Microbial metabolite butyrate activates sensory neurons through colonic serotonin release in a microfluidic co-culture model

Host microbiota interactions are essential regulators of gastrointestinal and systemic physiology. Disruptions in microbial composition can alter epithelial production of serotonin (5 hydroxytryptamine, 5-HT) and impair sensory neuronal signaling, contributing to metabolic, neurological, and cardiovascular disorders. Butyrate, produced by bacterial fermentation of dietary fiber, is a key microbial metabolite that modulates colonic epithelial cell (CEC) function, including 5-HT release. About 95% of 5-HT is produced by CECs which activates vagal and spinal afferent neurons expressing the 5-HT3a receptor, showing epithelial derived 5-HT as a mediator of rapid neuronal responses. Because these epithelial and neuronal pathways constitute a major component of the gut-brain axis, defining how microbial metabolites initiate this signaling is essential for understanding microbe–host neural communication. Mechanisms by which microbial metabolites drive epithelial to neuronal signaling remain insufficiently defined due to limitations of current in vivo and in vitro models. This study aimed to develop a microfluidic tri-culture model that enables directional signaling among CECs, sensory neurons, and microbial metabolites, allowing investigation of the epithelial-neuronal signaling cascade elicited by butyrate. The microfluidic device contains three parallel channels separated by 3 micrometer slits that permit molecular diffusion while maintaining strict physical compartmentalization. F11 neurons were seeded in the top channel, primary CECs were seeded in the middle channel, and Lactobacillus paracasei or sodium butyrate was introduced into the bottom channel to mimic luminal exposure. The model supported long term co-culture, controlled metabolite delivery, and selective receptor inhibition in each compartment. Functional neuronal activation was assessed by calcium imaging, allowing direct measurement of intracellular calcium changes in real time following butyrate stimulation. Epithelial 5-HT secretion was quantified using enzyme linked immunosorbent assay, and epithelial GPR41, GPR43, and neuronal 5-HT3a receptor involvement was evaluated pharmacologically using selective antagonists. Sodium butyrate elicited rapid, dose-dependent neuronal calcium responses in the model, with 1 millimolar (mM) producing significantly greater activation than both 30 micromolar (µM) and baseline (****p < 0.0001). These responses required epithelial involvement, as removing CECs reduced neuronal activation by 66%. Butyrate also increased epithelial 5-HT secretion, and pharmacological inhibition of epithelial GPR41 or GPR43 suppressed 5-HT release and abolished neuronal activation. Blocking the neuronal 5-HT3a receptor eliminated all butyrate-induced calcium responses. These findings show that the microfluidic tri-culture model replicates a multistep signaling pathway in which butyrate activates epithelial GPR41/GPR43, triggers rapid 5-HT release from CECs, and induces sensory neuronal activation through 5-HT3ar dependent calcium influx. This compartmentalized model provides a tool for studying microbial metabolite driven epithelial-neuronal communication in the gut. This work was supported by the National Heart, Lung, and Blood Institute grants R21HL179596 and R01HL152162. This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.