Quantification of neurod:EGFP depth in the OT for the experiment revealed 
in A (imply s.e.m. P < 0.01 n = 4 per group). D. WISH of MOnrg1IIatg-injected and the control MOnrg1IIatg5m-injected embryos for neurog1 and neurod mRNAs at 48 hpf. Scale bar, 100 m. E. A coronal section of WISH-stained embryos for nrg1 mRNA at 36 hpf. The approximate site of the section is shown in a lateral view in the inset. F. A schematic of a stacked image of confocal sections from a dorsal view and the higher magnification of the OT from a dorsal view (bottom). G. Impaired neurogenesis and a decrease in pH3-positive mitotic cells in the SVZ of MOnrg1IIatg-injected TgBAC(neurod:EGFP) embryos (nrg1IIatg) compared to the control embryos (nrg1IIatg5m) at 40 hpf. Scale bar, 20 m. H. Quantification of the number of pH3-positive cells in the VZ (blue) and the SVZ (red) of MOnrg1II-injected embryos and the control (white) embryos (mean s.e.m. P < 0.05, P < 0.01, P < 0.001 n = 6, 6, 4, for nrg1IIE1, n = 6, 5, 5 for nrg1IIE15m at 26, 33, 36 hpf, n = 3, 4 for nrg1IIatg, nrg1IIatg5m at 40 hpf, respectively)expression plasmid CMV:nrg1II-EGFP (S7 Fig). WISH showed increased expression of neurog1 and decreased expression of neurod in the optic tectum of MOnrg1II-injected embryos (Fig 4D), suggesting involvement of NRG1-II in production of post-mitotic neurons from neural progenitor cells. Expression of nrg1 was prominent in the apical region of the optic tectum (Fig 4E), suggesting production of NRG1 mainly by radial glial/neural progenitor cells in the ventricular zone.We therefore asked whether NRG1-II regulates either type of mitoses in the ventricular or subventricular zone (Fig 4FH). In 26-hpf embryos, in which neural stem/radial glial cells are a predominant population, the majority of mitoses occurred exclusively in the apical ventricular zone, and there was no significant difference in the number of mitotic cells between MOnrg1IIinjected and the control embryos (Fig 4H, 26 hpf). In contrast, the number of mitotic cells in MOnrg1II-injected embryos was significantly less than those in the controls at 33 and 36 hpf, both in the ventricular zone and in the sub-basal/sub-ventricular zone (Fig 4H, 33 and 36 hpf). Then at 40 hpf, while mitoses in the ventricular zone resumed, those in the sub-ventricular zone remained affected9399992 in MOnrg1II-injected embryos (Fig 4G and 4H, 40 hpf). These D-Glutamine results suggest that NRG1-II is involved in the production of neurons by stimulating neurogenic mitoses of neural progenitor cells in the sub-ventricular zone and promoting their mitoses in the ventricular zone.To examine which ErbB receptor participates in neuronal generation in the developing optic tectum, we introduced antisense MOs against ErbB receptors into embryos of Tg (pou4f1-hsp70l:GFP) and found that an antisense MO against erbb4 (MOerbb4 MOerbb4atg) impaired generation of post-mitotic neurons (Fig 5A and 5B phenotype ratios: erbb4atg, 0.66 0.05, n = 149, Standard control, 0.02 0.01, n = 161, p < 0.01).