For millennia, trillions of beneficial microbes in the human digestive tract have toiled in obscurity, munching on certain molecules and secreting others. In the last couple of decades, however, these tiny organisms, known collectively as the gut microbiome, have come into the spotlight. Scientists have been finding more and more links between the gut microbiome and a wide range of diseases and conditions, from autoimmune diseases to cancer.

The microbiome has even been implicated in mental health issues like depression and anxiety. That might seem surprising, until you consider the fact that microbes in the gut produce many neuroactive compounds, including neural transmitters like serotonin. Incredibly, about 95% of the serotonin in your body is produced by gut microbes.

“The gut is a major source of signals from the body to the brain,” said Kevin Bonham, assistant professor of medicine at Tufts University School of Medicine and a researcher in the division of gastroenterology at Tufts Medical Center. Bonham pointed out that strong communication between the brain and gut makes sense from an evolutionary perspective, since one of the brain’s primary survival duties is making sure we stay fed and well. But the communication goes beyond sensations of hunger, fullness, or nausea. 

To explore this connection further, Bonham is studying how microbiome development correlates to brain development in infants and toddlers, and whether changes in the microbiome support neural development.

Bonham came to Tufts in March as part of a microbiome cluster hire initiative between Tufts University School of Medicine and Tufts Medicine, along with physician-scientist Cammie Lesser, associate professor at the School of Medicine. “One of the goals of cluster hires is to bring the hospital and medical school research communities together,” Bonham said. This was the second cluster hire at Tufts this academic year. 

Baby’s Microbiome and Brain 

As a first step in understanding the relationship the microbiome might play in brain development, Bonham worked on a project while he was a senior research scientist in the Klepac-Ceraj Lab at Wellesley College to measure and describe the infant microbiome and how it transforms in the first months of life, using data from more than 1,800 babies in 12 countries. Each baby’s microbiome composition was determined by analyzing stool samples. 

“When infants shift from liquids to solid foods, we see this dramatic shift in the composition of the microbiome,” Bonham said, with certain species of microbes becoming more abundant and other species declining. Remarkably, the shift in species composition was largely consistent across the globe, despite cultural differences in lifestyle and diet. Based on these patterns, the team found they could predict a baby’s age to within three months by analyzing its microbiome.

Next, as part of a large worldwide study of children’s health called 1kD and funded by Wellcome Leap, Bonham and the research team collected microbiome and neural data from 194 typically developing children in South Africa at three times between the ages of 0 and 14 months. The neural data was collected via electroencephalogram “hairnets” placed on babies’ heads while they were shown a checkerboard pattern. Analyzing the types of brain waves produced can indicate how developed the brain’s visual cortex is. 

Bonham and his team found that the composition of the microbiome at four months of age was correlated to the level of visual cortex development between nine and 15 months. While this observation doesn’t prove that the microbiome composition is necessary for the development of the visual cortex, it raises intriguing possibilities. Bonham noted that some of the molecules produced by the microbe species that were abundant at four months are the same ones needed for the brain development seen several months later.

He plans to continue this line of inquiry at Tufts, and hopes to search for evidence that the changing microbiome composition in fact supports brain development. 

“We think that understanding the way the microbiome is related to typical development will give us insights into instances where there is some abnormality,” he said. In the future, it’s possible that analyzing the microbiome from a baby’s stool sample could be a way to screen for their risk of developmental delays.

Digesting the Data

Asking and answering these kinds of questions involves organizing and analyzing mind-blowing amounts of data. Scientists need to do DNA analysis on thousands of microbial species to know which neuroactive compounds the microbiome is producing. Adding to the complexity, individual microbes of the same species don’t always have the same DNA.

Bonham is up to the task: he has extensive training in statistics, machine learning, and computational modeling. He writes software to organize the data and uses tools from data science to analyze it. 

Bonham, who describes himself as a computational microbiologist with a background in experimental science, is particularly interested in engaging in an iterative process of discovery—looking for correlations in population data, then finding mechanisms in the lab that explain those observations, with the goal of discovering treatments that influence those mechanisms.

“It is challenging, methodologically, to take observations we have in humans and identify places where we can get an experimental handle on a particular mechanism,” he said. “But I think that strategy has a lot of promise for tackling some of these very complex aspects of human health.”

And Bonham knows Tufts is a good place to do that kind of work. “One thing that drew me to Tufts is the link between basic research and clinical research and patient populations,” he said.