Fig. 3

Pax2/8 act redundantly to specify inhibitory fates. Lateral views of in situ hybridisation for markers of glycinergic (A and D), GABAergic (B and E) glutamatergic (C and F) and acetylcholinergic (J and K) cell fates at 24 h in WT (A–C and J) and triple-knock-down (TKD) trunks (D–F and K). In (J and K) stars indicate cells expressing the acetylcholinergic marker, chat. In these photographs, some of the weaker cells are in a slightly different focal plane (on the other side of the spinal cord). In (B and E) the black line indicates the boundary between dorsal GABAergic cells within the Pax2/8 expression domain (rows 4–7) and more ventral GABAergic cells. Note that most remaining GABAergic cells in (E) are in the ventral region where Pax2/8 are not expressed and that none of the acetylcholinergic cells are within the Pax2/8 expression domain in either wild-type (J) or TKD (K) embryos. Rostal is left, dorsal is top. Scale bar = 40 μm. Immunohistochemistry for Pax2 in WT (G), noi (pax2a) mutant (H) and TKD (I) embryos. As Pax2 staining is present in (H) the antibody must recognise Pax2b. Therefore, lack of staining in (I) suggests that we have eliminated Pax2a + Pax2b protein in TKD embryos. (M and N) number of glycinergic and GABAergic cells at 24 h in Pax2/8 expression domain in WT, single-, double- and triple-knock-down embryos. Stars indicate statistically significant results (p < 0.05). For results in whole spinal cord see Supp. Data Figs. 4C and C′. For results of additional single and double knock-down experiments see Supp. Data Fig. 3 and Supp. Data Tables 1 and 2. (L and O–Q) show results for control and TKD embryos for all 4 neurotransmitters, broken down row by row. Rows not shown have a cell count of zero. In all cases, error bars denote standard deviation. The number of glycinergic and GABAergic cells is reduced in the Pax2/8 expression domain in TKD embryos but there is no significant change in the number of glutamatergic or acetylcholinergic cells in any region of the spinal cord.