The gut-brain axis embodies the bi-directional communication between the gastrointestinal tract and the central nervous system (CNS), where vagal afferent neurons (VANs) serve as sensors for a variety of gut-derived signals. The gut is colonized by a large and diverse population of microorganisms that communicate via small (effector) molecules, which also act on the VAN terminals situated in the gut viscera and consequently influence many CNS processes. However, the convoluted in vivo environment makes it difficult to study the causative impact of the effector molecules on VAN activation or desensitization. Here, we report on a VAN culture and its proof-of-principle demonstration as a cell-based sensor to monitor the influence of gastrointestinal effector molecules on neuronal behavior. We initially compared the effect of surface coatings (poly-L-lysine vs. Matrigel) and culture media composition (serum vs. growth factor supplement) on neurite growth as a surrogate of VAN regeneration following tissue harvesting, where the Matrigel coating, but not the media composition, played a significant role in the increased neurite growth. We then used both live-cell calcium imaging and extracellular electrophysiological recordings to show that the VANs responded to classical effector molecules of endogenous and exogenous origin (cholecystokinin serotonin and capsaicin) in a complex fashion. We expect this study to enable platforms for screening various effector molecules and their influence on VAN activity, assessed by their information-rich electrophysiological fingerprints.

Background

Vagal afferent neurons represent the key neurosensory branch of the gut-brain axis, which describes the bidirectional communication between the gastrointestinal system and the brain. These neurons are important for detecting and relaying sensory information from the periphery to the central nervous system to modulate feeding behavior, metabolism, and inflammation. Confounding variables complicate the process of isolating the role of the vagal afferents in mediating these physiological processes. Therefore, we developed a microfluidic model of the sensory branch of the gut-brain axis. We show that this microfluidic model successfully compartmentalizes the cell body and neurite terminals of the neurons, thereby simulates the anatomical layout of these neurons to more accurately study physiologically-relevant processes.

Methods

We implemented a primary rat vagal afferent neuron culture into a microfluidic platform consisting of two concentric chambers interconnected with radial microchannels. The microfluidic platform separated cell bodies from neurite terminals of vagal afferent neurons. We then introduced physiologically-relevant gastrointestinal effector molecules at the nerve terminals and assessed their retrograde transport along the neurite or capacity to elicit an electrophysiological response using live cell calcium imaging.

Results

The angle of microchannel outlets dictated the probability of neurites growing into a chamber versus tracking along chamber walls. When the neurite terminals were exposed to fluorescently-labeled cholera toxin subunit B, the proteins were taken up and retrogradely transported along the neurites over the course of 24 h. Additionally, mechanical perturbation (e.g., rinsing) of the neurite terminals significantly increased intracellular calcium concentration in the distal soma. Finally, membrane-displayed receptor for capsaicin was expressed and trafficked along newly projected neurites, as revealed by confocal microscopy.

Conclusions

In this work, we developed a microfluidic device that can recapitulate the anatomical layout of vagal afferent neurons in vitro. We demonstrated two physiologically-relevant applications of the platforms: retrograde transport and electrophysiological response. We expect this tool to enable controlled studies on the role of vagal afferent neurons in the gut-brain axis.