DEV028548 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Knockdown of REST protein levels in wild-type ES cells by RNAi. (A) Western blot analysis of REST protein levels in wild-type 46C ES cells transfected with siRNA targeting a control sequence (siCtrl) or Rest (siREST). siRNA was from Dharmacon/Thermo Scientific, Lafayette, CO, USA. ES cells were transfected with DharmaFECT 1 as recommended by the manufacturer and the cells harvested after 48 hours. (B) Western blot analysis of REST protein levels in wild-type OS25 ES cells transfected with shRNA constructs targeting either a control sequence (shCtrl) or Rest (shREST). pSuper(Neo-IRES-GFP)-based shRNA constructs were introduced into ES cells using Lipofectamine (Invitrogen, Paisley, UK) as described by the manufacturer. A mixture of two shRNA constructs was used to deplete REST (shREST: target sequence 1, GCGCTAAGAAGTTCTTTG and target sequence 2, GGAGAACGCCCGTATAAA) and the control construct (shCtrl) targets the sequence GCGCGCTTTGTAGGATTCG. Three days after transfection, GFP-positive cells were isolated by FACS. Cell populations were analysed for purity (>95%) after sorting. Anti-lamin antibody was used to demonstrate equal loading.

• Supplemental Figure S2 -

  Fig. S2. Replication timing of control genes in wild-type and Rest^−/− ES cells. Early (Oct4) and late (Myf5) replicating loci displayed maximal signal in G1−S1 and S4 fractions, respectively, in both wild-type (WT) and Rest^−/− ES cells. Bar charts show the relative amount of newly replicated locus-specific DNA in G1, four sequential fractions of S-phase (S1-S4) and G2 in wild-type and Rest^−/− ES cells. The error bars represent s.d. of two experiments.

• Supplemental Figure S3 -

  Fig. S3. Histone marks are retained in REST-deficient ES cells. Levels of modified histones were assessed by ChIP using antibodies against active histone marks (H3K4me3, H3K4me2 and H3K9ac) or H3K27me3, which is associated with repressed chromatin, in wild-type and Rest^−/− ES cells. Enrichment (relative to total histone H3) at the promoter of expressed (Oct4) and non-expressed (Myf5) genes are shown as controls for Math1 and Mash1. Bars show the average of two independent ChIP experiments and error bars indicate s.d.

• Supplemental Figure S4 -

  Fig. S4. Genes showing misregulation in REST-deficient ES cells are enriched for REST binding sites and display brain-specific expression. (A) Gene symbols, Affymetrix probe IDs, fold change (FC) and P-values for the 87 genes that are up- or downregulated by at least 1.4-fold in Rest^−/− and shREST-transfected ES cells as compared with controls. Affymetrix gene expression analysis was performed in triplicate for each sample (Rest^−/−, wild type, shREST and shCtrl) and analysed with Limma (Bioconductor) (Gentleman et al., 2004; Smyth, 2004). The raw data were normalised using Robust Multi-Array (RMA) (Irizarry et al., 2003) and filtered using the nsFilter (http://rss.acs.unt.edu/Rdoc/library/genefilter/html/00Index.html) with the IQR method and a cut-off of 0.5. Individual P-values for the remaining probe set (17,411 probes) were obtained by applying an empirical Bayes-moderated t-test (Smyth, 2004). A cut-off of 0.05 for the moderated P-value and of 0.5 for absolute log₂(fold change), which correspond to a 1.4-fold change, resulted in 594 unique genes for Rest^−/− versus wild type and in 429 unique genes for shREST versus shCtrl. Eighty-seven genes were either up- or downregulated in both Rest^−/− versus wild type and in shREST versus shCtrl. The analysis for enrichment of RE1 consensus sites at these 87 genes was based on the data published by Otto et al. (Otto et al., 2007). Genes containing an RE1 site within a region from −1 kb to +5 kb of the transcription start site covered by the Affymetrix Mouse 430 2.0 Array (21,248 unique gene symbols) were selected (511 genes in total). Twenty-five of 87 genes met these criteria (marked by × in the table), which corresponds to a significant association between REST-regulated genes that are differentially expressed in Rest^−/− versus wild type and shREST versus shCtrl and the presence of a RE1 site (P=2×10⁻²⁰, one-tailed Fisher’s exact test). (B) REST-regulated genes are significantly enriched for brain-expressed transcripts. Screenshot of expression analysis using the 'tissue expression' function in DAVID (http://david.abcc.ncifcrf.gov) with the 87 REST-regulated genes as input and all considered Affymetrix probes (17,411) as background. For each of the enriched expression categories, the proportion (Genes), number (Count) and fraction of REST-regulated genes (%) is displayed alongside the P-value before (P-Value) and after Bonferroni correction for multiple testing.

  References

  Gentleman, R. C., Carey, V. J., Bates, D. M., Bolstad, B., Dettling, M., Dudoit, S., Ellis, B., Gautier, L., Ge, Y., Gentry, J. et al. (2004). Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5, R80.

  Irizarry, R. A., Hobbs, B., Collin, F., Beazer-Barclay, Y. D., Antonellis, K. J., Scherf, U. and Speed, T. P. (2003). Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4, 249-264.

  Otto, S. J., McCorkle, S. R., Hover, J., Conaco, C., Han, J. J., Impey, S., Yochum, G. S., Dunn, J. J., Goodman, R. H. and Mandel, G. (2007). A new binding motif for the transcriptional repressor REST uncovers large gene networks devoted to neuronal functions. J. Neurosci. 27, 6729-6739.

  Smyth, G. K. (2004). Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat. Appl. Genet. Mol. Biol. 3, Article3.

• Supplemental Figure S5 -

  Fig. S5. Overexpression of Flag-REST in wild-type and Rest^−/− ES cells. Western blot of extracts from wild-type or Rest^−/− ES cells transfected with a Flag-REST expression construct or a control vector. Full-length REST protein is indicated with an arrow and the non-functional truncated REST peptides arising from the targeted REST allele are indicated by arrowheads. The Flag-REST expression construct or control vector was co-transfected with a GFP expression vector using Lipofectamine as described by the manufacturer (Invitrogen). Three days after transfection, the cells were harvested and GFP-expressing cells isolated by FACS.

• Supplemental Figure S6 -

  Fig. S6. Rest^−/− ES cells express pluripotency markers. FACS analysis of wild-type (WT), Rest^+/− and Rest^−/− ES cells stained for stem cell markers SSEA1 (left panels) or OCT4 (right panels) as previously described (Jorgensen et al., 2007). The black lines show staining with the relevant primary antibody and grey lines staining with control antibody. Staining of a B-cell line (B3) is included as a negative control.

• Supplemental Table S1 - Adobe PDF

• Supplemental Figure S7 -

  Fig. S7. Gene expression changes during embryoid body differentiation of wild-type and Rest^−/−ES cells. Levels of pluripotency- (Nanog) and differentiation-associated transcripts (Pax6, Ngn1, Mixl1, Sox17) detected in wild-type or Rest^−/− undifferentiated ES cells, and following embryoid body formation. Retinoic acid was added at day 4 of differentiation where indicated (RA). For comparison, the expression levels in control tissues (Ctrl), E15 heads (Pax6, Ngn1), ES cell-derived mesoderm (Mixl1), E15 liver (Sox17), are provided (grey). The expression levels were normalised to house keeping controls (Hmbs, Ywhaz, Gapdh). The average and s.d. from two to three experiments are shown.