Supporting Materials and Methods
Fluorescence-Activated Cell Sorting (FACS) Purification of Neural Progenitor Cells.
Retinal, cortical and cerebellar progenitor cells were purified by FACS by DNA content. All tissues were dissociated to a single cell suspension by a combination of trypsin incubation and mechanical trituration. After live staining with the DNA-binding dye Hoechst 33042, cells were sorted into 2N and 4N populations based on DNA content. To test the degree of enrichment for progenitor cells by this procedure, two independent sorts of retinal 4N cells were plated on poly(l-lysine)-coated slides, fixed and stained for the progenitor marker, nestin. To measure the percentage of nestin-positive/Hoechst-positive cells, a total of > 100 cells assayed from over 10 different regions of the slide were counted, indicating that > 95% of sorted cells were strongly nestin positive. Total RNA was extracted by the guanidinium-acid phenol method (Trizol, Life Technologies). This RNA was reverse transcribed to single-stranded cDNA by using the SMART system (Clontech) and amplified by PCR for 16-18 cycles.Microarray Production and Use.
cDNA inserts from » 11,136 clones from the Brain Molecular Anatomy Project clone set (kind gift of Bento Soares, University of Iowa, Ames) and > 600 clones of interest collected in our laboratory were PCR amplified by using T3 and T7 primers to the cloning vector, with a success rate of > 95%. PCR products were purified by ethanol precipitation and washing, and resuspended in 3´ SSC (1-4). Purified PCR products were printed on poly(l-lysine) coated glass slides (made according to the protocol described at www.microarrays.org) on a QArray microarrayer (Genetix, UK). Approximately 10 m g of cDNA prepared from either whole tissue or populations of cells was labeled by incorporation of Cy3- or Cy5-dCTP (Amersham Pharmacia) as described (5). Pairs of labeled probes were hybridized at 42°C overnight and washed in 0.2´ SSC/0.1%SDS at room temperature, followed by two room temperature washes in 0.2´ SSC. Slides were scanned in an Axon GenePix 4000B Scanner (Axon Instruments) and data extracted from the resulting images using the genepix software package (Axon Instruments). Data were stored locally in the Acuity database and analysis package (Axon Instruments).Data Analysis and Bioinformatics.
Data from each hybridization were normalized by print-tip using the lowess implementation in Acuity and the ratio of median gene expression values extracted for each feature. The ratio value used for these studies is the median foreground intensity minus the median local background. Expression ratios from features called absent by the spot finding algorithm and also poor quality features were removed from the data set. Poor quality features were identified using a set of empirically derived filtration criteria (sum of median intensities >500; percentage of spot pixels higher than local background plus one standard deviation of the background intensity in either channel greater than 50%).To focus the analysis on genes that show differences in expression between progenitor cells and their progeny, the remaining data were filtered to remove transcripts/spots whose expression was not 1.5-fold higher or lower than its reference sample in at least one of hybridizations in the data set, and whose expression was 1.3-fold higher or lower in at least one third of the hybridizations. Statistical analyses of repeated experiments to identify genes commonly expressed in samples or differentially expressed between groups of samples were carried out using the Significance Analysis of Microarrays package (SAM) (6). Data from SAM were extracted using the Samster tool (for documentation, see http://falkow.stanford.edu/whatwedo/software/programs/samster.pdf) before being hierarchically clustered. Hierarchical clustering was carried out on the log-2 transformed values using the Cluster, xcluster and treeview packages (7) (for a review of the different clustering techniques applied, see ref. 8). Annotation of clones was carried out by using NCBI’s UniGene database and the Stanford online SOURCE resource (www.genome-stanford.edu). Sequences of clones with no UniGene identifiers were compared to Ensembl using the SSAHA search to annotate them further, where possible. Extraction of Gene Ontology (GO) annotations and grouping of genes by GO category was carried out with GOMiner (9) or Fatigo (http://fatigo.bioinfo.cnio.es/) using UniGene symbols as gene identifiers.
In Situ Hybridization.
Non-radioactive in situ hybridization was carried out on tissue sections as described with minor modifications (10). Hybridization with digoxygenin-labeled cRNA probes was performed overnight at 70°C. Hybridized probe was detected with alkaline phosphatase-conjugated anti-DIG antibodies (Boehringer Mannheim), and NBT-BCIP substrate.1. Schena, M., Shalon, D., Davis, R. W. & Brown, P. O. (1995) Science 270, 467-470.
2. DeRisi, J., Penland, L., Brown, P. O., Bittner, M. L., Meltzer, P. S., Ray, M., Chen, Y., Su, Y. A. & Trent, J. M. (1996) Nat. Genet. 14, 457-460.
3. Shalon, D., Smith, S. J. & Brown, P. O. (1996) Genome Res. 6, 639-645.
4. Schena, M., Shalon, D., Heller, R., Chai, A., Brown, P. O. & Davis, R. W. (1996) Proc. Natl. Acad. Sci. USA 93, 10614-10619.
5. Livesey, F. J., Furukawa, T., Steffen, M. A., Church, G. M. & Cepko, C. L. (2000) Curr. Biol. 10, 301-310.
6. Tusher, V. G., Tibshirani, R. & Chu, G. (2001) Proc. Natl. Acad. Sci. USA 98, 5116-5121.
7. Eisen, M. B., Spellman, P. T., Brown, P. O. & Botstein, D. (1998) Proc. Natl. Acad. Sci. USA 95, 14863-14868.
8. Sherlock, G. (2000) Curr. Opin. Immunol. 12, 201-205.
9. Zeeberg, B. R., Feng, W., Wang, G., Wang, M. D., Fojo, A. T., Sunshine, M., Narasimhan, S., Kane, D. W., Reinhold, W. C., Lababidi, S., et al. (2003) Genome Biol. 4, R28.
10. Bao, Z. Z. & Cepko, C. L. (1997) J. Neurosci. 17, 1425-1434.