DEV022616 Supplementary Material

Files in this Data Supplement:

• Supplemental Table S1 - Adobe PDF

• Supplemental Table S2 - Adobe PDF

• Supplemental Figure S1 -

  Fig. S1. Expression of marker genes in single-cell cDNAs determined by Q-PCR. The markers included cell-cycle-related genes for distinguishing between progenitor and post-mitotic cells (Ki67, cyclin B1 and cyclin E1), SVZ genes as positional markers to distinguish SVZ from VZ cells (Svet1, Epha3 and HuB), and transcription factors that demonstrated whether a cell was undifferentiated or was differentiating into a neuron (Hes1, Neurog2, Pax6 and Sox2). We also examined Dlx1 expression (Panganiban and Rubenstein, 2002) to identify tangentially migrating cells from the ventral forebrain. The relative gene expression levels were calculated by 40-Ct values and are color-coded from red (high) to black (low). The Ct values were obtained from the Q-PCR analyses of the first PCR products of the single-cell cDNAs using the primer pairs listed in Table 1. If the expression of the gene was undetectable by Q-PCR, the Ct value was estimated to be 40. Based on these results, the cells were categorized into five groups (Groups A-E) (see Table 2).

  Reference

  Panganiban, G. and Rubenstein, J. L. (2002). Developmental functions of the Distal-less/Dlx homeobox genes. Development 129, 4371-4386.

• Supplemental Figure S2 -

  Fig. S2. Expression levels of amplified spiked RNAs are proportional to their copy numbers. The probe set IDs were AFFX-LysX-3_at, AFFX-DapX-3_at, AFFX-PheX-3_at, and AFFX-ThrX-3_at. Box plot of log-transformed expression levels, which consist of the smallest observation, lower quartile (Q1), median, upper quartile (Q3) and largest observation. RNAs spiked at 20 copies per cell or more were consistently and proportionally amplified by our method.

• Supplemental Figure S3 -

  Fig. S3. Cell-cycle phase and gene expression pattern in single-cell cDNAs. (A) Two-way cluster analysis of the expression levels of cell-cycle progression-related genes (528 probe sets, based on the gene ontogeny information) in cDNAs from 57 progenitor-derived single cells. Each column indicates one cell; a magnified view of the cell clusters is shown in F. Single-cell cDNAs that showed similar cell-cycle-related gene expression patterns were in a similar phase of the cell cycle. (B-C′) In situ hybridization for Rpa2 (B) and Ccnb2 (C) in the E14 mouse brain. The same sections were immunostained with the anti-BrdU antibody (B′,C′); BrdU had been administered 30 and 15 minutes (B,B′) (to label the cells in S phase) or 8 and 7.5 hours (C,C′) (to label the cells in G0/G1 phase) prior to fixation. Rpa2 (and Mcm2, not shown) was expressed in the 30-minute-labeled cells (B,B′), but not in the 8-hour-labeled cells (not shown), whereas Ccnb2 (and also Ccnb1, Cdc20, not shown) was not expressed in the 30-minute-labeled cells (not shown) but was expressed in the 8-hour-labeled cells (C,C′). Dotted line, apical surface. (E) Schematic presentation of the relationship of the mRNA expression levels of the five genes and the cell-cycle phases described above. (F) Tentative assignment of cell-cycle phase and gene expression. The cluster tree (A) is magnified, and labeled with cell-cycle phases as inferred from E. Since the Dll1 in situ hybridization pattern suggests that high-Dll1 Cluster II cells (B-22P, B-5B, B-17H and X-16K) are in the early G1 phase, three Cluster I cells (X-15i, B-17F, A-15M) near the Cluster II cells were assigned to the early G1 phase. The Hes1 expression level was low in these cells, consistent with the recent report showing that Hes1 expression is undetectable in early G1 phase (Shimojo et al., 2008). We also found variations in Dll1, Neurog2 and Cbf1 expression among Cluster I samples at similar cell-cycle phases.

  References

  Cho, R. J., Huang, M., Campbell, M. J., Dong, H., Steinmetz, L., Sapinoso, L., Hampton, G., Elledge, S. J., Davis R. W. and Lockhart, D. J. (2001). Transcriptional regulation and function during the human cell cycle. Nat. Genet. 27, 48-54.

  Whitfield, M. L., Sherlock, G., Saldanha, A. J., Murray, J. I., Ball, C. A., Alexander, K. E., Matese, J. C., Perou, C. M., Hurt, M. M., Brown, P. O. and Botstein, D. (2002). Identification of genes periodically expressed in the human cell cycle and their expression in tumors. Mol. Biol. Cell 13, 1977-2000.

• Supplemental Figure S4 -

  Fig. S4. Scatter diagrams of the gene expression levels of cDNAs from 70 single cells. (A-J) Scatter diagrams for the expression level of selected genes versus Hes1 with respect to cDNAs from 70 single cells. Each symbol indicates one cell. The probe set ID of each gene is shown in Fig. 5A. Gli3 (A) and Fzd8 (B) as well as Fgfr3 (Fig. 5B) represent genes showing invariably high expression levels in the Cluster I cells (black diamonds). Dll3 (H), Neurod1 (I) and Neurog1 (J) as well as Neurod2 (Fig. 5E) are examples of genes, the expression of which is invariably low in the Cluster I cells. Hes5, Ascl1 (Mash1) (E), Neurod6 (F) and Jag1 (G) show variable expression levels in the Cluster I cells, as observed for Neurog2 (Fig. 5C) and Dll1 (Fig. 5D). (K,L) Scatter diagrams for the expression level of Rbpj (Cbf1) versus Neurog2 (K) or Dll1 (L) with respect to cDNAs from 70 single cells.

• Supplemental Figure S5 -

  Fig. S5. Scatter diagrams of the gene expression levels of cDNAs from 70 single cells. (A-M) Scatter diagrams for the expression level of the selected genes versus Hes5 with respect to cDNAs from 70 single cells. Each symbol indicates one cell. Probe set ID of each gene is shown in Fig. 5A. The Cluster I cells are segregated into two groups according to high or low levels of Ascl1 (F). The same is true for Dll1 (G) and Hes5 (x-axis).

• Supplemental Figure S6 -

  Fig. S6. Combinations of Hes5, Ascl1 and Dll1 expression separate the Cluster I cells into four groups. (A-D) Scatter diagrams of the expression level of Hes5, Ascl1 and Dll1 for the 33 Cluster I cells. Each symbol indicates one Cluster I cell. In 32 of the 33 Cluster I cells examined, Hes5, Mash1 and Dll1 were expressed at either high or low levels. Consequently, these 32 Cluster I cells were classified into four groups according to the expression levels of these genes. The four groups showed the following respective combinations of Hes5, Ascl1 and Dll1 expression levels: seven cells, low, high, low (yellow cells); eight cells, high, low, low (green cells); nine cells, high, high, high (blue cells); and eight cells, high, high, low (purple cells). These four categories do not seem to reflect artifacts resulting from cDNA preparation or microarray procedures because only a limited set of the eight possible combinations were found.

• Supplemental Figure S7 -

  Fig. S7. Tbr2⁺ cells commit to become basal progenitor cells in the early G1 phase. (A,B) E14 CD1 cerebral wall double immunostained for Tbr2 (green) and BrdU (magenta). (A) A mouse was injected with BrdU for 2 hours (four injections at 0.5-hour intervals) to label the cells in the S, G2 and M phases. Arrowheads indicate Tbr2⁺ BrdU⁺ cells in the SVZ. (B) A mouse was injected with BrdU 8.0 hours and 7.5 hours before fixation to label the daughter cells in the G0/G1 phase. Arrows indicate Tbr2⁺ BrdU⁺ cells in the VZ. (C,C′) E14 CD1 cerebral wall triple stained for Tbr2 (green), cyclin A (magenta) and Ki67 (white). Tbr2⁺ Ki67⁺ cells in the VZ were cyclin A-negative (double arrowheads). In the SVZ, there were some Tbr2⁺ Ki67⁺ cyclin A⁺ cells (asterisks). (D) Egfp mRNA expression in the pTbr2:EGFP BAC transgenic mouse brain at E14, which mimics the Tbr2 mRNA expression (see Fig. 3Ac). (E) EGFP (green) and Tbr2 (magenta) expression in the cerebral wall of the E14 pTbr2:EGFP BAC transgenic mouse. In the VZ, all the EGFP⁺ cells were Tbr2⁺. In the IMZ, there were many EGFP⁺ cells with no Egfp mRNA expression (D), reflecting the stability of the EGFP protein. EGFP⁺ cell bodies were rarely seen in the VZ near the apical surface. Dotted line, apical surface. VZ, ventricular zone; SVZ, subventricular zone; IMZ, intermediate zone. pTbr2:EGFP (pEomes:EGFP) BAC transgenic mice were generated by the GENSAT Project, NINDS contract #N01NS02331 to The Rockefeller University (New York, NY).

• Supplemental Figure S8 -

  Fig. S8. Effect of Notch signaling inhibitor on monolayer cultured cells. (A-F) Effect of Notch inhibitor DAPT in E14 cerebral cell culture. DAPT (10 µM) (D-F) or DMSO vehicle (A-C) was present in the medium for 20 hours. Tbr2⁺ Ki67⁺ cells (arrowheads) were increased by the DAPT treatment. Anti-Ki67 (green); anti-Tbr2 (red); DAPI (blue). (G,H) Vcam1 immunoreactivity was almost abolished by DAPT treatment (H) compared with the DMSO control (G). Anti-Ki67 (green); anti-Vcam1 (red); DAPI (blue). (I) After 20 hours of DAPT treatment, the proportion of Ki67⁺ cells in the total cells was significantly decreased (P=0.0152, Mann-Whitney test, n=8 wells for DAPT, n=9 wells for DMSO), whereas the proportion of Tbr2⁺ Ki67⁺ cells in the total cells (green) increased significantly (P=0.0025, Mann-Whitney test, n=8 wells for DAPT and n=9 wells for DMSO); error bars indicate s.d. Cell death determined by DAPI staining did not significantly increase following the DAPT treatment (3.4±1.1% in DMSO, 4.6±1.1% in DAPT; mean±s.d., Mann-Whitney test) (not shown). (J) Most DAPT-treated progenitor cells underwent terminal mitosis in monolayer culture. E14 cerebral cells were cultured with DMSO or DAPT, and after 20 hours BrdU was added for 4 hours to label the progenitor cells. Cells were further cultured with DMSO or DAPT for an additional 20 hours, which allowed the progenitor cells to divide once, and double stained with anti-Ki67 and anti-BrdU. In the DAPT-treated culture, ∼90% of the BrdU-labeled cells were Ki67-negative (*P=0.042, two-tailed, Welch's t-test, n=3; error bars indicate s.d.).