Mechanisms underlying the gut-brain communication: How enterochromaffin (EC) cells activate vagal afferent nerve endings in the small intestine.

How the gastrointestinal tract communicates with the brain, via sensory nerves, is of significant interest for our understanding of human health and disease. Enterochromaffin (EC) cells in the gut mucosa release a variety of neurochemicals, including the largest quantity of 5-hydroxytryptamine (5-HT) in the body. How 5-HT and other substances released from EC cells activate sensory nerve endings in the gut wall remains a major unresolved mystery. We used in vivo anterograde tracing from nodose ganglia to determine the spatial relationship between 5-HT synthesizing and peptide-YY (PYY)-synthesizing EC cells and their proximity to vagal afferent nerve endings that project to the mucosa of mouse small intestine. The shortest mean distances between single 5-HT- and PYY-synthesizing EC cells and the nearest vagal afferent nerve endings in the mucosa were 33.1 ± 14.4 µm (n = 56; N = 6) and 70.3 ± 32.3 µm (n = 16; N = 6). No morphological evidence was found to suggest that 5-HT- or PYY-containing EC cells form close morphological associations with vagal afferents endings, or varicose axons of passage. The large distances between EC cells and vagal afferent endings are many hundreds of times greater than those known to underlie synaptic transmission in the nervous system (typically 10-15 nm). Taken together, the findings lead to the inescapable conclusion that communication between 5-HT-containing EC cells and vagal afferent nerve endings in the mucosa of the mouse small intestinal occurs in a paracrine fashion, via diffusion. New and Noteworthy None of the findings here are consistent with a view that close physical contacts occur between 5-HT-containing EC cells and vagal afferent nerve endings in mouse small intestine. Rather, the findings suggest that gut-brain communication between EC cells and vagal afferent endings occurs via passive diffusion. The morphological data presented do not support the view that EC cells are physically close enough to vagal afferent endings to communicate via fast synaptic transmission.

Identification of vagal afferent nerve endings in the mouse colon and their spatial relationship with enterochromaffin cells.

Understanding how the gut communicates with the brain, via sensory nerves, is of significant interest to medical science. Enteroendocrine cells (EEC) that line the mucosa of the gastrointestinal tract release neurochemicals, including the largest quantity of 5-hydroxytryptamine (5-HT). How the release of substances, like 5-HT, from enterochromaffin (EC) cells activates vagal afferent nerve endings is unresolved. We performed anterograde labelling from nodose ganglia in vivo and identified vagal afferent axons and nerve endings in the mucosa of whole-mount full-length preparations of mouse colon. We then determined the spatial relationship between mucosal-projecting vagal afferent nerve endings and EC cells in situ using 3D imaging. The mean distances between vagal afferent nerve endings in the mucosa, or nearest varicosities along vagal afferent axon branches, and the nearest EC cell were 29.6 ± 19.2 µm (n = 107, N = 6) and 25.7 ± 15.2 µm (n = 119, N = 6), respectively. No vagal afferent endings made close contacts with EC cells. The distances between EC cells and vagal afferent endings are many hundreds of times greater than known distances between pre- and post-synaptic membranes (typically 10-20 nm) that underlie synaptic transmission in vertebrates. The absence of any close physical contacts between 5-HT-containing EC cells and vagal afferent nerve endings in the mucosa leads to the inescapable conclusion that the mechanism by which 5-HT release from ECs in the colonic mucosa occurs in a paracrine fashion, to activate vagal afferents.

Anatomical distribution of CGRP-containing lumbosacral spinal afferent neurons in the mouse uterine horn.

Sensory stimuli from the uterus are detected by spinal afferent neurons whose cell bodies arise from thoracolumbar and lumbosacral dorsal root ganglia (DRG). Using an in vivo survival surgical technique developed in our laboratory to remove select DRG from live mice, we recently quantified the topographical distribution of thoracolumbar spinal afferents innervating the mouse uterine horn, revealed by loss of immunoreactivity to calcitonin gene-related peptide (CGRP). Here, we used the same technique to investigate the distribution of lumbosacral uterine spinal afferents, in which L5-S1 DRG were unilaterally removed from adult female C57BL/6J mice (N = 6). Following 10-12 days recovery, CGRP immunoreactivity was quantified along the length of uterine horns using fluorescence immunohistochemistry. Relative to myometrial thickness, overall CGRP density in uterine tissues ipsilateral to L5-S1 DRG removal was reduced compared to the DRG-intact, contralateral side (P = 0.0265). Regionally, however, myometrial CGRP density was unchanged in the cranial, mid, and caudal portions. Similarly, CGRP-expressing nerve fiber counts, network lengths, junctions, and the proportion of area occupied by CGRP immunoreactivity were unaffected by DRG removal (P ≥ 0.2438). Retrograde neuronal tracing from the caudal uterine horn revealed fewer spinal afferents here arise from lumbosacral than thoracolumbar DRG (P = 0.0442) (N = 4). These data indicate that, unlike thoracolumbar DRG, lumbosacral spinal afferent nerves supply relatively modest sensory innervation across the mouse uterine horn, with no regional specificity. We conclude most sensory information between the mouse uterine horn and central nervous system is likely relayed via thoracolumbar spinal afferents.

Disengaging spinal afferent nerve communication with the brain in live mice.

Our understanding of how abdominal organs (like the gut) communicate with the brain, via sensory nerves, has been limited by a lack of techniques to selectively activate or inhibit populations of spinal primary afferent neurons within dorsal root ganglia (DRG), of live animals. We report a survival surgery technique in mice, where select DRG are surgically removed (unilaterally or bilaterally), without interfering with other sensory or motor nerves. Using this approach, pain responses evoked by rectal distension were abolished by bilateral lumbosacral L5-S1 DRG removal, but not thoracolumbar T13-L1 DRG removal. However, animals lacking T13-L1 or L5-S1 DRG both showed reduced pain sensitivity to distal colonic distension. Removal of DRG led to selective loss of peripheral CGRP-expressing spinal afferent axons innervating visceral organs, arising from discrete spinal segments. This method thus allows spinal segment-specific determination of sensory pathway functions in conscious, free-to-move animals, without genetic modification.

The gut-brain axis: spatial relationship between spinal afferent nerves and 5-HT-containing enterochromaffin cells in mucosa of mouse colon.

Cross talk between the gastrointestinal tract and brain is of significant relevance for human health and disease. However, our understanding of how the gut and brain communicate has been limited by a lack of techniques to identify the precise spatial relationship between extrinsic nerve endings and their proximity to specific cell types that line the inner surface of the gastrointestinal tract. We used an in vivo anterograde tracing technique, previously developed in our laboratory, to selectively label single spinal afferent axons and their nerve endings in mouse colonic mucosa. The closest three-dimensional distances between spinal afferent nerve endings and axonal varicosities to enterochromaffin (EC) cells, which contain serotonin (5-hydroxytryptamine; 5-HT), were then measured. The mean distances (± standard deviation) between any varicosity along a spinal afferent axon or its nerve ending, and the nearest EC cell, were 5.7 ± 6.0 µm (median: 3.6 µm) and 26.9 ± 18.6 µm (median: 24.1 µm), respectively. Randomization of the spatial location of EC cells revealed similar results to this actual data. These distances are ∼200-1,000 times greater than those between pre- and postsynaptic membranes (15-25 nm) that underlie synaptic transmission in the vertebrate nervous system. Our findings suggest that colonic 5-HT-containing EC cells release substances to activate centrally projecting spinal afferent nerves likely via diffusion, as such signaling is unlikely to occur with the spatial fidelity of a synapse.NEW & NOTEWORTHY We show an absence of close physical contact between spinal afferent nerves and 5-HT-containing EC cells in mouse colonic mucosa. Similar relative distances were observed between randomized EC cells and spinal afferents compared with actual data. This spatial relationship suggests that substances released from colonic 5-HT-containing EC cells are unlikely to act via synaptic transmission to neighboring spinal afferents that relay sensory information from the gut lumen to the brain.

Novel intrinsic neurogenic and myogenic mechanisms underlying the formation of faecal pellets along the large intestine of guinea-pigs.

Soft faecal material is transformed into discrete, pellet-shaped faeces at the colonic flexure. Here, analysis of water content in natural faecal material revealed a decline from cecum to rectum without significant changes at the flexure. Thus, pellet formation is not explained by changes in viscosity alone. We then used video imaging of colonic wall movements with electromyography in isolated preparations containing guinea-pig proximal colon, colonic flexure and distal colon. To investigate the pellet formation process, the colonic segments were infused with artificial contents (Krebs solution and 4-6% methylcellulose) to simulate physiological faecal content flow. Remarkably, pellet formation took place in vitro, without extrinsic neural inputs. Infusion evoked slowly propagating neurogenic contractions, the proximal colon migrating motor complexes (∼0.6 cpm), which initiated pellet formation at the flexure. Lesion of the flexure, but not the proximal colon, disrupted the formation of normal individual pellets. In addition, a distinct myogenic mechanism was identified, whereby slow phasic contractions (∼1.9 cpm) initiated at the flexure and propagated short distances retrogradely into the proximal colon and antegradely into the distal colon. There were no detectable changes in the density or distribution of pacemaker-type interstitial cells of Cajal across the flexure. The findings provide new insights into how solid faecal content is generated, suggesting the major mechanisms underlying faecal pellet formation involve the unique interaction at the colonic flexure between antegrade proximal colon migrating motor complexes, organized by enteric neurons, and retrograde myogenic slow phasic contractions. Additional, as yet unidentified extrinsic and/or humoral influences appear to contribute to processing of faecal content in vivo. KEY POINTS: In herbivores, including guinea-pigs, clearly defined faecal pellets are formed at a distinct location along the large intestine (colonic flexure). The mechanism underlying the formation of these faecal pellets at this region has remained unknown. We reveal a progressive and gradual reduction in water content of faecal content along the bowel. Hence, the distinct transition from amorphous to pellet shaped faecal content could not be explained by a dramatic increase in water reabsorption from a specific site. We discovered patterns of anterograde neurogenic and retrograde myogenic motor activity that facilitate the formation of faecal pellets. The formation of 'pellet-like' boluses at the colonic flexure involves interaction of an antegrade migrating motor complex in the proximal colon and retrograde myogenic slow phasic contractions that emerge from the colonic flexure. The findings uncover intrinsic mechanisms responsible for the formation of discrete faecal scybala in the large intestine of a vertebrate.

Morphological identification of thoracolumbar spinal afferent nerve endings in mouse uterus.

Major sensory innervation to the uterus is provided by spinal afferent nerves, whose cell bodies lie predominantly in thoracolumbar dorsal root ganglia (DRG). While the origin of the cell bodies of uterine spinal afferents is clear, the identity of their sensory endings has remained unknown. Hence, our major aim was to identify the location, morphology, and calcitonin gene-related peptide (CGRP)-immunoreactivity of uterine spinal afferent endings supplied by thoracolumbar DRG. We also sought to determine the degree of uterine afferent innervation provided by the vagus nerve. Using an anterograde tracing technique, nulliparous female C57BL/6 mice were injected unilaterally with biotinylated dextran into thoracolumbar DRG (T13-L3). After 7-9 days, uterine horns were stained to visualize traced nerve axons and endings immunoreactive to CGRP. Whole uteri from a separate cohort of animals were injected with retrograde neuronal tracer (DiI) and dye uptake in nodose ganglia was examined. Anterogradely labeled axons innervated each uterine horn, these projected rostrally or caudally from their site of entry, branching to form varicose endings in the myometrium and/or vascular plexus. Most spinal afferent endings were CGRP-immunoreactive and morphologically classified as "simple-type." Rarely, uterine nerve cell bodies were labeled in nodose ganglia. Here, we provide the first detailed description of spinal afferent nerve endings in the uterus of a vertebrate. Distinct morphological types of spinal afferent nerve endings were identified throughout multiple anatomical layers of the uterine wall. Compared to other visceral organs, uterine spinal afferent endings displayed noticeably less morphological diversity. Few neurons in nodose ganglia innervate the uterus.

Sensory nerve endings arising from single spinal afferent neurons that innervate both circular muscle and myenteric ganglia in mouse colon: colon-brain axis.

There is considerable interest in understanding how contents within the gut wall (including microbiome) can activate sensory nerve endings in the gut that project to the central nervous system. However, we have only recently begun to understand the location and characteristics of extrinsic spinal afferent nerve endings that innervate the lower gastrointestinal (GI) tract. Our aim is to identify the nerve endings in the mouse distal colon that arise from single spinal afferent neurons. C57BL/6 mice were anaesthetised and single dorsal root ganglia (DRG) between lumbosacral L6-S1 were injected with dextran biotin. Mice recovered for 7 days. Animals were then euthanized and whole colons removed, fixed and stained for calcitonin-gene-related-peptide (CGRP). Single spinal afferent nerve axons were identified entering the distal colon that ramified along many rows of myenteric ganglia, often giving rise to varicose nerve endings. These same axons bifurcated in the circular muscle giving rise to 4-5 groups of branching-type intramuscular endings, where each group of endings was separated by ~ 370 µm in the rostro-caudal axis and projected 1.2 mm around the circumference. As spinal afferent axons bifurcated, their axons often showed dramatic reductions in diameter. Here, we identified in the distal colon, the characteristics of nerve endings that arise from single colorectal-projecting axons with cell bodies in DRG. These findings suggest that a population of sensory neurons in DRG can potentially detect sensory stimuli simultaneously via different morphological types of endings that lie in both colonic smooth muscle and myenteric ganglia.

Identification of spinal afferent nerve endings in the colonic mucosa and submucosa that communicate directly with the spinal cord: The gut-brain axis.

The major sensory nerve pathway between the colon and central nervous system (spinal cord and brain) that underlies the gut-brain axis, is via spinal afferent neurons, with cell bodies in dorsal root ganglia (DRG). Our aim was to identify the sensory nerve endings in the colon that arise from single colorectal-projecting DRG neurons. C57BL/6 mice were anesthetized and lumbosacral L6-S1 DRG injected with dextran biotin. Mice recovered for 7 days. The whole colon was then removed and stained to visualize single axons and nerve endings immunoreactive to calcitonin gene-related peptide (CGRP). Single axons arising from DRG were identified in the distal colon and their morphological features and CGRP immunoreactivity characterized. After entering the colon, single axons ramified rostrally or caudally along many rows of myenteric ganglia with little circumferential displacement, giving off varicose endings in multiple ganglia. Nerve endings arising from two classes of colorectal-projecting DRG neuron were identified. One class was peptidergic neurons that had nerve endings in circular muscle, myenteric ganglia, and submucosa. Another class of nonpeptidergic neurons innervated mucosal crypts, myenteric ganglia, and submucosa. Different morphological types of nerve endings which innervate different anatomical layers of colon can arise from the same axon and sensory neuron in DRG. These findings suggest single peptidergic and nonpeptidergic sensory neurons in DRG are potentially capable of detecting sensory stimuli from different anatomical layers of the colon, via different types of nerve endings.

Identification of different functional types of spinal afferent neurons innervating the mouse large intestine using a novel CGRPα transgenic reporter mouse.

Spinal afferent neurons detect noxious and physiological stimuli in visceral organs. Five functional classes of afferent terminals have been extensively characterized in the colorectum, primarily from axonal recordings. Little is known about the corresponding somata of these classes of afferents, including their morphology, neurochemistry, and electrophysiology. To address this, we made intracellular recordings from somata in L6/S1 dorsal root ganglia and applied intraluminal colonic distensions. A transgenic calcitonin gene-related peptide-α (CGRPα)-mCherry reporter mouse, which enabled rapid identification of soma neurochemistry and morphology following electrophysiological recordings, was developed. Three distinct classes of low-threshold distension-sensitive colorectal afferent neurons were characterized; an additional group was distension-insensitive. Two of three low-threshold classes expressed CGRPα. One class expressing CGRPα discharged phasically, with inflections on the rising phase of their action potentials, at low frequencies, to both physiological (<30 mmHg) and noxious (>30 mmHg) distensions. The second class expressed CGRPα and discharged tonically, with smooth, briefer action potentials and significantly greater distension sensitivity than phasically firing neurons. A third class that lacked CGRPα generated the highest-frequency firing to distension and had smaller somata. Thus, CGRPα expression in colorectal afferents was associated with lower distension sensitivity and firing rates and larger somata, while colorectal afferents that generated the highest firing frequencies to distension had the smallest somata and lacked CGRPα. These data fill significant gaps in our understanding of the different classes of colorectal afferent somata that give rise to distinct functional classes of colorectal afferents. In healthy mice, the majority of sensory neurons that respond to colorectal distension are low-threshold, wide-dynamic-range afferents, encoding both physiological and noxious ranges.