Sarkis Mazmanian Finds Gut-Brain Links to Parkinson’s Disease

Back in 2013, when The Scientist first profiled Sarkis Mazmanian, he had just won a MacArthur Foundation Fellowship, often called a “genius grant,” and was working at the cutting edge of microbiome research: the gut-brain connection, autoimmunity, and inflammatory disorders. His work in Dennis Kasper’s lab at Harvard Medical School had explored the links between the gut microbiome and immunity, showing that a specific bacterial strain could rewire immune development in germ-free mice. “This one organism was able to drive the maturation of the immune system,” Mazmanian said in his original interview.

Working at the California Institute of Technology ever since, Mazmanian has continued to make discoveries that have broadened scientists’ understanding of the effects of the microbiome on health, behavior, and immunity. For The Scientist’s 40^th Anniversary, Mazmanian shared some of his major discoveries, explained how his research has changed direction significantly, and cut through some of the hype that surrounds microbiome research. “[We] have now moved into Parkinson's disease, anxiety, feeding behaviors, and Alzheimer's disease, all [through] that same sort of lens of using gut bacteria to try to understand ways of accessing the brain or modulating brain function, so that we can restore health again, broadly speaking,” he said.

Exploring the Gut-Brain Connection in Parkinson’s Disease

Mazmanian’s work in Parkinson’s disease began when he delved into the literature and identified a key gap that he felt needed to be investigated. “Somewhere between 60 and 80 percent of Parkinson's patients report some form of gastrointestinal issue,” he explained. This association goes all the way back to James Parkinson’s 1817 essay in which he described the disease, then known as shaking palsy.¹

It was a mystery to Mazmanian why no one was investigating this clear association. When he spoke to neurologists, they often shrugged their shoulders and asked what he thought they should do about it or seemed to think that gut issues were irrelevant. “But the more Parkinson's patients I talked to, many of them—certainly not all, but many—reported that their gastrointestinal symptoms really eroded their quality of life, particularly early in the course of disease,” Mazmanian continued. “Now we know that gastrointestinal symptoms, primarily in the form of constipation, can precede the diagnosis of Parkinson’s by many years.”

To explore how the gut microbiome might be involved in Parkinson’s disease, Mazmanian turned to a mouse model of the condition. The animals overexpressed α-synuclein, a protein that forms amyloid plaques (or Lewy bodies), which accumulate in the brains of people with Parkinson’s disease and cause the loss of dopamine-producing neurons. This project led to the 2016 publication of what he describes as a simple but very impactful paper in Cell.² “If we remove the microbiome in a mouse model of Parkinson's, all the symptoms go away,” he said. “If we give the microbiome back, all the symptoms come back.”

To delve further, he and his team transplanted microbiomes from both people with Parkinson's disease and healthy donors into the mice. “If the fecal sample—meaning the microbiome—came from a Parkinson's patient, the symptoms were worse, suggesting that there's something in the [Parkinson’s] microbiome that may actually be pathogenic,” he explained. These results suggested that changes in the microbiome of people with Parkinson’s disease might not just be a consequence of the disease, but a contributor to its symptoms and development.

Microbiome Composition, Bacterial Proteins, and an Amyloid Cascade

More recently, Mazmanian’s team used a metagenomics approach to define the composition of Parkinson’s disease gut microbiomes.³ “Broadly speaking, [we found] overgrowth of pro-inflammatory organisms and a depletion of anti-inflammatory organisms… and a depletion in short-chain fatty acid-producing bacteria,” he said. They also noticed something that opened up a promising avenue of research: an increase in specific strains of E. coli that produce an amyloid protein called major curlin subunit (CsgA). “Amyloids are these self-aggregating proteins, and it's been known for some time that different mammalian amyloids can cross-seed each other,” Mazmanian explained.

This ability of amyloids to build on each other is why post-mortem brains of people with Parkinson’s disease show not just α-synuclein aggregation, but also the pathogenic proteins seen in other neurodegenerative diseases such as Huntington’s disease and Alzheimer's disease. Mazmanian wanted to know if bacterial amyloid could induce the aggregation of mammalian amyloid. While he emphasized that other triggers are also known to induce the cascade of α-synuclein in Parkinson’s disease brains, his team’s experiments in mice demonstrated that CsgA can too.

There are still gaps in their knowledge, but Mazmanian and his colleagues have developed a working hypothesis as to how bacteria in the distant gut can affect the brain in this way. Specialized, hormone-producing epithelial cells lining the gut, known as enteroendocrine cells, could be key players. “What we've shown in vitro that's not yet published is that the bacterial amyloid CsgA, or E. coli that carry CsgA, can induce α-synuclein aggregation in enteroendocrine cells,” said Mazmanian. “We're now looking [at that] in mice.”

Based on the work of other researchers studying the gut-brain axis, Mazmanian knew that these enteroendocrine cells can send signals to the brain by forming synapses with enteric neurons. “So, we think the chain is [that if there is] bacterial amyloid in the lumen, some of it gets into the enteroendocrine cells [and] induces α-synuclein aggregation,” Mazmanian said. “[Then] α-synuclein can pass trans-synaptically to neurons in the gut, and then after that, I think that's the signal [that] the train has now left the station. It can now pass from neuron to neuron, up the vagus nerve, into the brainstem, and then up into the brain.”

Harnessing the Microbiome to Combat Disease

While Mazmanian also researches the connection between the gut microbiome and other conditions, including autism and anxiety, he continues to primarily focus on Parkinson’s disease, which is the fastest-growing neurological disorder in the world. His team is currently putting their discoveries to use by creating novel therapies that may help prevent the development and progression of the disease. Their CsgA research, for example, led to the development of a therapeutic candidate that is currently in Phase 1 clinical trials.

Another promising therapeutic avenue is in one of the potent anti-inflammatory bacterial strains they identified as consistently depleted in the microbiomes of people with Parkinson’s disease: Faecalibacterium prausnitzii. In a recent study, Mazmanian and his team demonstrated that treating their Parkinson’s disease mice with this strain alleviated their motor symptoms, induced anti-inflammatory immune responses, and promoted protective gene expression profiles in the colons of the animals.⁴

However, Mazmanian remarked that in many cases, probiotics might not be the most efficient answer. “The microbiome is a source of drugs…but maybe the levels of the organisms that we could ever give back [through probiotics] are not enough,” he said. “If I knew which protein of F. prausnitzii was therapeutic in Parkinson's, I don't have to give [patients] F. prausnitzii. I could [just] give them high amounts of that protein…[Or] if I learned how F. prausnitzii was mediating its anti-inflammatory effect, like what bacterial molecule is interacting with what receptor on a cell type, then I can drug that.”

The chicken-or-egg question that remains, Mazmanian said, is whether or not the gastrointestinal symptoms seen in Parkinson’s disease are actually driving the pathophysiology of the condition. “I know a lot of people believe that it is. I'm open-minded. I would tend to believe it based on research, but I don't think that's conclusive yet. I think we need more research.”

References