Christina Szalinski
More than half the cells in the human body are not human — they’re microbes. Each microbe makes unique molecules, or metabolites, and studying them is crucial for understanding the microbiome’s role in health and disease.
But with so many microbes and metabolites floating around — literally trillions — how can you tell which microbe made which molecule? How do you know whether a molecule came from a microbe or the host?
Thanks to researchers from the University of California San Diego, answering those questions just got a lot easier. As reported in Nature Microbiology this month, they’ve developed a public search tool that matches microbes to the metabolites they produce within seconds.
No other tool like this exists, the researchers say, and it could be pivotal in advancing medical science and microbiome-based therapies.
“For the past few centuries, we’ve focused on the host-derived molecules,” said study author Pieter Dorrestein, PhD, pharmacology professor at UCSD. “So, we have a pretty good inventory of the host-derived or human-derived molecules, but a pretty poor inventory of the molecules that microbes produce.”
Developed from 100 million data points crowdsourced from scientists worldwide, the tool — called microbeMASST— searches a database of 60,000 microbe metabolites. Scientists can take a complex sample — plasma, tissue, feces — and based on data features of the molecules present in the sample, determine what microbes made those molecules.
“This is powerful,” said Eric B. Taylor, PhD, associate professor of molecular physiology and biophysics at the University of Iowa Carver College of Medicine, Iowa. (Taylor wasn’t involved in the study.) “This can advance medical science by informing new microbiomic mechanisms of health and disease that may be therapeutically modulated.” That could have implications for many diseases, “from nonalcoholic fatty liver disease to Alzheimer’s to diabetes to IBD.”
How the Tool Works
The molecules that microbes make can be involved in communication (signaling), metabolizing nutrients or drugs, or modulating inflammation, among other tasks. They can influence the local site where they reside, or they can travel — for instance, studies have shown that gut microbial molecules can affect the brain.
The study of these metabolites is known as metabolomics.
“The type of metabolomics we use is essentially a fancy scale,” Dorrestein said, “where we weigh all the molecules that are present in the sample. From that, we can begin to start inferring and understanding the structure.”
That “fancy scale” is tandem mass spectrometry, an analytical technique used to separate and measure the weight of molecules.
To use the search tool, researchers can input this mass spectrometry data, which tells the computer the signatures, or markers, of all the molecules in a sample. “You can almost think of them as barcodes that connect to specific molecules,” Dorrestein said. The search tool will then identify which microbes can produce the molecules present in the data.
To test the tool, the team searched for molecules using samples from human organs. After identifying the microbes they came from, the researchers used public data to trace the microbes’ origins. “We had a few hundred molecules that were found in the brain, for example,” Dorrestein said. The findings show an “exchange of metabolites from the gut to the brain.”
What’s Next
The project is part of a National Institutes of Health initiative to build an international repository of microbial metabolites and their functions, called the Collaborative Microbial Metabolite Center.
The data includes microbes from not just humans and animals but also plants, soils, oceans, and lakes. So, even though the focus of this research is on human health, it could help us understand other ecosystems as well.
By cross-referencing results with genomic data, the tool could be used to help identify which microbial genes produce specific metabolites as well, Taylor added.
Next, the researchers plan to link microbeMASST with data on the effect of medications, disease conditions, diet interventions, and age on the microbiome, Dorrestein said.
Leave a Reply