DEV032276 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Ectopic posterior clusters express the pontine-specific markers Nfix and Nfib. Among five candidates chosen from a subtractive screen for identifying late pontine-specific genes by Gesemann et al. (Gesemann et al., 2001), we found the Nuclear factor I (NFI) family members Nfix (NFI-X) and Nfib (NFI-B) named clone 47(1)3 in the original screen were expressed at a much higher level in PN than in the LRN or ECN from E16.5 onwards, making them suitable PN-specific markers (data not shown) (Kumbasar et al., 2009). Adjacent transverse sections from an E16.5 Cxcr4^−/− hindbrain were subjected to in situ hybridisation with Mbh2, Nfix and Nfib probes. Four axial levels are illustrated here. Anterior superficial (A, arrows) and posterior deep Mbh2-labelled (B, arrows) clusters showed strong Nfix and Nfib signals. The ECN (C, arrows) and LRN (D, arrows) expressed only low levels of Nfix and Nfib. Scale bar: 400 µm.

  References

  Gesemann, M., Litwack, E. D., Yee, K. T., Christen, U. and O'Leary, D. D. (2001). Identification of candidate genes for controlling development of the basilar pons by differential display PCR. Mol. Cell. Neurosci. 18, 1-12.

  Kumbasar, A., Plachez, C., Gronostajski, R. M., Richards, L. J. and Litwack, E. D. (2009). Absence of the transcription factor Nfib delays the formation of the basilar pontine and other mossy fiber nuclei. J. Comp. Neurol. 513, 98-112

• Supplemental Figure S2 -

  Fig. S2. Pontine neurons derailed from their marginal stream migrate circumferentially towards the midline. In utero electroporation delivered Egfp into the LRL of an E12.5 Cxcr4^−/− embryo and the labelled sample was analysed at E14.5. (A) A transverse section showing many Egfp-labelled cells migrating deep within the hindbrain parenchyma towards the midline anterior to the PES. Those that reached the midline (white vertical line) aggregated on both sides of the midline to form the deep ectopic pontine cluster. Laminin immunohistochemistry highlights the outline of the section. (B,C) Higher-magnification views of boxes b and c, respectively, in A, depicting regions both close to (B) and at a distance from (C) the LRL. In both regions, the cell bodies as well as the leading processes of migrating cells extended along the circumferential axis (arrowheads), suggesting that the derailed pontine neurons from their usual marginal stream migrated circumferentially without turning anteriorly. Scale bar: 400 µm in A; 80 µm in B,C.

• Supplemental Figure S3 -

  Fig. S3. Conditional knockout of Cxcr4 in Wnt1-labelled rhombic lip results in ectopic PN clusters that persist until at least P5. A Wnt1-Cre driver and a Cxcr4^fl/fl responder line were used to knockout Cxcr4 in PCN precursors at LRL. At P5, Nissl staining of a parasagittal section from a Wnt1-Cre; Cxcr4^fl/fl hindbrain showed PN at their presumptive position are almost missing (B); compare with the control hindbrain of Cxcr4^fl/fl (A). Instead, multiple ectopic clusters were formed between the pontine flexure and the IO (arrows in B). The ectopic clusters were PAX6-immunoreactive (arrows in D), confirming their PN identity. The validity of PAX6 as a PN marker at this stage was demonstrated in the control section (C). These results confirmed the cell-autonomous role of CXCR4 in PCN migration (see Fig. 6) and further showed that at least some ectopic pontine neurons persisted into postnatal stages. Scale bar: 400 µm.

• Supplemental Figure S4 -

  Fig. S4. SDF1 protein shows an anterior-high, posterior-low graded distribution underlying the anteriorly migrating pontine neurons. Parasagittal sections of E14.5 hindbrains with pial meninges attached were subjected to SDF1 and PAX6 double immunohistochemistry. (A) Ventral view of an E14.5 hindbrain with the AES (green) and the plane of parasagittal section (dashed lines) indicated. (B) Schematic of a section from the lateral hindbrain that captured the anterior migratory segment (green) of the AES. The boxed area approximately corresponds to the images in C-F. (C-F) Sections from two different brains showing merged images of SDF1 and PAX6 double signals (C,E), and images of SDF1 signal only (D,F). Adjacent to the PAX6-labelled anteriorly migrating pontine neurons, distribution of SDF1 protein appears high anteriorly and low posteriorly. The graded distribution of SDF1 was quantified using MetaMorph (version 6.1, Universal Imaging Corporation). First, an outline was drawn along the PAX6-labelled anterior migratory path at the pial surface of the hindbrain. Then the meningeal area underlying the outline was segmented into equal size bins, each bin with a width and depth of 50,×20 pixels and its average fluorescence measured. The normalized intensities of 13 bins numbered from the anterior end of the migratory path were plotted on a line graph (G). Quantifications of seven sections (derived from four different brains) are illustrated with the green curve representing the sample in C and the red curve the sample in E. Scale bar: 200 µm. A, anterior; CB, cerebellum; P, posterior.

• Supplemental Figure S5 -

  Fig. S5. Ectopically expressed SDF1 prevents pontine neurons from leaving LRL. Either Egfp alone or a mixture of Sfd1 and Egfp expression vectors was electroporated unilaterally into the LRL in utero at E12.5. The labelled samples were imaged at E15.5, and then subjected to whole-mount Mbh2 in situ hybridisation. (A) Electroporation of Egfp alone labelled both the PES (arrow) and AES (arrowhead) (n=5). (B) At the electroporated LRL, a region of reduced fluorescence (bracket) was often present, indicating that many electroporated PCN had left the LRL. (C,D) Mbh2 in situ hybridisation on these samples showed bilaterally symmetrical PN, LRN and ECN. (E) When Sdf1 was electroporated together with Egfp, either no AES or a tiny fraction of AES were labelled in seven out of ten samples. (I) The remaining three samples showed disrupted AES similar to that in Cxcr4^−/− (arrowheads). (F,J) The electroporated LRL of these samples showed intense fluorescence suggesting that many electroporated PCN could not leave this region (arrows). (G,K) Whole-mount Mbh2 in situ hybridisation on these samples showed that PN formation was almost abolished on the electroporated side (asterisks). The marked reduction in PN and AES suggests that most pontine neuron precursors, both electroporated and unelectroporated, were retained in the LRL, probably by the high concentration of SDF1 secreted from the electroporated LRL cells. The trapping of PCN could be visualized in some samples as clusters of Mbh2-labelled cells that remained at the LRL (arrow in H). The PES, by contrast, seemed to be able to migrate out of the LRL (E) and form LRN and ECN (G,H). This is most likely to be due to the fact that PES cells had already left the LRL by the time SDF1 expression began to accumulate several hours after E12.5, in contrast to the later birth date (around E13.5) of AES cells. Taken together, these data indicate that an ectopic source of SDF1 could prevent PCN from leaving this source, implicating a chemoattractive activity of SDF1 in vivo. Scale bar: 400 µm.