaccharide (LPS) levels in both the colon and serum. This reduction subsequently inhibits the TLR4/MyD88/NF-κB signaling pathway and its downstream proinflammatory mediators in colonic tissues (Zhao et al., 2021). Furthermore, some researchers have proposed the application of Artificial Intelligence -guided imaging sensors for non-invasive monitoring and diagnosis of PD (Li et al., 2024). However, this technology remains in its nascent stages and requires further research for optimization and refinement.

Notably, several critical parameters—including optimal timing, dosage, donor characteristics, and patient selection criteria may serve as confounding factors influencing FMT efficacy. To optimize FMT's therapeutic potential for PD treatment and achieve optimal clinical outcomes, ongoing research addressing these variables must be pursued (Donnelly et al., 2020). Furthermore, the efficacy of FMT is influenced not only by the gut microbiota composition of the donor, but also by that of the recipient. Additional factors, including the administration method and preservation techniques of FMT preparations, also significantly contribute to treatment outcomes (Ng S. C. et al., 2020). Furthermore, the therapeutic efficacy of FMT varies substantially across different administration routes when treating conditions such as recurrent Clostridioides difficile infection, inflammatory bowel disease (IBD), diarrhea, and constipation (McGowan et al., 2020). A 2020 domestic clinical study (Xue et al., 2020) conducted a self-controlled trial involving 15 PD patients receiving FMT, with 10 patients administered FMT via the colonic route and 5 via the nasoduodenal route. The results demonstrated that colonoscopic FMT significantly improved various clinical assessment scores in PD patients, whereas nasoduodenal administration showed suboptimal therapeutic efficacy. Consequently, modern clinical practice preferentially employs the colonic route for FMT in PD treatment to achieve satisfactory therapeutic outcomes. Although FMT has emerged as an innovative therapeutic approach for PD, current clinical evidence remains insufficient. Comprehensive investigations are still required to address several critical aspects, including donor selection criteria, standardized processing of transplant specimens, duration of therapeutic effects, and optimal intervention timing. These efforts aim to minimize potential safety concerns associated with FMT, particularly during the COVID-19 pandemic (Wu et al., 2020). Furthermore, a study investigating dietary interventions and FMT (Kennedy et al., 2025) revealed that fecal microbiota transplantation exhibits limited efficacy when administered in suboptimal dietary contexts. These findings suggest that optimal therapeutic outcomes may require the synergistic combination of FMT with appropriate dietary modifications, presenting a significant challenge for future research in this field. Moreover, distinct patterns of gut microbiota dysbiosis have been identified across different PD subtypes, necessitating systematic investigations to validate the feasibility and efficacy of personalized FMT protocols tailored to individual patient profiles.