Extended Data Fig. 7: Morphological features of V2a IN subtypes.

a, Esrrga⁺ V2a INs displayed only descending axonal projections. Arrow indicates continuing axonal projection. b, Lateral view of the morphology of two examples of Shox2⁺ V2a INs. c, vAChTa⁺ V2a INs mostly displayed bidirectional axonal projections. A specialized vAChTa⁺ V2a IN subtype showed located axonal projection and corresponds to the previously described escape V2a IN⁴¹. d, Box and whiskers plot of soma sizes of the three V2a INs subtypes (one-way ANOVA with Tukey’s post hoc multiple comparisons, ****p < 0.0001) (n = 12 Esrrga⁺; n = 12 Shox2⁺; n = 14 vAChTa⁺ V2a INs). e, Length of the axonal projections of Esrrga⁺ (magenta), Shox2⁺ (yellow) and vAChTa⁺ (blue) V2a INs (n = 5 neurons per subtype). f, V2a IN axons projected along topographically organized paths over the dorso-ventral and medio-lateral aspects of the spinal cord (n = 13 Esrrga⁺: magenta; n = 13 shox2⁺: yellow; n = 30 vAChTa⁺: blue). g, Recordings of a fast V2a IN displaying adapting firing that was recruited only at fast swimming frequency (>7-8 Hz). h, Top left: Example of dual patch-clamp recordings showing the lack of monosynaptic connection between a slow, Esrrga⁺ V2a IN and a fast MN. Bottom left: Morphological reconstruction of the recorded V2a IN-MN pair (black: soma and dendrites; red: V2a IN axons). Top right: Example of dual patch-clamp recording showing the absence of connections between a fast, vAChTa⁺ V2a IN and a slow MN. Bottom right: Morphological reconstruction of the recorded V2a IN-MN pair (black: soma and dendrites; red: V2a IN axons).

Source data