How a once-controversial idea is advancing understanding of Parkinson’s disease

Myriam Penninckx was having lunch at a restaurant in Brussels when her right arm began to shake. This was just the latest in a string of signs that something might be amiss. Her handwriting had suddenly become impossible to decipher, and she had noticed herself slowing down, physically and mentally. “I had always been Speedy Gonzales, doing a lot of things at the same time,” she says. “But that was becoming less possible.”

Penninckx was diagnosed with Parkinson’s disease shortly after her 60th birthday. She retired from her job and became depressed. “It was really hard to be someone who had a beautiful career, and then suddenly become a Parkinson’s patient,” she says. Her social life also ground to a halt. “If you have a tremor and you cannot cut your fish or eat your spaghetti, you don’t go out any more,” she says. “You close yourself to the world.”

When Penninckx heard about an unusual clinical trial being run at Ghent University in Belgium, she decided to enrol. In July 2021, she underwent a procedure that entailed flushing out her gut, and giving her a faecal transplant from a healthy donor to replace the community of microorganisms living there. The results, she says, “were amazing”. Her arm tremor went away and has not returned. She feels like the trial gave her “another shot at life”.

Nature Outlook: Parkinson’s disease

It might, at first, sound counterintuitive that such microbiome-targeting treatment could have marked benefits for people with Parkinson’s. Physicians have long been taught that the disease is a brain disorder that affects mainly the dopamine system, and that the hallmarks of the disease are Lewy bodies — misfolded clumps of a protein called α-synuclein that build up in neurons. But scientists have realized that the rest of the body is also involved, and the gut, in particular. For many individuals with Parkinson’s, evidence points to the gut’s involvement in the disease many years before diagnosis, says Per Borghammer, a neuroscientist at Aarhus University in Denmark.

Changes in the microbiome are also linked to some early symptoms and faster progression of the disease. These observations suggest that there are different subtypes of Parkinson’s, adds Filip Scheperjans, a neurologist at Helsinki University Hospital.

The gut’s seemingly central role in many Parkinson’s cases has implications not only for a better understanding of the disease’s origins and biology but also for tailoring treatments to an individual’s profile. “I make the comparison with what we see in oncology and cancer research,” says Arnout Bruggeman, the neurologist at Ghent University Hospital who oversaw Penninckx’s faecal transplant. Recognizing that various cancers exist as subtypes has allowed scientists to design vastly better treatments, he says — something he hopes will also be the case for Parkinson’s.

Brain or body first?

Hints that the gut is involved in Parkinson’s have been around since the disease was first described. People commonly have gastrointestinal issues, which sometimes start years before diagnosis. In 2003, researchers suggested that the α-synuclein pathology seen in the central nervous system might originate at peripheral sites, including the gut¹. The work “introduced the idea that Parkinson’s disease starts very localized and then spreads via the neural connections”, Borghammer says. “It’s a little analogous to cancer that starts in one place and then metastasizes.”

Dozens of papers subsequently revealed that the microbiomes of people with Parkinson’s differ from those of people without the disease. In 2016, Sarkis Mazmanian, a microbiologist at the California Institute of Technology in Pasadena, showed that the microbiome plays a crucial part in the development of Parkinson’s-like symptoms in mice engineered to overexpress α-synuclein². The animals’ symptoms went away when the researchers obliterated their microbiomes, and returned when they restored them. The team also transplanted human microbiomes into the mice, and found that rodents that received transplants from people with Parkinson’s had worse symptoms than did those who received transplants from healthy people.

In 2019, Borghammer and his colleague proposed that Parkinson’s can be divided into two main subtypes, according to where α-synuclein originates³. The ‘brain first’ subtype of the disease, he says, starts in the olfactory bulb, whereas a ‘body first’ version starts in the gut. This idea was controversial at first, and some researchers are still sceptical. But after years of mounting evidence, most are coming around to the idea.

Body-first Parkinson’s is thought to progress faster, and often manifests first with disturbances to the REM stage of sleep, gastrointestinal issues, depression, loss of smell and damage to the nerves that supply the heart (see ‘Teasing out distinct origins’). Brain-first Parkinson’s, by contrast, usually begins with motor symptoms. Eventually the subtypes converge — the end stage of the disease is the same, Borghammer says, it just starts in different places.

Identifying the culprits

For both brain-first and body-first subtypes, the majority of Parkinson’s cases are attributable to “something you came across in your life”, Mazmanian says. Environmental risk factors seem to include air pollution, heavy metals, pesticides, certain viral infections and possibly microplastics.

Genes also play a part, however, and there is no bright-line distinction between genetic and environmental causes; the two are often linked. Variants of a gene called GBA1 are most closely associated with body-first Parkinson’s. But only a minority of carriers of the gene actually develop the disease, leading researchers to wonder what external factors might also have a role.

A study in April explored the part that the microbiome might play in increasing that risk, by examining hundreds of participants in the United Kingdom and Italy who did and did not have Parkinson’s, and likewise did and did not have a GBA1 variant linked to the disease⁴. The researchers analysed the participants’ stool samples and confirmed that the microbiomes of people with Parkinson’s differed significantly from those of people without the disease, and that the more severe the participant’s disease, the more altered was their microbiome. Mirroring earlier findings in mice, they also confirmed that, regardless of whether they carried a variant, people with Parkinson’s had roughly the same altered microbiome. This suggests that the disease process does more to transform a gut microbiome than do a person’s genes, says Elisa Menozzi, a clinical neuroscientist now at Aarhus University, and a co-author of the study.

As they had expected, Menozzi and her colleagues also found that healthy people carrying a GBA1 variant had, on average, microbiomes shifted toward the Parkinson’s pattern. But given that both carriers and non-carriers who already had Parkinson’s had statistically indistinguishable microbiomes, it seems that the variant itself is not what drives this change. Furthermore, in healthy participants not every carrier had this microbiome shift, and carriers who showed early, subtle signs of disease had microbiomes that looked most like those of people with Parkinson’s. Roughly one in five healthy participants without the variant also had the same altered microbiomes and signs that could be early markers of disease. The researchers’ interpretation, Menozzi says, is that changes to the microbiome raise the risk of Parkinson’s independently of genetics.

Taken together, Menozzi says, the findings suggest that the microbiome could be a diagnostic marker of Parkinson’s before symptoms become noticeable. Confirming this will require following the participants over time to see whether any of them develop Parkinson’s — something the team is doing, says co-author Anthony Schapira, a neurologist at University College London.

Although the study did not address what is causing the microbiome to change, the researchers found a significant association with participants’ eating habits. Diets that were high in vegetables, fruits and polyunsaturated fats seemed to protect against a Parkinson’s-type microbiome, whereas those high in saturated fat and low in fibre were associated with a microbiome closer to that of someone with the disease-altered microbiome. So dietary intervention is worth investigating, Schapira says. “The first question to answer is, can you do anything about this — can you change the microbiome?” he says. “ The next question is, does it matter?”

Need for precision

Some evidence from human trials already indicates that harnessing the microbiome to treat Parkinson’s could be possible. When Penninckx told her family that she would be receiving a faecal transplant as part of a study, she encountered some scepticism. “My grandchildren said, ‘Are you going to eat it?’” she recalls. She was relieved that was not the case. Instead, the researchers inserted a plastic tube into her nose, snaking it through her stomach and into her small intestine. “You’re never in contact with the material, so it’s not like there is a bad smell or taste,” she says.

Penninckx was one of 22 participants in the 2024 study to receive a faecal transplant from a healthy donor, while another 24 people were given a placebo made up of their own gut flora⁵. Twelve months after treatment, those who received the active treatment had fewer motor symptoms and less constipation than did those in the control group. The researchers plan to launch a larger trial this autumn to replicate these findings, Bruggeman says. Future trials will also analyse the composition of the microbiomes of healthy donors, which he says is “a key question” for determining the source of a potential positive effect.

Not all the results have been this promising. Scheperjans and his colleagues did not see a significant improvement in 30 participants who received an active treatment in a double-blind, placebo-controlled faecal transplant trial⁶. They do not know why their results differed from those in Belgium, but they suspect it might have had something to do with the method of delivery. The team administered treatment into the colon of participants rather than into the small intestine, which contains more immune cells than the large intestine. “It may be that for immune-system regulation, the small bowel is more important,” he says.

Surprisingly, the placebo group in this trial experienced more benefits than did the active treatment group. Up to 12 months after treatment, those in the placebo group had less constipation and more relief of some motor symptoms, as well as more changes to their microbiomes. The researchers think that this might be related to the bowel cleansing that all participants underwent before treatment — a hypothesis that they plan to test in a follow-up study. “If you have an abnormal microbiome, maybe the bowel cleansing gives you an opportunity to get recolonized with more normal bacteria,” Scheperjans says. “We’ll know if that’s the case after the next trial.”

Borghammer suspects that the differing faecal transplant results might come down to how “incredibly crude” the trials so far have been, simply because the field is so new. For example, researchers do not yet have the knowledge that would allow them to select the best donors for participants. Given this, he does not think that inconclusive results should be discouraging. “It’s plausible that it can work,” he says. “We just don’t understand it well enough yet.”

Another potential problem for faecal-transplant trials — and for research on Parkinson’s interventions in general — is that studies usually lump together people with body-first and brain-first subtypes. This could be confounding the results, Borghammer says, because different forms of the disease are likely to be driven by different mechanistic pathways. In 2022, for example, a phase III trial of an immunotherapy treatment targeting α-synuclein failed to produce significant results. But when the researchers later analysed the data for participants only with the most severe Parkinson’s, they found that the treatment did seem to slow progression of motor symptoms⁷. “It looks like for the fast-progressing type, which is strongly associated with the body-first type, it might have actually worked,” Borghammer says. “But because the original study included a mixed group of unselected patients, the clinical trial might have been unable to show statistically significant results.”

Researchers are increasingly aware of these limitations, and some groups are looking more closely at the bacterial species and mechanisms involved — and at more targeted populations. Haydeh Payami, a geneticist at the University of Alabama at Birmingham, and her team found that the gut microbiomes of about one-quarter of people with Parkinson’s contain high levels of a protein that induces aggregation of α-synuclein⁸. That protein is produced only by the bacterium Escherichia coli, giving scientists “one defined target”, says Mazmanian, who collaborates with Payami. Vertero Therapeutics, a firm Mazmanian co-founded in Woburn, Massachusetts, have developed a drug that inhibits production of the E. coli protein, and in mice it prevented α-synuclein from aggregating. “You can just stop this process in its tracks,” Mazmanian says.

In May, Vertero began a phase I trial in people with Parkinson’s whose microbiome matches the bacterial profile targeted by the drug. “In the population that we can treat, our hope is to slow down or even stop Parkinson’s progression,” Mazmanian says. “That would be a huge victory.”

The ultimate goal, however, is to be able to identify people in the very early stages of the disease, and then use pharmaceuticals, faecal transplants, dietary changes or other interventions to alter their trajectory. “Fast forward a decade, I hope we’ll be able to treat people before they actually develop Parkinson’s,” Mazmanian says.

Five years on, Penninckx still feels empowered by the gains she experienced in the faecal-transplant trial. She keeps busy as a board member of Parkili, a Belgian charity that supports scientific research — including some of Bruggeman’s work — and arranges trips around the world for people with Parkinson’s. She has travelled to Jordan with the group, and in May, she and seven others represented it at the World Parkinson’s Conference in Phoenix, Arizona. “I’ve started a new life,” she says. “So my story is a happy one.”

Penninckx knows that her disease is not cured. But her experience hints at the potential benefits that the mounting appreciation of the Parkinson’s gut–brain connection could bring.