Fig. 3: Mucosal nerve endings are sensitive to 5-HT and ATP.

a, The MP was imaged while 5-HT or ATP was injected into a single villus. b,f, 5-HT (b) and ATP (f) activated a subset of neurons (arrowheads). c,g, Ca²⁺ transients in 5-HT (c) and ATP (g) responders. Agonists were applied at t = 10 s. d,h, Left, responders (cyan) to 5-HT (d) and ATP (h). Middle, immunolabelling for calbindin and nNOS. Right, overlays show that many 5-HT and ATP responders were calbindin⁺. e,i, Repeated 5-HT (e) or ATP (i) injections evoked reproducible responses under control conditions. e, The 5-HT response was abolished by ondansetron (Ond, 10 µM) in terms of the percentage of responders (top, two-way ANOVA, Šídák’s multiple comparisons test, n = 3 preparations examined) and the response amplitude (bottom, two-sided unpaired t-test). i, The ATP response was significantly reduced by TNP-ATP (5 µM) in terms of the percentage of responders (top, two-way ANOVA, Šídák’s multiple comparisons test, n = 4–5) and the response amplitude (bottom, two-sided unpaired t-test). j, Stimulating or antagonist solutions were perfused onto the mucosal surface. k, Left, the percentage of myenteric responders to mucosal high K⁺ was significantly reduced by ondansetron (10 µM), but not by TNP-ATP (5 µM), when compared with the corresponding control time point (two-way ANOVA, Sidak’s multiple comparisons test, n = 5–7). Right, response amplitudes were reduced by ondansetron (10 µM), but not by TNP-ATP (5 µM) (one-way ANOVA, Sidak’s multiple comparisons test). Data are mean ± s.e.m. Scale bars, 20 µm. Drawings in a and j are adapted from ref. ³⁰ (Wiley).

Source Data