Machine learning reveals link between metabolites and Alzheimer’s
A machine learning method recently shed light on how gut microbial metabolites interact with cells’ receptors and contribute to Alzheimer's disease.
Source: Getty Images
By - A team of researchers from Cleveland Clinic found that machine learning can help characterize how gut microbial metabolites and cell receptors interact to influence Alzheimer's disease, according to a recent study in Cell Reports.
The research team indicated that previous studies have demonstrated that Alzheimer's patients often experience gut bacteria changes as the condition progresses. However, many of the mechanisms that drive this “gut-brain axis” are unknown.
Gut metabolites are released by bacteria as they break down the food an individual has eaten, and these metabolites then go on to influence many cellular processes throughout the body. The researchers noted that these interactions can be helpful to human health in many cases, but links between metabolites and a host of conditions — like cancer and Parkinson’s disease, alongside Alzheimer's — have been documented.
The significant impact of the gut microbiome in these cases has led researchers to explore how drugs or other therapeutics could prevent harmful metabolite interactions.
"Gut metabolites are the key to many physiological processes in our bodies, and for every key there is a lock for human health and disease," explained Feixiong Cheng, PhD, inaugural director of the Cleveland Clinic Genome Center, in a news release. "The problem is that we have tens of thousands of receptors and thousands of metabolites in our system, so manually figuring out which key goes into which lock has been slow and costly."
To combat this, the research team used machine learning to study how metabolites and cell receptors interact in the context of Alzheimer's disease. By integrating information on metabolite shapes and receptor protein structures, genetic and proteomic data and metabolites’ known impact on patient-derived brain cells, the approach enabled researchers to analyze over 1.09 million potential metabolite-receptor pairs.
This data was used to rank metabolites and receptors based on the likelihood that they would interact with each other, which informed predictions around the likelihood that the pair would contribute to Alzheimer's.
From there, the metabolite-receptor pairs with the highest likelihood of influencing Alzheimer's were examined more closely.
One of the relevant metabolites, agmatine, is thought to protect brain cells from inflammation-associated damage. The analysis revealed that agmatine was most likely to interact with a receptor known as CA3R.
Further investigation indicated that agmatine and CA3R do influence one another.
Alzheimer's-affected neurons treated with agmatine reduced CA3R levels, while agmatine-treated neurons also demonstrated significantly lower levels of phosphorylated tau proteins – known markers of Alzheimer's disease.
The researchers underscored that this work highlights the potential for AI applications in other studies assessing the link between the gut microbiome and disease.
"We specifically focused on Alzheimer's disease, but metabolite-receptor interactions play a role in almost every disease that involves gut microbes," Cheng stated. "We hope that our methods can provide a framework to progress the entire field of metabolite-associated diseases and human health."
