Additional supporting information available online.

STEM_587_sm_suppinfo.doc53KSupporting Information
STEM_587_sm_suppinfoFigS1.tif3974KFigure S1. Characterization of the NKX2.1GFP/w HES3 cell line. (A) Flow cytometric analysis showing expression of stem cell surface markers (ECADHERIN, TRA-1-60, SSEA-4 and CD9) and intracellular OCT4. The percentage of positive cells is shown in the corresponding quadrant. (B) Intracellular flow cytometry showing staining for human NKX2.1 antibody is restricted to GFP+ cells. (C) Hematoxylin and eosin stained histological sections of teratomas generated by the NKX2.1GFP/w line showing derivatives of the three primary germ layers. (D) Metaphase chromosome spread of the cloned NKX2.1GFP/w line showing a normal female karyotype. (E-J) Bright field (BF) and GFP fluorescence images illustrating morphological structures associated with GFP+ cells in d24 EBs formed in the presence of FGF2 and RA. (K, L) RA time course experiment demonstrating the relationship between the time of RA (10-5 M) addition (days of differentiation, D) and the percentage of wells containing GFP+ cells (GFP+ wells) and/or total percentage of GFP+ cells measured at differentiation d24. (M-P) Q-PCR analysis of an RA titration experiment (analysed at differentiation d24) highlighting the different RA concentrations that correlated with expression of SHH, FOXG1, NKX2.1 and PAX6. RGE; relative gene expression. Scale bars: 100 μm (E-J).
STEM_587_sm_suppinfoFigS2.tif1094KFigure S2. Comparison of the transcriptional profiles of undifferentiated NKX2.1GFP/w, HES3 and MEL1 hESC lines. Scatter plots comparing the expression of 48,803 probe sets on Illumina HumanHT-12 v4 Expression BeadChips hybridized against labeled total RNA extracted from undifferentiated NKX2.1GFP/w, HES3 and MEL1 hESCs. (A-C) Scatter plots showing pairwise comparisons of relative signal intensity for (A) NKX2.1GFP/w and HES3, (B) NKX2.1GFP/w and MEL1 and (C) HES3 and MEL1 hESCs. Parallel lines mark 3-fold differences in the relative signal intensity and probe sets falling outside these lines are highlighted. The Venn diagrams adjacent to each panel graphically display the numbers of probe sets differing between each pair of cell lines.
STEM_587_sm_suppinfoFigS3.tif6146KFigure S3. In vivo differentiation of NKX2.1GFP/w RA-treated and FGF-only spin EBs. (A-F) Bright field (BF) and merged BF and GFP fluorescence images illustrating morphological structures associated with RA-treated d8 EBs transplanted under the kidney capsule. Note the presence of multiple GFP+ structures. White arrows indicate pigmented tissue. (G, H) BF and merged BF and GFP fluorescence images illustrating morphological structures associated with transplanted FGF-only EBs. Note the cystic phenotype associated with the graft as well as very little GFP expression. (I, J) Hematoxylin and eosin stained histological sections of grafts derived from an FGF2/RA-treated EBs showing a high frequency of pigmented tissue (I) and multiple neuroepithelial structures (J). (K, L) Immunofluorescence images showing that some of these neuroepithelial structures were composed of NKX2.1 expressing cells. (M, N) Immunofluorescence images showing extensive expression of the neural marker NESTIN within the RA-treated EB grafts. Magnification: A, B: 8.5X; C, D: 15X; E, F: 48X; G, H: 10X; I-N: Scale bar 50μM. These data were generated from 3 transplanted mice.
STEM_587_sm_suppinfoFigS4.tif1598KFigure S4. Differentiation of NKX2.1+ cells from unmanipulated hESC and iPS cell lines. (A-C) Analysis of the MEL1 hESCs differentiated with or without addition of RA between d7-10 (A) Flow cytometric analysis showing the relationship between RA treatment and the persistence of a high frequency of E-CAD expressing cells. FGF2 at 100 ng/ml was included for the first 7 days. The percentage of E-CAD-positive and -negative cells is indicated in each panel. (B) Gene expression analysis using Q-PCR demonstrated that the sequential activation of genes involved in neural induction and patterning paralleled that observed for the genetically modified NKX2.1GFP/w line. Relative gene expressions in cDNA samples from RA-treated cells are drawn as a green line and from non-RA treated cells as a red line. (C) Intracellular flow cytometric analysis of d24 cultures with the human NKX2.1 antibody revealed the presence of an NKX2.1+ population (∼4%). (D-I) Analysis of experiments comparing differentiated H9 and iPS (DF 19-9 7T) cells with NKX2.1GFP/w cells. (D, G, I) Intracellular flow cytometry in (D) differentiated NKX2.1GFP/w cells showing that staining for human NKX2.1 antibody correlates with GFP-expressing cells. A similar frequency of NKX2.1+ cells is observed in (G) differentiated H9 cells but lower numbers of NKX2.1+ cells are seen in (I) iPS cells. (E, F, H) Q-PCR analysis for the indicated genes showing that (E) upregulation of SHH precedes expression of NKX2.1 and ASCL1 in NKX2.1GFP/w cells. Q-PCR analysis for the indicated genes in differentiating (F) H9 and (H) iPS cells showing that the pattern of gene expression in differentiating H9 cells is similar to the NKX2.1GFP/w (see Fig 2H) and MEL1 cell lines but that the iPS cells do not transiently express high levels of PAX6 and SFRP5.
STEM_587_sm_suppinfoFigS5.tif4871KFigure S5. Gene Expression Analysis of Sorted NKX2.1-GFP+ cells. (A) Q-PCR analysis of d24 flow cytometrically sorted cells examining expression of genes associated with basal forebrain development and neurogenesis. RGE; relative gene expression. n = 3, mean ± s.e.m. (B) Bright field (BF) of NKX2.1-GFP+ cells 48 hr after FACS purification. (C-K) Fluorescence images from multiple fields showing that NKX2.1-GFP+ cells 48 hr after FACS purification do not express the telencephalic specific marker FOXG1. Scale bars: 50 μm.
STEM_587_sm_suppinfoFigS6.tif5787KFigure S6. In vivo proliferation and differentiation of GFP+ cells. (A-C) Bright field and GFP fluorescence images of grafts formed from flow cytometrically purified NKX2.1-GFP+ cells transplanted under the kidney capsule of immunodeficient mice, harvested 6 weeks following transplantation. (D-K) Immunofluorescence analysis using anti-TUBULIN and anti-NKX2.1 antibodies demonstrating that the engrafted cells retained an NKX2.1+ neuronal phenotype (boxed panels are magnified in H-K). Scale bars: 50μm in D-K; 20μm in H-K. These data were generated from 6 mice.
STEM_587_sm_suppinfoFigS7.tif5763KFigure S7. In vivo differentiation and in vitro migration of NKX2.1-GFP+ cells. (A-C) Immunofluorescence images documenting the fate of NKX2.1-GFP+ cells transplanted into the cortices of neonatal mice. NKX2.1-GFP+ cells co-expressed GABA and displayed a morphological phenotype that resembled that observed for interneurons. White arrows indicate GABA+ cells. Scale bars: 50 μm. (D) Bright field (BF) image demonstrating the extension of neuronal processes from cell aggregates 72 hr after plating onto ECM-coated tissue culture dishes. Single cells (arrowheads) were found up to 1 mm from the attached aggregate. The length of the side of each square is 100 μm. (E-H) BF and epifluorescence images illustrating the translocation of cell bodies (arrowheads) along neuronal processes of both GFP+ and GFP- cells (white box in E, F is magnified in G, H). (I-L) BF and epifluorescence images demonstrating the lack of migration of NKX2.1-GFP+ cells after 24 hr and 72 hr when plated onto gelatin-coated culture flasks. Scale bars: 100 μm in (F, L); 50 μm in (H).
STEM_587_sm_suppinfoFigS8.tif5029KFigure S8. In vitro differentiation of NKX2.1-GFP+ cells. (A-H) Immunofluorescence images showing the emergence of FOXG1+ cells from flow sorted d24 NKX2.1-GFP+ progenitors cultured for an additional 14 days. FOXG1 expression is seen in both GFP+ and GFP- cells. (I-P) Immunofluorescence images demonstrating the presence of FORSE-1+ cells in cultures 14 days post flow sorting. (Q) Flow cytometric analysis of ∼d90 cultures for FORSE-1 expression. Approximately 50% of cells maintain GFP expression in long-term cultures. (R) Q-PCR of d24 NKX2.1-GFP+ cells differentiated for a further ∼90 days and sorted on the basis of FORSE-1 and GFP. NKX2.1 expression is limited to the GFP+ fractions. FOXG1 is more highly expressed in the FORSE-1+ fractions but is also present in the FORSE-1- populations. The FORSE-1-GFP- fraction has the highest relative expression of OLP-associated genes, PDGFRα and OLIG2. Scale bars: 200 μm in (A-D); 100 μm in (E-H); 50 μm in (I-P).
STEM_587_sm_suppinfoTableS1.doc43KTable S1. Anti-human antibodies used for flow cytometry, immunofluorescence and immunohistochemistry.
STEM_587_sm_suppinfoTableS2.pdf62KTable S2.Gene probe sets varying ≥3-fold between undifferentiated NKX2.1GFP/w, HES3 and MEL1 cell lines.
STEM_587_sm_suppinfoTableS3.pdf819KTable S3. Gene probe sets for microarray comparisons between time points d0, d7 and d10 with or without RA treatment.
STEM_587_sm_suppinfoTableS4.pdf714KTable S4. Gene probe sets for microarray comparisons between d24 EB fractions sorted on E-CAD and GFP expression.
STEM_587_sm_suppinfoTableS5.pdf84KTable S5. Probe set for genes showing elevated expression levels in FORSE-1- GFP-PDGFRα+ cells compared to FORSE-1-GFP-PDGFRα- cells.
STEM_587_sm_suppinfoVideoS1.mp44857KVideo S1. Matrigel dependent migration of NKX2.1-GFP+ cells in vitro. Time lapse imaging movie (72 hr timespan) of merged bright field and fluorescent images showing the migration of neurons away from a reaggregated NKX2.1-GFP+ EB (shown in Fig. 4B, C). Note the formation of a meshwork of cells and processes surrounding the central GFP+ aggregate. Migrating cells are predominantly GFP+, with translocating cell bodies following leading neuronal processes.
STEM_587_sm_suppinfoVideoS2.mp46839KVideo S2. Matrigel dependent migration of NKX2.1-GFP+ cells in vitro. Time lapse imaging movie (72 hr timespan) of merged bright field and fluorescent images showing the migration of neurons away from a reaggregated NKX2.1-GFP+ EB. In comparison to supplemental Video S1, long extending neuronal processes are first to emerge from the NKX2.1-GFP+ EB. Multiple cell bodies track along the preceding neurites (seen clearly bottom left) and there is movement of GFP+ cells within the central cell mass.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.