The new research, led by Duke University School of Medicine neuroscientists Diego Bohórquez, PhD, and M. Maya Kaelberer, PhD, and published in Nature, centers on neuropods, tiny sensor cells lining the colon’s epithelium. These cells detect a common microbial protein and send rapid messages to the brain that help curb appetite.

But this is just the beginning. The team believes this neurobiotic sense may be a broader platform for understanding how gut detects microbes, influencing everything from eating habits to mood — and even how the brain might shape the microbiome in return. 

“We were curious whether the body could sense microbial patterns in real time and not just as an immune or inflammatory response, but as a neural response that guides behavior in real time,” said Bohórquez, a professor of medicine and neurobiology at Duke University School of Medicine and senior author of the study.  

The key player is flagellin, an ancient protein found in bacterial flagella, a tail-like structure that bacteria use to swim. When we eat, some gut bacteria release flagellin. Neuropods detect it, with help from a receptor called TLR5, and fire off a message through the vagus nerve – a major communication line of communication between the gut and the brain.  

The team, supported by the National Institutes of Health, proposed a bold idea: that bacterial flagellin in the colon could trigger neuropods to send an appetite-suppressing signal to the brain — a direct microbial influence on behavior.

The researchers tested this by fasting mice overnight, then giving them a small dose of flagellin directly to the colon. Those mice ate less.  

When researchers tried the same experiment in mice missing the TLR5 receptor, nothing changed. The mice kept eating and gained weight, a clue that the pathway helps regulate appetite. The findings suggest that flagellin sends a “we’ve had enough” signal through TLR5, allowing the gut to tell the brain it’s time to stop eating. Without that receptor, the message doesn’t get through.

The discovery was guided by lead study authors Winston Liu, MD, PhD, Emily Alway, both graduate students of the Medical Scientist Training Program, and postdoctoral fellow Naama Reicher, Ph.D. Their experiments reveal that disrupting the pathway altered eating habits in mice pointed to a deeper link between gut microbes and behavior.

“Looking ahead, I think this work will be especially helpful for the broader scientific community to explain how our behavior is influenced by microbes,” said Bohórquez. “One clear next step is to investigate how specific diets change the microbial landscape in the gut. That could be a key piece of the puzzle in conditions like obesity or psychiatric disorders.”

Cancer cells interact with this neighborhood -- which scientists term the tumor microenvironment -- in many ways, including obtaining extra resources needed to fuel their unchecked growth. Like a fishing trawler deploying its net, pancreatic ductal adenocarcinoma (PDAC) cells reform their cell surfaces to grab additional nutrients from the jelly-like substance between cells called the extracellular matrix.

This cellular scavenging process -- known as macropinocytosis -- affects the area surrounding the tumor, making the connective tissue stiffer and preventing immune cells from reaching the tumor.

Scientists at the NCI-Designated Cancer Center at Sanford Burnham Prebys published findings July 24, 2025, in Cancer Cell demonstrating that blocking macropinocytosis reshapes the tumor microenvironment to be less fibrous and to allow more access to immune cells. These changes made immunotherapy and chemotherapy more effective in treating PDAC tumors in mice.

The researchers started by observing cells in the tumor microenvironment called fibroblasts that typically form connective tissue and produce many components of the extracellular matrix that are captured during macropinocytosis. In the presence of a tumor, some nearby fibroblasts are coerced to become cancer-associated fibroblasts (CAF) that help tumors grow.

"These CAFs are among the cells surrounding the tumor, and they will support tumor growth by providing metabolites and growth signals, as well as helping in other ways," said Yijuan Zhang, PhD, a staff scientist at Sanford Burnham Prebys and lead author of the study.

The scientists found that blocking macropinocytosis exacerbated the metabolic stress experienced by CAFs that are deprived of glutamine, one of the 20 amino acids used to build proteins throughout the body. Because PDAC relies upon glutamine much more than other cancers, CAFs in the pancreatic cancer tumor microenvironment are routinely starved of glutamine. After preventing pancreatic CAFs from using the same scrounging strategy as PDAC tumors, the scientists observed a change to a different subtype of CAF marked by the expression of genes that promote inflammation.

"Most pancreatic CAFs are myofibroblasts that promote stiffness and density in the tumor microenvironment and make it more difficult for immune cells and drugs to reach the tumor," said Cosimo Commisso, PhD, senior author and interim director and deputy director of the institute's cancer center. "Our experiments led to a subtype reprogramming with fewer myofibroblasts and more inflammatory CAFs, and we wondered how this change would affect the overall tumor microenvironment."

The research team found that significant changes in the tumor neighborhood resulted from preventing macropinocytosis in CAFs.

"There were fewer deposits of collagen that make the tumor microenvironment stiff or fibrotic, more access for CD4+ and CD8+ T cells to infiltrate the tumor, and vascular expansion, which means a widening of blood vessels that can promote drug delivery," said Zhang.

The investigators then wanted to see how these tumor microenvironment modifications might make a difference for patients with PDAC and other cancers that rely on macropinocytosis for fuel. They tested the effects of combining a treatment to block macropinocytosis with immunotherapy and chemotherapy.

"Infiltrating T cells are rich in a cell surface protein called PD-1 that dampens the immune response, so we combined a macropinocytosis inhibitor called EIPA with an anti-PD-1 antibody," said Commisso. "We found it significantly suppressed tumor metastasis and prolonged mouse survival."

"Our findings were similar when using EIPA as a pre-treatment before using the chemotherapy gemcitabine," said Zhang. "In addition to synergistically suppressing tumor growth in mice with PDAC, it also reduced the spread of micrometastases in the lungs."

The scientists will continue to explore how to prevent tumors from scavenging energy to reshape the tumor microenvironment into one that makes cancer treatments more effective.

"We believe this is a very promising strategy to pursue for developing combination therapies for cancer patients," said Commisso. "Especially for pancreatic cancer that is the third leading causes of cancer deaths despite accounting for only three percent of cases."

Additional authors include:

• Li Ling, Rabi Murad, Swetha Maganti, Ambroise Manceau, Hannah A. Hetrick, Madelaine Neff, Cheska Marie Galapate, Shea F. Grenier, Florent Carrette, Karen Duong-Polk, Anindya Bagchi, David A. Scott, Yoav Altman, Jennifer L. Hope and Linda M. Bradley from Sanford Burnham Prebys
• Andrew M. Lowy from the University of California San Diego

The study was supported by the National Institutes of Health and the National Cancer Institute.