Years before the tremor: The gut clues Parkinson’s researchers are trying to decode

Richard Moore
13 hours ago
28 min read

The first sign was constipation.

For decades.

Not a tremor. Not stiffness. Not a change in walking. Just a sluggish bowel that, at the time, seemed ordinary enough to ignore.

Many people who later receive a Parkinson’s disease diagnosis can trace, in retrospect, a long-running familiarity with constipation, sometimes ten, fifteen or even twenty years before anything trembled or stiffened. Clinicians, looking back through the records, often see the same quiet trajectory (Pfeiffer et al., 2024).

It is a strange place for a story about a brain disease to begin.

But several lines of Parkinson’s research now point to the gut as an early site of interest: long-standing constipation, abnormal α-synuclein in gut tissue, vagus nerve studies, altered microbial metabolites and reproducible Parkinson’s-associated microbiome patterns (Stokholm et al., 2016; Wallen et al., 2022; Borghammer et al., 2024; Pfeiffer et al., 2024; Menozzi et al., 2026).

This is the first of two articles on Parkinson’s disease and the gut microbiome: the trillions of bacteria, viruses, fungi and other microorganisms that live in the gut. Here, the focus is the science. The aim is to separate findings supported by human data, such as constipation, altered gut microbial composition and abnormal α-synuclein in the gut, from hypotheses that remain under investigation, such as whether microbiome changes actively contribute to disease progression (Stokholm et al., 2016; Romano et al., 2021; Wallen et al., 2022; de Castro Fonseca et al., 2026).

The second article will move from mechanism to practice, looking at what this science may mean for food, fibre, probiotics, constipation, levodopa timing and daily life.

For anyone reading this with Parkinson’s, or caring for someone who is, this point needs saying clearly.

This is not a story about blame. Parkinson’s is not caused by eating the wrong foods, neglecting the gut, or failing to take probiotics. Current human intervention studies do not show that microbiome-directed treatments stop or slow Parkinson’s disease progression; the two published randomised faecal microbiota transplantation trials report mixed findings, and probiotic and prebiotic studies have mainly focused on constipation or gastrointestinal outcomes rather than disease modification (Bruggeman et al., 2024; Scheperjans et al., 2024; Pfeiffer et al., 2024).

The evidence leads to a narrower question: why do constipation, gut microbial changes, altered microbial metabolites and abnormal gut α-synuclein appear so often in Parkinson’s research, including in some people who may be in the earliest or pre-diagnostic stages of the disease (Stokholm et al., 2016; Huang et al., 2023; Pfeiffer et al., 2024; Menozzi et al., 2026)?

Parkinson’s does not stay in the brain

Parkinson’s disease is usually described as a brain disease, and for good reason. It is defined by the slow loss of dopamine-producing neurones in a region of the midbrain called the substantia nigra, Latin for “black substance”, so called because it looks dark to the naked eye due to a pigment called neuromelanin. Parkinson’s is also marked by the build-up of a misfolded form of a protein called α-synuclein (de Castro Fonseca et al., 2026).

Picture α-synuclein as a strand of correctly folded origami. In its healthy form, it sits neatly inside nerve endings, helping the brain release its chemical messengers. When one strand folds the wrong way, it can press against its neighbours and pull them into the same wrong shape. Eventually, these misfolded strands can clump together into deposits called Lewy bodies, one of the pathological hallmarks seen in Parkinson’s brains.

The movement symptoms most people associate with Parkinson’s, such as resting tremor, rigidity, slowed movement and changes in walking, usually appear only after a substantial fraction of dopamine-producing cells has already been lost (Menozzi et al., 2026).

But Parkinson’s does not behave like a disease confined to the brain. Gastrointestinal dysfunction, autonomic symptoms, sleep disturbance, reduced sense of smell and other non-motor features can occur before or alongside motor symptoms (Pfeiffer et al., 2024; Borghammer et al., 2024; Menozzi et al., 2026).

Between 60 and 80 percent of people with Parkinson’s have gastrointestinal symptoms, most often constipation, but also delayed stomach emptying, abdominal discomfort and increased intestinal permeability, meaning a change in how tightly the gut barrier controls what passes through it (de Castro Fonseca et al., 2026). Many of these symptoms appear years before any motor sign. In the large Birmingham-based cohort published by Wallen et al. (2022), chronic constipation was six times more common in people with Parkinson’s than in neurologically healthy controls of similar age.

Constipation is not a footnote in Parkinson’s. It is one of its earliest and most consistent features (Pfeiffer et al., 2024).

The gut symptoms matter because they suggest Parkinson’s also involves the autonomic and enteric nervous systems. The autonomic nervous system controls largely unconscious functions such as gut movement, heart rate and blood pressure. The enteric nervous system is the nerve network embedded in the gut wall itself. This has pushed researchers toward a sharper question: does the gut do something, or fail to do something, that is relevant to how Parkinson’s begins or progresses?

The Vagus Nerve clue

If the gut is involved early, the most obvious physical route between gut and brain is the vagus nerve.

The vagus is the body’s longest cranial nerve. It runs from the brainstem down through the chest and into the abdomen. In simple terms, it is a two-way communication line between the gut and the brain.

One of the most discussed human clues comes from a Swedish population-based study by Liu et al. (2017). The researchers looked at 9,430 people who had previously undergone vagotomy, an operation that cuts part or all of the vagus nerve and was historically used to treat peptic ulcers. They compared them with 377,200 matched controls.

The results were not simple. When all types of vagotomy were grouped together, people who had the surgery did not have a clearly lower risk of Parkinson’s disease. But when the researchers looked specifically at truncal vagotomy, the more complete operation that cuts the main trunk of the vagus nerve, a pattern appeared. People who had this more complete cut at least five years before any Parkinson’s diagnosis were less likely to be diagnosed with Parkinson’s later on (Liu et al., 2017).

In plain English, the more completely the gut-to-brain nerve route had been interrupted, the lower the later Parkinson’s risk appeared to be. That does not prove the vagus nerve carries Parkinson’s pathology from gut to brain, but it is one reason researchers take the gut-to-brain route seriously. A less complete operation, selective vagotomy, where more of the nerve is left intact, did not show the same association (Liu et al., 2017).

The finding catches the eye, but it cannot carry the argument by itself. It is observational, meaning it can show a pattern but cannot prove cause and effect. It also cannot rule out the possibility that people who needed vagotomy differed from controls in ways the researchers could not fully measure.

The reason this study matters is not that vagotomy prevents Parkinson’s. It does not prove that. The value of the finding is more specific: when the main gut-to-brain nerve route had been interrupted, later Parkinson’s risk appeared lower in one large observational study. That supports the idea that, in at least some people, signals or pathology connected to Parkinson’s may travel between the gut and the brain through the vagus nerve (Liu et al., 2017; Borghammer et al., 2024).

This becomes harder to dismiss when placed beside the imaging work by Horsager et al. (2020), which suggested that Parkinson’s may have at least two broad anatomical patterns. In one, described as brain-first, α-synuclein pathology appears to begin in the brain and spread outward to the autonomic system. In the other, described as body-first, it appears first in the enteric or peripheral autonomic nervous system and travels inward to the brain.

The neatest way to picture it is two trains starting at opposite ends of the same line and arriving at the same station. One leaves from the gut. The other leaves from the brainstem. Both may eventually arrive at motor symptoms. But the journey, and what shows up along the way, looks different.

Borghammer et al. (2024), updating this synthesis with newer autopsy and imaging data, concluded that the body-first / brain-first distinction is currently one of the best frameworks for thinking about why Parkinson’s looks different from person to person in its earliest phases. The body-first idea makes a testable prediction: people whose disease begins outside the brain should be more likely to develop autonomic and brainstem-related features before motor symptoms.

These features include rapid eye movement sleep behaviour disorder, a sleep disorder in which vivid dreams are physically acted out because the muscle paralysis that normally protects us during dreaming has failed; constipation; reduced sense of smell; and certain forms of depression and anxiety (Menozzi et al., 2026).

Taken together, the evidence on constipation, gut α-synuclein, vagal pathways and prodromal microbiome signatures makes the gut a serious part of Parkinson’s research, rather than a peripheral symptom category (Stokholm et al., 2016; Huang et al., 2023; Borghammer et al., 2024; Pfeiffer et al., 2024; Menozzi et al., 2026).

The Parkinson’s protein found in the gut

If the gut is upstream in some Parkinson’s, the disease’s signature protein should be findable there.

In some studies, it is.

Stokholm et al. (2016) examined archived gastrointestinal tissue from people who later went on to develop Parkinson’s, on average around seven years after the biopsy. Phosphorylated α-synuclein, the abnormal form associated with Lewy bodies in the brain, was present in 56 percent of these prodromal cases, compared with 26 percent of controls.

The sample was small and tissue staining methods have limits. But the finding remains important because it suggested that abnormal α-synuclein may be detectable in the gut before Parkinson’s is clinically diagnosed.

A newer study by Shin et al. (2024) added a useful detail. The researchers looked at gastrointestinal tissue from people with Parkinson’s and matched controls and found that α-synuclein accumulation was more often detected in the upper gastrointestinal tract, including the stomach, than in the lower gastrointestinal tract. In plain English, if researchers are looking for Parkinson’s-related α-synuclein in the gut, the stomach may be one of the more informative places to look. The study also reported that conventional tissue staining was not accurate enough to become a simple diagnostic test on its own (Shin et al., 2024).

Methods have since become sharper. A newer generation of laboratory tests, called seed amplification assays, can detect tiny quantities of misfolded α-synuclein by giving them the chance to “seed” the misfolding of normal α-synuclein in a test tube. The idea is similar to watching whether one misshapen paperclip can start bending a whole batch into the same shape.

Using this technique, Shin et al. (2025) found pathological α-synuclein seeding activity in stomach biopsies from 45 percent of people with early Parkinson’s and in none of the controls. Patients with more gastric seeding activity scored worse on cognitive testing, even at an early stage of disease. The numbers were small, 22 patients and 17 controls, and the findings need replication. Even so, they add sharper human evidence to the older observation: in at least some people, Parkinson’s-related α-synuclein pathology is not confined to the brain.

The reason this matters is simple. If Parkinson’s-related α-synuclein can be detected in gut tissue before diagnosis, then the gut may not only be affected after Parkinson’s develops. In some people, it may be one of the places where Parkinson’s-related biology is visible early. That does not prove the disease starts in the gut, but it supports the body-first model in which pathology may appear in the gut or autonomic nervous system before reaching the brain (Stokholm et al., 2016; Borghammer et al., 2024; Shin et al., 2024; Shin et al., 2025).

Animal studies go further, although they must be read as models, not as direct proof of human Parkinson’s disease. In animal models, when α-synuclein fibrils are injected into the gut, pathology can spread along the vagus nerve to the brain, and cutting the vagus nerve can stop it (de Castro Fonseca et al., 2026). When mice genetically engineered to overproduce human α-synuclein are kept germ-free, they have far fewer Parkinson’s-like motor and neuroinflammatory features than mice with normal gut microbes. Recolonisation reinstates the phenotype, and so does exposure to certain bacterial metabolites (Sampson et al., 2016). Specific bacterial amyloids, particularly curli produced by some strains of Escherichia coli, better known as E. coli, can accelerate α-synuclein aggregation and motor symptoms in those same mice (Sampson et al., 2020).

These animal experiments do not show that gut bacteria cause Parkinson’s in humans. They show something narrower and still important: gut microbes can influence α-synuclein biology, immune activation and Parkinson’s-like features in experimental systems. That makes the human microbiome findings harder to dismiss as incidental.

The microbial pattern that keeps appearing

When researchers sequence bacteria in the stool of people with Parkinson’s and compare them with controls, the findings can look messy at first.

One study reports one bacterial group as higher. Another reports it as lower. A third finds no clear difference. That can make the field look more confused than it really is.

The problem is that the microbiome is not like a blood test where one number goes up or down. It is an ecosystem. The same broad groups of microbes may be present in many people, but their proportions, genes and chemical outputs can differ. A forest is still a forest whether it contains more oak, more pine or more birch. But if the soil becomes poorer, the leaf litter changes and opportunistic plants begin to dominate, the ecology of the forest has shifted.

That is the clearer message from Parkinson’s microbiome research. The Parkinson’s gut microbiome is not defined by one bacterium. It is defined by a recurring change in microbial ecology and microbial function (Romano et al., 2021; Wallen et al., 2022; Boktor et al., 2023; Nishiwaki et al., 2024).

The first repeated finding is a reduction in bacteria that produce short-chain fatty acids. Short-chain fatty acids are small molecules, such as butyrate and acetate, that gut bacteria make when they ferment certain fibres in the colon. They help nourish the cells lining the bowel, support the mucus layer and influence immune signalling. Across Parkinson’s studies, several short-chain-fatty-acid-producing bacteria, including Faecalibacterium, Roseburia and members of the Lachnospiraceae family, are often reduced (Romano et al., 2021; Wallen et al., 2022; Nishiwaki et al., 2024).

For the reader, the important point is not the bacterial names. It is the function they represent. If these bacteria are lower, the gut may have less capacity to convert fermentable fibres into compounds that help maintain the bowel lining and regulate local inflammation. That fits with the wider pattern of constipation, altered gut barrier markers and immune activation seen in Parkinson’s research, although it does not prove that the microbiome change caused those features (Aho et al., 2021; Chen et al., 2022; de Castro Fonseca et al., 2026).

The second repeated finding is an increase in some opportunistic or mucin-degrading bacteria. “Mucin-degrading” means these bacteria can use mucus from the gut lining as an energy source. That is not automatically bad. Some mucus-associated bacteria can be part of a healthy gut ecosystem. But in Parkinson’s studies, increases in organisms such as Akkermansia and certain Lactobacillus species often appear alongside reductions in short-chain-fatty-acid-producing bacteria and other signs of a shifted gut environment (Romano et al., 2021; Wallen et al., 2022; Nishiwaki et al., 2024).

The third finding is that newer studies are moving beyond “who is there?” and asking “what can they do?” Earlier studies often used broad bacterial sequencing methods that could identify general microbial patterns but not much functional detail. Newer metagenomic studies can look more closely at microbial genes, giving researchers a better sense of what the gut microbiome may be capable of producing or breaking down (Boktor et al., 2023; Nishiwaki et al., 2024).

That shift matters for Parkinson’s because the functional findings are more biologically interesting than a list of bacterial names. Boktor et al. (2023), pooling metagenomic data from three continents, identified recurring functional changes in the Parkinson’s gut microbiome. Nishiwaki et al. (2024), in a later shotgun sequencing meta-analysis, reported that bacterial genes involved in making several B vitamins, including riboflavin and biotin, were depleted in Parkinson’s. Wallen et al. (2022), in the largest single-cohort metagenomic study to date, found that more than 30 percent of tested microbial species, genes and pathways differed between people with Parkinson’s and neurologically healthy elderly controls.

Put into plain English, the Parkinson’s microbiome appears to show three broad shifts: fewer microbes linked with fibre fermentation and short-chain fatty acid production; more microbes and microbial components that may be inflammatory or opportunistic; and altered microbial genes involved in metabolism, neuroactive signalling and nutrient-related pathways (Wallen et al., 2022; Boktor et al., 2023; Nishiwaki et al., 2024).

That does not mean the microbiome is causing Parkinson’s. It means the gut ecosystem in Parkinson’s repeatedly looks less like a stable, fibre-fermenting, barrier-supporting ecosystem and more like one under pressure. Whether that pressure is a driver of disease, a consequence of slow gut transit and medication, or both, is still being tested.

Before the diagnosis

Two recent studies have pushed the question backwards in time, asking whether this microbiome pattern shows up before Parkinson’s is diagnosed.

Huang et al. (2023) compared the gut microbiomes of people with early Parkinson’s, people with isolated rapid eye movement sleep behaviour disorder and first-degree relatives of people with rapid eye movement sleep behaviour disorder. This sleep disorder is one of the strongest known warning signs for a later synuclein-related disease, including Parkinson’s. A Parkinson’s-like dysbiosis, meaning an imbalanced microbial community, was already present in the rapid eye movement sleep behaviour disorder group and was emerging in their relatives, particularly younger relatives. The pattern was not identical to overt disease, but it travelled in the same direction (Huang et al., 2023).

Menozzi et al. (2026) went further. Their cohort included 271 people with Parkinson’s, 43 carriers of variants in the glucocerebrosidase gene, known as GBA1, who had no Parkinson’s symptoms, and 150 healthy controls. GBA1 is the most common genetic risk factor for Parkinson’s, although many carriers never develop the disease. About a quarter of the variation in the gut microbiome of unaffected GBA1 carriers sat between healthy controls and Parkinson’s patients, and the size of this shift tracked with prodromal symptom burden, including autonomic dysfunction, depression and sleep features. The same intermediate microbiome pattern showed up in about 20 percent of healthy controls without known genetic risk. The direction of these findings was reproduced across three independent cohorts from the United States, Turkey and Korea, involving a further 638 people with Parkinson’s and 319 controls (Menozzi et al., 2026).

These studies are important because they move the microbiome question earlier in the timeline. If Parkinson’s-like microbiome patterns are already present in people with rapid eye movement sleep behaviour disorder, unaffected GBA1 carriers or some healthy controls, then microbiome changes cannot be dismissed as only the result of years of Parkinson’s medication or advanced disease. The limitation is equally important: these studies do not yet prove that the microbiome predicts Parkinson’s or causes it. They show that the microbiome may be part of the pre-diagnostic biology in some people (Huang et al., 2023; Menozzi et al., 2026).

The Menozzi study is cross-sectional, meaning it takes a snapshot at one point in time. We do not yet know how many of the healthy controls with a Parkinson’s-like microbiome will go on to develop Parkinson’s. Longitudinal follow-up, where people are followed over time, is still needed. The finding should be taken seriously. What it means for one person sitting in a clinic is still unknown.

The chemical messages gut bacteria leave behind

It is tempting to think of the microbiome as a list of species. Functionally, what matters is what those species produce, and what they fail to produce.

Two families of microbial metabolites have drawn particular attention in Parkinson’s. Metabolites are small molecules produced during biological activity. In this case, they are chemicals made or modified by gut microbes. The first are short-chain fatty acids, especially butyrate. These fatty acids are, in a sense, the rent gut bacteria pay to live in the colon. They feed the cells lining the gut, support the mucus barrier and influence inflammation locally and systemically.

Aho et al. (2021) found reduced faecal short-chain fatty acids in people with Parkinson’s. Chen et al. (2022) added a more complicated picture: faecal short-chain fatty acids were lower in Parkinson’s, but blood-plasma short-chain fatty acids were higher, and faecal butyrate was lower in people with worse motor severity, longer disease duration and more severe constipation.

So the finding is not simply that people with Parkinson’s have “too little butyrate” or “not enough fibre fermentation.” The more complicated picture is that these bacterial products may be lower in stool but higher in blood. That makes simple supplement claims about fibre, butyrate or “microbiome support” too neat for the evidence (Chen et al., 2022).

The second family is bile acids. Bile acids are salts the liver makes to help digest fat. As they pass through the gut, microbes can modify them, a little like a postal sorting office changing the destination of a package after it has already been sent.

Yan et al. (2022), in a study that measured several biological layers at once, found changes in blood and stool bile acids in Parkinson’s that were statistically linked to specific microbial groups. Shen et al. (2024) took a rarer angle: they looked at blood metabolites measured, on average, eight years before Parkinson’s was diagnosed. Thirteen microbial metabolites, including bile acids, indoles and amino-acid-related compounds, were loosely associated with later Parkinson’s risk. However, none survived strict correction for multiple testing, which means the findings should not be treated as confirmed markers of future Parkinson’s (Shen et al., 2024).

The pattern suggested that gut microbial chemistry may be altered years before diagnosis, particularly in pathways linked to fibre fermentation, short-chain fatty acid production and amino-acid breakdown (Shen et al., 2024). Its value is the timing: metabolites were measured before Parkinson’s diagnosis, which gives the study a prospective element missing from most case-control microbiome research. But the findings are hypothesis-generating, not clinically actionable.

The wider message is that Parkinson’s microbiome research is moving beyond “which bacteria are present?” and towards “what chemicals are those bacteria producing?” That is important because microbial products such as short-chain fatty acids and bile acids can interact with the gut barrier, immune system and nervous system. The current evidence shows altered microbial chemistry in Parkinson’s; it does not yet show that correcting those chemical patterns changes the course of the disease (Aho et al., 2021; Chen et al., 2022; Yan et al., 2022; Shen et al., 2024; de Castro Fonseca et al., 2026).

The gut barrier question

The phrase “leaky gut” has been worked hard in popular health writing. In Parkinson’s research, intestinal permeability is a more specific and measurable idea.

The gut lining is a single layer of cells held tightly together by protein junctions, like rows of bricks held in place by mortar. When that mortar weakens, the wall still stands, but more substances may slip through the gaps than should. Schwiertz et al. (2018) measured faecal zonulin, a proposed marker of gut barrier function, and two markers linked with intestinal inflammation: faecal calprotectin and α1-antitrypsin. All three were elevated in people with Parkinson’s compared with controls. Aho et al. (2021) found related patterns in blood markers of gut barrier integrity.

The proposed mechanism is that altered gut barrier function may increase immune exposure to bacterial products such as lipopolysaccharide. Lipopolysaccharide is a component of the outer wall of certain bacteria that the immune system may read as an “invader” signal. This could link gut inflammation, peripheral immune activation and neuroinflammatory pathways in Parkinson’s disease, but it remains a mechanistic hypothesis rather than proven clinical causality (Aho et al., 2021; de Castro Fonseca et al., 2026).

This is where language matters. “Leaky gut” is often used online as if it explains almost every chronic disease. Parkinson’s research does not support that kind of claim. What it does support is more restrained: several studies have found markers consistent with altered gut barrier function and intestinal inflammation in Parkinson’s. Whether those changes help drive Parkinson’s biology, result from slowed gut transit and inflammation, or both, is still unresolved (Schwiertz et al., 2018; Aho et al., 2021; de Castro Fonseca et al., 2026).

When the gut talks to the immune system

Parkinson’s disease has a substantial immune component. People with the disease show heightened T-cell responses against α-synuclein fragments compared with healthy individuals. T cells are immune cells that help identify and respond to threats. Immune-related gene variants, especially in a region of the genome involved in immune recognition, are also associated with Parkinson’s risk (de Castro Fonseca et al., 2026).

Microglia, the brain’s resident immune cells, are activated in Parkinson’s. So are peripheral immune cells outside the brain. One way to picture this: the immune system is normally a careful security guard who knows the difference between a tenant and an intruder. In Parkinson’s, it appears to have spent too long studying a particular wanted poster, α-synuclein, and begins reacting as though some of the body’s own α-synuclein-containing cells are suspects. Whether the microbiome helps train the guard to behave this way is one of the major questions in the field.

In mice, certain gut microbial communities are required for α-synuclein-driven motor deficits and neuroinflammation; in their absence, those features are blunted (Sampson et al., 2016). In humans, the composition of the gut microbiome is known to shape systemic immunity, but Parkinson’s-specific causality is not yet proven. The 2026 Lancet Neurology Personal View argues that a gut–immune–brain axis is one of the most clinically relevant mechanisms currently under investigation: microbial changes may contribute to peripheral inflammation and T-cell activation that, in genetically susceptible people, could interact with neurodegenerative processes (de Castro Fonseca et al., 2026).

This is why immunity belongs in a microbiome article. The gut microbiome does not communicate with the brain only through nerves. It also communicates through the immune system. If gut microbes, gut barrier changes or bacterial products alter immune tone, they could theoretically influence the inflammatory environment around Parkinson’s-related α-synuclein pathology. That remains a hypothesis in humans, but it is one of the most biologically coherent explanations linking gut microbes to neurodegeneration (Sampson et al., 2016; Aho et al., 2021; de Castro Fonseca et al., 2026).

Why Parkinson’s microbiome studies disagree

Anyone who reads several Parkinson’s microbiome papers will notice the disagreements. One study finds a particular group of gut bacteria lower in Parkinson’s. Another study does not find the same pattern.

Trying to read the literature can feel like trying to read a paragraph through a window while someone taps on the glass. You can see most of the words, but it is hard to tell which jitters belong to the writing and which belong to the tapping.

The disagreement is partly methodological and partly biological: constipation, medication exposure, diet, geography, sequencing method and disease stage can all alter microbiome results (Wallen et al., 2022; Boktor et al., 2023; Kwon et al., 2024; Yoon et al., 2024). Constipation reshapes the microbiome. Slowed gut transit favours certain bacteria over others; constipation itself is associated with depletion of spore-forming and short-chain-fatty-acid-producing taxa, independent of Parkinson’s (Wallen et al., 2022). Some “Parkinson’s microbiome” signals are partly constipation signals.

Medication matters too. Levodopa, catechol-O-methyltransferase inhibitors, monoamine oxidase B inhibitors, laxatives, proton pump inhibitors and antibiotics can all affect the gut ecosystem. Diet matters: in Parkinson’s populations, dietary quality, fibre intake and added-sugar intake explain a meaningful share of microbiome variability (Kwon et al., 2024; Yoon et al., 2024). Geography matters because microbiomes in Helsinki, Birmingham, Shanghai, Sydney and Seoul do not start from identical baselines (Boktor et al., 2023). Sequencing method matters: broad bacterial sequencing methods cannot reliably resolve species, whereas newer whole-genome sequencing methods can show more detail about which microbes are present and what they may be capable of doing. Disease stage matters too; the microbiome of early, drug-naïve Parkinson’s is not identical to that of advanced Parkinson’s, and most studies are cross-sectional snapshots.

The Boktor et al. (2023) and Nishiwaki et al. (2024) meta-analyses, by pooling cohorts and explicitly modelling some of these variables, recover a more stable signal than most individual studies can. That is why the better question is no longer only “which bacteria are different?” but “which microbial functions keep appearing across cohorts once the obvious confounders are accounted for?”

Can a stool test predict Parkinson’s?

Not yet, and not in the way some commercial pitches imply.

Lubomski et al. (2022) combined microbiome features and dietary inputs in 103 people with Parkinson’s and 81 controls. Their model could distinguish Parkinson’s from control samples moderately well, but not with the level of accuracy needed for diagnosis.

Put plainly, the model was better than chance, but not accurate enough to diagnose Parkinson’s in real clinical life. It has also not yet been validated over time in people who may be in the earliest stages of the disease (Lubomski et al., 2022; Menozzi et al., 2026). Menozzi et al. (2026) extends the question into the prodromal phase, but the authors are explicit that longitudinal validation is required before any of this becomes a screening tool.

There is no current clinical test in which a stool sample tells you whether you will develop Parkinson’s, and no validated test that tells a clinician what kind of Parkinson’s a patient has based on the gut microbiome.

The distinction is important. A microbiome pattern can be useful for research without being ready for diagnosis. Researchers can use these patterns to explore disease mechanisms or identify subgroups. That is very different from telling an individual person that a stool test can predict Parkinson’s or prescribe a supplement plan. The current evidence does not support that clinical leap (Lubomski et al., 2022; Menozzi et al., 2026).

What is actually exciting here

What is exciting is not one single discovery. It is the convergence.

Constipation appears early. Parkinson’s-related α-synuclein can be found in gut tissue. The vagus nerve offers a plausible route between gut and brain. Microbiome studies repeatedly find altered microbial composition and function. Microbial metabolites and immune pathways are biologically relevant. Prodromal studies suggest some of these changes may appear before diagnosis.

None of these proves the gut causes Parkinson’s. Together, they make the gut a serious biological part of the Parkinson’s story (Stokholm et al., 2016; Liu et al., 2017; Aho et al., 2021; Wallen et al., 2022; Huang et al., 2023; Borghammer et al., 2024; Menozzi et al., 2026; de Castro Fonseca et al., 2026).

This is also where some of the newest work becomes interesting. The 2026 Nature Medicine study by Menozzi et al. did not simply compare people with Parkinson’s against controls. It looked at healthy people, genetically at-risk GBA1 carriers and diagnosed Parkinson’s patients, and found that the microbiome signature appeared to sit along a spectrum: least evident in controls, intermediate in some at-risk individuals and more pronounced in diagnosed Parkinson’s (Menozzi et al., 2026). That does not make the microbiome a diagnostic test, but it gives researchers a more detailed way of thinking about when gut changes may appear.

The 2026 Lancet Neurology Personal View by de Castro Fonseca et al. also matters because it connects the microbiome to systemic inflammation and autoimmunity rather than treating gut bacteria as an isolated topic. The gut, immune system and α-synuclein are increasingly being studied together, rather than as separate Parkinson’s subplots (de Castro Fonseca et al., 2026).

Together, these studies justify treating gut function and microbiome biology as part of Parkinson’s research, while still stopping short of proving that microbiome changes cause Parkinson’s or modify its progression (Romano et al., 2021; Wallen et al., 2022; Bruggeman et al., 2024; Scheperjans et al., 2024; de Castro Fonseca et al., 2026).

Where the science gets overstretched

The first limitation is timing.

Most Parkinson’s microbiome studies are cross-sectional, meaning they compare people with Parkinson’s and controls at one point in time (Romano et al., 2021; Wallen et al., 2022; Boktor et al., 2023; Nishiwaki et al., 2024). Those studies can show that the Parkinson’s gut microbiome is different, and increasingly they can describe the functional direction of that difference. They cannot, by themselves, show whether the microbial shift came before the disease process, followed it, or was partly shaped by features that travel with Parkinson’s, including constipation, slowed gut transit, medication exposure, diet, geography and disease stage (Wallen et al., 2022; Boktor et al., 2023; Kwon et al., 2024; Yoon et al., 2024).

The newer prodromal studies make the timing question more interesting, but they do not close it. Huang et al. (2023) found Parkinson’s-like microbiome changes in people with isolated rapid eye movement sleep behaviour disorder and their first-degree relatives, and Menozzi et al. (2026) found an intermediate Parkinson’s-like microbiome signature in unaffected GBA1 carriers and a subset of healthy controls. Those findings argue against the microbiome being only a late consequence of diagnosed Parkinson’s or long-term levodopa exposure. But because these studies do not yet show, over time, who later develops Parkinson’s and who does not, they cannot turn the microbiome into a validated prediction tool or prove that dysbiosis is driving the disease process (Huang et al., 2023; Menozzi et al., 2026).

The second limitation is intervention.

The two randomised faecal microbiota transplantation trials published so far do not give a consistent answer. GUT-PARFECT reported a modest motor benefit at 12 months, whereas the Finnish JAMA Neurology trial found microbiome change without clinically meaningful improvement over placebo on the outcomes the researchers had chosen before the trial began (Bruggeman et al., 2024; Scheperjans et al., 2024).

That is why claims that microbiome manipulation can alter Parkinson’s progression are ahead of the evidence. At present, the stronger clinical case is narrower: gut-directed approaches may be relevant for constipation, bowel function, medication response and quality of life, but disease modification remains unproven.

The third limitation is clinical translation.

The evidence does not currently support direct-to-consumer claims that stool microbiome tests can diagnose Parkinson’s, predict who will develop it, or generate validated supplement protocols. Current diagnostic microbiome models remain exploratory and require longitudinal validation before clinical use (Lubomski et al., 2022; Menozzi et al., 2026).

The evidence supports a more limited conclusion: gut and microbiome biology are relevant to Parkinson’s mechanisms and symptoms, but they do not justify blame-based explanations, diagnostic stool-test claims or promises of disease reversal (Lubomski et al., 2022; Bruggeman et al., 2024; Scheperjans et al., 2024; de Castro Fonseca et al., 2026).

So where does this leave us?

Parkinson’s disease is connected to the gut in several real, measurable ways: through gut symptoms that often appear first, through anatomical pathways such as the vagus nerve, through the presence of α-synuclein in gut tissue before diagnosis, through a reproducible pattern of gut microbial composition and function, through altered microbial metabolites, and through immune signalling that links the gut to the brain (Stokholm et al., 2016; Aho et al., 2021; Wallen et al., 2022; Borghammer et al., 2024; Menozzi et al., 2026).

None of this means Parkinson’s “starts in the gut” for everyone who has it. The brain-first / body-first framework remains the best current explanation for why some people fit the gut-first story more neatly than others (Horsager et al., 2020; Borghammer et al., 2024).

The fair conclusion is not “Parkinson’s starts in the gut” and not “the microbiome is irrelevant.” The fair conclusion sits between those extremes: the gut appears to be part of Parkinson’s biology in a real and measurable way, especially in relation to constipation, microbial metabolism, immune signalling and possibly body-first disease patterns. What remains unproven is whether changing the microbiome can change the long-term disease course.

The next question is more practical, and more difficult: what should anyone do with this information?

If the gut microbiome is altered in Parkinson’s, should people change what they eat? If constipation is part of the disease biology, how actively should it be managed? If microbial metabolites, fibre, bile acids and inflammation are part of the story, does that make probiotics useful, or are they being oversold? And where does levodopa timing fit into all of this?

These questions show up in daily care: bowel frequency, constipation management, meal timing, medication response, weight loss and swallowing difficulty are recognised clinical concerns in Parkinson’s care (Pfeiffer et al., 2024).

That is where the second article begins.

Coming next ...

Feeding the Parkinson’s gut: what the microbiome science does, and does not, support

Part two will look at food, fibre, plant diversity, probiotics, fermented foods, bowel routines, hydration, levodopa timing and lifestyle support, with the focus kept firmly on Parkinson’s disease, gut function and the microbiome.

It will also look at one of the more interesting recent intervention studies: a 2026 randomised trial testing resistant starch, a fermentable fibre, in people with Parkinson’s. That study reported microbiome changes, increases in Faecalibacterium and short-chain fatty acids, reductions in opportunistic pathogens and symptom improvements, but it still needs to be interpreted carefully before being turned into general advice (Petrov et al., 2026).

Support for Parkinson’s Disease at You Nutrition Clinic

At You Nutrition Clinic, we specialise in nutrition and lifestyle support for neurological diseases and disorders, including Parkinson’s disease and related neurodegenerative conditions.

Our Parkinson’s lead practitioner, Melody, is a highly experienced Registered Nutritional Therapy Practitioner who supports people living with Parkinson’s disease, as well as families and carers navigating the condition alongside them.

Melody works with clients to optimise nutrient intake around medication use, support neurological function, and address the symptoms that often affect daily quality of life. Her work includes support for both motor and non-motor features of Parkinson’s disease, including digestive symptoms, constipation, fatigue, sleep disturbance, changes in energy levels, appetite changes and the wider relationship between diet, digestion and brain chemistry.

🧩 Connect with us

For research updates, practical tips, and ongoing inspiration, follow us on Instagram:

👉 @nutritionandthebrain

👉 @younutritionkids

Stay curious. Stay hopeful. Support your brain. 🧠

Disclaimer: This article is for informational and educational purposes only and does not constitute medical advice. Always consult with a qualified, registered medical doctor (MD) for diagnosis and treatment decisions.

References

Aho, V. T. E., Houser, M. C., Pereira, P. A. B., Chang, J., Rudi, K., Paulin, L., Hertzberg, V., Auvinen, P., Tansey, M. G., & Scheperjans, F. (2021). Relationships of gut microbiota, short-chain fatty acids, inflammation, and the gut barrier in Parkinson’s disease. Molecular

Neurodegeneration, 16, 6. https://doi.org/10.1186/s13024-021-00427-6

Boktor, J. C., Sharon, G., Verhagen Metman, L. A., Hall, D. A., Engen, P. A., Zreloff, Z., Hakim, D. J., Bostick, J. W., Ousey, J., Lange, D., Humphrey, G., Knight, R., Keshavarzian, A., &

Mazmanian, S. K. (2023). Integrated multi-cohort analysis of the Parkinson’s disease gut metagenome. Movement Disorders, 38(3), 399–409. https://doi.org/10.1002/mds.29300

Borghammer, P., Just, M. K., Horsager, J., Skjærbæk, C., Raunio, A., Kok, E. H., Savola, S., Murayama, S., Saito, Y., Myllykangas, L., & Van Den Berge, N. (2024). Brain-first vs. body-first Parkinson’s disease: An update on recent evidence. Parkinsonism & Related Disorders, 122, 106101. https://doi.org/10.1016/j.parkreldis.2024.106101

Bruggeman, A., Vandendriessche, C., Crozier, T. W. M., De Pue, A., Vanwynsberghe, J., Spincemaille, S., De Decker, M., et al. (2024). Safety and efficacy of faecal microbiota transplantation in patients with mild to moderate Parkinson’s disease (GUT-PARFECT): A double-blind, placebo-controlled, randomised, phase 2 trial. eClinicalMedicine, 71, 102563. https://doi.org/10.1016/j.eclinm.2024.102563

Chen, S.-J., Chen, C.-C., Liao, H.-Y., Lin, Y.-T., Wu, Y.-W., Liou, J.-M., Wu, M.-S., Kuo, C.-H., & Lin, C.-H. (2022). Association of fecal and plasma levels of short-chain fatty acids with gut microbiota and clinical severity in patients with Parkinson disease. Neurology, 98(8), e848–e858. https://doi.org/10.1212/WNL.0000000000013225

de Castro Fonseca, M., Zanetti, L., Fanourakis, S., Sulzer, D. L., & Mazmanian, S. K. (2026). The gut microbiome, systemic inflammation, and autoimmunity in Parkinson’s disease. The Lancet Neurology, 25(1), 103–114. https://doi.org/10.1016/S1474-4422(25)00382-5

Horsager, J., Andersen, K. B., Knudsen, K., Skjærbæk, C., Fedorova, T. D., Okkels, N., Schaeffer, E., et al. (2020). Brain-first versus body-first Parkinson’s disease: A multimodal imaging case-control study. Brain, 143(10), 3077–3088. https://doi.org/10.1093/brain/awaa238

Huang, B., Chau, S. W. H., Liu, Y., Chan, J. W. Y., Wang, J., Ma, S. L., Lam, S. P., et al. (2023). Gut microbiome dysbiosis across early Parkinson’s disease, REM sleep behavior disorder and their first-degree relatives. Nature Communications, 14, 2501. https://doi.org/10.1038/s41467-023-38248-4

Kwon, D., Zhang, K., Paul, K. C., Folle, A. D., Del Rosario, I., Jacobs, J. P., Keener, A. M., Bronstein, J. M., & Ritz, B. (2024). Diet and the gut microbiome in patients with Parkinson’s disease. npj Parkinson’s Disease, 10, 89. https://doi.org/10.1038/s41531-024-00681-7

Liu, B., Fang, F., Pedersen, N. L., Tillander, A., Ludvigsson, J. F., Ekbom, A., Svenningsson, P., Chen, H., & Wirdefeldt, K. (2017). Vagotomy and Parkinson disease: A Swedish register-based matched-cohort study. Neurology, 88(21), 1996–2002. https://doi.org/10.1212/WNL.0000000000003961

Lubomski, M., Xu, X., Holmes, A. J., Muller, S., Yang, J. Y. H., Davis, R. L., & Sue, C. M. (2022). Nutritional intake and gut microbiome composition predict Parkinson’s disease. Frontiers in Aging Neuroscience, 14, 881872. https://doi.org/10.3389/fnagi.2022.881872

Menozzi, E., Ren, Y., Geiger, M., Schapira, A. H. V., Ehrlich, S. D., Meslier, V., et al. (2026). Microbiome signature of Parkinson’s disease in healthy and genetically at-risk individuals. Nature Medicine. Advance online publication. https://doi.org/10.1038/s41591-026-04318-5

Nishiwaki, H., Ito, M., Hamaguchi, T., Maeda, T., Kashihara, K., Tsuboi, Y., Ueyama, J., Yoshida, T., Hanada, H., Takeuchi, I., Katsuno, M., & Hirayama, M. (2024). Meta-analysis of shotgun sequencing of gut microbiota in Parkinson’s disease. npj Parkinson’s Disease, 10, 106. https://doi.org/10.1038/s41531-024-00724-z

Petrov, V. A., Schade, S., Laczny, C. C., Hällqvist, J., May, P., Jäger, C., Aho, V. T. E., Hickl, O., Halder, R., Lang, E., Caussin, J., Lebrun, L. A., Schulz, J., Unger, M. M., Mills, K., Mollenhauer, B., & Wilmes, P. (2026). Resistant starch improves Parkinson’s disease symptoms through restructuring of the gut microbiome and modulating inflammation. Brain, Behavior, and Immunity, 132, 106217. https://doi.org/10.1016/j.bbi.2025.106217

Pfeiffer, R. F., Isaacson, S. H., & Pahwa, R. (2024). Consensus practice recommendations for management of gastrointestinal dysfunction in Parkinson disease. Parkinsonism & Related

Disorders, 124, 106982. https://doi.org/10.1016/j.parkreldis.2024.106982

Romano, S., Savva, G. M., Bedarf, J. R., Charles, I. G., Hildebrand, F., & Narbad, A. (2021). Meta-analysis of the Parkinson’s disease gut microbiome suggests alterations linked to intestinal inflammation. npj Parkinson’s Disease, 7, 27. https://doi.org/10.1038/s41531-021-00156-z

Sampson, T. R., Debelius, J. W., Thron, T., Janssen, S., Shastri, G. G., Ilhan, Z. E., Challis, C., Schretter, C. E., Rocha, S., Gradinaru, V., Chesselet, M.-F., Keshavarzian, A., Shannon, K. M., Krajmalnik-Brown, R., Wittung-Stafshede, P., Knight, R., & Mazmanian, S. K. (2016). Gut microbiota regulate motor deficits and neuroinflammation in a model of Parkinson’s disease. Cell, 167(6), 1469–1480.e12. https://doi.org/10.1016/j.cell.2016.11.018

Sampson, T. R., Challis, C., Jain, N., Moiseyenko, A., Ladinsky, M. S., Shastri, G. G., Thron, T., Needham, B. D., Horvath, I., Debelius, J. W., Janssen, S., Knight, R., Wittung-Stafshede, P., Gradinaru, V., Chapman, M., & Mazmanian, S. K. (2020). A gut bacterial amyloid promotes α-synuclein aggregation and motor impairment in mice. eLife, 9, e53111. https://doi.org/10.7554/eLife.53111

Scheperjans, F., Levo, R., Bosch, B., Lääperi, M., Pereira, P. A. B., Smolander, O.-P., Aho, V. T. E., Vetkas, N., Toivio, L., Kainulainen, V., Fedorova, T. D., Lahtinen, P., Ortiz, R., Kaasinen, V., Satokari, R., & Arkkila, P. (2024). Fecal microbiota transplantation for treatment of Parkinson disease: A randomized clinical trial. JAMA Neurology, 81(9), 925–938. https://doi.org/10.1001/jamaneurol.2024.2305

Schwiertz, A., Spiegel, J., Dillmann, U., Grundmann, D., Bürmann, J., Faßbender, K., Schäfer, K. H., & Unger, M. M. (2018). Fecal markers of intestinal inflammation and intestinal permeability are elevated in Parkinson’s disease. Parkinsonism & Related Disorders, 50, 104–107. https://doi.org/10.1016/j.parkreldis.2018.02.022

Shen, X., Schaeffer, E., Berg, D., Heinzel, S., Dommershausen, N., Aho, V. T. E., Murros, K., Auvinen, P., Paulin, L., Hertzberg, V., Pereira, P. A. B., & Scheperjans, F. (2024). Gut microbial metabolites and future risk of Parkinson’s disease: A metabolome-wide association study. Movement Disorders, 39(12), 2143–2153. https://doi.org/10.1002/mds.30037

Shin, C., Kim, S.-I., Kim, H. J., Han, J.-Y., Park, S.-H., Yang, H.-K., Lee, H.-J., Kong, S.-H., Suh, Y.-S., Paek, S. H., Lee, J.-Y., Kim, J.-S., & Jeon, B. (2024). Diagnostic accuracy and predictors of alpha-synuclein accumulation in the gastrointestinal tract of Parkinson’s disease. npj

Parkinson’s Disease, 10, 154. https://doi.org/10.1038/s41531-024-00766-3

Shin, C., Im, J. P., Han, J.-Y., et al. (2025). Cognitive function correlates with gastric alpha-synuclein seeding activity in early Parkinson’s disease. npj Parkinson’s Disease, 11, 311. https://doi.org/10.1038/s41531-025-01152-3

Stokholm, M. G., Danielsen, E. H., Hamilton-Dutoit, S. J., & Borghammer, P. (2016). Pathological α-synuclein in gastrointestinal tissues from prodromal Parkinson disease patients. Annals of Neurology, 79(6), 940–949. https://doi.org/10.1002/ana.24648

Wallen, Z. D., Demirkan, A., Twa, G., Cohen, G., Dean, M. N., Standaert, D. G., Sampson, T. R., & Payami, H. (2022). Metagenomics of Parkinson’s disease implicates the gut microbiome in multiple disease mechanisms. Nature Communications, 13, 6958. https://doi.org/10.1038/s41467-022-34667-x

Yan, Z., Yang, F., Sun, L., Yu, J., Sun, L., Si, Y., Yao, L., & Sun, T. (2022). Multiomics implicate gut microbiota in altered lipid and energy metabolism in Parkinson’s disease. npj Parkinson’s Disease, 8, 43. https://doi.org/10.1038/s41531-022-00300-3

Yoon, H. J., Kang, W., Jo, S., Hwang, Y. S., Lee, J. H., Chung, S. J., & Park, Y. K. (2024). Dietary quality and the gut microbiome in early-stage Parkinson’s disease patients. Nutritional Neuroscience, 27(7), 754–767. https://doi.org/10.1080/1028415X.2023.2253025