Neuropod cells

The wiki’s clearest case of a placeholder being replaced by a mechanism.

What was assumed, and for how long

The brain senses body surfaces through innervated epithelial sensors: the Merkel cell synapses onto somatosensory afferents so we feel fine textures; the taste cell synapses onto facial and vagal nerves so we perceive flavour. The gut lumen — the largest sensory surface the body has — was assumed to work differently: luminal stimuli reached nerves only through the slow paracrine action of hormones (cholecystokinin, serotonin, peptide YY) diffusing to nearby fibres.

Berntson & Khalsa (2021): “Until 2015, to our knowledge, no direct innervation of an epithelial sensory cell had been described” in the gut.

What was found

The hormone-secreting enteroendocrine cell turns out also to be a wired sensory transducer. Transgenic mouse models built to identify these cells permitted profiling of chemo-, thermo- and mechanoreceptors within them, and then showed them forming synapses throughout the gastrointestinal tract:

• Over two-thirds of enteroendocrine cells contact PGP9.5-positive sensory nerves in the intestinal mucosa (Bohorquez et al. 2015).
• In the colon, they synapse using serotonin to transduce noxious stimuli onto spinal nerves (Bellono et al. 2017).
• They reconstitute their connections with peripheral sensory neurons even when isolated and co-cultured in vitro — the wiring is a property of the cells, not of the tissue context.

These innervated epithelial sensors are neuropod cells.

In 2018 (Kaelberer et al., Science) they were shown to synapse onto nodose neurons, forming a pathway from the intestinal lumen to the NTS across a single vagal synapse, transducing a luminal stimulus in as little as 60 ms. Glucose entering the neuropod cell triggers glutamate release onto ionotropic vagal receptors — a molecular basis for fast gut-to-brain transduction.

The behavioural result

The part that makes this more than anatomy (Buchanan et al. 2020, preprint at the time of the review):

• A mouse distinguishes sugars from an artificial sweetener guided by neuropod-cell input.
• Silence the neuropod cells and it can no longer tell sucrose from sucralose.
• Stimulate them and it consumes sucralose as if it were caloric sugar.
• The cell uses two different transmitters for the two classes — glutamate for sugars, ATP for sweeteners.

An innate preference, determined by a specific sensory cell of the gut.

Why the wiki keeps it

1. It gives gut interoception a fast, wired, transmitter-specific channel. The wiki’s gut material — microbiota-gut-brain-axis, disorders-of-brain-gut-interaction, the enteroendocrine relay of short-chain fatty acids — has been slow and humoral, with the EEC appearing as a diffuse chemical relay. It is also a synaptic one. Route 3 on the microbiota page (“via enteroendocrine cells”) is more literal than that page’s phrasing implied.

2. A preference is set below reward and below taste. The sucrose/sucralose result places the discrimination neither on the tongue nor in dopaminergic valuation but in a gut epithelial sensor. This is the sharpest instance the wiki holds of the Quigley et al. claim that interoception is a party to perception at the point of assembly rather than a channel that terminates and then influences things — and a concrete case for HRL-style arguments that regulation can proceed without anything resembling a felt bodily state. Nothing here requires an interocept.

3. It is a wiring diagram where the wiki usually has an inference. Selective silencing and stimulation of an identified cell type with a clean behavioural readout is a stronger design than nearly anything in the cardiac-perception literature — and, as with gut-hippocampal-memory, it exists because the work is in mice. That is the cross-species gap in its productive direction.

Caveats worth carrying

Entirely rodent. Held secondhand; none of the primary papers are in raw/. The sugar-preference result was unpublished at review time. And the review flags a limit on the pathway’s scope: this innate vagal signalling circuit “may be different from those involved in conditioned learning, that are vagus-independent” (Sclafani et al.) and could involve cortical and subcortical networks instead. So neuropod transduction is one gut-to-brain channel, not the gut-to-brain channel — the same lesson interoceptive-sensors teaches about the periphery generally.